{
  "best_global_step": 145440,
  "best_metric": 0.23126927018165588,
  "best_model_checkpoint": "saves_multiple/prompt-tuning/llama-3-8b-instruct/train_winogrande_101112_1760638068/checkpoint-145440",
  "epoch": 20.0,
  "eval_steps": 9090,
  "global_step": 181800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00055005500550055,
      "grad_norm": 53.25,
      "learning_rate": 6.6006600660066e-06,
      "loss": 7.573,
      "num_input_tokens_seen": 1056,
      "step": 5
    },
    {
      "epoch": 0.0011001100110011,
      "grad_norm": 69.0,
      "learning_rate": 1.4851485148514851e-05,
      "loss": 7.2714,
      "num_input_tokens_seen": 2208,
      "step": 10
    },
    {
      "epoch": 0.0016501650165016502,
      "grad_norm": 120.0,
      "learning_rate": 2.31023102310231e-05,
      "loss": 7.0963,
      "num_input_tokens_seen": 3264,
      "step": 15
    },
    {
      "epoch": 0.0022002200220022,
      "grad_norm": 76.0,
      "learning_rate": 3.135313531353135e-05,
      "loss": 6.3597,
      "num_input_tokens_seen": 4288,
      "step": 20
    },
    {
      "epoch": 0.0027502750275027505,
      "grad_norm": 53.75,
      "learning_rate": 3.9603960396039605e-05,
      "loss": 5.2557,
      "num_input_tokens_seen": 5408,
      "step": 25
    },
    {
      "epoch": 0.0033003300330033004,
      "grad_norm": 64.0,
      "learning_rate": 4.785478547854785e-05,
      "loss": 4.5747,
      "num_input_tokens_seen": 6464,
      "step": 30
    },
    {
      "epoch": 0.0038503850385038503,
      "grad_norm": 60.5,
      "learning_rate": 5.6105610561056106e-05,
      "loss": 4.0558,
      "num_input_tokens_seen": 7488,
      "step": 35
    },
    {
      "epoch": 0.0044004400440044,
      "grad_norm": 266.0,
      "learning_rate": 6.435643564356436e-05,
      "loss": 3.3519,
      "num_input_tokens_seen": 8576,
      "step": 40
    },
    {
      "epoch": 0.0049504950495049506,
      "grad_norm": 114.5,
      "learning_rate": 7.260726072607261e-05,
      "loss": 2.497,
      "num_input_tokens_seen": 9664,
      "step": 45
    },
    {
      "epoch": 0.005500550055005501,
      "grad_norm": 67.0,
      "learning_rate": 8.085808580858085e-05,
      "loss": 1.5064,
      "num_input_tokens_seen": 10752,
      "step": 50
    },
    {
      "epoch": 0.00605060506050605,
      "grad_norm": 19.25,
      "learning_rate": 8.91089108910891e-05,
      "loss": 0.4453,
      "num_input_tokens_seen": 11744,
      "step": 55
    },
    {
      "epoch": 0.006600660066006601,
      "grad_norm": 22.25,
      "learning_rate": 9.735973597359736e-05,
      "loss": 0.2919,
      "num_input_tokens_seen": 12800,
      "step": 60
    },
    {
      "epoch": 0.007150715071507151,
      "grad_norm": 35.0,
      "learning_rate": 0.0001056105610561056,
      "loss": 0.2299,
      "num_input_tokens_seen": 13888,
      "step": 65
    },
    {
      "epoch": 0.007700770077007701,
      "grad_norm": 66.0,
      "learning_rate": 0.00011386138613861385,
      "loss": 0.2327,
      "num_input_tokens_seen": 14944,
      "step": 70
    },
    {
      "epoch": 0.00825082508250825,
      "grad_norm": 55.25,
      "learning_rate": 0.0001221122112211221,
      "loss": 0.3055,
      "num_input_tokens_seen": 16032,
      "step": 75
    },
    {
      "epoch": 0.0088008800880088,
      "grad_norm": 15.8125,
      "learning_rate": 0.00013036303630363036,
      "loss": 0.29,
      "num_input_tokens_seen": 17088,
      "step": 80
    },
    {
      "epoch": 0.00935093509350935,
      "grad_norm": 61.75,
      "learning_rate": 0.0001386138613861386,
      "loss": 0.2153,
      "num_input_tokens_seen": 18112,
      "step": 85
    },
    {
      "epoch": 0.009900990099009901,
      "grad_norm": 23.75,
      "learning_rate": 0.00014686468646864687,
      "loss": 0.2414,
      "num_input_tokens_seen": 19168,
      "step": 90
    },
    {
      "epoch": 0.010451045104510451,
      "grad_norm": 9.5625,
      "learning_rate": 0.0001551155115511551,
      "loss": 0.3088,
      "num_input_tokens_seen": 20192,
      "step": 95
    },
    {
      "epoch": 0.011001100110011002,
      "grad_norm": 42.75,
      "learning_rate": 0.00016336633663366335,
      "loss": 0.2594,
      "num_input_tokens_seen": 21216,
      "step": 100
    },
    {
      "epoch": 0.01155115511551155,
      "grad_norm": 17.75,
      "learning_rate": 0.00017161716171617162,
      "loss": 0.2222,
      "num_input_tokens_seen": 22272,
      "step": 105
    },
    {
      "epoch": 0.0121012101210121,
      "grad_norm": 60.5,
      "learning_rate": 0.00017986798679867986,
      "loss": 0.1591,
      "num_input_tokens_seen": 23264,
      "step": 110
    },
    {
      "epoch": 0.012651265126512651,
      "grad_norm": 173.0,
      "learning_rate": 0.00018811881188118812,
      "loss": 0.3237,
      "num_input_tokens_seen": 24288,
      "step": 115
    },
    {
      "epoch": 0.013201320132013201,
      "grad_norm": 27.5,
      "learning_rate": 0.00019636963696369636,
      "loss": 0.1632,
      "num_input_tokens_seen": 25408,
      "step": 120
    },
    {
      "epoch": 0.013751375137513752,
      "grad_norm": 10.0,
      "learning_rate": 0.00020462046204620463,
      "loss": 0.2341,
      "num_input_tokens_seen": 26464,
      "step": 125
    },
    {
      "epoch": 0.014301430143014302,
      "grad_norm": 24.375,
      "learning_rate": 0.00021287128712871284,
      "loss": 0.2459,
      "num_input_tokens_seen": 27552,
      "step": 130
    },
    {
      "epoch": 0.01485148514851485,
      "grad_norm": 8.6875,
      "learning_rate": 0.0002211221122112211,
      "loss": 0.2354,
      "num_input_tokens_seen": 28608,
      "step": 135
    },
    {
      "epoch": 0.015401540154015401,
      "grad_norm": 32.75,
      "learning_rate": 0.00022937293729372935,
      "loss": 0.2266,
      "num_input_tokens_seen": 29664,
      "step": 140
    },
    {
      "epoch": 0.01595159515951595,
      "grad_norm": 13.1875,
      "learning_rate": 0.00023762376237623762,
      "loss": 0.2084,
      "num_input_tokens_seen": 30720,
      "step": 145
    },
    {
      "epoch": 0.0165016501650165,
      "grad_norm": 129.0,
      "learning_rate": 0.00024587458745874586,
      "loss": 0.5611,
      "num_input_tokens_seen": 31712,
      "step": 150
    },
    {
      "epoch": 0.017051705170517052,
      "grad_norm": 24.0,
      "learning_rate": 0.00025412541254125415,
      "loss": 0.3313,
      "num_input_tokens_seen": 32704,
      "step": 155
    },
    {
      "epoch": 0.0176017601760176,
      "grad_norm": 6.34375,
      "learning_rate": 0.00026237623762376234,
      "loss": 0.2729,
      "num_input_tokens_seen": 33696,
      "step": 160
    },
    {
      "epoch": 0.018151815181518153,
      "grad_norm": 4.71875,
      "learning_rate": 0.00027062706270627063,
      "loss": 0.2743,
      "num_input_tokens_seen": 34752,
      "step": 165
    },
    {
      "epoch": 0.0187018701870187,
      "grad_norm": 4.96875,
      "learning_rate": 0.00027887788778877887,
      "loss": 0.2669,
      "num_input_tokens_seen": 35872,
      "step": 170
    },
    {
      "epoch": 0.019251925192519254,
      "grad_norm": 4.5625,
      "learning_rate": 0.0002871287128712871,
      "loss": 0.3345,
      "num_input_tokens_seen": 36928,
      "step": 175
    },
    {
      "epoch": 0.019801980198019802,
      "grad_norm": 4.8125,
      "learning_rate": 0.00029537953795379535,
      "loss": 0.3014,
      "num_input_tokens_seen": 37952,
      "step": 180
    },
    {
      "epoch": 0.02035203520352035,
      "grad_norm": 5.0,
      "learning_rate": 0.0003036303630363036,
      "loss": 0.2291,
      "num_input_tokens_seen": 38944,
      "step": 185
    },
    {
      "epoch": 0.020902090209020903,
      "grad_norm": 2.84375,
      "learning_rate": 0.0003118811881188119,
      "loss": 0.2266,
      "num_input_tokens_seen": 40000,
      "step": 190
    },
    {
      "epoch": 0.02145214521452145,
      "grad_norm": 6.53125,
      "learning_rate": 0.00032013201320132013,
      "loss": 0.1716,
      "num_input_tokens_seen": 41056,
      "step": 195
    },
    {
      "epoch": 0.022002200220022004,
      "grad_norm": 23.125,
      "learning_rate": 0.0003283828382838284,
      "loss": 0.2566,
      "num_input_tokens_seen": 42080,
      "step": 200
    },
    {
      "epoch": 0.022552255225522552,
      "grad_norm": 7.78125,
      "learning_rate": 0.0003366336633663366,
      "loss": 0.1998,
      "num_input_tokens_seen": 43136,
      "step": 205
    },
    {
      "epoch": 0.0231023102310231,
      "grad_norm": 4.25,
      "learning_rate": 0.00034488448844884485,
      "loss": 0.2224,
      "num_input_tokens_seen": 44224,
      "step": 210
    },
    {
      "epoch": 0.023652365236523653,
      "grad_norm": 58.5,
      "learning_rate": 0.00035313531353135314,
      "loss": 0.1312,
      "num_input_tokens_seen": 45312,
      "step": 215
    },
    {
      "epoch": 0.0242024202420242,
      "grad_norm": 3.546875,
      "learning_rate": 0.0003613861386138614,
      "loss": 0.125,
      "num_input_tokens_seen": 46368,
      "step": 220
    },
    {
      "epoch": 0.024752475247524754,
      "grad_norm": 10.5625,
      "learning_rate": 0.0003696369636963696,
      "loss": 0.2023,
      "num_input_tokens_seen": 47488,
      "step": 225
    },
    {
      "epoch": 0.025302530253025302,
      "grad_norm": 14.0625,
      "learning_rate": 0.00037788778877887786,
      "loss": 1.7313,
      "num_input_tokens_seen": 48512,
      "step": 230
    },
    {
      "epoch": 0.02585258525852585,
      "grad_norm": 4.40625,
      "learning_rate": 0.0003861386138613861,
      "loss": 0.1979,
      "num_input_tokens_seen": 49568,
      "step": 235
    },
    {
      "epoch": 0.026402640264026403,
      "grad_norm": 33.0,
      "learning_rate": 0.0003943894389438944,
      "loss": 0.3624,
      "num_input_tokens_seen": 50656,
      "step": 240
    },
    {
      "epoch": 0.02695269526952695,
      "grad_norm": 3.125,
      "learning_rate": 0.00040264026402640264,
      "loss": 0.2496,
      "num_input_tokens_seen": 51712,
      "step": 245
    },
    {
      "epoch": 0.027502750275027504,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004108910891089109,
      "loss": 0.2112,
      "num_input_tokens_seen": 52736,
      "step": 250
    },
    {
      "epoch": 0.028052805280528052,
      "grad_norm": 27.0,
      "learning_rate": 0.0004191419141914191,
      "loss": 0.2054,
      "num_input_tokens_seen": 53824,
      "step": 255
    },
    {
      "epoch": 0.028602860286028604,
      "grad_norm": 12.3125,
      "learning_rate": 0.00042739273927392736,
      "loss": 0.267,
      "num_input_tokens_seen": 54880,
      "step": 260
    },
    {
      "epoch": 0.029152915291529153,
      "grad_norm": 4.21875,
      "learning_rate": 0.00043564356435643565,
      "loss": 0.1549,
      "num_input_tokens_seen": 56000,
      "step": 265
    },
    {
      "epoch": 0.0297029702970297,
      "grad_norm": 3.25,
      "learning_rate": 0.00044389438943894384,
      "loss": 0.2135,
      "num_input_tokens_seen": 56992,
      "step": 270
    },
    {
      "epoch": 0.030253025302530254,
      "grad_norm": 10.875,
      "learning_rate": 0.00045214521452145213,
      "loss": 0.2389,
      "num_input_tokens_seen": 58080,
      "step": 275
    },
    {
      "epoch": 0.030803080308030802,
      "grad_norm": 3.796875,
      "learning_rate": 0.00046039603960396037,
      "loss": 0.112,
      "num_input_tokens_seen": 59168,
      "step": 280
    },
    {
      "epoch": 0.03135313531353135,
      "grad_norm": 65.5,
      "learning_rate": 0.00046864686468646867,
      "loss": 1.226,
      "num_input_tokens_seen": 60288,
      "step": 285
    },
    {
      "epoch": 0.0319031903190319,
      "grad_norm": 232.0,
      "learning_rate": 0.00047689768976897685,
      "loss": 1.6272,
      "num_input_tokens_seen": 61280,
      "step": 290
    },
    {
      "epoch": 0.032453245324532455,
      "grad_norm": 26.75,
      "learning_rate": 0.0004851485148514851,
      "loss": 0.3609,
      "num_input_tokens_seen": 62272,
      "step": 295
    },
    {
      "epoch": 0.033003300330033,
      "grad_norm": 28.875,
      "learning_rate": 0.0004933993399339934,
      "loss": 0.3283,
      "num_input_tokens_seen": 63328,
      "step": 300
    },
    {
      "epoch": 0.03355335533553355,
      "grad_norm": 11.8125,
      "learning_rate": 0.0005016501650165016,
      "loss": 0.3531,
      "num_input_tokens_seen": 64352,
      "step": 305
    },
    {
      "epoch": 0.034103410341034104,
      "grad_norm": 22.25,
      "learning_rate": 0.0005099009900990098,
      "loss": 0.5056,
      "num_input_tokens_seen": 65440,
      "step": 310
    },
    {
      "epoch": 0.034653465346534656,
      "grad_norm": 28.25,
      "learning_rate": 0.0005181518151815181,
      "loss": 0.2751,
      "num_input_tokens_seen": 66528,
      "step": 315
    },
    {
      "epoch": 0.0352035203520352,
      "grad_norm": 4.5,
      "learning_rate": 0.0005264026402640264,
      "loss": 0.3104,
      "num_input_tokens_seen": 67552,
      "step": 320
    },
    {
      "epoch": 0.035753575357535754,
      "grad_norm": 5.90625,
      "learning_rate": 0.0005346534653465346,
      "loss": 0.2096,
      "num_input_tokens_seen": 68576,
      "step": 325
    },
    {
      "epoch": 0.036303630363036306,
      "grad_norm": 5.84375,
      "learning_rate": 0.0005429042904290429,
      "loss": 0.2349,
      "num_input_tokens_seen": 69600,
      "step": 330
    },
    {
      "epoch": 0.03685368536853685,
      "grad_norm": 30.25,
      "learning_rate": 0.0005511551155115512,
      "loss": 0.2485,
      "num_input_tokens_seen": 70624,
      "step": 335
    },
    {
      "epoch": 0.0374037403740374,
      "grad_norm": 19.5,
      "learning_rate": 0.0005594059405940595,
      "loss": 0.2845,
      "num_input_tokens_seen": 71680,
      "step": 340
    },
    {
      "epoch": 0.037953795379537955,
      "grad_norm": 5.46875,
      "learning_rate": 0.0005676567656765677,
      "loss": 0.283,
      "num_input_tokens_seen": 72672,
      "step": 345
    },
    {
      "epoch": 0.03850385038503851,
      "grad_norm": 980.0,
      "learning_rate": 0.0005759075907590758,
      "loss": 0.4062,
      "num_input_tokens_seen": 73696,
      "step": 350
    },
    {
      "epoch": 0.03905390539053905,
      "grad_norm": 7.03125,
      "learning_rate": 0.0005841584158415841,
      "loss": 2.4129,
      "num_input_tokens_seen": 74752,
      "step": 355
    },
    {
      "epoch": 0.039603960396039604,
      "grad_norm": 47.5,
      "learning_rate": 0.0005924092409240923,
      "loss": 0.4721,
      "num_input_tokens_seen": 75776,
      "step": 360
    },
    {
      "epoch": 0.040154015401540157,
      "grad_norm": 12.625,
      "learning_rate": 0.0006006600660066006,
      "loss": 0.2404,
      "num_input_tokens_seen": 76864,
      "step": 365
    },
    {
      "epoch": 0.0407040704070407,
      "grad_norm": 462.0,
      "learning_rate": 0.0006089108910891089,
      "loss": 0.7585,
      "num_input_tokens_seen": 77952,
      "step": 370
    },
    {
      "epoch": 0.041254125412541254,
      "grad_norm": 3.078125,
      "learning_rate": 0.0006171617161716172,
      "loss": 0.2401,
      "num_input_tokens_seen": 79040,
      "step": 375
    },
    {
      "epoch": 0.041804180418041806,
      "grad_norm": 18.0,
      "learning_rate": 0.0006254125412541254,
      "loss": 0.1257,
      "num_input_tokens_seen": 80096,
      "step": 380
    },
    {
      "epoch": 0.04235423542354235,
      "grad_norm": 28.375,
      "learning_rate": 0.0006336633663366337,
      "loss": 0.3759,
      "num_input_tokens_seen": 81184,
      "step": 385
    },
    {
      "epoch": 0.0429042904290429,
      "grad_norm": 8.5,
      "learning_rate": 0.000641914191419142,
      "loss": 0.5493,
      "num_input_tokens_seen": 82240,
      "step": 390
    },
    {
      "epoch": 0.043454345434543455,
      "grad_norm": 29.375,
      "learning_rate": 0.0006501650165016501,
      "loss": 0.2787,
      "num_input_tokens_seen": 83328,
      "step": 395
    },
    {
      "epoch": 0.04400440044004401,
      "grad_norm": 3.03125,
      "learning_rate": 0.0006584158415841584,
      "loss": 0.3091,
      "num_input_tokens_seen": 84352,
      "step": 400
    },
    {
      "epoch": 0.04455445544554455,
      "grad_norm": 129.0,
      "learning_rate": 0.0006666666666666666,
      "loss": 0.9364,
      "num_input_tokens_seen": 85408,
      "step": 405
    },
    {
      "epoch": 0.045104510451045104,
      "grad_norm": 2.078125,
      "learning_rate": 0.0006749174917491748,
      "loss": 0.3263,
      "num_input_tokens_seen": 86432,
      "step": 410
    },
    {
      "epoch": 0.04565456545654566,
      "grad_norm": 0.71484375,
      "learning_rate": 0.0006831683168316831,
      "loss": 0.2757,
      "num_input_tokens_seen": 87488,
      "step": 415
    },
    {
      "epoch": 0.0462046204620462,
      "grad_norm": 2.203125,
      "learning_rate": 0.0006914191419141914,
      "loss": 0.2653,
      "num_input_tokens_seen": 88480,
      "step": 420
    },
    {
      "epoch": 0.046754675467546754,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0006996699669966997,
      "loss": 0.2429,
      "num_input_tokens_seen": 89504,
      "step": 425
    },
    {
      "epoch": 0.047304730473047306,
      "grad_norm": 2.03125,
      "learning_rate": 0.0007079207920792079,
      "loss": 0.2297,
      "num_input_tokens_seen": 90560,
      "step": 430
    },
    {
      "epoch": 0.04785478547854786,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0007161716171617162,
      "loss": 0.244,
      "num_input_tokens_seen": 91584,
      "step": 435
    },
    {
      "epoch": 0.0484048404840484,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0007244224422442244,
      "loss": 0.243,
      "num_input_tokens_seen": 92640,
      "step": 440
    },
    {
      "epoch": 0.048954895489548955,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0007326732673267326,
      "loss": 0.294,
      "num_input_tokens_seen": 93664,
      "step": 445
    },
    {
      "epoch": 0.04950495049504951,
      "grad_norm": 1.109375,
      "learning_rate": 0.0007409240924092409,
      "loss": 0.2436,
      "num_input_tokens_seen": 94752,
      "step": 450
    },
    {
      "epoch": 0.05005500550055005,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0007491749174917492,
      "loss": 0.241,
      "num_input_tokens_seen": 95808,
      "step": 455
    },
    {
      "epoch": 0.050605060506050605,
      "grad_norm": 99.0,
      "learning_rate": 0.0007574257425742574,
      "loss": 0.2546,
      "num_input_tokens_seen": 96800,
      "step": 460
    },
    {
      "epoch": 0.05115511551155116,
      "grad_norm": 0.458984375,
      "learning_rate": 0.0007656765676567656,
      "loss": 0.2405,
      "num_input_tokens_seen": 97888,
      "step": 465
    },
    {
      "epoch": 0.0517051705170517,
      "grad_norm": 142.0,
      "learning_rate": 0.0007739273927392739,
      "loss": 0.2863,
      "num_input_tokens_seen": 99008,
      "step": 470
    },
    {
      "epoch": 0.052255225522552254,
      "grad_norm": 0.126953125,
      "learning_rate": 0.0007821782178217822,
      "loss": 0.2345,
      "num_input_tokens_seen": 100096,
      "step": 475
    },
    {
      "epoch": 0.052805280528052806,
      "grad_norm": 0.15625,
      "learning_rate": 0.0007904290429042904,
      "loss": 0.2351,
      "num_input_tokens_seen": 101152,
      "step": 480
    },
    {
      "epoch": 0.05335533553355336,
      "grad_norm": 0.349609375,
      "learning_rate": 0.0007986798679867986,
      "loss": 0.2344,
      "num_input_tokens_seen": 102208,
      "step": 485
    },
    {
      "epoch": 0.0539053905390539,
      "grad_norm": 0.216796875,
      "learning_rate": 0.0008069306930693069,
      "loss": 0.2289,
      "num_input_tokens_seen": 103232,
      "step": 490
    },
    {
      "epoch": 0.054455445544554455,
      "grad_norm": 0.27734375,
      "learning_rate": 0.0008151815181518151,
      "loss": 0.2333,
      "num_input_tokens_seen": 104320,
      "step": 495
    },
    {
      "epoch": 0.05500550055005501,
      "grad_norm": 0.0966796875,
      "learning_rate": 0.0008234323432343234,
      "loss": 0.2305,
      "num_input_tokens_seen": 105376,
      "step": 500
    },
    {
      "epoch": 0.05555555555555555,
      "grad_norm": 0.09326171875,
      "learning_rate": 0.0008316831683168317,
      "loss": 0.2346,
      "num_input_tokens_seen": 106432,
      "step": 505
    },
    {
      "epoch": 0.056105610561056105,
      "grad_norm": 0.478515625,
      "learning_rate": 0.00083993399339934,
      "loss": 0.2286,
      "num_input_tokens_seen": 107392,
      "step": 510
    },
    {
      "epoch": 0.05665566556655666,
      "grad_norm": 0.1181640625,
      "learning_rate": 0.0008481848184818481,
      "loss": 0.2491,
      "num_input_tokens_seen": 108416,
      "step": 515
    },
    {
      "epoch": 0.05720572057205721,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0008564356435643564,
      "loss": 0.2231,
      "num_input_tokens_seen": 109504,
      "step": 520
    },
    {
      "epoch": 0.057755775577557754,
      "grad_norm": 1.40625,
      "learning_rate": 0.0008646864686468647,
      "loss": 0.2349,
      "num_input_tokens_seen": 110528,
      "step": 525
    },
    {
      "epoch": 0.058305830583058306,
      "grad_norm": 0.251953125,
      "learning_rate": 0.0008729372937293728,
      "loss": 0.2393,
      "num_input_tokens_seen": 111520,
      "step": 530
    },
    {
      "epoch": 0.05885588558855886,
      "grad_norm": 0.2421875,
      "learning_rate": 0.0008811881188118811,
      "loss": 0.2325,
      "num_input_tokens_seen": 112512,
      "step": 535
    },
    {
      "epoch": 0.0594059405940594,
      "grad_norm": 0.26171875,
      "learning_rate": 0.0008894389438943894,
      "loss": 0.2394,
      "num_input_tokens_seen": 113536,
      "step": 540
    },
    {
      "epoch": 0.059955995599559955,
      "grad_norm": 0.44921875,
      "learning_rate": 0.0008976897689768977,
      "loss": 0.2364,
      "num_input_tokens_seen": 114592,
      "step": 545
    },
    {
      "epoch": 0.06050605060506051,
      "grad_norm": 0.427734375,
      "learning_rate": 0.0009059405940594059,
      "loss": 0.2447,
      "num_input_tokens_seen": 115680,
      "step": 550
    },
    {
      "epoch": 0.06105610561056106,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0009141914191419142,
      "loss": 0.2443,
      "num_input_tokens_seen": 116672,
      "step": 555
    },
    {
      "epoch": 0.061606160616061605,
      "grad_norm": 0.3515625,
      "learning_rate": 0.0009224422442244225,
      "loss": 0.2302,
      "num_input_tokens_seen": 117696,
      "step": 560
    },
    {
      "epoch": 0.06215621562156216,
      "grad_norm": 0.2373046875,
      "learning_rate": 0.0009306930693069307,
      "loss": 0.2403,
      "num_input_tokens_seen": 118720,
      "step": 565
    },
    {
      "epoch": 0.0627062706270627,
      "grad_norm": 0.2275390625,
      "learning_rate": 0.0009389438943894391,
      "loss": 0.2281,
      "num_input_tokens_seen": 119808,
      "step": 570
    },
    {
      "epoch": 0.06325632563256325,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000947194719471947,
      "loss": 0.2191,
      "num_input_tokens_seen": 120864,
      "step": 575
    },
    {
      "epoch": 0.0638063806380638,
      "grad_norm": 0.578125,
      "learning_rate": 0.0009554455445544553,
      "loss": 0.2396,
      "num_input_tokens_seen": 121952,
      "step": 580
    },
    {
      "epoch": 0.06435643564356436,
      "grad_norm": 0.35546875,
      "learning_rate": 0.0009636963696369636,
      "loss": 0.2343,
      "num_input_tokens_seen": 122976,
      "step": 585
    },
    {
      "epoch": 0.06490649064906491,
      "grad_norm": 15.9375,
      "learning_rate": 0.0009719471947194719,
      "loss": 0.4151,
      "num_input_tokens_seen": 123968,
      "step": 590
    },
    {
      "epoch": 0.06545654565456546,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0009801980198019802,
      "loss": 0.2638,
      "num_input_tokens_seen": 124992,
      "step": 595
    },
    {
      "epoch": 0.066006600660066,
      "grad_norm": 1.046875,
      "learning_rate": 0.0009884488448844884,
      "loss": 0.2713,
      "num_input_tokens_seen": 126048,
      "step": 600
    },
    {
      "epoch": 0.06655665566556655,
      "grad_norm": 0.640625,
      "learning_rate": 0.0009966996699669968,
      "loss": 0.2378,
      "num_input_tokens_seen": 127168,
      "step": 605
    },
    {
      "epoch": 0.0671067106710671,
      "grad_norm": 0.20703125,
      "learning_rate": 0.0010049504950495048,
      "loss": 0.2378,
      "num_input_tokens_seen": 128224,
      "step": 610
    },
    {
      "epoch": 0.06765676567656766,
      "grad_norm": 0.1162109375,
      "learning_rate": 0.0010132013201320132,
      "loss": 0.2363,
      "num_input_tokens_seen": 129280,
      "step": 615
    },
    {
      "epoch": 0.06820682068206821,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0010214521452145213,
      "loss": 0.2352,
      "num_input_tokens_seen": 130336,
      "step": 620
    },
    {
      "epoch": 0.06875687568756876,
      "grad_norm": 0.2001953125,
      "learning_rate": 0.0010297029702970298,
      "loss": 0.2247,
      "num_input_tokens_seen": 131424,
      "step": 625
    },
    {
      "epoch": 0.06930693069306931,
      "grad_norm": 0.27734375,
      "learning_rate": 0.001037953795379538,
      "loss": 0.2392,
      "num_input_tokens_seen": 132480,
      "step": 630
    },
    {
      "epoch": 0.06985698569856985,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0010462046204620463,
      "loss": 0.2445,
      "num_input_tokens_seen": 133536,
      "step": 635
    },
    {
      "epoch": 0.0704070407040704,
      "grad_norm": 0.34375,
      "learning_rate": 0.0010544554455445545,
      "loss": 0.2273,
      "num_input_tokens_seen": 134592,
      "step": 640
    },
    {
      "epoch": 0.07095709570957096,
      "grad_norm": 0.181640625,
      "learning_rate": 0.0010627062706270625,
      "loss": 0.2267,
      "num_input_tokens_seen": 135648,
      "step": 645
    },
    {
      "epoch": 0.07150715071507151,
      "grad_norm": 0.208984375,
      "learning_rate": 0.001070957095709571,
      "loss": 0.2284,
      "num_input_tokens_seen": 136768,
      "step": 650
    },
    {
      "epoch": 0.07205720572057206,
      "grad_norm": 0.62109375,
      "learning_rate": 0.001079207920792079,
      "loss": 0.2317,
      "num_input_tokens_seen": 137888,
      "step": 655
    },
    {
      "epoch": 0.07260726072607261,
      "grad_norm": 0.1455078125,
      "learning_rate": 0.0010874587458745875,
      "loss": 0.2295,
      "num_input_tokens_seen": 138944,
      "step": 660
    },
    {
      "epoch": 0.07315731573157316,
      "grad_norm": 0.146484375,
      "learning_rate": 0.0010957095709570957,
      "loss": 0.2117,
      "num_input_tokens_seen": 140000,
      "step": 665
    },
    {
      "epoch": 0.0737073707370737,
      "grad_norm": 0.345703125,
      "learning_rate": 0.001103960396039604,
      "loss": 0.2558,
      "num_input_tokens_seen": 141024,
      "step": 670
    },
    {
      "epoch": 0.07425742574257425,
      "grad_norm": 0.1845703125,
      "learning_rate": 0.001112211221122112,
      "loss": 0.2302,
      "num_input_tokens_seen": 142016,
      "step": 675
    },
    {
      "epoch": 0.0748074807480748,
      "grad_norm": 0.0751953125,
      "learning_rate": 0.0011204620462046204,
      "loss": 0.2312,
      "num_input_tokens_seen": 143040,
      "step": 680
    },
    {
      "epoch": 0.07535753575357536,
      "grad_norm": 0.2412109375,
      "learning_rate": 0.0011287128712871286,
      "loss": 0.2328,
      "num_input_tokens_seen": 144096,
      "step": 685
    },
    {
      "epoch": 0.07590759075907591,
      "grad_norm": 0.216796875,
      "learning_rate": 0.0011369636963696368,
      "loss": 0.2191,
      "num_input_tokens_seen": 145120,
      "step": 690
    },
    {
      "epoch": 0.07645764576457646,
      "grad_norm": 0.27734375,
      "learning_rate": 0.0011452145214521452,
      "loss": 0.2502,
      "num_input_tokens_seen": 146144,
      "step": 695
    },
    {
      "epoch": 0.07700770077007701,
      "grad_norm": 0.26171875,
      "learning_rate": 0.0011534653465346534,
      "loss": 0.2317,
      "num_input_tokens_seen": 147264,
      "step": 700
    },
    {
      "epoch": 0.07755775577557755,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0011617161716171618,
      "loss": 0.2377,
      "num_input_tokens_seen": 148384,
      "step": 705
    },
    {
      "epoch": 0.0781078107810781,
      "grad_norm": 0.3046875,
      "learning_rate": 0.0011699669966996698,
      "loss": 0.2394,
      "num_input_tokens_seen": 149376,
      "step": 710
    },
    {
      "epoch": 0.07865786578657866,
      "grad_norm": 0.359375,
      "learning_rate": 0.0011782178217821782,
      "loss": 0.2371,
      "num_input_tokens_seen": 150560,
      "step": 715
    },
    {
      "epoch": 0.07920792079207921,
      "grad_norm": 0.376953125,
      "learning_rate": 0.0011864686468646864,
      "loss": 0.2367,
      "num_input_tokens_seen": 151648,
      "step": 720
    },
    {
      "epoch": 0.07975797579757976,
      "grad_norm": 0.201171875,
      "learning_rate": 0.0011947194719471948,
      "loss": 0.2391,
      "num_input_tokens_seen": 152704,
      "step": 725
    },
    {
      "epoch": 0.08030803080308031,
      "grad_norm": 0.1494140625,
      "learning_rate": 0.001202970297029703,
      "loss": 0.2267,
      "num_input_tokens_seen": 153760,
      "step": 730
    },
    {
      "epoch": 0.08085808580858085,
      "grad_norm": 0.1064453125,
      "learning_rate": 0.0012112211221122111,
      "loss": 0.2343,
      "num_input_tokens_seen": 154752,
      "step": 735
    },
    {
      "epoch": 0.0814081408140814,
      "grad_norm": 0.251953125,
      "learning_rate": 0.0012194719471947195,
      "loss": 0.224,
      "num_input_tokens_seen": 155840,
      "step": 740
    },
    {
      "epoch": 0.08195819581958196,
      "grad_norm": 0.2373046875,
      "learning_rate": 0.0012277227722772275,
      "loss": 0.2178,
      "num_input_tokens_seen": 156896,
      "step": 745
    },
    {
      "epoch": 0.08250825082508251,
      "grad_norm": 0.11181640625,
      "learning_rate": 0.001235973597359736,
      "loss": 0.2469,
      "num_input_tokens_seen": 157984,
      "step": 750
    },
    {
      "epoch": 0.08305830583058306,
      "grad_norm": 0.470703125,
      "learning_rate": 0.001244224422442244,
      "loss": 0.2463,
      "num_input_tokens_seen": 159072,
      "step": 755
    },
    {
      "epoch": 0.08360836083608361,
      "grad_norm": 0.02587890625,
      "learning_rate": 0.0012524752475247525,
      "loss": 0.2385,
      "num_input_tokens_seen": 160160,
      "step": 760
    },
    {
      "epoch": 0.08415841584158416,
      "grad_norm": 0.06298828125,
      "learning_rate": 0.0012607260726072607,
      "loss": 0.2312,
      "num_input_tokens_seen": 161184,
      "step": 765
    },
    {
      "epoch": 0.0847084708470847,
      "grad_norm": 0.236328125,
      "learning_rate": 0.001268976897689769,
      "loss": 0.2381,
      "num_input_tokens_seen": 162208,
      "step": 770
    },
    {
      "epoch": 0.08525852585258525,
      "grad_norm": 0.044921875,
      "learning_rate": 0.0012772277227722773,
      "loss": 0.235,
      "num_input_tokens_seen": 163296,
      "step": 775
    },
    {
      "epoch": 0.0858085808580858,
      "grad_norm": 0.0419921875,
      "learning_rate": 0.0012854785478547852,
      "loss": 0.2646,
      "num_input_tokens_seen": 164416,
      "step": 780
    },
    {
      "epoch": 0.08635863586358636,
      "grad_norm": 0.259765625,
      "learning_rate": 0.0012937293729372936,
      "loss": 0.2319,
      "num_input_tokens_seen": 165440,
      "step": 785
    },
    {
      "epoch": 0.08690869086908691,
      "grad_norm": 0.0284423828125,
      "learning_rate": 0.0013019801980198018,
      "loss": 0.235,
      "num_input_tokens_seen": 166528,
      "step": 790
    },
    {
      "epoch": 0.08745874587458746,
      "grad_norm": 0.1328125,
      "learning_rate": 0.0013102310231023102,
      "loss": 0.2317,
      "num_input_tokens_seen": 167616,
      "step": 795
    },
    {
      "epoch": 0.08800880088008801,
      "grad_norm": 0.037109375,
      "learning_rate": 0.0013184818481848184,
      "loss": 0.2285,
      "num_input_tokens_seen": 168640,
      "step": 800
    },
    {
      "epoch": 0.08855885588558855,
      "grad_norm": 0.0986328125,
      "learning_rate": 0.0013267326732673268,
      "loss": 0.2312,
      "num_input_tokens_seen": 169696,
      "step": 805
    },
    {
      "epoch": 0.0891089108910891,
      "grad_norm": 0.0322265625,
      "learning_rate": 0.001334983498349835,
      "loss": 0.2441,
      "num_input_tokens_seen": 170720,
      "step": 810
    },
    {
      "epoch": 0.08965896589658966,
      "grad_norm": 0.10693359375,
      "learning_rate": 0.0013432343234323432,
      "loss": 0.2337,
      "num_input_tokens_seen": 171680,
      "step": 815
    },
    {
      "epoch": 0.09020902090209021,
      "grad_norm": 0.024658203125,
      "learning_rate": 0.0013514851485148514,
      "loss": 0.2358,
      "num_input_tokens_seen": 172800,
      "step": 820
    },
    {
      "epoch": 0.09075907590759076,
      "grad_norm": 0.0966796875,
      "learning_rate": 0.0013597359735973596,
      "loss": 0.2217,
      "num_input_tokens_seen": 173856,
      "step": 825
    },
    {
      "epoch": 0.09130913091309131,
      "grad_norm": 0.0830078125,
      "learning_rate": 0.001367986798679868,
      "loss": 0.2215,
      "num_input_tokens_seen": 174848,
      "step": 830
    },
    {
      "epoch": 0.09185918591859187,
      "grad_norm": 0.1357421875,
      "learning_rate": 0.0013762376237623762,
      "loss": 0.2327,
      "num_input_tokens_seen": 175936,
      "step": 835
    },
    {
      "epoch": 0.0924092409240924,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0013844884488448846,
      "loss": 0.3063,
      "num_input_tokens_seen": 176992,
      "step": 840
    },
    {
      "epoch": 0.09295929592959296,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0013927392739273927,
      "loss": 0.2241,
      "num_input_tokens_seen": 178048,
      "step": 845
    },
    {
      "epoch": 0.09350935093509351,
      "grad_norm": 0.14453125,
      "learning_rate": 0.001400990099009901,
      "loss": 0.2471,
      "num_input_tokens_seen": 179072,
      "step": 850
    },
    {
      "epoch": 0.09405940594059406,
      "grad_norm": 0.0751953125,
      "learning_rate": 0.0014092409240924091,
      "loss": 0.2306,
      "num_input_tokens_seen": 180096,
      "step": 855
    },
    {
      "epoch": 0.09460946094609461,
      "grad_norm": 0.232421875,
      "learning_rate": 0.0014174917491749175,
      "loss": 0.233,
      "num_input_tokens_seen": 181152,
      "step": 860
    },
    {
      "epoch": 0.09515951595159516,
      "grad_norm": 0.08203125,
      "learning_rate": 0.0014257425742574257,
      "loss": 0.2392,
      "num_input_tokens_seen": 182208,
      "step": 865
    },
    {
      "epoch": 0.09570957095709572,
      "grad_norm": 0.103515625,
      "learning_rate": 0.001433993399339934,
      "loss": 0.2283,
      "num_input_tokens_seen": 183232,
      "step": 870
    },
    {
      "epoch": 0.09625962596259625,
      "grad_norm": 0.1435546875,
      "learning_rate": 0.0014422442244224423,
      "loss": 0.2252,
      "num_input_tokens_seen": 184256,
      "step": 875
    },
    {
      "epoch": 0.0968096809680968,
      "grad_norm": 0.2275390625,
      "learning_rate": 0.0014504950495049503,
      "loss": 0.2287,
      "num_input_tokens_seen": 185280,
      "step": 880
    },
    {
      "epoch": 0.09735973597359736,
      "grad_norm": 0.333984375,
      "learning_rate": 0.0014587458745874587,
      "loss": 0.2424,
      "num_input_tokens_seen": 186336,
      "step": 885
    },
    {
      "epoch": 0.09790979097909791,
      "grad_norm": 0.2431640625,
      "learning_rate": 0.0014669966996699669,
      "loss": 0.2292,
      "num_input_tokens_seen": 187392,
      "step": 890
    },
    {
      "epoch": 0.09845984598459846,
      "grad_norm": 1.125,
      "learning_rate": 0.0014752475247524753,
      "loss": 0.1988,
      "num_input_tokens_seen": 188416,
      "step": 895
    },
    {
      "epoch": 0.09900990099009901,
      "grad_norm": 0.58984375,
      "learning_rate": 0.0014834983498349834,
      "loss": 0.2353,
      "num_input_tokens_seen": 189472,
      "step": 900
    },
    {
      "epoch": 0.09955995599559957,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0014917491749174918,
      "loss": 0.2548,
      "num_input_tokens_seen": 190528,
      "step": 905
    },
    {
      "epoch": 0.1001100110011001,
      "grad_norm": 56.25,
      "learning_rate": 0.0015,
      "loss": 1.5347,
      "num_input_tokens_seen": 191648,
      "step": 910
    },
    {
      "epoch": 0.10066006600660066,
      "grad_norm": 2.78125,
      "learning_rate": 0.001508250825082508,
      "loss": 0.3724,
      "num_input_tokens_seen": 192672,
      "step": 915
    },
    {
      "epoch": 0.10121012101210121,
      "grad_norm": 2.765625,
      "learning_rate": 0.0015165016501650164,
      "loss": 0.3302,
      "num_input_tokens_seen": 193760,
      "step": 920
    },
    {
      "epoch": 0.10176017601760176,
      "grad_norm": 0.201171875,
      "learning_rate": 0.0015247524752475246,
      "loss": 0.2349,
      "num_input_tokens_seen": 194848,
      "step": 925
    },
    {
      "epoch": 0.10231023102310231,
      "grad_norm": 0.25,
      "learning_rate": 0.001533003300330033,
      "loss": 0.2228,
      "num_input_tokens_seen": 195968,
      "step": 930
    },
    {
      "epoch": 0.10286028602860287,
      "grad_norm": 0.205078125,
      "learning_rate": 0.0015412541254125412,
      "loss": 0.2455,
      "num_input_tokens_seen": 196992,
      "step": 935
    },
    {
      "epoch": 0.1034103410341034,
      "grad_norm": 0.69921875,
      "learning_rate": 0.0015495049504950496,
      "loss": 0.2335,
      "num_input_tokens_seen": 198112,
      "step": 940
    },
    {
      "epoch": 0.10396039603960396,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0015577557755775578,
      "loss": 0.2767,
      "num_input_tokens_seen": 199136,
      "step": 945
    },
    {
      "epoch": 0.10451045104510451,
      "grad_norm": 0.1162109375,
      "learning_rate": 0.001566006600660066,
      "loss": 0.2537,
      "num_input_tokens_seen": 200224,
      "step": 950
    },
    {
      "epoch": 0.10506050605060506,
      "grad_norm": 0.140625,
      "learning_rate": 0.0015742574257425741,
      "loss": 0.2337,
      "num_input_tokens_seen": 201312,
      "step": 955
    },
    {
      "epoch": 0.10561056105610561,
      "grad_norm": 0.2099609375,
      "learning_rate": 0.0015825082508250823,
      "loss": 0.245,
      "num_input_tokens_seen": 202400,
      "step": 960
    },
    {
      "epoch": 0.10616061606160616,
      "grad_norm": 0.0478515625,
      "learning_rate": 0.0015907590759075907,
      "loss": 0.234,
      "num_input_tokens_seen": 203360,
      "step": 965
    },
    {
      "epoch": 0.10671067106710672,
      "grad_norm": 0.031494140625,
      "learning_rate": 0.001599009900990099,
      "loss": 0.2374,
      "num_input_tokens_seen": 204352,
      "step": 970
    },
    {
      "epoch": 0.10726072607260725,
      "grad_norm": 0.2060546875,
      "learning_rate": 0.0016072607260726073,
      "loss": 0.2329,
      "num_input_tokens_seen": 205376,
      "step": 975
    },
    {
      "epoch": 0.1078107810781078,
      "grad_norm": 0.1123046875,
      "learning_rate": 0.0016155115511551155,
      "loss": 0.2326,
      "num_input_tokens_seen": 206400,
      "step": 980
    },
    {
      "epoch": 0.10836083608360836,
      "grad_norm": 0.126953125,
      "learning_rate": 0.0016237623762376237,
      "loss": 0.2303,
      "num_input_tokens_seen": 207456,
      "step": 985
    },
    {
      "epoch": 0.10891089108910891,
      "grad_norm": 0.24609375,
      "learning_rate": 0.0016320132013201319,
      "loss": 0.2335,
      "num_input_tokens_seen": 208480,
      "step": 990
    },
    {
      "epoch": 0.10946094609460946,
      "grad_norm": 0.033447265625,
      "learning_rate": 0.0016402640264026403,
      "loss": 0.2235,
      "num_input_tokens_seen": 209536,
      "step": 995
    },
    {
      "epoch": 0.11001100110011001,
      "grad_norm": 0.10595703125,
      "learning_rate": 0.0016485148514851485,
      "loss": 0.242,
      "num_input_tokens_seen": 210560,
      "step": 1000
    },
    {
      "epoch": 0.11056105610561057,
      "grad_norm": 368.0,
      "learning_rate": 0.0016567656765676566,
      "loss": 0.4892,
      "num_input_tokens_seen": 211584,
      "step": 1005
    },
    {
      "epoch": 0.1111111111111111,
      "grad_norm": 0.1923828125,
      "learning_rate": 0.001665016501650165,
      "loss": 0.2328,
      "num_input_tokens_seen": 212672,
      "step": 1010
    },
    {
      "epoch": 0.11166116611661166,
      "grad_norm": 0.060791015625,
      "learning_rate": 0.0016732673267326732,
      "loss": 0.2371,
      "num_input_tokens_seen": 213760,
      "step": 1015
    },
    {
      "epoch": 0.11221122112211221,
      "grad_norm": 0.09814453125,
      "learning_rate": 0.0016815181518151814,
      "loss": 0.2359,
      "num_input_tokens_seen": 214816,
      "step": 1020
    },
    {
      "epoch": 0.11276127612761276,
      "grad_norm": 0.022705078125,
      "learning_rate": 0.0016897689768976896,
      "loss": 0.2296,
      "num_input_tokens_seen": 215904,
      "step": 1025
    },
    {
      "epoch": 0.11331133113311331,
      "grad_norm": 0.033203125,
      "learning_rate": 0.001698019801980198,
      "loss": 0.2328,
      "num_input_tokens_seen": 216992,
      "step": 1030
    },
    {
      "epoch": 0.11386138613861387,
      "grad_norm": 0.0179443359375,
      "learning_rate": 0.0017062706270627062,
      "loss": 0.2346,
      "num_input_tokens_seen": 218048,
      "step": 1035
    },
    {
      "epoch": 0.11441144114411442,
      "grad_norm": 0.02490234375,
      "learning_rate": 0.0017145214521452146,
      "loss": 0.2307,
      "num_input_tokens_seen": 219136,
      "step": 1040
    },
    {
      "epoch": 0.11496149614961496,
      "grad_norm": 0.028564453125,
      "learning_rate": 0.0017227722772277228,
      "loss": 0.228,
      "num_input_tokens_seen": 220128,
      "step": 1045
    },
    {
      "epoch": 0.11551155115511551,
      "grad_norm": 0.11669921875,
      "learning_rate": 0.001731023102310231,
      "loss": 0.2284,
      "num_input_tokens_seen": 221184,
      "step": 1050
    },
    {
      "epoch": 0.11606160616061606,
      "grad_norm": 0.0218505859375,
      "learning_rate": 0.0017392739273927392,
      "loss": 0.248,
      "num_input_tokens_seen": 222272,
      "step": 1055
    },
    {
      "epoch": 0.11661166116611661,
      "grad_norm": 0.08642578125,
      "learning_rate": 0.0017475247524752473,
      "loss": 0.24,
      "num_input_tokens_seen": 223296,
      "step": 1060
    },
    {
      "epoch": 0.11716171617161716,
      "grad_norm": 0.1396484375,
      "learning_rate": 0.0017557755775577557,
      "loss": 0.2326,
      "num_input_tokens_seen": 224352,
      "step": 1065
    },
    {
      "epoch": 0.11771177117711772,
      "grad_norm": 0.06787109375,
      "learning_rate": 0.001764026402640264,
      "loss": 0.2303,
      "num_input_tokens_seen": 225344,
      "step": 1070
    },
    {
      "epoch": 0.11826182618261827,
      "grad_norm": 0.021484375,
      "learning_rate": 0.0017722772277227723,
      "loss": 0.2347,
      "num_input_tokens_seen": 226400,
      "step": 1075
    },
    {
      "epoch": 0.1188118811881188,
      "grad_norm": 0.0859375,
      "learning_rate": 0.0017805280528052805,
      "loss": 0.2307,
      "num_input_tokens_seen": 227520,
      "step": 1080
    },
    {
      "epoch": 0.11936193619361936,
      "grad_norm": 0.0257568359375,
      "learning_rate": 0.0017887788778877887,
      "loss": 0.2253,
      "num_input_tokens_seen": 228608,
      "step": 1085
    },
    {
      "epoch": 0.11991199119911991,
      "grad_norm": 0.0260009765625,
      "learning_rate": 0.0017970297029702969,
      "loss": 0.2353,
      "num_input_tokens_seen": 229664,
      "step": 1090
    },
    {
      "epoch": 0.12046204620462046,
      "grad_norm": 0.0234375,
      "learning_rate": 0.001805280528052805,
      "loss": 0.2342,
      "num_input_tokens_seen": 230688,
      "step": 1095
    },
    {
      "epoch": 0.12101210121012101,
      "grad_norm": 0.0247802734375,
      "learning_rate": 0.0018135313531353135,
      "loss": 0.2268,
      "num_input_tokens_seen": 231776,
      "step": 1100
    },
    {
      "epoch": 0.12156215621562157,
      "grad_norm": 0.1552734375,
      "learning_rate": 0.0018217821782178217,
      "loss": 0.237,
      "num_input_tokens_seen": 232864,
      "step": 1105
    },
    {
      "epoch": 0.12211221122112212,
      "grad_norm": 0.0196533203125,
      "learning_rate": 0.00183003300330033,
      "loss": 0.2352,
      "num_input_tokens_seen": 233920,
      "step": 1110
    },
    {
      "epoch": 0.12266226622662266,
      "grad_norm": 0.0225830078125,
      "learning_rate": 0.0018382838283828383,
      "loss": 0.2315,
      "num_input_tokens_seen": 235040,
      "step": 1115
    },
    {
      "epoch": 0.12321232123212321,
      "grad_norm": 0.06884765625,
      "learning_rate": 0.0018465346534653464,
      "loss": 0.2337,
      "num_input_tokens_seen": 236128,
      "step": 1120
    },
    {
      "epoch": 0.12376237623762376,
      "grad_norm": 0.06494140625,
      "learning_rate": 0.0018547854785478546,
      "loss": 0.2379,
      "num_input_tokens_seen": 237120,
      "step": 1125
    },
    {
      "epoch": 0.12431243124312431,
      "grad_norm": 0.062255859375,
      "learning_rate": 0.001863036303630363,
      "loss": 0.2283,
      "num_input_tokens_seen": 238112,
      "step": 1130
    },
    {
      "epoch": 0.12486248624862487,
      "grad_norm": 0.01544189453125,
      "learning_rate": 0.0018712871287128712,
      "loss": 0.2366,
      "num_input_tokens_seen": 239168,
      "step": 1135
    },
    {
      "epoch": 0.1254125412541254,
      "grad_norm": 0.025390625,
      "learning_rate": 0.0018795379537953792,
      "loss": 0.2273,
      "num_input_tokens_seen": 240256,
      "step": 1140
    },
    {
      "epoch": 0.12596259625962597,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.0018877887788778878,
      "loss": 0.2348,
      "num_input_tokens_seen": 241280,
      "step": 1145
    },
    {
      "epoch": 0.1265126512651265,
      "grad_norm": 0.05859375,
      "learning_rate": 0.0018960396039603958,
      "loss": 0.2297,
      "num_input_tokens_seen": 242336,
      "step": 1150
    },
    {
      "epoch": 0.12706270627062707,
      "grad_norm": 0.0281982421875,
      "learning_rate": 0.0019042904290429044,
      "loss": 0.2317,
      "num_input_tokens_seen": 243456,
      "step": 1155
    },
    {
      "epoch": 0.1276127612761276,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.0019125412541254124,
      "loss": 0.2309,
      "num_input_tokens_seen": 244576,
      "step": 1160
    },
    {
      "epoch": 0.12816281628162815,
      "grad_norm": 0.055908203125,
      "learning_rate": 0.001920792079207921,
      "loss": 0.2319,
      "num_input_tokens_seen": 245664,
      "step": 1165
    },
    {
      "epoch": 0.12871287128712872,
      "grad_norm": 0.023681640625,
      "learning_rate": 0.001929042904290429,
      "loss": 0.2319,
      "num_input_tokens_seen": 246752,
      "step": 1170
    },
    {
      "epoch": 0.12926292629262925,
      "grad_norm": 0.0250244140625,
      "learning_rate": 0.0019372937293729371,
      "loss": 0.2328,
      "num_input_tokens_seen": 247744,
      "step": 1175
    },
    {
      "epoch": 0.12981298129812982,
      "grad_norm": 0.06005859375,
      "learning_rate": 0.0019455445544554455,
      "loss": 0.2286,
      "num_input_tokens_seen": 248768,
      "step": 1180
    },
    {
      "epoch": 0.13036303630363036,
      "grad_norm": 0.123046875,
      "learning_rate": 0.0019537953795379535,
      "loss": 0.2357,
      "num_input_tokens_seen": 249824,
      "step": 1185
    },
    {
      "epoch": 0.13091309130913092,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.001962046204620462,
      "loss": 0.2304,
      "num_input_tokens_seen": 250816,
      "step": 1190
    },
    {
      "epoch": 0.13146314631463146,
      "grad_norm": 0.02001953125,
      "learning_rate": 0.0019702970297029703,
      "loss": 0.2304,
      "num_input_tokens_seen": 251872,
      "step": 1195
    },
    {
      "epoch": 0.132013201320132,
      "grad_norm": 0.059814453125,
      "learning_rate": 0.0019785478547854787,
      "loss": 0.2315,
      "num_input_tokens_seen": 252960,
      "step": 1200
    },
    {
      "epoch": 0.13256325632563257,
      "grad_norm": 0.0634765625,
      "learning_rate": 0.0019867986798679867,
      "loss": 0.2346,
      "num_input_tokens_seen": 254048,
      "step": 1205
    },
    {
      "epoch": 0.1331133113311331,
      "grad_norm": 0.06103515625,
      "learning_rate": 0.0019950495049504947,
      "loss": 0.2307,
      "num_input_tokens_seen": 255168,
      "step": 1210
    },
    {
      "epoch": 0.13366336633663367,
      "grad_norm": 0.0164794921875,
      "learning_rate": 0.0020033003300330035,
      "loss": 0.2329,
      "num_input_tokens_seen": 256224,
      "step": 1215
    },
    {
      "epoch": 0.1342134213421342,
      "grad_norm": 0.0252685546875,
      "learning_rate": 0.0020115511551155115,
      "loss": 0.2297,
      "num_input_tokens_seen": 257248,
      "step": 1220
    },
    {
      "epoch": 0.13476347634763478,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.00201980198019802,
      "loss": 0.2274,
      "num_input_tokens_seen": 258336,
      "step": 1225
    },
    {
      "epoch": 0.1353135313531353,
      "grad_norm": 0.0157470703125,
      "learning_rate": 0.002028052805280528,
      "loss": 0.2323,
      "num_input_tokens_seen": 259360,
      "step": 1230
    },
    {
      "epoch": 0.13586358635863585,
      "grad_norm": 0.064453125,
      "learning_rate": 0.0020363036303630362,
      "loss": 0.2346,
      "num_input_tokens_seen": 260448,
      "step": 1235
    },
    {
      "epoch": 0.13641364136413642,
      "grad_norm": 0.05810546875,
      "learning_rate": 0.0020445544554455446,
      "loss": 0.2296,
      "num_input_tokens_seen": 261472,
      "step": 1240
    },
    {
      "epoch": 0.13696369636963696,
      "grad_norm": 0.07470703125,
      "learning_rate": 0.0020528052805280526,
      "loss": 0.2298,
      "num_input_tokens_seen": 262528,
      "step": 1245
    },
    {
      "epoch": 0.13751375137513752,
      "grad_norm": 0.056640625,
      "learning_rate": 0.002061056105610561,
      "loss": 0.2301,
      "num_input_tokens_seen": 263616,
      "step": 1250
    },
    {
      "epoch": 0.13806380638063806,
      "grad_norm": 0.024658203125,
      "learning_rate": 0.002069306930693069,
      "loss": 0.2322,
      "num_input_tokens_seen": 264640,
      "step": 1255
    },
    {
      "epoch": 0.13861386138613863,
      "grad_norm": 0.08740234375,
      "learning_rate": 0.002077557755775578,
      "loss": 0.2346,
      "num_input_tokens_seen": 265696,
      "step": 1260
    },
    {
      "epoch": 0.13916391639163916,
      "grad_norm": 0.057861328125,
      "learning_rate": 0.0020858085808580858,
      "loss": 0.2329,
      "num_input_tokens_seen": 266720,
      "step": 1265
    },
    {
      "epoch": 0.1397139713971397,
      "grad_norm": 0.1171875,
      "learning_rate": 0.002094059405940594,
      "loss": 0.2246,
      "num_input_tokens_seen": 267776,
      "step": 1270
    },
    {
      "epoch": 0.14026402640264027,
      "grad_norm": 0.0810546875,
      "learning_rate": 0.002102310231023102,
      "loss": 0.2287,
      "num_input_tokens_seen": 268832,
      "step": 1275
    },
    {
      "epoch": 0.1408140814081408,
      "grad_norm": 0.072265625,
      "learning_rate": 0.00211056105610561,
      "loss": 0.2517,
      "num_input_tokens_seen": 269952,
      "step": 1280
    },
    {
      "epoch": 0.14136413641364137,
      "grad_norm": 0.0194091796875,
      "learning_rate": 0.002118811881188119,
      "loss": 0.2311,
      "num_input_tokens_seen": 271008,
      "step": 1285
    },
    {
      "epoch": 0.1419141914191419,
      "grad_norm": 0.015869140625,
      "learning_rate": 0.002127062706270627,
      "loss": 0.2273,
      "num_input_tokens_seen": 272096,
      "step": 1290
    },
    {
      "epoch": 0.14246424642464248,
      "grad_norm": 0.1259765625,
      "learning_rate": 0.0021353135313531353,
      "loss": 0.2261,
      "num_input_tokens_seen": 273120,
      "step": 1295
    },
    {
      "epoch": 0.14301430143014301,
      "grad_norm": 0.051513671875,
      "learning_rate": 0.0021435643564356433,
      "loss": 0.2326,
      "num_input_tokens_seen": 274208,
      "step": 1300
    },
    {
      "epoch": 0.14356435643564355,
      "grad_norm": 0.052001953125,
      "learning_rate": 0.002151815181518152,
      "loss": 0.2248,
      "num_input_tokens_seen": 275264,
      "step": 1305
    },
    {
      "epoch": 0.14411441144114412,
      "grad_norm": 0.021484375,
      "learning_rate": 0.00216006600660066,
      "loss": 0.2405,
      "num_input_tokens_seen": 276256,
      "step": 1310
    },
    {
      "epoch": 0.14466446644664466,
      "grad_norm": 0.01953125,
      "learning_rate": 0.002168316831683168,
      "loss": 0.2329,
      "num_input_tokens_seen": 277280,
      "step": 1315
    },
    {
      "epoch": 0.14521452145214522,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.0021765676567656765,
      "loss": 0.2347,
      "num_input_tokens_seen": 278432,
      "step": 1320
    },
    {
      "epoch": 0.14576457645764576,
      "grad_norm": 0.1162109375,
      "learning_rate": 0.0021848184818481844,
      "loss": 0.237,
      "num_input_tokens_seen": 279488,
      "step": 1325
    },
    {
      "epoch": 0.14631463146314633,
      "grad_norm": 0.0537109375,
      "learning_rate": 0.0021930693069306933,
      "loss": 0.2325,
      "num_input_tokens_seen": 280576,
      "step": 1330
    },
    {
      "epoch": 0.14686468646864687,
      "grad_norm": 0.056640625,
      "learning_rate": 0.0022013201320132013,
      "loss": 0.2303,
      "num_input_tokens_seen": 281664,
      "step": 1335
    },
    {
      "epoch": 0.1474147414741474,
      "grad_norm": 0.059814453125,
      "learning_rate": 0.0022095709570957097,
      "loss": 0.2314,
      "num_input_tokens_seen": 282720,
      "step": 1340
    },
    {
      "epoch": 0.14796479647964797,
      "grad_norm": 0.103515625,
      "learning_rate": 0.0022178217821782176,
      "loss": 0.2316,
      "num_input_tokens_seen": 283776,
      "step": 1345
    },
    {
      "epoch": 0.1485148514851485,
      "grad_norm": 0.0556640625,
      "learning_rate": 0.002226072607260726,
      "loss": 0.2295,
      "num_input_tokens_seen": 284864,
      "step": 1350
    },
    {
      "epoch": 0.14906490649064907,
      "grad_norm": 0.015869140625,
      "learning_rate": 0.0022343234323432344,
      "loss": 0.2359,
      "num_input_tokens_seen": 285984,
      "step": 1355
    },
    {
      "epoch": 0.1496149614961496,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0022425742574257424,
      "loss": 0.2305,
      "num_input_tokens_seen": 287040,
      "step": 1360
    },
    {
      "epoch": 0.15016501650165018,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.002250825082508251,
      "loss": 0.2306,
      "num_input_tokens_seen": 288064,
      "step": 1365
    },
    {
      "epoch": 0.15071507150715072,
      "grad_norm": 0.044677734375,
      "learning_rate": 0.0022590759075907588,
      "loss": 0.2216,
      "num_input_tokens_seen": 289088,
      "step": 1370
    },
    {
      "epoch": 0.15126512651265125,
      "grad_norm": 0.043212890625,
      "learning_rate": 0.002267326732673267,
      "loss": 0.2269,
      "num_input_tokens_seen": 290048,
      "step": 1375
    },
    {
      "epoch": 0.15181518151815182,
      "grad_norm": 0.068359375,
      "learning_rate": 0.0022755775577557756,
      "loss": 0.2433,
      "num_input_tokens_seen": 291136,
      "step": 1380
    },
    {
      "epoch": 0.15236523652365236,
      "grad_norm": 0.0201416015625,
      "learning_rate": 0.0022838283828382835,
      "loss": 0.2348,
      "num_input_tokens_seen": 292192,
      "step": 1385
    },
    {
      "epoch": 0.15291529152915292,
      "grad_norm": 0.0654296875,
      "learning_rate": 0.002292079207920792,
      "loss": 0.227,
      "num_input_tokens_seen": 293184,
      "step": 1390
    },
    {
      "epoch": 0.15346534653465346,
      "grad_norm": 0.044677734375,
      "learning_rate": 0.0023003300330033003,
      "loss": 0.2341,
      "num_input_tokens_seen": 294272,
      "step": 1395
    },
    {
      "epoch": 0.15401540154015403,
      "grad_norm": 0.048583984375,
      "learning_rate": 0.0023085808580858088,
      "loss": 0.2331,
      "num_input_tokens_seen": 295328,
      "step": 1400
    },
    {
      "epoch": 0.15456545654565457,
      "grad_norm": 0.0625,
      "learning_rate": 0.0023168316831683167,
      "loss": 0.2352,
      "num_input_tokens_seen": 296416,
      "step": 1405
    },
    {
      "epoch": 0.1551155115511551,
      "grad_norm": 0.0673828125,
      "learning_rate": 0.0023250825082508247,
      "loss": 0.2307,
      "num_input_tokens_seen": 297504,
      "step": 1410
    },
    {
      "epoch": 0.15566556655665567,
      "grad_norm": 0.0703125,
      "learning_rate": 0.002333333333333333,
      "loss": 0.2337,
      "num_input_tokens_seen": 298496,
      "step": 1415
    },
    {
      "epoch": 0.1562156215621562,
      "grad_norm": 0.0177001953125,
      "learning_rate": 0.0023415841584158415,
      "loss": 0.2344,
      "num_input_tokens_seen": 299616,
      "step": 1420
    },
    {
      "epoch": 0.15676567656765678,
      "grad_norm": 0.10791015625,
      "learning_rate": 0.00234983498349835,
      "loss": 0.2314,
      "num_input_tokens_seen": 300672,
      "step": 1425
    },
    {
      "epoch": 0.1573157315731573,
      "grad_norm": 0.09423828125,
      "learning_rate": 0.002358085808580858,
      "loss": 0.2239,
      "num_input_tokens_seen": 301664,
      "step": 1430
    },
    {
      "epoch": 0.15786578657865788,
      "grad_norm": 0.0751953125,
      "learning_rate": 0.0023663366336633663,
      "loss": 0.2438,
      "num_input_tokens_seen": 302656,
      "step": 1435
    },
    {
      "epoch": 0.15841584158415842,
      "grad_norm": 0.09521484375,
      "learning_rate": 0.0023745874587458747,
      "loss": 0.2339,
      "num_input_tokens_seen": 303744,
      "step": 1440
    },
    {
      "epoch": 0.15896589658965896,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.0023828382838283826,
      "loss": 0.2352,
      "num_input_tokens_seen": 304768,
      "step": 1445
    },
    {
      "epoch": 0.15951595159515952,
      "grad_norm": 0.056640625,
      "learning_rate": 0.002391089108910891,
      "loss": 0.2292,
      "num_input_tokens_seen": 305792,
      "step": 1450
    },
    {
      "epoch": 0.16006600660066006,
      "grad_norm": 0.05419921875,
      "learning_rate": 0.002399339933993399,
      "loss": 0.2303,
      "num_input_tokens_seen": 306848,
      "step": 1455
    },
    {
      "epoch": 0.16061606160616063,
      "grad_norm": 0.052001953125,
      "learning_rate": 0.0024075907590759074,
      "loss": 0.2305,
      "num_input_tokens_seen": 307936,
      "step": 1460
    },
    {
      "epoch": 0.16116611661166116,
      "grad_norm": 0.017333984375,
      "learning_rate": 0.002415841584158416,
      "loss": 0.2305,
      "num_input_tokens_seen": 308960,
      "step": 1465
    },
    {
      "epoch": 0.1617161716171617,
      "grad_norm": 0.050537109375,
      "learning_rate": 0.0024240924092409242,
      "loss": 0.2359,
      "num_input_tokens_seen": 310080,
      "step": 1470
    },
    {
      "epoch": 0.16226622662266227,
      "grad_norm": 0.050537109375,
      "learning_rate": 0.002432343234323432,
      "loss": 0.2226,
      "num_input_tokens_seen": 311104,
      "step": 1475
    },
    {
      "epoch": 0.1628162816281628,
      "grad_norm": 0.08056640625,
      "learning_rate": 0.00244059405940594,
      "loss": 0.2407,
      "num_input_tokens_seen": 312192,
      "step": 1480
    },
    {
      "epoch": 0.16336633663366337,
      "grad_norm": 0.052734375,
      "learning_rate": 0.002448844884488449,
      "loss": 0.2315,
      "num_input_tokens_seen": 313248,
      "step": 1485
    },
    {
      "epoch": 0.1639163916391639,
      "grad_norm": 0.10693359375,
      "learning_rate": 0.002457095709570957,
      "loss": 0.2403,
      "num_input_tokens_seen": 314368,
      "step": 1490
    },
    {
      "epoch": 0.16446644664466448,
      "grad_norm": 0.055908203125,
      "learning_rate": 0.0024653465346534654,
      "loss": 0.2325,
      "num_input_tokens_seen": 315488,
      "step": 1495
    },
    {
      "epoch": 0.16501650165016502,
      "grad_norm": 0.0228271484375,
      "learning_rate": 0.0024735973597359733,
      "loss": 0.2314,
      "num_input_tokens_seen": 316608,
      "step": 1500
    },
    {
      "epoch": 0.16556655665566555,
      "grad_norm": 0.058349609375,
      "learning_rate": 0.0024818481848184817,
      "loss": 0.2324,
      "num_input_tokens_seen": 317664,
      "step": 1505
    },
    {
      "epoch": 0.16611661166116612,
      "grad_norm": 0.04052734375,
      "learning_rate": 0.00249009900990099,
      "loss": 0.232,
      "num_input_tokens_seen": 318784,
      "step": 1510
    },
    {
      "epoch": 0.16666666666666666,
      "grad_norm": 0.06591796875,
      "learning_rate": 0.002498349834983498,
      "loss": 0.2285,
      "num_input_tokens_seen": 319808,
      "step": 1515
    },
    {
      "epoch": 0.16721672167216722,
      "grad_norm": 0.0322265625,
      "learning_rate": 0.0025066006600660065,
      "loss": 0.2309,
      "num_input_tokens_seen": 320864,
      "step": 1520
    },
    {
      "epoch": 0.16776677667766776,
      "grad_norm": 0.053466796875,
      "learning_rate": 0.0025148514851485145,
      "loss": 0.2274,
      "num_input_tokens_seen": 321856,
      "step": 1525
    },
    {
      "epoch": 0.16831683168316833,
      "grad_norm": 0.09619140625,
      "learning_rate": 0.0025231023102310233,
      "loss": 0.2235,
      "num_input_tokens_seen": 323008,
      "step": 1530
    },
    {
      "epoch": 0.16886688668866887,
      "grad_norm": 0.0240478515625,
      "learning_rate": 0.0025313531353135313,
      "loss": 0.2398,
      "num_input_tokens_seen": 324096,
      "step": 1535
    },
    {
      "epoch": 0.1694169416941694,
      "grad_norm": 0.047607421875,
      "learning_rate": 0.0025396039603960397,
      "loss": 0.23,
      "num_input_tokens_seen": 325152,
      "step": 1540
    },
    {
      "epoch": 0.16996699669966997,
      "grad_norm": 0.029052734375,
      "learning_rate": 0.0025478547854785477,
      "loss": 0.2227,
      "num_input_tokens_seen": 326272,
      "step": 1545
    },
    {
      "epoch": 0.1705170517051705,
      "grad_norm": 0.0400390625,
      "learning_rate": 0.0025561056105610556,
      "loss": 0.2336,
      "num_input_tokens_seen": 327296,
      "step": 1550
    },
    {
      "epoch": 0.17106710671067107,
      "grad_norm": 0.11328125,
      "learning_rate": 0.0025643564356435645,
      "loss": 0.2298,
      "num_input_tokens_seen": 328288,
      "step": 1555
    },
    {
      "epoch": 0.1716171617161716,
      "grad_norm": 0.0869140625,
      "learning_rate": 0.0025726072607260724,
      "loss": 0.2351,
      "num_input_tokens_seen": 329408,
      "step": 1560
    },
    {
      "epoch": 0.17216721672167218,
      "grad_norm": 0.10400390625,
      "learning_rate": 0.002580858085808581,
      "loss": 0.2319,
      "num_input_tokens_seen": 330432,
      "step": 1565
    },
    {
      "epoch": 0.17271727172717272,
      "grad_norm": 0.134765625,
      "learning_rate": 0.002589108910891089,
      "loss": 0.23,
      "num_input_tokens_seen": 331488,
      "step": 1570
    },
    {
      "epoch": 0.17326732673267325,
      "grad_norm": 0.05810546875,
      "learning_rate": 0.0025973597359735976,
      "loss": 0.2369,
      "num_input_tokens_seen": 332480,
      "step": 1575
    },
    {
      "epoch": 0.17381738173817382,
      "grad_norm": 0.095703125,
      "learning_rate": 0.0026056105610561056,
      "loss": 0.2369,
      "num_input_tokens_seen": 333536,
      "step": 1580
    },
    {
      "epoch": 0.17436743674367436,
      "grad_norm": 0.0213623046875,
      "learning_rate": 0.0026138613861386136,
      "loss": 0.2326,
      "num_input_tokens_seen": 334560,
      "step": 1585
    },
    {
      "epoch": 0.17491749174917492,
      "grad_norm": 0.2373046875,
      "learning_rate": 0.002622112211221122,
      "loss": 0.2344,
      "num_input_tokens_seen": 335616,
      "step": 1590
    },
    {
      "epoch": 0.17546754675467546,
      "grad_norm": 0.10009765625,
      "learning_rate": 0.00263036303630363,
      "loss": 0.2349,
      "num_input_tokens_seen": 336640,
      "step": 1595
    },
    {
      "epoch": 0.17601760176017603,
      "grad_norm": 0.057861328125,
      "learning_rate": 0.002638613861386139,
      "loss": 0.2336,
      "num_input_tokens_seen": 337728,
      "step": 1600
    },
    {
      "epoch": 0.17656765676567657,
      "grad_norm": 0.06787109375,
      "learning_rate": 0.0026468646864686468,
      "loss": 0.2345,
      "num_input_tokens_seen": 338784,
      "step": 1605
    },
    {
      "epoch": 0.1771177117711771,
      "grad_norm": 0.0157470703125,
      "learning_rate": 0.002655115511551155,
      "loss": 0.2324,
      "num_input_tokens_seen": 339776,
      "step": 1610
    },
    {
      "epoch": 0.17766776677667767,
      "grad_norm": 0.061279296875,
      "learning_rate": 0.002663366336633663,
      "loss": 0.2291,
      "num_input_tokens_seen": 340768,
      "step": 1615
    },
    {
      "epoch": 0.1782178217821782,
      "grad_norm": 0.134765625,
      "learning_rate": 0.0026716171617161715,
      "loss": 0.2262,
      "num_input_tokens_seen": 341760,
      "step": 1620
    },
    {
      "epoch": 0.17876787678767878,
      "grad_norm": 0.04638671875,
      "learning_rate": 0.00267986798679868,
      "loss": 0.236,
      "num_input_tokens_seen": 342944,
      "step": 1625
    },
    {
      "epoch": 0.1793179317931793,
      "grad_norm": 0.0771484375,
      "learning_rate": 0.002688118811881188,
      "loss": 0.245,
      "num_input_tokens_seen": 344000,
      "step": 1630
    },
    {
      "epoch": 0.17986798679867988,
      "grad_norm": 0.0517578125,
      "learning_rate": 0.0026963696369636963,
      "loss": 0.2319,
      "num_input_tokens_seen": 345120,
      "step": 1635
    },
    {
      "epoch": 0.18041804180418042,
      "grad_norm": 0.0498046875,
      "learning_rate": 0.0027046204620462043,
      "loss": 0.2264,
      "num_input_tokens_seen": 346240,
      "step": 1640
    },
    {
      "epoch": 0.18096809680968096,
      "grad_norm": 0.087890625,
      "learning_rate": 0.002712871287128713,
      "loss": 0.2338,
      "num_input_tokens_seen": 347328,
      "step": 1645
    },
    {
      "epoch": 0.18151815181518152,
      "grad_norm": 0.06494140625,
      "learning_rate": 0.002721122112211221,
      "loss": 0.2329,
      "num_input_tokens_seen": 348416,
      "step": 1650
    },
    {
      "epoch": 0.18206820682068206,
      "grad_norm": 0.08056640625,
      "learning_rate": 0.002729372937293729,
      "loss": 0.2367,
      "num_input_tokens_seen": 349568,
      "step": 1655
    },
    {
      "epoch": 0.18261826182618263,
      "grad_norm": 0.09033203125,
      "learning_rate": 0.0027376237623762375,
      "loss": 0.2368,
      "num_input_tokens_seen": 350688,
      "step": 1660
    },
    {
      "epoch": 0.18316831683168316,
      "grad_norm": 0.126953125,
      "learning_rate": 0.002745874587458746,
      "loss": 0.2264,
      "num_input_tokens_seen": 351712,
      "step": 1665
    },
    {
      "epoch": 0.18371837183718373,
      "grad_norm": 0.044677734375,
      "learning_rate": 0.0027541254125412543,
      "loss": 0.2365,
      "num_input_tokens_seen": 352800,
      "step": 1670
    },
    {
      "epoch": 0.18426842684268427,
      "grad_norm": 0.056884765625,
      "learning_rate": 0.0027623762376237622,
      "loss": 0.2332,
      "num_input_tokens_seen": 353856,
      "step": 1675
    },
    {
      "epoch": 0.1848184818481848,
      "grad_norm": 0.0625,
      "learning_rate": 0.0027706270627062706,
      "loss": 0.2319,
      "num_input_tokens_seen": 354912,
      "step": 1680
    },
    {
      "epoch": 0.18536853685368537,
      "grad_norm": 0.10546875,
      "learning_rate": 0.0027788778877887786,
      "loss": 0.2338,
      "num_input_tokens_seen": 355968,
      "step": 1685
    },
    {
      "epoch": 0.1859185918591859,
      "grad_norm": 0.10498046875,
      "learning_rate": 0.002787128712871287,
      "loss": 0.2315,
      "num_input_tokens_seen": 357056,
      "step": 1690
    },
    {
      "epoch": 0.18646864686468648,
      "grad_norm": 0.054931640625,
      "learning_rate": 0.0027953795379537954,
      "loss": 0.2336,
      "num_input_tokens_seen": 358144,
      "step": 1695
    },
    {
      "epoch": 0.18701870187018702,
      "grad_norm": 0.072265625,
      "learning_rate": 0.0028036303630363034,
      "loss": 0.2356,
      "num_input_tokens_seen": 359168,
      "step": 1700
    },
    {
      "epoch": 0.18756875687568758,
      "grad_norm": 0.0517578125,
      "learning_rate": 0.0028118811881188118,
      "loss": 0.2303,
      "num_input_tokens_seen": 360224,
      "step": 1705
    },
    {
      "epoch": 0.18811881188118812,
      "grad_norm": 0.12890625,
      "learning_rate": 0.00282013201320132,
      "loss": 0.2356,
      "num_input_tokens_seen": 361248,
      "step": 1710
    },
    {
      "epoch": 0.18866886688668866,
      "grad_norm": 0.0439453125,
      "learning_rate": 0.0028283828382838286,
      "loss": 0.2274,
      "num_input_tokens_seen": 362272,
      "step": 1715
    },
    {
      "epoch": 0.18921892189218922,
      "grad_norm": 0.013671875,
      "learning_rate": 0.0028366336633663366,
      "loss": 0.236,
      "num_input_tokens_seen": 363392,
      "step": 1720
    },
    {
      "epoch": 0.18976897689768976,
      "grad_norm": 0.09765625,
      "learning_rate": 0.0028448844884488445,
      "loss": 0.2425,
      "num_input_tokens_seen": 364512,
      "step": 1725
    },
    {
      "epoch": 0.19031903190319033,
      "grad_norm": 0.013916015625,
      "learning_rate": 0.002853135313531353,
      "loss": 0.2296,
      "num_input_tokens_seen": 365568,
      "step": 1730
    },
    {
      "epoch": 0.19086908690869087,
      "grad_norm": 0.041259765625,
      "learning_rate": 0.0028613861386138613,
      "loss": 0.2305,
      "num_input_tokens_seen": 366720,
      "step": 1735
    },
    {
      "epoch": 0.19141914191419143,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0028696369636963697,
      "loss": 0.2318,
      "num_input_tokens_seen": 367776,
      "step": 1740
    },
    {
      "epoch": 0.19196919691969197,
      "grad_norm": 0.046142578125,
      "learning_rate": 0.0028778877887788777,
      "loss": 0.2277,
      "num_input_tokens_seen": 368832,
      "step": 1745
    },
    {
      "epoch": 0.1925192519251925,
      "grad_norm": 0.10546875,
      "learning_rate": 0.002886138613861386,
      "loss": 0.2322,
      "num_input_tokens_seen": 369888,
      "step": 1750
    },
    {
      "epoch": 0.19306930693069307,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.0028943894389438945,
      "loss": 0.2314,
      "num_input_tokens_seen": 370880,
      "step": 1755
    },
    {
      "epoch": 0.1936193619361936,
      "grad_norm": 0.01495361328125,
      "learning_rate": 0.0029026402640264025,
      "loss": 0.2278,
      "num_input_tokens_seen": 371904,
      "step": 1760
    },
    {
      "epoch": 0.19416941694169418,
      "grad_norm": 0.0830078125,
      "learning_rate": 0.002910891089108911,
      "loss": 0.2181,
      "num_input_tokens_seen": 372960,
      "step": 1765
    },
    {
      "epoch": 0.19471947194719472,
      "grad_norm": 0.041259765625,
      "learning_rate": 0.002919141914191419,
      "loss": 0.2066,
      "num_input_tokens_seen": 374048,
      "step": 1770
    },
    {
      "epoch": 0.19526952695269528,
      "grad_norm": 0.051025390625,
      "learning_rate": 0.0029273927392739272,
      "loss": 0.2283,
      "num_input_tokens_seen": 375104,
      "step": 1775
    },
    {
      "epoch": 0.19581958195819582,
      "grad_norm": 0.038818359375,
      "learning_rate": 0.0029356435643564356,
      "loss": 0.2256,
      "num_input_tokens_seen": 376224,
      "step": 1780
    },
    {
      "epoch": 0.19636963696369636,
      "grad_norm": 0.035400390625,
      "learning_rate": 0.0029438943894389436,
      "loss": 0.2447,
      "num_input_tokens_seen": 377344,
      "step": 1785
    },
    {
      "epoch": 0.19691969196919692,
      "grad_norm": 0.034912109375,
      "learning_rate": 0.002952145214521452,
      "loss": 0.261,
      "num_input_tokens_seen": 378368,
      "step": 1790
    },
    {
      "epoch": 0.19746974697469746,
      "grad_norm": 0.10107421875,
      "learning_rate": 0.00296039603960396,
      "loss": 0.2437,
      "num_input_tokens_seen": 379424,
      "step": 1795
    },
    {
      "epoch": 0.19801980198019803,
      "grad_norm": 0.08642578125,
      "learning_rate": 0.002968646864686469,
      "loss": 0.2248,
      "num_input_tokens_seen": 380448,
      "step": 1800
    },
    {
      "epoch": 0.19856985698569857,
      "grad_norm": 0.028076171875,
      "learning_rate": 0.002976897689768977,
      "loss": 0.2367,
      "num_input_tokens_seen": 381568,
      "step": 1805
    },
    {
      "epoch": 0.19911991199119913,
      "grad_norm": 0.07568359375,
      "learning_rate": 0.002985148514851485,
      "loss": 0.2272,
      "num_input_tokens_seen": 382688,
      "step": 1810
    },
    {
      "epoch": 0.19966996699669967,
      "grad_norm": 0.07763671875,
      "learning_rate": 0.002993399339933993,
      "loss": 0.2349,
      "num_input_tokens_seen": 383808,
      "step": 1815
    },
    {
      "epoch": 0.2002200220022002,
      "grad_norm": 0.0751953125,
      "learning_rate": 0.003001650165016501,
      "loss": 0.2358,
      "num_input_tokens_seen": 384928,
      "step": 1820
    },
    {
      "epoch": 0.20077007700770078,
      "grad_norm": 0.07421875,
      "learning_rate": 0.00300990099009901,
      "loss": 0.2334,
      "num_input_tokens_seen": 385920,
      "step": 1825
    },
    {
      "epoch": 0.20132013201320131,
      "grad_norm": 0.028564453125,
      "learning_rate": 0.003018151815181518,
      "loss": 0.2326,
      "num_input_tokens_seen": 386912,
      "step": 1830
    },
    {
      "epoch": 0.20187018701870188,
      "grad_norm": 0.0150146484375,
      "learning_rate": 0.0030264026402640263,
      "loss": 0.2303,
      "num_input_tokens_seen": 387968,
      "step": 1835
    },
    {
      "epoch": 0.20242024202420242,
      "grad_norm": 0.05224609375,
      "learning_rate": 0.0030346534653465343,
      "loss": 0.2338,
      "num_input_tokens_seen": 389024,
      "step": 1840
    },
    {
      "epoch": 0.20297029702970298,
      "grad_norm": 0.05419921875,
      "learning_rate": 0.003042904290429043,
      "loss": 0.2347,
      "num_input_tokens_seen": 390048,
      "step": 1845
    },
    {
      "epoch": 0.20352035203520352,
      "grad_norm": 0.018798828125,
      "learning_rate": 0.003051155115511551,
      "loss": 0.2303,
      "num_input_tokens_seen": 391008,
      "step": 1850
    },
    {
      "epoch": 0.20407040704070406,
      "grad_norm": 0.0556640625,
      "learning_rate": 0.003059405940594059,
      "loss": 0.2319,
      "num_input_tokens_seen": 392096,
      "step": 1855
    },
    {
      "epoch": 0.20462046204620463,
      "grad_norm": 0.08203125,
      "learning_rate": 0.0030676567656765675,
      "loss": 0.2341,
      "num_input_tokens_seen": 393184,
      "step": 1860
    },
    {
      "epoch": 0.20517051705170516,
      "grad_norm": 0.1142578125,
      "learning_rate": 0.0030759075907590755,
      "loss": 0.2243,
      "num_input_tokens_seen": 394272,
      "step": 1865
    },
    {
      "epoch": 0.20572057205720573,
      "grad_norm": 0.062255859375,
      "learning_rate": 0.0030841584158415843,
      "loss": 0.2259,
      "num_input_tokens_seen": 395328,
      "step": 1870
    },
    {
      "epoch": 0.20627062706270627,
      "grad_norm": 0.06591796875,
      "learning_rate": 0.0030924092409240923,
      "loss": 0.2316,
      "num_input_tokens_seen": 396384,
      "step": 1875
    },
    {
      "epoch": 0.2068206820682068,
      "grad_norm": 0.0791015625,
      "learning_rate": 0.0031006600660066007,
      "loss": 0.2447,
      "num_input_tokens_seen": 397472,
      "step": 1880
    },
    {
      "epoch": 0.20737073707370737,
      "grad_norm": 0.1328125,
      "learning_rate": 0.0031089108910891086,
      "loss": 0.2295,
      "num_input_tokens_seen": 398528,
      "step": 1885
    },
    {
      "epoch": 0.2079207920792079,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.003117161716171617,
      "loss": 0.2337,
      "num_input_tokens_seen": 399552,
      "step": 1890
    },
    {
      "epoch": 0.20847084708470848,
      "grad_norm": 0.046142578125,
      "learning_rate": 0.0031254125412541254,
      "loss": 0.2315,
      "num_input_tokens_seen": 400576,
      "step": 1895
    },
    {
      "epoch": 0.20902090209020902,
      "grad_norm": 0.052734375,
      "learning_rate": 0.0031336633663366334,
      "loss": 0.2366,
      "num_input_tokens_seen": 401600,
      "step": 1900
    },
    {
      "epoch": 0.20957095709570958,
      "grad_norm": 0.0174560546875,
      "learning_rate": 0.003141914191419142,
      "loss": 0.2283,
      "num_input_tokens_seen": 402656,
      "step": 1905
    },
    {
      "epoch": 0.21012101210121012,
      "grad_norm": 0.0556640625,
      "learning_rate": 0.0031501650165016498,
      "loss": 0.2325,
      "num_input_tokens_seen": 403648,
      "step": 1910
    },
    {
      "epoch": 0.21067106710671066,
      "grad_norm": 0.044677734375,
      "learning_rate": 0.0031584158415841586,
      "loss": 0.2314,
      "num_input_tokens_seen": 404672,
      "step": 1915
    },
    {
      "epoch": 0.21122112211221122,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.0031666666666666666,
      "loss": 0.2337,
      "num_input_tokens_seen": 405664,
      "step": 1920
    },
    {
      "epoch": 0.21177117711771176,
      "grad_norm": 0.08154296875,
      "learning_rate": 0.0031749174917491746,
      "loss": 0.2234,
      "num_input_tokens_seen": 406720,
      "step": 1925
    },
    {
      "epoch": 0.21232123212321233,
      "grad_norm": 0.04150390625,
      "learning_rate": 0.003183168316831683,
      "loss": 0.2336,
      "num_input_tokens_seen": 407776,
      "step": 1930
    },
    {
      "epoch": 0.21287128712871287,
      "grad_norm": 0.087890625,
      "learning_rate": 0.0031914191419141914,
      "loss": 0.2244,
      "num_input_tokens_seen": 408832,
      "step": 1935
    },
    {
      "epoch": 0.21342134213421343,
      "grad_norm": 0.0186767578125,
      "learning_rate": 0.0031996699669966998,
      "loss": 0.24,
      "num_input_tokens_seen": 409888,
      "step": 1940
    },
    {
      "epoch": 0.21397139713971397,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.0032079207920792077,
      "loss": 0.2341,
      "num_input_tokens_seen": 410944,
      "step": 1945
    },
    {
      "epoch": 0.2145214521452145,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.003216171617161716,
      "loss": 0.2325,
      "num_input_tokens_seen": 412096,
      "step": 1950
    },
    {
      "epoch": 0.21507150715071507,
      "grad_norm": 0.0308837890625,
      "learning_rate": 0.003224422442244224,
      "loss": 0.2303,
      "num_input_tokens_seen": 413184,
      "step": 1955
    },
    {
      "epoch": 0.2156215621562156,
      "grad_norm": 0.04541015625,
      "learning_rate": 0.0032326732673267325,
      "loss": 0.2262,
      "num_input_tokens_seen": 414176,
      "step": 1960
    },
    {
      "epoch": 0.21617161716171618,
      "grad_norm": 0.01251220703125,
      "learning_rate": 0.003240924092409241,
      "loss": 0.2348,
      "num_input_tokens_seen": 415200,
      "step": 1965
    },
    {
      "epoch": 0.21672167216721672,
      "grad_norm": 0.0732421875,
      "learning_rate": 0.003249174917491749,
      "loss": 0.2319,
      "num_input_tokens_seen": 416256,
      "step": 1970
    },
    {
      "epoch": 0.21727172717271728,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.0032574257425742573,
      "loss": 0.2329,
      "num_input_tokens_seen": 417280,
      "step": 1975
    },
    {
      "epoch": 0.21782178217821782,
      "grad_norm": 0.0157470703125,
      "learning_rate": 0.0032656765676567657,
      "loss": 0.2288,
      "num_input_tokens_seen": 418304,
      "step": 1980
    },
    {
      "epoch": 0.21837183718371836,
      "grad_norm": 0.041259765625,
      "learning_rate": 0.003273927392739274,
      "loss": 0.2405,
      "num_input_tokens_seen": 419456,
      "step": 1985
    },
    {
      "epoch": 0.21892189218921893,
      "grad_norm": 0.0400390625,
      "learning_rate": 0.003282178217821782,
      "loss": 0.2317,
      "num_input_tokens_seen": 420576,
      "step": 1990
    },
    {
      "epoch": 0.21947194719471946,
      "grad_norm": 0.07470703125,
      "learning_rate": 0.00329042904290429,
      "loss": 0.2315,
      "num_input_tokens_seen": 421664,
      "step": 1995
    },
    {
      "epoch": 0.22002200220022003,
      "grad_norm": 0.07470703125,
      "learning_rate": 0.0032986798679867984,
      "loss": 0.2324,
      "num_input_tokens_seen": 422688,
      "step": 2000
    },
    {
      "epoch": 0.22057205720572057,
      "grad_norm": 0.051513671875,
      "learning_rate": 0.003306930693069307,
      "loss": 0.2326,
      "num_input_tokens_seen": 423712,
      "step": 2005
    },
    {
      "epoch": 0.22112211221122113,
      "grad_norm": 0.0478515625,
      "learning_rate": 0.0033151815181518152,
      "loss": 0.2327,
      "num_input_tokens_seen": 424800,
      "step": 2010
    },
    {
      "epoch": 0.22167216721672167,
      "grad_norm": 0.0177001953125,
      "learning_rate": 0.003323432343234323,
      "loss": 0.3239,
      "num_input_tokens_seen": 425920,
      "step": 2015
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 0.040771484375,
      "learning_rate": 0.0033316831683168316,
      "loss": 0.2358,
      "num_input_tokens_seen": 427040,
      "step": 2020
    },
    {
      "epoch": 0.22277227722772278,
      "grad_norm": 0.017333984375,
      "learning_rate": 0.00333993399339934,
      "loss": 0.2294,
      "num_input_tokens_seen": 428128,
      "step": 2025
    },
    {
      "epoch": 0.22332233223322331,
      "grad_norm": 0.03564453125,
      "learning_rate": 0.003348184818481848,
      "loss": 0.2254,
      "num_input_tokens_seen": 429216,
      "step": 2030
    },
    {
      "epoch": 0.22387238723872388,
      "grad_norm": 0.0179443359375,
      "learning_rate": 0.0033564356435643564,
      "loss": 0.2371,
      "num_input_tokens_seen": 430336,
      "step": 2035
    },
    {
      "epoch": 0.22442244224422442,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0033646864686468644,
      "loss": 0.2371,
      "num_input_tokens_seen": 431424,
      "step": 2040
    },
    {
      "epoch": 0.22497249724972498,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.0033729372937293728,
      "loss": 0.226,
      "num_input_tokens_seen": 432448,
      "step": 2045
    },
    {
      "epoch": 0.22552255225522552,
      "grad_norm": 0.033203125,
      "learning_rate": 0.003381188118811881,
      "loss": 0.2289,
      "num_input_tokens_seen": 433504,
      "step": 2050
    },
    {
      "epoch": 0.22607260726072606,
      "grad_norm": 0.0751953125,
      "learning_rate": 0.0033894389438943896,
      "loss": 0.2289,
      "num_input_tokens_seen": 434592,
      "step": 2055
    },
    {
      "epoch": 0.22662266226622663,
      "grad_norm": 0.032470703125,
      "learning_rate": 0.0033976897689768975,
      "loss": 0.222,
      "num_input_tokens_seen": 435712,
      "step": 2060
    },
    {
      "epoch": 0.22717271727172716,
      "grad_norm": 0.033935546875,
      "learning_rate": 0.0034059405940594055,
      "loss": 0.2365,
      "num_input_tokens_seen": 436800,
      "step": 2065
    },
    {
      "epoch": 0.22772277227722773,
      "grad_norm": 0.09423828125,
      "learning_rate": 0.0034141914191419143,
      "loss": 0.2386,
      "num_input_tokens_seen": 437824,
      "step": 2070
    },
    {
      "epoch": 0.22827282728272827,
      "grad_norm": 0.047119140625,
      "learning_rate": 0.0034224422442244223,
      "loss": 0.2339,
      "num_input_tokens_seen": 438880,
      "step": 2075
    },
    {
      "epoch": 0.22882288228822883,
      "grad_norm": 0.01336669921875,
      "learning_rate": 0.0034306930693069307,
      "loss": 0.234,
      "num_input_tokens_seen": 439968,
      "step": 2080
    },
    {
      "epoch": 0.22937293729372937,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.0034389438943894387,
      "loss": 0.2317,
      "num_input_tokens_seen": 440992,
      "step": 2085
    },
    {
      "epoch": 0.2299229922992299,
      "grad_norm": 0.1123046875,
      "learning_rate": 0.003447194719471947,
      "loss": 0.236,
      "num_input_tokens_seen": 442080,
      "step": 2090
    },
    {
      "epoch": 0.23047304730473048,
      "grad_norm": 0.00830078125,
      "learning_rate": 0.0034554455445544555,
      "loss": 0.2389,
      "num_input_tokens_seen": 443136,
      "step": 2095
    },
    {
      "epoch": 0.23102310231023102,
      "grad_norm": 0.052001953125,
      "learning_rate": 0.0034636963696369634,
      "loss": 0.2294,
      "num_input_tokens_seen": 444192,
      "step": 2100
    },
    {
      "epoch": 0.23157315731573158,
      "grad_norm": 0.01495361328125,
      "learning_rate": 0.003471947194719472,
      "loss": 0.2316,
      "num_input_tokens_seen": 445216,
      "step": 2105
    },
    {
      "epoch": 0.23212321232123212,
      "grad_norm": 0.01275634765625,
      "learning_rate": 0.00348019801980198,
      "loss": 0.228,
      "num_input_tokens_seen": 446176,
      "step": 2110
    },
    {
      "epoch": 0.23267326732673269,
      "grad_norm": 0.06494140625,
      "learning_rate": 0.0034884488448844887,
      "loss": 0.2342,
      "num_input_tokens_seen": 447296,
      "step": 2115
    },
    {
      "epoch": 0.23322332233223322,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.0034966996699669966,
      "loss": 0.2228,
      "num_input_tokens_seen": 448320,
      "step": 2120
    },
    {
      "epoch": 0.23377337733773376,
      "grad_norm": 0.08740234375,
      "learning_rate": 0.003504950495049505,
      "loss": 0.2301,
      "num_input_tokens_seen": 449376,
      "step": 2125
    },
    {
      "epoch": 0.23432343234323433,
      "grad_norm": 0.0301513671875,
      "learning_rate": 0.003513201320132013,
      "loss": 0.2207,
      "num_input_tokens_seen": 450432,
      "step": 2130
    },
    {
      "epoch": 0.23487348734873487,
      "grad_norm": 0.11767578125,
      "learning_rate": 0.003521452145214521,
      "loss": 0.2139,
      "num_input_tokens_seen": 451520,
      "step": 2135
    },
    {
      "epoch": 0.23542354235423543,
      "grad_norm": 4.65625,
      "learning_rate": 0.00352970297029703,
      "loss": 0.3443,
      "num_input_tokens_seen": 452608,
      "step": 2140
    },
    {
      "epoch": 0.23597359735973597,
      "grad_norm": 0.2470703125,
      "learning_rate": 0.0035379537953795378,
      "loss": 0.3055,
      "num_input_tokens_seen": 453600,
      "step": 2145
    },
    {
      "epoch": 0.23652365236523654,
      "grad_norm": 0.1689453125,
      "learning_rate": 0.003546204620462046,
      "loss": 0.2422,
      "num_input_tokens_seen": 454656,
      "step": 2150
    },
    {
      "epoch": 0.23707370737073707,
      "grad_norm": 0.11181640625,
      "learning_rate": 0.003554455445544554,
      "loss": 0.2371,
      "num_input_tokens_seen": 455712,
      "step": 2155
    },
    {
      "epoch": 0.2376237623762376,
      "grad_norm": 1.203125,
      "learning_rate": 0.003562706270627063,
      "loss": 0.2472,
      "num_input_tokens_seen": 456768,
      "step": 2160
    },
    {
      "epoch": 0.23817381738173818,
      "grad_norm": 21.375,
      "learning_rate": 0.003570957095709571,
      "loss": 0.3387,
      "num_input_tokens_seen": 457760,
      "step": 2165
    },
    {
      "epoch": 0.23872387238723872,
      "grad_norm": 0.51171875,
      "learning_rate": 0.003579207920792079,
      "loss": 0.3038,
      "num_input_tokens_seen": 458784,
      "step": 2170
    },
    {
      "epoch": 0.23927392739273928,
      "grad_norm": 5.0625,
      "learning_rate": 0.0035874587458745873,
      "loss": 0.293,
      "num_input_tokens_seen": 459840,
      "step": 2175
    },
    {
      "epoch": 0.23982398239823982,
      "grad_norm": 26.625,
      "learning_rate": 0.0035957095709570953,
      "loss": 0.9275,
      "num_input_tokens_seen": 460896,
      "step": 2180
    },
    {
      "epoch": 0.2403740374037404,
      "grad_norm": 7.09375,
      "learning_rate": 0.003603960396039604,
      "loss": 0.5779,
      "num_input_tokens_seen": 461952,
      "step": 2185
    },
    {
      "epoch": 0.24092409240924093,
      "grad_norm": 1.3046875,
      "learning_rate": 0.003612211221122112,
      "loss": 0.4372,
      "num_input_tokens_seen": 463040,
      "step": 2190
    },
    {
      "epoch": 0.24147414741474146,
      "grad_norm": 0.484375,
      "learning_rate": 0.00362046204620462,
      "loss": 0.2632,
      "num_input_tokens_seen": 464064,
      "step": 2195
    },
    {
      "epoch": 0.24202420242024203,
      "grad_norm": 0.2578125,
      "learning_rate": 0.0036287128712871285,
      "loss": 0.2474,
      "num_input_tokens_seen": 465184,
      "step": 2200
    },
    {
      "epoch": 0.24257425742574257,
      "grad_norm": 0.1455078125,
      "learning_rate": 0.003636963696369637,
      "loss": 0.2392,
      "num_input_tokens_seen": 466240,
      "step": 2205
    },
    {
      "epoch": 0.24312431243124313,
      "grad_norm": 0.107421875,
      "learning_rate": 0.0036452145214521453,
      "loss": 0.2285,
      "num_input_tokens_seen": 467264,
      "step": 2210
    },
    {
      "epoch": 0.24367436743674367,
      "grad_norm": 0.1572265625,
      "learning_rate": 0.0036534653465346532,
      "loss": 0.2477,
      "num_input_tokens_seen": 468320,
      "step": 2215
    },
    {
      "epoch": 0.24422442244224424,
      "grad_norm": 0.0771484375,
      "learning_rate": 0.0036617161716171616,
      "loss": 0.2445,
      "num_input_tokens_seen": 469376,
      "step": 2220
    },
    {
      "epoch": 0.24477447744774478,
      "grad_norm": 0.1748046875,
      "learning_rate": 0.0036699669966996696,
      "loss": 0.2386,
      "num_input_tokens_seen": 470464,
      "step": 2225
    },
    {
      "epoch": 0.24532453245324531,
      "grad_norm": 0.1552734375,
      "learning_rate": 0.003678217821782178,
      "loss": 0.3195,
      "num_input_tokens_seen": 471552,
      "step": 2230
    },
    {
      "epoch": 0.24587458745874588,
      "grad_norm": 0.17578125,
      "learning_rate": 0.0036864686468646864,
      "loss": 0.2247,
      "num_input_tokens_seen": 472672,
      "step": 2235
    },
    {
      "epoch": 0.24642464246424642,
      "grad_norm": 0.408203125,
      "learning_rate": 0.0036947194719471944,
      "loss": 0.3447,
      "num_input_tokens_seen": 473664,
      "step": 2240
    },
    {
      "epoch": 0.24697469746974698,
      "grad_norm": 0.1279296875,
      "learning_rate": 0.003702970297029703,
      "loss": 0.2519,
      "num_input_tokens_seen": 474784,
      "step": 2245
    },
    {
      "epoch": 0.24752475247524752,
      "grad_norm": 0.240234375,
      "learning_rate": 0.003711221122112211,
      "loss": 0.2322,
      "num_input_tokens_seen": 475808,
      "step": 2250
    },
    {
      "epoch": 0.2480748074807481,
      "grad_norm": 0.1201171875,
      "learning_rate": 0.0037194719471947196,
      "loss": 0.2477,
      "num_input_tokens_seen": 476928,
      "step": 2255
    },
    {
      "epoch": 0.24862486248624863,
      "grad_norm": 0.1162109375,
      "learning_rate": 0.0037277227722772276,
      "loss": 0.229,
      "num_input_tokens_seen": 477984,
      "step": 2260
    },
    {
      "epoch": 0.24917491749174916,
      "grad_norm": 0.04150390625,
      "learning_rate": 0.0037359735973597355,
      "loss": 0.239,
      "num_input_tokens_seen": 479040,
      "step": 2265
    },
    {
      "epoch": 0.24972497249724973,
      "grad_norm": 0.1142578125,
      "learning_rate": 0.003744224422442244,
      "loss": 0.2251,
      "num_input_tokens_seen": 480160,
      "step": 2270
    },
    {
      "epoch": 0.25027502750275027,
      "grad_norm": 0.11279296875,
      "learning_rate": 0.0037524752475247528,
      "loss": 0.2387,
      "num_input_tokens_seen": 481216,
      "step": 2275
    },
    {
      "epoch": 0.2508250825082508,
      "grad_norm": 0.04931640625,
      "learning_rate": 0.0037607260726072607,
      "loss": 0.2269,
      "num_input_tokens_seen": 482272,
      "step": 2280
    },
    {
      "epoch": 0.2513751375137514,
      "grad_norm": 0.064453125,
      "learning_rate": 0.0037689768976897687,
      "loss": 0.2259,
      "num_input_tokens_seen": 483328,
      "step": 2285
    },
    {
      "epoch": 0.25192519251925194,
      "grad_norm": 0.1337890625,
      "learning_rate": 0.0037772277227722767,
      "loss": 0.2362,
      "num_input_tokens_seen": 484480,
      "step": 2290
    },
    {
      "epoch": 0.2524752475247525,
      "grad_norm": 0.042236328125,
      "learning_rate": 0.003785478547854786,
      "loss": 0.2265,
      "num_input_tokens_seen": 485600,
      "step": 2295
    },
    {
      "epoch": 0.253025302530253,
      "grad_norm": 0.0281982421875,
      "learning_rate": 0.003793729372937294,
      "loss": 0.2392,
      "num_input_tokens_seen": 486624,
      "step": 2300
    },
    {
      "epoch": 0.25357535753575355,
      "grad_norm": 0.08544921875,
      "learning_rate": 0.003801980198019802,
      "loss": 0.2392,
      "num_input_tokens_seen": 487712,
      "step": 2305
    },
    {
      "epoch": 0.25412541254125415,
      "grad_norm": 0.072265625,
      "learning_rate": 0.00381023102310231,
      "loss": 0.2299,
      "num_input_tokens_seen": 488768,
      "step": 2310
    },
    {
      "epoch": 0.2546754675467547,
      "grad_norm": 0.041015625,
      "learning_rate": 0.003818481848184818,
      "loss": 0.2293,
      "num_input_tokens_seen": 489856,
      "step": 2315
    },
    {
      "epoch": 0.2552255225522552,
      "grad_norm": 0.203125,
      "learning_rate": 0.003826732673267327,
      "loss": 0.2404,
      "num_input_tokens_seen": 490848,
      "step": 2320
    },
    {
      "epoch": 0.25577557755775576,
      "grad_norm": 0.1484375,
      "learning_rate": 0.003834983498349835,
      "loss": 0.2346,
      "num_input_tokens_seen": 491904,
      "step": 2325
    },
    {
      "epoch": 0.2563256325632563,
      "grad_norm": 0.035888671875,
      "learning_rate": 0.003843234323432343,
      "loss": 0.2307,
      "num_input_tokens_seen": 493056,
      "step": 2330
    },
    {
      "epoch": 0.2568756875687569,
      "grad_norm": 0.125,
      "learning_rate": 0.003851485148514851,
      "loss": 0.2282,
      "num_input_tokens_seen": 494112,
      "step": 2335
    },
    {
      "epoch": 0.25742574257425743,
      "grad_norm": 0.10546875,
      "learning_rate": 0.0038597359735973594,
      "loss": 0.2328,
      "num_input_tokens_seen": 495168,
      "step": 2340
    },
    {
      "epoch": 0.25797579757975797,
      "grad_norm": 0.0849609375,
      "learning_rate": 0.0038679867986798682,
      "loss": 0.2412,
      "num_input_tokens_seen": 496224,
      "step": 2345
    },
    {
      "epoch": 0.2585258525852585,
      "grad_norm": 0.076171875,
      "learning_rate": 0.003876237623762376,
      "loss": 0.2338,
      "num_input_tokens_seen": 497248,
      "step": 2350
    },
    {
      "epoch": 0.2590759075907591,
      "grad_norm": 0.044677734375,
      "learning_rate": 0.003884488448844884,
      "loss": 0.2305,
      "num_input_tokens_seen": 498240,
      "step": 2355
    },
    {
      "epoch": 0.25962596259625964,
      "grad_norm": 0.08740234375,
      "learning_rate": 0.003892739273927392,
      "loss": 0.2307,
      "num_input_tokens_seen": 499296,
      "step": 2360
    },
    {
      "epoch": 0.2601760176017602,
      "grad_norm": 0.07470703125,
      "learning_rate": 0.0039009900990099014,
      "loss": 0.2268,
      "num_input_tokens_seen": 500320,
      "step": 2365
    },
    {
      "epoch": 0.2607260726072607,
      "grad_norm": 0.0654296875,
      "learning_rate": 0.003909240924092409,
      "loss": 0.2285,
      "num_input_tokens_seen": 501376,
      "step": 2370
    },
    {
      "epoch": 0.26127612761276126,
      "grad_norm": 0.12890625,
      "learning_rate": 0.003917491749174918,
      "loss": 0.2359,
      "num_input_tokens_seen": 502400,
      "step": 2375
    },
    {
      "epoch": 0.26182618261826185,
      "grad_norm": 0.10986328125,
      "learning_rate": 0.003925742574257426,
      "loss": 0.2435,
      "num_input_tokens_seen": 503552,
      "step": 2380
    },
    {
      "epoch": 0.2623762376237624,
      "grad_norm": 0.12060546875,
      "learning_rate": 0.003933993399339934,
      "loss": 0.2262,
      "num_input_tokens_seen": 504640,
      "step": 2385
    },
    {
      "epoch": 0.2629262926292629,
      "grad_norm": 0.11328125,
      "learning_rate": 0.0039422442244224426,
      "loss": 0.2173,
      "num_input_tokens_seen": 505696,
      "step": 2390
    },
    {
      "epoch": 0.26347634763476346,
      "grad_norm": 0.107421875,
      "learning_rate": 0.0039504950495049505,
      "loss": 0.243,
      "num_input_tokens_seen": 506816,
      "step": 2395
    },
    {
      "epoch": 0.264026402640264,
      "grad_norm": 0.0859375,
      "learning_rate": 0.0039587458745874585,
      "loss": 0.2348,
      "num_input_tokens_seen": 507808,
      "step": 2400
    },
    {
      "epoch": 0.2645764576457646,
      "grad_norm": 0.1416015625,
      "learning_rate": 0.0039669966996699665,
      "loss": 0.2301,
      "num_input_tokens_seen": 508864,
      "step": 2405
    },
    {
      "epoch": 0.26512651265126513,
      "grad_norm": 0.078125,
      "learning_rate": 0.0039752475247524744,
      "loss": 0.2403,
      "num_input_tokens_seen": 509952,
      "step": 2410
    },
    {
      "epoch": 0.26567656765676567,
      "grad_norm": 0.060791015625,
      "learning_rate": 0.003983498349834983,
      "loss": 0.2321,
      "num_input_tokens_seen": 511008,
      "step": 2415
    },
    {
      "epoch": 0.2662266226622662,
      "grad_norm": 0.06640625,
      "learning_rate": 0.003991749174917492,
      "loss": 0.2291,
      "num_input_tokens_seen": 512032,
      "step": 2420
    },
    {
      "epoch": 0.2667766776677668,
      "grad_norm": 0.1796875,
      "learning_rate": 0.004,
      "loss": 0.2349,
      "num_input_tokens_seen": 513088,
      "step": 2425
    },
    {
      "epoch": 0.26732673267326734,
      "grad_norm": 0.1279296875,
      "learning_rate": 0.004008250825082508,
      "loss": 0.2247,
      "num_input_tokens_seen": 514176,
      "step": 2430
    },
    {
      "epoch": 0.2678767876787679,
      "grad_norm": 0.06494140625,
      "learning_rate": 0.004016501650165016,
      "loss": 0.2396,
      "num_input_tokens_seen": 515200,
      "step": 2435
    },
    {
      "epoch": 0.2684268426842684,
      "grad_norm": 0.052490234375,
      "learning_rate": 0.004024752475247525,
      "loss": 0.2343,
      "num_input_tokens_seen": 516256,
      "step": 2440
    },
    {
      "epoch": 0.26897689768976896,
      "grad_norm": 0.12890625,
      "learning_rate": 0.004033003300330033,
      "loss": 0.2365,
      "num_input_tokens_seen": 517312,
      "step": 2445
    },
    {
      "epoch": 0.26952695269526955,
      "grad_norm": 0.0966796875,
      "learning_rate": 0.004041254125412541,
      "loss": 0.2289,
      "num_input_tokens_seen": 518368,
      "step": 2450
    },
    {
      "epoch": 0.2700770077007701,
      "grad_norm": 0.12451171875,
      "learning_rate": 0.004049504950495049,
      "loss": 0.2241,
      "num_input_tokens_seen": 519424,
      "step": 2455
    },
    {
      "epoch": 0.2706270627062706,
      "grad_norm": 0.12109375,
      "learning_rate": 0.004057755775577558,
      "loss": 0.2254,
      "num_input_tokens_seen": 520512,
      "step": 2460
    },
    {
      "epoch": 0.27117711771177117,
      "grad_norm": 0.0517578125,
      "learning_rate": 0.0040660066006600664,
      "loss": 0.2518,
      "num_input_tokens_seen": 521504,
      "step": 2465
    },
    {
      "epoch": 0.2717271727172717,
      "grad_norm": 0.10498046875,
      "learning_rate": 0.004074257425742574,
      "loss": 0.2327,
      "num_input_tokens_seen": 522528,
      "step": 2470
    },
    {
      "epoch": 0.2722772277227723,
      "grad_norm": 0.08935546875,
      "learning_rate": 0.004082508250825082,
      "loss": 0.2354,
      "num_input_tokens_seen": 523616,
      "step": 2475
    },
    {
      "epoch": 0.27282728272827284,
      "grad_norm": 0.06787109375,
      "learning_rate": 0.00409075907590759,
      "loss": 0.2348,
      "num_input_tokens_seen": 524672,
      "step": 2480
    },
    {
      "epoch": 0.2733773377337734,
      "grad_norm": 0.024169921875,
      "learning_rate": 0.004099009900990099,
      "loss": 0.2356,
      "num_input_tokens_seen": 525696,
      "step": 2485
    },
    {
      "epoch": 0.2739273927392739,
      "grad_norm": 0.091796875,
      "learning_rate": 0.004107260726072607,
      "loss": 0.2324,
      "num_input_tokens_seen": 526784,
      "step": 2490
    },
    {
      "epoch": 0.27447744774477445,
      "grad_norm": 0.037841796875,
      "learning_rate": 0.004115511551155115,
      "loss": 0.2274,
      "num_input_tokens_seen": 527776,
      "step": 2495
    },
    {
      "epoch": 0.27502750275027504,
      "grad_norm": 0.060302734375,
      "learning_rate": 0.004123762376237623,
      "loss": 0.2302,
      "num_input_tokens_seen": 528832,
      "step": 2500
    },
    {
      "epoch": 0.2755775577557756,
      "grad_norm": 24.25,
      "learning_rate": 0.004132013201320131,
      "loss": 1.6856,
      "num_input_tokens_seen": 529920,
      "step": 2505
    },
    {
      "epoch": 0.2761276127612761,
      "grad_norm": 0.0751953125,
      "learning_rate": 0.004140264026402641,
      "loss": 1.2886,
      "num_input_tokens_seen": 530944,
      "step": 2510
    },
    {
      "epoch": 0.27667766776677666,
      "grad_norm": 0.044677734375,
      "learning_rate": 0.004148514851485149,
      "loss": 0.2413,
      "num_input_tokens_seen": 532032,
      "step": 2515
    },
    {
      "epoch": 0.27722772277227725,
      "grad_norm": 0.050537109375,
      "learning_rate": 0.004156765676567657,
      "loss": 0.2288,
      "num_input_tokens_seen": 533120,
      "step": 2520
    },
    {
      "epoch": 0.2777777777777778,
      "grad_norm": 0.0751953125,
      "learning_rate": 0.004165016501650165,
      "loss": 0.2367,
      "num_input_tokens_seen": 534208,
      "step": 2525
    },
    {
      "epoch": 0.27832783278327833,
      "grad_norm": 0.10400390625,
      "learning_rate": 0.0041732673267326735,
      "loss": 0.2341,
      "num_input_tokens_seen": 535232,
      "step": 2530
    },
    {
      "epoch": 0.27887788778877887,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0041815181518151815,
      "loss": 0.2366,
      "num_input_tokens_seen": 536288,
      "step": 2535
    },
    {
      "epoch": 0.2794279427942794,
      "grad_norm": 0.0546875,
      "learning_rate": 0.0041897689768976894,
      "loss": 0.2371,
      "num_input_tokens_seen": 537376,
      "step": 2540
    },
    {
      "epoch": 0.27997799779978,
      "grad_norm": 0.0242919921875,
      "learning_rate": 0.004198019801980197,
      "loss": 0.2325,
      "num_input_tokens_seen": 538432,
      "step": 2545
    },
    {
      "epoch": 0.28052805280528054,
      "grad_norm": 0.05615234375,
      "learning_rate": 0.004206270627062705,
      "loss": 0.2336,
      "num_input_tokens_seen": 539520,
      "step": 2550
    },
    {
      "epoch": 0.2810781078107811,
      "grad_norm": 0.0223388671875,
      "learning_rate": 0.004214521452145215,
      "loss": 0.2265,
      "num_input_tokens_seen": 540576,
      "step": 2555
    },
    {
      "epoch": 0.2816281628162816,
      "grad_norm": 0.038330078125,
      "learning_rate": 0.004222772277227723,
      "loss": 0.2211,
      "num_input_tokens_seen": 541632,
      "step": 2560
    },
    {
      "epoch": 0.28217821782178215,
      "grad_norm": 0.09521484375,
      "learning_rate": 0.004231023102310231,
      "loss": 0.2298,
      "num_input_tokens_seen": 542688,
      "step": 2565
    },
    {
      "epoch": 0.28272827282728275,
      "grad_norm": 0.08740234375,
      "learning_rate": 0.004239273927392739,
      "loss": 0.2218,
      "num_input_tokens_seen": 543744,
      "step": 2570
    },
    {
      "epoch": 0.2832783278327833,
      "grad_norm": 0.0673828125,
      "learning_rate": 0.004247524752475247,
      "loss": 0.2253,
      "num_input_tokens_seen": 544736,
      "step": 2575
    },
    {
      "epoch": 0.2838283828382838,
      "grad_norm": 0.1337890625,
      "learning_rate": 0.004255775577557756,
      "loss": 0.2609,
      "num_input_tokens_seen": 545824,
      "step": 2580
    },
    {
      "epoch": 0.28437843784378436,
      "grad_norm": 0.02197265625,
      "learning_rate": 0.004264026402640264,
      "loss": 0.2284,
      "num_input_tokens_seen": 546880,
      "step": 2585
    },
    {
      "epoch": 0.28492849284928495,
      "grad_norm": 0.0517578125,
      "learning_rate": 0.004272277227722772,
      "loss": 0.2327,
      "num_input_tokens_seen": 548000,
      "step": 2590
    },
    {
      "epoch": 0.2854785478547855,
      "grad_norm": 0.0849609375,
      "learning_rate": 0.00428052805280528,
      "loss": 0.2264,
      "num_input_tokens_seen": 549120,
      "step": 2595
    },
    {
      "epoch": 0.28602860286028603,
      "grad_norm": 0.0751953125,
      "learning_rate": 0.004288778877887789,
      "loss": 0.2316,
      "num_input_tokens_seen": 550208,
      "step": 2600
    },
    {
      "epoch": 0.28657865786578657,
      "grad_norm": 0.0244140625,
      "learning_rate": 0.004297029702970297,
      "loss": 0.2405,
      "num_input_tokens_seen": 551264,
      "step": 2605
    },
    {
      "epoch": 0.2871287128712871,
      "grad_norm": 0.049560546875,
      "learning_rate": 0.004305280528052805,
      "loss": 0.2358,
      "num_input_tokens_seen": 552320,
      "step": 2610
    },
    {
      "epoch": 0.2876787678767877,
      "grad_norm": 0.02001953125,
      "learning_rate": 0.004313531353135313,
      "loss": 0.2302,
      "num_input_tokens_seen": 553312,
      "step": 2615
    },
    {
      "epoch": 0.28822882288228824,
      "grad_norm": 0.01904296875,
      "learning_rate": 0.004321782178217821,
      "loss": 0.2349,
      "num_input_tokens_seen": 554368,
      "step": 2620
    },
    {
      "epoch": 0.2887788778877888,
      "grad_norm": 0.043212890625,
      "learning_rate": 0.00433003300330033,
      "loss": 0.2245,
      "num_input_tokens_seen": 555392,
      "step": 2625
    },
    {
      "epoch": 0.2893289328932893,
      "grad_norm": 0.0201416015625,
      "learning_rate": 0.004338283828382838,
      "loss": 0.2294,
      "num_input_tokens_seen": 556384,
      "step": 2630
    },
    {
      "epoch": 0.28987898789878985,
      "grad_norm": 0.032470703125,
      "learning_rate": 0.004346534653465346,
      "loss": 0.2309,
      "num_input_tokens_seen": 557472,
      "step": 2635
    },
    {
      "epoch": 0.29042904290429045,
      "grad_norm": 0.06396484375,
      "learning_rate": 0.004354785478547854,
      "loss": 0.2288,
      "num_input_tokens_seen": 558528,
      "step": 2640
    },
    {
      "epoch": 0.290979097909791,
      "grad_norm": 0.04345703125,
      "learning_rate": 0.004363036303630363,
      "loss": 0.235,
      "num_input_tokens_seen": 559520,
      "step": 2645
    },
    {
      "epoch": 0.2915291529152915,
      "grad_norm": 0.0712890625,
      "learning_rate": 0.004371287128712872,
      "loss": 0.235,
      "num_input_tokens_seen": 560512,
      "step": 2650
    },
    {
      "epoch": 0.29207920792079206,
      "grad_norm": 0.03662109375,
      "learning_rate": 0.00437953795379538,
      "loss": 0.2349,
      "num_input_tokens_seen": 561568,
      "step": 2655
    },
    {
      "epoch": 0.29262926292629265,
      "grad_norm": 0.03564453125,
      "learning_rate": 0.004387788778877888,
      "loss": 0.2285,
      "num_input_tokens_seen": 562528,
      "step": 2660
    },
    {
      "epoch": 0.2931793179317932,
      "grad_norm": 0.052734375,
      "learning_rate": 0.004396039603960396,
      "loss": 0.2319,
      "num_input_tokens_seen": 563552,
      "step": 2665
    },
    {
      "epoch": 0.29372937293729373,
      "grad_norm": 0.06787109375,
      "learning_rate": 0.0044042904290429044,
      "loss": 0.2359,
      "num_input_tokens_seen": 564576,
      "step": 2670
    },
    {
      "epoch": 0.29427942794279427,
      "grad_norm": 0.037109375,
      "learning_rate": 0.004412541254125412,
      "loss": 0.2398,
      "num_input_tokens_seen": 565696,
      "step": 2675
    },
    {
      "epoch": 0.2948294829482948,
      "grad_norm": 0.03955078125,
      "learning_rate": 0.00442079207920792,
      "loss": 0.2252,
      "num_input_tokens_seen": 566688,
      "step": 2680
    },
    {
      "epoch": 0.2953795379537954,
      "grad_norm": 0.058349609375,
      "learning_rate": 0.004429042904290428,
      "loss": 0.2369,
      "num_input_tokens_seen": 567712,
      "step": 2685
    },
    {
      "epoch": 0.29592959295929594,
      "grad_norm": 0.044921875,
      "learning_rate": 0.004437293729372937,
      "loss": 0.2373,
      "num_input_tokens_seen": 568864,
      "step": 2690
    },
    {
      "epoch": 0.2964796479647965,
      "grad_norm": 0.046142578125,
      "learning_rate": 0.004445544554455446,
      "loss": 0.2317,
      "num_input_tokens_seen": 569888,
      "step": 2695
    },
    {
      "epoch": 0.297029702970297,
      "grad_norm": 0.039306640625,
      "learning_rate": 0.004453795379537954,
      "loss": 0.2226,
      "num_input_tokens_seen": 570944,
      "step": 2700
    },
    {
      "epoch": 0.29757975797579755,
      "grad_norm": 0.08203125,
      "learning_rate": 0.004462046204620462,
      "loss": 0.2294,
      "num_input_tokens_seen": 572000,
      "step": 2705
    },
    {
      "epoch": 0.29812981298129815,
      "grad_norm": 0.034912109375,
      "learning_rate": 0.00447029702970297,
      "loss": 0.2353,
      "num_input_tokens_seen": 572992,
      "step": 2710
    },
    {
      "epoch": 0.2986798679867987,
      "grad_norm": 0.083984375,
      "learning_rate": 0.004478547854785478,
      "loss": 0.2431,
      "num_input_tokens_seen": 574048,
      "step": 2715
    },
    {
      "epoch": 0.2992299229922992,
      "grad_norm": 0.0625,
      "learning_rate": 0.004486798679867987,
      "loss": 0.2324,
      "num_input_tokens_seen": 575136,
      "step": 2720
    },
    {
      "epoch": 0.29977997799779976,
      "grad_norm": 0.07958984375,
      "learning_rate": 0.004495049504950495,
      "loss": 0.2297,
      "num_input_tokens_seen": 576160,
      "step": 2725
    },
    {
      "epoch": 0.30033003300330036,
      "grad_norm": 0.026611328125,
      "learning_rate": 0.004503300330033003,
      "loss": 0.232,
      "num_input_tokens_seen": 577184,
      "step": 2730
    },
    {
      "epoch": 0.3008800880088009,
      "grad_norm": 0.0289306640625,
      "learning_rate": 0.0045115511551155115,
      "loss": 0.2307,
      "num_input_tokens_seen": 578208,
      "step": 2735
    },
    {
      "epoch": 0.30143014301430143,
      "grad_norm": 0.053466796875,
      "learning_rate": 0.00451980198019802,
      "loss": 0.2316,
      "num_input_tokens_seen": 579296,
      "step": 2740
    },
    {
      "epoch": 0.30198019801980197,
      "grad_norm": 0.083984375,
      "learning_rate": 0.004528052805280528,
      "loss": 0.2305,
      "num_input_tokens_seen": 580320,
      "step": 2745
    },
    {
      "epoch": 0.3025302530253025,
      "grad_norm": 0.015869140625,
      "learning_rate": 0.004536303630363036,
      "loss": 0.2316,
      "num_input_tokens_seen": 581344,
      "step": 2750
    },
    {
      "epoch": 0.3030803080308031,
      "grad_norm": 0.0322265625,
      "learning_rate": 0.004544554455445544,
      "loss": 0.2339,
      "num_input_tokens_seen": 582368,
      "step": 2755
    },
    {
      "epoch": 0.30363036303630364,
      "grad_norm": 0.08349609375,
      "learning_rate": 0.004552805280528052,
      "loss": 0.2328,
      "num_input_tokens_seen": 583424,
      "step": 2760
    },
    {
      "epoch": 0.3041804180418042,
      "grad_norm": 0.027587890625,
      "learning_rate": 0.004561056105610561,
      "loss": 0.2285,
      "num_input_tokens_seen": 584512,
      "step": 2765
    },
    {
      "epoch": 0.3047304730473047,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.004569306930693069,
      "loss": 0.2295,
      "num_input_tokens_seen": 585536,
      "step": 2770
    },
    {
      "epoch": 0.30528052805280526,
      "grad_norm": 0.041259765625,
      "learning_rate": 0.004577557755775577,
      "loss": 0.2326,
      "num_input_tokens_seen": 586624,
      "step": 2775
    },
    {
      "epoch": 0.30583058305830585,
      "grad_norm": 0.037841796875,
      "learning_rate": 0.004585808580858086,
      "loss": 0.2327,
      "num_input_tokens_seen": 587712,
      "step": 2780
    },
    {
      "epoch": 0.3063806380638064,
      "grad_norm": 0.07958984375,
      "learning_rate": 0.004594059405940594,
      "loss": 0.2265,
      "num_input_tokens_seen": 588736,
      "step": 2785
    },
    {
      "epoch": 0.3069306930693069,
      "grad_norm": 0.01416015625,
      "learning_rate": 0.004602310231023103,
      "loss": 0.2297,
      "num_input_tokens_seen": 589792,
      "step": 2790
    },
    {
      "epoch": 0.30748074807480746,
      "grad_norm": 0.035888671875,
      "learning_rate": 0.004610561056105611,
      "loss": 0.2219,
      "num_input_tokens_seen": 590912,
      "step": 2795
    },
    {
      "epoch": 0.30803080308030806,
      "grad_norm": 0.056884765625,
      "learning_rate": 0.004618811881188119,
      "loss": 0.2282,
      "num_input_tokens_seen": 591968,
      "step": 2800
    },
    {
      "epoch": 0.3085808580858086,
      "grad_norm": 0.08203125,
      "learning_rate": 0.0046270627062706265,
      "loss": 0.2136,
      "num_input_tokens_seen": 593024,
      "step": 2805
    },
    {
      "epoch": 0.30913091309130913,
      "grad_norm": 0.154296875,
      "learning_rate": 0.004635313531353135,
      "loss": 0.2287,
      "num_input_tokens_seen": 594048,
      "step": 2810
    },
    {
      "epoch": 0.3096809680968097,
      "grad_norm": 0.07080078125,
      "learning_rate": 0.004643564356435643,
      "loss": 0.2558,
      "num_input_tokens_seen": 595072,
      "step": 2815
    },
    {
      "epoch": 0.3102310231023102,
      "grad_norm": 0.08349609375,
      "learning_rate": 0.004651815181518151,
      "loss": 0.2416,
      "num_input_tokens_seen": 596096,
      "step": 2820
    },
    {
      "epoch": 0.3107810781078108,
      "grad_norm": 0.09130859375,
      "learning_rate": 0.00466006600660066,
      "loss": 0.234,
      "num_input_tokens_seen": 597184,
      "step": 2825
    },
    {
      "epoch": 0.31133113311331134,
      "grad_norm": 0.041015625,
      "learning_rate": 0.004668316831683168,
      "loss": 0.2284,
      "num_input_tokens_seen": 598240,
      "step": 2830
    },
    {
      "epoch": 0.3118811881188119,
      "grad_norm": 0.04833984375,
      "learning_rate": 0.004676567656765677,
      "loss": 0.2335,
      "num_input_tokens_seen": 599328,
      "step": 2835
    },
    {
      "epoch": 0.3124312431243124,
      "grad_norm": 0.08349609375,
      "learning_rate": 0.004684818481848185,
      "loss": 0.2315,
      "num_input_tokens_seen": 600384,
      "step": 2840
    },
    {
      "epoch": 0.31298129812981296,
      "grad_norm": 0.0771484375,
      "learning_rate": 0.004693069306930693,
      "loss": 0.2316,
      "num_input_tokens_seen": 601472,
      "step": 2845
    },
    {
      "epoch": 0.31353135313531355,
      "grad_norm": 0.0159912109375,
      "learning_rate": 0.004701320132013201,
      "loss": 0.241,
      "num_input_tokens_seen": 602560,
      "step": 2850
    },
    {
      "epoch": 0.3140814081408141,
      "grad_norm": 0.0152587890625,
      "learning_rate": 0.004709570957095709,
      "loss": 0.2251,
      "num_input_tokens_seen": 603584,
      "step": 2855
    },
    {
      "epoch": 0.3146314631463146,
      "grad_norm": 0.0439453125,
      "learning_rate": 0.004717821782178218,
      "loss": 0.2348,
      "num_input_tokens_seen": 604640,
      "step": 2860
    },
    {
      "epoch": 0.31518151815181517,
      "grad_norm": 0.043212890625,
      "learning_rate": 0.004726072607260726,
      "loss": 0.2202,
      "num_input_tokens_seen": 605760,
      "step": 2865
    },
    {
      "epoch": 0.31573157315731576,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.0047343234323432345,
      "loss": 0.2393,
      "num_input_tokens_seen": 606848,
      "step": 2870
    },
    {
      "epoch": 0.3162816281628163,
      "grad_norm": 0.0654296875,
      "learning_rate": 0.0047425742574257424,
      "loss": 0.2227,
      "num_input_tokens_seen": 607904,
      "step": 2875
    },
    {
      "epoch": 0.31683168316831684,
      "grad_norm": 0.10205078125,
      "learning_rate": 0.00475082508250825,
      "loss": 0.2349,
      "num_input_tokens_seen": 608992,
      "step": 2880
    },
    {
      "epoch": 0.3173817381738174,
      "grad_norm": 0.031494140625,
      "learning_rate": 0.004759075907590759,
      "loss": 0.2316,
      "num_input_tokens_seen": 610016,
      "step": 2885
    },
    {
      "epoch": 0.3179317931793179,
      "grad_norm": 0.044189453125,
      "learning_rate": 0.004767326732673267,
      "loss": 0.2321,
      "num_input_tokens_seen": 611072,
      "step": 2890
    },
    {
      "epoch": 0.3184818481848185,
      "grad_norm": 0.02685546875,
      "learning_rate": 0.004775577557755775,
      "loss": 0.233,
      "num_input_tokens_seen": 612160,
      "step": 2895
    },
    {
      "epoch": 0.31903190319031904,
      "grad_norm": 0.042724609375,
      "learning_rate": 0.004783828382838283,
      "loss": 0.2356,
      "num_input_tokens_seen": 613216,
      "step": 2900
    },
    {
      "epoch": 0.3195819581958196,
      "grad_norm": 0.059814453125,
      "learning_rate": 0.004792079207920792,
      "loss": 0.2341,
      "num_input_tokens_seen": 614304,
      "step": 2905
    },
    {
      "epoch": 0.3201320132013201,
      "grad_norm": 0.0634765625,
      "learning_rate": 0.0048003300330033,
      "loss": 0.2327,
      "num_input_tokens_seen": 615360,
      "step": 2910
    },
    {
      "epoch": 0.32068206820682066,
      "grad_norm": 0.02294921875,
      "learning_rate": 0.004808580858085809,
      "loss": 0.2306,
      "num_input_tokens_seen": 616416,
      "step": 2915
    },
    {
      "epoch": 0.32123212321232125,
      "grad_norm": 0.037109375,
      "learning_rate": 0.004816831683168317,
      "loss": 0.2408,
      "num_input_tokens_seen": 617472,
      "step": 2920
    },
    {
      "epoch": 0.3217821782178218,
      "grad_norm": 0.0242919921875,
      "learning_rate": 0.004825082508250825,
      "loss": 0.2313,
      "num_input_tokens_seen": 618528,
      "step": 2925
    },
    {
      "epoch": 0.32233223322332233,
      "grad_norm": 0.0634765625,
      "learning_rate": 0.004833333333333334,
      "loss": 0.2284,
      "num_input_tokens_seen": 619648,
      "step": 2930
    },
    {
      "epoch": 0.32288228822882287,
      "grad_norm": 0.0322265625,
      "learning_rate": 0.0048415841584158415,
      "loss": 0.2337,
      "num_input_tokens_seen": 620704,
      "step": 2935
    },
    {
      "epoch": 0.3234323432343234,
      "grad_norm": 0.040771484375,
      "learning_rate": 0.0048498349834983495,
      "loss": 0.2372,
      "num_input_tokens_seen": 621760,
      "step": 2940
    },
    {
      "epoch": 0.323982398239824,
      "grad_norm": 0.06787109375,
      "learning_rate": 0.0048580858085808575,
      "loss": 0.2336,
      "num_input_tokens_seen": 622880,
      "step": 2945
    },
    {
      "epoch": 0.32453245324532454,
      "grad_norm": 0.034423828125,
      "learning_rate": 0.0048663366336633655,
      "loss": 0.2305,
      "num_input_tokens_seen": 623904,
      "step": 2950
    },
    {
      "epoch": 0.3250825082508251,
      "grad_norm": 0.0211181640625,
      "learning_rate": 0.004874587458745874,
      "loss": 0.2315,
      "num_input_tokens_seen": 625056,
      "step": 2955
    },
    {
      "epoch": 0.3256325632563256,
      "grad_norm": 0.03515625,
      "learning_rate": 0.004882838283828383,
      "loss": 0.2304,
      "num_input_tokens_seen": 626176,
      "step": 2960
    },
    {
      "epoch": 0.3261826182618262,
      "grad_norm": 0.040771484375,
      "learning_rate": 0.004891089108910891,
      "loss": 0.2329,
      "num_input_tokens_seen": 627200,
      "step": 2965
    },
    {
      "epoch": 0.32673267326732675,
      "grad_norm": 0.047607421875,
      "learning_rate": 0.004899339933993399,
      "loss": 0.2332,
      "num_input_tokens_seen": 628256,
      "step": 2970
    },
    {
      "epoch": 0.3272827282728273,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.004907590759075908,
      "loss": 0.2362,
      "num_input_tokens_seen": 629376,
      "step": 2975
    },
    {
      "epoch": 0.3278327832783278,
      "grad_norm": 0.043212890625,
      "learning_rate": 0.004915841584158416,
      "loss": 0.2337,
      "num_input_tokens_seen": 630432,
      "step": 2980
    },
    {
      "epoch": 0.32838283828382836,
      "grad_norm": 0.07666015625,
      "learning_rate": 0.004924092409240924,
      "loss": 0.2298,
      "num_input_tokens_seen": 631520,
      "step": 2985
    },
    {
      "epoch": 0.32893289328932895,
      "grad_norm": 0.07177734375,
      "learning_rate": 0.004932343234323432,
      "loss": 0.2395,
      "num_input_tokens_seen": 632608,
      "step": 2990
    },
    {
      "epoch": 0.3294829482948295,
      "grad_norm": 0.033203125,
      "learning_rate": 0.00494059405940594,
      "loss": 0.2392,
      "num_input_tokens_seen": 633664,
      "step": 2995
    },
    {
      "epoch": 0.33003300330033003,
      "grad_norm": 0.06201171875,
      "learning_rate": 0.004948844884488449,
      "loss": 0.2367,
      "num_input_tokens_seen": 634688,
      "step": 3000
    },
    {
      "epoch": 0.33058305830583057,
      "grad_norm": 0.035400390625,
      "learning_rate": 0.0049570957095709575,
      "loss": 0.2325,
      "num_input_tokens_seen": 635744,
      "step": 3005
    },
    {
      "epoch": 0.3311331133113311,
      "grad_norm": 0.03173828125,
      "learning_rate": 0.004965346534653465,
      "loss": 0.2313,
      "num_input_tokens_seen": 636864,
      "step": 3010
    },
    {
      "epoch": 0.3316831683168317,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.004973597359735973,
      "loss": 0.2324,
      "num_input_tokens_seen": 637888,
      "step": 3015
    },
    {
      "epoch": 0.33223322332233224,
      "grad_norm": 0.034912109375,
      "learning_rate": 0.004981848184818481,
      "loss": 0.2325,
      "num_input_tokens_seen": 638944,
      "step": 3020
    },
    {
      "epoch": 0.3327832783278328,
      "grad_norm": 0.030517578125,
      "learning_rate": 0.00499009900990099,
      "loss": 0.2284,
      "num_input_tokens_seen": 640000,
      "step": 3025
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 0.03466796875,
      "learning_rate": 0.004998349834983498,
      "loss": 0.2303,
      "num_input_tokens_seen": 641024,
      "step": 3030
    },
    {
      "epoch": 0.3338833883388339,
      "grad_norm": 0.031494140625,
      "learning_rate": 0.005006600660066006,
      "loss": 0.2303,
      "num_input_tokens_seen": 642080,
      "step": 3035
    },
    {
      "epoch": 0.33443344334433445,
      "grad_norm": 0.022705078125,
      "learning_rate": 0.005014851485148514,
      "loss": 0.2294,
      "num_input_tokens_seen": 643104,
      "step": 3040
    },
    {
      "epoch": 0.334983498349835,
      "grad_norm": 0.03564453125,
      "learning_rate": 0.005023102310231023,
      "loss": 0.2314,
      "num_input_tokens_seen": 644128,
      "step": 3045
    },
    {
      "epoch": 0.3355335533553355,
      "grad_norm": 0.03955078125,
      "learning_rate": 0.005031353135313532,
      "loss": 0.235,
      "num_input_tokens_seen": 645216,
      "step": 3050
    },
    {
      "epoch": 0.33608360836083606,
      "grad_norm": 0.0198974609375,
      "learning_rate": 0.00503960396039604,
      "loss": 0.2311,
      "num_input_tokens_seen": 646272,
      "step": 3055
    },
    {
      "epoch": 0.33663366336633666,
      "grad_norm": 3.09375,
      "learning_rate": 0.005047854785478548,
      "loss": 0.7249,
      "num_input_tokens_seen": 647392,
      "step": 3060
    },
    {
      "epoch": 0.3371837183718372,
      "grad_norm": 0.0703125,
      "learning_rate": 0.005056105610561056,
      "loss": 0.2337,
      "num_input_tokens_seen": 648448,
      "step": 3065
    },
    {
      "epoch": 0.33773377337733773,
      "grad_norm": 0.10302734375,
      "learning_rate": 0.0050643564356435645,
      "loss": 0.2425,
      "num_input_tokens_seen": 649472,
      "step": 3070
    },
    {
      "epoch": 0.33828382838283827,
      "grad_norm": 0.0167236328125,
      "learning_rate": 0.0050726072607260725,
      "loss": 0.2317,
      "num_input_tokens_seen": 650464,
      "step": 3075
    },
    {
      "epoch": 0.3388338833883388,
      "grad_norm": 0.014404296875,
      "learning_rate": 0.0050808580858085805,
      "loss": 0.2335,
      "num_input_tokens_seen": 651520,
      "step": 3080
    },
    {
      "epoch": 0.3393839383938394,
      "grad_norm": 0.1328125,
      "learning_rate": 0.005089108910891088,
      "loss": 0.2272,
      "num_input_tokens_seen": 652608,
      "step": 3085
    },
    {
      "epoch": 0.33993399339933994,
      "grad_norm": 0.10546875,
      "learning_rate": 0.005097359735973596,
      "loss": 0.2436,
      "num_input_tokens_seen": 653600,
      "step": 3090
    },
    {
      "epoch": 0.3404840484048405,
      "grad_norm": 0.05810546875,
      "learning_rate": 0.005105610561056106,
      "loss": 0.2349,
      "num_input_tokens_seen": 654656,
      "step": 3095
    },
    {
      "epoch": 0.341034103410341,
      "grad_norm": 0.01336669921875,
      "learning_rate": 0.005113861386138614,
      "loss": 0.2276,
      "num_input_tokens_seen": 655712,
      "step": 3100
    },
    {
      "epoch": 0.3415841584158416,
      "grad_norm": 0.016845703125,
      "learning_rate": 0.005122112211221122,
      "loss": 0.2166,
      "num_input_tokens_seen": 656736,
      "step": 3105
    },
    {
      "epoch": 0.34213421342134215,
      "grad_norm": 0.042236328125,
      "learning_rate": 0.00513036303630363,
      "loss": 0.1895,
      "num_input_tokens_seen": 657824,
      "step": 3110
    },
    {
      "epoch": 0.3426842684268427,
      "grad_norm": 0.03271484375,
      "learning_rate": 0.005138613861386139,
      "loss": 0.2478,
      "num_input_tokens_seen": 658944,
      "step": 3115
    },
    {
      "epoch": 0.3432343234323432,
      "grad_norm": 0.022216796875,
      "learning_rate": 0.005146864686468647,
      "loss": 0.2563,
      "num_input_tokens_seen": 660032,
      "step": 3120
    },
    {
      "epoch": 0.34378437843784376,
      "grad_norm": 0.057373046875,
      "learning_rate": 0.005155115511551155,
      "loss": 0.2437,
      "num_input_tokens_seen": 661088,
      "step": 3125
    },
    {
      "epoch": 0.34433443344334436,
      "grad_norm": 0.0203857421875,
      "learning_rate": 0.005163366336633663,
      "loss": 0.2329,
      "num_input_tokens_seen": 662112,
      "step": 3130
    },
    {
      "epoch": 0.3448844884488449,
      "grad_norm": 0.047119140625,
      "learning_rate": 0.005171617161716171,
      "loss": 0.2333,
      "num_input_tokens_seen": 663072,
      "step": 3135
    },
    {
      "epoch": 0.34543454345434543,
      "grad_norm": 0.08154296875,
      "learning_rate": 0.00517986798679868,
      "loss": 0.2303,
      "num_input_tokens_seen": 664128,
      "step": 3140
    },
    {
      "epoch": 0.34598459845984597,
      "grad_norm": 0.037841796875,
      "learning_rate": 0.005188118811881188,
      "loss": 0.2305,
      "num_input_tokens_seen": 665184,
      "step": 3145
    },
    {
      "epoch": 0.3465346534653465,
      "grad_norm": 0.054443359375,
      "learning_rate": 0.005196369636963696,
      "loss": 0.2389,
      "num_input_tokens_seen": 666240,
      "step": 3150
    },
    {
      "epoch": 0.3470847084708471,
      "grad_norm": 0.04931640625,
      "learning_rate": 0.005204620462046204,
      "loss": 0.2299,
      "num_input_tokens_seen": 667232,
      "step": 3155
    },
    {
      "epoch": 0.34763476347634764,
      "grad_norm": 0.046630859375,
      "learning_rate": 0.005212871287128712,
      "loss": 0.23,
      "num_input_tokens_seen": 668320,
      "step": 3160
    },
    {
      "epoch": 0.3481848184818482,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.005221122112211221,
      "loss": 0.2287,
      "num_input_tokens_seen": 669344,
      "step": 3165
    },
    {
      "epoch": 0.3487348734873487,
      "grad_norm": 0.01361083984375,
      "learning_rate": 0.005229372937293729,
      "loss": 0.2347,
      "num_input_tokens_seen": 670400,
      "step": 3170
    },
    {
      "epoch": 0.3492849284928493,
      "grad_norm": 0.039306640625,
      "learning_rate": 0.005237623762376237,
      "loss": 0.2346,
      "num_input_tokens_seen": 671488,
      "step": 3175
    },
    {
      "epoch": 0.34983498349834985,
      "grad_norm": 0.0390625,
      "learning_rate": 0.005245874587458745,
      "loss": 0.2323,
      "num_input_tokens_seen": 672576,
      "step": 3180
    },
    {
      "epoch": 0.3503850385038504,
      "grad_norm": 0.04296875,
      "learning_rate": 0.005254125412541255,
      "loss": 0.2306,
      "num_input_tokens_seen": 673696,
      "step": 3185
    },
    {
      "epoch": 0.3509350935093509,
      "grad_norm": 0.09375,
      "learning_rate": 0.005262376237623763,
      "loss": 0.2296,
      "num_input_tokens_seen": 674688,
      "step": 3190
    },
    {
      "epoch": 0.35148514851485146,
      "grad_norm": 0.01397705078125,
      "learning_rate": 0.005270627062706271,
      "loss": 0.2316,
      "num_input_tokens_seen": 675744,
      "step": 3195
    },
    {
      "epoch": 0.35203520352035206,
      "grad_norm": 0.030517578125,
      "learning_rate": 0.005278877887788779,
      "loss": 0.2324,
      "num_input_tokens_seen": 676800,
      "step": 3200
    },
    {
      "epoch": 0.3525852585258526,
      "grad_norm": 0.051513671875,
      "learning_rate": 0.005287128712871287,
      "loss": 0.2327,
      "num_input_tokens_seen": 677856,
      "step": 3205
    },
    {
      "epoch": 0.35313531353135313,
      "grad_norm": 0.047119140625,
      "learning_rate": 0.0052953795379537955,
      "loss": 0.2284,
      "num_input_tokens_seen": 678912,
      "step": 3210
    },
    {
      "epoch": 0.3536853685368537,
      "grad_norm": 0.03662109375,
      "learning_rate": 0.005303630363036303,
      "loss": 0.2305,
      "num_input_tokens_seen": 679904,
      "step": 3215
    },
    {
      "epoch": 0.3542354235423542,
      "grad_norm": 0.087890625,
      "learning_rate": 0.005311881188118811,
      "loss": 0.2835,
      "num_input_tokens_seen": 681024,
      "step": 3220
    },
    {
      "epoch": 0.3547854785478548,
      "grad_norm": 0.06689453125,
      "learning_rate": 0.005320132013201319,
      "loss": 0.2501,
      "num_input_tokens_seen": 682080,
      "step": 3225
    },
    {
      "epoch": 0.35533553355335534,
      "grad_norm": 0.0400390625,
      "learning_rate": 0.005328382838283828,
      "loss": 0.2381,
      "num_input_tokens_seen": 683136,
      "step": 3230
    },
    {
      "epoch": 0.3558855885588559,
      "grad_norm": 0.0390625,
      "learning_rate": 0.005336633663366337,
      "loss": 0.2316,
      "num_input_tokens_seen": 684256,
      "step": 3235
    },
    {
      "epoch": 0.3564356435643564,
      "grad_norm": 0.208984375,
      "learning_rate": 0.005344884488448845,
      "loss": 0.2291,
      "num_input_tokens_seen": 685344,
      "step": 3240
    },
    {
      "epoch": 0.356985698569857,
      "grad_norm": 0.034912109375,
      "learning_rate": 0.005353135313531353,
      "loss": 0.2584,
      "num_input_tokens_seen": 686304,
      "step": 3245
    },
    {
      "epoch": 0.35753575357535755,
      "grad_norm": 0.045166015625,
      "learning_rate": 0.005361386138613861,
      "loss": 0.2332,
      "num_input_tokens_seen": 687328,
      "step": 3250
    },
    {
      "epoch": 0.3580858085808581,
      "grad_norm": 0.01806640625,
      "learning_rate": 0.005369636963696369,
      "loss": 0.2189,
      "num_input_tokens_seen": 688320,
      "step": 3255
    },
    {
      "epoch": 0.3586358635863586,
      "grad_norm": 0.0322265625,
      "learning_rate": 0.005377887788778878,
      "loss": 0.2359,
      "num_input_tokens_seen": 689376,
      "step": 3260
    },
    {
      "epoch": 0.35918591859185917,
      "grad_norm": 0.051513671875,
      "learning_rate": 0.005386138613861386,
      "loss": 0.2364,
      "num_input_tokens_seen": 690336,
      "step": 3265
    },
    {
      "epoch": 0.35973597359735976,
      "grad_norm": 0.07958984375,
      "learning_rate": 0.005394389438943894,
      "loss": 0.238,
      "num_input_tokens_seen": 691456,
      "step": 3270
    },
    {
      "epoch": 0.3602860286028603,
      "grad_norm": 0.041259765625,
      "learning_rate": 0.0054026402640264025,
      "loss": 0.2347,
      "num_input_tokens_seen": 692512,
      "step": 3275
    },
    {
      "epoch": 0.36083608360836084,
      "grad_norm": 0.0751953125,
      "learning_rate": 0.005410891089108911,
      "loss": 0.2277,
      "num_input_tokens_seen": 693536,
      "step": 3280
    },
    {
      "epoch": 0.3613861386138614,
      "grad_norm": 0.064453125,
      "learning_rate": 0.005419141914191419,
      "loss": 0.234,
      "num_input_tokens_seen": 694592,
      "step": 3285
    },
    {
      "epoch": 0.3619361936193619,
      "grad_norm": 0.0147705078125,
      "learning_rate": 0.005427392739273927,
      "loss": 0.2415,
      "num_input_tokens_seen": 695648,
      "step": 3290
    },
    {
      "epoch": 0.3624862486248625,
      "grad_norm": 0.142578125,
      "learning_rate": 0.005435643564356435,
      "loss": 0.2336,
      "num_input_tokens_seen": 696768,
      "step": 3295
    },
    {
      "epoch": 0.36303630363036304,
      "grad_norm": 0.0830078125,
      "learning_rate": 0.005443894389438943,
      "loss": 0.236,
      "num_input_tokens_seen": 697824,
      "step": 3300
    },
    {
      "epoch": 0.3635863586358636,
      "grad_norm": 0.0703125,
      "learning_rate": 0.005452145214521452,
      "loss": 0.2298,
      "num_input_tokens_seen": 698912,
      "step": 3305
    },
    {
      "epoch": 0.3641364136413641,
      "grad_norm": 0.02392578125,
      "learning_rate": 0.00546039603960396,
      "loss": 0.2241,
      "num_input_tokens_seen": 699904,
      "step": 3310
    },
    {
      "epoch": 0.36468646864686466,
      "grad_norm": 0.07666015625,
      "learning_rate": 0.005468646864686468,
      "loss": 0.2376,
      "num_input_tokens_seen": 700928,
      "step": 3315
    },
    {
      "epoch": 0.36523652365236525,
      "grad_norm": 0.0556640625,
      "learning_rate": 0.005476897689768977,
      "loss": 0.237,
      "num_input_tokens_seen": 701952,
      "step": 3320
    },
    {
      "epoch": 0.3657865786578658,
      "grad_norm": 0.036865234375,
      "learning_rate": 0.005485148514851485,
      "loss": 0.2295,
      "num_input_tokens_seen": 702976,
      "step": 3325
    },
    {
      "epoch": 0.36633663366336633,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.005493399339933994,
      "loss": 0.2253,
      "num_input_tokens_seen": 704064,
      "step": 3330
    },
    {
      "epoch": 0.36688668866886687,
      "grad_norm": 0.03857421875,
      "learning_rate": 0.005501650165016502,
      "loss": 0.2424,
      "num_input_tokens_seen": 705152,
      "step": 3335
    },
    {
      "epoch": 0.36743674367436746,
      "grad_norm": 0.028564453125,
      "learning_rate": 0.00550990099009901,
      "loss": 0.2348,
      "num_input_tokens_seen": 706240,
      "step": 3340
    },
    {
      "epoch": 0.367986798679868,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.0055181518151815176,
      "loss": 0.2323,
      "num_input_tokens_seen": 707232,
      "step": 3345
    },
    {
      "epoch": 0.36853685368536854,
      "grad_norm": 0.03173828125,
      "learning_rate": 0.005526402640264026,
      "loss": 0.2323,
      "num_input_tokens_seen": 708352,
      "step": 3350
    },
    {
      "epoch": 0.3690869086908691,
      "grad_norm": 0.02978515625,
      "learning_rate": 0.005534653465346534,
      "loss": 0.2334,
      "num_input_tokens_seen": 709408,
      "step": 3355
    },
    {
      "epoch": 0.3696369636963696,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.005542904290429042,
      "loss": 0.2313,
      "num_input_tokens_seen": 710496,
      "step": 3360
    },
    {
      "epoch": 0.3701870187018702,
      "grad_norm": 0.029296875,
      "learning_rate": 0.005551155115511551,
      "loss": 0.2303,
      "num_input_tokens_seen": 711616,
      "step": 3365
    },
    {
      "epoch": 0.37073707370737075,
      "grad_norm": 0.0306396484375,
      "learning_rate": 0.005559405940594059,
      "loss": 0.2275,
      "num_input_tokens_seen": 712672,
      "step": 3370
    },
    {
      "epoch": 0.3712871287128713,
      "grad_norm": 0.0289306640625,
      "learning_rate": 0.005567656765676568,
      "loss": 0.2328,
      "num_input_tokens_seen": 713760,
      "step": 3375
    },
    {
      "epoch": 0.3718371837183718,
      "grad_norm": 0.01275634765625,
      "learning_rate": 0.005575907590759076,
      "loss": 0.236,
      "num_input_tokens_seen": 714912,
      "step": 3380
    },
    {
      "epoch": 0.37238723872387236,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.005584158415841584,
      "loss": 0.2287,
      "num_input_tokens_seen": 715968,
      "step": 3385
    },
    {
      "epoch": 0.37293729372937295,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.005592409240924092,
      "loss": 0.2329,
      "num_input_tokens_seen": 717024,
      "step": 3390
    },
    {
      "epoch": 0.3734873487348735,
      "grad_norm": 0.03369140625,
      "learning_rate": 0.0056006600660066,
      "loss": 0.237,
      "num_input_tokens_seen": 718112,
      "step": 3395
    },
    {
      "epoch": 0.37403740374037403,
      "grad_norm": 0.03173828125,
      "learning_rate": 0.005608910891089109,
      "loss": 0.2326,
      "num_input_tokens_seen": 719232,
      "step": 3400
    },
    {
      "epoch": 0.37458745874587457,
      "grad_norm": 0.02978515625,
      "learning_rate": 0.005617161716171617,
      "loss": 0.2282,
      "num_input_tokens_seen": 720288,
      "step": 3405
    },
    {
      "epoch": 0.37513751375137516,
      "grad_norm": 0.05859375,
      "learning_rate": 0.0056254125412541255,
      "loss": 0.2324,
      "num_input_tokens_seen": 721312,
      "step": 3410
    },
    {
      "epoch": 0.3756875687568757,
      "grad_norm": 0.055908203125,
      "learning_rate": 0.0056336633663366335,
      "loss": 0.2313,
      "num_input_tokens_seen": 722400,
      "step": 3415
    },
    {
      "epoch": 0.37623762376237624,
      "grad_norm": 0.0341796875,
      "learning_rate": 0.005641914191419142,
      "loss": 0.2304,
      "num_input_tokens_seen": 723392,
      "step": 3420
    },
    {
      "epoch": 0.3767876787678768,
      "grad_norm": 0.01446533203125,
      "learning_rate": 0.00565016501650165,
      "loss": 0.2303,
      "num_input_tokens_seen": 724448,
      "step": 3425
    },
    {
      "epoch": 0.3773377337733773,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.005658415841584158,
      "loss": 0.2313,
      "num_input_tokens_seen": 725440,
      "step": 3430
    },
    {
      "epoch": 0.3778877887788779,
      "grad_norm": 0.05322265625,
      "learning_rate": 0.005666666666666666,
      "loss": 0.2284,
      "num_input_tokens_seen": 726496,
      "step": 3435
    },
    {
      "epoch": 0.37843784378437845,
      "grad_norm": 0.0072021484375,
      "learning_rate": 0.005674917491749174,
      "loss": 0.2378,
      "num_input_tokens_seen": 727488,
      "step": 3440
    },
    {
      "epoch": 0.378987898789879,
      "grad_norm": 0.025146484375,
      "learning_rate": 0.005683168316831683,
      "loss": 0.2305,
      "num_input_tokens_seen": 728544,
      "step": 3445
    },
    {
      "epoch": 0.3795379537953795,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.005691419141914191,
      "loss": 0.2277,
      "num_input_tokens_seen": 729568,
      "step": 3450
    },
    {
      "epoch": 0.38008800880088006,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.0056996699669967,
      "loss": 0.2261,
      "num_input_tokens_seen": 730528,
      "step": 3455
    },
    {
      "epoch": 0.38063806380638066,
      "grad_norm": 0.051513671875,
      "learning_rate": 0.005707920792079208,
      "loss": 0.221,
      "num_input_tokens_seen": 731616,
      "step": 3460
    },
    {
      "epoch": 0.3811881188118812,
      "grad_norm": 0.036376953125,
      "learning_rate": 0.005716171617161716,
      "loss": 0.2393,
      "num_input_tokens_seen": 732800,
      "step": 3465
    },
    {
      "epoch": 0.38173817381738173,
      "grad_norm": 0.0263671875,
      "learning_rate": 0.005724422442244225,
      "loss": 0.2261,
      "num_input_tokens_seen": 733792,
      "step": 3470
    },
    {
      "epoch": 0.38228822882288227,
      "grad_norm": 0.0244140625,
      "learning_rate": 0.0057326732673267326,
      "loss": 0.2407,
      "num_input_tokens_seen": 734784,
      "step": 3475
    },
    {
      "epoch": 0.38283828382838286,
      "grad_norm": 0.03662109375,
      "learning_rate": 0.0057409240924092405,
      "loss": 0.2389,
      "num_input_tokens_seen": 735968,
      "step": 3480
    },
    {
      "epoch": 0.3833883388338834,
      "grad_norm": 0.02587890625,
      "learning_rate": 0.0057491749174917485,
      "loss": 0.24,
      "num_input_tokens_seen": 736928,
      "step": 3485
    },
    {
      "epoch": 0.38393839383938394,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.005757425742574257,
      "loss": 0.2298,
      "num_input_tokens_seen": 737920,
      "step": 3490
    },
    {
      "epoch": 0.3844884488448845,
      "grad_norm": 0.005859375,
      "learning_rate": 0.005765676567656765,
      "loss": 0.2244,
      "num_input_tokens_seen": 739008,
      "step": 3495
    },
    {
      "epoch": 0.385038503850385,
      "grad_norm": 0.027587890625,
      "learning_rate": 0.005773927392739274,
      "loss": 0.239,
      "num_input_tokens_seen": 740128,
      "step": 3500
    },
    {
      "epoch": 0.3855885588558856,
      "grad_norm": 0.029052734375,
      "learning_rate": 0.005782178217821782,
      "loss": 0.2335,
      "num_input_tokens_seen": 741248,
      "step": 3505
    },
    {
      "epoch": 0.38613861386138615,
      "grad_norm": 0.0281982421875,
      "learning_rate": 0.00579042904290429,
      "loss": 0.2345,
      "num_input_tokens_seen": 742304,
      "step": 3510
    },
    {
      "epoch": 0.3866886688668867,
      "grad_norm": 0.055419921875,
      "learning_rate": 0.005798679867986799,
      "loss": 0.2302,
      "num_input_tokens_seen": 743296,
      "step": 3515
    },
    {
      "epoch": 0.3872387238723872,
      "grad_norm": 0.032958984375,
      "learning_rate": 0.005806930693069307,
      "loss": 0.2377,
      "num_input_tokens_seen": 744288,
      "step": 3520
    },
    {
      "epoch": 0.38778877887788776,
      "grad_norm": 0.053955078125,
      "learning_rate": 0.005815181518151815,
      "loss": 0.2312,
      "num_input_tokens_seen": 745376,
      "step": 3525
    },
    {
      "epoch": 0.38833883388338836,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.005823432343234323,
      "loss": 0.2378,
      "num_input_tokens_seen": 746464,
      "step": 3530
    },
    {
      "epoch": 0.3888888888888889,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.005831683168316831,
      "loss": 0.2325,
      "num_input_tokens_seen": 747520,
      "step": 3535
    },
    {
      "epoch": 0.38943894389438943,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.00583993399339934,
      "loss": 0.235,
      "num_input_tokens_seen": 748544,
      "step": 3540
    },
    {
      "epoch": 0.38998899889988997,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.0058481848184818485,
      "loss": 0.2319,
      "num_input_tokens_seen": 749600,
      "step": 3545
    },
    {
      "epoch": 0.39053905390539057,
      "grad_norm": 0.007781982421875,
      "learning_rate": 0.005856435643564356,
      "loss": 0.237,
      "num_input_tokens_seen": 750688,
      "step": 3550
    },
    {
      "epoch": 0.3910891089108911,
      "grad_norm": 0.0303955078125,
      "learning_rate": 0.005864686468646864,
      "loss": 0.2339,
      "num_input_tokens_seen": 751712,
      "step": 3555
    },
    {
      "epoch": 0.39163916391639164,
      "grad_norm": 0.046630859375,
      "learning_rate": 0.005872937293729373,
      "loss": 0.2275,
      "num_input_tokens_seen": 752800,
      "step": 3560
    },
    {
      "epoch": 0.3921892189218922,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.005881188118811881,
      "loss": 0.2308,
      "num_input_tokens_seen": 753856,
      "step": 3565
    },
    {
      "epoch": 0.3927392739273927,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.005889438943894389,
      "loss": 0.236,
      "num_input_tokens_seen": 754848,
      "step": 3570
    },
    {
      "epoch": 0.3932893289328933,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.005897689768976897,
      "loss": 0.2287,
      "num_input_tokens_seen": 755872,
      "step": 3575
    },
    {
      "epoch": 0.39383938393839385,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.005905940594059405,
      "loss": 0.2316,
      "num_input_tokens_seen": 756896,
      "step": 3580
    },
    {
      "epoch": 0.3943894389438944,
      "grad_norm": 0.050537109375,
      "learning_rate": 0.005914191419141914,
      "loss": 0.2335,
      "num_input_tokens_seen": 758016,
      "step": 3585
    },
    {
      "epoch": 0.3949394939493949,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.005922442244224423,
      "loss": 0.2323,
      "num_input_tokens_seen": 759008,
      "step": 3590
    },
    {
      "epoch": 0.39548954895489546,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.005930693069306931,
      "loss": 0.2304,
      "num_input_tokens_seen": 760032,
      "step": 3595
    },
    {
      "epoch": 0.39603960396039606,
      "grad_norm": 0.0220947265625,
      "learning_rate": 0.005938943894389439,
      "loss": 0.2296,
      "num_input_tokens_seen": 761152,
      "step": 3600
    },
    {
      "epoch": 0.3965896589658966,
      "grad_norm": 0.009765625,
      "learning_rate": 0.005947194719471947,
      "loss": 0.2338,
      "num_input_tokens_seen": 762144,
      "step": 3605
    },
    {
      "epoch": 0.39713971397139713,
      "grad_norm": 0.02685546875,
      "learning_rate": 0.0059554455445544555,
      "loss": 0.2351,
      "num_input_tokens_seen": 763232,
      "step": 3610
    },
    {
      "epoch": 0.3976897689768977,
      "grad_norm": 0.05029296875,
      "learning_rate": 0.0059636963696369635,
      "loss": 0.2296,
      "num_input_tokens_seen": 764256,
      "step": 3615
    },
    {
      "epoch": 0.39823982398239827,
      "grad_norm": 0.0137939453125,
      "learning_rate": 0.0059719471947194715,
      "loss": 0.2287,
      "num_input_tokens_seen": 765280,
      "step": 3620
    },
    {
      "epoch": 0.3987898789878988,
      "grad_norm": 0.022216796875,
      "learning_rate": 0.0059801980198019794,
      "loss": 0.2359,
      "num_input_tokens_seen": 766368,
      "step": 3625
    },
    {
      "epoch": 0.39933993399339934,
      "grad_norm": 0.045654296875,
      "learning_rate": 0.005988448844884488,
      "loss": 0.2293,
      "num_input_tokens_seen": 767424,
      "step": 3630
    },
    {
      "epoch": 0.3998899889988999,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.005996699669966997,
      "loss": 0.2304,
      "num_input_tokens_seen": 768544,
      "step": 3635
    },
    {
      "epoch": 0.4004400440044004,
      "grad_norm": 0.047119140625,
      "learning_rate": 0.006004950495049505,
      "loss": 0.2326,
      "num_input_tokens_seen": 769632,
      "step": 3640
    },
    {
      "epoch": 0.400990099009901,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.006013201320132013,
      "loss": 0.2357,
      "num_input_tokens_seen": 770784,
      "step": 3645
    },
    {
      "epoch": 0.40154015401540155,
      "grad_norm": 0.02294921875,
      "learning_rate": 0.006021452145214521,
      "loss": 0.2284,
      "num_input_tokens_seen": 771840,
      "step": 3650
    },
    {
      "epoch": 0.4020902090209021,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.00602970297029703,
      "loss": 0.2327,
      "num_input_tokens_seen": 772928,
      "step": 3655
    },
    {
      "epoch": 0.40264026402640263,
      "grad_norm": 0.046875,
      "learning_rate": 0.006037953795379538,
      "loss": 0.2337,
      "num_input_tokens_seen": 774048,
      "step": 3660
    },
    {
      "epoch": 0.40319031903190317,
      "grad_norm": 0.02392578125,
      "learning_rate": 0.006046204620462046,
      "loss": 0.2283,
      "num_input_tokens_seen": 775104,
      "step": 3665
    },
    {
      "epoch": 0.40374037403740376,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.006054455445544554,
      "loss": 0.2253,
      "num_input_tokens_seen": 776160,
      "step": 3670
    },
    {
      "epoch": 0.4042904290429043,
      "grad_norm": 0.01544189453125,
      "learning_rate": 0.006062706270627062,
      "loss": 0.2314,
      "num_input_tokens_seen": 777216,
      "step": 3675
    },
    {
      "epoch": 0.40484048404840484,
      "grad_norm": 0.0216064453125,
      "learning_rate": 0.0060709570957095714,
      "loss": 0.2294,
      "num_input_tokens_seen": 778272,
      "step": 3680
    },
    {
      "epoch": 0.4053905390539054,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.006079207920792079,
      "loss": 0.2285,
      "num_input_tokens_seen": 779328,
      "step": 3685
    },
    {
      "epoch": 0.40594059405940597,
      "grad_norm": 0.0078125,
      "learning_rate": 0.006087458745874587,
      "loss": 0.234,
      "num_input_tokens_seen": 780480,
      "step": 3690
    },
    {
      "epoch": 0.4064906490649065,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.006095709570957095,
      "loss": 0.2371,
      "num_input_tokens_seen": 781632,
      "step": 3695
    },
    {
      "epoch": 0.40704070407040704,
      "grad_norm": 0.046142578125,
      "learning_rate": 0.006103960396039603,
      "loss": 0.237,
      "num_input_tokens_seen": 782720,
      "step": 3700
    },
    {
      "epoch": 0.4075907590759076,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.006112211221122112,
      "loss": 0.2349,
      "num_input_tokens_seen": 783712,
      "step": 3705
    },
    {
      "epoch": 0.4081408140814081,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.00612046204620462,
      "loss": 0.2285,
      "num_input_tokens_seen": 784768,
      "step": 3710
    },
    {
      "epoch": 0.4086908690869087,
      "grad_norm": 0.0238037109375,
      "learning_rate": 0.006128712871287128,
      "loss": 0.2314,
      "num_input_tokens_seen": 785888,
      "step": 3715
    },
    {
      "epoch": 0.40924092409240925,
      "grad_norm": 0.0244140625,
      "learning_rate": 0.006136963696369636,
      "loss": 0.2336,
      "num_input_tokens_seen": 786944,
      "step": 3720
    },
    {
      "epoch": 0.4097909790979098,
      "grad_norm": 0.0245361328125,
      "learning_rate": 0.006145214521452146,
      "loss": 0.2293,
      "num_input_tokens_seen": 788064,
      "step": 3725
    },
    {
      "epoch": 0.41034103410341033,
      "grad_norm": 0.02099609375,
      "learning_rate": 0.006153465346534654,
      "loss": 0.2284,
      "num_input_tokens_seen": 789088,
      "step": 3730
    },
    {
      "epoch": 0.41089108910891087,
      "grad_norm": 0.0206298828125,
      "learning_rate": 0.006161716171617162,
      "loss": 0.2285,
      "num_input_tokens_seen": 790144,
      "step": 3735
    },
    {
      "epoch": 0.41144114411441146,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00616996699669967,
      "loss": 0.2352,
      "num_input_tokens_seen": 791136,
      "step": 3740
    },
    {
      "epoch": 0.411991199119912,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.006178217821782178,
      "loss": 0.2261,
      "num_input_tokens_seen": 792160,
      "step": 3745
    },
    {
      "epoch": 0.41254125412541254,
      "grad_norm": 0.01416015625,
      "learning_rate": 0.0061864686468646865,
      "loss": 0.2254,
      "num_input_tokens_seen": 793216,
      "step": 3750
    },
    {
      "epoch": 0.4130913091309131,
      "grad_norm": 0.0255126953125,
      "learning_rate": 0.0061947194719471944,
      "loss": 0.2392,
      "num_input_tokens_seen": 794272,
      "step": 3755
    },
    {
      "epoch": 0.4136413641364136,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.006202970297029702,
      "loss": 0.2415,
      "num_input_tokens_seen": 795328,
      "step": 3760
    },
    {
      "epoch": 0.4141914191419142,
      "grad_norm": 0.0201416015625,
      "learning_rate": 0.00621122112211221,
      "loss": 0.2322,
      "num_input_tokens_seen": 796352,
      "step": 3765
    },
    {
      "epoch": 0.41474147414741475,
      "grad_norm": 0.0208740234375,
      "learning_rate": 0.006219471947194719,
      "loss": 0.231,
      "num_input_tokens_seen": 797376,
      "step": 3770
    },
    {
      "epoch": 0.4152915291529153,
      "grad_norm": 0.04150390625,
      "learning_rate": 0.006227722772277228,
      "loss": 0.232,
      "num_input_tokens_seen": 798432,
      "step": 3775
    },
    {
      "epoch": 0.4158415841584158,
      "grad_norm": 0.0223388671875,
      "learning_rate": 0.006235973597359736,
      "loss": 0.2329,
      "num_input_tokens_seen": 799552,
      "step": 3780
    },
    {
      "epoch": 0.4163916391639164,
      "grad_norm": 0.021240234375,
      "learning_rate": 0.006244224422442244,
      "loss": 0.2308,
      "num_input_tokens_seen": 800576,
      "step": 3785
    },
    {
      "epoch": 0.41694169416941695,
      "grad_norm": 0.0255126953125,
      "learning_rate": 0.006252475247524752,
      "loss": 0.2297,
      "num_input_tokens_seen": 801664,
      "step": 3790
    },
    {
      "epoch": 0.4174917491749175,
      "grad_norm": 0.021240234375,
      "learning_rate": 0.006260726072607261,
      "loss": 0.2289,
      "num_input_tokens_seen": 802624,
      "step": 3795
    },
    {
      "epoch": 0.41804180418041803,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006268976897689769,
      "loss": 0.2322,
      "num_input_tokens_seen": 803680,
      "step": 3800
    },
    {
      "epoch": 0.41859185918591857,
      "grad_norm": 0.042724609375,
      "learning_rate": 0.006277227722772277,
      "loss": 0.2257,
      "num_input_tokens_seen": 804768,
      "step": 3805
    },
    {
      "epoch": 0.41914191419141916,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.006285478547854785,
      "loss": 0.2333,
      "num_input_tokens_seen": 805760,
      "step": 3810
    },
    {
      "epoch": 0.4196919691969197,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.0062937293729372935,
      "loss": 0.2233,
      "num_input_tokens_seen": 806816,
      "step": 3815
    },
    {
      "epoch": 0.42024202420242024,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.006301980198019802,
      "loss": 0.2431,
      "num_input_tokens_seen": 807840,
      "step": 3820
    },
    {
      "epoch": 0.4207920792079208,
      "grad_norm": 0.0081787109375,
      "learning_rate": 0.00631023102310231,
      "loss": 0.2369,
      "num_input_tokens_seen": 808864,
      "step": 3825
    },
    {
      "epoch": 0.4213421342134213,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.006318481848184818,
      "loss": 0.2352,
      "num_input_tokens_seen": 809920,
      "step": 3830
    },
    {
      "epoch": 0.4218921892189219,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.006326732673267326,
      "loss": 0.2275,
      "num_input_tokens_seen": 811008,
      "step": 3835
    },
    {
      "epoch": 0.42244224422442245,
      "grad_norm": 0.0238037109375,
      "learning_rate": 0.006334983498349834,
      "loss": 0.2295,
      "num_input_tokens_seen": 812000,
      "step": 3840
    },
    {
      "epoch": 0.422992299229923,
      "grad_norm": 0.0267333984375,
      "learning_rate": 0.006343234323432343,
      "loss": 0.238,
      "num_input_tokens_seen": 813088,
      "step": 3845
    },
    {
      "epoch": 0.4235423542354235,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.006351485148514851,
      "loss": 0.2316,
      "num_input_tokens_seen": 814080,
      "step": 3850
    },
    {
      "epoch": 0.4240924092409241,
      "grad_norm": 0.02099609375,
      "learning_rate": 0.006359735973597359,
      "loss": 0.2314,
      "num_input_tokens_seen": 815104,
      "step": 3855
    },
    {
      "epoch": 0.42464246424642466,
      "grad_norm": 0.02099609375,
      "learning_rate": 0.006367986798679868,
      "loss": 0.2356,
      "num_input_tokens_seen": 816096,
      "step": 3860
    },
    {
      "epoch": 0.4251925192519252,
      "grad_norm": 0.022705078125,
      "learning_rate": 0.006376237623762377,
      "loss": 0.2336,
      "num_input_tokens_seen": 817120,
      "step": 3865
    },
    {
      "epoch": 0.42574257425742573,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.006384488448844885,
      "loss": 0.2326,
      "num_input_tokens_seen": 818176,
      "step": 3870
    },
    {
      "epoch": 0.42629262926292627,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.006392739273927393,
      "loss": 0.2316,
      "num_input_tokens_seen": 819168,
      "step": 3875
    },
    {
      "epoch": 0.42684268426842686,
      "grad_norm": 0.043212890625,
      "learning_rate": 0.006400990099009901,
      "loss": 0.2325,
      "num_input_tokens_seen": 820256,
      "step": 3880
    },
    {
      "epoch": 0.4273927392739274,
      "grad_norm": 0.0208740234375,
      "learning_rate": 0.006409240924092409,
      "loss": 0.2337,
      "num_input_tokens_seen": 821312,
      "step": 3885
    },
    {
      "epoch": 0.42794279427942794,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.006417491749174917,
      "loss": 0.2306,
      "num_input_tokens_seen": 822336,
      "step": 3890
    },
    {
      "epoch": 0.4284928492849285,
      "grad_norm": 0.0223388671875,
      "learning_rate": 0.006425742574257425,
      "loss": 0.2244,
      "num_input_tokens_seen": 823360,
      "step": 3895
    },
    {
      "epoch": 0.429042904290429,
      "grad_norm": 0.0220947265625,
      "learning_rate": 0.006433993399339933,
      "loss": 0.2349,
      "num_input_tokens_seen": 824416,
      "step": 3900
    },
    {
      "epoch": 0.4295929592959296,
      "grad_norm": 0.003814697265625,
      "learning_rate": 0.006442244224422442,
      "loss": 0.2267,
      "num_input_tokens_seen": 825408,
      "step": 3905
    },
    {
      "epoch": 0.43014301430143015,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.00645049504950495,
      "loss": 0.2307,
      "num_input_tokens_seen": 826464,
      "step": 3910
    },
    {
      "epoch": 0.4306930693069307,
      "grad_norm": 0.045654296875,
      "learning_rate": 0.006458745874587459,
      "loss": 0.2361,
      "num_input_tokens_seen": 827584,
      "step": 3915
    },
    {
      "epoch": 0.4312431243124312,
      "grad_norm": 0.02490234375,
      "learning_rate": 0.006466996699669967,
      "loss": 0.2378,
      "num_input_tokens_seen": 828640,
      "step": 3920
    },
    {
      "epoch": 0.4317931793179318,
      "grad_norm": 0.02197265625,
      "learning_rate": 0.006475247524752475,
      "loss": 0.2298,
      "num_input_tokens_seen": 829728,
      "step": 3925
    },
    {
      "epoch": 0.43234323432343236,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.006483498349834983,
      "loss": 0.2307,
      "num_input_tokens_seen": 830752,
      "step": 3930
    },
    {
      "epoch": 0.4328932893289329,
      "grad_norm": 0.041748046875,
      "learning_rate": 0.006491749174917492,
      "loss": 0.2223,
      "num_input_tokens_seen": 831808,
      "step": 3935
    },
    {
      "epoch": 0.43344334433443343,
      "grad_norm": 0.055419921875,
      "learning_rate": 0.0065,
      "loss": 0.2367,
      "num_input_tokens_seen": 832928,
      "step": 3940
    },
    {
      "epoch": 0.43399339933993397,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.006508250825082508,
      "loss": 0.2161,
      "num_input_tokens_seen": 833984,
      "step": 3945
    },
    {
      "epoch": 0.43454345434543457,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.0065165016501650165,
      "loss": 0.2312,
      "num_input_tokens_seen": 835008,
      "step": 3950
    },
    {
      "epoch": 0.4350935093509351,
      "grad_norm": 0.0576171875,
      "learning_rate": 0.0065247524752475245,
      "loss": 0.2557,
      "num_input_tokens_seen": 836032,
      "step": 3955
    },
    {
      "epoch": 0.43564356435643564,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.006533003300330033,
      "loss": 0.2385,
      "num_input_tokens_seen": 837056,
      "step": 3960
    },
    {
      "epoch": 0.4361936193619362,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.006541254125412541,
      "loss": 0.2202,
      "num_input_tokens_seen": 838112,
      "step": 3965
    },
    {
      "epoch": 0.4367436743674367,
      "grad_norm": 0.0247802734375,
      "learning_rate": 0.006549504950495049,
      "loss": 0.2325,
      "num_input_tokens_seen": 839104,
      "step": 3970
    },
    {
      "epoch": 0.4372937293729373,
      "grad_norm": 0.026611328125,
      "learning_rate": 0.006557755775577557,
      "loss": 0.2329,
      "num_input_tokens_seen": 840192,
      "step": 3975
    },
    {
      "epoch": 0.43784378437843785,
      "grad_norm": 0.0390625,
      "learning_rate": 0.006566006600660065,
      "loss": 0.2195,
      "num_input_tokens_seen": 841184,
      "step": 3980
    },
    {
      "epoch": 0.4383938393839384,
      "grad_norm": 0.0242919921875,
      "learning_rate": 0.006574257425742574,
      "loss": 0.2364,
      "num_input_tokens_seen": 842176,
      "step": 3985
    },
    {
      "epoch": 0.4389438943894389,
      "grad_norm": 0.04248046875,
      "learning_rate": 0.006582508250825082,
      "loss": 0.2364,
      "num_input_tokens_seen": 843232,
      "step": 3990
    },
    {
      "epoch": 0.4394939493949395,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.006590759075907591,
      "loss": 0.2402,
      "num_input_tokens_seen": 844288,
      "step": 3995
    },
    {
      "epoch": 0.44004400440044006,
      "grad_norm": 0.0194091796875,
      "learning_rate": 0.006599009900990099,
      "loss": 0.2358,
      "num_input_tokens_seen": 845408,
      "step": 4000
    },
    {
      "epoch": 0.4405940594059406,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.006607260726072608,
      "loss": 0.2314,
      "num_input_tokens_seen": 846400,
      "step": 4005
    },
    {
      "epoch": 0.44114411441144114,
      "grad_norm": 0.041259765625,
      "learning_rate": 0.006615511551155116,
      "loss": 0.2275,
      "num_input_tokens_seen": 847392,
      "step": 4010
    },
    {
      "epoch": 0.4416941694169417,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.006623762376237624,
      "loss": 0.2271,
      "num_input_tokens_seen": 848448,
      "step": 4015
    },
    {
      "epoch": 0.44224422442244227,
      "grad_norm": 0.12158203125,
      "learning_rate": 0.0066320132013201315,
      "loss": 0.2475,
      "num_input_tokens_seen": 849504,
      "step": 4020
    },
    {
      "epoch": 0.4427942794279428,
      "grad_norm": 0.2001953125,
      "learning_rate": 0.0066402640264026395,
      "loss": 0.2368,
      "num_input_tokens_seen": 850496,
      "step": 4025
    },
    {
      "epoch": 0.44334433443344334,
      "grad_norm": 0.08154296875,
      "learning_rate": 0.006648514851485148,
      "loss": 0.2278,
      "num_input_tokens_seen": 851520,
      "step": 4030
    },
    {
      "epoch": 0.4438943894389439,
      "grad_norm": 0.0458984375,
      "learning_rate": 0.006656765676567656,
      "loss": 0.2384,
      "num_input_tokens_seen": 852544,
      "step": 4035
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 0.01513671875,
      "learning_rate": 0.006665016501650165,
      "loss": 0.2282,
      "num_input_tokens_seen": 853600,
      "step": 4040
    },
    {
      "epoch": 0.444994499449945,
      "grad_norm": 0.0419921875,
      "learning_rate": 0.006673267326732673,
      "loss": 0.2406,
      "num_input_tokens_seen": 854656,
      "step": 4045
    },
    {
      "epoch": 0.44554455445544555,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.006681518151815181,
      "loss": 0.238,
      "num_input_tokens_seen": 855712,
      "step": 4050
    },
    {
      "epoch": 0.4460946094609461,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.00668976897689769,
      "loss": 0.2318,
      "num_input_tokens_seen": 856704,
      "step": 4055
    },
    {
      "epoch": 0.44664466446644663,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.006698019801980198,
      "loss": 0.2325,
      "num_input_tokens_seen": 857760,
      "step": 4060
    },
    {
      "epoch": 0.4471947194719472,
      "grad_norm": 0.0284423828125,
      "learning_rate": 0.006706270627062706,
      "loss": 0.2314,
      "num_input_tokens_seen": 858848,
      "step": 4065
    },
    {
      "epoch": 0.44774477447744776,
      "grad_norm": 0.052978515625,
      "learning_rate": 0.006714521452145214,
      "loss": 0.2284,
      "num_input_tokens_seen": 859872,
      "step": 4070
    },
    {
      "epoch": 0.4482948294829483,
      "grad_norm": 0.025146484375,
      "learning_rate": 0.006722772277227722,
      "loss": 0.2348,
      "num_input_tokens_seen": 860896,
      "step": 4075
    },
    {
      "epoch": 0.44884488448844884,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.006731023102310231,
      "loss": 0.2243,
      "num_input_tokens_seen": 861984,
      "step": 4080
    },
    {
      "epoch": 0.4493949394939494,
      "grad_norm": 0.016845703125,
      "learning_rate": 0.0067392739273927395,
      "loss": 0.2312,
      "num_input_tokens_seen": 862944,
      "step": 4085
    },
    {
      "epoch": 0.44994499449944997,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0067475247524752474,
      "loss": 0.237,
      "num_input_tokens_seen": 863968,
      "step": 4090
    },
    {
      "epoch": 0.4504950495049505,
      "grad_norm": 0.023193359375,
      "learning_rate": 0.006755775577557755,
      "loss": 0.2387,
      "num_input_tokens_seen": 865120,
      "step": 4095
    },
    {
      "epoch": 0.45104510451045104,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.006764026402640264,
      "loss": 0.239,
      "num_input_tokens_seen": 866112,
      "step": 4100
    },
    {
      "epoch": 0.4515951595159516,
      "grad_norm": 0.01544189453125,
      "learning_rate": 0.006772277227722772,
      "loss": 0.234,
      "num_input_tokens_seen": 867136,
      "step": 4105
    },
    {
      "epoch": 0.4521452145214521,
      "grad_norm": 0.0081787109375,
      "learning_rate": 0.00678052805280528,
      "loss": 0.2358,
      "num_input_tokens_seen": 868160,
      "step": 4110
    },
    {
      "epoch": 0.4526952695269527,
      "grad_norm": 0.048828125,
      "learning_rate": 0.006788778877887788,
      "loss": 0.2314,
      "num_input_tokens_seen": 869280,
      "step": 4115
    },
    {
      "epoch": 0.45324532453245325,
      "grad_norm": 0.0238037109375,
      "learning_rate": 0.006797029702970296,
      "loss": 0.2263,
      "num_input_tokens_seen": 870368,
      "step": 4120
    },
    {
      "epoch": 0.4537953795379538,
      "grad_norm": 0.04345703125,
      "learning_rate": 0.006805280528052805,
      "loss": 0.2285,
      "num_input_tokens_seen": 871360,
      "step": 4125
    },
    {
      "epoch": 0.45434543454345433,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.006813531353135314,
      "loss": 0.2225,
      "num_input_tokens_seen": 872448,
      "step": 4130
    },
    {
      "epoch": 0.45489548954895487,
      "grad_norm": 0.0279541015625,
      "learning_rate": 0.006821782178217822,
      "loss": 0.2352,
      "num_input_tokens_seen": 873536,
      "step": 4135
    },
    {
      "epoch": 0.45544554455445546,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.00683003300330033,
      "loss": 0.2459,
      "num_input_tokens_seen": 874592,
      "step": 4140
    },
    {
      "epoch": 0.455995599559956,
      "grad_norm": 0.0478515625,
      "learning_rate": 0.006838283828382838,
      "loss": 0.2379,
      "num_input_tokens_seen": 875584,
      "step": 4145
    },
    {
      "epoch": 0.45654565456545654,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.0068465346534653465,
      "loss": 0.2372,
      "num_input_tokens_seen": 876640,
      "step": 4150
    },
    {
      "epoch": 0.4570957095709571,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.0068547854785478545,
      "loss": 0.2327,
      "num_input_tokens_seen": 877728,
      "step": 4155
    },
    {
      "epoch": 0.45764576457645767,
      "grad_norm": 0.0311279296875,
      "learning_rate": 0.0068630363036303625,
      "loss": 0.2326,
      "num_input_tokens_seen": 878752,
      "step": 4160
    },
    {
      "epoch": 0.4581958195819582,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0068712871287128705,
      "loss": 0.2316,
      "num_input_tokens_seen": 879744,
      "step": 4165
    },
    {
      "epoch": 0.45874587458745875,
      "grad_norm": 0.0240478515625,
      "learning_rate": 0.006879537953795379,
      "loss": 0.2315,
      "num_input_tokens_seen": 880800,
      "step": 4170
    },
    {
      "epoch": 0.4592959295929593,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.006887788778877888,
      "loss": 0.2283,
      "num_input_tokens_seen": 881824,
      "step": 4175
    },
    {
      "epoch": 0.4598459845984598,
      "grad_norm": 0.04052734375,
      "learning_rate": 0.006896039603960396,
      "loss": 0.2315,
      "num_input_tokens_seen": 882848,
      "step": 4180
    },
    {
      "epoch": 0.4603960396039604,
      "grad_norm": 0.019775390625,
      "learning_rate": 0.006904290429042904,
      "loss": 0.2303,
      "num_input_tokens_seen": 883808,
      "step": 4185
    },
    {
      "epoch": 0.46094609460946095,
      "grad_norm": 0.038330078125,
      "learning_rate": 0.006912541254125412,
      "loss": 0.2346,
      "num_input_tokens_seen": 884832,
      "step": 4190
    },
    {
      "epoch": 0.4614961496149615,
      "grad_norm": 0.0213623046875,
      "learning_rate": 0.006920792079207921,
      "loss": 0.2313,
      "num_input_tokens_seen": 885856,
      "step": 4195
    },
    {
      "epoch": 0.46204620462046203,
      "grad_norm": 0.038818359375,
      "learning_rate": 0.006929042904290429,
      "loss": 0.2306,
      "num_input_tokens_seen": 886880,
      "step": 4200
    },
    {
      "epoch": 0.46259625962596257,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.006937293729372937,
      "loss": 0.2369,
      "num_input_tokens_seen": 887936,
      "step": 4205
    },
    {
      "epoch": 0.46314631463146316,
      "grad_norm": 0.01953125,
      "learning_rate": 0.006945544554455445,
      "loss": 0.2297,
      "num_input_tokens_seen": 888992,
      "step": 4210
    },
    {
      "epoch": 0.4636963696369637,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.006953795379537953,
      "loss": 0.2349,
      "num_input_tokens_seen": 890048,
      "step": 4215
    },
    {
      "epoch": 0.46424642464246424,
      "grad_norm": 0.0213623046875,
      "learning_rate": 0.0069620462046204624,
      "loss": 0.2328,
      "num_input_tokens_seen": 891104,
      "step": 4220
    },
    {
      "epoch": 0.4647964796479648,
      "grad_norm": 0.025634765625,
      "learning_rate": 0.00697029702970297,
      "loss": 0.2276,
      "num_input_tokens_seen": 892192,
      "step": 4225
    },
    {
      "epoch": 0.46534653465346537,
      "grad_norm": 0.0223388671875,
      "learning_rate": 0.006978547854785478,
      "loss": 0.2349,
      "num_input_tokens_seen": 893216,
      "step": 4230
    },
    {
      "epoch": 0.4658965896589659,
      "grad_norm": 0.0194091796875,
      "learning_rate": 0.006986798679867986,
      "loss": 0.2295,
      "num_input_tokens_seen": 894272,
      "step": 4235
    },
    {
      "epoch": 0.46644664466446645,
      "grad_norm": 0.022216796875,
      "learning_rate": 0.006995049504950495,
      "loss": 0.2316,
      "num_input_tokens_seen": 895360,
      "step": 4240
    },
    {
      "epoch": 0.466996699669967,
      "grad_norm": 0.0196533203125,
      "learning_rate": 0.007003300330033003,
      "loss": 0.2325,
      "num_input_tokens_seen": 896416,
      "step": 4245
    },
    {
      "epoch": 0.4675467546754675,
      "grad_norm": 0.020263671875,
      "learning_rate": 0.007011551155115511,
      "loss": 0.2315,
      "num_input_tokens_seen": 897440,
      "step": 4250
    },
    {
      "epoch": 0.4680968096809681,
      "grad_norm": 0.0390625,
      "learning_rate": 0.007019801980198019,
      "loss": 0.2348,
      "num_input_tokens_seen": 898528,
      "step": 4255
    },
    {
      "epoch": 0.46864686468646866,
      "grad_norm": 0.018310546875,
      "learning_rate": 0.007028052805280527,
      "loss": 0.2323,
      "num_input_tokens_seen": 899552,
      "step": 4260
    },
    {
      "epoch": 0.4691969196919692,
      "grad_norm": 0.0189208984375,
      "learning_rate": 0.007036303630363037,
      "loss": 0.2306,
      "num_input_tokens_seen": 900544,
      "step": 4265
    },
    {
      "epoch": 0.46974697469746973,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.007044554455445545,
      "loss": 0.2243,
      "num_input_tokens_seen": 901632,
      "step": 4270
    },
    {
      "epoch": 0.47029702970297027,
      "grad_norm": 0.0419921875,
      "learning_rate": 0.007052805280528053,
      "loss": 0.232,
      "num_input_tokens_seen": 902688,
      "step": 4275
    },
    {
      "epoch": 0.47084708470847086,
      "grad_norm": 0.0233154296875,
      "learning_rate": 0.007061056105610561,
      "loss": 0.2406,
      "num_input_tokens_seen": 903744,
      "step": 4280
    },
    {
      "epoch": 0.4713971397139714,
      "grad_norm": 0.036376953125,
      "learning_rate": 0.007069306930693069,
      "loss": 0.2394,
      "num_input_tokens_seen": 904800,
      "step": 4285
    },
    {
      "epoch": 0.47194719471947194,
      "grad_norm": 0.017333984375,
      "learning_rate": 0.0070775577557755775,
      "loss": 0.236,
      "num_input_tokens_seen": 905824,
      "step": 4290
    },
    {
      "epoch": 0.4724972497249725,
      "grad_norm": 0.00750732421875,
      "learning_rate": 0.0070858085808580855,
      "loss": 0.2368,
      "num_input_tokens_seen": 906848,
      "step": 4295
    },
    {
      "epoch": 0.4730473047304731,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.007094059405940593,
      "loss": 0.2357,
      "num_input_tokens_seen": 907904,
      "step": 4300
    },
    {
      "epoch": 0.4735973597359736,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.007102310231023101,
      "loss": 0.2346,
      "num_input_tokens_seen": 909024,
      "step": 4305
    },
    {
      "epoch": 0.47414741474147415,
      "grad_norm": 0.00726318359375,
      "learning_rate": 0.007110561056105611,
      "loss": 0.2358,
      "num_input_tokens_seen": 910176,
      "step": 4310
    },
    {
      "epoch": 0.4746974697469747,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.007118811881188119,
      "loss": 0.2327,
      "num_input_tokens_seen": 911296,
      "step": 4315
    },
    {
      "epoch": 0.4752475247524752,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.007127062706270627,
      "loss": 0.2357,
      "num_input_tokens_seen": 912352,
      "step": 4320
    },
    {
      "epoch": 0.4757975797579758,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.007135313531353135,
      "loss": 0.2318,
      "num_input_tokens_seen": 913408,
      "step": 4325
    },
    {
      "epoch": 0.47634763476347636,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.007143564356435643,
      "loss": 0.2295,
      "num_input_tokens_seen": 914464,
      "step": 4330
    },
    {
      "epoch": 0.4768976897689769,
      "grad_norm": 0.021240234375,
      "learning_rate": 0.007151815181518152,
      "loss": 0.2307,
      "num_input_tokens_seen": 915552,
      "step": 4335
    },
    {
      "epoch": 0.47744774477447743,
      "grad_norm": 0.0341796875,
      "learning_rate": 0.00716006600660066,
      "loss": 0.2326,
      "num_input_tokens_seen": 916640,
      "step": 4340
    },
    {
      "epoch": 0.47799779977997797,
      "grad_norm": 0.0045166015625,
      "learning_rate": 0.007168316831683168,
      "loss": 0.2317,
      "num_input_tokens_seen": 917696,
      "step": 4345
    },
    {
      "epoch": 0.47854785478547857,
      "grad_norm": 0.0205078125,
      "learning_rate": 0.007176567656765676,
      "loss": 0.2308,
      "num_input_tokens_seen": 918784,
      "step": 4350
    },
    {
      "epoch": 0.4790979097909791,
      "grad_norm": 0.0037994384765625,
      "learning_rate": 0.0071848184818481846,
      "loss": 0.2316,
      "num_input_tokens_seen": 919776,
      "step": 4355
    },
    {
      "epoch": 0.47964796479647964,
      "grad_norm": 0.032958984375,
      "learning_rate": 0.007193069306930693,
      "loss": 0.2327,
      "num_input_tokens_seen": 920896,
      "step": 4360
    },
    {
      "epoch": 0.4801980198019802,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.007201320132013201,
      "loss": 0.2297,
      "num_input_tokens_seen": 921856,
      "step": 4365
    },
    {
      "epoch": 0.4807480748074808,
      "grad_norm": 0.017578125,
      "learning_rate": 0.007209570957095709,
      "loss": 0.2295,
      "num_input_tokens_seen": 922912,
      "step": 4370
    },
    {
      "epoch": 0.4812981298129813,
      "grad_norm": 0.01904296875,
      "learning_rate": 0.007217821782178217,
      "loss": 0.2348,
      "num_input_tokens_seen": 923968,
      "step": 4375
    },
    {
      "epoch": 0.48184818481848185,
      "grad_norm": 0.0029144287109375,
      "learning_rate": 0.007226072607260726,
      "loss": 0.2316,
      "num_input_tokens_seen": 925024,
      "step": 4380
    },
    {
      "epoch": 0.4823982398239824,
      "grad_norm": 0.0191650390625,
      "learning_rate": 0.007234323432343234,
      "loss": 0.2337,
      "num_input_tokens_seen": 926080,
      "step": 4385
    },
    {
      "epoch": 0.4829482948294829,
      "grad_norm": 0.01806640625,
      "learning_rate": 0.007242574257425742,
      "loss": 0.2317,
      "num_input_tokens_seen": 927104,
      "step": 4390
    },
    {
      "epoch": 0.4834983498349835,
      "grad_norm": 0.020751953125,
      "learning_rate": 0.00725082508250825,
      "loss": 0.2346,
      "num_input_tokens_seen": 928128,
      "step": 4395
    },
    {
      "epoch": 0.48404840484048406,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.007259075907590759,
      "loss": 0.2306,
      "num_input_tokens_seen": 929184,
      "step": 4400
    },
    {
      "epoch": 0.4845984598459846,
      "grad_norm": 0.019775390625,
      "learning_rate": 0.007267326732673268,
      "loss": 0.2347,
      "num_input_tokens_seen": 930176,
      "step": 4405
    },
    {
      "epoch": 0.48514851485148514,
      "grad_norm": 0.018310546875,
      "learning_rate": 0.007275577557755776,
      "loss": 0.2326,
      "num_input_tokens_seen": 931136,
      "step": 4410
    },
    {
      "epoch": 0.4856985698569857,
      "grad_norm": 0.019775390625,
      "learning_rate": 0.007283828382838284,
      "loss": 0.2304,
      "num_input_tokens_seen": 932128,
      "step": 4415
    },
    {
      "epoch": 0.48624862486248627,
      "grad_norm": 0.01806640625,
      "learning_rate": 0.007292079207920792,
      "loss": 0.2303,
      "num_input_tokens_seen": 933184,
      "step": 4420
    },
    {
      "epoch": 0.4867986798679868,
      "grad_norm": 0.034423828125,
      "learning_rate": 0.0073003300330033,
      "loss": 0.2273,
      "num_input_tokens_seen": 934240,
      "step": 4425
    },
    {
      "epoch": 0.48734873487348734,
      "grad_norm": 0.01324462890625,
      "learning_rate": 0.007308580858085808,
      "loss": 0.2345,
      "num_input_tokens_seen": 935328,
      "step": 4430
    },
    {
      "epoch": 0.4878987898789879,
      "grad_norm": 0.0186767578125,
      "learning_rate": 0.007316831683168316,
      "loss": 0.2325,
      "num_input_tokens_seen": 936384,
      "step": 4435
    },
    {
      "epoch": 0.4884488448844885,
      "grad_norm": 0.021240234375,
      "learning_rate": 0.007325082508250824,
      "loss": 0.2335,
      "num_input_tokens_seen": 937408,
      "step": 4440
    },
    {
      "epoch": 0.488998899889989,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.007333333333333333,
      "loss": 0.2294,
      "num_input_tokens_seen": 938432,
      "step": 4445
    },
    {
      "epoch": 0.48954895489548955,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.007341584158415842,
      "loss": 0.2304,
      "num_input_tokens_seen": 939520,
      "step": 4450
    },
    {
      "epoch": 0.4900990099009901,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.00734983498349835,
      "loss": 0.2325,
      "num_input_tokens_seen": 940544,
      "step": 4455
    },
    {
      "epoch": 0.49064906490649063,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.007358085808580858,
      "loss": 0.2348,
      "num_input_tokens_seen": 941504,
      "step": 4460
    },
    {
      "epoch": 0.4911991199119912,
      "grad_norm": 0.03662109375,
      "learning_rate": 0.007366336633663366,
      "loss": 0.2279,
      "num_input_tokens_seen": 942560,
      "step": 4465
    },
    {
      "epoch": 0.49174917491749176,
      "grad_norm": 0.0211181640625,
      "learning_rate": 0.007374587458745874,
      "loss": 0.2305,
      "num_input_tokens_seen": 943680,
      "step": 4470
    },
    {
      "epoch": 0.4922992299229923,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.007382838283828383,
      "loss": 0.2321,
      "num_input_tokens_seen": 944768,
      "step": 4475
    },
    {
      "epoch": 0.49284928492849284,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.007391089108910891,
      "loss": 0.23,
      "num_input_tokens_seen": 945792,
      "step": 4480
    },
    {
      "epoch": 0.4933993399339934,
      "grad_norm": 0.0029296875,
      "learning_rate": 0.007399339933993399,
      "loss": 0.2326,
      "num_input_tokens_seen": 946848,
      "step": 4485
    },
    {
      "epoch": 0.49394939493949397,
      "grad_norm": 0.01904296875,
      "learning_rate": 0.0074075907590759075,
      "loss": 0.2332,
      "num_input_tokens_seen": 947968,
      "step": 4490
    },
    {
      "epoch": 0.4944994499449945,
      "grad_norm": 0.0213623046875,
      "learning_rate": 0.0074158415841584155,
      "loss": 0.2322,
      "num_input_tokens_seen": 949056,
      "step": 4495
    },
    {
      "epoch": 0.49504950495049505,
      "grad_norm": 0.0177001953125,
      "learning_rate": 0.007424092409240924,
      "loss": 0.2286,
      "num_input_tokens_seen": 950144,
      "step": 4500
    },
    {
      "epoch": 0.4955995599559956,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.007432343234323432,
      "loss": 0.2317,
      "num_input_tokens_seen": 951200,
      "step": 4505
    },
    {
      "epoch": 0.4961496149614962,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.00744059405940594,
      "loss": 0.2271,
      "num_input_tokens_seen": 952224,
      "step": 4510
    },
    {
      "epoch": 0.4966996699669967,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.007448844884488448,
      "loss": 0.238,
      "num_input_tokens_seen": 953280,
      "step": 4515
    },
    {
      "epoch": 0.49724972497249725,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.007457095709570956,
      "loss": 0.2328,
      "num_input_tokens_seen": 954368,
      "step": 4520
    },
    {
      "epoch": 0.4977997799779978,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.007465346534653465,
      "loss": 0.2291,
      "num_input_tokens_seen": 955424,
      "step": 4525
    },
    {
      "epoch": 0.49834983498349833,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.007473597359735973,
      "loss": 0.2306,
      "num_input_tokens_seen": 956416,
      "step": 4530
    },
    {
      "epoch": 0.4988998899889989,
      "grad_norm": 0.0341796875,
      "learning_rate": 0.007481848184818482,
      "loss": 0.2327,
      "num_input_tokens_seen": 957472,
      "step": 4535
    },
    {
      "epoch": 0.49944994499449946,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.00749009900990099,
      "loss": 0.2311,
      "num_input_tokens_seen": 958528,
      "step": 4540
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.007498349834983499,
      "loss": 0.2306,
      "num_input_tokens_seen": 959584,
      "step": 4545
    },
    {
      "epoch": 0.5005500550055005,
      "grad_norm": 0.007537841796875,
      "learning_rate": 0.007506600660066007,
      "loss": 0.2347,
      "num_input_tokens_seen": 960640,
      "step": 4550
    },
    {
      "epoch": 0.5011001100110011,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.007514851485148515,
      "loss": 0.2284,
      "num_input_tokens_seen": 961728,
      "step": 4555
    },
    {
      "epoch": 0.5016501650165016,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.0075231023102310226,
      "loss": 0.2305,
      "num_input_tokens_seen": 962784,
      "step": 4560
    },
    {
      "epoch": 0.5022002200220022,
      "grad_norm": 0.03466796875,
      "learning_rate": 0.0075313531353135305,
      "loss": 0.229,
      "num_input_tokens_seen": 963808,
      "step": 4565
    },
    {
      "epoch": 0.5027502750275028,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.0075396039603960385,
      "loss": 0.2345,
      "num_input_tokens_seen": 964928,
      "step": 4570
    },
    {
      "epoch": 0.5033003300330033,
      "grad_norm": 0.003997802734375,
      "learning_rate": 0.0075478547854785465,
      "loss": 0.2242,
      "num_input_tokens_seen": 966048,
      "step": 4575
    },
    {
      "epoch": 0.5038503850385039,
      "grad_norm": 0.0164794921875,
      "learning_rate": 0.007556105610561057,
      "loss": 0.2263,
      "num_input_tokens_seen": 967040,
      "step": 4580
    },
    {
      "epoch": 0.5044004400440044,
      "grad_norm": 0.0189208984375,
      "learning_rate": 0.007564356435643565,
      "loss": 0.2256,
      "num_input_tokens_seen": 968064,
      "step": 4585
    },
    {
      "epoch": 0.504950495049505,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.007572607260726073,
      "loss": 0.2336,
      "num_input_tokens_seen": 969088,
      "step": 4590
    },
    {
      "epoch": 0.5055005500550055,
      "grad_norm": 0.02490234375,
      "learning_rate": 0.007580858085808581,
      "loss": 0.2462,
      "num_input_tokens_seen": 970112,
      "step": 4595
    },
    {
      "epoch": 0.506050605060506,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.007589108910891089,
      "loss": 0.2298,
      "num_input_tokens_seen": 971136,
      "step": 4600
    },
    {
      "epoch": 0.5066006600660066,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.007597359735973597,
      "loss": 0.2352,
      "num_input_tokens_seen": 972160,
      "step": 4605
    },
    {
      "epoch": 0.5071507150715071,
      "grad_norm": 0.00726318359375,
      "learning_rate": 0.007605610561056105,
      "loss": 0.2357,
      "num_input_tokens_seen": 973184,
      "step": 4610
    },
    {
      "epoch": 0.5077007700770076,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.007613861386138613,
      "loss": 0.2408,
      "num_input_tokens_seen": 974240,
      "step": 4615
    },
    {
      "epoch": 0.5082508250825083,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.007622112211221121,
      "loss": 0.2319,
      "num_input_tokens_seen": 975264,
      "step": 4620
    },
    {
      "epoch": 0.5088008800880088,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.00763036303630363,
      "loss": 0.2306,
      "num_input_tokens_seen": 976256,
      "step": 4625
    },
    {
      "epoch": 0.5093509350935094,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.007638613861386139,
      "loss": 0.2368,
      "num_input_tokens_seen": 977248,
      "step": 4630
    },
    {
      "epoch": 0.5099009900990099,
      "grad_norm": 0.03369140625,
      "learning_rate": 0.007646864686468647,
      "loss": 0.2264,
      "num_input_tokens_seen": 978336,
      "step": 4635
    },
    {
      "epoch": 0.5104510451045104,
      "grad_norm": 0.0201416015625,
      "learning_rate": 0.007655115511551155,
      "loss": 0.2194,
      "num_input_tokens_seen": 979360,
      "step": 4640
    },
    {
      "epoch": 0.511001100110011,
      "grad_norm": 0.013671875,
      "learning_rate": 0.007663366336633663,
      "loss": 0.2298,
      "num_input_tokens_seen": 980416,
      "step": 4645
    },
    {
      "epoch": 0.5115511551155115,
      "grad_norm": 0.026123046875,
      "learning_rate": 0.007671617161716171,
      "loss": 0.2328,
      "num_input_tokens_seen": 981472,
      "step": 4650
    },
    {
      "epoch": 0.5121012101210121,
      "grad_norm": 0.0230712890625,
      "learning_rate": 0.007679867986798679,
      "loss": 0.2321,
      "num_input_tokens_seen": 982592,
      "step": 4655
    },
    {
      "epoch": 0.5126512651265126,
      "grad_norm": 0.0220947265625,
      "learning_rate": 0.007688118811881187,
      "loss": 0.2277,
      "num_input_tokens_seen": 983648,
      "step": 4660
    },
    {
      "epoch": 0.5132013201320133,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.007696369636963695,
      "loss": 0.2218,
      "num_input_tokens_seen": 984640,
      "step": 4665
    },
    {
      "epoch": 0.5137513751375138,
      "grad_norm": 0.0146484375,
      "learning_rate": 0.007704620462046204,
      "loss": 0.229,
      "num_input_tokens_seen": 985728,
      "step": 4670
    },
    {
      "epoch": 0.5143014301430143,
      "grad_norm": 0.020751953125,
      "learning_rate": 0.007712871287128714,
      "loss": 0.2387,
      "num_input_tokens_seen": 986816,
      "step": 4675
    },
    {
      "epoch": 0.5148514851485149,
      "grad_norm": 0.0267333984375,
      "learning_rate": 0.007721122112211222,
      "loss": 0.2385,
      "num_input_tokens_seen": 987904,
      "step": 4680
    },
    {
      "epoch": 0.5154015401540154,
      "grad_norm": 0.018310546875,
      "learning_rate": 0.00772937293729373,
      "loss": 0.2316,
      "num_input_tokens_seen": 988928,
      "step": 4685
    },
    {
      "epoch": 0.5159515951595159,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.0077376237623762376,
      "loss": 0.2375,
      "num_input_tokens_seen": 989920,
      "step": 4690
    },
    {
      "epoch": 0.5165016501650165,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0077458745874587455,
      "loss": 0.2318,
      "num_input_tokens_seen": 990944,
      "step": 4695
    },
    {
      "epoch": 0.517051705170517,
      "grad_norm": 0.021484375,
      "learning_rate": 0.0077541254125412535,
      "loss": 0.2316,
      "num_input_tokens_seen": 991936,
      "step": 4700
    },
    {
      "epoch": 0.5176017601760176,
      "grad_norm": 0.0206298828125,
      "learning_rate": 0.0077623762376237615,
      "loss": 0.2347,
      "num_input_tokens_seen": 992992,
      "step": 4705
    },
    {
      "epoch": 0.5181518151815182,
      "grad_norm": 0.0186767578125,
      "learning_rate": 0.007770627062706269,
      "loss": 0.2335,
      "num_input_tokens_seen": 994048,
      "step": 4710
    },
    {
      "epoch": 0.5187018701870187,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.007778877887788778,
      "loss": 0.2313,
      "num_input_tokens_seen": 995104,
      "step": 4715
    },
    {
      "epoch": 0.5192519251925193,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.007787128712871288,
      "loss": 0.2324,
      "num_input_tokens_seen": 996192,
      "step": 4720
    },
    {
      "epoch": 0.5198019801980198,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.007795379537953796,
      "loss": 0.2313,
      "num_input_tokens_seen": 997216,
      "step": 4725
    },
    {
      "epoch": 0.5203520352035204,
      "grad_norm": 0.038818359375,
      "learning_rate": 0.007803630363036304,
      "loss": 0.2314,
      "num_input_tokens_seen": 998208,
      "step": 4730
    },
    {
      "epoch": 0.5209020902090209,
      "grad_norm": 0.022705078125,
      "learning_rate": 0.007811881188118812,
      "loss": 0.2324,
      "num_input_tokens_seen": 999264,
      "step": 4735
    },
    {
      "epoch": 0.5214521452145214,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.00782013201320132,
      "loss": 0.2334,
      "num_input_tokens_seen": 1000352,
      "step": 4740
    },
    {
      "epoch": 0.522002200220022,
      "grad_norm": 0.0211181640625,
      "learning_rate": 0.007828382838283828,
      "loss": 0.2313,
      "num_input_tokens_seen": 1001408,
      "step": 4745
    },
    {
      "epoch": 0.5225522552255225,
      "grad_norm": 0.032958984375,
      "learning_rate": 0.007836633663366337,
      "loss": 0.2335,
      "num_input_tokens_seen": 1002528,
      "step": 4750
    },
    {
      "epoch": 0.523102310231023,
      "grad_norm": 0.033203125,
      "learning_rate": 0.007844884488448844,
      "loss": 0.2368,
      "num_input_tokens_seen": 1003552,
      "step": 4755
    },
    {
      "epoch": 0.5236523652365237,
      "grad_norm": 0.0172119140625,
      "learning_rate": 0.007853135313531353,
      "loss": 0.2285,
      "num_input_tokens_seen": 1004608,
      "step": 4760
    },
    {
      "epoch": 0.5242024202420242,
      "grad_norm": 0.00311279296875,
      "learning_rate": 0.00786138613861386,
      "loss": 0.2338,
      "num_input_tokens_seen": 1005664,
      "step": 4765
    },
    {
      "epoch": 0.5247524752475248,
      "grad_norm": 0.034912109375,
      "learning_rate": 0.00786963696369637,
      "loss": 0.2305,
      "num_input_tokens_seen": 1006624,
      "step": 4770
    },
    {
      "epoch": 0.5253025302530253,
      "grad_norm": 0.018798828125,
      "learning_rate": 0.007877887788778877,
      "loss": 0.2318,
      "num_input_tokens_seen": 1007680,
      "step": 4775
    },
    {
      "epoch": 0.5258525852585259,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.007886138613861386,
      "loss": 0.2336,
      "num_input_tokens_seen": 1008704,
      "step": 4780
    },
    {
      "epoch": 0.5264026402640264,
      "grad_norm": 0.033447265625,
      "learning_rate": 0.007894389438943895,
      "loss": 0.2377,
      "num_input_tokens_seen": 1009760,
      "step": 4785
    },
    {
      "epoch": 0.5269526952695269,
      "grad_norm": 0.0223388671875,
      "learning_rate": 0.007902640264026402,
      "loss": 0.2334,
      "num_input_tokens_seen": 1010816,
      "step": 4790
    },
    {
      "epoch": 0.5275027502750275,
      "grad_norm": 0.0194091796875,
      "learning_rate": 0.007910891089108911,
      "loss": 0.2337,
      "num_input_tokens_seen": 1011872,
      "step": 4795
    },
    {
      "epoch": 0.528052805280528,
      "grad_norm": 0.0191650390625,
      "learning_rate": 0.007919141914191418,
      "loss": 0.2358,
      "num_input_tokens_seen": 1012960,
      "step": 4800
    },
    {
      "epoch": 0.5286028602860287,
      "grad_norm": 0.015869140625,
      "learning_rate": 0.007927392739273927,
      "loss": 0.2243,
      "num_input_tokens_seen": 1014016,
      "step": 4805
    },
    {
      "epoch": 0.5291529152915292,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.007935643564356434,
      "loss": 0.2326,
      "num_input_tokens_seen": 1015040,
      "step": 4810
    },
    {
      "epoch": 0.5297029702970297,
      "grad_norm": 0.01275634765625,
      "learning_rate": 0.007943894389438945,
      "loss": 0.241,
      "num_input_tokens_seen": 1016064,
      "step": 4815
    },
    {
      "epoch": 0.5302530253025303,
      "grad_norm": 0.016845703125,
      "learning_rate": 0.007952145214521452,
      "loss": 0.2335,
      "num_input_tokens_seen": 1017088,
      "step": 4820
    },
    {
      "epoch": 0.5308030803080308,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.00796039603960396,
      "loss": 0.2254,
      "num_input_tokens_seen": 1018176,
      "step": 4825
    },
    {
      "epoch": 0.5313531353135313,
      "grad_norm": 0.0166015625,
      "learning_rate": 0.00796864686468647,
      "loss": 0.2293,
      "num_input_tokens_seen": 1019168,
      "step": 4830
    },
    {
      "epoch": 0.5319031903190319,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.007976897689768976,
      "loss": 0.2324,
      "num_input_tokens_seen": 1020224,
      "step": 4835
    },
    {
      "epoch": 0.5324532453245324,
      "grad_norm": 0.00726318359375,
      "learning_rate": 0.007985148514851485,
      "loss": 0.2303,
      "num_input_tokens_seen": 1021280,
      "step": 4840
    },
    {
      "epoch": 0.533003300330033,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.007993399339933992,
      "loss": 0.2313,
      "num_input_tokens_seen": 1022368,
      "step": 4845
    },
    {
      "epoch": 0.5335533553355336,
      "grad_norm": 0.0177001953125,
      "learning_rate": 0.008001650165016501,
      "loss": 0.2314,
      "num_input_tokens_seen": 1023424,
      "step": 4850
    },
    {
      "epoch": 0.5341034103410341,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.008009900990099008,
      "loss": 0.2323,
      "num_input_tokens_seen": 1024448,
      "step": 4855
    },
    {
      "epoch": 0.5346534653465347,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.008018151815181519,
      "loss": 0.2345,
      "num_input_tokens_seen": 1025504,
      "step": 4860
    },
    {
      "epoch": 0.5352035203520352,
      "grad_norm": 0.03271484375,
      "learning_rate": 0.008026402640264026,
      "loss": 0.2313,
      "num_input_tokens_seen": 1026560,
      "step": 4865
    },
    {
      "epoch": 0.5357535753575358,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.008034653465346535,
      "loss": 0.2346,
      "num_input_tokens_seen": 1027648,
      "step": 4870
    },
    {
      "epoch": 0.5363036303630363,
      "grad_norm": 0.0169677734375,
      "learning_rate": 0.008042904290429044,
      "loss": 0.2378,
      "num_input_tokens_seen": 1028736,
      "step": 4875
    },
    {
      "epoch": 0.5368536853685368,
      "grad_norm": 0.017333984375,
      "learning_rate": 0.00805115511551155,
      "loss": 0.2295,
      "num_input_tokens_seen": 1029760,
      "step": 4880
    },
    {
      "epoch": 0.5374037403740374,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00805940594059406,
      "loss": 0.2336,
      "num_input_tokens_seen": 1030848,
      "step": 4885
    },
    {
      "epoch": 0.5379537953795379,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.008067656765676567,
      "loss": 0.2337,
      "num_input_tokens_seen": 1032032,
      "step": 4890
    },
    {
      "epoch": 0.5385038503850385,
      "grad_norm": 0.018798828125,
      "learning_rate": 0.008075907590759076,
      "loss": 0.2316,
      "num_input_tokens_seen": 1033152,
      "step": 4895
    },
    {
      "epoch": 0.5390539053905391,
      "grad_norm": 0.00372314453125,
      "learning_rate": 0.008084158415841583,
      "loss": 0.2304,
      "num_input_tokens_seen": 1034176,
      "step": 4900
    },
    {
      "epoch": 0.5396039603960396,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.008092409240924092,
      "loss": 0.2336,
      "num_input_tokens_seen": 1035232,
      "step": 4905
    },
    {
      "epoch": 0.5401540154015402,
      "grad_norm": 0.0084228515625,
      "learning_rate": 0.0081006600660066,
      "loss": 0.2347,
      "num_input_tokens_seen": 1036352,
      "step": 4910
    },
    {
      "epoch": 0.5407040704070407,
      "grad_norm": 0.0169677734375,
      "learning_rate": 0.00810891089108911,
      "loss": 0.2305,
      "num_input_tokens_seen": 1037376,
      "step": 4915
    },
    {
      "epoch": 0.5412541254125413,
      "grad_norm": 0.0169677734375,
      "learning_rate": 0.008117161716171618,
      "loss": 0.2294,
      "num_input_tokens_seen": 1038464,
      "step": 4920
    },
    {
      "epoch": 0.5418041804180418,
      "grad_norm": 0.003936767578125,
      "learning_rate": 0.008125412541254125,
      "loss": 0.2305,
      "num_input_tokens_seen": 1039456,
      "step": 4925
    },
    {
      "epoch": 0.5423542354235423,
      "grad_norm": 0.015625,
      "learning_rate": 0.008133663366336634,
      "loss": 0.2287,
      "num_input_tokens_seen": 1040544,
      "step": 4930
    },
    {
      "epoch": 0.5429042904290429,
      "grad_norm": 0.03271484375,
      "learning_rate": 0.008141914191419141,
      "loss": 0.2381,
      "num_input_tokens_seen": 1041600,
      "step": 4935
    },
    {
      "epoch": 0.5434543454345434,
      "grad_norm": 0.0152587890625,
      "learning_rate": 0.00815016501650165,
      "loss": 0.2286,
      "num_input_tokens_seen": 1042752,
      "step": 4940
    },
    {
      "epoch": 0.5440044004400441,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.008158415841584157,
      "loss": 0.2297,
      "num_input_tokens_seen": 1043712,
      "step": 4945
    },
    {
      "epoch": 0.5445544554455446,
      "grad_norm": 0.0306396484375,
      "learning_rate": 0.008166666666666666,
      "loss": 0.2203,
      "num_input_tokens_seen": 1044768,
      "step": 4950
    },
    {
      "epoch": 0.5451045104510451,
      "grad_norm": 0.0299072265625,
      "learning_rate": 0.008174917491749175,
      "loss": 0.235,
      "num_input_tokens_seen": 1045824,
      "step": 4955
    },
    {
      "epoch": 0.5456545654565457,
      "grad_norm": 0.00750732421875,
      "learning_rate": 0.008183168316831683,
      "loss": 0.239,
      "num_input_tokens_seen": 1046880,
      "step": 4960
    },
    {
      "epoch": 0.5462046204620462,
      "grad_norm": 0.0185546875,
      "learning_rate": 0.008191419141914192,
      "loss": 0.2381,
      "num_input_tokens_seen": 1047904,
      "step": 4965
    },
    {
      "epoch": 0.5467546754675467,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.0081996699669967,
      "loss": 0.2342,
      "num_input_tokens_seen": 1048960,
      "step": 4970
    },
    {
      "epoch": 0.5473047304730473,
      "grad_norm": 0.032958984375,
      "learning_rate": 0.008207920792079208,
      "loss": 0.2314,
      "num_input_tokens_seen": 1049984,
      "step": 4975
    },
    {
      "epoch": 0.5478547854785478,
      "grad_norm": 0.0159912109375,
      "learning_rate": 0.008216171617161715,
      "loss": 0.2314,
      "num_input_tokens_seen": 1050976,
      "step": 4980
    },
    {
      "epoch": 0.5484048404840484,
      "grad_norm": 0.0177001953125,
      "learning_rate": 0.008224422442244224,
      "loss": 0.2304,
      "num_input_tokens_seen": 1052032,
      "step": 4985
    },
    {
      "epoch": 0.5489548954895489,
      "grad_norm": 0.03125,
      "learning_rate": 0.008232673267326731,
      "loss": 0.2314,
      "num_input_tokens_seen": 1053056,
      "step": 4990
    },
    {
      "epoch": 0.5495049504950495,
      "grad_norm": 0.031494140625,
      "learning_rate": 0.00824092409240924,
      "loss": 0.2346,
      "num_input_tokens_seen": 1054048,
      "step": 4995
    },
    {
      "epoch": 0.5500550055005501,
      "grad_norm": 0.03125,
      "learning_rate": 0.008249174917491747,
      "loss": 0.234,
      "num_input_tokens_seen": 1055104,
      "step": 5000
    },
    {
      "epoch": 0.5506050605060506,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.008257425742574258,
      "loss": 0.2326,
      "num_input_tokens_seen": 1056192,
      "step": 5005
    },
    {
      "epoch": 0.5511551155115512,
      "grad_norm": 0.0191650390625,
      "learning_rate": 0.008265676567656767,
      "loss": 0.2289,
      "num_input_tokens_seen": 1057312,
      "step": 5010
    },
    {
      "epoch": 0.5517051705170517,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.008273927392739274,
      "loss": 0.2369,
      "num_input_tokens_seen": 1058368,
      "step": 5015
    },
    {
      "epoch": 0.5522552255225522,
      "grad_norm": 0.015625,
      "learning_rate": 0.008282178217821783,
      "loss": 0.2358,
      "num_input_tokens_seen": 1059424,
      "step": 5020
    },
    {
      "epoch": 0.5528052805280528,
      "grad_norm": 0.0302734375,
      "learning_rate": 0.00829042904290429,
      "loss": 0.2295,
      "num_input_tokens_seen": 1060448,
      "step": 5025
    },
    {
      "epoch": 0.5533553355335533,
      "grad_norm": 0.033203125,
      "learning_rate": 0.008298679867986799,
      "loss": 0.2358,
      "num_input_tokens_seen": 1061472,
      "step": 5030
    },
    {
      "epoch": 0.5539053905390539,
      "grad_norm": 0.0045166015625,
      "learning_rate": 0.008306930693069306,
      "loss": 0.2274,
      "num_input_tokens_seen": 1062432,
      "step": 5035
    },
    {
      "epoch": 0.5544554455445545,
      "grad_norm": 0.0302734375,
      "learning_rate": 0.008315181518151814,
      "loss": 0.2308,
      "num_input_tokens_seen": 1063456,
      "step": 5040
    },
    {
      "epoch": 0.555005500550055,
      "grad_norm": 0.018798828125,
      "learning_rate": 0.008323432343234322,
      "loss": 0.237,
      "num_input_tokens_seen": 1064480,
      "step": 5045
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 0.016357421875,
      "learning_rate": 0.008331683168316832,
      "loss": 0.228,
      "num_input_tokens_seen": 1065536,
      "step": 5050
    },
    {
      "epoch": 0.5561056105610561,
      "grad_norm": 0.035400390625,
      "learning_rate": 0.008339933993399341,
      "loss": 0.2343,
      "num_input_tokens_seen": 1066592,
      "step": 5055
    },
    {
      "epoch": 0.5566556655665567,
      "grad_norm": 0.0045166015625,
      "learning_rate": 0.008348184818481848,
      "loss": 0.2321,
      "num_input_tokens_seen": 1067616,
      "step": 5060
    },
    {
      "epoch": 0.5572057205720572,
      "grad_norm": 0.029541015625,
      "learning_rate": 0.008356435643564357,
      "loss": 0.2234,
      "num_input_tokens_seen": 1068672,
      "step": 5065
    },
    {
      "epoch": 0.5577557755775577,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.008364686468646864,
      "loss": 0.2297,
      "num_input_tokens_seen": 1069696,
      "step": 5070
    },
    {
      "epoch": 0.5583058305830583,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.008372937293729373,
      "loss": 0.2381,
      "num_input_tokens_seen": 1070752,
      "step": 5075
    },
    {
      "epoch": 0.5588558855885588,
      "grad_norm": 0.0152587890625,
      "learning_rate": 0.00838118811881188,
      "loss": 0.2276,
      "num_input_tokens_seen": 1071840,
      "step": 5080
    },
    {
      "epoch": 0.5594059405940595,
      "grad_norm": 0.01708984375,
      "learning_rate": 0.008389438943894389,
      "loss": 0.2267,
      "num_input_tokens_seen": 1072928,
      "step": 5085
    },
    {
      "epoch": 0.55995599559956,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.008397689768976896,
      "loss": 0.2324,
      "num_input_tokens_seen": 1073888,
      "step": 5090
    },
    {
      "epoch": 0.5605060506050605,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008405940594059406,
      "loss": 0.2374,
      "num_input_tokens_seen": 1074944,
      "step": 5095
    },
    {
      "epoch": 0.5610561056105611,
      "grad_norm": 0.014892578125,
      "learning_rate": 0.008414191419141915,
      "loss": 0.227,
      "num_input_tokens_seen": 1076096,
      "step": 5100
    },
    {
      "epoch": 0.5616061606160616,
      "grad_norm": 0.01483154296875,
      "learning_rate": 0.008422442244224422,
      "loss": 0.2278,
      "num_input_tokens_seen": 1077088,
      "step": 5105
    },
    {
      "epoch": 0.5621562156215621,
      "grad_norm": 0.018310546875,
      "learning_rate": 0.008430693069306931,
      "loss": 0.231,
      "num_input_tokens_seen": 1078144,
      "step": 5110
    },
    {
      "epoch": 0.5627062706270627,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.008438943894389438,
      "loss": 0.2333,
      "num_input_tokens_seen": 1079200,
      "step": 5115
    },
    {
      "epoch": 0.5632563256325632,
      "grad_norm": 0.0191650390625,
      "learning_rate": 0.008447194719471947,
      "loss": 0.2308,
      "num_input_tokens_seen": 1080256,
      "step": 5120
    },
    {
      "epoch": 0.5638063806380638,
      "grad_norm": 0.027587890625,
      "learning_rate": 0.008455445544554454,
      "loss": 0.2309,
      "num_input_tokens_seen": 1081344,
      "step": 5125
    },
    {
      "epoch": 0.5643564356435643,
      "grad_norm": 0.0169677734375,
      "learning_rate": 0.008463696369636963,
      "loss": 0.2318,
      "num_input_tokens_seen": 1082400,
      "step": 5130
    },
    {
      "epoch": 0.564906490649065,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00847194719471947,
      "loss": 0.2329,
      "num_input_tokens_seen": 1083520,
      "step": 5135
    },
    {
      "epoch": 0.5654565456545655,
      "grad_norm": 0.0224609375,
      "learning_rate": 0.008480198019801979,
      "loss": 0.234,
      "num_input_tokens_seen": 1084576,
      "step": 5140
    },
    {
      "epoch": 0.566006600660066,
      "grad_norm": 0.0191650390625,
      "learning_rate": 0.00848844884488449,
      "loss": 0.238,
      "num_input_tokens_seen": 1085632,
      "step": 5145
    },
    {
      "epoch": 0.5665566556655666,
      "grad_norm": 0.019287109375,
      "learning_rate": 0.008496699669966997,
      "loss": 0.2368,
      "num_input_tokens_seen": 1086656,
      "step": 5150
    },
    {
      "epoch": 0.5671067106710671,
      "grad_norm": 0.03369140625,
      "learning_rate": 0.008504950495049506,
      "loss": 0.2304,
      "num_input_tokens_seen": 1087680,
      "step": 5155
    },
    {
      "epoch": 0.5676567656765676,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.008513201320132013,
      "loss": 0.2335,
      "num_input_tokens_seen": 1088704,
      "step": 5160
    },
    {
      "epoch": 0.5682068206820682,
      "grad_norm": 0.038818359375,
      "learning_rate": 0.008521452145214522,
      "loss": 0.2319,
      "num_input_tokens_seen": 1089728,
      "step": 5165
    },
    {
      "epoch": 0.5687568756875687,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008529702970297029,
      "loss": 0.2304,
      "num_input_tokens_seen": 1090816,
      "step": 5170
    },
    {
      "epoch": 0.5693069306930693,
      "grad_norm": 0.019287109375,
      "learning_rate": 0.008537953795379537,
      "loss": 0.2314,
      "num_input_tokens_seen": 1091872,
      "step": 5175
    },
    {
      "epoch": 0.5698569856985699,
      "grad_norm": 0.01904296875,
      "learning_rate": 0.008546204620462045,
      "loss": 0.2304,
      "num_input_tokens_seen": 1092992,
      "step": 5180
    },
    {
      "epoch": 0.5704070407040704,
      "grad_norm": 0.038330078125,
      "learning_rate": 0.008554455445544553,
      "loss": 0.2346,
      "num_input_tokens_seen": 1094016,
      "step": 5185
    },
    {
      "epoch": 0.570957095709571,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.008562706270627064,
      "loss": 0.2351,
      "num_input_tokens_seen": 1095104,
      "step": 5190
    },
    {
      "epoch": 0.5715071507150715,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.008570957095709571,
      "loss": 0.2309,
      "num_input_tokens_seen": 1096128,
      "step": 5195
    },
    {
      "epoch": 0.5720572057205721,
      "grad_norm": 0.0189208984375,
      "learning_rate": 0.00857920792079208,
      "loss": 0.2309,
      "num_input_tokens_seen": 1097152,
      "step": 5200
    },
    {
      "epoch": 0.5726072607260726,
      "grad_norm": 0.01806640625,
      "learning_rate": 0.008587458745874587,
      "loss": 0.2303,
      "num_input_tokens_seen": 1098176,
      "step": 5205
    },
    {
      "epoch": 0.5731573157315731,
      "grad_norm": 0.0211181640625,
      "learning_rate": 0.008595709570957096,
      "loss": 0.2325,
      "num_input_tokens_seen": 1099168,
      "step": 5210
    },
    {
      "epoch": 0.5737073707370737,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.008603960396039603,
      "loss": 0.2279,
      "num_input_tokens_seen": 1100160,
      "step": 5215
    },
    {
      "epoch": 0.5742574257425742,
      "grad_norm": 0.0191650390625,
      "learning_rate": 0.008612211221122112,
      "loss": 0.2209,
      "num_input_tokens_seen": 1101184,
      "step": 5220
    },
    {
      "epoch": 0.5748074807480749,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.008620462046204619,
      "loss": 0.2376,
      "num_input_tokens_seen": 1102272,
      "step": 5225
    },
    {
      "epoch": 0.5753575357535754,
      "grad_norm": 0.036376953125,
      "learning_rate": 0.008628712871287128,
      "loss": 0.2361,
      "num_input_tokens_seen": 1103328,
      "step": 5230
    },
    {
      "epoch": 0.5759075907590759,
      "grad_norm": 0.024658203125,
      "learning_rate": 0.008636963696369638,
      "loss": 0.2266,
      "num_input_tokens_seen": 1104384,
      "step": 5235
    },
    {
      "epoch": 0.5764576457645765,
      "grad_norm": 0.0230712890625,
      "learning_rate": 0.008645214521452145,
      "loss": 0.2349,
      "num_input_tokens_seen": 1105440,
      "step": 5240
    },
    {
      "epoch": 0.577007700770077,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.008653465346534654,
      "loss": 0.2358,
      "num_input_tokens_seen": 1106496,
      "step": 5245
    },
    {
      "epoch": 0.5775577557755776,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.008661716171617161,
      "loss": 0.2413,
      "num_input_tokens_seen": 1107552,
      "step": 5250
    },
    {
      "epoch": 0.5781078107810781,
      "grad_norm": 0.03173828125,
      "learning_rate": 0.00866996699669967,
      "loss": 0.2426,
      "num_input_tokens_seen": 1108640,
      "step": 5255
    },
    {
      "epoch": 0.5786578657865786,
      "grad_norm": 0.0140380859375,
      "learning_rate": 0.008678217821782177,
      "loss": 0.2325,
      "num_input_tokens_seen": 1109728,
      "step": 5260
    },
    {
      "epoch": 0.5792079207920792,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.008686468646864686,
      "loss": 0.2313,
      "num_input_tokens_seen": 1110784,
      "step": 5265
    },
    {
      "epoch": 0.5797579757975797,
      "grad_norm": 0.03515625,
      "learning_rate": 0.008694719471947193,
      "loss": 0.2265,
      "num_input_tokens_seen": 1111840,
      "step": 5270
    },
    {
      "epoch": 0.5803080308030804,
      "grad_norm": 0.0478515625,
      "learning_rate": 0.008702970297029702,
      "loss": 0.228,
      "num_input_tokens_seen": 1112832,
      "step": 5275
    },
    {
      "epoch": 0.5808580858085809,
      "grad_norm": 0.0634765625,
      "learning_rate": 0.00871122112211221,
      "loss": 0.2403,
      "num_input_tokens_seen": 1113920,
      "step": 5280
    },
    {
      "epoch": 0.5814081408140814,
      "grad_norm": 0.01251220703125,
      "learning_rate": 0.00871947194719472,
      "loss": 0.2255,
      "num_input_tokens_seen": 1115040,
      "step": 5285
    },
    {
      "epoch": 0.581958195819582,
      "grad_norm": 0.041259765625,
      "learning_rate": 0.008727722772277229,
      "loss": 0.2409,
      "num_input_tokens_seen": 1116064,
      "step": 5290
    },
    {
      "epoch": 0.5825082508250825,
      "grad_norm": 0.046875,
      "learning_rate": 0.008735973597359736,
      "loss": 0.2336,
      "num_input_tokens_seen": 1117120,
      "step": 5295
    },
    {
      "epoch": 0.583058305830583,
      "grad_norm": 0.01385498046875,
      "learning_rate": 0.008744224422442244,
      "loss": 0.2314,
      "num_input_tokens_seen": 1118176,
      "step": 5300
    },
    {
      "epoch": 0.5836083608360836,
      "grad_norm": 0.044677734375,
      "learning_rate": 0.008752475247524752,
      "loss": 0.2386,
      "num_input_tokens_seen": 1119232,
      "step": 5305
    },
    {
      "epoch": 0.5841584158415841,
      "grad_norm": 0.035888671875,
      "learning_rate": 0.00876072607260726,
      "loss": 0.2358,
      "num_input_tokens_seen": 1120320,
      "step": 5310
    },
    {
      "epoch": 0.5847084708470847,
      "grad_norm": 0.01324462890625,
      "learning_rate": 0.008768976897689768,
      "loss": 0.2312,
      "num_input_tokens_seen": 1121408,
      "step": 5315
    },
    {
      "epoch": 0.5852585258525853,
      "grad_norm": 0.024658203125,
      "learning_rate": 0.008777227722772276,
      "loss": 0.2291,
      "num_input_tokens_seen": 1122496,
      "step": 5320
    },
    {
      "epoch": 0.5858085808580858,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.008785478547854785,
      "loss": 0.2283,
      "num_input_tokens_seen": 1123456,
      "step": 5325
    },
    {
      "epoch": 0.5863586358635864,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.008793729372937294,
      "loss": 0.2356,
      "num_input_tokens_seen": 1124512,
      "step": 5330
    },
    {
      "epoch": 0.5869086908690869,
      "grad_norm": 0.0208740234375,
      "learning_rate": 0.008801980198019803,
      "loss": 0.2293,
      "num_input_tokens_seen": 1125568,
      "step": 5335
    },
    {
      "epoch": 0.5874587458745875,
      "grad_norm": 0.00726318359375,
      "learning_rate": 0.00881023102310231,
      "loss": 0.2302,
      "num_input_tokens_seen": 1126624,
      "step": 5340
    },
    {
      "epoch": 0.588008800880088,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.008818481848184819,
      "loss": 0.2322,
      "num_input_tokens_seen": 1127712,
      "step": 5345
    },
    {
      "epoch": 0.5885588558855885,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.008826732673267326,
      "loss": 0.2323,
      "num_input_tokens_seen": 1128736,
      "step": 5350
    },
    {
      "epoch": 0.5891089108910891,
      "grad_norm": 0.03759765625,
      "learning_rate": 0.008834983498349835,
      "loss": 0.2313,
      "num_input_tokens_seen": 1129760,
      "step": 5355
    },
    {
      "epoch": 0.5896589658965896,
      "grad_norm": 0.033935546875,
      "learning_rate": 0.008843234323432342,
      "loss": 0.2302,
      "num_input_tokens_seen": 1130880,
      "step": 5360
    },
    {
      "epoch": 0.5902090209020903,
      "grad_norm": 0.019287109375,
      "learning_rate": 0.00885148514851485,
      "loss": 0.2292,
      "num_input_tokens_seen": 1131904,
      "step": 5365
    },
    {
      "epoch": 0.5907590759075908,
      "grad_norm": 0.019287109375,
      "learning_rate": 0.00885973597359736,
      "loss": 0.2303,
      "num_input_tokens_seen": 1132960,
      "step": 5370
    },
    {
      "epoch": 0.5913091309130913,
      "grad_norm": 0.017822265625,
      "learning_rate": 0.008867986798679867,
      "loss": 0.2323,
      "num_input_tokens_seen": 1133952,
      "step": 5375
    },
    {
      "epoch": 0.5918591859185919,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.008876237623762377,
      "loss": 0.2334,
      "num_input_tokens_seen": 1135008,
      "step": 5380
    },
    {
      "epoch": 0.5924092409240924,
      "grad_norm": 0.0172119140625,
      "learning_rate": 0.008884488448844884,
      "loss": 0.2301,
      "num_input_tokens_seen": 1136032,
      "step": 5385
    },
    {
      "epoch": 0.592959295929593,
      "grad_norm": 0.0033416748046875,
      "learning_rate": 0.008892739273927393,
      "loss": 0.2301,
      "num_input_tokens_seen": 1136992,
      "step": 5390
    },
    {
      "epoch": 0.5935093509350935,
      "grad_norm": 0.0159912109375,
      "learning_rate": 0.0089009900990099,
      "loss": 0.2344,
      "num_input_tokens_seen": 1138112,
      "step": 5395
    },
    {
      "epoch": 0.594059405940594,
      "grad_norm": 0.003997802734375,
      "learning_rate": 0.008909240924092409,
      "loss": 0.2334,
      "num_input_tokens_seen": 1139136,
      "step": 5400
    },
    {
      "epoch": 0.5946094609460946,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.008917491749174916,
      "loss": 0.2282,
      "num_input_tokens_seen": 1140128,
      "step": 5405
    },
    {
      "epoch": 0.5951595159515951,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.008925742574257425,
      "loss": 0.2273,
      "num_input_tokens_seen": 1141216,
      "step": 5410
    },
    {
      "epoch": 0.5957095709570958,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.008933993399339934,
      "loss": 0.2265,
      "num_input_tokens_seen": 1142272,
      "step": 5415
    },
    {
      "epoch": 0.5962596259625963,
      "grad_norm": 0.037841796875,
      "learning_rate": 0.008942244224422441,
      "loss": 0.2238,
      "num_input_tokens_seen": 1143328,
      "step": 5420
    },
    {
      "epoch": 0.5968096809680968,
      "grad_norm": 0.0279541015625,
      "learning_rate": 0.008950495049504952,
      "loss": 0.2468,
      "num_input_tokens_seen": 1144320,
      "step": 5425
    },
    {
      "epoch": 0.5973597359735974,
      "grad_norm": 0.014892578125,
      "learning_rate": 0.008958745874587459,
      "loss": 0.2257,
      "num_input_tokens_seen": 1145376,
      "step": 5430
    },
    {
      "epoch": 0.5979097909790979,
      "grad_norm": 0.02392578125,
      "learning_rate": 0.008966996699669967,
      "loss": 0.2371,
      "num_input_tokens_seen": 1146400,
      "step": 5435
    },
    {
      "epoch": 0.5984598459845984,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.008975247524752475,
      "loss": 0.2354,
      "num_input_tokens_seen": 1147392,
      "step": 5440
    },
    {
      "epoch": 0.599009900990099,
      "grad_norm": 0.034912109375,
      "learning_rate": 0.008983498349834983,
      "loss": 0.2411,
      "num_input_tokens_seen": 1148448,
      "step": 5445
    },
    {
      "epoch": 0.5995599559955995,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.00899174917491749,
      "loss": 0.2304,
      "num_input_tokens_seen": 1149504,
      "step": 5450
    },
    {
      "epoch": 0.6001100110011001,
      "grad_norm": 0.0166015625,
      "learning_rate": 0.009,
      "loss": 0.2274,
      "num_input_tokens_seen": 1150592,
      "step": 5455
    },
    {
      "epoch": 0.6006600660066007,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.009008250825082508,
      "loss": 0.2314,
      "num_input_tokens_seen": 1151648,
      "step": 5460
    },
    {
      "epoch": 0.6012101210121013,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.009016501650165015,
      "loss": 0.2314,
      "num_input_tokens_seen": 1152672,
      "step": 5465
    },
    {
      "epoch": 0.6017601760176018,
      "grad_norm": 0.030517578125,
      "learning_rate": 0.009024752475247526,
      "loss": 0.2335,
      "num_input_tokens_seen": 1153728,
      "step": 5470
    },
    {
      "epoch": 0.6023102310231023,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.009033003300330033,
      "loss": 0.2324,
      "num_input_tokens_seen": 1154752,
      "step": 5475
    },
    {
      "epoch": 0.6028602860286029,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.009041254125412542,
      "loss": 0.2324,
      "num_input_tokens_seen": 1155840,
      "step": 5480
    },
    {
      "epoch": 0.6034103410341034,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.009049504950495049,
      "loss": 0.2293,
      "num_input_tokens_seen": 1156928,
      "step": 5485
    },
    {
      "epoch": 0.6039603960396039,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.009057755775577558,
      "loss": 0.2303,
      "num_input_tokens_seen": 1157984,
      "step": 5490
    },
    {
      "epoch": 0.6045104510451045,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.009066006600660065,
      "loss": 0.2313,
      "num_input_tokens_seen": 1159072,
      "step": 5495
    },
    {
      "epoch": 0.605060506050605,
      "grad_norm": 0.014892578125,
      "learning_rate": 0.009074257425742574,
      "loss": 0.2313,
      "num_input_tokens_seen": 1160160,
      "step": 5500
    },
    {
      "epoch": 0.6056105610561056,
      "grad_norm": 0.0150146484375,
      "learning_rate": 0.009082508250825082,
      "loss": 0.2313,
      "num_input_tokens_seen": 1161248,
      "step": 5505
    },
    {
      "epoch": 0.6061606160616062,
      "grad_norm": 0.0291748046875,
      "learning_rate": 0.00909075907590759,
      "loss": 0.2304,
      "num_input_tokens_seen": 1162272,
      "step": 5510
    },
    {
      "epoch": 0.6067106710671067,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.009099009900990098,
      "loss": 0.2304,
      "num_input_tokens_seen": 1163296,
      "step": 5515
    },
    {
      "epoch": 0.6072607260726073,
      "grad_norm": 0.003936767578125,
      "learning_rate": 0.009107260726072607,
      "loss": 0.2335,
      "num_input_tokens_seen": 1164352,
      "step": 5520
    },
    {
      "epoch": 0.6078107810781078,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.009115511551155116,
      "loss": 0.2314,
      "num_input_tokens_seen": 1165344,
      "step": 5525
    },
    {
      "epoch": 0.6083608360836084,
      "grad_norm": 0.0169677734375,
      "learning_rate": 0.009123762376237623,
      "loss": 0.2284,
      "num_input_tokens_seen": 1166432,
      "step": 5530
    },
    {
      "epoch": 0.6089108910891089,
      "grad_norm": 0.01397705078125,
      "learning_rate": 0.009132013201320132,
      "loss": 0.2307,
      "num_input_tokens_seen": 1167456,
      "step": 5535
    },
    {
      "epoch": 0.6094609460946094,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.009140264026402639,
      "loss": 0.2298,
      "num_input_tokens_seen": 1168512,
      "step": 5540
    },
    {
      "epoch": 0.61001100110011,
      "grad_norm": 0.0191650390625,
      "learning_rate": 0.009148514851485148,
      "loss": 0.2381,
      "num_input_tokens_seen": 1169600,
      "step": 5545
    },
    {
      "epoch": 0.6105610561056105,
      "grad_norm": 0.0166015625,
      "learning_rate": 0.009156765676567657,
      "loss": 0.234,
      "num_input_tokens_seen": 1170624,
      "step": 5550
    },
    {
      "epoch": 0.6111111111111112,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.009165016501650164,
      "loss": 0.2359,
      "num_input_tokens_seen": 1171680,
      "step": 5555
    },
    {
      "epoch": 0.6116611661166117,
      "grad_norm": 0.0157470703125,
      "learning_rate": 0.009173267326732673,
      "loss": 0.2347,
      "num_input_tokens_seen": 1172704,
      "step": 5560
    },
    {
      "epoch": 0.6122112211221122,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.009181518151815182,
      "loss": 0.2293,
      "num_input_tokens_seen": 1173760,
      "step": 5565
    },
    {
      "epoch": 0.6127612761276128,
      "grad_norm": 0.0146484375,
      "learning_rate": 0.00918976897689769,
      "loss": 0.2303,
      "num_input_tokens_seen": 1174848,
      "step": 5570
    },
    {
      "epoch": 0.6133113311331133,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.009198019801980198,
      "loss": 0.2325,
      "num_input_tokens_seen": 1175872,
      "step": 5575
    },
    {
      "epoch": 0.6138613861386139,
      "grad_norm": 0.01385498046875,
      "learning_rate": 0.009206270627062706,
      "loss": 0.2313,
      "num_input_tokens_seen": 1176864,
      "step": 5580
    },
    {
      "epoch": 0.6144114411441144,
      "grad_norm": 0.01519775390625,
      "learning_rate": 0.009214521452145213,
      "loss": 0.2324,
      "num_input_tokens_seen": 1177984,
      "step": 5585
    },
    {
      "epoch": 0.6149614961496149,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.009222772277227722,
      "loss": 0.2324,
      "num_input_tokens_seen": 1179072,
      "step": 5590
    },
    {
      "epoch": 0.6155115511551155,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.009231023102310231,
      "loss": 0.2313,
      "num_input_tokens_seen": 1180128,
      "step": 5595
    },
    {
      "epoch": 0.6160616061606161,
      "grad_norm": 0.004364013671875,
      "learning_rate": 0.009239273927392738,
      "loss": 0.2314,
      "num_input_tokens_seen": 1181184,
      "step": 5600
    },
    {
      "epoch": 0.6166116611661167,
      "grad_norm": 0.01422119140625,
      "learning_rate": 0.009247524752475247,
      "loss": 0.2315,
      "num_input_tokens_seen": 1182208,
      "step": 5605
    },
    {
      "epoch": 0.6171617161716172,
      "grad_norm": 0.0035400390625,
      "learning_rate": 0.009255775577557756,
      "loss": 0.2305,
      "num_input_tokens_seen": 1183264,
      "step": 5610
    },
    {
      "epoch": 0.6177117711771177,
      "grad_norm": 0.004302978515625,
      "learning_rate": 0.009264026402640265,
      "loss": 0.2303,
      "num_input_tokens_seen": 1184256,
      "step": 5615
    },
    {
      "epoch": 0.6182618261826183,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.009272277227722772,
      "loss": 0.2315,
      "num_input_tokens_seen": 1185280,
      "step": 5620
    },
    {
      "epoch": 0.6188118811881188,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00928052805280528,
      "loss": 0.2348,
      "num_input_tokens_seen": 1186368,
      "step": 5625
    },
    {
      "epoch": 0.6193619361936193,
      "grad_norm": 0.00396728515625,
      "learning_rate": 0.009288778877887788,
      "loss": 0.2368,
      "num_input_tokens_seen": 1187424,
      "step": 5630
    },
    {
      "epoch": 0.6199119911991199,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.009297029702970297,
      "loss": 0.2314,
      "num_input_tokens_seen": 1188448,
      "step": 5635
    },
    {
      "epoch": 0.6204620462046204,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.009305280528052805,
      "loss": 0.2325,
      "num_input_tokens_seen": 1189504,
      "step": 5640
    },
    {
      "epoch": 0.621012101210121,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009313531353135313,
      "loss": 0.2303,
      "num_input_tokens_seen": 1190592,
      "step": 5645
    },
    {
      "epoch": 0.6215621562156216,
      "grad_norm": 0.01434326171875,
      "learning_rate": 0.009321782178217821,
      "loss": 0.2314,
      "num_input_tokens_seen": 1191648,
      "step": 5650
    },
    {
      "epoch": 0.6221122112211221,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.009330033003300328,
      "loss": 0.2324,
      "num_input_tokens_seen": 1192704,
      "step": 5655
    },
    {
      "epoch": 0.6226622662266227,
      "grad_norm": 0.01361083984375,
      "learning_rate": 0.009338283828382839,
      "loss": 0.2345,
      "num_input_tokens_seen": 1193728,
      "step": 5660
    },
    {
      "epoch": 0.6232123212321232,
      "grad_norm": 0.004547119140625,
      "learning_rate": 0.009346534653465346,
      "loss": 0.2303,
      "num_input_tokens_seen": 1194784,
      "step": 5665
    },
    {
      "epoch": 0.6237623762376238,
      "grad_norm": 0.004302978515625,
      "learning_rate": 0.009354785478547855,
      "loss": 0.2313,
      "num_input_tokens_seen": 1195808,
      "step": 5670
    },
    {
      "epoch": 0.6243124312431243,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.009363036303630362,
      "loss": 0.2345,
      "num_input_tokens_seen": 1196864,
      "step": 5675
    },
    {
      "epoch": 0.6248624862486248,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.009371287128712871,
      "loss": 0.2293,
      "num_input_tokens_seen": 1197952,
      "step": 5680
    },
    {
      "epoch": 0.6254125412541254,
      "grad_norm": 0.013916015625,
      "learning_rate": 0.00937953795379538,
      "loss": 0.2273,
      "num_input_tokens_seen": 1198976,
      "step": 5685
    },
    {
      "epoch": 0.6259625962596259,
      "grad_norm": 0.0274658203125,
      "learning_rate": 0.009387788778877887,
      "loss": 0.2357,
      "num_input_tokens_seen": 1200032,
      "step": 5690
    },
    {
      "epoch": 0.6265126512651266,
      "grad_norm": 0.0137939453125,
      "learning_rate": 0.009396039603960396,
      "loss": 0.2316,
      "num_input_tokens_seen": 1201120,
      "step": 5695
    },
    {
      "epoch": 0.6270627062706271,
      "grad_norm": 0.01300048828125,
      "learning_rate": 0.009404290429042903,
      "loss": 0.2305,
      "num_input_tokens_seen": 1202144,
      "step": 5700
    },
    {
      "epoch": 0.6276127612761276,
      "grad_norm": 0.0037384033203125,
      "learning_rate": 0.009412541254125413,
      "loss": 0.2325,
      "num_input_tokens_seen": 1203200,
      "step": 5705
    },
    {
      "epoch": 0.6281628162816282,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.00942079207920792,
      "loss": 0.2304,
      "num_input_tokens_seen": 1204320,
      "step": 5710
    },
    {
      "epoch": 0.6287128712871287,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.00942904290429043,
      "loss": 0.2273,
      "num_input_tokens_seen": 1205344,
      "step": 5715
    },
    {
      "epoch": 0.6292629262926293,
      "grad_norm": 0.01300048828125,
      "learning_rate": 0.009437293729372936,
      "loss": 0.2316,
      "num_input_tokens_seen": 1206336,
      "step": 5720
    },
    {
      "epoch": 0.6298129812981298,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.009445544554455445,
      "loss": 0.2337,
      "num_input_tokens_seen": 1207392,
      "step": 5725
    },
    {
      "epoch": 0.6303630363036303,
      "grad_norm": 0.0155029296875,
      "learning_rate": 0.009453795379537954,
      "loss": 0.2337,
      "num_input_tokens_seen": 1208352,
      "step": 5730
    },
    {
      "epoch": 0.6309130913091309,
      "grad_norm": 0.0135498046875,
      "learning_rate": 0.009462046204620461,
      "loss": 0.2274,
      "num_input_tokens_seen": 1209408,
      "step": 5735
    },
    {
      "epoch": 0.6314631463146315,
      "grad_norm": 0.028076171875,
      "learning_rate": 0.00947029702970297,
      "loss": 0.2343,
      "num_input_tokens_seen": 1210432,
      "step": 5740
    },
    {
      "epoch": 0.6320132013201321,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.009478547854785477,
      "loss": 0.2347,
      "num_input_tokens_seen": 1211520,
      "step": 5745
    },
    {
      "epoch": 0.6325632563256326,
      "grad_norm": 0.0169677734375,
      "learning_rate": 0.009486798679867986,
      "loss": 0.2296,
      "num_input_tokens_seen": 1212608,
      "step": 5750
    },
    {
      "epoch": 0.6331133113311331,
      "grad_norm": 0.0145263671875,
      "learning_rate": 0.009495049504950495,
      "loss": 0.2311,
      "num_input_tokens_seen": 1213600,
      "step": 5755
    },
    {
      "epoch": 0.6336633663366337,
      "grad_norm": 0.0262451171875,
      "learning_rate": 0.009503300330033004,
      "loss": 0.2294,
      "num_input_tokens_seen": 1214656,
      "step": 5760
    },
    {
      "epoch": 0.6342134213421342,
      "grad_norm": 0.0167236328125,
      "learning_rate": 0.00951155115511551,
      "loss": 0.2358,
      "num_input_tokens_seen": 1215712,
      "step": 5765
    },
    {
      "epoch": 0.6347634763476347,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.00951980198019802,
      "loss": 0.2326,
      "num_input_tokens_seen": 1216768,
      "step": 5770
    },
    {
      "epoch": 0.6353135313531353,
      "grad_norm": 0.0147705078125,
      "learning_rate": 0.009528052805280528,
      "loss": 0.2326,
      "num_input_tokens_seen": 1217856,
      "step": 5775
    },
    {
      "epoch": 0.6358635863586358,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009536303630363036,
      "loss": 0.2338,
      "num_input_tokens_seen": 1218880,
      "step": 5780
    },
    {
      "epoch": 0.6364136413641364,
      "grad_norm": 0.0135498046875,
      "learning_rate": 0.009544554455445544,
      "loss": 0.2389,
      "num_input_tokens_seen": 1219904,
      "step": 5785
    },
    {
      "epoch": 0.636963696369637,
      "grad_norm": 0.0267333984375,
      "learning_rate": 0.009552805280528051,
      "loss": 0.2336,
      "num_input_tokens_seen": 1220928,
      "step": 5790
    },
    {
      "epoch": 0.6375137513751375,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.00956105610561056,
      "loss": 0.2314,
      "num_input_tokens_seen": 1221952,
      "step": 5795
    },
    {
      "epoch": 0.6380638063806381,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.009569306930693069,
      "loss": 0.2336,
      "num_input_tokens_seen": 1223008,
      "step": 5800
    },
    {
      "epoch": 0.6386138613861386,
      "grad_norm": 0.0142822265625,
      "learning_rate": 0.009577557755775578,
      "loss": 0.2347,
      "num_input_tokens_seen": 1224032,
      "step": 5805
    },
    {
      "epoch": 0.6391639163916392,
      "grad_norm": 0.01300048828125,
      "learning_rate": 0.009585808580858085,
      "loss": 0.2254,
      "num_input_tokens_seen": 1225056,
      "step": 5810
    },
    {
      "epoch": 0.6397139713971397,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.009594059405940594,
      "loss": 0.2226,
      "num_input_tokens_seen": 1226176,
      "step": 5815
    },
    {
      "epoch": 0.6402640264026402,
      "grad_norm": 0.007080078125,
      "learning_rate": 0.009602310231023103,
      "loss": 0.2247,
      "num_input_tokens_seen": 1227232,
      "step": 5820
    },
    {
      "epoch": 0.6408140814081408,
      "grad_norm": 0.01300048828125,
      "learning_rate": 0.00961056105610561,
      "loss": 0.2244,
      "num_input_tokens_seen": 1228256,
      "step": 5825
    },
    {
      "epoch": 0.6413641364136413,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.009618811881188119,
      "loss": 0.238,
      "num_input_tokens_seen": 1229344,
      "step": 5830
    },
    {
      "epoch": 0.641914191419142,
      "grad_norm": 0.01409912109375,
      "learning_rate": 0.009627062706270626,
      "loss": 0.2389,
      "num_input_tokens_seen": 1230432,
      "step": 5835
    },
    {
      "epoch": 0.6424642464246425,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.009635313531353135,
      "loss": 0.2369,
      "num_input_tokens_seen": 1231488,
      "step": 5840
    },
    {
      "epoch": 0.643014301430143,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.009643564356435643,
      "loss": 0.2329,
      "num_input_tokens_seen": 1232512,
      "step": 5845
    },
    {
      "epoch": 0.6435643564356436,
      "grad_norm": 0.018310546875,
      "learning_rate": 0.009651815181518152,
      "loss": 0.2423,
      "num_input_tokens_seen": 1233632,
      "step": 5850
    },
    {
      "epoch": 0.6441144114411441,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.00966006600660066,
      "loss": 0.235,
      "num_input_tokens_seen": 1234624,
      "step": 5855
    },
    {
      "epoch": 0.6446644664466447,
      "grad_norm": 0.017333984375,
      "learning_rate": 0.009668316831683168,
      "loss": 0.2358,
      "num_input_tokens_seen": 1235680,
      "step": 5860
    },
    {
      "epoch": 0.6452145214521452,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.009676567656765677,
      "loss": 0.2346,
      "num_input_tokens_seen": 1236768,
      "step": 5865
    },
    {
      "epoch": 0.6457645764576457,
      "grad_norm": 0.030029296875,
      "learning_rate": 0.009684818481848184,
      "loss": 0.2323,
      "num_input_tokens_seen": 1237856,
      "step": 5870
    },
    {
      "epoch": 0.6463146314631463,
      "grad_norm": 0.007537841796875,
      "learning_rate": 0.009693069306930693,
      "loss": 0.2293,
      "num_input_tokens_seen": 1238880,
      "step": 5875
    },
    {
      "epoch": 0.6468646864686468,
      "grad_norm": 0.056640625,
      "learning_rate": 0.0097013201320132,
      "loss": 0.2243,
      "num_input_tokens_seen": 1239968,
      "step": 5880
    },
    {
      "epoch": 0.6474147414741475,
      "grad_norm": 0.021484375,
      "learning_rate": 0.009709570957095709,
      "loss": 0.2162,
      "num_input_tokens_seen": 1241024,
      "step": 5885
    },
    {
      "epoch": 0.647964796479648,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.009717821782178216,
      "loss": 0.2315,
      "num_input_tokens_seen": 1242048,
      "step": 5890
    },
    {
      "epoch": 0.6485148514851485,
      "grad_norm": 0.04931640625,
      "learning_rate": 0.009726072607260727,
      "loss": 0.2665,
      "num_input_tokens_seen": 1243040,
      "step": 5895
    },
    {
      "epoch": 0.6490649064906491,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.009734323432343234,
      "loss": 0.2372,
      "num_input_tokens_seen": 1244064,
      "step": 5900
    },
    {
      "epoch": 0.6496149614961496,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.009742574257425743,
      "loss": 0.2314,
      "num_input_tokens_seen": 1245088,
      "step": 5905
    },
    {
      "epoch": 0.6501650165016502,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.009750825082508251,
      "loss": 0.2323,
      "num_input_tokens_seen": 1246144,
      "step": 5910
    },
    {
      "epoch": 0.6507150715071507,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.009759075907590758,
      "loss": 0.2333,
      "num_input_tokens_seen": 1247232,
      "step": 5915
    },
    {
      "epoch": 0.6512651265126512,
      "grad_norm": 0.0196533203125,
      "learning_rate": 0.009767326732673267,
      "loss": 0.2313,
      "num_input_tokens_seen": 1248288,
      "step": 5920
    },
    {
      "epoch": 0.6518151815181518,
      "grad_norm": 0.019287109375,
      "learning_rate": 0.009775577557755774,
      "loss": 0.2242,
      "num_input_tokens_seen": 1249376,
      "step": 5925
    },
    {
      "epoch": 0.6523652365236524,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.009783828382838283,
      "loss": 0.2359,
      "num_input_tokens_seen": 1250432,
      "step": 5930
    },
    {
      "epoch": 0.652915291529153,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00979207920792079,
      "loss": 0.2338,
      "num_input_tokens_seen": 1251552,
      "step": 5935
    },
    {
      "epoch": 0.6534653465346535,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.009800330033003301,
      "loss": 0.2283,
      "num_input_tokens_seen": 1252608,
      "step": 5940
    },
    {
      "epoch": 0.654015401540154,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.009808580858085808,
      "loss": 0.2345,
      "num_input_tokens_seen": 1253664,
      "step": 5945
    },
    {
      "epoch": 0.6545654565456546,
      "grad_norm": 0.00408935546875,
      "learning_rate": 0.009816831683168317,
      "loss": 0.2303,
      "num_input_tokens_seen": 1254720,
      "step": 5950
    },
    {
      "epoch": 0.6551155115511551,
      "grad_norm": 0.034912109375,
      "learning_rate": 0.009825082508250826,
      "loss": 0.2302,
      "num_input_tokens_seen": 1255872,
      "step": 5955
    },
    {
      "epoch": 0.6556655665566556,
      "grad_norm": 0.0341796875,
      "learning_rate": 0.009833333333333333,
      "loss": 0.2323,
      "num_input_tokens_seen": 1256928,
      "step": 5960
    },
    {
      "epoch": 0.6562156215621562,
      "grad_norm": 0.0035247802734375,
      "learning_rate": 0.009841584158415842,
      "loss": 0.2323,
      "num_input_tokens_seen": 1257888,
      "step": 5965
    },
    {
      "epoch": 0.6567656765676567,
      "grad_norm": 0.0152587890625,
      "learning_rate": 0.009849834983498349,
      "loss": 0.2303,
      "num_input_tokens_seen": 1258944,
      "step": 5970
    },
    {
      "epoch": 0.6573157315731574,
      "grad_norm": 0.004150390625,
      "learning_rate": 0.009858085808580858,
      "loss": 0.2323,
      "num_input_tokens_seen": 1259936,
      "step": 5975
    },
    {
      "epoch": 0.6578657865786579,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.009866336633663365,
      "loss": 0.2314,
      "num_input_tokens_seen": 1260992,
      "step": 5980
    },
    {
      "epoch": 0.6584158415841584,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.009874587458745875,
      "loss": 0.2274,
      "num_input_tokens_seen": 1262016,
      "step": 5985
    },
    {
      "epoch": 0.658965896589659,
      "grad_norm": 0.0179443359375,
      "learning_rate": 0.009882838283828382,
      "loss": 0.2356,
      "num_input_tokens_seen": 1263072,
      "step": 5990
    },
    {
      "epoch": 0.6595159515951595,
      "grad_norm": 0.0035247802734375,
      "learning_rate": 0.009891089108910891,
      "loss": 0.2335,
      "num_input_tokens_seen": 1264160,
      "step": 5995
    },
    {
      "epoch": 0.6600660066006601,
      "grad_norm": 0.014892578125,
      "learning_rate": 0.0098993399339934,
      "loss": 0.2312,
      "num_input_tokens_seen": 1265152,
      "step": 6000
    },
    {
      "epoch": 0.6606160616061606,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.009907590759075907,
      "loss": 0.2292,
      "num_input_tokens_seen": 1266208,
      "step": 6005
    },
    {
      "epoch": 0.6611661166116611,
      "grad_norm": 0.01519775390625,
      "learning_rate": 0.009915841584158416,
      "loss": 0.2303,
      "num_input_tokens_seen": 1267264,
      "step": 6010
    },
    {
      "epoch": 0.6617161716171617,
      "grad_norm": 0.007171630859375,
      "learning_rate": 0.009924092409240923,
      "loss": 0.2344,
      "num_input_tokens_seen": 1268320,
      "step": 6015
    },
    {
      "epoch": 0.6622662266226622,
      "grad_norm": 0.0284423828125,
      "learning_rate": 0.009932343234323432,
      "loss": 0.2312,
      "num_input_tokens_seen": 1269312,
      "step": 6020
    },
    {
      "epoch": 0.6628162816281629,
      "grad_norm": 0.0031280517578125,
      "learning_rate": 0.009940594059405939,
      "loss": 0.2335,
      "num_input_tokens_seen": 1270400,
      "step": 6025
    },
    {
      "epoch": 0.6633663366336634,
      "grad_norm": 0.0035552978515625,
      "learning_rate": 0.009948844884488448,
      "loss": 0.2294,
      "num_input_tokens_seen": 1271488,
      "step": 6030
    },
    {
      "epoch": 0.6639163916391639,
      "grad_norm": 0.0294189453125,
      "learning_rate": 0.009957095709570957,
      "loss": 0.2263,
      "num_input_tokens_seen": 1272544,
      "step": 6035
    },
    {
      "epoch": 0.6644664466446645,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.009965346534653466,
      "loss": 0.2294,
      "num_input_tokens_seen": 1273568,
      "step": 6040
    },
    {
      "epoch": 0.665016501650165,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.009973597359735974,
      "loss": 0.2296,
      "num_input_tokens_seen": 1274656,
      "step": 6045
    },
    {
      "epoch": 0.6655665566556656,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.009981848184818481,
      "loss": 0.238,
      "num_input_tokens_seen": 1275712,
      "step": 6050
    },
    {
      "epoch": 0.6661166116611661,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.00999009900990099,
      "loss": 0.2285,
      "num_input_tokens_seen": 1276736,
      "step": 6055
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.00335693359375,
      "learning_rate": 0.009998349834983497,
      "loss": 0.2324,
      "num_input_tokens_seen": 1277760,
      "step": 6060
    },
    {
      "epoch": 0.6672167216721672,
      "grad_norm": 0.0040283203125,
      "learning_rate": 0.010006600660066006,
      "loss": 0.2302,
      "num_input_tokens_seen": 1278816,
      "step": 6065
    },
    {
      "epoch": 0.6677667766776678,
      "grad_norm": 0.0155029296875,
      "learning_rate": 0.010014851485148513,
      "loss": 0.2334,
      "num_input_tokens_seen": 1279872,
      "step": 6070
    },
    {
      "epoch": 0.6683168316831684,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.010023102310231022,
      "loss": 0.2325,
      "num_input_tokens_seen": 1281024,
      "step": 6075
    },
    {
      "epoch": 0.6688668866886689,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.010031353135313531,
      "loss": 0.2304,
      "num_input_tokens_seen": 1282048,
      "step": 6080
    },
    {
      "epoch": 0.6694169416941694,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.01003960396039604,
      "loss": 0.2273,
      "num_input_tokens_seen": 1283104,
      "step": 6085
    },
    {
      "epoch": 0.66996699669967,
      "grad_norm": 0.01513671875,
      "learning_rate": 0.010047854785478549,
      "loss": 0.2335,
      "num_input_tokens_seen": 1284160,
      "step": 6090
    },
    {
      "epoch": 0.6705170517051705,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.010056105610561056,
      "loss": 0.2346,
      "num_input_tokens_seen": 1285152,
      "step": 6095
    },
    {
      "epoch": 0.671067106710671,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.010064356435643565,
      "loss": 0.2323,
      "num_input_tokens_seen": 1286240,
      "step": 6100
    },
    {
      "epoch": 0.6716171617161716,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.010072607260726072,
      "loss": 0.2313,
      "num_input_tokens_seen": 1287328,
      "step": 6105
    },
    {
      "epoch": 0.6721672167216721,
      "grad_norm": 0.004486083984375,
      "learning_rate": 0.01008085808580858,
      "loss": 0.2346,
      "num_input_tokens_seen": 1288416,
      "step": 6110
    },
    {
      "epoch": 0.6727172717271728,
      "grad_norm": 0.01470947265625,
      "learning_rate": 0.010089108910891088,
      "loss": 0.2264,
      "num_input_tokens_seen": 1289536,
      "step": 6115
    },
    {
      "epoch": 0.6732673267326733,
      "grad_norm": 0.01544189453125,
      "learning_rate": 0.010097359735973596,
      "loss": 0.2338,
      "num_input_tokens_seen": 1290592,
      "step": 6120
    },
    {
      "epoch": 0.6738173817381738,
      "grad_norm": 0.0264892578125,
      "learning_rate": 0.010105610561056104,
      "loss": 0.2336,
      "num_input_tokens_seen": 1291616,
      "step": 6125
    },
    {
      "epoch": 0.6743674367436744,
      "grad_norm": 0.01519775390625,
      "learning_rate": 0.010113861386138614,
      "loss": 0.2314,
      "num_input_tokens_seen": 1292672,
      "step": 6130
    },
    {
      "epoch": 0.6749174917491749,
      "grad_norm": 0.00445556640625,
      "learning_rate": 0.010122112211221123,
      "loss": 0.2293,
      "num_input_tokens_seen": 1293728,
      "step": 6135
    },
    {
      "epoch": 0.6754675467546755,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01013036303630363,
      "loss": 0.2282,
      "num_input_tokens_seen": 1294784,
      "step": 6140
    },
    {
      "epoch": 0.676017601760176,
      "grad_norm": 0.013671875,
      "learning_rate": 0.010138613861386139,
      "loss": 0.2313,
      "num_input_tokens_seen": 1295872,
      "step": 6145
    },
    {
      "epoch": 0.6765676567656765,
      "grad_norm": 0.013916015625,
      "learning_rate": 0.010146864686468646,
      "loss": 0.2334,
      "num_input_tokens_seen": 1296896,
      "step": 6150
    },
    {
      "epoch": 0.6771177117711771,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.010155115511551155,
      "loss": 0.2312,
      "num_input_tokens_seen": 1297952,
      "step": 6155
    },
    {
      "epoch": 0.6776677667766776,
      "grad_norm": 0.013671875,
      "learning_rate": 0.010163366336633662,
      "loss": 0.2315,
      "num_input_tokens_seen": 1298976,
      "step": 6160
    },
    {
      "epoch": 0.6782178217821783,
      "grad_norm": 0.0264892578125,
      "learning_rate": 0.01017161716171617,
      "loss": 0.2324,
      "num_input_tokens_seen": 1300096,
      "step": 6165
    },
    {
      "epoch": 0.6787678767876788,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.010179867986798678,
      "loss": 0.2336,
      "num_input_tokens_seen": 1301152,
      "step": 6170
    },
    {
      "epoch": 0.6793179317931793,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.010188118811881188,
      "loss": 0.2348,
      "num_input_tokens_seen": 1302176,
      "step": 6175
    },
    {
      "epoch": 0.6798679867986799,
      "grad_norm": 0.015625,
      "learning_rate": 0.010196369636963697,
      "loss": 0.2254,
      "num_input_tokens_seen": 1303232,
      "step": 6180
    },
    {
      "epoch": 0.6804180418041804,
      "grad_norm": 0.01458740234375,
      "learning_rate": 0.010204620462046204,
      "loss": 0.2326,
      "num_input_tokens_seen": 1304320,
      "step": 6185
    },
    {
      "epoch": 0.680968096809681,
      "grad_norm": 0.0264892578125,
      "learning_rate": 0.010212871287128713,
      "loss": 0.2326,
      "num_input_tokens_seen": 1305312,
      "step": 6190
    },
    {
      "epoch": 0.6815181518151815,
      "grad_norm": 0.014404296875,
      "learning_rate": 0.01022112211221122,
      "loss": 0.2314,
      "num_input_tokens_seen": 1306304,
      "step": 6195
    },
    {
      "epoch": 0.682068206820682,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.01022937293729373,
      "loss": 0.2324,
      "num_input_tokens_seen": 1307328,
      "step": 6200
    },
    {
      "epoch": 0.6826182618261826,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.010237623762376236,
      "loss": 0.2356,
      "num_input_tokens_seen": 1308384,
      "step": 6205
    },
    {
      "epoch": 0.6831683168316832,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.010245874587458745,
      "loss": 0.2335,
      "num_input_tokens_seen": 1309440,
      "step": 6210
    },
    {
      "epoch": 0.6837183718371838,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.010254125412541252,
      "loss": 0.2272,
      "num_input_tokens_seen": 1310528,
      "step": 6215
    },
    {
      "epoch": 0.6842684268426843,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.010262376237623763,
      "loss": 0.2324,
      "num_input_tokens_seen": 1311584,
      "step": 6220
    },
    {
      "epoch": 0.6848184818481848,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010270627062706272,
      "loss": 0.2335,
      "num_input_tokens_seen": 1312640,
      "step": 6225
    },
    {
      "epoch": 0.6853685368536854,
      "grad_norm": 0.0283203125,
      "learning_rate": 0.010278877887788779,
      "loss": 0.2357,
      "num_input_tokens_seen": 1313696,
      "step": 6230
    },
    {
      "epoch": 0.6859185918591859,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010287128712871288,
      "loss": 0.2252,
      "num_input_tokens_seen": 1314784,
      "step": 6235
    },
    {
      "epoch": 0.6864686468646864,
      "grad_norm": 0.0159912109375,
      "learning_rate": 0.010295379537953795,
      "loss": 0.2335,
      "num_input_tokens_seen": 1315872,
      "step": 6240
    },
    {
      "epoch": 0.687018701870187,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.010303630363036304,
      "loss": 0.2314,
      "num_input_tokens_seen": 1316960,
      "step": 6245
    },
    {
      "epoch": 0.6875687568756875,
      "grad_norm": 0.0147705078125,
      "learning_rate": 0.01031188118811881,
      "loss": 0.2315,
      "num_input_tokens_seen": 1318016,
      "step": 6250
    },
    {
      "epoch": 0.6881188118811881,
      "grad_norm": 0.0157470703125,
      "learning_rate": 0.01032013201320132,
      "loss": 0.2314,
      "num_input_tokens_seen": 1319072,
      "step": 6255
    },
    {
      "epoch": 0.6886688668866887,
      "grad_norm": 0.018310546875,
      "learning_rate": 0.010328382838283827,
      "loss": 0.2323,
      "num_input_tokens_seen": 1320128,
      "step": 6260
    },
    {
      "epoch": 0.6892189218921893,
      "grad_norm": 0.0137939453125,
      "learning_rate": 0.010336633663366335,
      "loss": 0.2313,
      "num_input_tokens_seen": 1321184,
      "step": 6265
    },
    {
      "epoch": 0.6897689768976898,
      "grad_norm": 0.0137939453125,
      "learning_rate": 0.010344884488448846,
      "loss": 0.2345,
      "num_input_tokens_seen": 1322240,
      "step": 6270
    },
    {
      "epoch": 0.6903190319031903,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.010353135313531353,
      "loss": 0.2293,
      "num_input_tokens_seen": 1323296,
      "step": 6275
    },
    {
      "epoch": 0.6908690869086909,
      "grad_norm": 0.0147705078125,
      "learning_rate": 0.010361386138613862,
      "loss": 0.2305,
      "num_input_tokens_seen": 1324320,
      "step": 6280
    },
    {
      "epoch": 0.6914191419141914,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.010369636963696369,
      "loss": 0.2316,
      "num_input_tokens_seen": 1325408,
      "step": 6285
    },
    {
      "epoch": 0.6919691969196919,
      "grad_norm": 0.017333984375,
      "learning_rate": 0.010377887788778878,
      "loss": 0.2317,
      "num_input_tokens_seen": 1326496,
      "step": 6290
    },
    {
      "epoch": 0.6925192519251925,
      "grad_norm": 0.016357421875,
      "learning_rate": 0.010386138613861385,
      "loss": 0.2341,
      "num_input_tokens_seen": 1327584,
      "step": 6295
    },
    {
      "epoch": 0.693069306930693,
      "grad_norm": 0.0260009765625,
      "learning_rate": 0.010394389438943894,
      "loss": 0.2317,
      "num_input_tokens_seen": 1328640,
      "step": 6300
    },
    {
      "epoch": 0.6936193619361937,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.010402640264026401,
      "loss": 0.2349,
      "num_input_tokens_seen": 1329728,
      "step": 6305
    },
    {
      "epoch": 0.6941694169416942,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.01041089108910891,
      "loss": 0.2277,
      "num_input_tokens_seen": 1330816,
      "step": 6310
    },
    {
      "epoch": 0.6947194719471947,
      "grad_norm": 0.01531982421875,
      "learning_rate": 0.01041914191419142,
      "loss": 0.2306,
      "num_input_tokens_seen": 1331872,
      "step": 6315
    },
    {
      "epoch": 0.6952695269526953,
      "grad_norm": 0.01397705078125,
      "learning_rate": 0.010427392739273927,
      "loss": 0.2233,
      "num_input_tokens_seen": 1332896,
      "step": 6320
    },
    {
      "epoch": 0.6958195819581958,
      "grad_norm": 0.0150146484375,
      "learning_rate": 0.010435643564356436,
      "loss": 0.2371,
      "num_input_tokens_seen": 1333952,
      "step": 6325
    },
    {
      "epoch": 0.6963696369636964,
      "grad_norm": 0.01409912109375,
      "learning_rate": 0.010443894389438943,
      "loss": 0.2325,
      "num_input_tokens_seen": 1335040,
      "step": 6330
    },
    {
      "epoch": 0.6969196919691969,
      "grad_norm": 0.0135498046875,
      "learning_rate": 0.010452145214521452,
      "loss": 0.2275,
      "num_input_tokens_seen": 1336096,
      "step": 6335
    },
    {
      "epoch": 0.6974697469746974,
      "grad_norm": 0.016845703125,
      "learning_rate": 0.01046039603960396,
      "loss": 0.2287,
      "num_input_tokens_seen": 1337152,
      "step": 6340
    },
    {
      "epoch": 0.698019801980198,
      "grad_norm": 0.00396728515625,
      "learning_rate": 0.010468646864686468,
      "loss": 0.2298,
      "num_input_tokens_seen": 1338272,
      "step": 6345
    },
    {
      "epoch": 0.6985698569856986,
      "grad_norm": 0.01470947265625,
      "learning_rate": 0.010476897689768975,
      "loss": 0.239,
      "num_input_tokens_seen": 1339296,
      "step": 6350
    },
    {
      "epoch": 0.6991199119911992,
      "grad_norm": 0.013671875,
      "learning_rate": 0.010485148514851484,
      "loss": 0.2265,
      "num_input_tokens_seen": 1340320,
      "step": 6355
    },
    {
      "epoch": 0.6996699669966997,
      "grad_norm": 0.01416015625,
      "learning_rate": 0.010493399339933995,
      "loss": 0.2337,
      "num_input_tokens_seen": 1341408,
      "step": 6360
    },
    {
      "epoch": 0.7002200220022002,
      "grad_norm": 0.0257568359375,
      "learning_rate": 0.010501650165016502,
      "loss": 0.2276,
      "num_input_tokens_seen": 1342432,
      "step": 6365
    },
    {
      "epoch": 0.7007700770077008,
      "grad_norm": 0.004150390625,
      "learning_rate": 0.01050990099009901,
      "loss": 0.2306,
      "num_input_tokens_seen": 1343456,
      "step": 6370
    },
    {
      "epoch": 0.7013201320132013,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.010518151815181518,
      "loss": 0.2326,
      "num_input_tokens_seen": 1344512,
      "step": 6375
    },
    {
      "epoch": 0.7018701870187019,
      "grad_norm": 0.01446533203125,
      "learning_rate": 0.010526402640264027,
      "loss": 0.2337,
      "num_input_tokens_seen": 1345504,
      "step": 6380
    },
    {
      "epoch": 0.7024202420242024,
      "grad_norm": 0.01361083984375,
      "learning_rate": 0.010534653465346534,
      "loss": 0.2327,
      "num_input_tokens_seen": 1346560,
      "step": 6385
    },
    {
      "epoch": 0.7029702970297029,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.010542904290429042,
      "loss": 0.2296,
      "num_input_tokens_seen": 1347648,
      "step": 6390
    },
    {
      "epoch": 0.7035203520352035,
      "grad_norm": 0.003875732421875,
      "learning_rate": 0.01055115511551155,
      "loss": 0.2298,
      "num_input_tokens_seen": 1348704,
      "step": 6395
    },
    {
      "epoch": 0.7040704070407041,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.010559405940594058,
      "loss": 0.2339,
      "num_input_tokens_seen": 1349824,
      "step": 6400
    },
    {
      "epoch": 0.7046204620462047,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.010567656765676567,
      "loss": 0.2272,
      "num_input_tokens_seen": 1350816,
      "step": 6405
    },
    {
      "epoch": 0.7051705170517052,
      "grad_norm": 0.0264892578125,
      "learning_rate": 0.010575907590759076,
      "loss": 0.2268,
      "num_input_tokens_seen": 1351840,
      "step": 6410
    },
    {
      "epoch": 0.7057205720572057,
      "grad_norm": 0.016845703125,
      "learning_rate": 0.010584158415841585,
      "loss": 0.2322,
      "num_input_tokens_seen": 1352896,
      "step": 6415
    },
    {
      "epoch": 0.7062706270627063,
      "grad_norm": 0.01422119140625,
      "learning_rate": 0.010592409240924092,
      "loss": 0.2253,
      "num_input_tokens_seen": 1354016,
      "step": 6420
    },
    {
      "epoch": 0.7068206820682068,
      "grad_norm": 0.01239013671875,
      "learning_rate": 0.0106006600660066,
      "loss": 0.2432,
      "num_input_tokens_seen": 1355104,
      "step": 6425
    },
    {
      "epoch": 0.7073707370737073,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.010608910891089108,
      "loss": 0.2405,
      "num_input_tokens_seen": 1356128,
      "step": 6430
    },
    {
      "epoch": 0.7079207920792079,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.010617161716171617,
      "loss": 0.2297,
      "num_input_tokens_seen": 1357184,
      "step": 6435
    },
    {
      "epoch": 0.7084708470847084,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.010625412541254124,
      "loss": 0.2317,
      "num_input_tokens_seen": 1358208,
      "step": 6440
    },
    {
      "epoch": 0.7090209020902091,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.010633663366336633,
      "loss": 0.2294,
      "num_input_tokens_seen": 1359232,
      "step": 6445
    },
    {
      "epoch": 0.7095709570957096,
      "grad_norm": 0.0123291015625,
      "learning_rate": 0.010641914191419142,
      "loss": 0.2314,
      "num_input_tokens_seen": 1360288,
      "step": 6450
    },
    {
      "epoch": 0.7101210121012101,
      "grad_norm": 0.003997802734375,
      "learning_rate": 0.01065016501650165,
      "loss": 0.2324,
      "num_input_tokens_seen": 1361344,
      "step": 6455
    },
    {
      "epoch": 0.7106710671067107,
      "grad_norm": 0.00421142578125,
      "learning_rate": 0.01065841584158416,
      "loss": 0.2345,
      "num_input_tokens_seen": 1362432,
      "step": 6460
    },
    {
      "epoch": 0.7112211221122112,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.010666666666666666,
      "loss": 0.2325,
      "num_input_tokens_seen": 1363488,
      "step": 6465
    },
    {
      "epoch": 0.7117711771177118,
      "grad_norm": 0.00372314453125,
      "learning_rate": 0.010674917491749175,
      "loss": 0.2345,
      "num_input_tokens_seen": 1364608,
      "step": 6470
    },
    {
      "epoch": 0.7123212321232123,
      "grad_norm": 0.00445556640625,
      "learning_rate": 0.010683168316831682,
      "loss": 0.2314,
      "num_input_tokens_seen": 1365600,
      "step": 6475
    },
    {
      "epoch": 0.7128712871287128,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.010691419141914191,
      "loss": 0.2356,
      "num_input_tokens_seen": 1366624,
      "step": 6480
    },
    {
      "epoch": 0.7134213421342134,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.010699669966996698,
      "loss": 0.2336,
      "num_input_tokens_seen": 1367616,
      "step": 6485
    },
    {
      "epoch": 0.713971397139714,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.010707920792079207,
      "loss": 0.2314,
      "num_input_tokens_seen": 1368736,
      "step": 6490
    },
    {
      "epoch": 0.7145214521452146,
      "grad_norm": 0.01385498046875,
      "learning_rate": 0.010716171617161716,
      "loss": 0.2336,
      "num_input_tokens_seen": 1369792,
      "step": 6495
    },
    {
      "epoch": 0.7150715071507151,
      "grad_norm": 0.02685546875,
      "learning_rate": 0.010724422442244225,
      "loss": 0.2335,
      "num_input_tokens_seen": 1370880,
      "step": 6500
    },
    {
      "epoch": 0.7156215621562156,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010732673267326734,
      "loss": 0.2294,
      "num_input_tokens_seen": 1371968,
      "step": 6505
    },
    {
      "epoch": 0.7161716171617162,
      "grad_norm": 0.01300048828125,
      "learning_rate": 0.01074092409240924,
      "loss": 0.2283,
      "num_input_tokens_seen": 1373088,
      "step": 6510
    },
    {
      "epoch": 0.7167216721672167,
      "grad_norm": 0.004547119140625,
      "learning_rate": 0.01074917491749175,
      "loss": 0.2334,
      "num_input_tokens_seen": 1374112,
      "step": 6515
    },
    {
      "epoch": 0.7172717271727173,
      "grad_norm": 0.025146484375,
      "learning_rate": 0.010757425742574257,
      "loss": 0.2283,
      "num_input_tokens_seen": 1375200,
      "step": 6520
    },
    {
      "epoch": 0.7178217821782178,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.010765676567656765,
      "loss": 0.2293,
      "num_input_tokens_seen": 1376224,
      "step": 6525
    },
    {
      "epoch": 0.7183718371837183,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.010773927392739273,
      "loss": 0.2315,
      "num_input_tokens_seen": 1377344,
      "step": 6530
    },
    {
      "epoch": 0.7189218921892189,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.010782178217821781,
      "loss": 0.2245,
      "num_input_tokens_seen": 1378400,
      "step": 6535
    },
    {
      "epoch": 0.7194719471947195,
      "grad_norm": 0.0164794921875,
      "learning_rate": 0.01079042904290429,
      "loss": 0.2251,
      "num_input_tokens_seen": 1379520,
      "step": 6540
    },
    {
      "epoch": 0.7200220022002201,
      "grad_norm": 0.007171630859375,
      "learning_rate": 0.010798679867986797,
      "loss": 0.2323,
      "num_input_tokens_seen": 1380544,
      "step": 6545
    },
    {
      "epoch": 0.7205720572057206,
      "grad_norm": 0.015625,
      "learning_rate": 0.010806930693069308,
      "loss": 0.214,
      "num_input_tokens_seen": 1381600,
      "step": 6550
    },
    {
      "epoch": 0.7211221122112211,
      "grad_norm": 0.023681640625,
      "learning_rate": 0.010815181518151815,
      "loss": 0.244,
      "num_input_tokens_seen": 1382688,
      "step": 6555
    },
    {
      "epoch": 0.7216721672167217,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.010823432343234324,
      "loss": 0.2219,
      "num_input_tokens_seen": 1383776,
      "step": 6560
    },
    {
      "epoch": 0.7222222222222222,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.010831683168316831,
      "loss": 0.2374,
      "num_input_tokens_seen": 1384736,
      "step": 6565
    },
    {
      "epoch": 0.7227722772277227,
      "grad_norm": 0.0152587890625,
      "learning_rate": 0.01083993399339934,
      "loss": 0.2144,
      "num_input_tokens_seen": 1385760,
      "step": 6570
    },
    {
      "epoch": 0.7233223322332233,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.010848184818481847,
      "loss": 0.2374,
      "num_input_tokens_seen": 1386816,
      "step": 6575
    },
    {
      "epoch": 0.7238723872387238,
      "grad_norm": 0.0257568359375,
      "learning_rate": 0.010856435643564356,
      "loss": 0.2272,
      "num_input_tokens_seen": 1387872,
      "step": 6580
    },
    {
      "epoch": 0.7244224422442245,
      "grad_norm": 0.0159912109375,
      "learning_rate": 0.010864686468646865,
      "loss": 0.2377,
      "num_input_tokens_seen": 1388896,
      "step": 6585
    },
    {
      "epoch": 0.724972497249725,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.010872937293729372,
      "loss": 0.235,
      "num_input_tokens_seen": 1390016,
      "step": 6590
    },
    {
      "epoch": 0.7255225522552256,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.010881188118811882,
      "loss": 0.2364,
      "num_input_tokens_seen": 1391072,
      "step": 6595
    },
    {
      "epoch": 0.7260726072607261,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.01088943894389439,
      "loss": 0.2188,
      "num_input_tokens_seen": 1392128,
      "step": 6600
    },
    {
      "epoch": 0.7266226622662266,
      "grad_norm": 0.00421142578125,
      "learning_rate": 0.010897689768976898,
      "loss": 0.2527,
      "num_input_tokens_seen": 1393216,
      "step": 6605
    },
    {
      "epoch": 0.7271727172717272,
      "grad_norm": 0.004486083984375,
      "learning_rate": 0.010905940594059405,
      "loss": 0.2309,
      "num_input_tokens_seen": 1394304,
      "step": 6610
    },
    {
      "epoch": 0.7277227722772277,
      "grad_norm": 0.0038299560546875,
      "learning_rate": 0.010914191419141914,
      "loss": 0.2338,
      "num_input_tokens_seen": 1395392,
      "step": 6615
    },
    {
      "epoch": 0.7282728272827282,
      "grad_norm": 0.00433349609375,
      "learning_rate": 0.010922442244224421,
      "loss": 0.2265,
      "num_input_tokens_seen": 1396480,
      "step": 6620
    },
    {
      "epoch": 0.7288228822882288,
      "grad_norm": 0.0172119140625,
      "learning_rate": 0.01093069306930693,
      "loss": 0.2316,
      "num_input_tokens_seen": 1397536,
      "step": 6625
    },
    {
      "epoch": 0.7293729372937293,
      "grad_norm": 0.0034942626953125,
      "learning_rate": 0.010938943894389439,
      "loss": 0.2317,
      "num_input_tokens_seen": 1398624,
      "step": 6630
    },
    {
      "epoch": 0.72992299229923,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010947194719471946,
      "loss": 0.2348,
      "num_input_tokens_seen": 1399712,
      "step": 6635
    },
    {
      "epoch": 0.7304730473047305,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.010955445544554455,
      "loss": 0.24,
      "num_input_tokens_seen": 1400768,
      "step": 6640
    },
    {
      "epoch": 0.731023102310231,
      "grad_norm": 0.03369140625,
      "learning_rate": 0.010963696369636964,
      "loss": 0.2344,
      "num_input_tokens_seen": 1401824,
      "step": 6645
    },
    {
      "epoch": 0.7315731573157316,
      "grad_norm": 0.021484375,
      "learning_rate": 0.010971947194719472,
      "loss": 0.2335,
      "num_input_tokens_seen": 1402912,
      "step": 6650
    },
    {
      "epoch": 0.7321232123212321,
      "grad_norm": 0.0189208984375,
      "learning_rate": 0.01098019801980198,
      "loss": 0.2303,
      "num_input_tokens_seen": 1403904,
      "step": 6655
    },
    {
      "epoch": 0.7326732673267327,
      "grad_norm": 0.01953125,
      "learning_rate": 0.010988448844884488,
      "loss": 0.2323,
      "num_input_tokens_seen": 1404992,
      "step": 6660
    },
    {
      "epoch": 0.7332233223322332,
      "grad_norm": 0.0198974609375,
      "learning_rate": 0.010996699669966995,
      "loss": 0.2344,
      "num_input_tokens_seen": 1406080,
      "step": 6665
    },
    {
      "epoch": 0.7337733773377337,
      "grad_norm": 0.0035858154296875,
      "learning_rate": 0.011004950495049504,
      "loss": 0.2322,
      "num_input_tokens_seen": 1407136,
      "step": 6670
    },
    {
      "epoch": 0.7343234323432343,
      "grad_norm": 0.015380859375,
      "learning_rate": 0.011013201320132013,
      "loss": 0.2291,
      "num_input_tokens_seen": 1408192,
      "step": 6675
    },
    {
      "epoch": 0.7348734873487349,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.01102145214521452,
      "loss": 0.2302,
      "num_input_tokens_seen": 1409184,
      "step": 6680
    },
    {
      "epoch": 0.7354235423542355,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011029702970297029,
      "loss": 0.2323,
      "num_input_tokens_seen": 1410240,
      "step": 6685
    },
    {
      "epoch": 0.735973597359736,
      "grad_norm": 0.015625,
      "learning_rate": 0.011037953795379538,
      "loss": 0.2293,
      "num_input_tokens_seen": 1411296,
      "step": 6690
    },
    {
      "epoch": 0.7365236523652365,
      "grad_norm": 0.01708984375,
      "learning_rate": 0.011046204620462047,
      "loss": 0.2324,
      "num_input_tokens_seen": 1412320,
      "step": 6695
    },
    {
      "epoch": 0.7370737073707371,
      "grad_norm": 0.014892578125,
      "learning_rate": 0.011054455445544554,
      "loss": 0.2325,
      "num_input_tokens_seen": 1413376,
      "step": 6700
    },
    {
      "epoch": 0.7376237623762376,
      "grad_norm": 0.01556396484375,
      "learning_rate": 0.011062706270627063,
      "loss": 0.2302,
      "num_input_tokens_seen": 1414368,
      "step": 6705
    },
    {
      "epoch": 0.7381738173817382,
      "grad_norm": 0.01544189453125,
      "learning_rate": 0.01107095709570957,
      "loss": 0.2312,
      "num_input_tokens_seen": 1415456,
      "step": 6710
    },
    {
      "epoch": 0.7387238723872387,
      "grad_norm": 0.00335693359375,
      "learning_rate": 0.011079207920792079,
      "loss": 0.2291,
      "num_input_tokens_seen": 1416576,
      "step": 6715
    },
    {
      "epoch": 0.7392739273927392,
      "grad_norm": 0.01544189453125,
      "learning_rate": 0.011087458745874587,
      "loss": 0.2322,
      "num_input_tokens_seen": 1417632,
      "step": 6720
    },
    {
      "epoch": 0.7398239823982399,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.011095709570957095,
      "loss": 0.2301,
      "num_input_tokens_seen": 1418592,
      "step": 6725
    },
    {
      "epoch": 0.7403740374037404,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.011103960396039603,
      "loss": 0.2323,
      "num_input_tokens_seen": 1419712,
      "step": 6730
    },
    {
      "epoch": 0.740924092409241,
      "grad_norm": 0.0157470703125,
      "learning_rate": 0.011112211221122112,
      "loss": 0.2333,
      "num_input_tokens_seen": 1420832,
      "step": 6735
    },
    {
      "epoch": 0.7414741474147415,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.011120462046204621,
      "loss": 0.2322,
      "num_input_tokens_seen": 1421920,
      "step": 6740
    },
    {
      "epoch": 0.742024202420242,
      "grad_norm": 0.0169677734375,
      "learning_rate": 0.011128712871287128,
      "loss": 0.2312,
      "num_input_tokens_seen": 1423008,
      "step": 6745
    },
    {
      "epoch": 0.7425742574257426,
      "grad_norm": 0.0042724609375,
      "learning_rate": 0.011136963696369637,
      "loss": 0.2292,
      "num_input_tokens_seen": 1424000,
      "step": 6750
    },
    {
      "epoch": 0.7431243124312431,
      "grad_norm": 0.0218505859375,
      "learning_rate": 0.011145214521452144,
      "loss": 0.2345,
      "num_input_tokens_seen": 1424992,
      "step": 6755
    },
    {
      "epoch": 0.7436743674367436,
      "grad_norm": 0.0186767578125,
      "learning_rate": 0.011153465346534653,
      "loss": 0.2284,
      "num_input_tokens_seen": 1426016,
      "step": 6760
    },
    {
      "epoch": 0.7442244224422442,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.011161716171617162,
      "loss": 0.2286,
      "num_input_tokens_seen": 1427008,
      "step": 6765
    },
    {
      "epoch": 0.7447744774477447,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.011169966996699669,
      "loss": 0.2333,
      "num_input_tokens_seen": 1428096,
      "step": 6770
    },
    {
      "epoch": 0.7453245324532454,
      "grad_norm": 0.049072265625,
      "learning_rate": 0.011178217821782178,
      "loss": 0.2343,
      "num_input_tokens_seen": 1429216,
      "step": 6775
    },
    {
      "epoch": 0.7458745874587459,
      "grad_norm": 0.0218505859375,
      "learning_rate": 0.011186468646864685,
      "loss": 0.2246,
      "num_input_tokens_seen": 1430272,
      "step": 6780
    },
    {
      "epoch": 0.7464246424642464,
      "grad_norm": 0.0263671875,
      "learning_rate": 0.011194719471947195,
      "loss": 0.2373,
      "num_input_tokens_seen": 1431296,
      "step": 6785
    },
    {
      "epoch": 0.746974697469747,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.011202970297029703,
      "loss": 0.2337,
      "num_input_tokens_seen": 1432384,
      "step": 6790
    },
    {
      "epoch": 0.7475247524752475,
      "grad_norm": 0.016357421875,
      "learning_rate": 0.011211221122112211,
      "loss": 0.2334,
      "num_input_tokens_seen": 1433440,
      "step": 6795
    },
    {
      "epoch": 0.7480748074807481,
      "grad_norm": 0.0291748046875,
      "learning_rate": 0.011219471947194718,
      "loss": 0.2313,
      "num_input_tokens_seen": 1434432,
      "step": 6800
    },
    {
      "epoch": 0.7486248624862486,
      "grad_norm": 0.01361083984375,
      "learning_rate": 0.011227722772277227,
      "loss": 0.2314,
      "num_input_tokens_seen": 1435488,
      "step": 6805
    },
    {
      "epoch": 0.7491749174917491,
      "grad_norm": 0.0034942626953125,
      "learning_rate": 0.011235973597359736,
      "loss": 0.2337,
      "num_input_tokens_seen": 1436512,
      "step": 6810
    },
    {
      "epoch": 0.7497249724972497,
      "grad_norm": 0.01324462890625,
      "learning_rate": 0.011244224422442243,
      "loss": 0.237,
      "num_input_tokens_seen": 1437504,
      "step": 6815
    },
    {
      "epoch": 0.7502750275027503,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.011252475247524752,
      "loss": 0.2306,
      "num_input_tokens_seen": 1438592,
      "step": 6820
    },
    {
      "epoch": 0.7508250825082509,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01126072607260726,
      "loss": 0.2285,
      "num_input_tokens_seen": 1439648,
      "step": 6825
    },
    {
      "epoch": 0.7513751375137514,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01126897689768977,
      "loss": 0.2276,
      "num_input_tokens_seen": 1440640,
      "step": 6830
    },
    {
      "epoch": 0.7519251925192519,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.011277227722772277,
      "loss": 0.2288,
      "num_input_tokens_seen": 1441696,
      "step": 6835
    },
    {
      "epoch": 0.7524752475247525,
      "grad_norm": 0.025634765625,
      "learning_rate": 0.011285478547854786,
      "loss": 0.2308,
      "num_input_tokens_seen": 1442720,
      "step": 6840
    },
    {
      "epoch": 0.753025302530253,
      "grad_norm": 0.0030517578125,
      "learning_rate": 0.011293729372937293,
      "loss": 0.2297,
      "num_input_tokens_seen": 1443744,
      "step": 6845
    },
    {
      "epoch": 0.7535753575357536,
      "grad_norm": 0.003936767578125,
      "learning_rate": 0.011301980198019802,
      "loss": 0.2329,
      "num_input_tokens_seen": 1444928,
      "step": 6850
    },
    {
      "epoch": 0.7541254125412541,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.01131023102310231,
      "loss": 0.232,
      "num_input_tokens_seen": 1446048,
      "step": 6855
    },
    {
      "epoch": 0.7546754675467546,
      "grad_norm": 0.013916015625,
      "learning_rate": 0.011318481848184818,
      "loss": 0.2292,
      "num_input_tokens_seen": 1447040,
      "step": 6860
    },
    {
      "epoch": 0.7552255225522553,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.011326732673267326,
      "loss": 0.2355,
      "num_input_tokens_seen": 1448096,
      "step": 6865
    },
    {
      "epoch": 0.7557755775577558,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.011334983498349833,
      "loss": 0.2215,
      "num_input_tokens_seen": 1449184,
      "step": 6870
    },
    {
      "epoch": 0.7563256325632564,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.011343234323432344,
      "loss": 0.2342,
      "num_input_tokens_seen": 1450240,
      "step": 6875
    },
    {
      "epoch": 0.7568756875687569,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.011351485148514851,
      "loss": 0.2327,
      "num_input_tokens_seen": 1451296,
      "step": 6880
    },
    {
      "epoch": 0.7574257425742574,
      "grad_norm": 0.0157470703125,
      "learning_rate": 0.01135973597359736,
      "loss": 0.2305,
      "num_input_tokens_seen": 1452352,
      "step": 6885
    },
    {
      "epoch": 0.757975797579758,
      "grad_norm": 0.0291748046875,
      "learning_rate": 0.011367986798679867,
      "loss": 0.2343,
      "num_input_tokens_seen": 1453408,
      "step": 6890
    },
    {
      "epoch": 0.7585258525852585,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.011376237623762376,
      "loss": 0.2325,
      "num_input_tokens_seen": 1454592,
      "step": 6895
    },
    {
      "epoch": 0.759075907590759,
      "grad_norm": 0.02587890625,
      "learning_rate": 0.011384488448844885,
      "loss": 0.2276,
      "num_input_tokens_seen": 1455584,
      "step": 6900
    },
    {
      "epoch": 0.7596259625962596,
      "grad_norm": 0.007598876953125,
      "learning_rate": 0.011392739273927392,
      "loss": 0.2293,
      "num_input_tokens_seen": 1456608,
      "step": 6905
    },
    {
      "epoch": 0.7601760176017601,
      "grad_norm": 0.0029296875,
      "learning_rate": 0.0114009900990099,
      "loss": 0.2371,
      "num_input_tokens_seen": 1457760,
      "step": 6910
    },
    {
      "epoch": 0.7607260726072608,
      "grad_norm": 0.0257568359375,
      "learning_rate": 0.011409240924092408,
      "loss": 0.2257,
      "num_input_tokens_seen": 1458752,
      "step": 6915
    },
    {
      "epoch": 0.7612761276127613,
      "grad_norm": 0.0289306640625,
      "learning_rate": 0.011417491749174917,
      "loss": 0.2356,
      "num_input_tokens_seen": 1459808,
      "step": 6920
    },
    {
      "epoch": 0.7618261826182618,
      "grad_norm": 0.0128173828125,
      "learning_rate": 0.011425742574257425,
      "loss": 0.2266,
      "num_input_tokens_seen": 1460832,
      "step": 6925
    },
    {
      "epoch": 0.7623762376237624,
      "grad_norm": 0.0150146484375,
      "learning_rate": 0.011433993399339934,
      "loss": 0.234,
      "num_input_tokens_seen": 1461888,
      "step": 6930
    },
    {
      "epoch": 0.7629262926292629,
      "grad_norm": 0.0252685546875,
      "learning_rate": 0.011442244224422441,
      "loss": 0.227,
      "num_input_tokens_seen": 1462912,
      "step": 6935
    },
    {
      "epoch": 0.7634763476347635,
      "grad_norm": 0.013916015625,
      "learning_rate": 0.01145049504950495,
      "loss": 0.2287,
      "num_input_tokens_seen": 1463936,
      "step": 6940
    },
    {
      "epoch": 0.764026402640264,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.011458745874587459,
      "loss": 0.2308,
      "num_input_tokens_seen": 1464992,
      "step": 6945
    },
    {
      "epoch": 0.7645764576457645,
      "grad_norm": 0.026123046875,
      "learning_rate": 0.011466996699669966,
      "loss": 0.2243,
      "num_input_tokens_seen": 1466048,
      "step": 6950
    },
    {
      "epoch": 0.7651265126512651,
      "grad_norm": 0.0172119140625,
      "learning_rate": 0.011475247524752475,
      "loss": 0.2386,
      "num_input_tokens_seen": 1467072,
      "step": 6955
    },
    {
      "epoch": 0.7656765676567657,
      "grad_norm": 0.014404296875,
      "learning_rate": 0.011483498349834982,
      "loss": 0.2324,
      "num_input_tokens_seen": 1468064,
      "step": 6960
    },
    {
      "epoch": 0.7662266226622663,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.011491749174917491,
      "loss": 0.2238,
      "num_input_tokens_seen": 1469152,
      "step": 6965
    },
    {
      "epoch": 0.7667766776677668,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.0115,
      "loss": 0.2301,
      "num_input_tokens_seen": 1470208,
      "step": 6970
    },
    {
      "epoch": 0.7673267326732673,
      "grad_norm": 0.00433349609375,
      "learning_rate": 0.011508250825082509,
      "loss": 0.2309,
      "num_input_tokens_seen": 1471296,
      "step": 6975
    },
    {
      "epoch": 0.7678767876787679,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.011516501650165016,
      "loss": 0.2312,
      "num_input_tokens_seen": 1472384,
      "step": 6980
    },
    {
      "epoch": 0.7684268426842684,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.011524752475247525,
      "loss": 0.2291,
      "num_input_tokens_seen": 1473440,
      "step": 6985
    },
    {
      "epoch": 0.768976897689769,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.011533003300330033,
      "loss": 0.2348,
      "num_input_tokens_seen": 1474464,
      "step": 6990
    },
    {
      "epoch": 0.7695269526952695,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.01154125412541254,
      "loss": 0.2284,
      "num_input_tokens_seen": 1475520,
      "step": 6995
    },
    {
      "epoch": 0.77007700770077,
      "grad_norm": 0.0203857421875,
      "learning_rate": 0.01154950495049505,
      "loss": 0.239,
      "num_input_tokens_seen": 1476544,
      "step": 7000
    },
    {
      "epoch": 0.7706270627062707,
      "grad_norm": 0.01513671875,
      "learning_rate": 0.011557755775577556,
      "loss": 0.2333,
      "num_input_tokens_seen": 1477600,
      "step": 7005
    },
    {
      "epoch": 0.7711771177117712,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.011566006600660065,
      "loss": 0.2267,
      "num_input_tokens_seen": 1478624,
      "step": 7010
    },
    {
      "epoch": 0.7717271727172718,
      "grad_norm": 0.0040283203125,
      "learning_rate": 0.011574257425742572,
      "loss": 0.227,
      "num_input_tokens_seen": 1479712,
      "step": 7015
    },
    {
      "epoch": 0.7722772277227723,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.011582508250825083,
      "loss": 0.2372,
      "num_input_tokens_seen": 1480768,
      "step": 7020
    },
    {
      "epoch": 0.7728272827282728,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.01159075907590759,
      "loss": 0.2277,
      "num_input_tokens_seen": 1481856,
      "step": 7025
    },
    {
      "epoch": 0.7733773377337734,
      "grad_norm": 0.015869140625,
      "learning_rate": 0.011599009900990099,
      "loss": 0.233,
      "num_input_tokens_seen": 1482912,
      "step": 7030
    },
    {
      "epoch": 0.7739273927392739,
      "grad_norm": 0.0284423828125,
      "learning_rate": 0.011607260726072608,
      "loss": 0.2277,
      "num_input_tokens_seen": 1483936,
      "step": 7035
    },
    {
      "epoch": 0.7744774477447744,
      "grad_norm": 0.0172119140625,
      "learning_rate": 0.011615511551155115,
      "loss": 0.2239,
      "num_input_tokens_seen": 1484960,
      "step": 7040
    },
    {
      "epoch": 0.775027502750275,
      "grad_norm": 0.031005859375,
      "learning_rate": 0.011623762376237624,
      "loss": 0.2424,
      "num_input_tokens_seen": 1485984,
      "step": 7045
    },
    {
      "epoch": 0.7755775577557755,
      "grad_norm": 0.01397705078125,
      "learning_rate": 0.01163201320132013,
      "loss": 0.2282,
      "num_input_tokens_seen": 1487040,
      "step": 7050
    },
    {
      "epoch": 0.7761276127612762,
      "grad_norm": 0.0255126953125,
      "learning_rate": 0.01164026402640264,
      "loss": 0.228,
      "num_input_tokens_seen": 1488096,
      "step": 7055
    },
    {
      "epoch": 0.7766776677667767,
      "grad_norm": 0.016845703125,
      "learning_rate": 0.011648514851485147,
      "loss": 0.2321,
      "num_input_tokens_seen": 1489120,
      "step": 7060
    },
    {
      "epoch": 0.7772277227722773,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.011656765676567657,
      "loss": 0.2312,
      "num_input_tokens_seen": 1490208,
      "step": 7065
    },
    {
      "epoch": 0.7777777777777778,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.011665016501650164,
      "loss": 0.2314,
      "num_input_tokens_seen": 1491232,
      "step": 7070
    },
    {
      "epoch": 0.7783278327832783,
      "grad_norm": 0.015869140625,
      "learning_rate": 0.011673267326732673,
      "loss": 0.2378,
      "num_input_tokens_seen": 1492320,
      "step": 7075
    },
    {
      "epoch": 0.7788778877887789,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.011681518151815182,
      "loss": 0.2384,
      "num_input_tokens_seen": 1493344,
      "step": 7080
    },
    {
      "epoch": 0.7794279427942794,
      "grad_norm": 0.0142822265625,
      "learning_rate": 0.01168976897689769,
      "loss": 0.2287,
      "num_input_tokens_seen": 1494400,
      "step": 7085
    },
    {
      "epoch": 0.7799779977997799,
      "grad_norm": 0.01409912109375,
      "learning_rate": 0.011698019801980198,
      "loss": 0.2297,
      "num_input_tokens_seen": 1495456,
      "step": 7090
    },
    {
      "epoch": 0.7805280528052805,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.011706270627062705,
      "loss": 0.2392,
      "num_input_tokens_seen": 1496480,
      "step": 7095
    },
    {
      "epoch": 0.7810781078107811,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.011714521452145214,
      "loss": 0.2286,
      "num_input_tokens_seen": 1497504,
      "step": 7100
    },
    {
      "epoch": 0.7816281628162817,
      "grad_norm": 0.0031890869140625,
      "learning_rate": 0.011722772277227721,
      "loss": 0.2306,
      "num_input_tokens_seen": 1498560,
      "step": 7105
    },
    {
      "epoch": 0.7821782178217822,
      "grad_norm": 0.01434326171875,
      "learning_rate": 0.011731023102310232,
      "loss": 0.2279,
      "num_input_tokens_seen": 1499552,
      "step": 7110
    },
    {
      "epoch": 0.7827282728272827,
      "grad_norm": 0.01202392578125,
      "learning_rate": 0.011739273927392739,
      "loss": 0.2305,
      "num_input_tokens_seen": 1500576,
      "step": 7115
    },
    {
      "epoch": 0.7832783278327833,
      "grad_norm": 0.0135498046875,
      "learning_rate": 0.011747524752475248,
      "loss": 0.2342,
      "num_input_tokens_seen": 1501632,
      "step": 7120
    },
    {
      "epoch": 0.7838283828382838,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011755775577557756,
      "loss": 0.2336,
      "num_input_tokens_seen": 1502656,
      "step": 7125
    },
    {
      "epoch": 0.7843784378437844,
      "grad_norm": 0.01513671875,
      "learning_rate": 0.011764026402640263,
      "loss": 0.2284,
      "num_input_tokens_seen": 1503712,
      "step": 7130
    },
    {
      "epoch": 0.7849284928492849,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.011772277227722772,
      "loss": 0.231,
      "num_input_tokens_seen": 1504832,
      "step": 7135
    },
    {
      "epoch": 0.7854785478547854,
      "grad_norm": 0.026123046875,
      "learning_rate": 0.01178052805280528,
      "loss": 0.2368,
      "num_input_tokens_seen": 1505856,
      "step": 7140
    },
    {
      "epoch": 0.786028602860286,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.011788778877887788,
      "loss": 0.2315,
      "num_input_tokens_seen": 1506912,
      "step": 7145
    },
    {
      "epoch": 0.7865786578657866,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.011797029702970295,
      "loss": 0.2308,
      "num_input_tokens_seen": 1508032,
      "step": 7150
    },
    {
      "epoch": 0.7871287128712872,
      "grad_norm": 0.0042724609375,
      "learning_rate": 0.011805280528052804,
      "loss": 0.2309,
      "num_input_tokens_seen": 1509056,
      "step": 7155
    },
    {
      "epoch": 0.7876787678767877,
      "grad_norm": 0.0264892578125,
      "learning_rate": 0.011813531353135313,
      "loss": 0.2314,
      "num_input_tokens_seen": 1510112,
      "step": 7160
    },
    {
      "epoch": 0.7882288228822882,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.011821782178217822,
      "loss": 0.2289,
      "num_input_tokens_seen": 1511168,
      "step": 7165
    },
    {
      "epoch": 0.7887788778877888,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.01183003300330033,
      "loss": 0.2317,
      "num_input_tokens_seen": 1512256,
      "step": 7170
    },
    {
      "epoch": 0.7893289328932893,
      "grad_norm": 0.0322265625,
      "learning_rate": 0.011838283828382838,
      "loss": 0.2245,
      "num_input_tokens_seen": 1513344,
      "step": 7175
    },
    {
      "epoch": 0.7898789878987899,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.011846534653465347,
      "loss": 0.2282,
      "num_input_tokens_seen": 1514400,
      "step": 7180
    },
    {
      "epoch": 0.7904290429042904,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.011854785478547854,
      "loss": 0.2302,
      "num_input_tokens_seen": 1515424,
      "step": 7185
    },
    {
      "epoch": 0.7909790979097909,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.011863036303630363,
      "loss": 0.2465,
      "num_input_tokens_seen": 1516448,
      "step": 7190
    },
    {
      "epoch": 0.7915291529152916,
      "grad_norm": 0.0157470703125,
      "learning_rate": 0.01187128712871287,
      "loss": 0.2383,
      "num_input_tokens_seen": 1517472,
      "step": 7195
    },
    {
      "epoch": 0.7920792079207921,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.011879537953795379,
      "loss": 0.2308,
      "num_input_tokens_seen": 1518496,
      "step": 7200
    },
    {
      "epoch": 0.7926292629262927,
      "grad_norm": 0.02783203125,
      "learning_rate": 0.011887788778877887,
      "loss": 0.2337,
      "num_input_tokens_seen": 1519520,
      "step": 7205
    },
    {
      "epoch": 0.7931793179317932,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.011896039603960396,
      "loss": 0.2336,
      "num_input_tokens_seen": 1520608,
      "step": 7210
    },
    {
      "epoch": 0.7937293729372937,
      "grad_norm": 0.0142822265625,
      "learning_rate": 0.011904290429042905,
      "loss": 0.2324,
      "num_input_tokens_seen": 1521664,
      "step": 7215
    },
    {
      "epoch": 0.7942794279427943,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.011912541254125412,
      "loss": 0.2302,
      "num_input_tokens_seen": 1522688,
      "step": 7220
    },
    {
      "epoch": 0.7948294829482948,
      "grad_norm": 0.0150146484375,
      "learning_rate": 0.011920792079207921,
      "loss": 0.2334,
      "num_input_tokens_seen": 1523776,
      "step": 7225
    },
    {
      "epoch": 0.7953795379537953,
      "grad_norm": 0.01544189453125,
      "learning_rate": 0.011929042904290428,
      "loss": 0.2313,
      "num_input_tokens_seen": 1524864,
      "step": 7230
    },
    {
      "epoch": 0.7959295929592959,
      "grad_norm": 0.0137939453125,
      "learning_rate": 0.011937293729372937,
      "loss": 0.2283,
      "num_input_tokens_seen": 1525888,
      "step": 7235
    },
    {
      "epoch": 0.7964796479647965,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.011945544554455444,
      "loss": 0.2285,
      "num_input_tokens_seen": 1526976,
      "step": 7240
    },
    {
      "epoch": 0.7970297029702971,
      "grad_norm": 0.023193359375,
      "learning_rate": 0.011953795379537953,
      "loss": 0.2319,
      "num_input_tokens_seen": 1527968,
      "step": 7245
    },
    {
      "epoch": 0.7975797579757976,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.011962046204620462,
      "loss": 0.2355,
      "num_input_tokens_seen": 1528992,
      "step": 7250
    },
    {
      "epoch": 0.7981298129812981,
      "grad_norm": 0.01202392578125,
      "learning_rate": 0.01197029702970297,
      "loss": 0.2258,
      "num_input_tokens_seen": 1530016,
      "step": 7255
    },
    {
      "epoch": 0.7986798679867987,
      "grad_norm": 0.024658203125,
      "learning_rate": 0.01197854785478548,
      "loss": 0.2442,
      "num_input_tokens_seen": 1531104,
      "step": 7260
    },
    {
      "epoch": 0.7992299229922992,
      "grad_norm": 0.01422119140625,
      "learning_rate": 0.011986798679867986,
      "loss": 0.2328,
      "num_input_tokens_seen": 1532192,
      "step": 7265
    },
    {
      "epoch": 0.7997799779977998,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.011995049504950495,
      "loss": 0.2318,
      "num_input_tokens_seen": 1533216,
      "step": 7270
    },
    {
      "epoch": 0.8003300330033003,
      "grad_norm": 0.007720947265625,
      "learning_rate": 0.012003300330033002,
      "loss": 0.2243,
      "num_input_tokens_seen": 1534240,
      "step": 7275
    },
    {
      "epoch": 0.8008800880088008,
      "grad_norm": 0.02734375,
      "learning_rate": 0.012011551155115511,
      "loss": 0.2294,
      "num_input_tokens_seen": 1535296,
      "step": 7280
    },
    {
      "epoch": 0.8014301430143014,
      "grad_norm": 0.01361083984375,
      "learning_rate": 0.012019801980198018,
      "loss": 0.2314,
      "num_input_tokens_seen": 1536416,
      "step": 7285
    },
    {
      "epoch": 0.801980198019802,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.012028052805280527,
      "loss": 0.2314,
      "num_input_tokens_seen": 1537472,
      "step": 7290
    },
    {
      "epoch": 0.8025302530253026,
      "grad_norm": 0.003204345703125,
      "learning_rate": 0.012036303630363034,
      "loss": 0.2324,
      "num_input_tokens_seen": 1538464,
      "step": 7295
    },
    {
      "epoch": 0.8030803080308031,
      "grad_norm": 0.018310546875,
      "learning_rate": 0.012044554455445545,
      "loss": 0.2313,
      "num_input_tokens_seen": 1539488,
      "step": 7300
    },
    {
      "epoch": 0.8036303630363036,
      "grad_norm": 0.015869140625,
      "learning_rate": 0.012052805280528054,
      "loss": 0.2334,
      "num_input_tokens_seen": 1540576,
      "step": 7305
    },
    {
      "epoch": 0.8041804180418042,
      "grad_norm": 0.02685546875,
      "learning_rate": 0.01206105610561056,
      "loss": 0.2334,
      "num_input_tokens_seen": 1541632,
      "step": 7310
    },
    {
      "epoch": 0.8047304730473047,
      "grad_norm": 0.025146484375,
      "learning_rate": 0.01206930693069307,
      "loss": 0.2272,
      "num_input_tokens_seen": 1542656,
      "step": 7315
    },
    {
      "epoch": 0.8052805280528053,
      "grad_norm": 0.0123291015625,
      "learning_rate": 0.012077557755775577,
      "loss": 0.2294,
      "num_input_tokens_seen": 1543744,
      "step": 7320
    },
    {
      "epoch": 0.8058305830583058,
      "grad_norm": 0.02392578125,
      "learning_rate": 0.012085808580858086,
      "loss": 0.2255,
      "num_input_tokens_seen": 1544896,
      "step": 7325
    },
    {
      "epoch": 0.8063806380638063,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.012094059405940593,
      "loss": 0.2255,
      "num_input_tokens_seen": 1545920,
      "step": 7330
    },
    {
      "epoch": 0.806930693069307,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.012102310231023101,
      "loss": 0.2357,
      "num_input_tokens_seen": 1547040,
      "step": 7335
    },
    {
      "epoch": 0.8074807480748075,
      "grad_norm": 0.0167236328125,
      "learning_rate": 0.012110561056105609,
      "loss": 0.2385,
      "num_input_tokens_seen": 1548128,
      "step": 7340
    },
    {
      "epoch": 0.8080308030803081,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.01211881188118812,
      "loss": 0.2362,
      "num_input_tokens_seen": 1549216,
      "step": 7345
    },
    {
      "epoch": 0.8085808580858086,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.012127062706270628,
      "loss": 0.2299,
      "num_input_tokens_seen": 1550240,
      "step": 7350
    },
    {
      "epoch": 0.8091309130913091,
      "grad_norm": 0.0128173828125,
      "learning_rate": 0.012135313531353135,
      "loss": 0.2246,
      "num_input_tokens_seen": 1551264,
      "step": 7355
    },
    {
      "epoch": 0.8096809680968097,
      "grad_norm": 0.024658203125,
      "learning_rate": 0.012143564356435644,
      "loss": 0.2318,
      "num_input_tokens_seen": 1552288,
      "step": 7360
    },
    {
      "epoch": 0.8102310231023102,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.012151815181518151,
      "loss": 0.2318,
      "num_input_tokens_seen": 1553344,
      "step": 7365
    },
    {
      "epoch": 0.8107810781078107,
      "grad_norm": 0.0294189453125,
      "learning_rate": 0.01216006600660066,
      "loss": 0.2331,
      "num_input_tokens_seen": 1554432,
      "step": 7370
    },
    {
      "epoch": 0.8113311331133113,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.012168316831683167,
      "loss": 0.234,
      "num_input_tokens_seen": 1555456,
      "step": 7375
    },
    {
      "epoch": 0.8118811881188119,
      "grad_norm": 0.0147705078125,
      "learning_rate": 0.012176567656765676,
      "loss": 0.2391,
      "num_input_tokens_seen": 1556544,
      "step": 7380
    },
    {
      "epoch": 0.8124312431243125,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.012184818481848183,
      "loss": 0.2265,
      "num_input_tokens_seen": 1557600,
      "step": 7385
    },
    {
      "epoch": 0.812981298129813,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.012193069306930692,
      "loss": 0.2296,
      "num_input_tokens_seen": 1558624,
      "step": 7390
    },
    {
      "epoch": 0.8135313531353136,
      "grad_norm": 0.00390625,
      "learning_rate": 0.012201320132013202,
      "loss": 0.2399,
      "num_input_tokens_seen": 1559680,
      "step": 7395
    },
    {
      "epoch": 0.8140814081408141,
      "grad_norm": 0.0145263671875,
      "learning_rate": 0.01220957095709571,
      "loss": 0.2337,
      "num_input_tokens_seen": 1560768,
      "step": 7400
    },
    {
      "epoch": 0.8146314631463146,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.012217821782178218,
      "loss": 0.2324,
      "num_input_tokens_seen": 1561824,
      "step": 7405
    },
    {
      "epoch": 0.8151815181518152,
      "grad_norm": 0.017822265625,
      "learning_rate": 0.012226072607260725,
      "loss": 0.2308,
      "num_input_tokens_seen": 1562880,
      "step": 7410
    },
    {
      "epoch": 0.8157315731573157,
      "grad_norm": 0.01495361328125,
      "learning_rate": 0.012234323432343234,
      "loss": 0.2299,
      "num_input_tokens_seen": 1563936,
      "step": 7415
    },
    {
      "epoch": 0.8162816281628162,
      "grad_norm": 0.0166015625,
      "learning_rate": 0.012242574257425741,
      "loss": 0.2374,
      "num_input_tokens_seen": 1564960,
      "step": 7420
    },
    {
      "epoch": 0.8168316831683168,
      "grad_norm": 0.01397705078125,
      "learning_rate": 0.01225082508250825,
      "loss": 0.2309,
      "num_input_tokens_seen": 1565984,
      "step": 7425
    },
    {
      "epoch": 0.8173817381738174,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.012259075907590757,
      "loss": 0.2335,
      "num_input_tokens_seen": 1567008,
      "step": 7430
    },
    {
      "epoch": 0.817931793179318,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.012267326732673266,
      "loss": 0.2314,
      "num_input_tokens_seen": 1568064,
      "step": 7435
    },
    {
      "epoch": 0.8184818481848185,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.012275577557755777,
      "loss": 0.2299,
      "num_input_tokens_seen": 1569184,
      "step": 7440
    },
    {
      "epoch": 0.819031903190319,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.012283828382838284,
      "loss": 0.2309,
      "num_input_tokens_seen": 1570336,
      "step": 7445
    },
    {
      "epoch": 0.8195819581958196,
      "grad_norm": 0.0145263671875,
      "learning_rate": 0.012292079207920793,
      "loss": 0.2324,
      "num_input_tokens_seen": 1571328,
      "step": 7450
    },
    {
      "epoch": 0.8201320132013201,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0123003300330033,
      "loss": 0.2319,
      "num_input_tokens_seen": 1572448,
      "step": 7455
    },
    {
      "epoch": 0.8206820682068207,
      "grad_norm": 0.0234375,
      "learning_rate": 0.012308580858085809,
      "loss": 0.2304,
      "num_input_tokens_seen": 1573568,
      "step": 7460
    },
    {
      "epoch": 0.8212321232123212,
      "grad_norm": 0.0128173828125,
      "learning_rate": 0.012316831683168316,
      "loss": 0.2293,
      "num_input_tokens_seen": 1574592,
      "step": 7465
    },
    {
      "epoch": 0.8217821782178217,
      "grad_norm": 0.01300048828125,
      "learning_rate": 0.012325082508250824,
      "loss": 0.2315,
      "num_input_tokens_seen": 1575680,
      "step": 7470
    },
    {
      "epoch": 0.8223322332233224,
      "grad_norm": 0.02587890625,
      "learning_rate": 0.012333333333333332,
      "loss": 0.2346,
      "num_input_tokens_seen": 1576736,
      "step": 7475
    },
    {
      "epoch": 0.8228822882288229,
      "grad_norm": 0.0123291015625,
      "learning_rate": 0.01234158415841584,
      "loss": 0.2352,
      "num_input_tokens_seen": 1577760,
      "step": 7480
    },
    {
      "epoch": 0.8234323432343235,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.012349834983498351,
      "loss": 0.2308,
      "num_input_tokens_seen": 1578816,
      "step": 7485
    },
    {
      "epoch": 0.823982398239824,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.012358085808580858,
      "loss": 0.2324,
      "num_input_tokens_seen": 1579904,
      "step": 7490
    },
    {
      "epoch": 0.8245324532453245,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.012366336633663367,
      "loss": 0.2313,
      "num_input_tokens_seen": 1581024,
      "step": 7495
    },
    {
      "epoch": 0.8250825082508251,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.012374587458745874,
      "loss": 0.2298,
      "num_input_tokens_seen": 1582048,
      "step": 7500
    },
    {
      "epoch": 0.8256325632563256,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.012382838283828383,
      "loss": 0.2298,
      "num_input_tokens_seen": 1583136,
      "step": 7505
    },
    {
      "epoch": 0.8261826182618262,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01239108910891089,
      "loss": 0.2293,
      "num_input_tokens_seen": 1584096,
      "step": 7510
    },
    {
      "epoch": 0.8267326732673267,
      "grad_norm": 0.029296875,
      "learning_rate": 0.012399339933993399,
      "loss": 0.2263,
      "num_input_tokens_seen": 1585120,
      "step": 7515
    },
    {
      "epoch": 0.8272827282728272,
      "grad_norm": 0.06884765625,
      "learning_rate": 0.012407590759075906,
      "loss": 0.2319,
      "num_input_tokens_seen": 1586144,
      "step": 7520
    },
    {
      "epoch": 0.8278327832783279,
      "grad_norm": 0.0252685546875,
      "learning_rate": 0.012415841584158415,
      "loss": 0.2346,
      "num_input_tokens_seen": 1587136,
      "step": 7525
    },
    {
      "epoch": 0.8283828382838284,
      "grad_norm": 0.01171875,
      "learning_rate": 0.012424092409240924,
      "loss": 0.2388,
      "num_input_tokens_seen": 1588192,
      "step": 7530
    },
    {
      "epoch": 0.828932893289329,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.012432343234323432,
      "loss": 0.2293,
      "num_input_tokens_seen": 1589216,
      "step": 7535
    },
    {
      "epoch": 0.8294829482948295,
      "grad_norm": 0.04345703125,
      "learning_rate": 0.012440594059405941,
      "loss": 0.2368,
      "num_input_tokens_seen": 1590304,
      "step": 7540
    },
    {
      "epoch": 0.83003300330033,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.012448844884488448,
      "loss": 0.2321,
      "num_input_tokens_seen": 1591456,
      "step": 7545
    },
    {
      "epoch": 0.8305830583058306,
      "grad_norm": 0.056640625,
      "learning_rate": 0.012457095709570957,
      "loss": 0.2327,
      "num_input_tokens_seen": 1592576,
      "step": 7550
    },
    {
      "epoch": 0.8311331133113311,
      "grad_norm": 0.1240234375,
      "learning_rate": 0.012465346534653464,
      "loss": 0.2391,
      "num_input_tokens_seen": 1593664,
      "step": 7555
    },
    {
      "epoch": 0.8316831683168316,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.012473597359735973,
      "loss": 0.2381,
      "num_input_tokens_seen": 1594688,
      "step": 7560
    },
    {
      "epoch": 0.8322332233223322,
      "grad_norm": 0.033935546875,
      "learning_rate": 0.01248184818481848,
      "loss": 0.2333,
      "num_input_tokens_seen": 1595712,
      "step": 7565
    },
    {
      "epoch": 0.8327832783278328,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.012490099009900989,
      "loss": 0.2325,
      "num_input_tokens_seen": 1596768,
      "step": 7570
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 0.01556396484375,
      "learning_rate": 0.012498349834983498,
      "loss": 0.2335,
      "num_input_tokens_seen": 1597856,
      "step": 7575
    },
    {
      "epoch": 0.8338833883388339,
      "grad_norm": 0.0159912109375,
      "learning_rate": 0.012506600660066007,
      "loss": 0.2284,
      "num_input_tokens_seen": 1598944,
      "step": 7580
    },
    {
      "epoch": 0.8344334433443344,
      "grad_norm": 0.015625,
      "learning_rate": 0.012514851485148516,
      "loss": 0.2327,
      "num_input_tokens_seen": 1599936,
      "step": 7585
    },
    {
      "epoch": 0.834983498349835,
      "grad_norm": 0.003997802734375,
      "learning_rate": 0.012523102310231023,
      "loss": 0.2295,
      "num_input_tokens_seen": 1600992,
      "step": 7590
    },
    {
      "epoch": 0.8355335533553355,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.012531353135313531,
      "loss": 0.2315,
      "num_input_tokens_seen": 1602016,
      "step": 7595
    },
    {
      "epoch": 0.8360836083608361,
      "grad_norm": 0.0146484375,
      "learning_rate": 0.012539603960396039,
      "loss": 0.2272,
      "num_input_tokens_seen": 1603040,
      "step": 7600
    },
    {
      "epoch": 0.8366336633663366,
      "grad_norm": 0.0155029296875,
      "learning_rate": 0.012547854785478547,
      "loss": 0.2334,
      "num_input_tokens_seen": 1604064,
      "step": 7605
    },
    {
      "epoch": 0.8371837183718371,
      "grad_norm": 0.0157470703125,
      "learning_rate": 0.012556105610561055,
      "loss": 0.2312,
      "num_input_tokens_seen": 1605120,
      "step": 7610
    },
    {
      "epoch": 0.8377337733773378,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.012564356435643563,
      "loss": 0.2333,
      "num_input_tokens_seen": 1606176,
      "step": 7615
    },
    {
      "epoch": 0.8382838283828383,
      "grad_norm": 0.0296630859375,
      "learning_rate": 0.012572607260726072,
      "loss": 0.2333,
      "num_input_tokens_seen": 1607232,
      "step": 7620
    },
    {
      "epoch": 0.8388338833883389,
      "grad_norm": 0.015625,
      "learning_rate": 0.012580858085808581,
      "loss": 0.2301,
      "num_input_tokens_seen": 1608256,
      "step": 7625
    },
    {
      "epoch": 0.8393839383938394,
      "grad_norm": 0.01513671875,
      "learning_rate": 0.01258910891089109,
      "loss": 0.2292,
      "num_input_tokens_seen": 1609312,
      "step": 7630
    },
    {
      "epoch": 0.8399339933993399,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.012597359735973597,
      "loss": 0.2313,
      "num_input_tokens_seen": 1610400,
      "step": 7635
    },
    {
      "epoch": 0.8404840484048405,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.012605610561056106,
      "loss": 0.2334,
      "num_input_tokens_seen": 1611520,
      "step": 7640
    },
    {
      "epoch": 0.841034103410341,
      "grad_norm": 0.01324462890625,
      "learning_rate": 0.012613861386138613,
      "loss": 0.2322,
      "num_input_tokens_seen": 1612576,
      "step": 7645
    },
    {
      "epoch": 0.8415841584158416,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.012622112211221122,
      "loss": 0.2312,
      "num_input_tokens_seen": 1613664,
      "step": 7650
    },
    {
      "epoch": 0.8421342134213421,
      "grad_norm": 0.007598876953125,
      "learning_rate": 0.012630363036303629,
      "loss": 0.2314,
      "num_input_tokens_seen": 1614720,
      "step": 7655
    },
    {
      "epoch": 0.8426842684268426,
      "grad_norm": 0.00445556640625,
      "learning_rate": 0.012638613861386138,
      "loss": 0.2323,
      "num_input_tokens_seen": 1615840,
      "step": 7660
    },
    {
      "epoch": 0.8432343234323433,
      "grad_norm": 0.0147705078125,
      "learning_rate": 0.012646864686468647,
      "loss": 0.2323,
      "num_input_tokens_seen": 1616960,
      "step": 7665
    },
    {
      "epoch": 0.8437843784378438,
      "grad_norm": 0.00433349609375,
      "learning_rate": 0.012655115511551154,
      "loss": 0.2302,
      "num_input_tokens_seen": 1617984,
      "step": 7670
    },
    {
      "epoch": 0.8443344334433444,
      "grad_norm": 0.015380859375,
      "learning_rate": 0.012663366336633664,
      "loss": 0.2302,
      "num_input_tokens_seen": 1619040,
      "step": 7675
    },
    {
      "epoch": 0.8448844884488449,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.012671617161716171,
      "loss": 0.2313,
      "num_input_tokens_seen": 1620128,
      "step": 7680
    },
    {
      "epoch": 0.8454345434543454,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.01267986798679868,
      "loss": 0.2282,
      "num_input_tokens_seen": 1621152,
      "step": 7685
    },
    {
      "epoch": 0.845984598459846,
      "grad_norm": 0.01513671875,
      "learning_rate": 0.012688118811881187,
      "loss": 0.2345,
      "num_input_tokens_seen": 1622208,
      "step": 7690
    },
    {
      "epoch": 0.8465346534653465,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.012696369636963696,
      "loss": 0.2323,
      "num_input_tokens_seen": 1623360,
      "step": 7695
    },
    {
      "epoch": 0.847084708470847,
      "grad_norm": 0.0142822265625,
      "learning_rate": 0.012704620462046203,
      "loss": 0.2323,
      "num_input_tokens_seen": 1624448,
      "step": 7700
    },
    {
      "epoch": 0.8476347634763476,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.012712871287128712,
      "loss": 0.2334,
      "num_input_tokens_seen": 1625504,
      "step": 7705
    },
    {
      "epoch": 0.8481848184818482,
      "grad_norm": 0.01361083984375,
      "learning_rate": 0.01272112211221122,
      "loss": 0.2323,
      "num_input_tokens_seen": 1626688,
      "step": 7710
    },
    {
      "epoch": 0.8487348734873488,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012729372937293728,
      "loss": 0.2312,
      "num_input_tokens_seen": 1627712,
      "step": 7715
    },
    {
      "epoch": 0.8492849284928493,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.012737623762376239,
      "loss": 0.2324,
      "num_input_tokens_seen": 1628704,
      "step": 7720
    },
    {
      "epoch": 0.8498349834983498,
      "grad_norm": 0.015869140625,
      "learning_rate": 0.012745874587458746,
      "loss": 0.2313,
      "num_input_tokens_seen": 1629824,
      "step": 7725
    },
    {
      "epoch": 0.8503850385038504,
      "grad_norm": 0.015869140625,
      "learning_rate": 0.012754125412541254,
      "loss": 0.2322,
      "num_input_tokens_seen": 1630880,
      "step": 7730
    },
    {
      "epoch": 0.8509350935093509,
      "grad_norm": 0.004302978515625,
      "learning_rate": 0.012762376237623762,
      "loss": 0.2338,
      "num_input_tokens_seen": 1631904,
      "step": 7735
    },
    {
      "epoch": 0.8514851485148515,
      "grad_norm": 0.004547119140625,
      "learning_rate": 0.01277062706270627,
      "loss": 0.2328,
      "num_input_tokens_seen": 1632992,
      "step": 7740
    },
    {
      "epoch": 0.852035203520352,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.012778877887788778,
      "loss": 0.2318,
      "num_input_tokens_seen": 1634080,
      "step": 7745
    },
    {
      "epoch": 0.8525852585258525,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.012787128712871286,
      "loss": 0.2234,
      "num_input_tokens_seen": 1635104,
      "step": 7750
    },
    {
      "epoch": 0.8531353135313532,
      "grad_norm": 0.0264892578125,
      "learning_rate": 0.012795379537953795,
      "loss": 0.2348,
      "num_input_tokens_seen": 1636096,
      "step": 7755
    },
    {
      "epoch": 0.8536853685368537,
      "grad_norm": 0.003662109375,
      "learning_rate": 0.012803630363036302,
      "loss": 0.2285,
      "num_input_tokens_seen": 1637056,
      "step": 7760
    },
    {
      "epoch": 0.8542354235423543,
      "grad_norm": 0.002838134765625,
      "learning_rate": 0.012811881188118813,
      "loss": 0.2296,
      "num_input_tokens_seen": 1638176,
      "step": 7765
    },
    {
      "epoch": 0.8547854785478548,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.01282013201320132,
      "loss": 0.2232,
      "num_input_tokens_seen": 1639232,
      "step": 7770
    },
    {
      "epoch": 0.8553355335533553,
      "grad_norm": 0.01544189453125,
      "learning_rate": 0.012828382838283829,
      "loss": 0.2349,
      "num_input_tokens_seen": 1640320,
      "step": 7775
    },
    {
      "epoch": 0.8558855885588559,
      "grad_norm": 0.00299072265625,
      "learning_rate": 0.012836633663366336,
      "loss": 0.2214,
      "num_input_tokens_seen": 1641344,
      "step": 7780
    },
    {
      "epoch": 0.8564356435643564,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.012844884488448845,
      "loss": 0.229,
      "num_input_tokens_seen": 1642464,
      "step": 7785
    },
    {
      "epoch": 0.856985698569857,
      "grad_norm": 0.02685546875,
      "learning_rate": 0.012853135313531352,
      "loss": 0.2262,
      "num_input_tokens_seen": 1643552,
      "step": 7790
    },
    {
      "epoch": 0.8575357535753575,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.01286138613861386,
      "loss": 0.2389,
      "num_input_tokens_seen": 1644608,
      "step": 7795
    },
    {
      "epoch": 0.858085808580858,
      "grad_norm": 0.0034942626953125,
      "learning_rate": 0.01286963696369637,
      "loss": 0.2305,
      "num_input_tokens_seen": 1645664,
      "step": 7800
    },
    {
      "epoch": 0.8586358635863587,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.012877887788778877,
      "loss": 0.2368,
      "num_input_tokens_seen": 1646688,
      "step": 7805
    },
    {
      "epoch": 0.8591859185918592,
      "grad_norm": 0.014404296875,
      "learning_rate": 0.012886138613861385,
      "loss": 0.2319,
      "num_input_tokens_seen": 1647712,
      "step": 7810
    },
    {
      "epoch": 0.8597359735973598,
      "grad_norm": 0.014892578125,
      "learning_rate": 0.012894389438943894,
      "loss": 0.2321,
      "num_input_tokens_seen": 1648704,
      "step": 7815
    },
    {
      "epoch": 0.8602860286028603,
      "grad_norm": 0.0230712890625,
      "learning_rate": 0.012902640264026403,
      "loss": 0.2277,
      "num_input_tokens_seen": 1649728,
      "step": 7820
    },
    {
      "epoch": 0.8608360836083608,
      "grad_norm": 0.01470947265625,
      "learning_rate": 0.01291089108910891,
      "loss": 0.2339,
      "num_input_tokens_seen": 1650784,
      "step": 7825
    },
    {
      "epoch": 0.8613861386138614,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.012919141914191419,
      "loss": 0.2319,
      "num_input_tokens_seen": 1651872,
      "step": 7830
    },
    {
      "epoch": 0.8619361936193619,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.012927392739273926,
      "loss": 0.238,
      "num_input_tokens_seen": 1652896,
      "step": 7835
    },
    {
      "epoch": 0.8624862486248625,
      "grad_norm": 0.0140380859375,
      "learning_rate": 0.012935643564356435,
      "loss": 0.2307,
      "num_input_tokens_seen": 1653952,
      "step": 7840
    },
    {
      "epoch": 0.863036303630363,
      "grad_norm": 0.02490234375,
      "learning_rate": 0.012943894389438944,
      "loss": 0.2368,
      "num_input_tokens_seen": 1654976,
      "step": 7845
    },
    {
      "epoch": 0.8635863586358636,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.012952145214521451,
      "loss": 0.2306,
      "num_input_tokens_seen": 1656064,
      "step": 7850
    },
    {
      "epoch": 0.8641364136413642,
      "grad_norm": 0.01361083984375,
      "learning_rate": 0.01296039603960396,
      "loss": 0.2346,
      "num_input_tokens_seen": 1657120,
      "step": 7855
    },
    {
      "epoch": 0.8646864686468647,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.012968646864686469,
      "loss": 0.2305,
      "num_input_tokens_seen": 1658144,
      "step": 7860
    },
    {
      "epoch": 0.8652365236523653,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.012976897689768977,
      "loss": 0.2326,
      "num_input_tokens_seen": 1659136,
      "step": 7865
    },
    {
      "epoch": 0.8657865786578658,
      "grad_norm": 0.01300048828125,
      "learning_rate": 0.012985148514851485,
      "loss": 0.2353,
      "num_input_tokens_seen": 1660192,
      "step": 7870
    },
    {
      "epoch": 0.8663366336633663,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.012993399339933993,
      "loss": 0.2358,
      "num_input_tokens_seen": 1661280,
      "step": 7875
    },
    {
      "epoch": 0.8668866886688669,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.0130016501650165,
      "loss": 0.2305,
      "num_input_tokens_seen": 1662336,
      "step": 7880
    },
    {
      "epoch": 0.8674367436743674,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.01300990099009901,
      "loss": 0.2326,
      "num_input_tokens_seen": 1663392,
      "step": 7885
    },
    {
      "epoch": 0.8679867986798679,
      "grad_norm": 0.01324462890625,
      "learning_rate": 0.013018151815181518,
      "loss": 0.2372,
      "num_input_tokens_seen": 1664416,
      "step": 7890
    },
    {
      "epoch": 0.8685368536853685,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.013026402640264025,
      "loss": 0.2324,
      "num_input_tokens_seen": 1665504,
      "step": 7895
    },
    {
      "epoch": 0.8690869086908691,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.013034653465346534,
      "loss": 0.2293,
      "num_input_tokens_seen": 1666592,
      "step": 7900
    },
    {
      "epoch": 0.8696369636963697,
      "grad_norm": 0.03857421875,
      "learning_rate": 0.013042904290429041,
      "loss": 0.2305,
      "num_input_tokens_seen": 1667680,
      "step": 7905
    },
    {
      "epoch": 0.8701870187018702,
      "grad_norm": 0.0205078125,
      "learning_rate": 0.013051155115511552,
      "loss": 0.2414,
      "num_input_tokens_seen": 1668736,
      "step": 7910
    },
    {
      "epoch": 0.8707370737073707,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.013059405940594059,
      "loss": 0.235,
      "num_input_tokens_seen": 1669824,
      "step": 7915
    },
    {
      "epoch": 0.8712871287128713,
      "grad_norm": 0.02734375,
      "learning_rate": 0.013067656765676568,
      "loss": 0.2316,
      "num_input_tokens_seen": 1670880,
      "step": 7920
    },
    {
      "epoch": 0.8718371837183718,
      "grad_norm": 0.0159912109375,
      "learning_rate": 0.013075907590759075,
      "loss": 0.2335,
      "num_input_tokens_seen": 1671936,
      "step": 7925
    },
    {
      "epoch": 0.8723872387238724,
      "grad_norm": 0.002838134765625,
      "learning_rate": 0.013084158415841584,
      "loss": 0.2325,
      "num_input_tokens_seen": 1672928,
      "step": 7930
    },
    {
      "epoch": 0.8729372937293729,
      "grad_norm": 0.0179443359375,
      "learning_rate": 0.013092409240924092,
      "loss": 0.231,
      "num_input_tokens_seen": 1673984,
      "step": 7935
    },
    {
      "epoch": 0.8734873487348734,
      "grad_norm": 0.023681640625,
      "learning_rate": 0.0131006600660066,
      "loss": 0.2315,
      "num_input_tokens_seen": 1675040,
      "step": 7940
    },
    {
      "epoch": 0.8740374037403741,
      "grad_norm": 0.0137939453125,
      "learning_rate": 0.013108910891089108,
      "loss": 0.2336,
      "num_input_tokens_seen": 1676096,
      "step": 7945
    },
    {
      "epoch": 0.8745874587458746,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.013117161716171616,
      "loss": 0.2298,
      "num_input_tokens_seen": 1677184,
      "step": 7950
    },
    {
      "epoch": 0.8751375137513752,
      "grad_norm": 0.0223388671875,
      "learning_rate": 0.013125412541254126,
      "loss": 0.233,
      "num_input_tokens_seen": 1678304,
      "step": 7955
    },
    {
      "epoch": 0.8756875687568757,
      "grad_norm": 0.0234375,
      "learning_rate": 0.013133663366336633,
      "loss": 0.2319,
      "num_input_tokens_seen": 1679360,
      "step": 7960
    },
    {
      "epoch": 0.8762376237623762,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.013141914191419142,
      "loss": 0.2341,
      "num_input_tokens_seen": 1680384,
      "step": 7965
    },
    {
      "epoch": 0.8767876787678768,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.013150165016501649,
      "loss": 0.2314,
      "num_input_tokens_seen": 1681440,
      "step": 7970
    },
    {
      "epoch": 0.8773377337733773,
      "grad_norm": 0.00323486328125,
      "learning_rate": 0.013158415841584158,
      "loss": 0.2319,
      "num_input_tokens_seen": 1682464,
      "step": 7975
    },
    {
      "epoch": 0.8778877887788779,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.013166666666666667,
      "loss": 0.2319,
      "num_input_tokens_seen": 1683552,
      "step": 7980
    },
    {
      "epoch": 0.8784378437843784,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.013174917491749174,
      "loss": 0.2314,
      "num_input_tokens_seen": 1684576,
      "step": 7985
    },
    {
      "epoch": 0.878987898789879,
      "grad_norm": 0.0123291015625,
      "learning_rate": 0.013183168316831683,
      "loss": 0.2304,
      "num_input_tokens_seen": 1685664,
      "step": 7990
    },
    {
      "epoch": 0.8795379537953796,
      "grad_norm": 0.0036468505859375,
      "learning_rate": 0.01319141914191419,
      "loss": 0.2274,
      "num_input_tokens_seen": 1686720,
      "step": 7995
    },
    {
      "epoch": 0.8800880088008801,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.0131996699669967,
      "loss": 0.2332,
      "num_input_tokens_seen": 1687744,
      "step": 8000
    },
    {
      "epoch": 0.8806380638063807,
      "grad_norm": 0.02392578125,
      "learning_rate": 0.013207920792079208,
      "loss": 0.2287,
      "num_input_tokens_seen": 1688800,
      "step": 8005
    },
    {
      "epoch": 0.8811881188118812,
      "grad_norm": 0.00335693359375,
      "learning_rate": 0.013216171617161716,
      "loss": 0.2333,
      "num_input_tokens_seen": 1689856,
      "step": 8010
    },
    {
      "epoch": 0.8817381738173817,
      "grad_norm": 0.0140380859375,
      "learning_rate": 0.013224422442244223,
      "loss": 0.2355,
      "num_input_tokens_seen": 1690912,
      "step": 8015
    },
    {
      "epoch": 0.8822882288228823,
      "grad_norm": 0.01336669921875,
      "learning_rate": 0.013232673267326732,
      "loss": 0.2354,
      "num_input_tokens_seen": 1691936,
      "step": 8020
    },
    {
      "epoch": 0.8828382838283828,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.013240924092409241,
      "loss": 0.2331,
      "num_input_tokens_seen": 1692960,
      "step": 8025
    },
    {
      "epoch": 0.8833883388338833,
      "grad_norm": 0.00396728515625,
      "learning_rate": 0.013249174917491748,
      "loss": 0.2309,
      "num_input_tokens_seen": 1694016,
      "step": 8030
    },
    {
      "epoch": 0.8839383938393839,
      "grad_norm": 0.0033111572265625,
      "learning_rate": 0.013257425742574257,
      "loss": 0.2324,
      "num_input_tokens_seen": 1695104,
      "step": 8035
    },
    {
      "epoch": 0.8844884488448845,
      "grad_norm": 0.0030059814453125,
      "learning_rate": 0.013265676567656764,
      "loss": 0.2298,
      "num_input_tokens_seen": 1696192,
      "step": 8040
    },
    {
      "epoch": 0.8850385038503851,
      "grad_norm": 0.00335693359375,
      "learning_rate": 0.013273927392739273,
      "loss": 0.2324,
      "num_input_tokens_seen": 1697280,
      "step": 8045
    },
    {
      "epoch": 0.8855885588558856,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.013282178217821782,
      "loss": 0.2314,
      "num_input_tokens_seen": 1698240,
      "step": 8050
    },
    {
      "epoch": 0.8861386138613861,
      "grad_norm": 0.00421142578125,
      "learning_rate": 0.01329042904290429,
      "loss": 0.2319,
      "num_input_tokens_seen": 1699360,
      "step": 8055
    },
    {
      "epoch": 0.8866886688668867,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.013298679867986798,
      "loss": 0.2304,
      "num_input_tokens_seen": 1700416,
      "step": 8060
    },
    {
      "epoch": 0.8872387238723872,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.013306930693069307,
      "loss": 0.2319,
      "num_input_tokens_seen": 1701504,
      "step": 8065
    },
    {
      "epoch": 0.8877887788778878,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.013315181518151815,
      "loss": 0.2319,
      "num_input_tokens_seen": 1702560,
      "step": 8070
    },
    {
      "epoch": 0.8883388338833883,
      "grad_norm": 0.003936767578125,
      "learning_rate": 0.013323432343234323,
      "loss": 0.2309,
      "num_input_tokens_seen": 1703584,
      "step": 8075
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.013331683168316831,
      "loss": 0.2314,
      "num_input_tokens_seen": 1704576,
      "step": 8080
    },
    {
      "epoch": 0.8894389438943895,
      "grad_norm": 0.01953125,
      "learning_rate": 0.013339933993399338,
      "loss": 0.2319,
      "num_input_tokens_seen": 1705728,
      "step": 8085
    },
    {
      "epoch": 0.88998899889989,
      "grad_norm": 0.002960205078125,
      "learning_rate": 0.013348184818481847,
      "loss": 0.2314,
      "num_input_tokens_seen": 1706784,
      "step": 8090
    },
    {
      "epoch": 0.8905390539053906,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.013356435643564356,
      "loss": 0.2298,
      "num_input_tokens_seen": 1707936,
      "step": 8095
    },
    {
      "epoch": 0.8910891089108911,
      "grad_norm": 0.01171875,
      "learning_rate": 0.013364686468646865,
      "loss": 0.233,
      "num_input_tokens_seen": 1709056,
      "step": 8100
    },
    {
      "epoch": 0.8916391639163916,
      "grad_norm": 0.01202392578125,
      "learning_rate": 0.013372937293729372,
      "loss": 0.2363,
      "num_input_tokens_seen": 1710144,
      "step": 8105
    },
    {
      "epoch": 0.8921892189218922,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.013381188118811881,
      "loss": 0.2362,
      "num_input_tokens_seen": 1711232,
      "step": 8110
    },
    {
      "epoch": 0.8927392739273927,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01338943894389439,
      "loss": 0.232,
      "num_input_tokens_seen": 1712320,
      "step": 8115
    },
    {
      "epoch": 0.8932893289328933,
      "grad_norm": 0.003448486328125,
      "learning_rate": 0.013397689768976897,
      "loss": 0.2315,
      "num_input_tokens_seen": 1713376,
      "step": 8120
    },
    {
      "epoch": 0.8938393839383938,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.013405940594059406,
      "loss": 0.2336,
      "num_input_tokens_seen": 1714432,
      "step": 8125
    },
    {
      "epoch": 0.8943894389438944,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.013414191419141913,
      "loss": 0.2319,
      "num_input_tokens_seen": 1715392,
      "step": 8130
    },
    {
      "epoch": 0.894939493949395,
      "grad_norm": 0.01171875,
      "learning_rate": 0.013422442244224422,
      "loss": 0.2319,
      "num_input_tokens_seen": 1716352,
      "step": 8135
    },
    {
      "epoch": 0.8954895489548955,
      "grad_norm": 0.020751953125,
      "learning_rate": 0.01343069306930693,
      "loss": 0.233,
      "num_input_tokens_seen": 1717376,
      "step": 8140
    },
    {
      "epoch": 0.8960396039603961,
      "grad_norm": 0.003753662109375,
      "learning_rate": 0.01343894389438944,
      "loss": 0.2325,
      "num_input_tokens_seen": 1718496,
      "step": 8145
    },
    {
      "epoch": 0.8965896589658966,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.013447194719471946,
      "loss": 0.233,
      "num_input_tokens_seen": 1719552,
      "step": 8150
    },
    {
      "epoch": 0.8971397139713971,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.013455445544554455,
      "loss": 0.2294,
      "num_input_tokens_seen": 1720640,
      "step": 8155
    },
    {
      "epoch": 0.8976897689768977,
      "grad_norm": 0.022705078125,
      "learning_rate": 0.013463696369636964,
      "loss": 0.2325,
      "num_input_tokens_seen": 1721728,
      "step": 8160
    },
    {
      "epoch": 0.8982398239823982,
      "grad_norm": 0.00445556640625,
      "learning_rate": 0.013471947194719471,
      "loss": 0.232,
      "num_input_tokens_seen": 1722752,
      "step": 8165
    },
    {
      "epoch": 0.8987898789878987,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.01348019801980198,
      "loss": 0.232,
      "num_input_tokens_seen": 1723872,
      "step": 8170
    },
    {
      "epoch": 0.8993399339933993,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.013488448844884487,
      "loss": 0.233,
      "num_input_tokens_seen": 1724896,
      "step": 8175
    },
    {
      "epoch": 0.8998899889988999,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013496699669966996,
      "loss": 0.2304,
      "num_input_tokens_seen": 1725984,
      "step": 8180
    },
    {
      "epoch": 0.9004400440044005,
      "grad_norm": 0.00421142578125,
      "learning_rate": 0.013504950495049503,
      "loss": 0.2279,
      "num_input_tokens_seen": 1727040,
      "step": 8185
    },
    {
      "epoch": 0.900990099009901,
      "grad_norm": 0.01171875,
      "learning_rate": 0.013513201320132014,
      "loss": 0.232,
      "num_input_tokens_seen": 1728064,
      "step": 8190
    },
    {
      "epoch": 0.9015401540154016,
      "grad_norm": 0.023193359375,
      "learning_rate": 0.01352145214521452,
      "loss": 0.2289,
      "num_input_tokens_seen": 1729088,
      "step": 8195
    },
    {
      "epoch": 0.9020902090209021,
      "grad_norm": 0.0123291015625,
      "learning_rate": 0.01352970297029703,
      "loss": 0.2321,
      "num_input_tokens_seen": 1730208,
      "step": 8200
    },
    {
      "epoch": 0.9026402640264026,
      "grad_norm": 0.023681640625,
      "learning_rate": 0.013537953795379538,
      "loss": 0.2332,
      "num_input_tokens_seen": 1731200,
      "step": 8205
    },
    {
      "epoch": 0.9031903190319032,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.013546204620462046,
      "loss": 0.2305,
      "num_input_tokens_seen": 1732224,
      "step": 8210
    },
    {
      "epoch": 0.9037403740374037,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.013554455445544554,
      "loss": 0.2299,
      "num_input_tokens_seen": 1733280,
      "step": 8215
    },
    {
      "epoch": 0.9042904290429042,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.013562706270627061,
      "loss": 0.2299,
      "num_input_tokens_seen": 1734336,
      "step": 8220
    },
    {
      "epoch": 0.9048404840484049,
      "grad_norm": 0.003448486328125,
      "learning_rate": 0.01357095709570957,
      "loss": 0.2331,
      "num_input_tokens_seen": 1735360,
      "step": 8225
    },
    {
      "epoch": 0.9053905390539054,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.013579207920792077,
      "loss": 0.233,
      "num_input_tokens_seen": 1736448,
      "step": 8230
    },
    {
      "epoch": 0.905940594059406,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.013587458745874588,
      "loss": 0.233,
      "num_input_tokens_seen": 1737536,
      "step": 8235
    },
    {
      "epoch": 0.9064906490649065,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.013595709570957095,
      "loss": 0.2319,
      "num_input_tokens_seen": 1738528,
      "step": 8240
    },
    {
      "epoch": 0.907040704070407,
      "grad_norm": 0.0038604736328125,
      "learning_rate": 0.013603960396039604,
      "loss": 0.2324,
      "num_input_tokens_seen": 1739584,
      "step": 8245
    },
    {
      "epoch": 0.9075907590759076,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.013612211221122113,
      "loss": 0.2303,
      "num_input_tokens_seen": 1740608,
      "step": 8250
    },
    {
      "epoch": 0.9081408140814081,
      "grad_norm": 0.0030670166015625,
      "learning_rate": 0.01362046204620462,
      "loss": 0.2283,
      "num_input_tokens_seen": 1741664,
      "step": 8255
    },
    {
      "epoch": 0.9086908690869087,
      "grad_norm": 0.009765625,
      "learning_rate": 0.013628712871287129,
      "loss": 0.2328,
      "num_input_tokens_seen": 1742656,
      "step": 8260
    },
    {
      "epoch": 0.9092409240924092,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.013636963696369636,
      "loss": 0.2352,
      "num_input_tokens_seen": 1743648,
      "step": 8265
    },
    {
      "epoch": 0.9097909790979097,
      "grad_norm": 0.003082275390625,
      "learning_rate": 0.013645214521452145,
      "loss": 0.2361,
      "num_input_tokens_seen": 1744704,
      "step": 8270
    },
    {
      "epoch": 0.9103410341034104,
      "grad_norm": 0.0037994384765625,
      "learning_rate": 0.013653465346534652,
      "loss": 0.234,
      "num_input_tokens_seen": 1745760,
      "step": 8275
    },
    {
      "epoch": 0.9108910891089109,
      "grad_norm": 0.0123291015625,
      "learning_rate": 0.01366171617161716,
      "loss": 0.2387,
      "num_input_tokens_seen": 1746880,
      "step": 8280
    },
    {
      "epoch": 0.9114411441144115,
      "grad_norm": 0.0031890869140625,
      "learning_rate": 0.01366996699669967,
      "loss": 0.2338,
      "num_input_tokens_seen": 1747968,
      "step": 8285
    },
    {
      "epoch": 0.911991199119912,
      "grad_norm": 0.01171875,
      "learning_rate": 0.013678217821782178,
      "loss": 0.2264,
      "num_input_tokens_seen": 1748992,
      "step": 8290
    },
    {
      "epoch": 0.9125412541254125,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.013686468646864687,
      "loss": 0.2305,
      "num_input_tokens_seen": 1750048,
      "step": 8295
    },
    {
      "epoch": 0.9130913091309131,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.013694719471947194,
      "loss": 0.2325,
      "num_input_tokens_seen": 1751072,
      "step": 8300
    },
    {
      "epoch": 0.9136413641364136,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.013702970297029703,
      "loss": 0.2336,
      "num_input_tokens_seen": 1752096,
      "step": 8305
    },
    {
      "epoch": 0.9141914191419142,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.01371122112211221,
      "loss": 0.2303,
      "num_input_tokens_seen": 1753152,
      "step": 8310
    },
    {
      "epoch": 0.9147414741474147,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.013719471947194719,
      "loss": 0.2324,
      "num_input_tokens_seen": 1754240,
      "step": 8315
    },
    {
      "epoch": 0.9152915291529153,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.013727722772277226,
      "loss": 0.2314,
      "num_input_tokens_seen": 1755328,
      "step": 8320
    },
    {
      "epoch": 0.9158415841584159,
      "grad_norm": 0.01239013671875,
      "learning_rate": 0.013735973597359735,
      "loss": 0.2314,
      "num_input_tokens_seen": 1756384,
      "step": 8325
    },
    {
      "epoch": 0.9163916391639164,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.013744224422442244,
      "loss": 0.2324,
      "num_input_tokens_seen": 1757376,
      "step": 8330
    },
    {
      "epoch": 0.916941694169417,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.013752475247524753,
      "loss": 0.2325,
      "num_input_tokens_seen": 1758432,
      "step": 8335
    },
    {
      "epoch": 0.9174917491749175,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.013760726072607261,
      "loss": 0.2285,
      "num_input_tokens_seen": 1759424,
      "step": 8340
    },
    {
      "epoch": 0.918041804180418,
      "grad_norm": 0.0035552978515625,
      "learning_rate": 0.013768976897689768,
      "loss": 0.2286,
      "num_input_tokens_seen": 1760480,
      "step": 8345
    },
    {
      "epoch": 0.9185918591859186,
      "grad_norm": 0.00372314453125,
      "learning_rate": 0.013777227722772277,
      "loss": 0.2357,
      "num_input_tokens_seen": 1761536,
      "step": 8350
    },
    {
      "epoch": 0.9191419141914191,
      "grad_norm": 0.002777099609375,
      "learning_rate": 0.013785478547854784,
      "loss": 0.2284,
      "num_input_tokens_seen": 1762624,
      "step": 8355
    },
    {
      "epoch": 0.9196919691969196,
      "grad_norm": 0.01263427734375,
      "learning_rate": 0.013793729372937293,
      "loss": 0.2286,
      "num_input_tokens_seen": 1763648,
      "step": 8360
    },
    {
      "epoch": 0.9202420242024203,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.0138019801980198,
      "loss": 0.2348,
      "num_input_tokens_seen": 1764672,
      "step": 8365
    },
    {
      "epoch": 0.9207920792079208,
      "grad_norm": 0.0234375,
      "learning_rate": 0.01381023102310231,
      "loss": 0.2337,
      "num_input_tokens_seen": 1765728,
      "step": 8370
    },
    {
      "epoch": 0.9213421342134214,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.013818481848184818,
      "loss": 0.2294,
      "num_input_tokens_seen": 1766784,
      "step": 8375
    },
    {
      "epoch": 0.9218921892189219,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.013826732673267327,
      "loss": 0.2273,
      "num_input_tokens_seen": 1767904,
      "step": 8380
    },
    {
      "epoch": 0.9224422442244224,
      "grad_norm": 0.0033721923828125,
      "learning_rate": 0.013834983498349836,
      "loss": 0.2315,
      "num_input_tokens_seen": 1768928,
      "step": 8385
    },
    {
      "epoch": 0.922992299229923,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.013843234323432343,
      "loss": 0.2303,
      "num_input_tokens_seen": 1770016,
      "step": 8390
    },
    {
      "epoch": 0.9235423542354235,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.013851485148514852,
      "loss": 0.2378,
      "num_input_tokens_seen": 1771040,
      "step": 8395
    },
    {
      "epoch": 0.9240924092409241,
      "grad_norm": 0.0224609375,
      "learning_rate": 0.013859735973597359,
      "loss": 0.2294,
      "num_input_tokens_seen": 1772096,
      "step": 8400
    },
    {
      "epoch": 0.9246424642464246,
      "grad_norm": 0.0034637451171875,
      "learning_rate": 0.013867986798679868,
      "loss": 0.2305,
      "num_input_tokens_seen": 1773184,
      "step": 8405
    },
    {
      "epoch": 0.9251925192519251,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.013876237623762375,
      "loss": 0.2307,
      "num_input_tokens_seen": 1774240,
      "step": 8410
    },
    {
      "epoch": 0.9257425742574258,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013884488448844884,
      "loss": 0.237,
      "num_input_tokens_seen": 1775360,
      "step": 8415
    },
    {
      "epoch": 0.9262926292629263,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.01389273927392739,
      "loss": 0.2257,
      "num_input_tokens_seen": 1776480,
      "step": 8420
    },
    {
      "epoch": 0.9268426842684269,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.013900990099009901,
      "loss": 0.2262,
      "num_input_tokens_seen": 1777536,
      "step": 8425
    },
    {
      "epoch": 0.9273927392739274,
      "grad_norm": 0.0250244140625,
      "learning_rate": 0.01390924092409241,
      "loss": 0.2334,
      "num_input_tokens_seen": 1778624,
      "step": 8430
    },
    {
      "epoch": 0.9279427942794279,
      "grad_norm": 0.003509521484375,
      "learning_rate": 0.013917491749174917,
      "loss": 0.2319,
      "num_input_tokens_seen": 1779616,
      "step": 8435
    },
    {
      "epoch": 0.9284928492849285,
      "grad_norm": 0.017333984375,
      "learning_rate": 0.013925742574257426,
      "loss": 0.2457,
      "num_input_tokens_seen": 1780672,
      "step": 8440
    },
    {
      "epoch": 0.929042904290429,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.013933993399339933,
      "loss": 0.2253,
      "num_input_tokens_seen": 1781792,
      "step": 8445
    },
    {
      "epoch": 0.9295929592959296,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013942244224422442,
      "loss": 0.2302,
      "num_input_tokens_seen": 1782816,
      "step": 8450
    },
    {
      "epoch": 0.9301430143014301,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.013950495049504949,
      "loss": 0.229,
      "num_input_tokens_seen": 1783936,
      "step": 8455
    },
    {
      "epoch": 0.9306930693069307,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.013958745874587458,
      "loss": 0.232,
      "num_input_tokens_seen": 1784992,
      "step": 8460
    },
    {
      "epoch": 0.9312431243124313,
      "grad_norm": 0.0296630859375,
      "learning_rate": 0.013966996699669965,
      "loss": 0.2336,
      "num_input_tokens_seen": 1786112,
      "step": 8465
    },
    {
      "epoch": 0.9317931793179318,
      "grad_norm": 0.00390625,
      "learning_rate": 0.013975247524752476,
      "loss": 0.2308,
      "num_input_tokens_seen": 1787136,
      "step": 8470
    },
    {
      "epoch": 0.9323432343234324,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.013983498349834984,
      "loss": 0.2328,
      "num_input_tokens_seen": 1788160,
      "step": 8475
    },
    {
      "epoch": 0.9328932893289329,
      "grad_norm": 0.0140380859375,
      "learning_rate": 0.013991749174917491,
      "loss": 0.2358,
      "num_input_tokens_seen": 1789184,
      "step": 8480
    },
    {
      "epoch": 0.9334433443344334,
      "grad_norm": 0.01416015625,
      "learning_rate": 0.014,
      "loss": 0.2299,
      "num_input_tokens_seen": 1790176,
      "step": 8485
    },
    {
      "epoch": 0.933993399339934,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.014008250825082507,
      "loss": 0.2309,
      "num_input_tokens_seen": 1791232,
      "step": 8490
    },
    {
      "epoch": 0.9345434543454345,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.014016501650165016,
      "loss": 0.2335,
      "num_input_tokens_seen": 1792224,
      "step": 8495
    },
    {
      "epoch": 0.935093509350935,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.014024752475247523,
      "loss": 0.233,
      "num_input_tokens_seen": 1793248,
      "step": 8500
    },
    {
      "epoch": 0.9356435643564357,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.014033003300330032,
      "loss": 0.2314,
      "num_input_tokens_seen": 1794336,
      "step": 8505
    },
    {
      "epoch": 0.9361936193619362,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.01404125412541254,
      "loss": 0.2326,
      "num_input_tokens_seen": 1795424,
      "step": 8510
    },
    {
      "epoch": 0.9367436743674368,
      "grad_norm": 0.01507568359375,
      "learning_rate": 0.01404950495049505,
      "loss": 0.2347,
      "num_input_tokens_seen": 1796480,
      "step": 8515
    },
    {
      "epoch": 0.9372937293729373,
      "grad_norm": 0.0233154296875,
      "learning_rate": 0.014057755775577559,
      "loss": 0.232,
      "num_input_tokens_seen": 1797600,
      "step": 8520
    },
    {
      "epoch": 0.9378437843784379,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.014066006600660066,
      "loss": 0.2336,
      "num_input_tokens_seen": 1798656,
      "step": 8525
    },
    {
      "epoch": 0.9383938393839384,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.014074257425742575,
      "loss": 0.231,
      "num_input_tokens_seen": 1799744,
      "step": 8530
    },
    {
      "epoch": 0.9389438943894389,
      "grad_norm": 0.0030517578125,
      "learning_rate": 0.014082508250825082,
      "loss": 0.2304,
      "num_input_tokens_seen": 1800768,
      "step": 8535
    },
    {
      "epoch": 0.9394939493949395,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.01409075907590759,
      "loss": 0.231,
      "num_input_tokens_seen": 1801792,
      "step": 8540
    },
    {
      "epoch": 0.94004400440044,
      "grad_norm": 0.0255126953125,
      "learning_rate": 0.014099009900990098,
      "loss": 0.2336,
      "num_input_tokens_seen": 1802848,
      "step": 8545
    },
    {
      "epoch": 0.9405940594059405,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.014107260726072606,
      "loss": 0.232,
      "num_input_tokens_seen": 1803840,
      "step": 8550
    },
    {
      "epoch": 0.9411441144114412,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.014115511551155114,
      "loss": 0.232,
      "num_input_tokens_seen": 1804928,
      "step": 8555
    },
    {
      "epoch": 0.9416941694169417,
      "grad_norm": 0.004180908203125,
      "learning_rate": 0.014123762376237622,
      "loss": 0.2319,
      "num_input_tokens_seen": 1805888,
      "step": 8560
    },
    {
      "epoch": 0.9422442244224423,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.014132013201320133,
      "loss": 0.234,
      "num_input_tokens_seen": 1807008,
      "step": 8565
    },
    {
      "epoch": 0.9427942794279428,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.01414026402640264,
      "loss": 0.2308,
      "num_input_tokens_seen": 1808000,
      "step": 8570
    },
    {
      "epoch": 0.9433443344334433,
      "grad_norm": 0.0034637451171875,
      "learning_rate": 0.014148514851485149,
      "loss": 0.2308,
      "num_input_tokens_seen": 1809088,
      "step": 8575
    },
    {
      "epoch": 0.9438943894389439,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.014156765676567656,
      "loss": 0.2319,
      "num_input_tokens_seen": 1810208,
      "step": 8580
    },
    {
      "epoch": 0.9444444444444444,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.014165016501650165,
      "loss": 0.2314,
      "num_input_tokens_seen": 1811264,
      "step": 8585
    },
    {
      "epoch": 0.944994499449945,
      "grad_norm": 0.01953125,
      "learning_rate": 0.014173267326732672,
      "loss": 0.2313,
      "num_input_tokens_seen": 1812384,
      "step": 8590
    },
    {
      "epoch": 0.9455445544554455,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01418151815181518,
      "loss": 0.2293,
      "num_input_tokens_seen": 1813472,
      "step": 8595
    },
    {
      "epoch": 0.9460946094609461,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.014189768976897688,
      "loss": 0.2319,
      "num_input_tokens_seen": 1814592,
      "step": 8600
    },
    {
      "epoch": 0.9466446644664467,
      "grad_norm": 0.0213623046875,
      "learning_rate": 0.014198019801980197,
      "loss": 0.2341,
      "num_input_tokens_seen": 1815584,
      "step": 8605
    },
    {
      "epoch": 0.9471947194719472,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.014206270627062707,
      "loss": 0.2315,
      "num_input_tokens_seen": 1816640,
      "step": 8610
    },
    {
      "epoch": 0.9477447744774478,
      "grad_norm": 0.009765625,
      "learning_rate": 0.014214521452145214,
      "loss": 0.2284,
      "num_input_tokens_seen": 1817696,
      "step": 8615
    },
    {
      "epoch": 0.9482948294829483,
      "grad_norm": 0.01275634765625,
      "learning_rate": 0.014222772277227723,
      "loss": 0.233,
      "num_input_tokens_seen": 1818848,
      "step": 8620
    },
    {
      "epoch": 0.9488448844884488,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.01423102310231023,
      "loss": 0.2283,
      "num_input_tokens_seen": 1819936,
      "step": 8625
    },
    {
      "epoch": 0.9493949394939494,
      "grad_norm": 0.004364013671875,
      "learning_rate": 0.01423927392739274,
      "loss": 0.2304,
      "num_input_tokens_seen": 1821024,
      "step": 8630
    },
    {
      "epoch": 0.9499449944994499,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.014247524752475246,
      "loss": 0.2319,
      "num_input_tokens_seen": 1822048,
      "step": 8635
    },
    {
      "epoch": 0.9504950495049505,
      "grad_norm": 0.003204345703125,
      "learning_rate": 0.014255775577557755,
      "loss": 0.233,
      "num_input_tokens_seen": 1823168,
      "step": 8640
    },
    {
      "epoch": 0.9510451045104511,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.014264026402640262,
      "loss": 0.2324,
      "num_input_tokens_seen": 1824256,
      "step": 8645
    },
    {
      "epoch": 0.9515951595159516,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.014272277227722771,
      "loss": 0.2288,
      "num_input_tokens_seen": 1825376,
      "step": 8650
    },
    {
      "epoch": 0.9521452145214522,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01428052805280528,
      "loss": 0.2314,
      "num_input_tokens_seen": 1826400,
      "step": 8655
    },
    {
      "epoch": 0.9526952695269527,
      "grad_norm": 0.00390625,
      "learning_rate": 0.014288778877887789,
      "loss": 0.2303,
      "num_input_tokens_seen": 1827520,
      "step": 8660
    },
    {
      "epoch": 0.9532453245324533,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.014297029702970298,
      "loss": 0.2309,
      "num_input_tokens_seen": 1828544,
      "step": 8665
    },
    {
      "epoch": 0.9537953795379538,
      "grad_norm": 0.0037384033203125,
      "learning_rate": 0.014305280528052805,
      "loss": 0.2352,
      "num_input_tokens_seen": 1829568,
      "step": 8670
    },
    {
      "epoch": 0.9543454345434543,
      "grad_norm": 0.00421142578125,
      "learning_rate": 0.014313531353135314,
      "loss": 0.233,
      "num_input_tokens_seen": 1830656,
      "step": 8675
    },
    {
      "epoch": 0.9548954895489549,
      "grad_norm": 0.0186767578125,
      "learning_rate": 0.01432178217821782,
      "loss": 0.2283,
      "num_input_tokens_seen": 1831680,
      "step": 8680
    },
    {
      "epoch": 0.9554455445544554,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.01433003300330033,
      "loss": 0.2361,
      "num_input_tokens_seen": 1832736,
      "step": 8685
    },
    {
      "epoch": 0.9559955995599559,
      "grad_norm": 0.0037994384765625,
      "learning_rate": 0.014338283828382837,
      "loss": 0.233,
      "num_input_tokens_seen": 1833856,
      "step": 8690
    },
    {
      "epoch": 0.9565456545654566,
      "grad_norm": 0.003936767578125,
      "learning_rate": 0.014346534653465345,
      "loss": 0.2325,
      "num_input_tokens_seen": 1834848,
      "step": 8695
    },
    {
      "epoch": 0.9570957095709571,
      "grad_norm": 0.004180908203125,
      "learning_rate": 0.014354785478547854,
      "loss": 0.2304,
      "num_input_tokens_seen": 1835936,
      "step": 8700
    },
    {
      "epoch": 0.9576457645764577,
      "grad_norm": 0.00408935546875,
      "learning_rate": 0.014363036303630363,
      "loss": 0.2325,
      "num_input_tokens_seen": 1836992,
      "step": 8705
    },
    {
      "epoch": 0.9581958195819582,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.014371287128712872,
      "loss": 0.2335,
      "num_input_tokens_seen": 1837984,
      "step": 8710
    },
    {
      "epoch": 0.9587458745874587,
      "grad_norm": 0.0186767578125,
      "learning_rate": 0.014379537953795379,
      "loss": 0.2319,
      "num_input_tokens_seen": 1839040,
      "step": 8715
    },
    {
      "epoch": 0.9592959295929593,
      "grad_norm": 0.0031890869140625,
      "learning_rate": 0.014387788778877888,
      "loss": 0.2324,
      "num_input_tokens_seen": 1840128,
      "step": 8720
    },
    {
      "epoch": 0.9598459845984598,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.014396039603960395,
      "loss": 0.2287,
      "num_input_tokens_seen": 1841152,
      "step": 8725
    },
    {
      "epoch": 0.9603960396039604,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.014404290429042904,
      "loss": 0.2303,
      "num_input_tokens_seen": 1842272,
      "step": 8730
    },
    {
      "epoch": 0.9609460946094609,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.014412541254125411,
      "loss": 0.2325,
      "num_input_tokens_seen": 1843296,
      "step": 8735
    },
    {
      "epoch": 0.9614961496149615,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.01442079207920792,
      "loss": 0.232,
      "num_input_tokens_seen": 1844352,
      "step": 8740
    },
    {
      "epoch": 0.9620462046204621,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.014429042904290429,
      "loss": 0.2299,
      "num_input_tokens_seen": 1845408,
      "step": 8745
    },
    {
      "epoch": 0.9625962596259626,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.014437293729372937,
      "loss": 0.2284,
      "num_input_tokens_seen": 1846400,
      "step": 8750
    },
    {
      "epoch": 0.9631463146314632,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.014445544554455446,
      "loss": 0.2325,
      "num_input_tokens_seen": 1847456,
      "step": 8755
    },
    {
      "epoch": 0.9636963696369637,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.014453795379537953,
      "loss": 0.2309,
      "num_input_tokens_seen": 1848416,
      "step": 8760
    },
    {
      "epoch": 0.9642464246424642,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.014462046204620462,
      "loss": 0.2331,
      "num_input_tokens_seen": 1849472,
      "step": 8765
    },
    {
      "epoch": 0.9647964796479648,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.01447029702970297,
      "loss": 0.232,
      "num_input_tokens_seen": 1850528,
      "step": 8770
    },
    {
      "epoch": 0.9653465346534653,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.014478547854785478,
      "loss": 0.2309,
      "num_input_tokens_seen": 1851680,
      "step": 8775
    },
    {
      "epoch": 0.9658965896589659,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.014486798679867985,
      "loss": 0.231,
      "num_input_tokens_seen": 1852672,
      "step": 8780
    },
    {
      "epoch": 0.9664466446644664,
      "grad_norm": 0.0036163330078125,
      "learning_rate": 0.014495049504950494,
      "loss": 0.2351,
      "num_input_tokens_seen": 1853696,
      "step": 8785
    },
    {
      "epoch": 0.966996699669967,
      "grad_norm": 0.0038299560546875,
      "learning_rate": 0.014503300330033003,
      "loss": 0.2361,
      "num_input_tokens_seen": 1854688,
      "step": 8790
    },
    {
      "epoch": 0.9675467546754676,
      "grad_norm": 0.019287109375,
      "learning_rate": 0.01451155115511551,
      "loss": 0.2293,
      "num_input_tokens_seen": 1855712,
      "step": 8795
    },
    {
      "epoch": 0.9680968096809681,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01451980198019802,
      "loss": 0.2308,
      "num_input_tokens_seen": 1856832,
      "step": 8800
    },
    {
      "epoch": 0.9686468646864687,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.014528052805280528,
      "loss": 0.2313,
      "num_input_tokens_seen": 1857920,
      "step": 8805
    },
    {
      "epoch": 0.9691969196919692,
      "grad_norm": 0.0196533203125,
      "learning_rate": 0.014536303630363036,
      "loss": 0.2318,
      "num_input_tokens_seen": 1858944,
      "step": 8810
    },
    {
      "epoch": 0.9697469746974697,
      "grad_norm": 0.004547119140625,
      "learning_rate": 0.014544554455445544,
      "loss": 0.2318,
      "num_input_tokens_seen": 1860000,
      "step": 8815
    },
    {
      "epoch": 0.9702970297029703,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.014552805280528052,
      "loss": 0.2313,
      "num_input_tokens_seen": 1861056,
      "step": 8820
    },
    {
      "epoch": 0.9708470847084708,
      "grad_norm": 0.02001953125,
      "learning_rate": 0.01456105610561056,
      "loss": 0.2324,
      "num_input_tokens_seen": 1862080,
      "step": 8825
    },
    {
      "epoch": 0.9713971397139713,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014569306930693068,
      "loss": 0.2292,
      "num_input_tokens_seen": 1863104,
      "step": 8830
    },
    {
      "epoch": 0.971947194719472,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.014577557755775577,
      "loss": 0.2303,
      "num_input_tokens_seen": 1864128,
      "step": 8835
    },
    {
      "epoch": 0.9724972497249725,
      "grad_norm": 0.0030517578125,
      "learning_rate": 0.014585808580858084,
      "loss": 0.2313,
      "num_input_tokens_seen": 1865184,
      "step": 8840
    },
    {
      "epoch": 0.9730473047304731,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.014594059405940595,
      "loss": 0.2308,
      "num_input_tokens_seen": 1866240,
      "step": 8845
    },
    {
      "epoch": 0.9735973597359736,
      "grad_norm": 0.00439453125,
      "learning_rate": 0.014602310231023102,
      "loss": 0.2313,
      "num_input_tokens_seen": 1867360,
      "step": 8850
    },
    {
      "epoch": 0.9741474147414741,
      "grad_norm": 0.0037689208984375,
      "learning_rate": 0.01461056105610561,
      "loss": 0.2319,
      "num_input_tokens_seen": 1868480,
      "step": 8855
    },
    {
      "epoch": 0.9746974697469747,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.014618811881188118,
      "loss": 0.2318,
      "num_input_tokens_seen": 1869440,
      "step": 8860
    },
    {
      "epoch": 0.9752475247524752,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.014627062706270627,
      "loss": 0.2324,
      "num_input_tokens_seen": 1870496,
      "step": 8865
    },
    {
      "epoch": 0.9757975797579758,
      "grad_norm": 0.009765625,
      "learning_rate": 0.014635313531353134,
      "loss": 0.2287,
      "num_input_tokens_seen": 1871616,
      "step": 8870
    },
    {
      "epoch": 0.9763476347634763,
      "grad_norm": 0.004241943359375,
      "learning_rate": 0.014643564356435643,
      "loss": 0.2313,
      "num_input_tokens_seen": 1872672,
      "step": 8875
    },
    {
      "epoch": 0.976897689768977,
      "grad_norm": 0.019775390625,
      "learning_rate": 0.014651815181518152,
      "loss": 0.2308,
      "num_input_tokens_seen": 1873760,
      "step": 8880
    },
    {
      "epoch": 0.9774477447744775,
      "grad_norm": 0.019775390625,
      "learning_rate": 0.014660066006600659,
      "loss": 0.2334,
      "num_input_tokens_seen": 1874784,
      "step": 8885
    },
    {
      "epoch": 0.977997799779978,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.01466831683168317,
      "loss": 0.2324,
      "num_input_tokens_seen": 1875808,
      "step": 8890
    },
    {
      "epoch": 0.9785478547854786,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.014676567656765676,
      "loss": 0.2308,
      "num_input_tokens_seen": 1876832,
      "step": 8895
    },
    {
      "epoch": 0.9790979097909791,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.014684818481848185,
      "loss": 0.2308,
      "num_input_tokens_seen": 1877792,
      "step": 8900
    },
    {
      "epoch": 0.9796479647964796,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.014693069306930692,
      "loss": 0.2324,
      "num_input_tokens_seen": 1878880,
      "step": 8905
    },
    {
      "epoch": 0.9801980198019802,
      "grad_norm": 0.00335693359375,
      "learning_rate": 0.014701320132013201,
      "loss": 0.2308,
      "num_input_tokens_seen": 1879936,
      "step": 8910
    },
    {
      "epoch": 0.9807480748074807,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.014709570957095708,
      "loss": 0.2324,
      "num_input_tokens_seen": 1880992,
      "step": 8915
    },
    {
      "epoch": 0.9812981298129813,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.014717821782178217,
      "loss": 0.2314,
      "num_input_tokens_seen": 1882048,
      "step": 8920
    },
    {
      "epoch": 0.9818481848184818,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.014726072607260726,
      "loss": 0.2309,
      "num_input_tokens_seen": 1883104,
      "step": 8925
    },
    {
      "epoch": 0.9823982398239824,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.014734323432343233,
      "loss": 0.233,
      "num_input_tokens_seen": 1884224,
      "step": 8930
    },
    {
      "epoch": 0.982948294829483,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.014742574257425742,
      "loss": 0.2325,
      "num_input_tokens_seen": 1885248,
      "step": 8935
    },
    {
      "epoch": 0.9834983498349835,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.01475082508250825,
      "loss": 0.2314,
      "num_input_tokens_seen": 1886304,
      "step": 8940
    },
    {
      "epoch": 0.9840484048404841,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.01475907590759076,
      "loss": 0.2319,
      "num_input_tokens_seen": 1887360,
      "step": 8945
    },
    {
      "epoch": 0.9845984598459846,
      "grad_norm": 0.0029296875,
      "learning_rate": 0.014767326732673267,
      "loss": 0.2298,
      "num_input_tokens_seen": 1888416,
      "step": 8950
    },
    {
      "epoch": 0.9851485148514851,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.014775577557755775,
      "loss": 0.2314,
      "num_input_tokens_seen": 1889472,
      "step": 8955
    },
    {
      "epoch": 0.9856985698569857,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.014783828382838283,
      "loss": 0.2309,
      "num_input_tokens_seen": 1890592,
      "step": 8960
    },
    {
      "epoch": 0.9862486248624862,
      "grad_norm": 0.002655029296875,
      "learning_rate": 0.014792079207920791,
      "loss": 0.2293,
      "num_input_tokens_seen": 1891648,
      "step": 8965
    },
    {
      "epoch": 0.9867986798679867,
      "grad_norm": 0.0189208984375,
      "learning_rate": 0.0148003300330033,
      "loss": 0.2314,
      "num_input_tokens_seen": 1892672,
      "step": 8970
    },
    {
      "epoch": 0.9873487348734874,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.014808580858085807,
      "loss": 0.2308,
      "num_input_tokens_seen": 1893760,
      "step": 8975
    },
    {
      "epoch": 0.9878987898789879,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.014816831683168316,
      "loss": 0.2319,
      "num_input_tokens_seen": 1894816,
      "step": 8980
    },
    {
      "epoch": 0.9884488448844885,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.014825082508250825,
      "loss": 0.2308,
      "num_input_tokens_seen": 1895904,
      "step": 8985
    },
    {
      "epoch": 0.988998899889989,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.014833333333333334,
      "loss": 0.2308,
      "num_input_tokens_seen": 1896992,
      "step": 8990
    },
    {
      "epoch": 0.9895489548954896,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.014841584158415841,
      "loss": 0.2324,
      "num_input_tokens_seen": 1898016,
      "step": 8995
    },
    {
      "epoch": 0.9900990099009901,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.01484983498349835,
      "loss": 0.2298,
      "num_input_tokens_seen": 1899104,
      "step": 9000
    },
    {
      "epoch": 0.9906490649064906,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.014858085808580857,
      "loss": 0.2288,
      "num_input_tokens_seen": 1900128,
      "step": 9005
    },
    {
      "epoch": 0.9911991199119912,
      "grad_norm": 0.029296875,
      "learning_rate": 0.014866336633663366,
      "loss": 0.227,
      "num_input_tokens_seen": 1901184,
      "step": 9010
    },
    {
      "epoch": 0.9917491749174917,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014874587458745874,
      "loss": 0.2329,
      "num_input_tokens_seen": 1902272,
      "step": 9015
    },
    {
      "epoch": 0.9922992299229924,
      "grad_norm": 0.025634765625,
      "learning_rate": 0.014882838283828382,
      "loss": 0.2305,
      "num_input_tokens_seen": 1903264,
      "step": 9020
    },
    {
      "epoch": 0.9928492849284929,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.01489108910891089,
      "loss": 0.2281,
      "num_input_tokens_seen": 1904288,
      "step": 9025
    },
    {
      "epoch": 0.9933993399339934,
      "grad_norm": 0.08154296875,
      "learning_rate": 0.014899339933993398,
      "loss": 0.2228,
      "num_input_tokens_seen": 1905408,
      "step": 9030
    },
    {
      "epoch": 0.993949394939494,
      "grad_norm": 0.1279296875,
      "learning_rate": 0.014907590759075908,
      "loss": 0.2338,
      "num_input_tokens_seen": 1906464,
      "step": 9035
    },
    {
      "epoch": 0.9944994499449945,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.014915841584158415,
      "loss": 0.223,
      "num_input_tokens_seen": 1907552,
      "step": 9040
    },
    {
      "epoch": 0.995049504950495,
      "grad_norm": 0.01324462890625,
      "learning_rate": 0.014924092409240924,
      "loss": 0.235,
      "num_input_tokens_seen": 1908576,
      "step": 9045
    },
    {
      "epoch": 0.9955995599559956,
      "grad_norm": 0.0038604736328125,
      "learning_rate": 0.014932343234323431,
      "loss": 0.2311,
      "num_input_tokens_seen": 1909664,
      "step": 9050
    },
    {
      "epoch": 0.9961496149614961,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01494059405940594,
      "loss": 0.2287,
      "num_input_tokens_seen": 1910720,
      "step": 9055
    },
    {
      "epoch": 0.9966996699669967,
      "grad_norm": 0.01361083984375,
      "learning_rate": 0.014948844884488449,
      "loss": 0.2361,
      "num_input_tokens_seen": 1911840,
      "step": 9060
    },
    {
      "epoch": 0.9972497249724972,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.014957095709570956,
      "loss": 0.2308,
      "num_input_tokens_seen": 1912896,
      "step": 9065
    },
    {
      "epoch": 0.9977997799779978,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.014965346534653465,
      "loss": 0.2286,
      "num_input_tokens_seen": 1913952,
      "step": 9070
    },
    {
      "epoch": 0.9983498349834984,
      "grad_norm": 0.014892578125,
      "learning_rate": 0.014973597359735972,
      "loss": 0.2411,
      "num_input_tokens_seen": 1915008,
      "step": 9075
    },
    {
      "epoch": 0.9988998899889989,
      "grad_norm": 0.027587890625,
      "learning_rate": 0.014981848184818482,
      "loss": 0.2325,
      "num_input_tokens_seen": 1916064,
      "step": 9080
    },
    {
      "epoch": 0.9994499449944995,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.01499009900990099,
      "loss": 0.2304,
      "num_input_tokens_seen": 1917056,
      "step": 9085
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.029296875,
      "learning_rate": 0.014998349834983498,
      "loss": 0.2313,
      "num_input_tokens_seen": 1917952,
      "step": 9090
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.23137536644935608,
      "eval_runtime": 60.5584,
      "eval_samples_per_second": 66.712,
      "eval_steps_per_second": 16.678,
      "num_input_tokens_seen": 1917952,
      "step": 9090
    },
    {
      "epoch": 1.0005500550055006,
      "grad_norm": 0.026611328125,
      "learning_rate": 0.015006600660066004,
      "loss": 0.2315,
      "num_input_tokens_seen": 1919008,
      "step": 9095
    },
    {
      "epoch": 1.001100110011001,
      "grad_norm": 0.0272216796875,
      "learning_rate": 0.015014851485148514,
      "loss": 0.2326,
      "num_input_tokens_seen": 1920032,
      "step": 9100
    },
    {
      "epoch": 1.0016501650165017,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.015023102310231025,
      "loss": 0.2293,
      "num_input_tokens_seen": 1921056,
      "step": 9105
    },
    {
      "epoch": 1.0022002200220022,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.01503135313531353,
      "loss": 0.2283,
      "num_input_tokens_seen": 1922112,
      "step": 9110
    },
    {
      "epoch": 1.0027502750275028,
      "grad_norm": 0.0147705078125,
      "learning_rate": 0.01503960396039604,
      "loss": 0.2304,
      "num_input_tokens_seen": 1923136,
      "step": 9115
    },
    {
      "epoch": 1.0033003300330032,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.015047854785478546,
      "loss": 0.2302,
      "num_input_tokens_seen": 1924160,
      "step": 9120
    },
    {
      "epoch": 1.0038503850385039,
      "grad_norm": 0.01397705078125,
      "learning_rate": 0.015056105610561057,
      "loss": 0.2314,
      "num_input_tokens_seen": 1925248,
      "step": 9125
    },
    {
      "epoch": 1.0044004400440043,
      "grad_norm": 0.025634765625,
      "learning_rate": 0.015064356435643562,
      "loss": 0.2302,
      "num_input_tokens_seen": 1926240,
      "step": 9130
    },
    {
      "epoch": 1.004950495049505,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.015072607260726073,
      "loss": 0.2263,
      "num_input_tokens_seen": 1927296,
      "step": 9135
    },
    {
      "epoch": 1.0055005500550056,
      "grad_norm": 0.01300048828125,
      "learning_rate": 0.015080858085808578,
      "loss": 0.2303,
      "num_input_tokens_seen": 1928320,
      "step": 9140
    },
    {
      "epoch": 1.006050605060506,
      "grad_norm": 0.0240478515625,
      "learning_rate": 0.015089108910891089,
      "loss": 0.2306,
      "num_input_tokens_seen": 1929408,
      "step": 9145
    },
    {
      "epoch": 1.0066006600660067,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.0150973597359736,
      "loss": 0.2314,
      "num_input_tokens_seen": 1930496,
      "step": 9150
    },
    {
      "epoch": 1.007150715071507,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.015105610561056105,
      "loss": 0.2292,
      "num_input_tokens_seen": 1931584,
      "step": 9155
    },
    {
      "epoch": 1.0077007700770078,
      "grad_norm": 0.0033111572265625,
      "learning_rate": 0.015113861386138615,
      "loss": 0.237,
      "num_input_tokens_seen": 1932576,
      "step": 9160
    },
    {
      "epoch": 1.0082508250825082,
      "grad_norm": 0.0198974609375,
      "learning_rate": 0.01512211221122112,
      "loss": 0.222,
      "num_input_tokens_seen": 1933600,
      "step": 9165
    },
    {
      "epoch": 1.0088008800880088,
      "grad_norm": 0.0035858154296875,
      "learning_rate": 0.015130363036303631,
      "loss": 0.2381,
      "num_input_tokens_seen": 1934592,
      "step": 9170
    },
    {
      "epoch": 1.0093509350935093,
      "grad_norm": 0.0038909912109375,
      "learning_rate": 0.015138613861386136,
      "loss": 0.2345,
      "num_input_tokens_seen": 1935584,
      "step": 9175
    },
    {
      "epoch": 1.00990099009901,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.015146864686468647,
      "loss": 0.2283,
      "num_input_tokens_seen": 1936640,
      "step": 9180
    },
    {
      "epoch": 1.0104510451045106,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.015155115511551152,
      "loss": 0.2333,
      "num_input_tokens_seen": 1937664,
      "step": 9185
    },
    {
      "epoch": 1.011001100110011,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.015163366336633663,
      "loss": 0.2303,
      "num_input_tokens_seen": 1938720,
      "step": 9190
    },
    {
      "epoch": 1.0115511551155116,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.015171617161716174,
      "loss": 0.2339,
      "num_input_tokens_seen": 1939744,
      "step": 9195
    },
    {
      "epoch": 1.012101210121012,
      "grad_norm": 0.0030975341796875,
      "learning_rate": 0.015179867986798679,
      "loss": 0.2308,
      "num_input_tokens_seen": 1940864,
      "step": 9200
    },
    {
      "epoch": 1.0126512651265127,
      "grad_norm": 0.0220947265625,
      "learning_rate": 0.01518811881188119,
      "loss": 0.2422,
      "num_input_tokens_seen": 1941888,
      "step": 9205
    },
    {
      "epoch": 1.0132013201320131,
      "grad_norm": 0.0036773681640625,
      "learning_rate": 0.015196369636963695,
      "loss": 0.2342,
      "num_input_tokens_seen": 1942912,
      "step": 9210
    },
    {
      "epoch": 1.0137513751375138,
      "grad_norm": 0.00439453125,
      "learning_rate": 0.015204620462046205,
      "loss": 0.2325,
      "num_input_tokens_seen": 1943968,
      "step": 9215
    },
    {
      "epoch": 1.0143014301430142,
      "grad_norm": 0.0247802734375,
      "learning_rate": 0.01521287128712871,
      "loss": 0.2319,
      "num_input_tokens_seen": 1944992,
      "step": 9220
    },
    {
      "epoch": 1.0148514851485149,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.015221122112211221,
      "loss": 0.2314,
      "num_input_tokens_seen": 1946048,
      "step": 9225
    },
    {
      "epoch": 1.0154015401540153,
      "grad_norm": 0.02294921875,
      "learning_rate": 0.015229372937293727,
      "loss": 0.2314,
      "num_input_tokens_seen": 1947008,
      "step": 9230
    },
    {
      "epoch": 1.015951595159516,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.015237623762376237,
      "loss": 0.2329,
      "num_input_tokens_seen": 1948032,
      "step": 9235
    },
    {
      "epoch": 1.0165016501650166,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.015245874587458744,
      "loss": 0.2319,
      "num_input_tokens_seen": 1949056,
      "step": 9240
    },
    {
      "epoch": 1.017051705170517,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.015254125412541253,
      "loss": 0.2278,
      "num_input_tokens_seen": 1950112,
      "step": 9245
    },
    {
      "epoch": 1.0176017601760177,
      "grad_norm": 0.01495361328125,
      "learning_rate": 0.015262376237623764,
      "loss": 0.2338,
      "num_input_tokens_seen": 1951200,
      "step": 9250
    },
    {
      "epoch": 1.018151815181518,
      "grad_norm": 0.028564453125,
      "learning_rate": 0.01527062706270627,
      "loss": 0.2386,
      "num_input_tokens_seen": 1952224,
      "step": 9255
    },
    {
      "epoch": 1.0187018701870187,
      "grad_norm": 0.005859375,
      "learning_rate": 0.01527887788778878,
      "loss": 0.2322,
      "num_input_tokens_seen": 1953312,
      "step": 9260
    },
    {
      "epoch": 1.0192519251925192,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.015287128712871285,
      "loss": 0.234,
      "num_input_tokens_seen": 1954336,
      "step": 9265
    },
    {
      "epoch": 1.0198019801980198,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.015295379537953796,
      "loss": 0.2319,
      "num_input_tokens_seen": 1955392,
      "step": 9270
    },
    {
      "epoch": 1.0203520352035202,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.015303630363036301,
      "loss": 0.2319,
      "num_input_tokens_seen": 1956416,
      "step": 9275
    },
    {
      "epoch": 1.020902090209021,
      "grad_norm": 0.003997802734375,
      "learning_rate": 0.015311881188118812,
      "loss": 0.2293,
      "num_input_tokens_seen": 1957408,
      "step": 9280
    },
    {
      "epoch": 1.0214521452145215,
      "grad_norm": 0.004058837890625,
      "learning_rate": 0.015320132013201319,
      "loss": 0.2294,
      "num_input_tokens_seen": 1958464,
      "step": 9285
    },
    {
      "epoch": 1.022002200220022,
      "grad_norm": 0.003082275390625,
      "learning_rate": 0.015328382838283828,
      "loss": 0.233,
      "num_input_tokens_seen": 1959552,
      "step": 9290
    },
    {
      "epoch": 1.0225522552255226,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.015336633663366338,
      "loss": 0.233,
      "num_input_tokens_seen": 1960544,
      "step": 9295
    },
    {
      "epoch": 1.023102310231023,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.015344884488448843,
      "loss": 0.2283,
      "num_input_tokens_seen": 1961600,
      "step": 9300
    },
    {
      "epoch": 1.0236523652365237,
      "grad_norm": 0.0033721923828125,
      "learning_rate": 0.015353135313531354,
      "loss": 0.233,
      "num_input_tokens_seen": 1962688,
      "step": 9305
    },
    {
      "epoch": 1.0242024202420241,
      "grad_norm": 0.003173828125,
      "learning_rate": 0.01536138613861386,
      "loss": 0.2314,
      "num_input_tokens_seen": 1963744,
      "step": 9310
    },
    {
      "epoch": 1.0247524752475248,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01536963696369637,
      "loss": 0.2329,
      "num_input_tokens_seen": 1964832,
      "step": 9315
    },
    {
      "epoch": 1.0253025302530252,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.015377887788778875,
      "loss": 0.2324,
      "num_input_tokens_seen": 1965920,
      "step": 9320
    },
    {
      "epoch": 1.0258525852585259,
      "grad_norm": 0.00927734375,
      "learning_rate": 0.015386138613861386,
      "loss": 0.2314,
      "num_input_tokens_seen": 1966912,
      "step": 9325
    },
    {
      "epoch": 1.0264026402640265,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.015394389438943893,
      "loss": 0.2298,
      "num_input_tokens_seen": 1968096,
      "step": 9330
    },
    {
      "epoch": 1.026952695269527,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.015402640264026402,
      "loss": 0.23,
      "num_input_tokens_seen": 1969216,
      "step": 9335
    },
    {
      "epoch": 1.0275027502750276,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.015410891089108912,
      "loss": 0.2326,
      "num_input_tokens_seen": 1970304,
      "step": 9340
    },
    {
      "epoch": 1.028052805280528,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.015419141914191418,
      "loss": 0.2249,
      "num_input_tokens_seen": 1971360,
      "step": 9345
    },
    {
      "epoch": 1.0286028602860287,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.015427392739273928,
      "loss": 0.2334,
      "num_input_tokens_seen": 1972416,
      "step": 9350
    },
    {
      "epoch": 1.029152915291529,
      "grad_norm": 0.01904296875,
      "learning_rate": 0.015435643564356434,
      "loss": 0.2354,
      "num_input_tokens_seen": 1973440,
      "step": 9355
    },
    {
      "epoch": 1.0297029702970297,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.015443894389438944,
      "loss": 0.2327,
      "num_input_tokens_seen": 1974432,
      "step": 9360
    },
    {
      "epoch": 1.0302530253025302,
      "grad_norm": 0.0038604736328125,
      "learning_rate": 0.01545214521452145,
      "loss": 0.2346,
      "num_input_tokens_seen": 1975488,
      "step": 9365
    },
    {
      "epoch": 1.0308030803080308,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.01546039603960396,
      "loss": 0.2304,
      "num_input_tokens_seen": 1976576,
      "step": 9370
    },
    {
      "epoch": 1.0313531353135315,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.015468646864686467,
      "loss": 0.2335,
      "num_input_tokens_seen": 1977632,
      "step": 9375
    },
    {
      "epoch": 1.0319031903190319,
      "grad_norm": 0.00390625,
      "learning_rate": 0.015476897689768976,
      "loss": 0.2309,
      "num_input_tokens_seen": 1978720,
      "step": 9380
    },
    {
      "epoch": 1.0324532453245325,
      "grad_norm": 0.018798828125,
      "learning_rate": 0.015485148514851487,
      "loss": 0.2335,
      "num_input_tokens_seen": 1979744,
      "step": 9385
    },
    {
      "epoch": 1.033003300330033,
      "grad_norm": 0.00341796875,
      "learning_rate": 0.015493399339933992,
      "loss": 0.2351,
      "num_input_tokens_seen": 1980832,
      "step": 9390
    },
    {
      "epoch": 1.0335533553355336,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.015501650165016503,
      "loss": 0.2324,
      "num_input_tokens_seen": 1981888,
      "step": 9395
    },
    {
      "epoch": 1.034103410341034,
      "grad_norm": 0.0030975341796875,
      "learning_rate": 0.015509900990099008,
      "loss": 0.2319,
      "num_input_tokens_seen": 1982944,
      "step": 9400
    },
    {
      "epoch": 1.0346534653465347,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.015518151815181519,
      "loss": 0.2309,
      "num_input_tokens_seen": 1984000,
      "step": 9405
    },
    {
      "epoch": 1.0352035203520351,
      "grad_norm": 0.0035247802734375,
      "learning_rate": 0.015526402640264024,
      "loss": 0.2319,
      "num_input_tokens_seen": 1985024,
      "step": 9410
    },
    {
      "epoch": 1.0357535753575358,
      "grad_norm": 0.00927734375,
      "learning_rate": 0.015534653465346535,
      "loss": 0.2298,
      "num_input_tokens_seen": 1986080,
      "step": 9415
    },
    {
      "epoch": 1.0363036303630364,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.015542904290429042,
      "loss": 0.2319,
      "num_input_tokens_seen": 1987104,
      "step": 9420
    },
    {
      "epoch": 1.0368536853685368,
      "grad_norm": 0.0185546875,
      "learning_rate": 0.01555115511551155,
      "loss": 0.234,
      "num_input_tokens_seen": 1988192,
      "step": 9425
    },
    {
      "epoch": 1.0374037403740375,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.015559405940594061,
      "loss": 0.2304,
      "num_input_tokens_seen": 1989216,
      "step": 9430
    },
    {
      "epoch": 1.037953795379538,
      "grad_norm": 0.009765625,
      "learning_rate": 0.015567656765676566,
      "loss": 0.2314,
      "num_input_tokens_seen": 1990272,
      "step": 9435
    },
    {
      "epoch": 1.0385038503850386,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.015575907590759077,
      "loss": 0.2304,
      "num_input_tokens_seen": 1991264,
      "step": 9440
    },
    {
      "epoch": 1.039053905390539,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.015584158415841582,
      "loss": 0.231,
      "num_input_tokens_seen": 1992288,
      "step": 9445
    },
    {
      "epoch": 1.0396039603960396,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.015592409240924093,
      "loss": 0.233,
      "num_input_tokens_seen": 1993344,
      "step": 9450
    },
    {
      "epoch": 1.04015401540154,
      "grad_norm": 0.00396728515625,
      "learning_rate": 0.015600660066006598,
      "loss": 0.2294,
      "num_input_tokens_seen": 1994464,
      "step": 9455
    },
    {
      "epoch": 1.0407040704070407,
      "grad_norm": 0.0030517578125,
      "learning_rate": 0.015608910891089109,
      "loss": 0.2356,
      "num_input_tokens_seen": 1995584,
      "step": 9460
    },
    {
      "epoch": 1.0412541254125411,
      "grad_norm": 0.0031585693359375,
      "learning_rate": 0.015617161716171616,
      "loss": 0.2298,
      "num_input_tokens_seen": 1996640,
      "step": 9465
    },
    {
      "epoch": 1.0418041804180418,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.015625412541254127,
      "loss": 0.2298,
      "num_input_tokens_seen": 1997696,
      "step": 9470
    },
    {
      "epoch": 1.0423542354235424,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.01563366336633663,
      "loss": 0.2314,
      "num_input_tokens_seen": 1998720,
      "step": 9475
    },
    {
      "epoch": 1.0429042904290429,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.01564191419141914,
      "loss": 0.2299,
      "num_input_tokens_seen": 1999840,
      "step": 9480
    },
    {
      "epoch": 1.0434543454345435,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.01565016501650165,
      "loss": 0.2315,
      "num_input_tokens_seen": 2000832,
      "step": 9485
    },
    {
      "epoch": 1.044004400440044,
      "grad_norm": 0.0035400390625,
      "learning_rate": 0.01565841584158416,
      "loss": 0.2299,
      "num_input_tokens_seen": 2001952,
      "step": 9490
    },
    {
      "epoch": 1.0445544554455446,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.015666666666666666,
      "loss": 0.2338,
      "num_input_tokens_seen": 2002976,
      "step": 9495
    },
    {
      "epoch": 1.045104510451045,
      "grad_norm": 0.0029144287109375,
      "learning_rate": 0.015674917491749173,
      "loss": 0.2347,
      "num_input_tokens_seen": 2004000,
      "step": 9500
    },
    {
      "epoch": 1.0456545654565457,
      "grad_norm": 0.004547119140625,
      "learning_rate": 0.015683168316831683,
      "loss": 0.23,
      "num_input_tokens_seen": 2005088,
      "step": 9505
    },
    {
      "epoch": 1.046204620462046,
      "grad_norm": 0.003662109375,
      "learning_rate": 0.01569141914191419,
      "loss": 0.2327,
      "num_input_tokens_seen": 2006144,
      "step": 9510
    },
    {
      "epoch": 1.0467546754675467,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0156996699669967,
      "loss": 0.2362,
      "num_input_tokens_seen": 2007200,
      "step": 9515
    },
    {
      "epoch": 1.0473047304730474,
      "grad_norm": 0.00421142578125,
      "learning_rate": 0.015707920792079205,
      "loss": 0.2264,
      "num_input_tokens_seen": 2008288,
      "step": 9520
    },
    {
      "epoch": 1.0478547854785478,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.015716171617161715,
      "loss": 0.2289,
      "num_input_tokens_seen": 2009344,
      "step": 9525
    },
    {
      "epoch": 1.0484048404840485,
      "grad_norm": 0.0030517578125,
      "learning_rate": 0.015724422442244226,
      "loss": 0.2326,
      "num_input_tokens_seen": 2010432,
      "step": 9530
    },
    {
      "epoch": 1.048954895489549,
      "grad_norm": 0.0208740234375,
      "learning_rate": 0.015732673267326733,
      "loss": 0.229,
      "num_input_tokens_seen": 2011392,
      "step": 9535
    },
    {
      "epoch": 1.0495049504950495,
      "grad_norm": 0.009765625,
      "learning_rate": 0.01574092409240924,
      "loss": 0.2331,
      "num_input_tokens_seen": 2012416,
      "step": 9540
    },
    {
      "epoch": 1.05005500550055,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.015749174917491747,
      "loss": 0.2238,
      "num_input_tokens_seen": 2013440,
      "step": 9545
    },
    {
      "epoch": 1.0506050605060506,
      "grad_norm": 0.00341796875,
      "learning_rate": 0.015757425742574258,
      "loss": 0.2323,
      "num_input_tokens_seen": 2014464,
      "step": 9550
    },
    {
      "epoch": 1.051155115511551,
      "grad_norm": 0.021240234375,
      "learning_rate": 0.015765676567656765,
      "loss": 0.2252,
      "num_input_tokens_seen": 2015584,
      "step": 9555
    },
    {
      "epoch": 1.0517051705170517,
      "grad_norm": 0.0269775390625,
      "learning_rate": 0.015773927392739275,
      "loss": 0.241,
      "num_input_tokens_seen": 2016576,
      "step": 9560
    },
    {
      "epoch": 1.0522552255225524,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.01578217821782178,
      "loss": 0.2275,
      "num_input_tokens_seen": 2017568,
      "step": 9565
    },
    {
      "epoch": 1.0528052805280528,
      "grad_norm": 0.01397705078125,
      "learning_rate": 0.01579042904290429,
      "loss": 0.2326,
      "num_input_tokens_seen": 2018656,
      "step": 9570
    },
    {
      "epoch": 1.0533553355335534,
      "grad_norm": 0.0135498046875,
      "learning_rate": 0.0157986798679868,
      "loss": 0.2358,
      "num_input_tokens_seen": 2019680,
      "step": 9575
    },
    {
      "epoch": 1.0539053905390539,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.015806930693069307,
      "loss": 0.2258,
      "num_input_tokens_seen": 2020736,
      "step": 9580
    },
    {
      "epoch": 1.0544554455445545,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.015815181518151814,
      "loss": 0.2278,
      "num_input_tokens_seen": 2021792,
      "step": 9585
    },
    {
      "epoch": 1.055005500550055,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.01582343234323432,
      "loss": 0.2315,
      "num_input_tokens_seen": 2022880,
      "step": 9590
    },
    {
      "epoch": 1.0555555555555556,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.015831683168316832,
      "loss": 0.2293,
      "num_input_tokens_seen": 2023904,
      "step": 9595
    },
    {
      "epoch": 1.056105610561056,
      "grad_norm": 0.00372314453125,
      "learning_rate": 0.01583993399339934,
      "loss": 0.2345,
      "num_input_tokens_seen": 2024992,
      "step": 9600
    },
    {
      "epoch": 1.0566556655665567,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.01584818481848185,
      "loss": 0.2341,
      "num_input_tokens_seen": 2026048,
      "step": 9605
    },
    {
      "epoch": 1.0572057205720573,
      "grad_norm": 0.002716064453125,
      "learning_rate": 0.015856435643564353,
      "loss": 0.2345,
      "num_input_tokens_seen": 2027040,
      "step": 9610
    },
    {
      "epoch": 1.0577557755775577,
      "grad_norm": 0.0032806396484375,
      "learning_rate": 0.015864686468646864,
      "loss": 0.2282,
      "num_input_tokens_seen": 2028128,
      "step": 9615
    },
    {
      "epoch": 1.0583058305830584,
      "grad_norm": 0.02001953125,
      "learning_rate": 0.015872937293729374,
      "loss": 0.2245,
      "num_input_tokens_seen": 2029184,
      "step": 9620
    },
    {
      "epoch": 1.0588558855885588,
      "grad_norm": 0.021240234375,
      "learning_rate": 0.01588118811881188,
      "loss": 0.2241,
      "num_input_tokens_seen": 2030304,
      "step": 9625
    },
    {
      "epoch": 1.0594059405940595,
      "grad_norm": 0.003173828125,
      "learning_rate": 0.01588943894389439,
      "loss": 0.2379,
      "num_input_tokens_seen": 2031424,
      "step": 9630
    },
    {
      "epoch": 1.0599559955995599,
      "grad_norm": 0.022705078125,
      "learning_rate": 0.015897689768976896,
      "loss": 0.2306,
      "num_input_tokens_seen": 2032384,
      "step": 9635
    },
    {
      "epoch": 1.0605060506050605,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.015905940594059406,
      "loss": 0.2347,
      "num_input_tokens_seen": 2033472,
      "step": 9640
    },
    {
      "epoch": 1.061056105610561,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.015914191419141913,
      "loss": 0.2405,
      "num_input_tokens_seen": 2034496,
      "step": 9645
    },
    {
      "epoch": 1.0616061606160616,
      "grad_norm": 0.020751953125,
      "learning_rate": 0.015922442244224424,
      "loss": 0.2369,
      "num_input_tokens_seen": 2035552,
      "step": 9650
    },
    {
      "epoch": 1.0621562156215623,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.015930693069306927,
      "loss": 0.2299,
      "num_input_tokens_seen": 2036576,
      "step": 9655
    },
    {
      "epoch": 1.0627062706270627,
      "grad_norm": 0.00927734375,
      "learning_rate": 0.015938943894389438,
      "loss": 0.2319,
      "num_input_tokens_seen": 2037664,
      "step": 9660
    },
    {
      "epoch": 1.0632563256325633,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01594719471947195,
      "loss": 0.2303,
      "num_input_tokens_seen": 2038656,
      "step": 9665
    },
    {
      "epoch": 1.0638063806380638,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.015955445544554456,
      "loss": 0.2298,
      "num_input_tokens_seen": 2039744,
      "step": 9670
    },
    {
      "epoch": 1.0643564356435644,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.015963696369636963,
      "loss": 0.2314,
      "num_input_tokens_seen": 2040736,
      "step": 9675
    },
    {
      "epoch": 1.0649064906490648,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.01597194719471947,
      "loss": 0.2298,
      "num_input_tokens_seen": 2041728,
      "step": 9680
    },
    {
      "epoch": 1.0654565456545655,
      "grad_norm": 0.00927734375,
      "learning_rate": 0.01598019801980198,
      "loss": 0.2324,
      "num_input_tokens_seen": 2042720,
      "step": 9685
    },
    {
      "epoch": 1.066006600660066,
      "grad_norm": 0.0030517578125,
      "learning_rate": 0.015988448844884488,
      "loss": 0.2314,
      "num_input_tokens_seen": 2043808,
      "step": 9690
    },
    {
      "epoch": 1.0665566556655666,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.015996699669966998,
      "loss": 0.2304,
      "num_input_tokens_seen": 2044832,
      "step": 9695
    },
    {
      "epoch": 1.0671067106710672,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.016004950495049502,
      "loss": 0.2314,
      "num_input_tokens_seen": 2045920,
      "step": 9700
    },
    {
      "epoch": 1.0676567656765676,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.016013201320132012,
      "loss": 0.2324,
      "num_input_tokens_seen": 2046944,
      "step": 9705
    },
    {
      "epoch": 1.0682068206820683,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.016021452145214523,
      "loss": 0.2325,
      "num_input_tokens_seen": 2048000,
      "step": 9710
    },
    {
      "epoch": 1.0687568756875687,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.01602970297029703,
      "loss": 0.2324,
      "num_input_tokens_seen": 2049056,
      "step": 9715
    },
    {
      "epoch": 1.0693069306930694,
      "grad_norm": 0.004486083984375,
      "learning_rate": 0.016037953795379537,
      "loss": 0.2329,
      "num_input_tokens_seen": 2050176,
      "step": 9720
    },
    {
      "epoch": 1.0698569856985698,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.016046204620462044,
      "loss": 0.2319,
      "num_input_tokens_seen": 2051296,
      "step": 9725
    },
    {
      "epoch": 1.0704070407040704,
      "grad_norm": 0.017578125,
      "learning_rate": 0.016054455445544555,
      "loss": 0.2308,
      "num_input_tokens_seen": 2052352,
      "step": 9730
    },
    {
      "epoch": 1.0709570957095709,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.016062706270627062,
      "loss": 0.2308,
      "num_input_tokens_seen": 2053408,
      "step": 9735
    },
    {
      "epoch": 1.0715071507150715,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.016070957095709573,
      "loss": 0.2324,
      "num_input_tokens_seen": 2054400,
      "step": 9740
    },
    {
      "epoch": 1.0720572057205722,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.016079207920792076,
      "loss": 0.2324,
      "num_input_tokens_seen": 2055488,
      "step": 9745
    },
    {
      "epoch": 1.0726072607260726,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.016087458745874587,
      "loss": 0.2303,
      "num_input_tokens_seen": 2056544,
      "step": 9750
    },
    {
      "epoch": 1.0731573157315732,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.016095709570957094,
      "loss": 0.2313,
      "num_input_tokens_seen": 2057600,
      "step": 9755
    },
    {
      "epoch": 1.0737073707370737,
      "grad_norm": 0.002960205078125,
      "learning_rate": 0.016103960396039604,
      "loss": 0.2329,
      "num_input_tokens_seen": 2058656,
      "step": 9760
    },
    {
      "epoch": 1.0742574257425743,
      "grad_norm": 0.00286865234375,
      "learning_rate": 0.01611221122112211,
      "loss": 0.2324,
      "num_input_tokens_seen": 2059680,
      "step": 9765
    },
    {
      "epoch": 1.0748074807480748,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.01612046204620462,
      "loss": 0.2329,
      "num_input_tokens_seen": 2060736,
      "step": 9770
    },
    {
      "epoch": 1.0753575357535754,
      "grad_norm": 0.0028839111328125,
      "learning_rate": 0.01612871287128713,
      "loss": 0.2303,
      "num_input_tokens_seen": 2061792,
      "step": 9775
    },
    {
      "epoch": 1.0759075907590758,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.016136963696369636,
      "loss": 0.2293,
      "num_input_tokens_seen": 2062784,
      "step": 9780
    },
    {
      "epoch": 1.0764576457645765,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.016145214521452147,
      "loss": 0.2319,
      "num_input_tokens_seen": 2063776,
      "step": 9785
    },
    {
      "epoch": 1.0770077007700771,
      "grad_norm": 0.0185546875,
      "learning_rate": 0.01615346534653465,
      "loss": 0.2309,
      "num_input_tokens_seen": 2064864,
      "step": 9790
    },
    {
      "epoch": 1.0775577557755776,
      "grad_norm": 0.00897216796875,
      "learning_rate": 0.01616171617161716,
      "loss": 0.232,
      "num_input_tokens_seen": 2065920,
      "step": 9795
    },
    {
      "epoch": 1.0781078107810782,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.016169966996699668,
      "loss": 0.2299,
      "num_input_tokens_seen": 2066976,
      "step": 9800
    },
    {
      "epoch": 1.0786578657865786,
      "grad_norm": 0.0035858154296875,
      "learning_rate": 0.01617821782178218,
      "loss": 0.2288,
      "num_input_tokens_seen": 2068096,
      "step": 9805
    },
    {
      "epoch": 1.0792079207920793,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.016186468646864686,
      "loss": 0.2341,
      "num_input_tokens_seen": 2069184,
      "step": 9810
    },
    {
      "epoch": 1.0797579757975797,
      "grad_norm": 0.003265380859375,
      "learning_rate": 0.016194719471947193,
      "loss": 0.2314,
      "num_input_tokens_seen": 2070240,
      "step": 9815
    },
    {
      "epoch": 1.0803080308030804,
      "grad_norm": 0.003326416015625,
      "learning_rate": 0.016202970297029703,
      "loss": 0.231,
      "num_input_tokens_seen": 2071328,
      "step": 9820
    },
    {
      "epoch": 1.0808580858085808,
      "grad_norm": 0.0042724609375,
      "learning_rate": 0.01621122112211221,
      "loss": 0.2304,
      "num_input_tokens_seen": 2072352,
      "step": 9825
    },
    {
      "epoch": 1.0814081408140814,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.01621947194719472,
      "loss": 0.2315,
      "num_input_tokens_seen": 2073376,
      "step": 9830
    },
    {
      "epoch": 1.0819581958195819,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.016227722772277225,
      "loss": 0.2309,
      "num_input_tokens_seen": 2074400,
      "step": 9835
    },
    {
      "epoch": 1.0825082508250825,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.016235973597359735,
      "loss": 0.2295,
      "num_input_tokens_seen": 2075584,
      "step": 9840
    },
    {
      "epoch": 1.0830583058305832,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.016244224422442242,
      "loss": 0.232,
      "num_input_tokens_seen": 2076672,
      "step": 9845
    },
    {
      "epoch": 1.0836083608360836,
      "grad_norm": 0.01806640625,
      "learning_rate": 0.016252475247524753,
      "loss": 0.2268,
      "num_input_tokens_seen": 2077696,
      "step": 9850
    },
    {
      "epoch": 1.0841584158415842,
      "grad_norm": 0.0038604736328125,
      "learning_rate": 0.01626072607260726,
      "loss": 0.2335,
      "num_input_tokens_seen": 2078784,
      "step": 9855
    },
    {
      "epoch": 1.0847084708470847,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.016268976897689767,
      "loss": 0.2299,
      "num_input_tokens_seen": 2079808,
      "step": 9860
    },
    {
      "epoch": 1.0852585258525853,
      "grad_norm": 0.0045166015625,
      "learning_rate": 0.016277227722772278,
      "loss": 0.2325,
      "num_input_tokens_seen": 2080896,
      "step": 9865
    },
    {
      "epoch": 1.0858085808580857,
      "grad_norm": 0.01806640625,
      "learning_rate": 0.016285478547854785,
      "loss": 0.2278,
      "num_input_tokens_seen": 2081952,
      "step": 9870
    },
    {
      "epoch": 1.0863586358635864,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.016293729372937295,
      "loss": 0.2273,
      "num_input_tokens_seen": 2083072,
      "step": 9875
    },
    {
      "epoch": 1.0869086908690868,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.0163019801980198,
      "loss": 0.2362,
      "num_input_tokens_seen": 2084096,
      "step": 9880
    },
    {
      "epoch": 1.0874587458745875,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.01631023102310231,
      "loss": 0.2289,
      "num_input_tokens_seen": 2085184,
      "step": 9885
    },
    {
      "epoch": 1.0880088008800881,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.016318481848184817,
      "loss": 0.2305,
      "num_input_tokens_seen": 2086336,
      "step": 9890
    },
    {
      "epoch": 1.0885588558855885,
      "grad_norm": 0.0040283203125,
      "learning_rate": 0.016326732673267327,
      "loss": 0.2269,
      "num_input_tokens_seen": 2087360,
      "step": 9895
    },
    {
      "epoch": 1.0891089108910892,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.016334983498349834,
      "loss": 0.2306,
      "num_input_tokens_seen": 2088384,
      "step": 9900
    },
    {
      "epoch": 1.0896589658965896,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.01634323432343234,
      "loss": 0.227,
      "num_input_tokens_seen": 2089408,
      "step": 9905
    },
    {
      "epoch": 1.0902090209020903,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.016351485148514852,
      "loss": 0.2338,
      "num_input_tokens_seen": 2090464,
      "step": 9910
    },
    {
      "epoch": 1.0907590759075907,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01635973597359736,
      "loss": 0.2248,
      "num_input_tokens_seen": 2091552,
      "step": 9915
    },
    {
      "epoch": 1.0913091309130913,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.01636798679867987,
      "loss": 0.2377,
      "num_input_tokens_seen": 2092576,
      "step": 9920
    },
    {
      "epoch": 1.0918591859185918,
      "grad_norm": 0.01239013671875,
      "learning_rate": 0.016376237623762373,
      "loss": 0.2334,
      "num_input_tokens_seen": 2093664,
      "step": 9925
    },
    {
      "epoch": 1.0924092409240924,
      "grad_norm": 0.0198974609375,
      "learning_rate": 0.016384488448844884,
      "loss": 0.2224,
      "num_input_tokens_seen": 2094720,
      "step": 9930
    },
    {
      "epoch": 1.0929592959295928,
      "grad_norm": 0.0033111572265625,
      "learning_rate": 0.01639273927392739,
      "loss": 0.2255,
      "num_input_tokens_seen": 2095776,
      "step": 9935
    },
    {
      "epoch": 1.0935093509350935,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0164009900990099,
      "loss": 0.2299,
      "num_input_tokens_seen": 2096896,
      "step": 9940
    },
    {
      "epoch": 1.0940594059405941,
      "grad_norm": 0.00445556640625,
      "learning_rate": 0.01640924092409241,
      "loss": 0.2304,
      "num_input_tokens_seen": 2097920,
      "step": 9945
    },
    {
      "epoch": 1.0946094609460946,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.016417491749174916,
      "loss": 0.2381,
      "num_input_tokens_seen": 2098976,
      "step": 9950
    },
    {
      "epoch": 1.0951595159515952,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.016425742574257426,
      "loss": 0.241,
      "num_input_tokens_seen": 2100032,
      "step": 9955
    },
    {
      "epoch": 1.0957095709570956,
      "grad_norm": 0.020751953125,
      "learning_rate": 0.016433993399339934,
      "loss": 0.2256,
      "num_input_tokens_seen": 2101024,
      "step": 9960
    },
    {
      "epoch": 1.0962596259625963,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.016442244224422444,
      "loss": 0.2333,
      "num_input_tokens_seen": 2102048,
      "step": 9965
    },
    {
      "epoch": 1.0968096809680967,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.016450495049504948,
      "loss": 0.2306,
      "num_input_tokens_seen": 2103136,
      "step": 9970
    },
    {
      "epoch": 1.0973597359735974,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.01645874587458746,
      "loss": 0.2317,
      "num_input_tokens_seen": 2104192,
      "step": 9975
    },
    {
      "epoch": 1.0979097909790978,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.016466996699669965,
      "loss": 0.2322,
      "num_input_tokens_seen": 2105280,
      "step": 9980
    },
    {
      "epoch": 1.0984598459845984,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.016475247524752476,
      "loss": 0.2321,
      "num_input_tokens_seen": 2106368,
      "step": 9985
    },
    {
      "epoch": 1.099009900990099,
      "grad_norm": 0.00897216796875,
      "learning_rate": 0.01648349834983498,
      "loss": 0.23,
      "num_input_tokens_seen": 2107360,
      "step": 9990
    },
    {
      "epoch": 1.0995599559955995,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.01649174917491749,
      "loss": 0.2326,
      "num_input_tokens_seen": 2108480,
      "step": 9995
    },
    {
      "epoch": 1.1001100110011002,
      "grad_norm": 0.0038909912109375,
      "learning_rate": 0.0165,
      "loss": 0.2341,
      "num_input_tokens_seen": 2109504,
      "step": 10000
    },
    {
      "epoch": 1.1006600660066006,
      "grad_norm": 0.002960205078125,
      "learning_rate": 0.016508250825082508,
      "loss": 0.232,
      "num_input_tokens_seen": 2110592,
      "step": 10005
    },
    {
      "epoch": 1.1012101210121013,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.01651650165016502,
      "loss": 0.2331,
      "num_input_tokens_seen": 2111648,
      "step": 10010
    },
    {
      "epoch": 1.1017601760176017,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.016524752475247522,
      "loss": 0.2309,
      "num_input_tokens_seen": 2112640,
      "step": 10015
    },
    {
      "epoch": 1.1023102310231023,
      "grad_norm": 0.004180908203125,
      "learning_rate": 0.016533003300330033,
      "loss": 0.2335,
      "num_input_tokens_seen": 2113696,
      "step": 10020
    },
    {
      "epoch": 1.1028602860286028,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.01654125412541254,
      "loss": 0.2299,
      "num_input_tokens_seen": 2114752,
      "step": 10025
    },
    {
      "epoch": 1.1034103410341034,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.01654950495049505,
      "loss": 0.2299,
      "num_input_tokens_seen": 2115808,
      "step": 10030
    },
    {
      "epoch": 1.103960396039604,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.016557755775577554,
      "loss": 0.2331,
      "num_input_tokens_seen": 2116864,
      "step": 10035
    },
    {
      "epoch": 1.1045104510451045,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.016566006600660065,
      "loss": 0.231,
      "num_input_tokens_seen": 2117920,
      "step": 10040
    },
    {
      "epoch": 1.1050605060506051,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.016574257425742575,
      "loss": 0.2321,
      "num_input_tokens_seen": 2118944,
      "step": 10045
    },
    {
      "epoch": 1.1056105610561056,
      "grad_norm": 0.003326416015625,
      "learning_rate": 0.016582508250825082,
      "loss": 0.2326,
      "num_input_tokens_seen": 2119936,
      "step": 10050
    },
    {
      "epoch": 1.1061606160616062,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.016590759075907593,
      "loss": 0.2278,
      "num_input_tokens_seen": 2121056,
      "step": 10055
    },
    {
      "epoch": 1.1067106710671066,
      "grad_norm": 0.0037384033203125,
      "learning_rate": 0.016599009900990096,
      "loss": 0.2306,
      "num_input_tokens_seen": 2122144,
      "step": 10060
    },
    {
      "epoch": 1.1072607260726073,
      "grad_norm": 0.0038909912109375,
      "learning_rate": 0.016607260726072607,
      "loss": 0.2285,
      "num_input_tokens_seen": 2123200,
      "step": 10065
    },
    {
      "epoch": 1.1078107810781077,
      "grad_norm": 0.0244140625,
      "learning_rate": 0.016615511551155114,
      "loss": 0.2273,
      "num_input_tokens_seen": 2124256,
      "step": 10070
    },
    {
      "epoch": 1.1083608360836084,
      "grad_norm": 0.016845703125,
      "learning_rate": 0.016623762376237625,
      "loss": 0.2344,
      "num_input_tokens_seen": 2125376,
      "step": 10075
    },
    {
      "epoch": 1.108910891089109,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.01663201320132013,
      "loss": 0.2365,
      "num_input_tokens_seen": 2126464,
      "step": 10080
    },
    {
      "epoch": 1.1094609460946094,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.01664026402640264,
      "loss": 0.2233,
      "num_input_tokens_seen": 2127520,
      "step": 10085
    },
    {
      "epoch": 1.11001100110011,
      "grad_norm": 0.01519775390625,
      "learning_rate": 0.01664851485148515,
      "loss": 0.2334,
      "num_input_tokens_seen": 2128576,
      "step": 10090
    },
    {
      "epoch": 1.1105610561056105,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.016656765676567657,
      "loss": 0.2337,
      "num_input_tokens_seen": 2129696,
      "step": 10095
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 0.003021240234375,
      "learning_rate": 0.016665016501650167,
      "loss": 0.2354,
      "num_input_tokens_seen": 2130752,
      "step": 10100
    },
    {
      "epoch": 1.1116611661166116,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.01667326732673267,
      "loss": 0.2359,
      "num_input_tokens_seen": 2131936,
      "step": 10105
    },
    {
      "epoch": 1.1122112211221122,
      "grad_norm": 0.01806640625,
      "learning_rate": 0.01668151815181518,
      "loss": 0.2314,
      "num_input_tokens_seen": 2133024,
      "step": 10110
    },
    {
      "epoch": 1.1127612761276127,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.01668976897689769,
      "loss": 0.234,
      "num_input_tokens_seen": 2134176,
      "step": 10115
    },
    {
      "epoch": 1.1133113311331133,
      "grad_norm": 0.0036468505859375,
      "learning_rate": 0.0166980198019802,
      "loss": 0.2324,
      "num_input_tokens_seen": 2135264,
      "step": 10120
    },
    {
      "epoch": 1.113861386138614,
      "grad_norm": 0.00286865234375,
      "learning_rate": 0.016706270627062703,
      "loss": 0.2318,
      "num_input_tokens_seen": 2136256,
      "step": 10125
    },
    {
      "epoch": 1.1144114411441144,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.016714521452145213,
      "loss": 0.2303,
      "num_input_tokens_seen": 2137312,
      "step": 10130
    },
    {
      "epoch": 1.114961496149615,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.016722772277227724,
      "loss": 0.2287,
      "num_input_tokens_seen": 2138464,
      "step": 10135
    },
    {
      "epoch": 1.1155115511551155,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.01673102310231023,
      "loss": 0.2324,
      "num_input_tokens_seen": 2139488,
      "step": 10140
    },
    {
      "epoch": 1.1160616061606161,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.01673927392739274,
      "loss": 0.2314,
      "num_input_tokens_seen": 2140480,
      "step": 10145
    },
    {
      "epoch": 1.1166116611661165,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.016747524752475245,
      "loss": 0.2335,
      "num_input_tokens_seen": 2141504,
      "step": 10150
    },
    {
      "epoch": 1.1171617161716172,
      "grad_norm": 0.004364013671875,
      "learning_rate": 0.016755775577557756,
      "loss": 0.2324,
      "num_input_tokens_seen": 2142560,
      "step": 10155
    },
    {
      "epoch": 1.1177117711771176,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.016764026402640263,
      "loss": 0.2293,
      "num_input_tokens_seen": 2143616,
      "step": 10160
    },
    {
      "epoch": 1.1182618261826183,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.016772277227722773,
      "loss": 0.2298,
      "num_input_tokens_seen": 2144640,
      "step": 10165
    },
    {
      "epoch": 1.118811881188119,
      "grad_norm": 0.0033416748046875,
      "learning_rate": 0.016780528052805277,
      "loss": 0.2308,
      "num_input_tokens_seen": 2145696,
      "step": 10170
    },
    {
      "epoch": 1.1193619361936193,
      "grad_norm": 0.003814697265625,
      "learning_rate": 0.016788778877887787,
      "loss": 0.2308,
      "num_input_tokens_seen": 2146784,
      "step": 10175
    },
    {
      "epoch": 1.11991199119912,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.016797029702970298,
      "loss": 0.2329,
      "num_input_tokens_seen": 2147744,
      "step": 10180
    },
    {
      "epoch": 1.1204620462046204,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.016805280528052805,
      "loss": 0.2298,
      "num_input_tokens_seen": 2148896,
      "step": 10185
    },
    {
      "epoch": 1.121012101210121,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.016813531353135316,
      "loss": 0.234,
      "num_input_tokens_seen": 2149952,
      "step": 10190
    },
    {
      "epoch": 1.1215621562156215,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.01682178217821782,
      "loss": 0.2329,
      "num_input_tokens_seen": 2150976,
      "step": 10195
    },
    {
      "epoch": 1.1221122112211221,
      "grad_norm": 0.0174560546875,
      "learning_rate": 0.01683003300330033,
      "loss": 0.2318,
      "num_input_tokens_seen": 2152000,
      "step": 10200
    },
    {
      "epoch": 1.1226622662266226,
      "grad_norm": 0.0029754638671875,
      "learning_rate": 0.016838283828382837,
      "loss": 0.2318,
      "num_input_tokens_seen": 2153056,
      "step": 10205
    },
    {
      "epoch": 1.1232123212321232,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.016846534653465348,
      "loss": 0.2318,
      "num_input_tokens_seen": 2154144,
      "step": 10210
    },
    {
      "epoch": 1.1237623762376239,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.01685478547854785,
      "loss": 0.2303,
      "num_input_tokens_seen": 2155168,
      "step": 10215
    },
    {
      "epoch": 1.1243124312431243,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.016863036303630362,
      "loss": 0.2319,
      "num_input_tokens_seen": 2156160,
      "step": 10220
    },
    {
      "epoch": 1.124862486248625,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.016871287128712872,
      "loss": 0.2308,
      "num_input_tokens_seen": 2157152,
      "step": 10225
    },
    {
      "epoch": 1.1254125412541254,
      "grad_norm": 0.01806640625,
      "learning_rate": 0.01687953795379538,
      "loss": 0.2356,
      "num_input_tokens_seen": 2158208,
      "step": 10230
    },
    {
      "epoch": 1.125962596259626,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01688778877887789,
      "loss": 0.2313,
      "num_input_tokens_seen": 2159232,
      "step": 10235
    },
    {
      "epoch": 1.1265126512651265,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.016896039603960394,
      "loss": 0.2308,
      "num_input_tokens_seen": 2160320,
      "step": 10240
    },
    {
      "epoch": 1.127062706270627,
      "grad_norm": 0.009765625,
      "learning_rate": 0.016904290429042904,
      "loss": 0.2303,
      "num_input_tokens_seen": 2161472,
      "step": 10245
    },
    {
      "epoch": 1.1276127612761275,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.01691254125412541,
      "loss": 0.2313,
      "num_input_tokens_seen": 2162496,
      "step": 10250
    },
    {
      "epoch": 1.1281628162816282,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.016920792079207922,
      "loss": 0.2335,
      "num_input_tokens_seen": 2163488,
      "step": 10255
    },
    {
      "epoch": 1.1287128712871288,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.016929042904290426,
      "loss": 0.2319,
      "num_input_tokens_seen": 2164544,
      "step": 10260
    },
    {
      "epoch": 1.1292629262926293,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.016937293729372936,
      "loss": 0.2324,
      "num_input_tokens_seen": 2165600,
      "step": 10265
    },
    {
      "epoch": 1.12981298129813,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.016945544554455443,
      "loss": 0.2319,
      "num_input_tokens_seen": 2166624,
      "step": 10270
    },
    {
      "epoch": 1.1303630363036303,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.016953795379537954,
      "loss": 0.2334,
      "num_input_tokens_seen": 2167648,
      "step": 10275
    },
    {
      "epoch": 1.130913091309131,
      "grad_norm": 0.00897216796875,
      "learning_rate": 0.016962046204620464,
      "loss": 0.2329,
      "num_input_tokens_seen": 2168704,
      "step": 10280
    },
    {
      "epoch": 1.1314631463146314,
      "grad_norm": 0.003082275390625,
      "learning_rate": 0.016970297029702968,
      "loss": 0.2329,
      "num_input_tokens_seen": 2169760,
      "step": 10285
    },
    {
      "epoch": 1.132013201320132,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.01697854785478548,
      "loss": 0.2308,
      "num_input_tokens_seen": 2170816,
      "step": 10290
    },
    {
      "epoch": 1.1325632563256325,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.016986798679867986,
      "loss": 0.2318,
      "num_input_tokens_seen": 2171872,
      "step": 10295
    },
    {
      "epoch": 1.1331133113311331,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.016995049504950496,
      "loss": 0.2308,
      "num_input_tokens_seen": 2172896,
      "step": 10300
    },
    {
      "epoch": 1.1336633663366338,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.017003300330033,
      "loss": 0.2304,
      "num_input_tokens_seen": 2173984,
      "step": 10305
    },
    {
      "epoch": 1.1342134213421342,
      "grad_norm": 0.0174560546875,
      "learning_rate": 0.01701155115511551,
      "loss": 0.2283,
      "num_input_tokens_seen": 2175072,
      "step": 10310
    },
    {
      "epoch": 1.1347634763476349,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.017019801980198018,
      "loss": 0.2326,
      "num_input_tokens_seen": 2176128,
      "step": 10315
    },
    {
      "epoch": 1.1353135313531353,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.017028052805280528,
      "loss": 0.2357,
      "num_input_tokens_seen": 2177184,
      "step": 10320
    },
    {
      "epoch": 1.135863586358636,
      "grad_norm": 0.01904296875,
      "learning_rate": 0.01703630363036304,
      "loss": 0.2325,
      "num_input_tokens_seen": 2178336,
      "step": 10325
    },
    {
      "epoch": 1.1364136413641364,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.017044554455445542,
      "loss": 0.2284,
      "num_input_tokens_seen": 2179360,
      "step": 10330
    },
    {
      "epoch": 1.136963696369637,
      "grad_norm": 0.017333984375,
      "learning_rate": 0.017052805280528053,
      "loss": 0.2305,
      "num_input_tokens_seen": 2180448,
      "step": 10335
    },
    {
      "epoch": 1.1375137513751374,
      "grad_norm": 0.0034027099609375,
      "learning_rate": 0.01706105610561056,
      "loss": 0.23,
      "num_input_tokens_seen": 2181440,
      "step": 10340
    },
    {
      "epoch": 1.138063806380638,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.01706930693069307,
      "loss": 0.2296,
      "num_input_tokens_seen": 2182528,
      "step": 10345
    },
    {
      "epoch": 1.1386138613861387,
      "grad_norm": 0.00897216796875,
      "learning_rate": 0.017077557755775574,
      "loss": 0.2275,
      "num_input_tokens_seen": 2183552,
      "step": 10350
    },
    {
      "epoch": 1.1391639163916392,
      "grad_norm": 0.002838134765625,
      "learning_rate": 0.017085808580858085,
      "loss": 0.2281,
      "num_input_tokens_seen": 2184640,
      "step": 10355
    },
    {
      "epoch": 1.1397139713971396,
      "grad_norm": 0.003631591796875,
      "learning_rate": 0.017094059405940592,
      "loss": 0.2335,
      "num_input_tokens_seen": 2185760,
      "step": 10360
    },
    {
      "epoch": 1.1402640264026402,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.017102310231023102,
      "loss": 0.2299,
      "num_input_tokens_seen": 2186848,
      "step": 10365
    },
    {
      "epoch": 1.140814081408141,
      "grad_norm": 0.003997802734375,
      "learning_rate": 0.017110561056105613,
      "loss": 0.232,
      "num_input_tokens_seen": 2187904,
      "step": 10370
    },
    {
      "epoch": 1.1413641364136413,
      "grad_norm": 0.003204345703125,
      "learning_rate": 0.017118811881188117,
      "loss": 0.2346,
      "num_input_tokens_seen": 2188960,
      "step": 10375
    },
    {
      "epoch": 1.141914191419142,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.017127062706270627,
      "loss": 0.2335,
      "num_input_tokens_seen": 2190048,
      "step": 10380
    },
    {
      "epoch": 1.1424642464246424,
      "grad_norm": 0.0036773681640625,
      "learning_rate": 0.017135313531353134,
      "loss": 0.2329,
      "num_input_tokens_seen": 2191136,
      "step": 10385
    },
    {
      "epoch": 1.143014301430143,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.017143564356435645,
      "loss": 0.2359,
      "num_input_tokens_seen": 2192160,
      "step": 10390
    },
    {
      "epoch": 1.1435643564356435,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.01715181518151815,
      "loss": 0.2316,
      "num_input_tokens_seen": 2193216,
      "step": 10395
    },
    {
      "epoch": 1.1441144114411441,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01716006600660066,
      "loss": 0.2326,
      "num_input_tokens_seen": 2194304,
      "step": 10400
    },
    {
      "epoch": 1.1446644664466445,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.017168316831683166,
      "loss": 0.2294,
      "num_input_tokens_seen": 2195392,
      "step": 10405
    },
    {
      "epoch": 1.1452145214521452,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.017176567656765677,
      "loss": 0.2305,
      "num_input_tokens_seen": 2196512,
      "step": 10410
    },
    {
      "epoch": 1.1457645764576458,
      "grad_norm": 0.00897216796875,
      "learning_rate": 0.017184818481848187,
      "loss": 0.2325,
      "num_input_tokens_seen": 2197632,
      "step": 10415
    },
    {
      "epoch": 1.1463146314631463,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.01719306930693069,
      "loss": 0.2346,
      "num_input_tokens_seen": 2198784,
      "step": 10420
    },
    {
      "epoch": 1.146864686468647,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.0172013201320132,
      "loss": 0.2304,
      "num_input_tokens_seen": 2199744,
      "step": 10425
    },
    {
      "epoch": 1.1474147414741473,
      "grad_norm": 0.009765625,
      "learning_rate": 0.01720957095709571,
      "loss": 0.233,
      "num_input_tokens_seen": 2200736,
      "step": 10430
    },
    {
      "epoch": 1.147964796479648,
      "grad_norm": 0.0208740234375,
      "learning_rate": 0.01721782178217822,
      "loss": 0.2319,
      "num_input_tokens_seen": 2201728,
      "step": 10435
    },
    {
      "epoch": 1.1485148514851484,
      "grad_norm": 0.0029754638671875,
      "learning_rate": 0.017226072607260723,
      "loss": 0.232,
      "num_input_tokens_seen": 2202752,
      "step": 10440
    },
    {
      "epoch": 1.149064906490649,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.017234323432343233,
      "loss": 0.2358,
      "num_input_tokens_seen": 2203872,
      "step": 10445
    },
    {
      "epoch": 1.1496149614961495,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.01724257425742574,
      "loss": 0.2322,
      "num_input_tokens_seen": 2204896,
      "step": 10450
    },
    {
      "epoch": 1.1501650165016502,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.01725082508250825,
      "loss": 0.2331,
      "num_input_tokens_seen": 2205888,
      "step": 10455
    },
    {
      "epoch": 1.1507150715071508,
      "grad_norm": 0.0194091796875,
      "learning_rate": 0.01725907590759076,
      "loss": 0.2273,
      "num_input_tokens_seen": 2206912,
      "step": 10460
    },
    {
      "epoch": 1.1512651265126512,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.017267326732673265,
      "loss": 0.23,
      "num_input_tokens_seen": 2207968,
      "step": 10465
    },
    {
      "epoch": 1.1518151815181519,
      "grad_norm": 0.0038909912109375,
      "learning_rate": 0.017275577557755776,
      "loss": 0.2291,
      "num_input_tokens_seen": 2208992,
      "step": 10470
    },
    {
      "epoch": 1.1523652365236523,
      "grad_norm": 0.003662109375,
      "learning_rate": 0.017283828382838283,
      "loss": 0.2312,
      "num_input_tokens_seen": 2210048,
      "step": 10475
    },
    {
      "epoch": 1.152915291529153,
      "grad_norm": 0.0038909912109375,
      "learning_rate": 0.017292079207920794,
      "loss": 0.2345,
      "num_input_tokens_seen": 2211136,
      "step": 10480
    },
    {
      "epoch": 1.1534653465346534,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.017300330033003297,
      "loss": 0.2257,
      "num_input_tokens_seen": 2212160,
      "step": 10485
    },
    {
      "epoch": 1.154015401540154,
      "grad_norm": 0.0260009765625,
      "learning_rate": 0.017308580858085808,
      "loss": 0.2387,
      "num_input_tokens_seen": 2213184,
      "step": 10490
    },
    {
      "epoch": 1.1545654565456545,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.017316831683168315,
      "loss": 0.2282,
      "num_input_tokens_seen": 2214208,
      "step": 10495
    },
    {
      "epoch": 1.155115511551155,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.017325082508250825,
      "loss": 0.2287,
      "num_input_tokens_seen": 2215200,
      "step": 10500
    },
    {
      "epoch": 1.1556655665566558,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.017333333333333333,
      "loss": 0.2257,
      "num_input_tokens_seen": 2216320,
      "step": 10505
    },
    {
      "epoch": 1.1562156215621562,
      "grad_norm": 0.0196533203125,
      "learning_rate": 0.01734158415841584,
      "loss": 0.2328,
      "num_input_tokens_seen": 2217376,
      "step": 10510
    },
    {
      "epoch": 1.1567656765676568,
      "grad_norm": 0.024169921875,
      "learning_rate": 0.01734983498349835,
      "loss": 0.227,
      "num_input_tokens_seen": 2218400,
      "step": 10515
    },
    {
      "epoch": 1.1573157315731573,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.017358085808580857,
      "loss": 0.2261,
      "num_input_tokens_seen": 2219488,
      "step": 10520
    },
    {
      "epoch": 1.157865786578658,
      "grad_norm": 0.04248046875,
      "learning_rate": 0.017366336633663368,
      "loss": 0.2445,
      "num_input_tokens_seen": 2220480,
      "step": 10525
    },
    {
      "epoch": 1.1584158415841583,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.01737458745874587,
      "loss": 0.2362,
      "num_input_tokens_seen": 2221536,
      "step": 10530
    },
    {
      "epoch": 1.158965896589659,
      "grad_norm": 0.00286865234375,
      "learning_rate": 0.017382838283828382,
      "loss": 0.2287,
      "num_input_tokens_seen": 2222528,
      "step": 10535
    },
    {
      "epoch": 1.1595159515951594,
      "grad_norm": 0.0033721923828125,
      "learning_rate": 0.01739108910891089,
      "loss": 0.227,
      "num_input_tokens_seen": 2223648,
      "step": 10540
    },
    {
      "epoch": 1.16006600660066,
      "grad_norm": 0.003570556640625,
      "learning_rate": 0.0173993399339934,
      "loss": 0.2337,
      "num_input_tokens_seen": 2224672,
      "step": 10545
    },
    {
      "epoch": 1.1606160616061607,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.017407590759075907,
      "loss": 0.2332,
      "num_input_tokens_seen": 2225728,
      "step": 10550
    },
    {
      "epoch": 1.1611661166116611,
      "grad_norm": 0.009765625,
      "learning_rate": 0.017415841584158414,
      "loss": 0.2295,
      "num_input_tokens_seen": 2226784,
      "step": 10555
    },
    {
      "epoch": 1.1617161716171618,
      "grad_norm": 0.0213623046875,
      "learning_rate": 0.017424092409240925,
      "loss": 0.2306,
      "num_input_tokens_seen": 2227872,
      "step": 10560
    },
    {
      "epoch": 1.1622662266226622,
      "grad_norm": 0.0036773681640625,
      "learning_rate": 0.01743234323432343,
      "loss": 0.2286,
      "num_input_tokens_seen": 2228864,
      "step": 10565
    },
    {
      "epoch": 1.1628162816281629,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.017440594059405942,
      "loss": 0.2224,
      "num_input_tokens_seen": 2229920,
      "step": 10570
    },
    {
      "epoch": 1.1633663366336633,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.017448844884488446,
      "loss": 0.2401,
      "num_input_tokens_seen": 2231008,
      "step": 10575
    },
    {
      "epoch": 1.163916391639164,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.017457095709570956,
      "loss": 0.2407,
      "num_input_tokens_seen": 2232064,
      "step": 10580
    },
    {
      "epoch": 1.1644664466446644,
      "grad_norm": 0.017822265625,
      "learning_rate": 0.017465346534653464,
      "loss": 0.2308,
      "num_input_tokens_seen": 2233152,
      "step": 10585
    },
    {
      "epoch": 1.165016501650165,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.017473597359735974,
      "loss": 0.2358,
      "num_input_tokens_seen": 2234176,
      "step": 10590
    },
    {
      "epoch": 1.1655665566556657,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.01748184818481848,
      "loss": 0.2301,
      "num_input_tokens_seen": 2235264,
      "step": 10595
    },
    {
      "epoch": 1.166116611661166,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.01749009900990099,
      "loss": 0.2284,
      "num_input_tokens_seen": 2236352,
      "step": 10600
    },
    {
      "epoch": 1.1666666666666667,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.0174983498349835,
      "loss": 0.2301,
      "num_input_tokens_seen": 2237472,
      "step": 10605
    },
    {
      "epoch": 1.1672167216721672,
      "grad_norm": 0.017333984375,
      "learning_rate": 0.017506600660066006,
      "loss": 0.2274,
      "num_input_tokens_seen": 2238528,
      "step": 10610
    },
    {
      "epoch": 1.1677667766776678,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.017514851485148517,
      "loss": 0.2312,
      "num_input_tokens_seen": 2239616,
      "step": 10615
    },
    {
      "epoch": 1.1683168316831682,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.01752310231023102,
      "loss": 0.2385,
      "num_input_tokens_seen": 2240640,
      "step": 10620
    },
    {
      "epoch": 1.168866886688669,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.01753135313531353,
      "loss": 0.2291,
      "num_input_tokens_seen": 2241664,
      "step": 10625
    },
    {
      "epoch": 1.1694169416941693,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.017539603960396038,
      "loss": 0.2364,
      "num_input_tokens_seen": 2242720,
      "step": 10630
    },
    {
      "epoch": 1.16996699669967,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01754785478547855,
      "loss": 0.2369,
      "num_input_tokens_seen": 2243808,
      "step": 10635
    },
    {
      "epoch": 1.1705170517051706,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.017556105610561056,
      "loss": 0.2346,
      "num_input_tokens_seen": 2244864,
      "step": 10640
    },
    {
      "epoch": 1.171067106710671,
      "grad_norm": 0.0042724609375,
      "learning_rate": 0.017564356435643563,
      "loss": 0.2314,
      "num_input_tokens_seen": 2245920,
      "step": 10645
    },
    {
      "epoch": 1.1716171617161717,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.017572607260726073,
      "loss": 0.2304,
      "num_input_tokens_seen": 2246976,
      "step": 10650
    },
    {
      "epoch": 1.1721672167216721,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01758085808580858,
      "loss": 0.2319,
      "num_input_tokens_seen": 2248064,
      "step": 10655
    },
    {
      "epoch": 1.1727172717271728,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.01758910891089109,
      "loss": 0.2335,
      "num_input_tokens_seen": 2249152,
      "step": 10660
    },
    {
      "epoch": 1.1732673267326732,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.017597359735973594,
      "loss": 0.233,
      "num_input_tokens_seen": 2250240,
      "step": 10665
    },
    {
      "epoch": 1.1738173817381738,
      "grad_norm": 0.003021240234375,
      "learning_rate": 0.017605610561056105,
      "loss": 0.2304,
      "num_input_tokens_seen": 2251328,
      "step": 10670
    },
    {
      "epoch": 1.1743674367436743,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.017613861386138612,
      "loss": 0.2309,
      "num_input_tokens_seen": 2252352,
      "step": 10675
    },
    {
      "epoch": 1.174917491749175,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.017622112211221123,
      "loss": 0.2324,
      "num_input_tokens_seen": 2253408,
      "step": 10680
    },
    {
      "epoch": 1.1754675467546756,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.01763036303630363,
      "loss": 0.2314,
      "num_input_tokens_seen": 2254464,
      "step": 10685
    },
    {
      "epoch": 1.176017601760176,
      "grad_norm": 0.00299072265625,
      "learning_rate": 0.017638613861386137,
      "loss": 0.2314,
      "num_input_tokens_seen": 2255552,
      "step": 10690
    },
    {
      "epoch": 1.1765676567656767,
      "grad_norm": 0.00897216796875,
      "learning_rate": 0.017646864686468648,
      "loss": 0.2303,
      "num_input_tokens_seen": 2256544,
      "step": 10695
    },
    {
      "epoch": 1.177117711771177,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.017655115511551155,
      "loss": 0.2293,
      "num_input_tokens_seen": 2257632,
      "step": 10700
    },
    {
      "epoch": 1.1776677667766777,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.017663366336633665,
      "loss": 0.2314,
      "num_input_tokens_seen": 2258720,
      "step": 10705
    },
    {
      "epoch": 1.1782178217821782,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01767161716171617,
      "loss": 0.2314,
      "num_input_tokens_seen": 2259776,
      "step": 10710
    },
    {
      "epoch": 1.1787678767876788,
      "grad_norm": 0.0035247802734375,
      "learning_rate": 0.01767986798679868,
      "loss": 0.2319,
      "num_input_tokens_seen": 2260896,
      "step": 10715
    },
    {
      "epoch": 1.1793179317931792,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.017688118811881186,
      "loss": 0.2273,
      "num_input_tokens_seen": 2261984,
      "step": 10720
    },
    {
      "epoch": 1.1798679867986799,
      "grad_norm": 0.0032958984375,
      "learning_rate": 0.017696369636963697,
      "loss": 0.232,
      "num_input_tokens_seen": 2263072,
      "step": 10725
    },
    {
      "epoch": 1.1804180418041805,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.017704620462046204,
      "loss": 0.2325,
      "num_input_tokens_seen": 2264096,
      "step": 10730
    },
    {
      "epoch": 1.180968096809681,
      "grad_norm": 0.0027313232421875,
      "learning_rate": 0.01771287128712871,
      "loss": 0.2299,
      "num_input_tokens_seen": 2265152,
      "step": 10735
    },
    {
      "epoch": 1.1815181518151816,
      "grad_norm": 0.0035247802734375,
      "learning_rate": 0.01772112211221122,
      "loss": 0.2263,
      "num_input_tokens_seen": 2266176,
      "step": 10740
    },
    {
      "epoch": 1.182068206820682,
      "grad_norm": 0.0322265625,
      "learning_rate": 0.01772937293729373,
      "loss": 0.2339,
      "num_input_tokens_seen": 2267200,
      "step": 10745
    },
    {
      "epoch": 1.1826182618261827,
      "grad_norm": 0.0029449462890625,
      "learning_rate": 0.01773762376237624,
      "loss": 0.2376,
      "num_input_tokens_seen": 2268256,
      "step": 10750
    },
    {
      "epoch": 1.183168316831683,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.017745874587458743,
      "loss": 0.231,
      "num_input_tokens_seen": 2269312,
      "step": 10755
    },
    {
      "epoch": 1.1837183718371838,
      "grad_norm": 0.01708984375,
      "learning_rate": 0.017754125412541254,
      "loss": 0.2274,
      "num_input_tokens_seen": 2270400,
      "step": 10760
    },
    {
      "epoch": 1.1842684268426842,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.01776237623762376,
      "loss": 0.2306,
      "num_input_tokens_seen": 2271392,
      "step": 10765
    },
    {
      "epoch": 1.1848184818481848,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.01777062706270627,
      "loss": 0.2301,
      "num_input_tokens_seen": 2272448,
      "step": 10770
    },
    {
      "epoch": 1.1853685368536855,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.01777887788778878,
      "loss": 0.2384,
      "num_input_tokens_seen": 2273504,
      "step": 10775
    },
    {
      "epoch": 1.185918591859186,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.017787128712871286,
      "loss": 0.2275,
      "num_input_tokens_seen": 2274592,
      "step": 10780
    },
    {
      "epoch": 1.1864686468646866,
      "grad_norm": 0.003265380859375,
      "learning_rate": 0.017795379537953793,
      "loss": 0.2362,
      "num_input_tokens_seen": 2275648,
      "step": 10785
    },
    {
      "epoch": 1.187018701870187,
      "grad_norm": 0.018798828125,
      "learning_rate": 0.017803630363036303,
      "loss": 0.2331,
      "num_input_tokens_seen": 2276736,
      "step": 10790
    },
    {
      "epoch": 1.1875687568756876,
      "grad_norm": 0.00390625,
      "learning_rate": 0.017811881188118814,
      "loss": 0.2335,
      "num_input_tokens_seen": 2277856,
      "step": 10795
    },
    {
      "epoch": 1.188118811881188,
      "grad_norm": 0.00927734375,
      "learning_rate": 0.017820132013201317,
      "loss": 0.2319,
      "num_input_tokens_seen": 2278944,
      "step": 10800
    },
    {
      "epoch": 1.1886688668866887,
      "grad_norm": 0.004119873046875,
      "learning_rate": 0.017828382838283828,
      "loss": 0.2309,
      "num_input_tokens_seen": 2279936,
      "step": 10805
    },
    {
      "epoch": 1.1892189218921891,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.017836633663366335,
      "loss": 0.2299,
      "num_input_tokens_seen": 2280960,
      "step": 10810
    },
    {
      "epoch": 1.1897689768976898,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.017844884488448846,
      "loss": 0.231,
      "num_input_tokens_seen": 2282048,
      "step": 10815
    },
    {
      "epoch": 1.1903190319031904,
      "grad_norm": 0.0028076171875,
      "learning_rate": 0.017853135313531353,
      "loss": 0.2332,
      "num_input_tokens_seen": 2283008,
      "step": 10820
    },
    {
      "epoch": 1.1908690869086909,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.01786138613861386,
      "loss": 0.2285,
      "num_input_tokens_seen": 2284128,
      "step": 10825
    },
    {
      "epoch": 1.1914191419141915,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.017869636963696367,
      "loss": 0.2301,
      "num_input_tokens_seen": 2285152,
      "step": 10830
    },
    {
      "epoch": 1.191969196919692,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.017877887788778878,
      "loss": 0.2276,
      "num_input_tokens_seen": 2286208,
      "step": 10835
    },
    {
      "epoch": 1.1925192519251926,
      "grad_norm": 0.026611328125,
      "learning_rate": 0.017886138613861388,
      "loss": 0.2406,
      "num_input_tokens_seen": 2287296,
      "step": 10840
    },
    {
      "epoch": 1.193069306930693,
      "grad_norm": 0.002960205078125,
      "learning_rate": 0.017894389438943892,
      "loss": 0.2347,
      "num_input_tokens_seen": 2288288,
      "step": 10845
    },
    {
      "epoch": 1.1936193619361937,
      "grad_norm": 0.0036468505859375,
      "learning_rate": 0.017902640264026402,
      "loss": 0.2355,
      "num_input_tokens_seen": 2289248,
      "step": 10850
    },
    {
      "epoch": 1.194169416941694,
      "grad_norm": 0.0029449462890625,
      "learning_rate": 0.01791089108910891,
      "loss": 0.2313,
      "num_input_tokens_seen": 2290240,
      "step": 10855
    },
    {
      "epoch": 1.1947194719471947,
      "grad_norm": 0.0040283203125,
      "learning_rate": 0.01791914191419142,
      "loss": 0.2335,
      "num_input_tokens_seen": 2291360,
      "step": 10860
    },
    {
      "epoch": 1.1952695269526954,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.017927392739273927,
      "loss": 0.2329,
      "num_input_tokens_seen": 2292480,
      "step": 10865
    },
    {
      "epoch": 1.1958195819581958,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.017935643564356434,
      "loss": 0.2319,
      "num_input_tokens_seen": 2293504,
      "step": 10870
    },
    {
      "epoch": 1.1963696369636962,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.01794389438943894,
      "loss": 0.235,
      "num_input_tokens_seen": 2294528,
      "step": 10875
    },
    {
      "epoch": 1.196919691969197,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.017952145214521452,
      "loss": 0.2314,
      "num_input_tokens_seen": 2295616,
      "step": 10880
    },
    {
      "epoch": 1.1974697469746975,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.017960396039603962,
      "loss": 0.2314,
      "num_input_tokens_seen": 2296736,
      "step": 10885
    },
    {
      "epoch": 1.198019801980198,
      "grad_norm": 0.003143310546875,
      "learning_rate": 0.017968646864686466,
      "loss": 0.2298,
      "num_input_tokens_seen": 2297824,
      "step": 10890
    },
    {
      "epoch": 1.1985698569856986,
      "grad_norm": 0.016357421875,
      "learning_rate": 0.017976897689768977,
      "loss": 0.2319,
      "num_input_tokens_seen": 2298880,
      "step": 10895
    },
    {
      "epoch": 1.199119911991199,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.017985148514851484,
      "loss": 0.2319,
      "num_input_tokens_seen": 2299968,
      "step": 10900
    },
    {
      "epoch": 1.1996699669966997,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.017993399339933994,
      "loss": 0.2314,
      "num_input_tokens_seen": 2301024,
      "step": 10905
    },
    {
      "epoch": 1.2002200220022001,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.0180016501650165,
      "loss": 0.2288,
      "num_input_tokens_seen": 2302048,
      "step": 10910
    },
    {
      "epoch": 1.2007700770077008,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.01800990099009901,
      "loss": 0.2335,
      "num_input_tokens_seen": 2303072,
      "step": 10915
    },
    {
      "epoch": 1.2013201320132012,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.018018151815181516,
      "loss": 0.233,
      "num_input_tokens_seen": 2304160,
      "step": 10920
    },
    {
      "epoch": 1.2018701870187019,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.018026402640264026,
      "loss": 0.2346,
      "num_input_tokens_seen": 2305184,
      "step": 10925
    },
    {
      "epoch": 1.2024202420242025,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.018034653465346537,
      "loss": 0.2314,
      "num_input_tokens_seen": 2306176,
      "step": 10930
    },
    {
      "epoch": 1.202970297029703,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.01804290429042904,
      "loss": 0.2329,
      "num_input_tokens_seen": 2307232,
      "step": 10935
    },
    {
      "epoch": 1.2035203520352036,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.01805115511551155,
      "loss": 0.2319,
      "num_input_tokens_seen": 2308320,
      "step": 10940
    },
    {
      "epoch": 1.204070407040704,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.018059405940594058,
      "loss": 0.2313,
      "num_input_tokens_seen": 2309344,
      "step": 10945
    },
    {
      "epoch": 1.2046204620462047,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.01806765676567657,
      "loss": 0.2318,
      "num_input_tokens_seen": 2310368,
      "step": 10950
    },
    {
      "epoch": 1.205170517051705,
      "grad_norm": 0.0081787109375,
      "learning_rate": 0.018075907590759076,
      "loss": 0.2314,
      "num_input_tokens_seen": 2311360,
      "step": 10955
    },
    {
      "epoch": 1.2057205720572057,
      "grad_norm": 0.0084228515625,
      "learning_rate": 0.018084158415841583,
      "loss": 0.2334,
      "num_input_tokens_seen": 2312416,
      "step": 10960
    },
    {
      "epoch": 1.2062706270627062,
      "grad_norm": 0.00775146484375,
      "learning_rate": 0.01809240924092409,
      "loss": 0.2308,
      "num_input_tokens_seen": 2313408,
      "step": 10965
    },
    {
      "epoch": 1.2068206820682068,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.0181006600660066,
      "loss": 0.2313,
      "num_input_tokens_seen": 2314400,
      "step": 10970
    },
    {
      "epoch": 1.2073707370737075,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.01810891089108911,
      "loss": 0.2308,
      "num_input_tokens_seen": 2315424,
      "step": 10975
    },
    {
      "epoch": 1.2079207920792079,
      "grad_norm": 0.0159912109375,
      "learning_rate": 0.018117161716171615,
      "loss": 0.2308,
      "num_input_tokens_seen": 2316576,
      "step": 10980
    },
    {
      "epoch": 1.2084708470847085,
      "grad_norm": 0.0032958984375,
      "learning_rate": 0.018125412541254125,
      "loss": 0.2319,
      "num_input_tokens_seen": 2317664,
      "step": 10985
    },
    {
      "epoch": 1.209020902090209,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.018133663366336632,
      "loss": 0.2318,
      "num_input_tokens_seen": 2318688,
      "step": 10990
    },
    {
      "epoch": 1.2095709570957096,
      "grad_norm": 0.0038604736328125,
      "learning_rate": 0.018141914191419143,
      "loss": 0.2308,
      "num_input_tokens_seen": 2319776,
      "step": 10995
    },
    {
      "epoch": 1.21012101210121,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.01815016501650165,
      "loss": 0.2298,
      "num_input_tokens_seen": 2320896,
      "step": 11000
    },
    {
      "epoch": 1.2106710671067107,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.018158415841584157,
      "loss": 0.2294,
      "num_input_tokens_seen": 2322016,
      "step": 11005
    },
    {
      "epoch": 1.2112211221122111,
      "grad_norm": 0.0030975341796875,
      "learning_rate": 0.018166666666666664,
      "loss": 0.2299,
      "num_input_tokens_seen": 2323072,
      "step": 11010
    },
    {
      "epoch": 1.2117711771177118,
      "grad_norm": 0.01544189453125,
      "learning_rate": 0.018174917491749175,
      "loss": 0.2288,
      "num_input_tokens_seen": 2324096,
      "step": 11015
    },
    {
      "epoch": 1.2123212321232124,
      "grad_norm": 0.007720947265625,
      "learning_rate": 0.018183168316831682,
      "loss": 0.2315,
      "num_input_tokens_seen": 2325120,
      "step": 11020
    },
    {
      "epoch": 1.2128712871287128,
      "grad_norm": 0.004180908203125,
      "learning_rate": 0.01819141914191419,
      "loss": 0.2341,
      "num_input_tokens_seen": 2326144,
      "step": 11025
    },
    {
      "epoch": 1.2134213421342135,
      "grad_norm": 0.0032501220703125,
      "learning_rate": 0.0181996699669967,
      "loss": 0.2335,
      "num_input_tokens_seen": 2327296,
      "step": 11030
    },
    {
      "epoch": 1.213971397139714,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.018207920792079207,
      "loss": 0.2299,
      "num_input_tokens_seen": 2328384,
      "step": 11035
    },
    {
      "epoch": 1.2145214521452146,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.018216171617161717,
      "loss": 0.2294,
      "num_input_tokens_seen": 2329408,
      "step": 11040
    },
    {
      "epoch": 1.215071507150715,
      "grad_norm": 0.00927734375,
      "learning_rate": 0.018224422442244224,
      "loss": 0.2325,
      "num_input_tokens_seen": 2330464,
      "step": 11045
    },
    {
      "epoch": 1.2156215621562156,
      "grad_norm": 0.0076904296875,
      "learning_rate": 0.01823267326732673,
      "loss": 0.2315,
      "num_input_tokens_seen": 2331488,
      "step": 11050
    },
    {
      "epoch": 1.216171617161716,
      "grad_norm": 0.003021240234375,
      "learning_rate": 0.01824092409240924,
      "loss": 0.2309,
      "num_input_tokens_seen": 2332608,
      "step": 11055
    },
    {
      "epoch": 1.2167216721672167,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.01824917491749175,
      "loss": 0.2361,
      "num_input_tokens_seen": 2333760,
      "step": 11060
    },
    {
      "epoch": 1.2172717271727174,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.018257425742574256,
      "loss": 0.2351,
      "num_input_tokens_seen": 2334784,
      "step": 11065
    },
    {
      "epoch": 1.2178217821782178,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.018265676567656763,
      "loss": 0.2309,
      "num_input_tokens_seen": 2335808,
      "step": 11070
    },
    {
      "epoch": 1.2183718371837184,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.018273927392739274,
      "loss": 0.2303,
      "num_input_tokens_seen": 2336864,
      "step": 11075
    },
    {
      "epoch": 1.2189218921892189,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.01828217821782178,
      "loss": 0.2319,
      "num_input_tokens_seen": 2337856,
      "step": 11080
    },
    {
      "epoch": 1.2194719471947195,
      "grad_norm": 0.01544189453125,
      "learning_rate": 0.01829042904290429,
      "loss": 0.2298,
      "num_input_tokens_seen": 2338912,
      "step": 11085
    },
    {
      "epoch": 1.22002200220022,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.0182986798679868,
      "loss": 0.2319,
      "num_input_tokens_seen": 2339968,
      "step": 11090
    },
    {
      "epoch": 1.2205720572057206,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.018306930693069306,
      "loss": 0.2303,
      "num_input_tokens_seen": 2341120,
      "step": 11095
    },
    {
      "epoch": 1.221122112211221,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.018315181518151813,
      "loss": 0.2304,
      "num_input_tokens_seen": 2342144,
      "step": 11100
    },
    {
      "epoch": 1.2216721672167217,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.018323432343234324,
      "loss": 0.2309,
      "num_input_tokens_seen": 2343168,
      "step": 11105
    },
    {
      "epoch": 1.2222222222222223,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.01833168316831683,
      "loss": 0.2325,
      "num_input_tokens_seen": 2344256,
      "step": 11110
    },
    {
      "epoch": 1.2227722772277227,
      "grad_norm": 0.002716064453125,
      "learning_rate": 0.018339933993399338,
      "loss": 0.2346,
      "num_input_tokens_seen": 2345280,
      "step": 11115
    },
    {
      "epoch": 1.2233223322332234,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.01834818481848185,
      "loss": 0.2319,
      "num_input_tokens_seen": 2346272,
      "step": 11120
    },
    {
      "epoch": 1.2238723872387238,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.018356435643564355,
      "loss": 0.2314,
      "num_input_tokens_seen": 2347296,
      "step": 11125
    },
    {
      "epoch": 1.2244224422442245,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.018364686468646866,
      "loss": 0.2319,
      "num_input_tokens_seen": 2348352,
      "step": 11130
    },
    {
      "epoch": 1.224972497249725,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.018372937293729373,
      "loss": 0.2308,
      "num_input_tokens_seen": 2349440,
      "step": 11135
    },
    {
      "epoch": 1.2255225522552256,
      "grad_norm": 0.0084228515625,
      "learning_rate": 0.01838118811881188,
      "loss": 0.2324,
      "num_input_tokens_seen": 2350560,
      "step": 11140
    },
    {
      "epoch": 1.226072607260726,
      "grad_norm": 0.00457763671875,
      "learning_rate": 0.018389438943894387,
      "loss": 0.2303,
      "num_input_tokens_seen": 2351616,
      "step": 11145
    },
    {
      "epoch": 1.2266226622662266,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.018397689768976898,
      "loss": 0.2303,
      "num_input_tokens_seen": 2352704,
      "step": 11150
    },
    {
      "epoch": 1.2271727172717273,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.018405940594059405,
      "loss": 0.2298,
      "num_input_tokens_seen": 2353728,
      "step": 11155
    },
    {
      "epoch": 1.2277227722772277,
      "grad_norm": 0.00830078125,
      "learning_rate": 0.018414191419141912,
      "loss": 0.2298,
      "num_input_tokens_seen": 2354784,
      "step": 11160
    },
    {
      "epoch": 1.2282728272827284,
      "grad_norm": 0.003326416015625,
      "learning_rate": 0.018422442244224423,
      "loss": 0.2329,
      "num_input_tokens_seen": 2355776,
      "step": 11165
    },
    {
      "epoch": 1.2288228822882288,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.01843069306930693,
      "loss": 0.2309,
      "num_input_tokens_seen": 2356768,
      "step": 11170
    },
    {
      "epoch": 1.2293729372937294,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.01843894389438944,
      "loss": 0.231,
      "num_input_tokens_seen": 2357824,
      "step": 11175
    },
    {
      "epoch": 1.2299229922992299,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.018447194719471947,
      "loss": 0.2301,
      "num_input_tokens_seen": 2358816,
      "step": 11180
    },
    {
      "epoch": 1.2304730473047305,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.018455445544554454,
      "loss": 0.2354,
      "num_input_tokens_seen": 2359776,
      "step": 11185
    },
    {
      "epoch": 1.231023102310231,
      "grad_norm": 0.00286865234375,
      "learning_rate": 0.01846369636963696,
      "loss": 0.2343,
      "num_input_tokens_seen": 2360864,
      "step": 11190
    },
    {
      "epoch": 1.2315731573157316,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.018471947194719472,
      "loss": 0.229,
      "num_input_tokens_seen": 2361984,
      "step": 11195
    },
    {
      "epoch": 1.2321232123212322,
      "grad_norm": 0.007720947265625,
      "learning_rate": 0.01848019801980198,
      "loss": 0.2318,
      "num_input_tokens_seen": 2363104,
      "step": 11200
    },
    {
      "epoch": 1.2326732673267327,
      "grad_norm": 0.00927734375,
      "learning_rate": 0.018488448844884486,
      "loss": 0.2353,
      "num_input_tokens_seen": 2364224,
      "step": 11205
    },
    {
      "epoch": 1.2332233223322333,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.018496699669966997,
      "loss": 0.23,
      "num_input_tokens_seen": 2365280,
      "step": 11210
    },
    {
      "epoch": 1.2337733773377337,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.018504950495049504,
      "loss": 0.2326,
      "num_input_tokens_seen": 2366400,
      "step": 11215
    },
    {
      "epoch": 1.2343234323432344,
      "grad_norm": 0.0036773681640625,
      "learning_rate": 0.018513201320132015,
      "loss": 0.2284,
      "num_input_tokens_seen": 2367456,
      "step": 11220
    },
    {
      "epoch": 1.2348734873487348,
      "grad_norm": 0.00347900390625,
      "learning_rate": 0.01852145214521452,
      "loss": 0.2294,
      "num_input_tokens_seen": 2368608,
      "step": 11225
    },
    {
      "epoch": 1.2354235423542355,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.01852970297029703,
      "loss": 0.2331,
      "num_input_tokens_seen": 2369632,
      "step": 11230
    },
    {
      "epoch": 1.2359735973597359,
      "grad_norm": 0.00830078125,
      "learning_rate": 0.018537953795379536,
      "loss": 0.231,
      "num_input_tokens_seen": 2370720,
      "step": 11235
    },
    {
      "epoch": 1.2365236523652365,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.018546204620462046,
      "loss": 0.2273,
      "num_input_tokens_seen": 2371744,
      "step": 11240
    },
    {
      "epoch": 1.2370737073707372,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.018554455445544554,
      "loss": 0.232,
      "num_input_tokens_seen": 2372768,
      "step": 11245
    },
    {
      "epoch": 1.2376237623762376,
      "grad_norm": 0.0031280517578125,
      "learning_rate": 0.01856270627062706,
      "loss": 0.2341,
      "num_input_tokens_seen": 2373792,
      "step": 11250
    },
    {
      "epoch": 1.2381738173817383,
      "grad_norm": 0.007659912109375,
      "learning_rate": 0.018570957095709568,
      "loss": 0.2315,
      "num_input_tokens_seen": 2374848,
      "step": 11255
    },
    {
      "epoch": 1.2387238723872387,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.01857920792079208,
      "loss": 0.2309,
      "num_input_tokens_seen": 2375872,
      "step": 11260
    },
    {
      "epoch": 1.2392739273927393,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.01858745874587459,
      "loss": 0.2325,
      "num_input_tokens_seen": 2376960,
      "step": 11265
    },
    {
      "epoch": 1.2398239823982398,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.018595709570957096,
      "loss": 0.2309,
      "num_input_tokens_seen": 2378016,
      "step": 11270
    },
    {
      "epoch": 1.2403740374037404,
      "grad_norm": 0.0045166015625,
      "learning_rate": 0.018603960396039603,
      "loss": 0.2288,
      "num_input_tokens_seen": 2379104,
      "step": 11275
    },
    {
      "epoch": 1.2409240924092408,
      "grad_norm": 0.00384521484375,
      "learning_rate": 0.01861221122112211,
      "loss": 0.2284,
      "num_input_tokens_seen": 2380192,
      "step": 11280
    },
    {
      "epoch": 1.2414741474147415,
      "grad_norm": 0.0174560546875,
      "learning_rate": 0.01862046204620462,
      "loss": 0.2208,
      "num_input_tokens_seen": 2381216,
      "step": 11285
    },
    {
      "epoch": 1.2420242024202421,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.018628712871287128,
      "loss": 0.235,
      "num_input_tokens_seen": 2382304,
      "step": 11290
    },
    {
      "epoch": 1.2425742574257426,
      "grad_norm": 0.0198974609375,
      "learning_rate": 0.018636963696369635,
      "loss": 0.2308,
      "num_input_tokens_seen": 2383424,
      "step": 11295
    },
    {
      "epoch": 1.2431243124312432,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.018645214521452142,
      "loss": 0.2216,
      "num_input_tokens_seen": 2384480,
      "step": 11300
    },
    {
      "epoch": 1.2436743674367436,
      "grad_norm": 0.01953125,
      "learning_rate": 0.018653465346534653,
      "loss": 0.2288,
      "num_input_tokens_seen": 2385472,
      "step": 11305
    },
    {
      "epoch": 1.2442244224422443,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.018661716171617163,
      "loss": 0.2301,
      "num_input_tokens_seen": 2386528,
      "step": 11310
    },
    {
      "epoch": 1.2447744774477447,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.01866996699669967,
      "loss": 0.2303,
      "num_input_tokens_seen": 2387616,
      "step": 11315
    },
    {
      "epoch": 1.2453245324532454,
      "grad_norm": 0.02197265625,
      "learning_rate": 0.018678217821782177,
      "loss": 0.2234,
      "num_input_tokens_seen": 2388608,
      "step": 11320
    },
    {
      "epoch": 1.2458745874587458,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.018686468646864685,
      "loss": 0.2349,
      "num_input_tokens_seen": 2389664,
      "step": 11325
    },
    {
      "epoch": 1.2464246424642464,
      "grad_norm": 0.01708984375,
      "learning_rate": 0.018694719471947195,
      "loss": 0.2373,
      "num_input_tokens_seen": 2390656,
      "step": 11330
    },
    {
      "epoch": 1.246974697469747,
      "grad_norm": 0.003997802734375,
      "learning_rate": 0.018702970297029702,
      "loss": 0.2361,
      "num_input_tokens_seen": 2391744,
      "step": 11335
    },
    {
      "epoch": 1.2475247524752475,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.01871122112211221,
      "loss": 0.2303,
      "num_input_tokens_seen": 2392800,
      "step": 11340
    },
    {
      "epoch": 1.2480748074807482,
      "grad_norm": 0.015625,
      "learning_rate": 0.018719471947194716,
      "loss": 0.2322,
      "num_input_tokens_seen": 2393856,
      "step": 11345
    },
    {
      "epoch": 1.2486248624862486,
      "grad_norm": 0.0042724609375,
      "learning_rate": 0.018727722772277227,
      "loss": 0.2317,
      "num_input_tokens_seen": 2394944,
      "step": 11350
    },
    {
      "epoch": 1.2491749174917492,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.018735973597359738,
      "loss": 0.2263,
      "num_input_tokens_seen": 2395968,
      "step": 11355
    },
    {
      "epoch": 1.2497249724972497,
      "grad_norm": 0.00396728515625,
      "learning_rate": 0.018744224422442245,
      "loss": 0.2325,
      "num_input_tokens_seen": 2396992,
      "step": 11360
    },
    {
      "epoch": 1.2502750275027503,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.018752475247524752,
      "loss": 0.2424,
      "num_input_tokens_seen": 2398080,
      "step": 11365
    },
    {
      "epoch": 1.2508250825082508,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.01876072607260726,
      "loss": 0.2244,
      "num_input_tokens_seen": 2399200,
      "step": 11370
    },
    {
      "epoch": 1.2513751375137514,
      "grad_norm": 0.0123291015625,
      "learning_rate": 0.01876897689768977,
      "loss": 0.2384,
      "num_input_tokens_seen": 2400224,
      "step": 11375
    },
    {
      "epoch": 1.251925192519252,
      "grad_norm": 0.019287109375,
      "learning_rate": 0.018777227722772277,
      "loss": 0.2347,
      "num_input_tokens_seen": 2401216,
      "step": 11380
    },
    {
      "epoch": 1.2524752475247525,
      "grad_norm": 0.0390625,
      "learning_rate": 0.018785478547854784,
      "loss": 0.2304,
      "num_input_tokens_seen": 2402240,
      "step": 11385
    },
    {
      "epoch": 1.253025302530253,
      "grad_norm": 0.0128173828125,
      "learning_rate": 0.01879372937293729,
      "loss": 0.2295,
      "num_input_tokens_seen": 2403392,
      "step": 11390
    },
    {
      "epoch": 1.2535753575357536,
      "grad_norm": 0.031494140625,
      "learning_rate": 0.0188019801980198,
      "loss": 0.2308,
      "num_input_tokens_seen": 2404480,
      "step": 11395
    },
    {
      "epoch": 1.2541254125412542,
      "grad_norm": 0.0537109375,
      "learning_rate": 0.018810231023102312,
      "loss": 0.2392,
      "num_input_tokens_seen": 2405568,
      "step": 11400
    },
    {
      "epoch": 1.2546754675467546,
      "grad_norm": 0.031494140625,
      "learning_rate": 0.01881848184818482,
      "loss": 0.2234,
      "num_input_tokens_seen": 2406688,
      "step": 11405
    },
    {
      "epoch": 1.2552255225522553,
      "grad_norm": 0.023681640625,
      "learning_rate": 0.018826732673267326,
      "loss": 0.2413,
      "num_input_tokens_seen": 2407744,
      "step": 11410
    },
    {
      "epoch": 1.2557755775577557,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.018834983498349833,
      "loss": 0.2314,
      "num_input_tokens_seen": 2408800,
      "step": 11415
    },
    {
      "epoch": 1.2563256325632564,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.018843234323432344,
      "loss": 0.2324,
      "num_input_tokens_seen": 2409888,
      "step": 11420
    },
    {
      "epoch": 1.256875687568757,
      "grad_norm": 0.023681640625,
      "learning_rate": 0.01885148514851485,
      "loss": 0.2304,
      "num_input_tokens_seen": 2410944,
      "step": 11425
    },
    {
      "epoch": 1.2574257425742574,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.018859735973597358,
      "loss": 0.2299,
      "num_input_tokens_seen": 2412000,
      "step": 11430
    },
    {
      "epoch": 1.2579757975797579,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.018867986798679865,
      "loss": 0.2372,
      "num_input_tokens_seen": 2413024,
      "step": 11435
    },
    {
      "epoch": 1.2585258525852585,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.018876237623762376,
      "loss": 0.2314,
      "num_input_tokens_seen": 2414144,
      "step": 11440
    },
    {
      "epoch": 1.2590759075907592,
      "grad_norm": 0.01190185546875,
      "learning_rate": 0.018884488448844886,
      "loss": 0.2319,
      "num_input_tokens_seen": 2415168,
      "step": 11445
    },
    {
      "epoch": 1.2596259625962596,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.018892739273927393,
      "loss": 0.2308,
      "num_input_tokens_seen": 2416256,
      "step": 11450
    },
    {
      "epoch": 1.2601760176017602,
      "grad_norm": 0.00341796875,
      "learning_rate": 0.0189009900990099,
      "loss": 0.234,
      "num_input_tokens_seen": 2417280,
      "step": 11455
    },
    {
      "epoch": 1.2607260726072607,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.018909240924092408,
      "loss": 0.2319,
      "num_input_tokens_seen": 2418304,
      "step": 11460
    },
    {
      "epoch": 1.2612761276127613,
      "grad_norm": 0.00347900390625,
      "learning_rate": 0.018917491749174918,
      "loss": 0.2314,
      "num_input_tokens_seen": 2419328,
      "step": 11465
    },
    {
      "epoch": 1.261826182618262,
      "grad_norm": 0.0031890869140625,
      "learning_rate": 0.018925742574257425,
      "loss": 0.232,
      "num_input_tokens_seen": 2420384,
      "step": 11470
    },
    {
      "epoch": 1.2623762376237624,
      "grad_norm": 0.0218505859375,
      "learning_rate": 0.018933993399339932,
      "loss": 0.233,
      "num_input_tokens_seen": 2421440,
      "step": 11475
    },
    {
      "epoch": 1.2629262926292628,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.01894224422442244,
      "loss": 0.2299,
      "num_input_tokens_seen": 2422464,
      "step": 11480
    },
    {
      "epoch": 1.2634763476347635,
      "grad_norm": 0.003265380859375,
      "learning_rate": 0.01895049504950495,
      "loss": 0.2316,
      "num_input_tokens_seen": 2423520,
      "step": 11485
    },
    {
      "epoch": 1.2640264026402641,
      "grad_norm": 0.009765625,
      "learning_rate": 0.018958745874587457,
      "loss": 0.2291,
      "num_input_tokens_seen": 2424672,
      "step": 11490
    },
    {
      "epoch": 1.2645764576457645,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.018966996699669968,
      "loss": 0.236,
      "num_input_tokens_seen": 2425696,
      "step": 11495
    },
    {
      "epoch": 1.2651265126512652,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.018975247524752475,
      "loss": 0.2307,
      "num_input_tokens_seen": 2426656,
      "step": 11500
    },
    {
      "epoch": 1.2656765676567656,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.018983498349834982,
      "loss": 0.2327,
      "num_input_tokens_seen": 2427776,
      "step": 11505
    },
    {
      "epoch": 1.2662266226622663,
      "grad_norm": 0.0038299560546875,
      "learning_rate": 0.018991749174917492,
      "loss": 0.232,
      "num_input_tokens_seen": 2428832,
      "step": 11510
    },
    {
      "epoch": 1.266776677667767,
      "grad_norm": 0.003265380859375,
      "learning_rate": 0.019,
      "loss": 0.2325,
      "num_input_tokens_seen": 2429888,
      "step": 11515
    },
    {
      "epoch": 1.2673267326732673,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.019008250825082507,
      "loss": 0.2299,
      "num_input_tokens_seen": 2430976,
      "step": 11520
    },
    {
      "epoch": 1.2678767876787678,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.019016501650165014,
      "loss": 0.2325,
      "num_input_tokens_seen": 2432032,
      "step": 11525
    },
    {
      "epoch": 1.2684268426842684,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.019024752475247524,
      "loss": 0.2289,
      "num_input_tokens_seen": 2433120,
      "step": 11530
    },
    {
      "epoch": 1.268976897689769,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.01903300330033003,
      "loss": 0.2292,
      "num_input_tokens_seen": 2434240,
      "step": 11535
    },
    {
      "epoch": 1.2695269526952695,
      "grad_norm": 0.0038909912109375,
      "learning_rate": 0.019041254125412542,
      "loss": 0.2288,
      "num_input_tokens_seen": 2435296,
      "step": 11540
    },
    {
      "epoch": 1.2700770077007701,
      "grad_norm": 0.00299072265625,
      "learning_rate": 0.01904950495049505,
      "loss": 0.2326,
      "num_input_tokens_seen": 2436384,
      "step": 11545
    },
    {
      "epoch": 1.2706270627062706,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.019057755775577556,
      "loss": 0.2303,
      "num_input_tokens_seen": 2437408,
      "step": 11550
    },
    {
      "epoch": 1.2711771177117712,
      "grad_norm": 0.0038604736328125,
      "learning_rate": 0.019066006600660067,
      "loss": 0.2394,
      "num_input_tokens_seen": 2438464,
      "step": 11555
    },
    {
      "epoch": 1.2717271727172716,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.019074257425742574,
      "loss": 0.2282,
      "num_input_tokens_seen": 2439488,
      "step": 11560
    },
    {
      "epoch": 1.2722772277227723,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01908250825082508,
      "loss": 0.2364,
      "num_input_tokens_seen": 2440544,
      "step": 11565
    },
    {
      "epoch": 1.2728272827282727,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.019090759075907588,
      "loss": 0.2301,
      "num_input_tokens_seen": 2441600,
      "step": 11570
    },
    {
      "epoch": 1.2733773377337734,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.0190990099009901,
      "loss": 0.2305,
      "num_input_tokens_seen": 2442656,
      "step": 11575
    },
    {
      "epoch": 1.273927392739274,
      "grad_norm": 0.019287109375,
      "learning_rate": 0.019107260726072606,
      "loss": 0.2321,
      "num_input_tokens_seen": 2443680,
      "step": 11580
    },
    {
      "epoch": 1.2744774477447744,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.019115511551155116,
      "loss": 0.2275,
      "num_input_tokens_seen": 2444736,
      "step": 11585
    },
    {
      "epoch": 1.275027502750275,
      "grad_norm": 0.004180908203125,
      "learning_rate": 0.019123762376237623,
      "loss": 0.2291,
      "num_input_tokens_seen": 2445824,
      "step": 11590
    },
    {
      "epoch": 1.2755775577557755,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.01913201320132013,
      "loss": 0.2359,
      "num_input_tokens_seen": 2446816,
      "step": 11595
    },
    {
      "epoch": 1.2761276127612762,
      "grad_norm": 0.004302978515625,
      "learning_rate": 0.01914026402640264,
      "loss": 0.2306,
      "num_input_tokens_seen": 2447904,
      "step": 11600
    },
    {
      "epoch": 1.2766776677667766,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.019148514851485148,
      "loss": 0.23,
      "num_input_tokens_seen": 2448960,
      "step": 11605
    },
    {
      "epoch": 1.2772277227722773,
      "grad_norm": 0.02734375,
      "learning_rate": 0.019156765676567655,
      "loss": 0.2322,
      "num_input_tokens_seen": 2450016,
      "step": 11610
    },
    {
      "epoch": 1.2777777777777777,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.019165016501650162,
      "loss": 0.2284,
      "num_input_tokens_seen": 2451072,
      "step": 11615
    },
    {
      "epoch": 1.2783278327832783,
      "grad_norm": 0.0029754638671875,
      "learning_rate": 0.019173267326732673,
      "loss": 0.2324,
      "num_input_tokens_seen": 2452128,
      "step": 11620
    },
    {
      "epoch": 1.278877887788779,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.01918151815181518,
      "loss": 0.237,
      "num_input_tokens_seen": 2453216,
      "step": 11625
    },
    {
      "epoch": 1.2794279427942794,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.01918976897689769,
      "loss": 0.2286,
      "num_input_tokens_seen": 2454304,
      "step": 11630
    },
    {
      "epoch": 1.27997799779978,
      "grad_norm": 0.02880859375,
      "learning_rate": 0.019198019801980198,
      "loss": 0.2338,
      "num_input_tokens_seen": 2455392,
      "step": 11635
    },
    {
      "epoch": 1.2805280528052805,
      "grad_norm": 0.0225830078125,
      "learning_rate": 0.019206270627062705,
      "loss": 0.231,
      "num_input_tokens_seen": 2456512,
      "step": 11640
    },
    {
      "epoch": 1.2810781078107811,
      "grad_norm": 0.0205078125,
      "learning_rate": 0.019214521452145215,
      "loss": 0.2297,
      "num_input_tokens_seen": 2457504,
      "step": 11645
    },
    {
      "epoch": 1.2816281628162816,
      "grad_norm": 0.0157470703125,
      "learning_rate": 0.019222772277227722,
      "loss": 0.2349,
      "num_input_tokens_seen": 2458592,
      "step": 11650
    },
    {
      "epoch": 1.2821782178217822,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.01923102310231023,
      "loss": 0.2293,
      "num_input_tokens_seen": 2459616,
      "step": 11655
    },
    {
      "epoch": 1.2827282728272826,
      "grad_norm": 0.0284423828125,
      "learning_rate": 0.019239273927392737,
      "loss": 0.2253,
      "num_input_tokens_seen": 2460704,
      "step": 11660
    },
    {
      "epoch": 1.2832783278327833,
      "grad_norm": 0.03955078125,
      "learning_rate": 0.019247524752475247,
      "loss": 0.2329,
      "num_input_tokens_seen": 2461760,
      "step": 11665
    },
    {
      "epoch": 1.283828382838284,
      "grad_norm": 0.0281982421875,
      "learning_rate": 0.019255775577557754,
      "loss": 0.2309,
      "num_input_tokens_seen": 2462816,
      "step": 11670
    },
    {
      "epoch": 1.2843784378437844,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.019264026402640265,
      "loss": 0.2339,
      "num_input_tokens_seen": 2463936,
      "step": 11675
    },
    {
      "epoch": 1.284928492849285,
      "grad_norm": 0.0299072265625,
      "learning_rate": 0.019272277227722772,
      "loss": 0.2363,
      "num_input_tokens_seen": 2464928,
      "step": 11680
    },
    {
      "epoch": 1.2854785478547854,
      "grad_norm": 0.0263671875,
      "learning_rate": 0.01928052805280528,
      "loss": 0.2319,
      "num_input_tokens_seen": 2466016,
      "step": 11685
    },
    {
      "epoch": 1.286028602860286,
      "grad_norm": 0.0252685546875,
      "learning_rate": 0.01928877887788779,
      "loss": 0.2337,
      "num_input_tokens_seen": 2467168,
      "step": 11690
    },
    {
      "epoch": 1.2865786578657865,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.019297029702970297,
      "loss": 0.2299,
      "num_input_tokens_seen": 2468192,
      "step": 11695
    },
    {
      "epoch": 1.2871287128712872,
      "grad_norm": 0.014404296875,
      "learning_rate": 0.019305280528052804,
      "loss": 0.2314,
      "num_input_tokens_seen": 2469248,
      "step": 11700
    },
    {
      "epoch": 1.2876787678767876,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.01931353135313531,
      "loss": 0.2305,
      "num_input_tokens_seen": 2470272,
      "step": 11705
    },
    {
      "epoch": 1.2882288228822882,
      "grad_norm": 0.0072021484375,
      "learning_rate": 0.01932178217821782,
      "loss": 0.2311,
      "num_input_tokens_seen": 2471264,
      "step": 11710
    },
    {
      "epoch": 1.2887788778877889,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.01933003300330033,
      "loss": 0.2291,
      "num_input_tokens_seen": 2472384,
      "step": 11715
    },
    {
      "epoch": 1.2893289328932893,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.01933828382838284,
      "loss": 0.2291,
      "num_input_tokens_seen": 2473472,
      "step": 11720
    },
    {
      "epoch": 1.2898789878987897,
      "grad_norm": 0.01507568359375,
      "learning_rate": 0.019346534653465346,
      "loss": 0.2282,
      "num_input_tokens_seen": 2474560,
      "step": 11725
    },
    {
      "epoch": 1.2904290429042904,
      "grad_norm": 0.017578125,
      "learning_rate": 0.019354785478547853,
      "loss": 0.2377,
      "num_input_tokens_seen": 2475616,
      "step": 11730
    },
    {
      "epoch": 1.290979097909791,
      "grad_norm": 0.01239013671875,
      "learning_rate": 0.019363036303630364,
      "loss": 0.2271,
      "num_input_tokens_seen": 2476608,
      "step": 11735
    },
    {
      "epoch": 1.2915291529152915,
      "grad_norm": 0.013916015625,
      "learning_rate": 0.01937128712871287,
      "loss": 0.2371,
      "num_input_tokens_seen": 2477664,
      "step": 11740
    },
    {
      "epoch": 1.2920792079207921,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.019379537953795378,
      "loss": 0.2301,
      "num_input_tokens_seen": 2478720,
      "step": 11745
    },
    {
      "epoch": 1.2926292629262925,
      "grad_norm": 0.00897216796875,
      "learning_rate": 0.019387788778877885,
      "loss": 0.231,
      "num_input_tokens_seen": 2479776,
      "step": 11750
    },
    {
      "epoch": 1.2931793179317932,
      "grad_norm": 0.0228271484375,
      "learning_rate": 0.019396039603960396,
      "loss": 0.2335,
      "num_input_tokens_seen": 2480800,
      "step": 11755
    },
    {
      "epoch": 1.2937293729372938,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.019404290429042903,
      "loss": 0.2308,
      "num_input_tokens_seen": 2481824,
      "step": 11760
    },
    {
      "epoch": 1.2942794279427943,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.019412541254125414,
      "loss": 0.2319,
      "num_input_tokens_seen": 2482912,
      "step": 11765
    },
    {
      "epoch": 1.2948294829482947,
      "grad_norm": 0.0035400390625,
      "learning_rate": 0.019420792079207917,
      "loss": 0.2334,
      "num_input_tokens_seen": 2483904,
      "step": 11770
    },
    {
      "epoch": 1.2953795379537953,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.019429042904290428,
      "loss": 0.2314,
      "num_input_tokens_seen": 2484928,
      "step": 11775
    },
    {
      "epoch": 1.295929592959296,
      "grad_norm": 0.00433349609375,
      "learning_rate": 0.01943729372937294,
      "loss": 0.2308,
      "num_input_tokens_seen": 2486080,
      "step": 11780
    },
    {
      "epoch": 1.2964796479647964,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.019445544554455445,
      "loss": 0.2298,
      "num_input_tokens_seen": 2487104,
      "step": 11785
    },
    {
      "epoch": 1.297029702970297,
      "grad_norm": 0.01422119140625,
      "learning_rate": 0.019453795379537953,
      "loss": 0.235,
      "num_input_tokens_seen": 2488128,
      "step": 11790
    },
    {
      "epoch": 1.2975797579757975,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.01946204620462046,
      "loss": 0.2329,
      "num_input_tokens_seen": 2489184,
      "step": 11795
    },
    {
      "epoch": 1.2981298129812981,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.01947029702970297,
      "loss": 0.2298,
      "num_input_tokens_seen": 2490336,
      "step": 11800
    },
    {
      "epoch": 1.2986798679867988,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.019478547854785477,
      "loss": 0.2314,
      "num_input_tokens_seen": 2491488,
      "step": 11805
    },
    {
      "epoch": 1.2992299229922992,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.019486798679867988,
      "loss": 0.2284,
      "num_input_tokens_seen": 2492480,
      "step": 11810
    },
    {
      "epoch": 1.2997799779977997,
      "grad_norm": 0.0238037109375,
      "learning_rate": 0.01949504950495049,
      "loss": 0.2311,
      "num_input_tokens_seen": 2493568,
      "step": 11815
    },
    {
      "epoch": 1.3003300330033003,
      "grad_norm": 0.020751953125,
      "learning_rate": 0.019503300330033002,
      "loss": 0.2291,
      "num_input_tokens_seen": 2494656,
      "step": 11820
    },
    {
      "epoch": 1.300880088008801,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.019511551155115513,
      "loss": 0.2282,
      "num_input_tokens_seen": 2495680,
      "step": 11825
    },
    {
      "epoch": 1.3014301430143014,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.01951980198019802,
      "loss": 0.2311,
      "num_input_tokens_seen": 2496704,
      "step": 11830
    },
    {
      "epoch": 1.301980198019802,
      "grad_norm": 0.00433349609375,
      "learning_rate": 0.019528052805280527,
      "loss": 0.2365,
      "num_input_tokens_seen": 2497824,
      "step": 11835
    },
    {
      "epoch": 1.3025302530253025,
      "grad_norm": 0.01361083984375,
      "learning_rate": 0.019536303630363034,
      "loss": 0.2316,
      "num_input_tokens_seen": 2498912,
      "step": 11840
    },
    {
      "epoch": 1.303080308030803,
      "grad_norm": 0.0142822265625,
      "learning_rate": 0.019544554455445545,
      "loss": 0.2366,
      "num_input_tokens_seen": 2500032,
      "step": 11845
    },
    {
      "epoch": 1.3036303630363038,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.01955280528052805,
      "loss": 0.2322,
      "num_input_tokens_seen": 2501088,
      "step": 11850
    },
    {
      "epoch": 1.3041804180418042,
      "grad_norm": 0.01202392578125,
      "learning_rate": 0.019561056105610562,
      "loss": 0.229,
      "num_input_tokens_seen": 2502144,
      "step": 11855
    },
    {
      "epoch": 1.3047304730473046,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.019569306930693066,
      "loss": 0.2336,
      "num_input_tokens_seen": 2503168,
      "step": 11860
    },
    {
      "epoch": 1.3052805280528053,
      "grad_norm": 0.0128173828125,
      "learning_rate": 0.019577557755775576,
      "loss": 0.2346,
      "num_input_tokens_seen": 2504192,
      "step": 11865
    },
    {
      "epoch": 1.305830583058306,
      "grad_norm": 0.0218505859375,
      "learning_rate": 0.019585808580858087,
      "loss": 0.2335,
      "num_input_tokens_seen": 2505216,
      "step": 11870
    },
    {
      "epoch": 1.3063806380638063,
      "grad_norm": 0.01190185546875,
      "learning_rate": 0.019594059405940594,
      "loss": 0.2319,
      "num_input_tokens_seen": 2506272,
      "step": 11875
    },
    {
      "epoch": 1.306930693069307,
      "grad_norm": 0.01409912109375,
      "learning_rate": 0.0196023102310231,
      "loss": 0.2308,
      "num_input_tokens_seen": 2507360,
      "step": 11880
    },
    {
      "epoch": 1.3074807480748074,
      "grad_norm": 0.01336669921875,
      "learning_rate": 0.01961056105610561,
      "loss": 0.2298,
      "num_input_tokens_seen": 2508416,
      "step": 11885
    },
    {
      "epoch": 1.308030803080308,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.01961881188118812,
      "loss": 0.2319,
      "num_input_tokens_seen": 2509440,
      "step": 11890
    },
    {
      "epoch": 1.3085808580858087,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.019627062706270626,
      "loss": 0.234,
      "num_input_tokens_seen": 2510496,
      "step": 11895
    },
    {
      "epoch": 1.3091309130913091,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.019635313531353137,
      "loss": 0.2335,
      "num_input_tokens_seen": 2511488,
      "step": 11900
    },
    {
      "epoch": 1.3096809680968096,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.01964356435643564,
      "loss": 0.2329,
      "num_input_tokens_seen": 2512544,
      "step": 11905
    },
    {
      "epoch": 1.3102310231023102,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.01965181518151815,
      "loss": 0.233,
      "num_input_tokens_seen": 2513600,
      "step": 11910
    },
    {
      "epoch": 1.3107810781078109,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.01966006600660066,
      "loss": 0.2303,
      "num_input_tokens_seen": 2514624,
      "step": 11915
    },
    {
      "epoch": 1.3113311331133113,
      "grad_norm": 0.004119873046875,
      "learning_rate": 0.01966831683168317,
      "loss": 0.2314,
      "num_input_tokens_seen": 2515712,
      "step": 11920
    },
    {
      "epoch": 1.311881188118812,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.019676567656765676,
      "loss": 0.2319,
      "num_input_tokens_seen": 2516768,
      "step": 11925
    },
    {
      "epoch": 1.3124312431243124,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.019684818481848183,
      "loss": 0.2293,
      "num_input_tokens_seen": 2517824,
      "step": 11930
    },
    {
      "epoch": 1.312981298129813,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.019693069306930693,
      "loss": 0.2324,
      "num_input_tokens_seen": 2518912,
      "step": 11935
    },
    {
      "epoch": 1.3135313531353137,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.0197013201320132,
      "loss": 0.2319,
      "num_input_tokens_seen": 2519968,
      "step": 11940
    },
    {
      "epoch": 1.314081408140814,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.01970957095709571,
      "loss": 0.2308,
      "num_input_tokens_seen": 2521024,
      "step": 11945
    },
    {
      "epoch": 1.3146314631463145,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.019717821782178215,
      "loss": 0.2319,
      "num_input_tokens_seen": 2522048,
      "step": 11950
    },
    {
      "epoch": 1.3151815181518152,
      "grad_norm": 0.014404296875,
      "learning_rate": 0.019726072607260725,
      "loss": 0.2329,
      "num_input_tokens_seen": 2523104,
      "step": 11955
    },
    {
      "epoch": 1.3157315731573158,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.019734323432343236,
      "loss": 0.2298,
      "num_input_tokens_seen": 2524160,
      "step": 11960
    },
    {
      "epoch": 1.3162816281628162,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.019742574257425743,
      "loss": 0.2314,
      "num_input_tokens_seen": 2525248,
      "step": 11965
    },
    {
      "epoch": 1.316831683168317,
      "grad_norm": 0.0038909912109375,
      "learning_rate": 0.01975082508250825,
      "loss": 0.2334,
      "num_input_tokens_seen": 2526304,
      "step": 11970
    },
    {
      "epoch": 1.3173817381738173,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.019759075907590757,
      "loss": 0.2309,
      "num_input_tokens_seen": 2527360,
      "step": 11975
    },
    {
      "epoch": 1.317931793179318,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.019767326732673268,
      "loss": 0.2298,
      "num_input_tokens_seen": 2528416,
      "step": 11980
    },
    {
      "epoch": 1.3184818481848186,
      "grad_norm": 0.0228271484375,
      "learning_rate": 0.019775577557755775,
      "loss": 0.2273,
      "num_input_tokens_seen": 2529440,
      "step": 11985
    },
    {
      "epoch": 1.319031903190319,
      "grad_norm": 0.0147705078125,
      "learning_rate": 0.019783828382838285,
      "loss": 0.2289,
      "num_input_tokens_seen": 2530432,
      "step": 11990
    },
    {
      "epoch": 1.3195819581958195,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01979207920792079,
      "loss": 0.2337,
      "num_input_tokens_seen": 2531424,
      "step": 11995
    },
    {
      "epoch": 1.3201320132013201,
      "grad_norm": 0.0235595703125,
      "learning_rate": 0.0198003300330033,
      "loss": 0.2361,
      "num_input_tokens_seen": 2532480,
      "step": 12000
    },
    {
      "epoch": 1.3206820682068208,
      "grad_norm": 0.0142822265625,
      "learning_rate": 0.019808580858085807,
      "loss": 0.2319,
      "num_input_tokens_seen": 2533568,
      "step": 12005
    },
    {
      "epoch": 1.3212321232123212,
      "grad_norm": 0.01202392578125,
      "learning_rate": 0.019816831683168317,
      "loss": 0.2325,
      "num_input_tokens_seen": 2534560,
      "step": 12010
    },
    {
      "epoch": 1.3217821782178218,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.019825082508250824,
      "loss": 0.2304,
      "num_input_tokens_seen": 2535584,
      "step": 12015
    },
    {
      "epoch": 1.3223322332233223,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.01983333333333333,
      "loss": 0.2319,
      "num_input_tokens_seen": 2536576,
      "step": 12020
    },
    {
      "epoch": 1.322882288228823,
      "grad_norm": 0.004425048828125,
      "learning_rate": 0.019841584158415842,
      "loss": 0.2308,
      "num_input_tokens_seen": 2537568,
      "step": 12025
    },
    {
      "epoch": 1.3234323432343233,
      "grad_norm": 0.01470947265625,
      "learning_rate": 0.01984983498349835,
      "loss": 0.2288,
      "num_input_tokens_seen": 2538592,
      "step": 12030
    },
    {
      "epoch": 1.323982398239824,
      "grad_norm": 0.01275634765625,
      "learning_rate": 0.01985808580858086,
      "loss": 0.2313,
      "num_input_tokens_seen": 2539680,
      "step": 12035
    },
    {
      "epoch": 1.3245324532453244,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019866336633663363,
      "loss": 0.2324,
      "num_input_tokens_seen": 2540736,
      "step": 12040
    },
    {
      "epoch": 1.325082508250825,
      "grad_norm": 0.01397705078125,
      "learning_rate": 0.019874587458745874,
      "loss": 0.2319,
      "num_input_tokens_seen": 2541824,
      "step": 12045
    },
    {
      "epoch": 1.3256325632563257,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.01988283828382838,
      "loss": 0.2351,
      "num_input_tokens_seen": 2542912,
      "step": 12050
    },
    {
      "epoch": 1.3261826182618262,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.01989108910891089,
      "loss": 0.2329,
      "num_input_tokens_seen": 2544064,
      "step": 12055
    },
    {
      "epoch": 1.3267326732673268,
      "grad_norm": 0.01458740234375,
      "learning_rate": 0.0198993399339934,
      "loss": 0.2329,
      "num_input_tokens_seen": 2545088,
      "step": 12060
    },
    {
      "epoch": 1.3272827282728272,
      "grad_norm": 0.0167236328125,
      "learning_rate": 0.019907590759075906,
      "loss": 0.2298,
      "num_input_tokens_seen": 2546144,
      "step": 12065
    },
    {
      "epoch": 1.3278327832783279,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.019915841584158416,
      "loss": 0.227,
      "num_input_tokens_seen": 2547136,
      "step": 12070
    },
    {
      "epoch": 1.3283828382838283,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.019924092409240923,
      "loss": 0.2303,
      "num_input_tokens_seen": 2548160,
      "step": 12075
    },
    {
      "epoch": 1.328932893289329,
      "grad_norm": 0.007171630859375,
      "learning_rate": 0.019932343234323434,
      "loss": 0.2307,
      "num_input_tokens_seen": 2549216,
      "step": 12080
    },
    {
      "epoch": 1.3294829482948294,
      "grad_norm": 0.019287109375,
      "learning_rate": 0.019940594059405937,
      "loss": 0.2098,
      "num_input_tokens_seen": 2550240,
      "step": 12085
    },
    {
      "epoch": 1.33003300330033,
      "grad_norm": 0.07763671875,
      "learning_rate": 0.019948844884488448,
      "loss": 0.2463,
      "num_input_tokens_seen": 2551296,
      "step": 12090
    },
    {
      "epoch": 1.3305830583058307,
      "grad_norm": 0.439453125,
      "learning_rate": 0.019957095709570955,
      "loss": 0.5053,
      "num_input_tokens_seen": 2552352,
      "step": 12095
    },
    {
      "epoch": 1.331133113311331,
      "grad_norm": 0.0169677734375,
      "learning_rate": 0.019965346534653466,
      "loss": 0.239,
      "num_input_tokens_seen": 2553376,
      "step": 12100
    },
    {
      "epoch": 1.3316831683168318,
      "grad_norm": 0.0233154296875,
      "learning_rate": 0.019973597359735973,
      "loss": 0.2306,
      "num_input_tokens_seen": 2554400,
      "step": 12105
    },
    {
      "epoch": 1.3322332233223322,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.01998184818481848,
      "loss": 0.2329,
      "num_input_tokens_seen": 2555456,
      "step": 12110
    },
    {
      "epoch": 1.3327832783278328,
      "grad_norm": 0.212890625,
      "learning_rate": 0.01999009900990099,
      "loss": 0.2295,
      "num_input_tokens_seen": 2556608,
      "step": 12115
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.0751953125,
      "learning_rate": 0.019998349834983498,
      "loss": 0.2562,
      "num_input_tokens_seen": 2557696,
      "step": 12120
    },
    {
      "epoch": 1.333883388338834,
      "grad_norm": 0.036865234375,
      "learning_rate": 0.020006600660066008,
      "loss": 0.2565,
      "num_input_tokens_seen": 2558816,
      "step": 12125
    },
    {
      "epoch": 1.3344334433443343,
      "grad_norm": 0.054931640625,
      "learning_rate": 0.020014851485148512,
      "loss": 0.2375,
      "num_input_tokens_seen": 2559808,
      "step": 12130
    },
    {
      "epoch": 1.334983498349835,
      "grad_norm": 0.025390625,
      "learning_rate": 0.020023102310231022,
      "loss": 0.2308,
      "num_input_tokens_seen": 2560864,
      "step": 12135
    },
    {
      "epoch": 1.3355335533553356,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02003135313531353,
      "loss": 0.2327,
      "num_input_tokens_seen": 2561952,
      "step": 12140
    },
    {
      "epoch": 1.336083608360836,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02003960396039604,
      "loss": 0.2284,
      "num_input_tokens_seen": 2563008,
      "step": 12145
    },
    {
      "epoch": 1.3366336633663367,
      "grad_norm": 0.10107421875,
      "learning_rate": 0.020047854785478547,
      "loss": 0.2336,
      "num_input_tokens_seen": 2564032,
      "step": 12150
    },
    {
      "epoch": 1.3371837183718371,
      "grad_norm": 0.0233154296875,
      "learning_rate": 0.020056105610561054,
      "loss": 0.2431,
      "num_input_tokens_seen": 2565088,
      "step": 12155
    },
    {
      "epoch": 1.3377337733773378,
      "grad_norm": 0.0198974609375,
      "learning_rate": 0.020064356435643565,
      "loss": 0.2326,
      "num_input_tokens_seen": 2566144,
      "step": 12160
    },
    {
      "epoch": 1.3382838283828382,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.020072607260726072,
      "loss": 0.2337,
      "num_input_tokens_seen": 2567200,
      "step": 12165
    },
    {
      "epoch": 1.3388338833883389,
      "grad_norm": 0.0216064453125,
      "learning_rate": 0.020080858085808583,
      "loss": 0.2301,
      "num_input_tokens_seen": 2568288,
      "step": 12170
    },
    {
      "epoch": 1.3393839383938393,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.020089108910891086,
      "loss": 0.2299,
      "num_input_tokens_seen": 2569408,
      "step": 12175
    },
    {
      "epoch": 1.33993399339934,
      "grad_norm": 0.0194091796875,
      "learning_rate": 0.020097359735973597,
      "loss": 0.2381,
      "num_input_tokens_seen": 2570496,
      "step": 12180
    },
    {
      "epoch": 1.3404840484048406,
      "grad_norm": 0.00421142578125,
      "learning_rate": 0.020105610561056104,
      "loss": 0.2354,
      "num_input_tokens_seen": 2571520,
      "step": 12185
    },
    {
      "epoch": 1.341034103410341,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.020113861386138614,
      "loss": 0.232,
      "num_input_tokens_seen": 2572576,
      "step": 12190
    },
    {
      "epoch": 1.3415841584158417,
      "grad_norm": 0.016845703125,
      "learning_rate": 0.02012211221122112,
      "loss": 0.2299,
      "num_input_tokens_seen": 2573632,
      "step": 12195
    },
    {
      "epoch": 1.342134213421342,
      "grad_norm": 0.0152587890625,
      "learning_rate": 0.02013036303630363,
      "loss": 0.2272,
      "num_input_tokens_seen": 2574752,
      "step": 12200
    },
    {
      "epoch": 1.3426842684268427,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.02013861386138614,
      "loss": 0.2279,
      "num_input_tokens_seen": 2575808,
      "step": 12205
    },
    {
      "epoch": 1.3432343234323432,
      "grad_norm": 0.0174560546875,
      "learning_rate": 0.020146864686468646,
      "loss": 0.2434,
      "num_input_tokens_seen": 2576864,
      "step": 12210
    },
    {
      "epoch": 1.3437843784378438,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.020155115511551157,
      "loss": 0.2401,
      "num_input_tokens_seen": 2577952,
      "step": 12215
    },
    {
      "epoch": 1.3443344334433442,
      "grad_norm": 0.0211181640625,
      "learning_rate": 0.02016336633663366,
      "loss": 0.2335,
      "num_input_tokens_seen": 2578976,
      "step": 12220
    },
    {
      "epoch": 1.344884488448845,
      "grad_norm": 0.004150390625,
      "learning_rate": 0.02017161716171617,
      "loss": 0.2348,
      "num_input_tokens_seen": 2580032,
      "step": 12225
    },
    {
      "epoch": 1.3454345434543455,
      "grad_norm": 0.0478515625,
      "learning_rate": 0.020179867986798678,
      "loss": 0.2338,
      "num_input_tokens_seen": 2581056,
      "step": 12230
    },
    {
      "epoch": 1.345984598459846,
      "grad_norm": 0.0390625,
      "learning_rate": 0.02018811881188119,
      "loss": 0.2346,
      "num_input_tokens_seen": 2582112,
      "step": 12235
    },
    {
      "epoch": 1.3465346534653464,
      "grad_norm": 0.0208740234375,
      "learning_rate": 0.020196369636963696,
      "loss": 0.2321,
      "num_input_tokens_seen": 2583168,
      "step": 12240
    },
    {
      "epoch": 1.347084708470847,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.020204620462046203,
      "loss": 0.2278,
      "num_input_tokens_seen": 2584160,
      "step": 12245
    },
    {
      "epoch": 1.3476347634763477,
      "grad_norm": 0.0150146484375,
      "learning_rate": 0.020212871287128713,
      "loss": 0.2361,
      "num_input_tokens_seen": 2585152,
      "step": 12250
    },
    {
      "epoch": 1.3481848184818481,
      "grad_norm": 0.015625,
      "learning_rate": 0.02022112211221122,
      "loss": 0.2483,
      "num_input_tokens_seen": 2586240,
      "step": 12255
    },
    {
      "epoch": 1.3487348734873488,
      "grad_norm": 0.01708984375,
      "learning_rate": 0.02022937293729373,
      "loss": 0.2264,
      "num_input_tokens_seen": 2587296,
      "step": 12260
    },
    {
      "epoch": 1.3492849284928492,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.020237623762376235,
      "loss": 0.2349,
      "num_input_tokens_seen": 2588352,
      "step": 12265
    },
    {
      "epoch": 1.3498349834983498,
      "grad_norm": 0.03564453125,
      "learning_rate": 0.020245874587458745,
      "loss": 0.2385,
      "num_input_tokens_seen": 2589408,
      "step": 12270
    },
    {
      "epoch": 1.3503850385038505,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.020254125412541252,
      "loss": 0.2322,
      "num_input_tokens_seen": 2590400,
      "step": 12275
    },
    {
      "epoch": 1.350935093509351,
      "grad_norm": 0.0164794921875,
      "learning_rate": 0.020262376237623763,
      "loss": 0.2314,
      "num_input_tokens_seen": 2591424,
      "step": 12280
    },
    {
      "epoch": 1.3514851485148514,
      "grad_norm": 0.015625,
      "learning_rate": 0.020270627062706267,
      "loss": 0.233,
      "num_input_tokens_seen": 2592480,
      "step": 12285
    },
    {
      "epoch": 1.352035203520352,
      "grad_norm": 0.0033111572265625,
      "learning_rate": 0.020278877887788777,
      "loss": 0.2346,
      "num_input_tokens_seen": 2593536,
      "step": 12290
    },
    {
      "epoch": 1.3525852585258527,
      "grad_norm": 0.0152587890625,
      "learning_rate": 0.020287128712871288,
      "loss": 0.2294,
      "num_input_tokens_seen": 2594560,
      "step": 12295
    },
    {
      "epoch": 1.353135313531353,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.020295379537953795,
      "loss": 0.2312,
      "num_input_tokens_seen": 2595648,
      "step": 12300
    },
    {
      "epoch": 1.3536853685368537,
      "grad_norm": 0.03662109375,
      "learning_rate": 0.020303630363036305,
      "loss": 0.2192,
      "num_input_tokens_seen": 2596672,
      "step": 12305
    },
    {
      "epoch": 1.3542354235423542,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02031188118811881,
      "loss": 0.2425,
      "num_input_tokens_seen": 2597696,
      "step": 12310
    },
    {
      "epoch": 1.3547854785478548,
      "grad_norm": 0.003997802734375,
      "learning_rate": 0.02032013201320132,
      "loss": 0.2261,
      "num_input_tokens_seen": 2598688,
      "step": 12315
    },
    {
      "epoch": 1.3553355335533555,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020328382838283827,
      "loss": 0.2334,
      "num_input_tokens_seen": 2599808,
      "step": 12320
    },
    {
      "epoch": 1.3558855885588559,
      "grad_norm": 0.01300048828125,
      "learning_rate": 0.020336633663366337,
      "loss": 0.23,
      "num_input_tokens_seen": 2600896,
      "step": 12325
    },
    {
      "epoch": 1.3564356435643563,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.02034488448844884,
      "loss": 0.232,
      "num_input_tokens_seen": 2601920,
      "step": 12330
    },
    {
      "epoch": 1.356985698569857,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02035313531353135,
      "loss": 0.2231,
      "num_input_tokens_seen": 2602944,
      "step": 12335
    },
    {
      "epoch": 1.3575357535753576,
      "grad_norm": 0.038330078125,
      "learning_rate": 0.020361386138613862,
      "loss": 0.2422,
      "num_input_tokens_seen": 2604032,
      "step": 12340
    },
    {
      "epoch": 1.358085808580858,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02036963696369637,
      "loss": 0.2255,
      "num_input_tokens_seen": 2605120,
      "step": 12345
    },
    {
      "epoch": 1.3586358635863587,
      "grad_norm": 0.0135498046875,
      "learning_rate": 0.02037788778877888,
      "loss": 0.2253,
      "num_input_tokens_seen": 2606144,
      "step": 12350
    },
    {
      "epoch": 1.359185918591859,
      "grad_norm": 0.039794921875,
      "learning_rate": 0.020386138613861383,
      "loss": 0.2487,
      "num_input_tokens_seen": 2607168,
      "step": 12355
    },
    {
      "epoch": 1.3597359735973598,
      "grad_norm": 0.0030059814453125,
      "learning_rate": 0.020394389438943894,
      "loss": 0.2288,
      "num_input_tokens_seen": 2608224,
      "step": 12360
    },
    {
      "epoch": 1.3602860286028604,
      "grad_norm": 0.0274658203125,
      "learning_rate": 0.0204026402640264,
      "loss": 0.2336,
      "num_input_tokens_seen": 2609248,
      "step": 12365
    },
    {
      "epoch": 1.3608360836083608,
      "grad_norm": 0.0751953125,
      "learning_rate": 0.02041089108910891,
      "loss": 0.2305,
      "num_input_tokens_seen": 2610304,
      "step": 12370
    },
    {
      "epoch": 1.3613861386138613,
      "grad_norm": 0.0283203125,
      "learning_rate": 0.020419141914191415,
      "loss": 0.232,
      "num_input_tokens_seen": 2611296,
      "step": 12375
    },
    {
      "epoch": 1.361936193619362,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.020427392739273926,
      "loss": 0.2302,
      "num_input_tokens_seen": 2612320,
      "step": 12380
    },
    {
      "epoch": 1.3624862486248626,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.020435643564356436,
      "loss": 0.223,
      "num_input_tokens_seen": 2613472,
      "step": 12385
    },
    {
      "epoch": 1.363036303630363,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.020443894389438944,
      "loss": 0.2205,
      "num_input_tokens_seen": 2614560,
      "step": 12390
    },
    {
      "epoch": 1.3635863586358636,
      "grad_norm": 0.01708984375,
      "learning_rate": 0.020452145214521454,
      "loss": 0.2288,
      "num_input_tokens_seen": 2615648,
      "step": 12395
    },
    {
      "epoch": 1.364136413641364,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.020460396039603958,
      "loss": 0.2295,
      "num_input_tokens_seen": 2616736,
      "step": 12400
    },
    {
      "epoch": 1.3646864686468647,
      "grad_norm": 0.02978515625,
      "learning_rate": 0.02046864686468647,
      "loss": 0.2447,
      "num_input_tokens_seen": 2617760,
      "step": 12405
    },
    {
      "epoch": 1.3652365236523654,
      "grad_norm": 0.014404296875,
      "learning_rate": 0.020476897689768975,
      "loss": 0.2269,
      "num_input_tokens_seen": 2618816,
      "step": 12410
    },
    {
      "epoch": 1.3657865786578658,
      "grad_norm": 0.00445556640625,
      "learning_rate": 0.020485148514851486,
      "loss": 0.2354,
      "num_input_tokens_seen": 2619840,
      "step": 12415
    },
    {
      "epoch": 1.3663366336633662,
      "grad_norm": 0.0033416748046875,
      "learning_rate": 0.02049339933993399,
      "loss": 0.2245,
      "num_input_tokens_seen": 2620864,
      "step": 12420
    },
    {
      "epoch": 1.3668866886688669,
      "grad_norm": 0.016357421875,
      "learning_rate": 0.0205016501650165,
      "loss": 0.235,
      "num_input_tokens_seen": 2621888,
      "step": 12425
    },
    {
      "epoch": 1.3674367436743675,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02050990099009901,
      "loss": 0.2276,
      "num_input_tokens_seen": 2622880,
      "step": 12430
    },
    {
      "epoch": 1.367986798679868,
      "grad_norm": 0.01422119140625,
      "learning_rate": 0.020518151815181518,
      "loss": 0.2313,
      "num_input_tokens_seen": 2623904,
      "step": 12435
    },
    {
      "epoch": 1.3685368536853686,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02052640264026403,
      "loss": 0.2318,
      "num_input_tokens_seen": 2624960,
      "step": 12440
    },
    {
      "epoch": 1.369086908690869,
      "grad_norm": 0.014892578125,
      "learning_rate": 0.020534653465346532,
      "loss": 0.2251,
      "num_input_tokens_seen": 2626048,
      "step": 12445
    },
    {
      "epoch": 1.3696369636963697,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.020542904290429043,
      "loss": 0.2355,
      "num_input_tokens_seen": 2627040,
      "step": 12450
    },
    {
      "epoch": 1.3701870187018703,
      "grad_norm": 0.03076171875,
      "learning_rate": 0.02055115511551155,
      "loss": 0.2407,
      "num_input_tokens_seen": 2628096,
      "step": 12455
    },
    {
      "epoch": 1.3707370737073707,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.02055940594059406,
      "loss": 0.2324,
      "num_input_tokens_seen": 2629088,
      "step": 12460
    },
    {
      "epoch": 1.3712871287128712,
      "grad_norm": 0.0264892578125,
      "learning_rate": 0.020567656765676564,
      "loss": 0.2269,
      "num_input_tokens_seen": 2630240,
      "step": 12465
    },
    {
      "epoch": 1.3718371837183718,
      "grad_norm": 0.01324462890625,
      "learning_rate": 0.020575907590759075,
      "loss": 0.2333,
      "num_input_tokens_seen": 2631360,
      "step": 12470
    },
    {
      "epoch": 1.3723872387238725,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.020584158415841585,
      "loss": 0.2307,
      "num_input_tokens_seen": 2632416,
      "step": 12475
    },
    {
      "epoch": 1.372937293729373,
      "grad_norm": 0.02880859375,
      "learning_rate": 0.020592409240924092,
      "loss": 0.2428,
      "num_input_tokens_seen": 2633440,
      "step": 12480
    },
    {
      "epoch": 1.3734873487348735,
      "grad_norm": 0.0147705078125,
      "learning_rate": 0.020600660066006603,
      "loss": 0.2337,
      "num_input_tokens_seen": 2634432,
      "step": 12485
    },
    {
      "epoch": 1.374037403740374,
      "grad_norm": 0.01416015625,
      "learning_rate": 0.020608910891089106,
      "loss": 0.232,
      "num_input_tokens_seen": 2635520,
      "step": 12490
    },
    {
      "epoch": 1.3745874587458746,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.020617161716171617,
      "loss": 0.231,
      "num_input_tokens_seen": 2636576,
      "step": 12495
    },
    {
      "epoch": 1.3751375137513753,
      "grad_norm": 0.0279541015625,
      "learning_rate": 0.020625412541254124,
      "loss": 0.2379,
      "num_input_tokens_seen": 2637568,
      "step": 12500
    },
    {
      "epoch": 1.3756875687568757,
      "grad_norm": 0.0277099609375,
      "learning_rate": 0.020633663366336635,
      "loss": 0.2315,
      "num_input_tokens_seen": 2638656,
      "step": 12505
    },
    {
      "epoch": 1.3762376237623761,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.02064191419141914,
      "loss": 0.2346,
      "num_input_tokens_seen": 2639648,
      "step": 12510
    },
    {
      "epoch": 1.3767876787678768,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.02065016501650165,
      "loss": 0.2347,
      "num_input_tokens_seen": 2640672,
      "step": 12515
    },
    {
      "epoch": 1.3773377337733774,
      "grad_norm": 0.01483154296875,
      "learning_rate": 0.020658415841584156,
      "loss": 0.2324,
      "num_input_tokens_seen": 2641696,
      "step": 12520
    },
    {
      "epoch": 1.3778877887788779,
      "grad_norm": 0.0137939453125,
      "learning_rate": 0.020666666666666667,
      "loss": 0.2314,
      "num_input_tokens_seen": 2642816,
      "step": 12525
    },
    {
      "epoch": 1.3784378437843785,
      "grad_norm": 0.01300048828125,
      "learning_rate": 0.020674917491749177,
      "loss": 0.2293,
      "num_input_tokens_seen": 2643776,
      "step": 12530
    },
    {
      "epoch": 1.378987898789879,
      "grad_norm": 0.0240478515625,
      "learning_rate": 0.02068316831683168,
      "loss": 0.2264,
      "num_input_tokens_seen": 2644800,
      "step": 12535
    },
    {
      "epoch": 1.3795379537953796,
      "grad_norm": 0.01263427734375,
      "learning_rate": 0.02069141914191419,
      "loss": 0.2298,
      "num_input_tokens_seen": 2645856,
      "step": 12540
    },
    {
      "epoch": 1.38008800880088,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.0206996699669967,
      "loss": 0.2315,
      "num_input_tokens_seen": 2646944,
      "step": 12545
    },
    {
      "epoch": 1.3806380638063807,
      "grad_norm": 0.016357421875,
      "learning_rate": 0.02070792079207921,
      "loss": 0.2349,
      "num_input_tokens_seen": 2648000,
      "step": 12550
    },
    {
      "epoch": 1.381188118811881,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.020716171617161713,
      "loss": 0.2266,
      "num_input_tokens_seen": 2649056,
      "step": 12555
    },
    {
      "epoch": 1.3817381738173817,
      "grad_norm": 0.019287109375,
      "learning_rate": 0.020724422442244223,
      "loss": 0.237,
      "num_input_tokens_seen": 2650112,
      "step": 12560
    },
    {
      "epoch": 1.3822882288228824,
      "grad_norm": 0.01708984375,
      "learning_rate": 0.02073267326732673,
      "loss": 0.2288,
      "num_input_tokens_seen": 2651168,
      "step": 12565
    },
    {
      "epoch": 1.3828382838283828,
      "grad_norm": 0.02734375,
      "learning_rate": 0.02074092409240924,
      "loss": 0.2537,
      "num_input_tokens_seen": 2652224,
      "step": 12570
    },
    {
      "epoch": 1.3833883388338835,
      "grad_norm": 0.0245361328125,
      "learning_rate": 0.02074917491749175,
      "loss": 0.2317,
      "num_input_tokens_seen": 2653248,
      "step": 12575
    },
    {
      "epoch": 1.3839383938393839,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.020757425742574255,
      "loss": 0.2341,
      "num_input_tokens_seen": 2654272,
      "step": 12580
    },
    {
      "epoch": 1.3844884488448845,
      "grad_norm": 0.02294921875,
      "learning_rate": 0.020765676567656766,
      "loss": 0.2254,
      "num_input_tokens_seen": 2655328,
      "step": 12585
    },
    {
      "epoch": 1.385038503850385,
      "grad_norm": 0.01556396484375,
      "learning_rate": 0.020773927392739273,
      "loss": 0.2385,
      "num_input_tokens_seen": 2656416,
      "step": 12590
    },
    {
      "epoch": 1.3855885588558856,
      "grad_norm": 0.01422119140625,
      "learning_rate": 0.020782178217821783,
      "loss": 0.2381,
      "num_input_tokens_seen": 2657440,
      "step": 12595
    },
    {
      "epoch": 1.386138613861386,
      "grad_norm": 0.0038299560546875,
      "learning_rate": 0.020790429042904287,
      "loss": 0.2308,
      "num_input_tokens_seen": 2658560,
      "step": 12600
    },
    {
      "epoch": 1.3866886688668867,
      "grad_norm": 0.024169921875,
      "learning_rate": 0.020798679867986797,
      "loss": 0.2286,
      "num_input_tokens_seen": 2659552,
      "step": 12605
    },
    {
      "epoch": 1.3872387238723873,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.020806930693069305,
      "loss": 0.2265,
      "num_input_tokens_seen": 2660672,
      "step": 12610
    },
    {
      "epoch": 1.3877887788778878,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020815181518151815,
      "loss": 0.2338,
      "num_input_tokens_seen": 2661664,
      "step": 12615
    },
    {
      "epoch": 1.3883388338833884,
      "grad_norm": 0.0213623046875,
      "learning_rate": 0.020823432343234326,
      "loss": 0.2296,
      "num_input_tokens_seen": 2662624,
      "step": 12620
    },
    {
      "epoch": 1.3888888888888888,
      "grad_norm": 0.004150390625,
      "learning_rate": 0.02083168316831683,
      "loss": 0.2319,
      "num_input_tokens_seen": 2663680,
      "step": 12625
    },
    {
      "epoch": 1.3894389438943895,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02083993399339934,
      "loss": 0.236,
      "num_input_tokens_seen": 2664704,
      "step": 12630
    },
    {
      "epoch": 1.38998899889989,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.020848184818481847,
      "loss": 0.2319,
      "num_input_tokens_seen": 2665760,
      "step": 12635
    },
    {
      "epoch": 1.3905390539053906,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.020856435643564358,
      "loss": 0.2316,
      "num_input_tokens_seen": 2666880,
      "step": 12640
    },
    {
      "epoch": 1.391089108910891,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.02086468646864686,
      "loss": 0.2287,
      "num_input_tokens_seen": 2667904,
      "step": 12645
    },
    {
      "epoch": 1.3916391639163916,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.020872937293729372,
      "loss": 0.2267,
      "num_input_tokens_seen": 2669024,
      "step": 12650
    },
    {
      "epoch": 1.3921892189218923,
      "grad_norm": 0.02392578125,
      "learning_rate": 0.02088118811881188,
      "loss": 0.2357,
      "num_input_tokens_seen": 2670112,
      "step": 12655
    },
    {
      "epoch": 1.3927392739273927,
      "grad_norm": 0.003082275390625,
      "learning_rate": 0.02088943894389439,
      "loss": 0.2356,
      "num_input_tokens_seen": 2671136,
      "step": 12660
    },
    {
      "epoch": 1.3932893289328934,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.0208976897689769,
      "loss": 0.2273,
      "num_input_tokens_seen": 2672128,
      "step": 12665
    },
    {
      "epoch": 1.3938393839383938,
      "grad_norm": 0.003082275390625,
      "learning_rate": 0.020905940594059404,
      "loss": 0.2366,
      "num_input_tokens_seen": 2673120,
      "step": 12670
    },
    {
      "epoch": 1.3943894389438944,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.020914191419141914,
      "loss": 0.236,
      "num_input_tokens_seen": 2674208,
      "step": 12675
    },
    {
      "epoch": 1.3949394939493949,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.02092244224422442,
      "loss": 0.2326,
      "num_input_tokens_seen": 2675264,
      "step": 12680
    },
    {
      "epoch": 1.3954895489548955,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.020930693069306932,
      "loss": 0.2331,
      "num_input_tokens_seen": 2676256,
      "step": 12685
    },
    {
      "epoch": 1.396039603960396,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.020938943894389436,
      "loss": 0.2259,
      "num_input_tokens_seen": 2677312,
      "step": 12690
    },
    {
      "epoch": 1.3965896589658966,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.020947194719471946,
      "loss": 0.2348,
      "num_input_tokens_seen": 2678368,
      "step": 12695
    },
    {
      "epoch": 1.3971397139713972,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.020955445544554453,
      "loss": 0.2325,
      "num_input_tokens_seen": 2679456,
      "step": 12700
    },
    {
      "epoch": 1.3976897689768977,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.020963696369636964,
      "loss": 0.2313,
      "num_input_tokens_seen": 2680544,
      "step": 12705
    },
    {
      "epoch": 1.3982398239823983,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.020971947194719474,
      "loss": 0.2313,
      "num_input_tokens_seen": 2681568,
      "step": 12710
    },
    {
      "epoch": 1.3987898789878987,
      "grad_norm": 0.021728515625,
      "learning_rate": 0.020980198019801978,
      "loss": 0.2313,
      "num_input_tokens_seen": 2682656,
      "step": 12715
    },
    {
      "epoch": 1.3993399339933994,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.02098844884488449,
      "loss": 0.2303,
      "num_input_tokens_seen": 2683680,
      "step": 12720
    },
    {
      "epoch": 1.3998899889988998,
      "grad_norm": 0.003662109375,
      "learning_rate": 0.020996699669966996,
      "loss": 0.2315,
      "num_input_tokens_seen": 2684704,
      "step": 12725
    },
    {
      "epoch": 1.4004400440044005,
      "grad_norm": 0.003692626953125,
      "learning_rate": 0.021004950495049506,
      "loss": 0.2293,
      "num_input_tokens_seen": 2685792,
      "step": 12730
    },
    {
      "epoch": 1.400990099009901,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.02101320132013201,
      "loss": 0.2304,
      "num_input_tokens_seen": 2686848,
      "step": 12735
    },
    {
      "epoch": 1.4015401540154016,
      "grad_norm": 0.01171875,
      "learning_rate": 0.02102145214521452,
      "loss": 0.2314,
      "num_input_tokens_seen": 2687904,
      "step": 12740
    },
    {
      "epoch": 1.4020902090209022,
      "grad_norm": 0.003265380859375,
      "learning_rate": 0.021029702970297028,
      "loss": 0.2301,
      "num_input_tokens_seen": 2688992,
      "step": 12745
    },
    {
      "epoch": 1.4026402640264026,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.021037953795379538,
      "loss": 0.2313,
      "num_input_tokens_seen": 2690080,
      "step": 12750
    },
    {
      "epoch": 1.403190319031903,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.021046204620462045,
      "loss": 0.2257,
      "num_input_tokens_seen": 2691200,
      "step": 12755
    },
    {
      "epoch": 1.4037403740374037,
      "grad_norm": 0.01275634765625,
      "learning_rate": 0.021054455445544552,
      "loss": 0.2513,
      "num_input_tokens_seen": 2692224,
      "step": 12760
    },
    {
      "epoch": 1.4042904290429044,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.021062706270627063,
      "loss": 0.2331,
      "num_input_tokens_seen": 2693280,
      "step": 12765
    },
    {
      "epoch": 1.4048404840484048,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.02107095709570957,
      "loss": 0.2285,
      "num_input_tokens_seen": 2694368,
      "step": 12770
    },
    {
      "epoch": 1.4053905390539054,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.02107920792079208,
      "loss": 0.2347,
      "num_input_tokens_seen": 2695392,
      "step": 12775
    },
    {
      "epoch": 1.4059405940594059,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.021087458745874584,
      "loss": 0.2303,
      "num_input_tokens_seen": 2696416,
      "step": 12780
    },
    {
      "epoch": 1.4064906490649065,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.021095709570957095,
      "loss": 0.2304,
      "num_input_tokens_seen": 2697472,
      "step": 12785
    },
    {
      "epoch": 1.4070407040704072,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.021103960396039602,
      "loss": 0.2296,
      "num_input_tokens_seen": 2698432,
      "step": 12790
    },
    {
      "epoch": 1.4075907590759076,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.021112211221122112,
      "loss": 0.2245,
      "num_input_tokens_seen": 2699456,
      "step": 12795
    },
    {
      "epoch": 1.408140814081408,
      "grad_norm": 0.00341796875,
      "learning_rate": 0.02112046204620462,
      "loss": 0.2319,
      "num_input_tokens_seen": 2700512,
      "step": 12800
    },
    {
      "epoch": 1.4086908690869087,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.021128712871287127,
      "loss": 0.231,
      "num_input_tokens_seen": 2701536,
      "step": 12805
    },
    {
      "epoch": 1.4092409240924093,
      "grad_norm": 0.023193359375,
      "learning_rate": 0.021136963696369637,
      "loss": 0.2352,
      "num_input_tokens_seen": 2702624,
      "step": 12810
    },
    {
      "epoch": 1.4097909790979097,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.021145214521452144,
      "loss": 0.2403,
      "num_input_tokens_seen": 2703712,
      "step": 12815
    },
    {
      "epoch": 1.4103410341034104,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.021153465346534655,
      "loss": 0.2246,
      "num_input_tokens_seen": 2704704,
      "step": 12820
    },
    {
      "epoch": 1.4108910891089108,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.02116171617161716,
      "loss": 0.2281,
      "num_input_tokens_seen": 2705760,
      "step": 12825
    },
    {
      "epoch": 1.4114411441144115,
      "grad_norm": 0.0220947265625,
      "learning_rate": 0.02116996699669967,
      "loss": 0.2407,
      "num_input_tokens_seen": 2706752,
      "step": 12830
    },
    {
      "epoch": 1.411991199119912,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.021178217821782176,
      "loss": 0.2282,
      "num_input_tokens_seen": 2707808,
      "step": 12835
    },
    {
      "epoch": 1.4125412541254125,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.021186468646864687,
      "loss": 0.2406,
      "num_input_tokens_seen": 2708800,
      "step": 12840
    },
    {
      "epoch": 1.413091309130913,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.021194719471947194,
      "loss": 0.2294,
      "num_input_tokens_seen": 2709856,
      "step": 12845
    },
    {
      "epoch": 1.4136413641364136,
      "grad_norm": 0.004058837890625,
      "learning_rate": 0.0212029702970297,
      "loss": 0.2299,
      "num_input_tokens_seen": 2710912,
      "step": 12850
    },
    {
      "epoch": 1.4141914191419143,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.02121122112211221,
      "loss": 0.2336,
      "num_input_tokens_seen": 2711936,
      "step": 12855
    },
    {
      "epoch": 1.4147414741474147,
      "grad_norm": 0.001953125,
      "learning_rate": 0.02121947194719472,
      "loss": 0.2315,
      "num_input_tokens_seen": 2713056,
      "step": 12860
    },
    {
      "epoch": 1.4152915291529153,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.02122772277227723,
      "loss": 0.2325,
      "num_input_tokens_seen": 2714208,
      "step": 12865
    },
    {
      "epoch": 1.4158415841584158,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.021235973597359733,
      "loss": 0.2325,
      "num_input_tokens_seen": 2715296,
      "step": 12870
    },
    {
      "epoch": 1.4163916391639164,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.021244224422442243,
      "loss": 0.2327,
      "num_input_tokens_seen": 2716384,
      "step": 12875
    },
    {
      "epoch": 1.416941694169417,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02125247524752475,
      "loss": 0.2327,
      "num_input_tokens_seen": 2717408,
      "step": 12880
    },
    {
      "epoch": 1.4174917491749175,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02126072607260726,
      "loss": 0.2305,
      "num_input_tokens_seen": 2718464,
      "step": 12885
    },
    {
      "epoch": 1.418041804180418,
      "grad_norm": 0.0031280517578125,
      "learning_rate": 0.021268976897689768,
      "loss": 0.2314,
      "num_input_tokens_seen": 2719552,
      "step": 12890
    },
    {
      "epoch": 1.4185918591859186,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.021277227722772275,
      "loss": 0.2324,
      "num_input_tokens_seen": 2720608,
      "step": 12895
    },
    {
      "epoch": 1.4191419141914192,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.021285478547854786,
      "loss": 0.2314,
      "num_input_tokens_seen": 2721664,
      "step": 12900
    },
    {
      "epoch": 1.4196919691969196,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.021293729372937293,
      "loss": 0.2295,
      "num_input_tokens_seen": 2722688,
      "step": 12905
    },
    {
      "epoch": 1.4202420242024203,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.021301980198019804,
      "loss": 0.2296,
      "num_input_tokens_seen": 2723808,
      "step": 12910
    },
    {
      "epoch": 1.4207920792079207,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.021310231023102307,
      "loss": 0.2347,
      "num_input_tokens_seen": 2724896,
      "step": 12915
    },
    {
      "epoch": 1.4213421342134214,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.021318481848184818,
      "loss": 0.2325,
      "num_input_tokens_seen": 2725920,
      "step": 12920
    },
    {
      "epoch": 1.421892189218922,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.021326732673267325,
      "loss": 0.2335,
      "num_input_tokens_seen": 2726976,
      "step": 12925
    },
    {
      "epoch": 1.4224422442244224,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.021334983498349835,
      "loss": 0.2304,
      "num_input_tokens_seen": 2728032,
      "step": 12930
    },
    {
      "epoch": 1.4229922992299229,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.021343234323432343,
      "loss": 0.2324,
      "num_input_tokens_seen": 2729056,
      "step": 12935
    },
    {
      "epoch": 1.4235423542354235,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02135148514851485,
      "loss": 0.2285,
      "num_input_tokens_seen": 2730112,
      "step": 12940
    },
    {
      "epoch": 1.4240924092409242,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.02135973597359736,
      "loss": 0.239,
      "num_input_tokens_seen": 2731200,
      "step": 12945
    },
    {
      "epoch": 1.4246424642464246,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.021367986798679867,
      "loss": 0.2328,
      "num_input_tokens_seen": 2732224,
      "step": 12950
    },
    {
      "epoch": 1.4251925192519252,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.021376237623762378,
      "loss": 0.2338,
      "num_input_tokens_seen": 2733248,
      "step": 12955
    },
    {
      "epoch": 1.4257425742574257,
      "grad_norm": 0.00927734375,
      "learning_rate": 0.02138448844884488,
      "loss": 0.2317,
      "num_input_tokens_seen": 2734336,
      "step": 12960
    },
    {
      "epoch": 1.4262926292629263,
      "grad_norm": 0.00286865234375,
      "learning_rate": 0.021392739273927392,
      "loss": 0.2264,
      "num_input_tokens_seen": 2735424,
      "step": 12965
    },
    {
      "epoch": 1.426842684268427,
      "grad_norm": 0.0194091796875,
      "learning_rate": 0.0214009900990099,
      "loss": 0.2275,
      "num_input_tokens_seen": 2736480,
      "step": 12970
    },
    {
      "epoch": 1.4273927392739274,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.02140924092409241,
      "loss": 0.2286,
      "num_input_tokens_seen": 2737472,
      "step": 12975
    },
    {
      "epoch": 1.4279427942794278,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.021417491749174917,
      "loss": 0.2307,
      "num_input_tokens_seen": 2738592,
      "step": 12980
    },
    {
      "epoch": 1.4284928492849285,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.021425742574257424,
      "loss": 0.2225,
      "num_input_tokens_seen": 2739616,
      "step": 12985
    },
    {
      "epoch": 1.4290429042904291,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.021433993399339935,
      "loss": 0.2318,
      "num_input_tokens_seen": 2740608,
      "step": 12990
    },
    {
      "epoch": 1.4295929592959296,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.02144224422442244,
      "loss": 0.237,
      "num_input_tokens_seen": 2741728,
      "step": 12995
    },
    {
      "epoch": 1.4301430143014302,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.021450495049504952,
      "loss": 0.2307,
      "num_input_tokens_seen": 2742752,
      "step": 13000
    },
    {
      "epoch": 1.4306930693069306,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.021458745874587456,
      "loss": 0.2296,
      "num_input_tokens_seen": 2743808,
      "step": 13005
    },
    {
      "epoch": 1.4312431243124313,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.021466996699669966,
      "loss": 0.2337,
      "num_input_tokens_seen": 2744864,
      "step": 13010
    },
    {
      "epoch": 1.431793179317932,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.021475247524752474,
      "loss": 0.23,
      "num_input_tokens_seen": 2745856,
      "step": 13015
    },
    {
      "epoch": 1.4323432343234324,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.021483498349834984,
      "loss": 0.2295,
      "num_input_tokens_seen": 2746848,
      "step": 13020
    },
    {
      "epoch": 1.4328932893289328,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02149174917491749,
      "loss": 0.2294,
      "num_input_tokens_seen": 2747904,
      "step": 13025
    },
    {
      "epoch": 1.4334433443344334,
      "grad_norm": 0.0194091796875,
      "learning_rate": 0.0215,
      "loss": 0.2335,
      "num_input_tokens_seen": 2748896,
      "step": 13030
    },
    {
      "epoch": 1.433993399339934,
      "grad_norm": 0.0189208984375,
      "learning_rate": 0.021508250825082505,
      "loss": 0.2299,
      "num_input_tokens_seen": 2749920,
      "step": 13035
    },
    {
      "epoch": 1.4345434543454345,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.021516501650165016,
      "loss": 0.2325,
      "num_input_tokens_seen": 2750944,
      "step": 13040
    },
    {
      "epoch": 1.4350935093509352,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.021524752475247527,
      "loss": 0.2317,
      "num_input_tokens_seen": 2752000,
      "step": 13045
    },
    {
      "epoch": 1.4356435643564356,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.02153300330033003,
      "loss": 0.2358,
      "num_input_tokens_seen": 2753088,
      "step": 13050
    },
    {
      "epoch": 1.4361936193619362,
      "grad_norm": 0.0186767578125,
      "learning_rate": 0.02154125412541254,
      "loss": 0.2283,
      "num_input_tokens_seen": 2754176,
      "step": 13055
    },
    {
      "epoch": 1.4367436743674367,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.021549504950495048,
      "loss": 0.2314,
      "num_input_tokens_seen": 2755264,
      "step": 13060
    },
    {
      "epoch": 1.4372937293729373,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.02155775577557756,
      "loss": 0.2305,
      "num_input_tokens_seen": 2756320,
      "step": 13065
    },
    {
      "epoch": 1.4378437843784377,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.021566006600660066,
      "loss": 0.2295,
      "num_input_tokens_seen": 2757440,
      "step": 13070
    },
    {
      "epoch": 1.4383938393839384,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.021574257425742573,
      "loss": 0.2328,
      "num_input_tokens_seen": 2758496,
      "step": 13075
    },
    {
      "epoch": 1.438943894389439,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.02158250825082508,
      "loss": 0.2318,
      "num_input_tokens_seen": 2759552,
      "step": 13080
    },
    {
      "epoch": 1.4394939493949395,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.02159075907590759,
      "loss": 0.2308,
      "num_input_tokens_seen": 2760672,
      "step": 13085
    },
    {
      "epoch": 1.4400440044004401,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0215990099009901,
      "loss": 0.2245,
      "num_input_tokens_seen": 2761664,
      "step": 13090
    },
    {
      "epoch": 1.4405940594059405,
      "grad_norm": 0.003204345703125,
      "learning_rate": 0.021607260726072604,
      "loss": 0.234,
      "num_input_tokens_seen": 2762688,
      "step": 13095
    },
    {
      "epoch": 1.4411441144114412,
      "grad_norm": 0.003204345703125,
      "learning_rate": 0.021615511551155115,
      "loss": 0.2246,
      "num_input_tokens_seen": 2763680,
      "step": 13100
    },
    {
      "epoch": 1.4416941694169416,
      "grad_norm": 0.0218505859375,
      "learning_rate": 0.021623762376237622,
      "loss": 0.234,
      "num_input_tokens_seen": 2764672,
      "step": 13105
    },
    {
      "epoch": 1.4422442244224423,
      "grad_norm": 0.019287109375,
      "learning_rate": 0.021632013201320133,
      "loss": 0.2323,
      "num_input_tokens_seen": 2765792,
      "step": 13110
    },
    {
      "epoch": 1.4427942794279427,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.02164026402640264,
      "loss": 0.2272,
      "num_input_tokens_seen": 2766816,
      "step": 13115
    },
    {
      "epoch": 1.4433443344334433,
      "grad_norm": 0.0191650390625,
      "learning_rate": 0.021648514851485147,
      "loss": 0.2333,
      "num_input_tokens_seen": 2767872,
      "step": 13120
    },
    {
      "epoch": 1.443894389438944,
      "grad_norm": 0.01263427734375,
      "learning_rate": 0.021656765676567654,
      "loss": 0.2346,
      "num_input_tokens_seen": 2768928,
      "step": 13125
    },
    {
      "epoch": 1.4444444444444444,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.021665016501650165,
      "loss": 0.2212,
      "num_input_tokens_seen": 2769952,
      "step": 13130
    },
    {
      "epoch": 1.444994499449945,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.021673267326732675,
      "loss": 0.2369,
      "num_input_tokens_seen": 2770976,
      "step": 13135
    },
    {
      "epoch": 1.4455445544554455,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.02168151815181518,
      "loss": 0.2363,
      "num_input_tokens_seen": 2772032,
      "step": 13140
    },
    {
      "epoch": 1.4460946094609461,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.02168976897689769,
      "loss": 0.2314,
      "num_input_tokens_seen": 2773088,
      "step": 13145
    },
    {
      "epoch": 1.4466446644664466,
      "grad_norm": 0.0028533935546875,
      "learning_rate": 0.021698019801980196,
      "loss": 0.2388,
      "num_input_tokens_seen": 2774176,
      "step": 13150
    },
    {
      "epoch": 1.4471947194719472,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.021706270627062707,
      "loss": 0.234,
      "num_input_tokens_seen": 2775168,
      "step": 13155
    },
    {
      "epoch": 1.4477447744774476,
      "grad_norm": 0.0205078125,
      "learning_rate": 0.021714521452145214,
      "loss": 0.2306,
      "num_input_tokens_seen": 2776192,
      "step": 13160
    },
    {
      "epoch": 1.4482948294829483,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.02172277227722772,
      "loss": 0.2325,
      "num_input_tokens_seen": 2777184,
      "step": 13165
    },
    {
      "epoch": 1.448844884488449,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.02173102310231023,
      "loss": 0.2336,
      "num_input_tokens_seen": 2778240,
      "step": 13170
    },
    {
      "epoch": 1.4493949394939494,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.02173927392739274,
      "loss": 0.2315,
      "num_input_tokens_seen": 2779296,
      "step": 13175
    },
    {
      "epoch": 1.44994499449945,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.02174752475247525,
      "loss": 0.2327,
      "num_input_tokens_seen": 2780320,
      "step": 13180
    },
    {
      "epoch": 1.4504950495049505,
      "grad_norm": 0.0036163330078125,
      "learning_rate": 0.021755775577557753,
      "loss": 0.2347,
      "num_input_tokens_seen": 2781440,
      "step": 13185
    },
    {
      "epoch": 1.451045104510451,
      "grad_norm": 0.009765625,
      "learning_rate": 0.021764026402640264,
      "loss": 0.2325,
      "num_input_tokens_seen": 2782464,
      "step": 13190
    },
    {
      "epoch": 1.4515951595159515,
      "grad_norm": 0.0031280517578125,
      "learning_rate": 0.02177227722772277,
      "loss": 0.2342,
      "num_input_tokens_seen": 2783488,
      "step": 13195
    },
    {
      "epoch": 1.4521452145214522,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.02178052805280528,
      "loss": 0.2347,
      "num_input_tokens_seen": 2784576,
      "step": 13200
    },
    {
      "epoch": 1.4526952695269526,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02178877887788779,
      "loss": 0.2297,
      "num_input_tokens_seen": 2785632,
      "step": 13205
    },
    {
      "epoch": 1.4532453245324533,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.021797029702970296,
      "loss": 0.2189,
      "num_input_tokens_seen": 2786624,
      "step": 13210
    },
    {
      "epoch": 1.453795379537954,
      "grad_norm": 0.0194091796875,
      "learning_rate": 0.021805280528052803,
      "loss": 0.2213,
      "num_input_tokens_seen": 2787712,
      "step": 13215
    },
    {
      "epoch": 1.4543454345434543,
      "grad_norm": 0.0145263671875,
      "learning_rate": 0.021813531353135313,
      "loss": 0.256,
      "num_input_tokens_seen": 2788768,
      "step": 13220
    },
    {
      "epoch": 1.4548954895489548,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.021821782178217824,
      "loss": 0.2298,
      "num_input_tokens_seen": 2789792,
      "step": 13225
    },
    {
      "epoch": 1.4554455445544554,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.021830033003300327,
      "loss": 0.2303,
      "num_input_tokens_seen": 2790880,
      "step": 13230
    },
    {
      "epoch": 1.455995599559956,
      "grad_norm": 0.0029754638671875,
      "learning_rate": 0.021838283828382838,
      "loss": 0.2428,
      "num_input_tokens_seen": 2791872,
      "step": 13235
    },
    {
      "epoch": 1.4565456545654565,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.021846534653465345,
      "loss": 0.2304,
      "num_input_tokens_seen": 2792864,
      "step": 13240
    },
    {
      "epoch": 1.4570957095709571,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.021854785478547856,
      "loss": 0.2126,
      "num_input_tokens_seen": 2793920,
      "step": 13245
    },
    {
      "epoch": 1.4576457645764576,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.021863036303630363,
      "loss": 0.2414,
      "num_input_tokens_seen": 2795008,
      "step": 13250
    },
    {
      "epoch": 1.4581958195819582,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.02187128712871287,
      "loss": 0.2249,
      "num_input_tokens_seen": 2796064,
      "step": 13255
    },
    {
      "epoch": 1.4587458745874589,
      "grad_norm": 0.00897216796875,
      "learning_rate": 0.021879537953795377,
      "loss": 0.2404,
      "num_input_tokens_seen": 2797120,
      "step": 13260
    },
    {
      "epoch": 1.4592959295929593,
      "grad_norm": 0.01263427734375,
      "learning_rate": 0.021887788778877888,
      "loss": 0.2296,
      "num_input_tokens_seen": 2798176,
      "step": 13265
    },
    {
      "epoch": 1.4598459845984597,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.021896039603960395,
      "loss": 0.2345,
      "num_input_tokens_seen": 2799168,
      "step": 13270
    },
    {
      "epoch": 1.4603960396039604,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.021904290429042902,
      "loss": 0.2237,
      "num_input_tokens_seen": 2800192,
      "step": 13275
    },
    {
      "epoch": 1.460946094609461,
      "grad_norm": 0.022216796875,
      "learning_rate": 0.021912541254125412,
      "loss": 0.2301,
      "num_input_tokens_seen": 2801280,
      "step": 13280
    },
    {
      "epoch": 1.4614961496149614,
      "grad_norm": 0.004180908203125,
      "learning_rate": 0.02192079207920792,
      "loss": 0.2335,
      "num_input_tokens_seen": 2802336,
      "step": 13285
    },
    {
      "epoch": 1.462046204620462,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.02192904290429043,
      "loss": 0.2324,
      "num_input_tokens_seen": 2803328,
      "step": 13290
    },
    {
      "epoch": 1.4625962596259625,
      "grad_norm": 0.01202392578125,
      "learning_rate": 0.021937293729372937,
      "loss": 0.232,
      "num_input_tokens_seen": 2804352,
      "step": 13295
    },
    {
      "epoch": 1.4631463146314632,
      "grad_norm": 0.01239013671875,
      "learning_rate": 0.021945544554455444,
      "loss": 0.235,
      "num_input_tokens_seen": 2805472,
      "step": 13300
    },
    {
      "epoch": 1.4636963696369638,
      "grad_norm": 0.0028839111328125,
      "learning_rate": 0.02195379537953795,
      "loss": 0.2318,
      "num_input_tokens_seen": 2806464,
      "step": 13305
    },
    {
      "epoch": 1.4642464246424642,
      "grad_norm": 0.0230712890625,
      "learning_rate": 0.021962046204620462,
      "loss": 0.2339,
      "num_input_tokens_seen": 2807552,
      "step": 13310
    },
    {
      "epoch": 1.4647964796479647,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.02197029702970297,
      "loss": 0.2327,
      "num_input_tokens_seen": 2808672,
      "step": 13315
    },
    {
      "epoch": 1.4653465346534653,
      "grad_norm": 0.01190185546875,
      "learning_rate": 0.021978547854785476,
      "loss": 0.2283,
      "num_input_tokens_seen": 2809696,
      "step": 13320
    },
    {
      "epoch": 1.465896589658966,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.021986798679867987,
      "loss": 0.2293,
      "num_input_tokens_seen": 2810720,
      "step": 13325
    },
    {
      "epoch": 1.4664466446644664,
      "grad_norm": 0.0030059814453125,
      "learning_rate": 0.021995049504950494,
      "loss": 0.2314,
      "num_input_tokens_seen": 2811840,
      "step": 13330
    },
    {
      "epoch": 1.466996699669967,
      "grad_norm": 0.0157470703125,
      "learning_rate": 0.022003300330033004,
      "loss": 0.2301,
      "num_input_tokens_seen": 2812864,
      "step": 13335
    },
    {
      "epoch": 1.4675467546754675,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02201155115511551,
      "loss": 0.2316,
      "num_input_tokens_seen": 2813920,
      "step": 13340
    },
    {
      "epoch": 1.4680968096809681,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.02201980198019802,
      "loss": 0.2401,
      "num_input_tokens_seen": 2814912,
      "step": 13345
    },
    {
      "epoch": 1.4686468646864688,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.022028052805280526,
      "loss": 0.2303,
      "num_input_tokens_seen": 2816064,
      "step": 13350
    },
    {
      "epoch": 1.4691969196919692,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.022036303630363036,
      "loss": 0.2293,
      "num_input_tokens_seen": 2817184,
      "step": 13355
    },
    {
      "epoch": 1.4697469746974696,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.022044554455445543,
      "loss": 0.2294,
      "num_input_tokens_seen": 2818240,
      "step": 13360
    },
    {
      "epoch": 1.4702970297029703,
      "grad_norm": 0.0233154296875,
      "learning_rate": 0.02205280528052805,
      "loss": 0.2255,
      "num_input_tokens_seen": 2819264,
      "step": 13365
    },
    {
      "epoch": 1.470847084708471,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.02206105610561056,
      "loss": 0.2363,
      "num_input_tokens_seen": 2820352,
      "step": 13370
    },
    {
      "epoch": 1.4713971397139713,
      "grad_norm": 0.0035552978515625,
      "learning_rate": 0.022069306930693068,
      "loss": 0.2397,
      "num_input_tokens_seen": 2821344,
      "step": 13375
    },
    {
      "epoch": 1.471947194719472,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.02207755775577558,
      "loss": 0.2339,
      "num_input_tokens_seen": 2822368,
      "step": 13380
    },
    {
      "epoch": 1.4724972497249724,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.022085808580858086,
      "loss": 0.2285,
      "num_input_tokens_seen": 2823424,
      "step": 13385
    },
    {
      "epoch": 1.473047304730473,
      "grad_norm": 0.003326416015625,
      "learning_rate": 0.022094059405940593,
      "loss": 0.2325,
      "num_input_tokens_seen": 2824512,
      "step": 13390
    },
    {
      "epoch": 1.4735973597359737,
      "grad_norm": 0.0031890869140625,
      "learning_rate": 0.0221023102310231,
      "loss": 0.2335,
      "num_input_tokens_seen": 2825568,
      "step": 13395
    },
    {
      "epoch": 1.4741474147414741,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.02211056105610561,
      "loss": 0.2294,
      "num_input_tokens_seen": 2826528,
      "step": 13400
    },
    {
      "epoch": 1.4746974697469746,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.022118811881188118,
      "loss": 0.2284,
      "num_input_tokens_seen": 2827648,
      "step": 13405
    },
    {
      "epoch": 1.4752475247524752,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.022127062706270625,
      "loss": 0.2327,
      "num_input_tokens_seen": 2828672,
      "step": 13410
    },
    {
      "epoch": 1.4757975797579759,
      "grad_norm": 0.0186767578125,
      "learning_rate": 0.022135313531353135,
      "loss": 0.241,
      "num_input_tokens_seen": 2829760,
      "step": 13415
    },
    {
      "epoch": 1.4763476347634763,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.022143564356435642,
      "loss": 0.2317,
      "num_input_tokens_seen": 2830784,
      "step": 13420
    },
    {
      "epoch": 1.476897689768977,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.022151815181518153,
      "loss": 0.2348,
      "num_input_tokens_seen": 2831776,
      "step": 13425
    },
    {
      "epoch": 1.4774477447744774,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02216006600660066,
      "loss": 0.2325,
      "num_input_tokens_seen": 2832800,
      "step": 13430
    },
    {
      "epoch": 1.477997799779978,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.022168316831683167,
      "loss": 0.2305,
      "num_input_tokens_seen": 2833856,
      "step": 13435
    },
    {
      "epoch": 1.4785478547854787,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.022176567656765674,
      "loss": 0.2327,
      "num_input_tokens_seen": 2834848,
      "step": 13440
    },
    {
      "epoch": 1.479097909790979,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.022184818481848185,
      "loss": 0.2294,
      "num_input_tokens_seen": 2835968,
      "step": 13445
    },
    {
      "epoch": 1.4796479647964795,
      "grad_norm": 0.018310546875,
      "learning_rate": 0.022193069306930692,
      "loss": 0.2296,
      "num_input_tokens_seen": 2837024,
      "step": 13450
    },
    {
      "epoch": 1.4801980198019802,
      "grad_norm": 0.01708984375,
      "learning_rate": 0.0222013201320132,
      "loss": 0.2285,
      "num_input_tokens_seen": 2838144,
      "step": 13455
    },
    {
      "epoch": 1.4807480748074808,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.02220957095709571,
      "loss": 0.2294,
      "num_input_tokens_seen": 2839200,
      "step": 13460
    },
    {
      "epoch": 1.4812981298129813,
      "grad_norm": 0.0174560546875,
      "learning_rate": 0.022217821782178217,
      "loss": 0.2275,
      "num_input_tokens_seen": 2840256,
      "step": 13465
    },
    {
      "epoch": 1.481848184818482,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.022226072607260727,
      "loss": 0.2287,
      "num_input_tokens_seen": 2841376,
      "step": 13470
    },
    {
      "epoch": 1.4823982398239823,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.022234323432343234,
      "loss": 0.2298,
      "num_input_tokens_seen": 2842400,
      "step": 13475
    },
    {
      "epoch": 1.482948294829483,
      "grad_norm": 0.017333984375,
      "learning_rate": 0.02224257425742574,
      "loss": 0.2226,
      "num_input_tokens_seen": 2843392,
      "step": 13480
    },
    {
      "epoch": 1.4834983498349836,
      "grad_norm": 0.00433349609375,
      "learning_rate": 0.02225082508250825,
      "loss": 0.2357,
      "num_input_tokens_seen": 2844416,
      "step": 13485
    },
    {
      "epoch": 1.484048404840484,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.02225907590759076,
      "loss": 0.2328,
      "num_input_tokens_seen": 2845504,
      "step": 13490
    },
    {
      "epoch": 1.4845984598459845,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.022267326732673266,
      "loss": 0.2313,
      "num_input_tokens_seen": 2846560,
      "step": 13495
    },
    {
      "epoch": 1.4851485148514851,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.022275577557755773,
      "loss": 0.2312,
      "num_input_tokens_seen": 2847648,
      "step": 13500
    },
    {
      "epoch": 1.4856985698569858,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.022283828382838284,
      "loss": 0.238,
      "num_input_tokens_seen": 2848768,
      "step": 13505
    },
    {
      "epoch": 1.4862486248624862,
      "grad_norm": 0.01708984375,
      "learning_rate": 0.02229207920792079,
      "loss": 0.2247,
      "num_input_tokens_seen": 2849792,
      "step": 13510
    },
    {
      "epoch": 1.4867986798679869,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.0223003300330033,
      "loss": 0.2258,
      "num_input_tokens_seen": 2850880,
      "step": 13515
    },
    {
      "epoch": 1.4873487348734873,
      "grad_norm": 0.02001953125,
      "learning_rate": 0.02230858085808581,
      "loss": 0.2383,
      "num_input_tokens_seen": 2851936,
      "step": 13520
    },
    {
      "epoch": 1.487898789878988,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.022316831683168316,
      "loss": 0.2378,
      "num_input_tokens_seen": 2852928,
      "step": 13525
    },
    {
      "epoch": 1.4884488448844886,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.022325082508250823,
      "loss": 0.2351,
      "num_input_tokens_seen": 2853984,
      "step": 13530
    },
    {
      "epoch": 1.488998899889989,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.022333333333333334,
      "loss": 0.2331,
      "num_input_tokens_seen": 2855040,
      "step": 13535
    },
    {
      "epoch": 1.4895489548954894,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.02234158415841584,
      "loss": 0.2311,
      "num_input_tokens_seen": 2856032,
      "step": 13540
    },
    {
      "epoch": 1.49009900990099,
      "grad_norm": 0.0186767578125,
      "learning_rate": 0.022349834983498348,
      "loss": 0.2342,
      "num_input_tokens_seen": 2857088,
      "step": 13545
    },
    {
      "epoch": 1.4906490649064907,
      "grad_norm": 0.018310546875,
      "learning_rate": 0.022358085808580855,
      "loss": 0.2304,
      "num_input_tokens_seen": 2858176,
      "step": 13550
    },
    {
      "epoch": 1.4911991199119912,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.022366336633663365,
      "loss": 0.2335,
      "num_input_tokens_seen": 2859232,
      "step": 13555
    },
    {
      "epoch": 1.4917491749174918,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.022374587458745876,
      "loss": 0.2305,
      "num_input_tokens_seen": 2860288,
      "step": 13560
    },
    {
      "epoch": 1.4922992299229922,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.022382838283828383,
      "loss": 0.2285,
      "num_input_tokens_seen": 2861312,
      "step": 13565
    },
    {
      "epoch": 1.492849284928493,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.02239108910891089,
      "loss": 0.2348,
      "num_input_tokens_seen": 2862368,
      "step": 13570
    },
    {
      "epoch": 1.4933993399339933,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.022399339933993397,
      "loss": 0.2306,
      "num_input_tokens_seen": 2863392,
      "step": 13575
    },
    {
      "epoch": 1.493949394939494,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.022407590759075908,
      "loss": 0.233,
      "num_input_tokens_seen": 2864480,
      "step": 13580
    },
    {
      "epoch": 1.4944994499449944,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.022415841584158415,
      "loss": 0.2325,
      "num_input_tokens_seen": 2865504,
      "step": 13585
    },
    {
      "epoch": 1.495049504950495,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.022424092409240922,
      "loss": 0.231,
      "num_input_tokens_seen": 2866560,
      "step": 13590
    },
    {
      "epoch": 1.4955995599559957,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.02243234323432343,
      "loss": 0.2299,
      "num_input_tokens_seen": 2867680,
      "step": 13595
    },
    {
      "epoch": 1.4961496149614961,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.02244059405940594,
      "loss": 0.2309,
      "num_input_tokens_seen": 2868736,
      "step": 13600
    },
    {
      "epoch": 1.4966996699669968,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02244884488448845,
      "loss": 0.2315,
      "num_input_tokens_seen": 2869824,
      "step": 13605
    },
    {
      "epoch": 1.4972497249724972,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.022457095709570957,
      "loss": 0.2295,
      "num_input_tokens_seen": 2870912,
      "step": 13610
    },
    {
      "epoch": 1.4977997799779978,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.022465346534653464,
      "loss": 0.2341,
      "num_input_tokens_seen": 2871936,
      "step": 13615
    },
    {
      "epoch": 1.4983498349834983,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.02247359735973597,
      "loss": 0.232,
      "num_input_tokens_seen": 2873024,
      "step": 13620
    },
    {
      "epoch": 1.498899889988999,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.022481848184818482,
      "loss": 0.233,
      "num_input_tokens_seen": 2874048,
      "step": 13625
    },
    {
      "epoch": 1.4994499449944994,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.02249009900990099,
      "loss": 0.2325,
      "num_input_tokens_seen": 2875136,
      "step": 13630
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.022498349834983496,
      "loss": 0.2309,
      "num_input_tokens_seen": 2876224,
      "step": 13635
    },
    {
      "epoch": 1.5005500550055006,
      "grad_norm": 0.018310546875,
      "learning_rate": 0.022506600660066003,
      "loss": 0.2314,
      "num_input_tokens_seen": 2877248,
      "step": 13640
    },
    {
      "epoch": 1.501100110011001,
      "grad_norm": 0.00897216796875,
      "learning_rate": 0.022514851485148514,
      "loss": 0.2325,
      "num_input_tokens_seen": 2878336,
      "step": 13645
    },
    {
      "epoch": 1.5016501650165015,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.022523102310231025,
      "loss": 0.233,
      "num_input_tokens_seen": 2879360,
      "step": 13650
    },
    {
      "epoch": 1.5022002200220022,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02253135313531353,
      "loss": 0.2341,
      "num_input_tokens_seen": 2880416,
      "step": 13655
    },
    {
      "epoch": 1.5027502750275028,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.02253960396039604,
      "loss": 0.2346,
      "num_input_tokens_seen": 2881504,
      "step": 13660
    },
    {
      "epoch": 1.5033003300330035,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.022547854785478546,
      "loss": 0.2309,
      "num_input_tokens_seen": 2882528,
      "step": 13665
    },
    {
      "epoch": 1.5038503850385039,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.022556105610561056,
      "loss": 0.2304,
      "num_input_tokens_seen": 2883616,
      "step": 13670
    },
    {
      "epoch": 1.5044004400440043,
      "grad_norm": 0.0084228515625,
      "learning_rate": 0.022564356435643564,
      "loss": 0.2319,
      "num_input_tokens_seen": 2884736,
      "step": 13675
    },
    {
      "epoch": 1.504950495049505,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.02257260726072607,
      "loss": 0.2314,
      "num_input_tokens_seen": 2885760,
      "step": 13680
    },
    {
      "epoch": 1.5055005500550056,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.022580858085808578,
      "loss": 0.2309,
      "num_input_tokens_seen": 2886848,
      "step": 13685
    },
    {
      "epoch": 1.506050605060506,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.02258910891089109,
      "loss": 0.2299,
      "num_input_tokens_seen": 2887904,
      "step": 13690
    },
    {
      "epoch": 1.5066006600660065,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.0225973597359736,
      "loss": 0.2325,
      "num_input_tokens_seen": 2889024,
      "step": 13695
    },
    {
      "epoch": 1.507150715071507,
      "grad_norm": 0.00897216796875,
      "learning_rate": 0.022605610561056106,
      "loss": 0.233,
      "num_input_tokens_seen": 2890080,
      "step": 13700
    },
    {
      "epoch": 1.5077007700770078,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.022613861386138613,
      "loss": 0.232,
      "num_input_tokens_seen": 2891104,
      "step": 13705
    },
    {
      "epoch": 1.5082508250825084,
      "grad_norm": 0.01904296875,
      "learning_rate": 0.02262211221122112,
      "loss": 0.231,
      "num_input_tokens_seen": 2892128,
      "step": 13710
    },
    {
      "epoch": 1.5088008800880088,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.02263036303630363,
      "loss": 0.2315,
      "num_input_tokens_seen": 2893120,
      "step": 13715
    },
    {
      "epoch": 1.5093509350935093,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.022638613861386138,
      "loss": 0.2325,
      "num_input_tokens_seen": 2894176,
      "step": 13720
    },
    {
      "epoch": 1.50990099009901,
      "grad_norm": 0.0032806396484375,
      "learning_rate": 0.022646864686468645,
      "loss": 0.2319,
      "num_input_tokens_seen": 2895296,
      "step": 13725
    },
    {
      "epoch": 1.5104510451045106,
      "grad_norm": 0.00347900390625,
      "learning_rate": 0.022655115511551152,
      "loss": 0.2324,
      "num_input_tokens_seen": 2896352,
      "step": 13730
    },
    {
      "epoch": 1.511001100110011,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.022663366336633663,
      "loss": 0.2319,
      "num_input_tokens_seen": 2897408,
      "step": 13735
    },
    {
      "epoch": 1.5115511551155114,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.022671617161716173,
      "loss": 0.2314,
      "num_input_tokens_seen": 2898528,
      "step": 13740
    },
    {
      "epoch": 1.512101210121012,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.02267986798679868,
      "loss": 0.2334,
      "num_input_tokens_seen": 2899616,
      "step": 13745
    },
    {
      "epoch": 1.5126512651265127,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.022688118811881187,
      "loss": 0.2319,
      "num_input_tokens_seen": 2900608,
      "step": 13750
    },
    {
      "epoch": 1.5132013201320134,
      "grad_norm": 0.003997802734375,
      "learning_rate": 0.022696369636963695,
      "loss": 0.2313,
      "num_input_tokens_seen": 2901600,
      "step": 13755
    },
    {
      "epoch": 1.5137513751375138,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.022704620462046205,
      "loss": 0.2319,
      "num_input_tokens_seen": 2902592,
      "step": 13760
    },
    {
      "epoch": 1.5143014301430142,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.022712871287128712,
      "loss": 0.2298,
      "num_input_tokens_seen": 2903648,
      "step": 13765
    },
    {
      "epoch": 1.5148514851485149,
      "grad_norm": 0.0038299560546875,
      "learning_rate": 0.02272112211221122,
      "loss": 0.2305,
      "num_input_tokens_seen": 2904704,
      "step": 13770
    },
    {
      "epoch": 1.5154015401540155,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.022729372937293726,
      "loss": 0.2286,
      "num_input_tokens_seen": 2905728,
      "step": 13775
    },
    {
      "epoch": 1.515951595159516,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.022737623762376237,
      "loss": 0.2315,
      "num_input_tokens_seen": 2906784,
      "step": 13780
    },
    {
      "epoch": 1.5165016501650164,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.022745874587458744,
      "loss": 0.2274,
      "num_input_tokens_seen": 2907840,
      "step": 13785
    },
    {
      "epoch": 1.517051705170517,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.022754125412541255,
      "loss": 0.2333,
      "num_input_tokens_seen": 2908928,
      "step": 13790
    },
    {
      "epoch": 1.5176017601760177,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.022762376237623762,
      "loss": 0.2252,
      "num_input_tokens_seen": 2909984,
      "step": 13795
    },
    {
      "epoch": 1.5181518151815183,
      "grad_norm": 0.0033111572265625,
      "learning_rate": 0.02277062706270627,
      "loss": 0.2317,
      "num_input_tokens_seen": 2910976,
      "step": 13800
    },
    {
      "epoch": 1.5187018701870187,
      "grad_norm": 0.0185546875,
      "learning_rate": 0.02277887788778878,
      "loss": 0.2374,
      "num_input_tokens_seen": 2912032,
      "step": 13805
    },
    {
      "epoch": 1.5192519251925192,
      "grad_norm": 0.004150390625,
      "learning_rate": 0.022787128712871287,
      "loss": 0.2313,
      "num_input_tokens_seen": 2913088,
      "step": 13810
    },
    {
      "epoch": 1.5198019801980198,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.022795379537953794,
      "loss": 0.2403,
      "num_input_tokens_seen": 2914144,
      "step": 13815
    },
    {
      "epoch": 1.5203520352035205,
      "grad_norm": 0.00384521484375,
      "learning_rate": 0.0228036303630363,
      "loss": 0.2415,
      "num_input_tokens_seen": 2915232,
      "step": 13820
    },
    {
      "epoch": 1.520902090209021,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.02281188118811881,
      "loss": 0.2265,
      "num_input_tokens_seen": 2916288,
      "step": 13825
    },
    {
      "epoch": 1.5214521452145213,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.02282013201320132,
      "loss": 0.2295,
      "num_input_tokens_seen": 2917376,
      "step": 13830
    },
    {
      "epoch": 1.522002200220022,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.02282838283828383,
      "loss": 0.2326,
      "num_input_tokens_seen": 2918432,
      "step": 13835
    },
    {
      "epoch": 1.5225522552255226,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.022836633663366336,
      "loss": 0.2335,
      "num_input_tokens_seen": 2919552,
      "step": 13840
    },
    {
      "epoch": 1.523102310231023,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.022844884488448843,
      "loss": 0.2324,
      "num_input_tokens_seen": 2920640,
      "step": 13845
    },
    {
      "epoch": 1.5236523652365237,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.022853135313531354,
      "loss": 0.2324,
      "num_input_tokens_seen": 2921696,
      "step": 13850
    },
    {
      "epoch": 1.5242024202420241,
      "grad_norm": 0.01806640625,
      "learning_rate": 0.02286138613861386,
      "loss": 0.2302,
      "num_input_tokens_seen": 2922720,
      "step": 13855
    },
    {
      "epoch": 1.5247524752475248,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.022869636963696368,
      "loss": 0.2334,
      "num_input_tokens_seen": 2923776,
      "step": 13860
    },
    {
      "epoch": 1.5253025302530254,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.022877887788778875,
      "loss": 0.2303,
      "num_input_tokens_seen": 2924768,
      "step": 13865
    },
    {
      "epoch": 1.5258525852585259,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.022886138613861386,
      "loss": 0.2273,
      "num_input_tokens_seen": 2925856,
      "step": 13870
    },
    {
      "epoch": 1.5264026402640263,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.022894389438943893,
      "loss": 0.2283,
      "num_input_tokens_seen": 2926976,
      "step": 13875
    },
    {
      "epoch": 1.526952695269527,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.022902640264026403,
      "loss": 0.2304,
      "num_input_tokens_seen": 2928000,
      "step": 13880
    },
    {
      "epoch": 1.5275027502750276,
      "grad_norm": 0.0185546875,
      "learning_rate": 0.02291089108910891,
      "loss": 0.2304,
      "num_input_tokens_seen": 2929056,
      "step": 13885
    },
    {
      "epoch": 1.528052805280528,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.022919141914191418,
      "loss": 0.2272,
      "num_input_tokens_seen": 2930208,
      "step": 13890
    },
    {
      "epoch": 1.5286028602860287,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.022927392739273928,
      "loss": 0.2264,
      "num_input_tokens_seen": 2931200,
      "step": 13895
    },
    {
      "epoch": 1.529152915291529,
      "grad_norm": 0.002593994140625,
      "learning_rate": 0.022935643564356435,
      "loss": 0.2305,
      "num_input_tokens_seen": 2932288,
      "step": 13900
    },
    {
      "epoch": 1.5297029702970297,
      "grad_norm": 0.0203857421875,
      "learning_rate": 0.022943894389438942,
      "loss": 0.2286,
      "num_input_tokens_seen": 2933280,
      "step": 13905
    },
    {
      "epoch": 1.5302530253025304,
      "grad_norm": 0.0277099609375,
      "learning_rate": 0.02295214521452145,
      "loss": 0.2405,
      "num_input_tokens_seen": 2934336,
      "step": 13910
    },
    {
      "epoch": 1.5308030803080308,
      "grad_norm": 0.0184326171875,
      "learning_rate": 0.02296039603960396,
      "loss": 0.232,
      "num_input_tokens_seen": 2935360,
      "step": 13915
    },
    {
      "epoch": 1.5313531353135312,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.022968646864686467,
      "loss": 0.2358,
      "num_input_tokens_seen": 2936352,
      "step": 13920
    },
    {
      "epoch": 1.5319031903190319,
      "grad_norm": 0.004058837890625,
      "learning_rate": 0.022976897689768978,
      "loss": 0.2327,
      "num_input_tokens_seen": 2937408,
      "step": 13925
    },
    {
      "epoch": 1.5324532453245325,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.022985148514851485,
      "loss": 0.2317,
      "num_input_tokens_seen": 2938560,
      "step": 13930
    },
    {
      "epoch": 1.533003300330033,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.022993399339933992,
      "loss": 0.2327,
      "num_input_tokens_seen": 2939584,
      "step": 13935
    },
    {
      "epoch": 1.5335533553355336,
      "grad_norm": 0.0031280517578125,
      "learning_rate": 0.023001650165016502,
      "loss": 0.2325,
      "num_input_tokens_seen": 2940640,
      "step": 13940
    },
    {
      "epoch": 1.534103410341034,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.02300990099009901,
      "loss": 0.2336,
      "num_input_tokens_seen": 2941664,
      "step": 13945
    },
    {
      "epoch": 1.5346534653465347,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.023018151815181517,
      "loss": 0.2319,
      "num_input_tokens_seen": 2942720,
      "step": 13950
    },
    {
      "epoch": 1.5352035203520353,
      "grad_norm": 0.0177001953125,
      "learning_rate": 0.023026402640264024,
      "loss": 0.2318,
      "num_input_tokens_seen": 2943744,
      "step": 13955
    },
    {
      "epoch": 1.5357535753575358,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.023034653465346534,
      "loss": 0.2314,
      "num_input_tokens_seen": 2944704,
      "step": 13960
    },
    {
      "epoch": 1.5363036303630362,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.02304290429042904,
      "loss": 0.2288,
      "num_input_tokens_seen": 2945760,
      "step": 13965
    },
    {
      "epoch": 1.5368536853685368,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.023051155115511552,
      "loss": 0.2364,
      "num_input_tokens_seen": 2946720,
      "step": 13970
    },
    {
      "epoch": 1.5374037403740375,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.02305940594059406,
      "loss": 0.2265,
      "num_input_tokens_seen": 2947776,
      "step": 13975
    },
    {
      "epoch": 1.537953795379538,
      "grad_norm": 0.00390625,
      "learning_rate": 0.023067656765676566,
      "loss": 0.2332,
      "num_input_tokens_seen": 2948800,
      "step": 13980
    },
    {
      "epoch": 1.5385038503850383,
      "grad_norm": 0.0028076171875,
      "learning_rate": 0.023075907590759077,
      "loss": 0.2339,
      "num_input_tokens_seen": 2949824,
      "step": 13985
    },
    {
      "epoch": 1.539053905390539,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.023084158415841584,
      "loss": 0.2359,
      "num_input_tokens_seen": 2950912,
      "step": 13990
    },
    {
      "epoch": 1.5396039603960396,
      "grad_norm": 0.0037689208984375,
      "learning_rate": 0.02309240924092409,
      "loss": 0.2358,
      "num_input_tokens_seen": 2951968,
      "step": 13995
    },
    {
      "epoch": 1.5401540154015403,
      "grad_norm": 0.009765625,
      "learning_rate": 0.023100660066006598,
      "loss": 0.2325,
      "num_input_tokens_seen": 2953024,
      "step": 14000
    },
    {
      "epoch": 1.5407040704070407,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.02310891089108911,
      "loss": 0.2319,
      "num_input_tokens_seen": 2954048,
      "step": 14005
    },
    {
      "epoch": 1.5412541254125411,
      "grad_norm": 0.01806640625,
      "learning_rate": 0.023117161716171616,
      "loss": 0.234,
      "num_input_tokens_seen": 2955040,
      "step": 14010
    },
    {
      "epoch": 1.5418041804180418,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.023125412541254126,
      "loss": 0.2309,
      "num_input_tokens_seen": 2956064,
      "step": 14015
    },
    {
      "epoch": 1.5423542354235424,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.02313366336633663,
      "loss": 0.2308,
      "num_input_tokens_seen": 2957152,
      "step": 14020
    },
    {
      "epoch": 1.5429042904290429,
      "grad_norm": 0.00286865234375,
      "learning_rate": 0.02314191419141914,
      "loss": 0.2329,
      "num_input_tokens_seen": 2958208,
      "step": 14025
    },
    {
      "epoch": 1.5434543454345433,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02315016501650165,
      "loss": 0.2314,
      "num_input_tokens_seen": 2959200,
      "step": 14030
    },
    {
      "epoch": 1.544004400440044,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.023158415841584158,
      "loss": 0.2325,
      "num_input_tokens_seen": 2960256,
      "step": 14035
    },
    {
      "epoch": 1.5445544554455446,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.023166666666666665,
      "loss": 0.2308,
      "num_input_tokens_seen": 2961312,
      "step": 14040
    },
    {
      "epoch": 1.5451045104510452,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.023174917491749172,
      "loss": 0.2314,
      "num_input_tokens_seen": 2962400,
      "step": 14045
    },
    {
      "epoch": 1.5456545654565457,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.023183168316831683,
      "loss": 0.2309,
      "num_input_tokens_seen": 2963488,
      "step": 14050
    },
    {
      "epoch": 1.546204620462046,
      "grad_norm": 0.01806640625,
      "learning_rate": 0.02319141914191419,
      "loss": 0.2314,
      "num_input_tokens_seen": 2964448,
      "step": 14055
    },
    {
      "epoch": 1.5467546754675467,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.0231996699669967,
      "loss": 0.2324,
      "num_input_tokens_seen": 2965568,
      "step": 14060
    },
    {
      "epoch": 1.5473047304730474,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.023207920792079204,
      "loss": 0.2324,
      "num_input_tokens_seen": 2966624,
      "step": 14065
    },
    {
      "epoch": 1.5478547854785478,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.023216171617161715,
      "loss": 0.232,
      "num_input_tokens_seen": 2967648,
      "step": 14070
    },
    {
      "epoch": 1.5484048404840483,
      "grad_norm": 0.0172119140625,
      "learning_rate": 0.023224422442244225,
      "loss": 0.2304,
      "num_input_tokens_seen": 2968672,
      "step": 14075
    },
    {
      "epoch": 1.548954895489549,
      "grad_norm": 0.00927734375,
      "learning_rate": 0.023232673267326732,
      "loss": 0.2345,
      "num_input_tokens_seen": 2969696,
      "step": 14080
    },
    {
      "epoch": 1.5495049504950495,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.02324092409240924,
      "loss": 0.2329,
      "num_input_tokens_seen": 2970720,
      "step": 14085
    },
    {
      "epoch": 1.5500550055005502,
      "grad_norm": 0.00927734375,
      "learning_rate": 0.023249174917491747,
      "loss": 0.2309,
      "num_input_tokens_seen": 2971744,
      "step": 14090
    },
    {
      "epoch": 1.5506050605060506,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.023257425742574257,
      "loss": 0.233,
      "num_input_tokens_seen": 2972768,
      "step": 14095
    },
    {
      "epoch": 1.551155115511551,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.023265676567656764,
      "loss": 0.2324,
      "num_input_tokens_seen": 2973760,
      "step": 14100
    },
    {
      "epoch": 1.5517051705170517,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.023273927392739275,
      "loss": 0.2303,
      "num_input_tokens_seen": 2974784,
      "step": 14105
    },
    {
      "epoch": 1.5522552255225524,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.02328217821782178,
      "loss": 0.232,
      "num_input_tokens_seen": 2975840,
      "step": 14110
    },
    {
      "epoch": 1.5528052805280528,
      "grad_norm": 0.0179443359375,
      "learning_rate": 0.02329042904290429,
      "loss": 0.2299,
      "num_input_tokens_seen": 2976832,
      "step": 14115
    },
    {
      "epoch": 1.5533553355335532,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.0232986798679868,
      "loss": 0.2294,
      "num_input_tokens_seen": 2977792,
      "step": 14120
    },
    {
      "epoch": 1.5539053905390539,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.023306930693069307,
      "loss": 0.2336,
      "num_input_tokens_seen": 2978784,
      "step": 14125
    },
    {
      "epoch": 1.5544554455445545,
      "grad_norm": 0.0186767578125,
      "learning_rate": 0.023315181518151814,
      "loss": 0.2284,
      "num_input_tokens_seen": 2979840,
      "step": 14130
    },
    {
      "epoch": 1.5550055005500552,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.02332343234323432,
      "loss": 0.2346,
      "num_input_tokens_seen": 2980928,
      "step": 14135
    },
    {
      "epoch": 1.5555555555555556,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.02333168316831683,
      "loss": 0.2362,
      "num_input_tokens_seen": 2982016,
      "step": 14140
    },
    {
      "epoch": 1.556105610561056,
      "grad_norm": 0.017333984375,
      "learning_rate": 0.02333993399339934,
      "loss": 0.2329,
      "num_input_tokens_seen": 2983136,
      "step": 14145
    },
    {
      "epoch": 1.5566556655665567,
      "grad_norm": 0.0033111572265625,
      "learning_rate": 0.02334818481848185,
      "loss": 0.2258,
      "num_input_tokens_seen": 2984192,
      "step": 14150
    },
    {
      "epoch": 1.5572057205720573,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.023356435643564353,
      "loss": 0.2328,
      "num_input_tokens_seen": 2985280,
      "step": 14155
    },
    {
      "epoch": 1.5577557755775577,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.023364686468646863,
      "loss": 0.2327,
      "num_input_tokens_seen": 2986400,
      "step": 14160
    },
    {
      "epoch": 1.5583058305830582,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.023372937293729374,
      "loss": 0.2308,
      "num_input_tokens_seen": 2987456,
      "step": 14165
    },
    {
      "epoch": 1.5588558855885588,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.02338118811881188,
      "loss": 0.2275,
      "num_input_tokens_seen": 2988448,
      "step": 14170
    },
    {
      "epoch": 1.5594059405940595,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.023389438943894388,
      "loss": 0.236,
      "num_input_tokens_seen": 2989504,
      "step": 14175
    },
    {
      "epoch": 1.55995599559956,
      "grad_norm": 0.0026397705078125,
      "learning_rate": 0.023397689768976895,
      "loss": 0.2244,
      "num_input_tokens_seen": 2990528,
      "step": 14180
    },
    {
      "epoch": 1.5605060506050605,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.023405940594059406,
      "loss": 0.2318,
      "num_input_tokens_seen": 2991552,
      "step": 14185
    },
    {
      "epoch": 1.561056105610561,
      "grad_norm": 0.002838134765625,
      "learning_rate": 0.023414191419141913,
      "loss": 0.2308,
      "num_input_tokens_seen": 2992576,
      "step": 14190
    },
    {
      "epoch": 1.5616061606160616,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.023422442244224424,
      "loss": 0.233,
      "num_input_tokens_seen": 2993696,
      "step": 14195
    },
    {
      "epoch": 1.5621562156215623,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.023430693069306927,
      "loss": 0.2364,
      "num_input_tokens_seen": 2994816,
      "step": 14200
    },
    {
      "epoch": 1.5627062706270627,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.023438943894389438,
      "loss": 0.2319,
      "num_input_tokens_seen": 2995872,
      "step": 14205
    },
    {
      "epoch": 1.5632563256325631,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.02344719471947195,
      "loss": 0.2371,
      "num_input_tokens_seen": 2996928,
      "step": 14210
    },
    {
      "epoch": 1.5638063806380638,
      "grad_norm": 0.009765625,
      "learning_rate": 0.023455445544554455,
      "loss": 0.2286,
      "num_input_tokens_seen": 2998016,
      "step": 14215
    },
    {
      "epoch": 1.5643564356435644,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.023463696369636963,
      "loss": 0.235,
      "num_input_tokens_seen": 2999104,
      "step": 14220
    },
    {
      "epoch": 1.564906490649065,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02347194719471947,
      "loss": 0.2339,
      "num_input_tokens_seen": 3000224,
      "step": 14225
    },
    {
      "epoch": 1.5654565456545655,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.02348019801980198,
      "loss": 0.2333,
      "num_input_tokens_seen": 3001216,
      "step": 14230
    },
    {
      "epoch": 1.566006600660066,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.023488448844884487,
      "loss": 0.23,
      "num_input_tokens_seen": 3002272,
      "step": 14235
    },
    {
      "epoch": 1.5665566556655666,
      "grad_norm": 0.00830078125,
      "learning_rate": 0.023496699669966998,
      "loss": 0.2305,
      "num_input_tokens_seen": 3003296,
      "step": 14240
    },
    {
      "epoch": 1.5671067106710672,
      "grad_norm": 0.0027923583984375,
      "learning_rate": 0.0235049504950495,
      "loss": 0.2325,
      "num_input_tokens_seen": 3004320,
      "step": 14245
    },
    {
      "epoch": 1.5676567656765676,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.023513201320132012,
      "loss": 0.2319,
      "num_input_tokens_seen": 3005440,
      "step": 14250
    },
    {
      "epoch": 1.568206820682068,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.023521452145214523,
      "loss": 0.2314,
      "num_input_tokens_seen": 3006496,
      "step": 14255
    },
    {
      "epoch": 1.5687568756875687,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.02352970297029703,
      "loss": 0.2289,
      "num_input_tokens_seen": 3007616,
      "step": 14260
    },
    {
      "epoch": 1.5693069306930694,
      "grad_norm": 0.017333984375,
      "learning_rate": 0.023537953795379537,
      "loss": 0.2379,
      "num_input_tokens_seen": 3008640,
      "step": 14265
    },
    {
      "epoch": 1.56985698569857,
      "grad_norm": 0.00311279296875,
      "learning_rate": 0.023546204620462044,
      "loss": 0.2312,
      "num_input_tokens_seen": 3009728,
      "step": 14270
    },
    {
      "epoch": 1.5704070407040704,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.023554455445544555,
      "loss": 0.2279,
      "num_input_tokens_seen": 3010752,
      "step": 14275
    },
    {
      "epoch": 1.5709570957095709,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.02356270627062706,
      "loss": 0.2311,
      "num_input_tokens_seen": 3011776,
      "step": 14280
    },
    {
      "epoch": 1.5715071507150715,
      "grad_norm": 0.0078125,
      "learning_rate": 0.023570957095709572,
      "loss": 0.2327,
      "num_input_tokens_seen": 3012928,
      "step": 14285
    },
    {
      "epoch": 1.5720572057205722,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.023579207920792076,
      "loss": 0.2322,
      "num_input_tokens_seen": 3013920,
      "step": 14290
    },
    {
      "epoch": 1.5726072607260726,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.023587458745874586,
      "loss": 0.227,
      "num_input_tokens_seen": 3014976,
      "step": 14295
    },
    {
      "epoch": 1.573157315731573,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.023595709570957094,
      "loss": 0.2344,
      "num_input_tokens_seen": 3016064,
      "step": 14300
    },
    {
      "epoch": 1.5737073707370737,
      "grad_norm": 0.0159912109375,
      "learning_rate": 0.023603960396039604,
      "loss": 0.229,
      "num_input_tokens_seen": 3017152,
      "step": 14305
    },
    {
      "epoch": 1.5742574257425743,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.02361221122112211,
      "loss": 0.2316,
      "num_input_tokens_seen": 3018208,
      "step": 14310
    },
    {
      "epoch": 1.574807480748075,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.02362046204620462,
      "loss": 0.2321,
      "num_input_tokens_seen": 3019296,
      "step": 14315
    },
    {
      "epoch": 1.5753575357535754,
      "grad_norm": 0.0081787109375,
      "learning_rate": 0.02362871287128713,
      "loss": 0.2336,
      "num_input_tokens_seen": 3020320,
      "step": 14320
    },
    {
      "epoch": 1.5759075907590758,
      "grad_norm": 0.01708984375,
      "learning_rate": 0.023636963696369636,
      "loss": 0.2326,
      "num_input_tokens_seen": 3021344,
      "step": 14325
    },
    {
      "epoch": 1.5764576457645765,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.023645214521452147,
      "loss": 0.2299,
      "num_input_tokens_seen": 3022368,
      "step": 14330
    },
    {
      "epoch": 1.5770077007700771,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.02365346534653465,
      "loss": 0.2288,
      "num_input_tokens_seen": 3023456,
      "step": 14335
    },
    {
      "epoch": 1.5775577557755776,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.02366171617161716,
      "loss": 0.2305,
      "num_input_tokens_seen": 3024512,
      "step": 14340
    },
    {
      "epoch": 1.578107810781078,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.023669966996699668,
      "loss": 0.232,
      "num_input_tokens_seen": 3025600,
      "step": 14345
    },
    {
      "epoch": 1.5786578657865786,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.02367821782178218,
      "loss": 0.2325,
      "num_input_tokens_seen": 3026688,
      "step": 14350
    },
    {
      "epoch": 1.5792079207920793,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.023686468646864686,
      "loss": 0.232,
      "num_input_tokens_seen": 3027808,
      "step": 14355
    },
    {
      "epoch": 1.5797579757975797,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.023694719471947193,
      "loss": 0.2335,
      "num_input_tokens_seen": 3028832,
      "step": 14360
    },
    {
      "epoch": 1.5803080308030804,
      "grad_norm": 0.00323486328125,
      "learning_rate": 0.023702970297029703,
      "loss": 0.2294,
      "num_input_tokens_seen": 3029856,
      "step": 14365
    },
    {
      "epoch": 1.5808580858085808,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.02371122112211221,
      "loss": 0.2304,
      "num_input_tokens_seen": 3030912,
      "step": 14370
    },
    {
      "epoch": 1.5814081408140814,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.02371947194719472,
      "loss": 0.2309,
      "num_input_tokens_seen": 3031968,
      "step": 14375
    },
    {
      "epoch": 1.581958195819582,
      "grad_norm": 0.0172119140625,
      "learning_rate": 0.023727722772277225,
      "loss": 0.232,
      "num_input_tokens_seen": 3033056,
      "step": 14380
    },
    {
      "epoch": 1.5825082508250825,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.023735973597359735,
      "loss": 0.229,
      "num_input_tokens_seen": 3034080,
      "step": 14385
    },
    {
      "epoch": 1.583058305830583,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.023744224422442242,
      "loss": 0.2327,
      "num_input_tokens_seen": 3035168,
      "step": 14390
    },
    {
      "epoch": 1.5836083608360836,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.023752475247524753,
      "loss": 0.2338,
      "num_input_tokens_seen": 3036256,
      "step": 14395
    },
    {
      "epoch": 1.5841584158415842,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.02376072607260726,
      "loss": 0.2327,
      "num_input_tokens_seen": 3037280,
      "step": 14400
    },
    {
      "epoch": 1.5847084708470847,
      "grad_norm": 0.00927734375,
      "learning_rate": 0.023768976897689767,
      "loss": 0.229,
      "num_input_tokens_seen": 3038304,
      "step": 14405
    },
    {
      "epoch": 1.5852585258525853,
      "grad_norm": 0.0030059814453125,
      "learning_rate": 0.023777227722772278,
      "loss": 0.2322,
      "num_input_tokens_seen": 3039296,
      "step": 14410
    },
    {
      "epoch": 1.5858085808580857,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.023785478547854785,
      "loss": 0.2316,
      "num_input_tokens_seen": 3040352,
      "step": 14415
    },
    {
      "epoch": 1.5863586358635864,
      "grad_norm": 0.017578125,
      "learning_rate": 0.023793729372937295,
      "loss": 0.2358,
      "num_input_tokens_seen": 3041408,
      "step": 14420
    },
    {
      "epoch": 1.586908690869087,
      "grad_norm": 0.00927734375,
      "learning_rate": 0.0238019801980198,
      "loss": 0.2284,
      "num_input_tokens_seen": 3042432,
      "step": 14425
    },
    {
      "epoch": 1.5874587458745875,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.02381023102310231,
      "loss": 0.2315,
      "num_input_tokens_seen": 3043392,
      "step": 14430
    },
    {
      "epoch": 1.588008800880088,
      "grad_norm": 0.0078125,
      "learning_rate": 0.023818481848184817,
      "loss": 0.2299,
      "num_input_tokens_seen": 3044384,
      "step": 14435
    },
    {
      "epoch": 1.5885588558855885,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.023826732673267327,
      "loss": 0.2346,
      "num_input_tokens_seen": 3045376,
      "step": 14440
    },
    {
      "epoch": 1.5891089108910892,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.023834983498349834,
      "loss": 0.2284,
      "num_input_tokens_seen": 3046400,
      "step": 14445
    },
    {
      "epoch": 1.5896589658965896,
      "grad_norm": 0.007781982421875,
      "learning_rate": 0.02384323432343234,
      "loss": 0.2311,
      "num_input_tokens_seen": 3047488,
      "step": 14450
    },
    {
      "epoch": 1.5902090209020903,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.023851485148514852,
      "loss": 0.228,
      "num_input_tokens_seen": 3048512,
      "step": 14455
    },
    {
      "epoch": 1.5907590759075907,
      "grad_norm": 0.0078125,
      "learning_rate": 0.02385973597359736,
      "loss": 0.2317,
      "num_input_tokens_seen": 3049600,
      "step": 14460
    },
    {
      "epoch": 1.5913091309130913,
      "grad_norm": 0.017822265625,
      "learning_rate": 0.02386798679867987,
      "loss": 0.2361,
      "num_input_tokens_seen": 3050656,
      "step": 14465
    },
    {
      "epoch": 1.591859185918592,
      "grad_norm": 0.017822265625,
      "learning_rate": 0.023876237623762373,
      "loss": 0.2355,
      "num_input_tokens_seen": 3051744,
      "step": 14470
    },
    {
      "epoch": 1.5924092409240924,
      "grad_norm": 0.0091552734375,
      "learning_rate": 0.023884488448844884,
      "loss": 0.2358,
      "num_input_tokens_seen": 3052736,
      "step": 14475
    },
    {
      "epoch": 1.5929592959295928,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.02389273927392739,
      "loss": 0.2316,
      "num_input_tokens_seen": 3053824,
      "step": 14480
    },
    {
      "epoch": 1.5935093509350935,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.0239009900990099,
      "loss": 0.23,
      "num_input_tokens_seen": 3054880,
      "step": 14485
    },
    {
      "epoch": 1.5940594059405941,
      "grad_norm": 0.00927734375,
      "learning_rate": 0.02390924092409241,
      "loss": 0.2315,
      "num_input_tokens_seen": 3055936,
      "step": 14490
    },
    {
      "epoch": 1.5946094609460946,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.023917491749174916,
      "loss": 0.2315,
      "num_input_tokens_seen": 3056992,
      "step": 14495
    },
    {
      "epoch": 1.595159515951595,
      "grad_norm": 0.01708984375,
      "learning_rate": 0.023925742574257426,
      "loss": 0.2325,
      "num_input_tokens_seen": 3058048,
      "step": 14500
    },
    {
      "epoch": 1.5957095709570956,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.023933993399339933,
      "loss": 0.2314,
      "num_input_tokens_seen": 3059072,
      "step": 14505
    },
    {
      "epoch": 1.5962596259625963,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.023942244224422444,
      "loss": 0.2314,
      "num_input_tokens_seen": 3060224,
      "step": 14510
    },
    {
      "epoch": 1.596809680968097,
      "grad_norm": 0.0169677734375,
      "learning_rate": 0.023950495049504947,
      "loss": 0.2309,
      "num_input_tokens_seen": 3061312,
      "step": 14515
    },
    {
      "epoch": 1.5973597359735974,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.023958745874587458,
      "loss": 0.2346,
      "num_input_tokens_seen": 3062368,
      "step": 14520
    },
    {
      "epoch": 1.5979097909790978,
      "grad_norm": 0.0084228515625,
      "learning_rate": 0.023966996699669965,
      "loss": 0.2309,
      "num_input_tokens_seen": 3063392,
      "step": 14525
    },
    {
      "epoch": 1.5984598459845984,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.023975247524752476,
      "loss": 0.233,
      "num_input_tokens_seen": 3064384,
      "step": 14530
    },
    {
      "epoch": 1.599009900990099,
      "grad_norm": 0.00830078125,
      "learning_rate": 0.02398349834983498,
      "loss": 0.2299,
      "num_input_tokens_seen": 3065376,
      "step": 14535
    },
    {
      "epoch": 1.5995599559955995,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.02399174917491749,
      "loss": 0.2314,
      "num_input_tokens_seen": 3066464,
      "step": 14540
    },
    {
      "epoch": 1.6001100110011,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.024,
      "loss": 0.2314,
      "num_input_tokens_seen": 3067424,
      "step": 14545
    },
    {
      "epoch": 1.6006600660066006,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.024008250825082508,
      "loss": 0.2299,
      "num_input_tokens_seen": 3068512,
      "step": 14550
    },
    {
      "epoch": 1.6012101210121013,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.024016501650165018,
      "loss": 0.2299,
      "num_input_tokens_seen": 3069568,
      "step": 14555
    },
    {
      "epoch": 1.601760176017602,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.024024752475247522,
      "loss": 0.2299,
      "num_input_tokens_seen": 3070624,
      "step": 14560
    },
    {
      "epoch": 1.6023102310231023,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.024033003300330032,
      "loss": 0.2351,
      "num_input_tokens_seen": 3071712,
      "step": 14565
    },
    {
      "epoch": 1.6028602860286028,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.02404125412541254,
      "loss": 0.2304,
      "num_input_tokens_seen": 3072768,
      "step": 14570
    },
    {
      "epoch": 1.6034103410341034,
      "grad_norm": 0.00286865234375,
      "learning_rate": 0.02404950495049505,
      "loss": 0.2289,
      "num_input_tokens_seen": 3073792,
      "step": 14575
    },
    {
      "epoch": 1.603960396039604,
      "grad_norm": 0.0084228515625,
      "learning_rate": 0.024057755775577554,
      "loss": 0.2325,
      "num_input_tokens_seen": 3074880,
      "step": 14580
    },
    {
      "epoch": 1.6045104510451045,
      "grad_norm": 0.016357421875,
      "learning_rate": 0.024066006600660064,
      "loss": 0.2299,
      "num_input_tokens_seen": 3076000,
      "step": 14585
    },
    {
      "epoch": 1.605060506050605,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.024074257425742575,
      "loss": 0.2299,
      "num_input_tokens_seen": 3077056,
      "step": 14590
    },
    {
      "epoch": 1.6056105610561056,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.024082508250825082,
      "loss": 0.231,
      "num_input_tokens_seen": 3078144,
      "step": 14595
    },
    {
      "epoch": 1.6061606160616062,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.024090759075907592,
      "loss": 0.2336,
      "num_input_tokens_seen": 3079200,
      "step": 14600
    },
    {
      "epoch": 1.6067106710671069,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.024099009900990096,
      "loss": 0.2331,
      "num_input_tokens_seen": 3080256,
      "step": 14605
    },
    {
      "epoch": 1.6072607260726073,
      "grad_norm": 0.0030364990234375,
      "learning_rate": 0.024107260726072607,
      "loss": 0.2336,
      "num_input_tokens_seen": 3081376,
      "step": 14610
    },
    {
      "epoch": 1.6078107810781077,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.024115511551155114,
      "loss": 0.2273,
      "num_input_tokens_seen": 3082464,
      "step": 14615
    },
    {
      "epoch": 1.6083608360836084,
      "grad_norm": 0.0177001953125,
      "learning_rate": 0.024123762376237624,
      "loss": 0.2321,
      "num_input_tokens_seen": 3083456,
      "step": 14620
    },
    {
      "epoch": 1.608910891089109,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.024132013201320128,
      "loss": 0.2295,
      "num_input_tokens_seen": 3084512,
      "step": 14625
    },
    {
      "epoch": 1.6094609460946094,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.02414026402640264,
      "loss": 0.2311,
      "num_input_tokens_seen": 3085568,
      "step": 14630
    },
    {
      "epoch": 1.6100110011001099,
      "grad_norm": 0.003204345703125,
      "learning_rate": 0.02414851485148515,
      "loss": 0.2322,
      "num_input_tokens_seen": 3086592,
      "step": 14635
    },
    {
      "epoch": 1.6105610561056105,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.024156765676567656,
      "loss": 0.2338,
      "num_input_tokens_seen": 3087680,
      "step": 14640
    },
    {
      "epoch": 1.6111111111111112,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.024165016501650167,
      "loss": 0.2302,
      "num_input_tokens_seen": 3088736,
      "step": 14645
    },
    {
      "epoch": 1.6116611661166118,
      "grad_norm": 0.003570556640625,
      "learning_rate": 0.02417326732673267,
      "loss": 0.2297,
      "num_input_tokens_seen": 3089792,
      "step": 14650
    },
    {
      "epoch": 1.6122112211221122,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.02418151815181518,
      "loss": 0.2334,
      "num_input_tokens_seen": 3090816,
      "step": 14655
    },
    {
      "epoch": 1.6127612761276127,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.024189768976897688,
      "loss": 0.2239,
      "num_input_tokens_seen": 3091872,
      "step": 14660
    },
    {
      "epoch": 1.6133113311331133,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.0241980198019802,
      "loss": 0.2277,
      "num_input_tokens_seen": 3092960,
      "step": 14665
    },
    {
      "epoch": 1.613861386138614,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.024206270627062702,
      "loss": 0.2338,
      "num_input_tokens_seen": 3093984,
      "step": 14670
    },
    {
      "epoch": 1.6144114411441144,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.024214521452145213,
      "loss": 0.23,
      "num_input_tokens_seen": 3095008,
      "step": 14675
    },
    {
      "epoch": 1.6149614961496148,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.024222772277227723,
      "loss": 0.2323,
      "num_input_tokens_seen": 3096064,
      "step": 14680
    },
    {
      "epoch": 1.6155115511551155,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.02423102310231023,
      "loss": 0.2415,
      "num_input_tokens_seen": 3097088,
      "step": 14685
    },
    {
      "epoch": 1.6160616061606161,
      "grad_norm": 0.00775146484375,
      "learning_rate": 0.02423927392739274,
      "loss": 0.2309,
      "num_input_tokens_seen": 3098144,
      "step": 14690
    },
    {
      "epoch": 1.6166116611661168,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.024247524752475245,
      "loss": 0.2318,
      "num_input_tokens_seen": 3099168,
      "step": 14695
    },
    {
      "epoch": 1.6171617161716172,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.024255775577557755,
      "loss": 0.2296,
      "num_input_tokens_seen": 3100288,
      "step": 14700
    },
    {
      "epoch": 1.6177117711771176,
      "grad_norm": 0.0033111572265625,
      "learning_rate": 0.024264026402640262,
      "loss": 0.2286,
      "num_input_tokens_seen": 3101344,
      "step": 14705
    },
    {
      "epoch": 1.6182618261826183,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.024272277227722773,
      "loss": 0.2349,
      "num_input_tokens_seen": 3102400,
      "step": 14710
    },
    {
      "epoch": 1.618811881188119,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.024280528052805277,
      "loss": 0.2323,
      "num_input_tokens_seen": 3103488,
      "step": 14715
    },
    {
      "epoch": 1.6193619361936193,
      "grad_norm": 0.0030364990234375,
      "learning_rate": 0.024288778877887787,
      "loss": 0.2265,
      "num_input_tokens_seen": 3104512,
      "step": 14720
    },
    {
      "epoch": 1.6199119911991198,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.024297029702970298,
      "loss": 0.2355,
      "num_input_tokens_seen": 3105536,
      "step": 14725
    },
    {
      "epoch": 1.6204620462046204,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.024305280528052805,
      "loss": 0.2313,
      "num_input_tokens_seen": 3106528,
      "step": 14730
    },
    {
      "epoch": 1.621012101210121,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.024313531353135315,
      "loss": 0.2335,
      "num_input_tokens_seen": 3107584,
      "step": 14735
    },
    {
      "epoch": 1.6215621562156217,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.02432178217821782,
      "loss": 0.2391,
      "num_input_tokens_seen": 3108704,
      "step": 14740
    },
    {
      "epoch": 1.6221122112211221,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.02433003300330033,
      "loss": 0.2285,
      "num_input_tokens_seen": 3109760,
      "step": 14745
    },
    {
      "epoch": 1.6226622662266226,
      "grad_norm": 0.0179443359375,
      "learning_rate": 0.024338283828382837,
      "loss": 0.2316,
      "num_input_tokens_seen": 3110816,
      "step": 14750
    },
    {
      "epoch": 1.6232123212321232,
      "grad_norm": 0.0081787109375,
      "learning_rate": 0.024346534653465347,
      "loss": 0.2289,
      "num_input_tokens_seen": 3111808,
      "step": 14755
    },
    {
      "epoch": 1.6237623762376239,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.02435478547854785,
      "loss": 0.2331,
      "num_input_tokens_seen": 3112928,
      "step": 14760
    },
    {
      "epoch": 1.6243124312431243,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.02436303630363036,
      "loss": 0.2321,
      "num_input_tokens_seen": 3114048,
      "step": 14765
    },
    {
      "epoch": 1.6248624862486247,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.024371287128712872,
      "loss": 0.2253,
      "num_input_tokens_seen": 3115136,
      "step": 14770
    },
    {
      "epoch": 1.6254125412541254,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.02437953795379538,
      "loss": 0.2291,
      "num_input_tokens_seen": 3116192,
      "step": 14775
    },
    {
      "epoch": 1.625962596259626,
      "grad_norm": 0.0185546875,
      "learning_rate": 0.02438778877887789,
      "loss": 0.235,
      "num_input_tokens_seen": 3117248,
      "step": 14780
    },
    {
      "epoch": 1.6265126512651267,
      "grad_norm": 0.0030975341796875,
      "learning_rate": 0.024396039603960393,
      "loss": 0.235,
      "num_input_tokens_seen": 3118304,
      "step": 14785
    },
    {
      "epoch": 1.627062706270627,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.024404290429042904,
      "loss": 0.2241,
      "num_input_tokens_seen": 3119392,
      "step": 14790
    },
    {
      "epoch": 1.6276127612761275,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.02441254125412541,
      "loss": 0.2232,
      "num_input_tokens_seen": 3120544,
      "step": 14795
    },
    {
      "epoch": 1.6281628162816282,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.02442079207920792,
      "loss": 0.2308,
      "num_input_tokens_seen": 3121600,
      "step": 14800
    },
    {
      "epoch": 1.6287128712871288,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.024429042904290425,
      "loss": 0.2378,
      "num_input_tokens_seen": 3122656,
      "step": 14805
    },
    {
      "epoch": 1.6292629262926293,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.024437293729372936,
      "loss": 0.2318,
      "num_input_tokens_seen": 3123712,
      "step": 14810
    },
    {
      "epoch": 1.6298129812981297,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.024445544554455443,
      "loss": 0.2248,
      "num_input_tokens_seen": 3124736,
      "step": 14815
    },
    {
      "epoch": 1.6303630363036303,
      "grad_norm": 0.0026397705078125,
      "learning_rate": 0.024453795379537954,
      "loss": 0.2401,
      "num_input_tokens_seen": 3125728,
      "step": 14820
    },
    {
      "epoch": 1.630913091309131,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.024462046204620464,
      "loss": 0.2306,
      "num_input_tokens_seen": 3126816,
      "step": 14825
    },
    {
      "epoch": 1.6314631463146316,
      "grad_norm": 0.007781982421875,
      "learning_rate": 0.024470297029702968,
      "loss": 0.2321,
      "num_input_tokens_seen": 3127904,
      "step": 14830
    },
    {
      "epoch": 1.632013201320132,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.02447854785478548,
      "loss": 0.2369,
      "num_input_tokens_seen": 3128960,
      "step": 14835
    },
    {
      "epoch": 1.6325632563256325,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.024486798679867985,
      "loss": 0.2309,
      "num_input_tokens_seen": 3129984,
      "step": 14840
    },
    {
      "epoch": 1.6331133113311331,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.024495049504950496,
      "loss": 0.2398,
      "num_input_tokens_seen": 3131040,
      "step": 14845
    },
    {
      "epoch": 1.6336633663366338,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.024503300330033,
      "loss": 0.2254,
      "num_input_tokens_seen": 3132096,
      "step": 14850
    },
    {
      "epoch": 1.6342134213421342,
      "grad_norm": 0.00341796875,
      "learning_rate": 0.02451155115511551,
      "loss": 0.2312,
      "num_input_tokens_seen": 3133152,
      "step": 14855
    },
    {
      "epoch": 1.6347634763476346,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.024519801980198017,
      "loss": 0.2323,
      "num_input_tokens_seen": 3134272,
      "step": 14860
    },
    {
      "epoch": 1.6353135313531353,
      "grad_norm": 0.0030670166015625,
      "learning_rate": 0.024528052805280528,
      "loss": 0.2296,
      "num_input_tokens_seen": 3135360,
      "step": 14865
    },
    {
      "epoch": 1.635863586358636,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.02453630363036304,
      "loss": 0.2369,
      "num_input_tokens_seen": 3136480,
      "step": 14870
    },
    {
      "epoch": 1.6364136413641364,
      "grad_norm": 0.00775146484375,
      "learning_rate": 0.024544554455445542,
      "loss": 0.2291,
      "num_input_tokens_seen": 3137504,
      "step": 14875
    },
    {
      "epoch": 1.636963696369637,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.024552805280528053,
      "loss": 0.2337,
      "num_input_tokens_seen": 3138560,
      "step": 14880
    },
    {
      "epoch": 1.6375137513751374,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.02456105610561056,
      "loss": 0.2295,
      "num_input_tokens_seen": 3139648,
      "step": 14885
    },
    {
      "epoch": 1.638063806380638,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.02456930693069307,
      "loss": 0.23,
      "num_input_tokens_seen": 3140704,
      "step": 14890
    },
    {
      "epoch": 1.6386138613861387,
      "grad_norm": 0.0167236328125,
      "learning_rate": 0.024577557755775574,
      "loss": 0.2356,
      "num_input_tokens_seen": 3141760,
      "step": 14895
    },
    {
      "epoch": 1.6391639163916392,
      "grad_norm": 0.00927734375,
      "learning_rate": 0.024585808580858085,
      "loss": 0.2335,
      "num_input_tokens_seen": 3142752,
      "step": 14900
    },
    {
      "epoch": 1.6397139713971396,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.02459405940594059,
      "loss": 0.2319,
      "num_input_tokens_seen": 3143808,
      "step": 14905
    },
    {
      "epoch": 1.6402640264026402,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.024602310231023102,
      "loss": 0.2314,
      "num_input_tokens_seen": 3144864,
      "step": 14910
    },
    {
      "epoch": 1.640814081408141,
      "grad_norm": 0.0084228515625,
      "learning_rate": 0.024610561056105613,
      "loss": 0.2303,
      "num_input_tokens_seen": 3145888,
      "step": 14915
    },
    {
      "epoch": 1.6413641364136413,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.024618811881188116,
      "loss": 0.232,
      "num_input_tokens_seen": 3146880,
      "step": 14920
    },
    {
      "epoch": 1.641914191419142,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.024627062706270627,
      "loss": 0.2311,
      "num_input_tokens_seen": 3147936,
      "step": 14925
    },
    {
      "epoch": 1.6424642464246424,
      "grad_norm": 0.007568359375,
      "learning_rate": 0.024635313531353134,
      "loss": 0.2313,
      "num_input_tokens_seen": 3149088,
      "step": 14930
    },
    {
      "epoch": 1.643014301430143,
      "grad_norm": 0.002716064453125,
      "learning_rate": 0.024643564356435645,
      "loss": 0.2313,
      "num_input_tokens_seen": 3150176,
      "step": 14935
    },
    {
      "epoch": 1.6435643564356437,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.02465181518151815,
      "loss": 0.2381,
      "num_input_tokens_seen": 3151328,
      "step": 14940
    },
    {
      "epoch": 1.6441144114411441,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.02466006600660066,
      "loss": 0.2329,
      "num_input_tokens_seen": 3152384,
      "step": 14945
    },
    {
      "epoch": 1.6446644664466445,
      "grad_norm": 0.015625,
      "learning_rate": 0.024668316831683166,
      "loss": 0.2333,
      "num_input_tokens_seen": 3153504,
      "step": 14950
    },
    {
      "epoch": 1.6452145214521452,
      "grad_norm": 0.0076904296875,
      "learning_rate": 0.024676567656765677,
      "loss": 0.2331,
      "num_input_tokens_seen": 3154560,
      "step": 14955
    },
    {
      "epoch": 1.6457645764576458,
      "grad_norm": 0.007781982421875,
      "learning_rate": 0.024684818481848187,
      "loss": 0.2284,
      "num_input_tokens_seen": 3155616,
      "step": 14960
    },
    {
      "epoch": 1.6463146314631463,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.02469306930693069,
      "loss": 0.23,
      "num_input_tokens_seen": 3156704,
      "step": 14965
    },
    {
      "epoch": 1.6468646864686467,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.0247013201320132,
      "loss": 0.2301,
      "num_input_tokens_seen": 3157728,
      "step": 14970
    },
    {
      "epoch": 1.6474147414741473,
      "grad_norm": 0.007476806640625,
      "learning_rate": 0.02470957095709571,
      "loss": 0.2324,
      "num_input_tokens_seen": 3158752,
      "step": 14975
    },
    {
      "epoch": 1.647964796479648,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.02471782178217822,
      "loss": 0.2291,
      "num_input_tokens_seen": 3159776,
      "step": 14980
    },
    {
      "epoch": 1.6485148514851486,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.024726072607260723,
      "loss": 0.2266,
      "num_input_tokens_seen": 3160832,
      "step": 14985
    },
    {
      "epoch": 1.649064906490649,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.024734323432343233,
      "loss": 0.2324,
      "num_input_tokens_seen": 3161920,
      "step": 14990
    },
    {
      "epoch": 1.6496149614961495,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.02474257425742574,
      "loss": 0.2303,
      "num_input_tokens_seen": 3163072,
      "step": 14995
    },
    {
      "epoch": 1.6501650165016502,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.02475082508250825,
      "loss": 0.2435,
      "num_input_tokens_seen": 3164160,
      "step": 15000
    },
    {
      "epoch": 1.6507150715071508,
      "grad_norm": 0.01513671875,
      "learning_rate": 0.02475907590759076,
      "loss": 0.2215,
      "num_input_tokens_seen": 3165216,
      "step": 15005
    },
    {
      "epoch": 1.6512651265126512,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.024767326732673265,
      "loss": 0.2303,
      "num_input_tokens_seen": 3166304,
      "step": 15010
    },
    {
      "epoch": 1.6518151815181517,
      "grad_norm": 0.0152587890625,
      "learning_rate": 0.024775577557755776,
      "loss": 0.2308,
      "num_input_tokens_seen": 3167360,
      "step": 15015
    },
    {
      "epoch": 1.6523652365236523,
      "grad_norm": 0.007720947265625,
      "learning_rate": 0.024783828382838283,
      "loss": 0.2396,
      "num_input_tokens_seen": 3168480,
      "step": 15020
    },
    {
      "epoch": 1.652915291529153,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.024792079207920793,
      "loss": 0.2384,
      "num_input_tokens_seen": 3169568,
      "step": 15025
    },
    {
      "epoch": 1.6534653465346536,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.024800330033003297,
      "loss": 0.2304,
      "num_input_tokens_seen": 3170592,
      "step": 15030
    },
    {
      "epoch": 1.654015401540154,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.024808580858085807,
      "loss": 0.2319,
      "num_input_tokens_seen": 3171648,
      "step": 15035
    },
    {
      "epoch": 1.6545654565456545,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.024816831683168315,
      "loss": 0.2283,
      "num_input_tokens_seen": 3172736,
      "step": 15040
    },
    {
      "epoch": 1.655115511551155,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.024825082508250825,
      "loss": 0.232,
      "num_input_tokens_seen": 3173792,
      "step": 15045
    },
    {
      "epoch": 1.6556655665566558,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.024833333333333332,
      "loss": 0.231,
      "num_input_tokens_seen": 3174880,
      "step": 15050
    },
    {
      "epoch": 1.6562156215621562,
      "grad_norm": 0.002838134765625,
      "learning_rate": 0.02484158415841584,
      "loss": 0.2314,
      "num_input_tokens_seen": 3175968,
      "step": 15055
    },
    {
      "epoch": 1.6567656765676566,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.02484983498349835,
      "loss": 0.2367,
      "num_input_tokens_seen": 3177056,
      "step": 15060
    },
    {
      "epoch": 1.6573157315731573,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.024858085808580857,
      "loss": 0.2341,
      "num_input_tokens_seen": 3178144,
      "step": 15065
    },
    {
      "epoch": 1.657865786578658,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.024866336633663368,
      "loss": 0.2303,
      "num_input_tokens_seen": 3179168,
      "step": 15070
    },
    {
      "epoch": 1.6584158415841586,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.02487458745874587,
      "loss": 0.2283,
      "num_input_tokens_seen": 3180224,
      "step": 15075
    },
    {
      "epoch": 1.658965896589659,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.024882838283828382,
      "loss": 0.2304,
      "num_input_tokens_seen": 3181248,
      "step": 15080
    },
    {
      "epoch": 1.6595159515951594,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.02489108910891089,
      "loss": 0.2316,
      "num_input_tokens_seen": 3182336,
      "step": 15085
    },
    {
      "epoch": 1.66006600660066,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.0248993399339934,
      "loss": 0.2255,
      "num_input_tokens_seen": 3183392,
      "step": 15090
    },
    {
      "epoch": 1.6606160616061607,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.024907590759075907,
      "loss": 0.2257,
      "num_input_tokens_seen": 3184480,
      "step": 15095
    },
    {
      "epoch": 1.6611661166116611,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.024915841584158414,
      "loss": 0.2361,
      "num_input_tokens_seen": 3185472,
      "step": 15100
    },
    {
      "epoch": 1.6617161716171616,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.024924092409240924,
      "loss": 0.2302,
      "num_input_tokens_seen": 3186528,
      "step": 15105
    },
    {
      "epoch": 1.6622662266226622,
      "grad_norm": 0.00335693359375,
      "learning_rate": 0.02493234323432343,
      "loss": 0.2335,
      "num_input_tokens_seen": 3187520,
      "step": 15110
    },
    {
      "epoch": 1.6628162816281629,
      "grad_norm": 0.003173828125,
      "learning_rate": 0.024940594059405942,
      "loss": 0.2103,
      "num_input_tokens_seen": 3188544,
      "step": 15115
    },
    {
      "epoch": 1.6633663366336635,
      "grad_norm": 0.003387451171875,
      "learning_rate": 0.024948844884488446,
      "loss": 0.2402,
      "num_input_tokens_seen": 3189632,
      "step": 15120
    },
    {
      "epoch": 1.663916391639164,
      "grad_norm": 0.0030364990234375,
      "learning_rate": 0.024957095709570956,
      "loss": 0.235,
      "num_input_tokens_seen": 3190688,
      "step": 15125
    },
    {
      "epoch": 1.6644664466446644,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.024965346534653463,
      "loss": 0.2453,
      "num_input_tokens_seen": 3191712,
      "step": 15130
    },
    {
      "epoch": 1.665016501650165,
      "grad_norm": 0.002960205078125,
      "learning_rate": 0.024973597359735974,
      "loss": 0.2284,
      "num_input_tokens_seen": 3192736,
      "step": 15135
    },
    {
      "epoch": 1.6655665566556657,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.02498184818481848,
      "loss": 0.2241,
      "num_input_tokens_seen": 3193728,
      "step": 15140
    },
    {
      "epoch": 1.666116611661166,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.024990099009900988,
      "loss": 0.2392,
      "num_input_tokens_seen": 3194784,
      "step": 15145
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.00830078125,
      "learning_rate": 0.0249983498349835,
      "loss": 0.2318,
      "num_input_tokens_seen": 3195840,
      "step": 15150
    },
    {
      "epoch": 1.6672167216721672,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.025006600660066006,
      "loss": 0.2402,
      "num_input_tokens_seen": 3196928,
      "step": 15155
    },
    {
      "epoch": 1.6677667766776678,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.025014851485148516,
      "loss": 0.2427,
      "num_input_tokens_seen": 3198016,
      "step": 15160
    },
    {
      "epoch": 1.6683168316831685,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.02502310231023102,
      "loss": 0.2273,
      "num_input_tokens_seen": 3199072,
      "step": 15165
    },
    {
      "epoch": 1.668866886688669,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.02503135313531353,
      "loss": 0.2324,
      "num_input_tokens_seen": 3200128,
      "step": 15170
    },
    {
      "epoch": 1.6694169416941693,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.025039603960396038,
      "loss": 0.2314,
      "num_input_tokens_seen": 3201120,
      "step": 15175
    },
    {
      "epoch": 1.66996699669967,
      "grad_norm": 0.015869140625,
      "learning_rate": 0.025047854785478548,
      "loss": 0.2319,
      "num_input_tokens_seen": 3202112,
      "step": 15180
    },
    {
      "epoch": 1.6705170517051706,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.025056105610561055,
      "loss": 0.2314,
      "num_input_tokens_seen": 3203136,
      "step": 15185
    },
    {
      "epoch": 1.671067106710671,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.025064356435643562,
      "loss": 0.2345,
      "num_input_tokens_seen": 3204224,
      "step": 15190
    },
    {
      "epoch": 1.6716171617161715,
      "grad_norm": 0.0081787109375,
      "learning_rate": 0.025072607260726073,
      "loss": 0.2314,
      "num_input_tokens_seen": 3205344,
      "step": 15195
    },
    {
      "epoch": 1.6721672167216721,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.02508085808580858,
      "loss": 0.2314,
      "num_input_tokens_seen": 3206336,
      "step": 15200
    },
    {
      "epoch": 1.6727172717271728,
      "grad_norm": 0.0081787109375,
      "learning_rate": 0.02508910891089109,
      "loss": 0.2319,
      "num_input_tokens_seen": 3207456,
      "step": 15205
    },
    {
      "epoch": 1.6732673267326734,
      "grad_norm": 0.0081787109375,
      "learning_rate": 0.025097359735973594,
      "loss": 0.2298,
      "num_input_tokens_seen": 3208480,
      "step": 15210
    },
    {
      "epoch": 1.6738173817381738,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.025105610561056105,
      "loss": 0.2314,
      "num_input_tokens_seen": 3209536,
      "step": 15215
    },
    {
      "epoch": 1.6743674367436743,
      "grad_norm": 0.016357421875,
      "learning_rate": 0.025113861386138612,
      "loss": 0.2313,
      "num_input_tokens_seen": 3210592,
      "step": 15220
    },
    {
      "epoch": 1.674917491749175,
      "grad_norm": 0.0084228515625,
      "learning_rate": 0.025122112211221122,
      "loss": 0.2292,
      "num_input_tokens_seen": 3211584,
      "step": 15225
    },
    {
      "epoch": 1.6754675467546756,
      "grad_norm": 0.016357421875,
      "learning_rate": 0.02513036303630363,
      "loss": 0.2299,
      "num_input_tokens_seen": 3212640,
      "step": 15230
    },
    {
      "epoch": 1.676017601760176,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.025138613861386137,
      "loss": 0.2283,
      "num_input_tokens_seen": 3213664,
      "step": 15235
    },
    {
      "epoch": 1.6765676567656764,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.025146864686468647,
      "loss": 0.232,
      "num_input_tokens_seen": 3214784,
      "step": 15240
    },
    {
      "epoch": 1.677117711771177,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.025155115511551154,
      "loss": 0.2337,
      "num_input_tokens_seen": 3215872,
      "step": 15245
    },
    {
      "epoch": 1.6776677667766777,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.025163366336633665,
      "loss": 0.2326,
      "num_input_tokens_seen": 3216896,
      "step": 15250
    },
    {
      "epoch": 1.6782178217821784,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.02517161716171617,
      "loss": 0.2315,
      "num_input_tokens_seen": 3217920,
      "step": 15255
    },
    {
      "epoch": 1.6787678767876788,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.02517986798679868,
      "loss": 0.2299,
      "num_input_tokens_seen": 3218944,
      "step": 15260
    },
    {
      "epoch": 1.6793179317931792,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.025188118811881186,
      "loss": 0.231,
      "num_input_tokens_seen": 3219904,
      "step": 15265
    },
    {
      "epoch": 1.6798679867986799,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.025196369636963697,
      "loss": 0.2346,
      "num_input_tokens_seen": 3221056,
      "step": 15270
    },
    {
      "epoch": 1.6804180418041805,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.025204620462046204,
      "loss": 0.2294,
      "num_input_tokens_seen": 3222080,
      "step": 15275
    },
    {
      "epoch": 1.680968096809681,
      "grad_norm": 0.0157470703125,
      "learning_rate": 0.02521287128712871,
      "loss": 0.2361,
      "num_input_tokens_seen": 3223232,
      "step": 15280
    },
    {
      "epoch": 1.6815181518151814,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.025221122112211218,
      "loss": 0.2313,
      "num_input_tokens_seen": 3224256,
      "step": 15285
    },
    {
      "epoch": 1.682068206820682,
      "grad_norm": 0.015625,
      "learning_rate": 0.02522937293729373,
      "loss": 0.2298,
      "num_input_tokens_seen": 3225312,
      "step": 15290
    },
    {
      "epoch": 1.6826182618261827,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.02523762376237624,
      "loss": 0.2319,
      "num_input_tokens_seen": 3226368,
      "step": 15295
    },
    {
      "epoch": 1.6831683168316833,
      "grad_norm": 0.007720947265625,
      "learning_rate": 0.025245874587458743,
      "loss": 0.2314,
      "num_input_tokens_seen": 3227424,
      "step": 15300
    },
    {
      "epoch": 1.6837183718371838,
      "grad_norm": 0.0074462890625,
      "learning_rate": 0.025254125412541253,
      "loss": 0.2314,
      "num_input_tokens_seen": 3228512,
      "step": 15305
    },
    {
      "epoch": 1.6842684268426842,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.02526237623762376,
      "loss": 0.2284,
      "num_input_tokens_seen": 3229504,
      "step": 15310
    },
    {
      "epoch": 1.6848184818481848,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.02527062706270627,
      "loss": 0.2295,
      "num_input_tokens_seen": 3230560,
      "step": 15315
    },
    {
      "epoch": 1.6853685368536855,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.025278877887788778,
      "loss": 0.2327,
      "num_input_tokens_seen": 3231648,
      "step": 15320
    },
    {
      "epoch": 1.685918591859186,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.025287128712871285,
      "loss": 0.2326,
      "num_input_tokens_seen": 3232704,
      "step": 15325
    },
    {
      "epoch": 1.6864686468646863,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.025295379537953792,
      "loss": 0.2342,
      "num_input_tokens_seen": 3233824,
      "step": 15330
    },
    {
      "epoch": 1.687018701870187,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.025303630363036303,
      "loss": 0.23,
      "num_input_tokens_seen": 3234848,
      "step": 15335
    },
    {
      "epoch": 1.6875687568756876,
      "grad_norm": 0.007598876953125,
      "learning_rate": 0.025311881188118814,
      "loss": 0.2315,
      "num_input_tokens_seen": 3235968,
      "step": 15340
    },
    {
      "epoch": 1.688118811881188,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.025320132013201317,
      "loss": 0.2289,
      "num_input_tokens_seen": 3237024,
      "step": 15345
    },
    {
      "epoch": 1.6886688668866887,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.025328382838283828,
      "loss": 0.2332,
      "num_input_tokens_seen": 3238112,
      "step": 15350
    },
    {
      "epoch": 1.6892189218921891,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.025336633663366335,
      "loss": 0.2294,
      "num_input_tokens_seen": 3239136,
      "step": 15355
    },
    {
      "epoch": 1.6897689768976898,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.025344884488448845,
      "loss": 0.2332,
      "num_input_tokens_seen": 3240160,
      "step": 15360
    },
    {
      "epoch": 1.6903190319031904,
      "grad_norm": 0.015380859375,
      "learning_rate": 0.025353135313531353,
      "loss": 0.2301,
      "num_input_tokens_seen": 3241280,
      "step": 15365
    },
    {
      "epoch": 1.6908690869086909,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.02536138613861386,
      "loss": 0.2311,
      "num_input_tokens_seen": 3242304,
      "step": 15370
    },
    {
      "epoch": 1.6914191419141913,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.025369636963696367,
      "loss": 0.2322,
      "num_input_tokens_seen": 3243328,
      "step": 15375
    },
    {
      "epoch": 1.691969196919692,
      "grad_norm": 0.007415771484375,
      "learning_rate": 0.025377887788778877,
      "loss": 0.2264,
      "num_input_tokens_seen": 3244384,
      "step": 15380
    },
    {
      "epoch": 1.6925192519251926,
      "grad_norm": 0.0028533935546875,
      "learning_rate": 0.025386138613861388,
      "loss": 0.2316,
      "num_input_tokens_seen": 3245472,
      "step": 15385
    },
    {
      "epoch": 1.693069306930693,
      "grad_norm": 0.0167236328125,
      "learning_rate": 0.02539438943894389,
      "loss": 0.2405,
      "num_input_tokens_seen": 3246528,
      "step": 15390
    },
    {
      "epoch": 1.6936193619361937,
      "grad_norm": 0.007598876953125,
      "learning_rate": 0.025402640264026402,
      "loss": 0.2289,
      "num_input_tokens_seen": 3247616,
      "step": 15395
    },
    {
      "epoch": 1.694169416941694,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.02541089108910891,
      "loss": 0.2324,
      "num_input_tokens_seen": 3248736,
      "step": 15400
    },
    {
      "epoch": 1.6947194719471947,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.02541914191419142,
      "loss": 0.2324,
      "num_input_tokens_seen": 3249824,
      "step": 15405
    },
    {
      "epoch": 1.6952695269526954,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.025427392739273927,
      "loss": 0.2329,
      "num_input_tokens_seen": 3250880,
      "step": 15410
    },
    {
      "epoch": 1.6958195819581958,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.025435643564356434,
      "loss": 0.2283,
      "num_input_tokens_seen": 3251872,
      "step": 15415
    },
    {
      "epoch": 1.6963696369636962,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.02544389438943894,
      "loss": 0.2325,
      "num_input_tokens_seen": 3252928,
      "step": 15420
    },
    {
      "epoch": 1.696919691969197,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.02545214521452145,
      "loss": 0.2309,
      "num_input_tokens_seen": 3253984,
      "step": 15425
    },
    {
      "epoch": 1.6974697469746975,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.025460396039603962,
      "loss": 0.2294,
      "num_input_tokens_seen": 3255008,
      "step": 15430
    },
    {
      "epoch": 1.698019801980198,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.025468646864686466,
      "loss": 0.2304,
      "num_input_tokens_seen": 3256096,
      "step": 15435
    },
    {
      "epoch": 1.6985698569856986,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.025476897689768976,
      "loss": 0.2332,
      "num_input_tokens_seen": 3257184,
      "step": 15440
    },
    {
      "epoch": 1.699119911991199,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.025485148514851483,
      "loss": 0.2286,
      "num_input_tokens_seen": 3258208,
      "step": 15445
    },
    {
      "epoch": 1.6996699669966997,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.025493399339933994,
      "loss": 0.2323,
      "num_input_tokens_seen": 3259328,
      "step": 15450
    },
    {
      "epoch": 1.7002200220022003,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.0255016501650165,
      "loss": 0.2316,
      "num_input_tokens_seen": 3260416,
      "step": 15455
    },
    {
      "epoch": 1.7007700770077008,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.02550990099009901,
      "loss": 0.2317,
      "num_input_tokens_seen": 3261536,
      "step": 15460
    },
    {
      "epoch": 1.7013201320132012,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.025518151815181515,
      "loss": 0.2349,
      "num_input_tokens_seen": 3262720,
      "step": 15465
    },
    {
      "epoch": 1.7018701870187019,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.025526402640264026,
      "loss": 0.2332,
      "num_input_tokens_seen": 3263808,
      "step": 15470
    },
    {
      "epoch": 1.7024202420242025,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.025534653465346537,
      "loss": 0.232,
      "num_input_tokens_seen": 3264800,
      "step": 15475
    },
    {
      "epoch": 1.702970297029703,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.02554290429042904,
      "loss": 0.2335,
      "num_input_tokens_seen": 3265952,
      "step": 15480
    },
    {
      "epoch": 1.7035203520352034,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.02555115511551155,
      "loss": 0.2335,
      "num_input_tokens_seen": 3266976,
      "step": 15485
    },
    {
      "epoch": 1.704070407040704,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.025559405940594058,
      "loss": 0.2309,
      "num_input_tokens_seen": 3267968,
      "step": 15490
    },
    {
      "epoch": 1.7046204620462047,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.02556765676567657,
      "loss": 0.2324,
      "num_input_tokens_seen": 3269056,
      "step": 15495
    },
    {
      "epoch": 1.7051705170517053,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.025575907590759075,
      "loss": 0.2308,
      "num_input_tokens_seen": 3270080,
      "step": 15500
    },
    {
      "epoch": 1.7057205720572057,
      "grad_norm": 0.0081787109375,
      "learning_rate": 0.025584158415841583,
      "loss": 0.2314,
      "num_input_tokens_seen": 3271104,
      "step": 15505
    },
    {
      "epoch": 1.7062706270627062,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.02559240924092409,
      "loss": 0.2298,
      "num_input_tokens_seen": 3272128,
      "step": 15510
    },
    {
      "epoch": 1.7068206820682068,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.0256006600660066,
      "loss": 0.2304,
      "num_input_tokens_seen": 3273248,
      "step": 15515
    },
    {
      "epoch": 1.7073707370737075,
      "grad_norm": 0.002593994140625,
      "learning_rate": 0.02560891089108911,
      "loss": 0.233,
      "num_input_tokens_seen": 3274272,
      "step": 15520
    },
    {
      "epoch": 1.7079207920792079,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.025617161716171614,
      "loss": 0.2319,
      "num_input_tokens_seen": 3275232,
      "step": 15525
    },
    {
      "epoch": 1.7084708470847083,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.025625412541254125,
      "loss": 0.2313,
      "num_input_tokens_seen": 3276256,
      "step": 15530
    },
    {
      "epoch": 1.709020902090209,
      "grad_norm": 0.0078125,
      "learning_rate": 0.025633663366336632,
      "loss": 0.2324,
      "num_input_tokens_seen": 3277280,
      "step": 15535
    },
    {
      "epoch": 1.7095709570957096,
      "grad_norm": 0.015869140625,
      "learning_rate": 0.025641914191419143,
      "loss": 0.2324,
      "num_input_tokens_seen": 3278272,
      "step": 15540
    },
    {
      "epoch": 1.7101210121012103,
      "grad_norm": 0.0155029296875,
      "learning_rate": 0.02565016501650165,
      "loss": 0.2319,
      "num_input_tokens_seen": 3279360,
      "step": 15545
    },
    {
      "epoch": 1.7106710671067107,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.025658415841584157,
      "loss": 0.2319,
      "num_input_tokens_seen": 3280448,
      "step": 15550
    },
    {
      "epoch": 1.7112211221122111,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.025666666666666664,
      "loss": 0.2319,
      "num_input_tokens_seen": 3281472,
      "step": 15555
    },
    {
      "epoch": 1.7117711771177118,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.025674917491749175,
      "loss": 0.233,
      "num_input_tokens_seen": 3282528,
      "step": 15560
    },
    {
      "epoch": 1.7123212321232124,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.02568316831683168,
      "loss": 0.2332,
      "num_input_tokens_seen": 3283584,
      "step": 15565
    },
    {
      "epoch": 1.7128712871287128,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.02569141914191419,
      "loss": 0.2313,
      "num_input_tokens_seen": 3284608,
      "step": 15570
    },
    {
      "epoch": 1.7134213421342133,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.0256996699669967,
      "loss": 0.236,
      "num_input_tokens_seen": 3285600,
      "step": 15575
    },
    {
      "epoch": 1.713971397139714,
      "grad_norm": 0.01458740234375,
      "learning_rate": 0.025707920792079206,
      "loss": 0.2352,
      "num_input_tokens_seen": 3286688,
      "step": 15580
    },
    {
      "epoch": 1.7145214521452146,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.025716171617161717,
      "loss": 0.2315,
      "num_input_tokens_seen": 3287808,
      "step": 15585
    },
    {
      "epoch": 1.7150715071507152,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.025724422442244224,
      "loss": 0.231,
      "num_input_tokens_seen": 3288864,
      "step": 15590
    },
    {
      "epoch": 1.7156215621562156,
      "grad_norm": 0.007781982421875,
      "learning_rate": 0.02573267326732673,
      "loss": 0.232,
      "num_input_tokens_seen": 3289920,
      "step": 15595
    },
    {
      "epoch": 1.716171617161716,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.02574092409240924,
      "loss": 0.233,
      "num_input_tokens_seen": 3290976,
      "step": 15600
    },
    {
      "epoch": 1.7167216721672167,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.02574917491749175,
      "loss": 0.2288,
      "num_input_tokens_seen": 3292000,
      "step": 15605
    },
    {
      "epoch": 1.7172717271727174,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.025757425742574256,
      "loss": 0.2324,
      "num_input_tokens_seen": 3293056,
      "step": 15610
    },
    {
      "epoch": 1.7178217821782178,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.025765676567656763,
      "loss": 0.2325,
      "num_input_tokens_seen": 3294080,
      "step": 15615
    },
    {
      "epoch": 1.7183718371837182,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.025773927392739274,
      "loss": 0.2314,
      "num_input_tokens_seen": 3295136,
      "step": 15620
    },
    {
      "epoch": 1.7189218921892189,
      "grad_norm": 0.01495361328125,
      "learning_rate": 0.02578217821782178,
      "loss": 0.233,
      "num_input_tokens_seen": 3296288,
      "step": 15625
    },
    {
      "epoch": 1.7194719471947195,
      "grad_norm": 0.01519775390625,
      "learning_rate": 0.02579042904290429,
      "loss": 0.2319,
      "num_input_tokens_seen": 3297344,
      "step": 15630
    },
    {
      "epoch": 1.7200220022002202,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0257986798679868,
      "loss": 0.2329,
      "num_input_tokens_seen": 3298432,
      "step": 15635
    },
    {
      "epoch": 1.7205720572057206,
      "grad_norm": 0.01531982421875,
      "learning_rate": 0.025806930693069306,
      "loss": 0.2324,
      "num_input_tokens_seen": 3299424,
      "step": 15640
    },
    {
      "epoch": 1.721122112211221,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.025815181518151813,
      "loss": 0.2313,
      "num_input_tokens_seen": 3300480,
      "step": 15645
    },
    {
      "epoch": 1.7216721672167217,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.025823432343234323,
      "loss": 0.2314,
      "num_input_tokens_seen": 3301504,
      "step": 15650
    },
    {
      "epoch": 1.7222222222222223,
      "grad_norm": 0.007415771484375,
      "learning_rate": 0.02583168316831683,
      "loss": 0.2303,
      "num_input_tokens_seen": 3302656,
      "step": 15655
    },
    {
      "epoch": 1.7227722772277227,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.025839933993399337,
      "loss": 0.2335,
      "num_input_tokens_seen": 3303712,
      "step": 15660
    },
    {
      "epoch": 1.7233223322332232,
      "grad_norm": 0.0076904296875,
      "learning_rate": 0.025848184818481848,
      "loss": 0.2319,
      "num_input_tokens_seen": 3304800,
      "step": 15665
    },
    {
      "epoch": 1.7238723872387238,
      "grad_norm": 0.007476806640625,
      "learning_rate": 0.025856435643564355,
      "loss": 0.2303,
      "num_input_tokens_seen": 3305792,
      "step": 15670
    },
    {
      "epoch": 1.7244224422442245,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.025864686468646866,
      "loss": 0.2288,
      "num_input_tokens_seen": 3306848,
      "step": 15675
    },
    {
      "epoch": 1.7249724972497251,
      "grad_norm": 0.00897216796875,
      "learning_rate": 0.025872937293729373,
      "loss": 0.24,
      "num_input_tokens_seen": 3307872,
      "step": 15680
    },
    {
      "epoch": 1.7255225522552256,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.02588118811881188,
      "loss": 0.2354,
      "num_input_tokens_seen": 3308928,
      "step": 15685
    },
    {
      "epoch": 1.726072607260726,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.025889438943894387,
      "loss": 0.2307,
      "num_input_tokens_seen": 3309984,
      "step": 15690
    },
    {
      "epoch": 1.7266226622662266,
      "grad_norm": 0.01458740234375,
      "learning_rate": 0.025897689768976898,
      "loss": 0.2225,
      "num_input_tokens_seen": 3311040,
      "step": 15695
    },
    {
      "epoch": 1.7271727172717273,
      "grad_norm": 0.00726318359375,
      "learning_rate": 0.025905940594059405,
      "loss": 0.2308,
      "num_input_tokens_seen": 3312192,
      "step": 15700
    },
    {
      "epoch": 1.7277227722772277,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.025914191419141912,
      "loss": 0.2382,
      "num_input_tokens_seen": 3313216,
      "step": 15705
    },
    {
      "epoch": 1.7282728272827281,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.025922442244224422,
      "loss": 0.2335,
      "num_input_tokens_seen": 3314240,
      "step": 15710
    },
    {
      "epoch": 1.7288228822882288,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.02593069306930693,
      "loss": 0.2318,
      "num_input_tokens_seen": 3315296,
      "step": 15715
    },
    {
      "epoch": 1.7293729372937294,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.02593894389438944,
      "loss": 0.2365,
      "num_input_tokens_seen": 3316288,
      "step": 15720
    },
    {
      "epoch": 1.72992299229923,
      "grad_norm": 0.015625,
      "learning_rate": 0.025947194719471947,
      "loss": 0.2337,
      "num_input_tokens_seen": 3317376,
      "step": 15725
    },
    {
      "epoch": 1.7304730473047305,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.025955445544554454,
      "loss": 0.2279,
      "num_input_tokens_seen": 3318464,
      "step": 15730
    },
    {
      "epoch": 1.731023102310231,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.02596369636963696,
      "loss": 0.2301,
      "num_input_tokens_seen": 3319520,
      "step": 15735
    },
    {
      "epoch": 1.7315731573157316,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.025971947194719472,
      "loss": 0.2321,
      "num_input_tokens_seen": 3320576,
      "step": 15740
    },
    {
      "epoch": 1.7321232123212322,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.02598019801980198,
      "loss": 0.2325,
      "num_input_tokens_seen": 3321536,
      "step": 15745
    },
    {
      "epoch": 1.7326732673267327,
      "grad_norm": 0.0146484375,
      "learning_rate": 0.025988448844884486,
      "loss": 0.2311,
      "num_input_tokens_seen": 3322592,
      "step": 15750
    },
    {
      "epoch": 1.733223322332233,
      "grad_norm": 0.0078125,
      "learning_rate": 0.025996699669966997,
      "loss": 0.2362,
      "num_input_tokens_seen": 3323712,
      "step": 15755
    },
    {
      "epoch": 1.7337733773377337,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.026004950495049504,
      "loss": 0.2299,
      "num_input_tokens_seen": 3324768,
      "step": 15760
    },
    {
      "epoch": 1.7343234323432344,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.026013201320132014,
      "loss": 0.2304,
      "num_input_tokens_seen": 3325824,
      "step": 15765
    },
    {
      "epoch": 1.734873487348735,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.02602145214521452,
      "loss": 0.2304,
      "num_input_tokens_seen": 3326880,
      "step": 15770
    },
    {
      "epoch": 1.7354235423542355,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.02602970297029703,
      "loss": 0.2335,
      "num_input_tokens_seen": 3327904,
      "step": 15775
    },
    {
      "epoch": 1.7359735973597359,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.026037953795379536,
      "loss": 0.2325,
      "num_input_tokens_seen": 3328960,
      "step": 15780
    },
    {
      "epoch": 1.7365236523652365,
      "grad_norm": 0.0072021484375,
      "learning_rate": 0.026046204620462046,
      "loss": 0.2309,
      "num_input_tokens_seen": 3330016,
      "step": 15785
    },
    {
      "epoch": 1.7370737073707372,
      "grad_norm": 0.007598876953125,
      "learning_rate": 0.026054455445544553,
      "loss": 0.2309,
      "num_input_tokens_seen": 3331040,
      "step": 15790
    },
    {
      "epoch": 1.7376237623762376,
      "grad_norm": 0.00726318359375,
      "learning_rate": 0.02606270627062706,
      "loss": 0.2299,
      "num_input_tokens_seen": 3332128,
      "step": 15795
    },
    {
      "epoch": 1.738173817381738,
      "grad_norm": 0.0147705078125,
      "learning_rate": 0.026070957095709568,
      "loss": 0.2325,
      "num_input_tokens_seen": 3333216,
      "step": 15800
    },
    {
      "epoch": 1.7387238723872387,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.026079207920792078,
      "loss": 0.232,
      "num_input_tokens_seen": 3334240,
      "step": 15805
    },
    {
      "epoch": 1.7392739273927393,
      "grad_norm": 0.007537841796875,
      "learning_rate": 0.02608745874587459,
      "loss": 0.2315,
      "num_input_tokens_seen": 3335328,
      "step": 15810
    },
    {
      "epoch": 1.73982398239824,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.026095709570957096,
      "loss": 0.2284,
      "num_input_tokens_seen": 3336288,
      "step": 15815
    },
    {
      "epoch": 1.7403740374037404,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.026103960396039603,
      "loss": 0.2295,
      "num_input_tokens_seen": 3337344,
      "step": 15820
    },
    {
      "epoch": 1.7409240924092408,
      "grad_norm": 0.007415771484375,
      "learning_rate": 0.02611221122112211,
      "loss": 0.2348,
      "num_input_tokens_seen": 3338368,
      "step": 15825
    },
    {
      "epoch": 1.7414741474147415,
      "grad_norm": 0.00384521484375,
      "learning_rate": 0.02612046204620462,
      "loss": 0.2322,
      "num_input_tokens_seen": 3339424,
      "step": 15830
    },
    {
      "epoch": 1.7420242024202421,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.026128712871287128,
      "loss": 0.2312,
      "num_input_tokens_seen": 3340512,
      "step": 15835
    },
    {
      "epoch": 1.7425742574257426,
      "grad_norm": 0.014892578125,
      "learning_rate": 0.026136963696369635,
      "loss": 0.2291,
      "num_input_tokens_seen": 3341568,
      "step": 15840
    },
    {
      "epoch": 1.743124312431243,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.026145214521452142,
      "loss": 0.2342,
      "num_input_tokens_seen": 3342688,
      "step": 15845
    },
    {
      "epoch": 1.7436743674367436,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.026153465346534652,
      "loss": 0.2302,
      "num_input_tokens_seen": 3343680,
      "step": 15850
    },
    {
      "epoch": 1.7442244224422443,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.026161716171617163,
      "loss": 0.2373,
      "num_input_tokens_seen": 3344736,
      "step": 15855
    },
    {
      "epoch": 1.7447744774477447,
      "grad_norm": 0.01531982421875,
      "learning_rate": 0.02616996699669967,
      "loss": 0.2352,
      "num_input_tokens_seen": 3345760,
      "step": 15860
    },
    {
      "epoch": 1.7453245324532454,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.026178217821782177,
      "loss": 0.2314,
      "num_input_tokens_seen": 3346816,
      "step": 15865
    },
    {
      "epoch": 1.7458745874587458,
      "grad_norm": 0.0076904296875,
      "learning_rate": 0.026186468646864684,
      "loss": 0.234,
      "num_input_tokens_seen": 3347872,
      "step": 15870
    },
    {
      "epoch": 1.7464246424642464,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.026194719471947195,
      "loss": 0.2309,
      "num_input_tokens_seen": 3348928,
      "step": 15875
    },
    {
      "epoch": 1.746974697469747,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.026202970297029702,
      "loss": 0.2319,
      "num_input_tokens_seen": 3350048,
      "step": 15880
    },
    {
      "epoch": 1.7475247524752475,
      "grad_norm": 0.007415771484375,
      "learning_rate": 0.02621122112211221,
      "loss": 0.2319,
      "num_input_tokens_seen": 3351040,
      "step": 15885
    },
    {
      "epoch": 1.748074807480748,
      "grad_norm": 0.0028076171875,
      "learning_rate": 0.026219471947194716,
      "loss": 0.2308,
      "num_input_tokens_seen": 3352096,
      "step": 15890
    },
    {
      "epoch": 1.7486248624862486,
      "grad_norm": 0.007415771484375,
      "learning_rate": 0.026227722772277227,
      "loss": 0.2314,
      "num_input_tokens_seen": 3353184,
      "step": 15895
    },
    {
      "epoch": 1.7491749174917492,
      "grad_norm": 0.007720947265625,
      "learning_rate": 0.026235973597359737,
      "loss": 0.2308,
      "num_input_tokens_seen": 3354272,
      "step": 15900
    },
    {
      "epoch": 1.7497249724972497,
      "grad_norm": 0.003082275390625,
      "learning_rate": 0.026244224422442244,
      "loss": 0.2319,
      "num_input_tokens_seen": 3355328,
      "step": 15905
    },
    {
      "epoch": 1.7502750275027503,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.02625247524752475,
      "loss": 0.2319,
      "num_input_tokens_seen": 3356416,
      "step": 15910
    },
    {
      "epoch": 1.7508250825082508,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02626072607260726,
      "loss": 0.2314,
      "num_input_tokens_seen": 3357440,
      "step": 15915
    },
    {
      "epoch": 1.7513751375137514,
      "grad_norm": 0.00750732421875,
      "learning_rate": 0.02626897689768977,
      "loss": 0.2298,
      "num_input_tokens_seen": 3358560,
      "step": 15920
    },
    {
      "epoch": 1.751925192519252,
      "grad_norm": 0.0147705078125,
      "learning_rate": 0.026277227722772276,
      "loss": 0.2314,
      "num_input_tokens_seen": 3359648,
      "step": 15925
    },
    {
      "epoch": 1.7524752475247525,
      "grad_norm": 0.0150146484375,
      "learning_rate": 0.026285478547854783,
      "loss": 0.2319,
      "num_input_tokens_seen": 3360672,
      "step": 15930
    },
    {
      "epoch": 1.753025302530253,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.02629372937293729,
      "loss": 0.2319,
      "num_input_tokens_seen": 3361696,
      "step": 15935
    },
    {
      "epoch": 1.7535753575357536,
      "grad_norm": 0.007781982421875,
      "learning_rate": 0.0263019801980198,
      "loss": 0.2329,
      "num_input_tokens_seen": 3362720,
      "step": 15940
    },
    {
      "epoch": 1.7541254125412542,
      "grad_norm": 0.01519775390625,
      "learning_rate": 0.02631023102310231,
      "loss": 0.2319,
      "num_input_tokens_seen": 3363776,
      "step": 15945
    },
    {
      "epoch": 1.7546754675467546,
      "grad_norm": 0.0076904296875,
      "learning_rate": 0.02631848184818482,
      "loss": 0.2308,
      "num_input_tokens_seen": 3364832,
      "step": 15950
    },
    {
      "epoch": 1.7552255225522553,
      "grad_norm": 0.007781982421875,
      "learning_rate": 0.026326732673267326,
      "loss": 0.2329,
      "num_input_tokens_seen": 3365856,
      "step": 15955
    },
    {
      "epoch": 1.7557755775577557,
      "grad_norm": 0.01495361328125,
      "learning_rate": 0.026334983498349833,
      "loss": 0.2319,
      "num_input_tokens_seen": 3366976,
      "step": 15960
    },
    {
      "epoch": 1.7563256325632564,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.026343234323432344,
      "loss": 0.2324,
      "num_input_tokens_seen": 3368032,
      "step": 15965
    },
    {
      "epoch": 1.756875687568757,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.02635148514851485,
      "loss": 0.2335,
      "num_input_tokens_seen": 3369088,
      "step": 15970
    },
    {
      "epoch": 1.7574257425742574,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.026359735973597358,
      "loss": 0.2324,
      "num_input_tokens_seen": 3370144,
      "step": 15975
    },
    {
      "epoch": 1.7579757975797579,
      "grad_norm": 0.0030975341796875,
      "learning_rate": 0.026367986798679865,
      "loss": 0.2324,
      "num_input_tokens_seen": 3371200,
      "step": 15980
    },
    {
      "epoch": 1.7585258525852585,
      "grad_norm": 0.007415771484375,
      "learning_rate": 0.026376237623762375,
      "loss": 0.2314,
      "num_input_tokens_seen": 3372256,
      "step": 15985
    },
    {
      "epoch": 1.7590759075907592,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.026384488448844886,
      "loss": 0.2315,
      "num_input_tokens_seen": 3373280,
      "step": 15990
    },
    {
      "epoch": 1.7596259625962596,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.026392739273927393,
      "loss": 0.2311,
      "num_input_tokens_seen": 3374336,
      "step": 15995
    },
    {
      "epoch": 1.76017601760176,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.0264009900990099,
      "loss": 0.2322,
      "num_input_tokens_seen": 3375360,
      "step": 16000
    },
    {
      "epoch": 1.7607260726072607,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.026409240924092407,
      "loss": 0.2353,
      "num_input_tokens_seen": 3376416,
      "step": 16005
    },
    {
      "epoch": 1.7612761276127613,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.026417491749174918,
      "loss": 0.2358,
      "num_input_tokens_seen": 3377472,
      "step": 16010
    },
    {
      "epoch": 1.761826182618262,
      "grad_norm": 0.003692626953125,
      "learning_rate": 0.026425742574257425,
      "loss": 0.232,
      "num_input_tokens_seen": 3378528,
      "step": 16015
    },
    {
      "epoch": 1.7623762376237624,
      "grad_norm": 0.003936767578125,
      "learning_rate": 0.026433993399339932,
      "loss": 0.2335,
      "num_input_tokens_seen": 3379648,
      "step": 16020
    },
    {
      "epoch": 1.7629262926292628,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.02644224422442244,
      "loss": 0.2314,
      "num_input_tokens_seen": 3380768,
      "step": 16025
    },
    {
      "epoch": 1.7634763476347635,
      "grad_norm": 0.0031890869140625,
      "learning_rate": 0.02645049504950495,
      "loss": 0.2303,
      "num_input_tokens_seen": 3381920,
      "step": 16030
    },
    {
      "epoch": 1.7640264026402641,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.026458745874587457,
      "loss": 0.2298,
      "num_input_tokens_seen": 3382976,
      "step": 16035
    },
    {
      "epoch": 1.7645764576457645,
      "grad_norm": 0.003509521484375,
      "learning_rate": 0.026466996699669967,
      "loss": 0.2293,
      "num_input_tokens_seen": 3384096,
      "step": 16040
    },
    {
      "epoch": 1.765126512651265,
      "grad_norm": 0.01416015625,
      "learning_rate": 0.026475247524752474,
      "loss": 0.2299,
      "num_input_tokens_seen": 3385120,
      "step": 16045
    },
    {
      "epoch": 1.7656765676567656,
      "grad_norm": 0.01434326171875,
      "learning_rate": 0.02648349834983498,
      "loss": 0.228,
      "num_input_tokens_seen": 3386208,
      "step": 16050
    },
    {
      "epoch": 1.7662266226622663,
      "grad_norm": 0.01409912109375,
      "learning_rate": 0.026491749174917492,
      "loss": 0.2308,
      "num_input_tokens_seen": 3387232,
      "step": 16055
    },
    {
      "epoch": 1.766776677667767,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.0265,
      "loss": 0.2299,
      "num_input_tokens_seen": 3388256,
      "step": 16060
    },
    {
      "epoch": 1.7673267326732673,
      "grad_norm": 0.0032501220703125,
      "learning_rate": 0.026508250825082506,
      "loss": 0.2346,
      "num_input_tokens_seen": 3389344,
      "step": 16065
    },
    {
      "epoch": 1.7678767876787678,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.026516501650165013,
      "loss": 0.2279,
      "num_input_tokens_seen": 3390400,
      "step": 16070
    },
    {
      "epoch": 1.7684268426842684,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.026524752475247524,
      "loss": 0.2341,
      "num_input_tokens_seen": 3391456,
      "step": 16075
    },
    {
      "epoch": 1.768976897689769,
      "grad_norm": 0.00897216796875,
      "learning_rate": 0.02653300330033003,
      "loss": 0.23,
      "num_input_tokens_seen": 3392576,
      "step": 16080
    },
    {
      "epoch": 1.7695269526952695,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.02654125412541254,
      "loss": 0.231,
      "num_input_tokens_seen": 3393696,
      "step": 16085
    },
    {
      "epoch": 1.77007700770077,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.02654950495049505,
      "loss": 0.2367,
      "num_input_tokens_seen": 3394688,
      "step": 16090
    },
    {
      "epoch": 1.7706270627062706,
      "grad_norm": 0.01416015625,
      "learning_rate": 0.026557755775577556,
      "loss": 0.2329,
      "num_input_tokens_seen": 3395712,
      "step": 16095
    },
    {
      "epoch": 1.7711771177117712,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.026566006600660066,
      "loss": 0.2348,
      "num_input_tokens_seen": 3396768,
      "step": 16100
    },
    {
      "epoch": 1.7717271727172719,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.026574257425742574,
      "loss": 0.2307,
      "num_input_tokens_seen": 3397792,
      "step": 16105
    },
    {
      "epoch": 1.7722772277227723,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.02658250825082508,
      "loss": 0.2359,
      "num_input_tokens_seen": 3398848,
      "step": 16110
    },
    {
      "epoch": 1.7728272827282727,
      "grad_norm": 0.0030364990234375,
      "learning_rate": 0.026590759075907588,
      "loss": 0.2306,
      "num_input_tokens_seen": 3399968,
      "step": 16115
    },
    {
      "epoch": 1.7733773377337734,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.0265990099009901,
      "loss": 0.2317,
      "num_input_tokens_seen": 3401024,
      "step": 16120
    },
    {
      "epoch": 1.773927392739274,
      "grad_norm": 0.0081787109375,
      "learning_rate": 0.026607260726072605,
      "loss": 0.2306,
      "num_input_tokens_seen": 3402048,
      "step": 16125
    },
    {
      "epoch": 1.7744774477447744,
      "grad_norm": 0.007080078125,
      "learning_rate": 0.026615511551155116,
      "loss": 0.2353,
      "num_input_tokens_seen": 3403136,
      "step": 16130
    },
    {
      "epoch": 1.7750275027502749,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.026623762376237623,
      "loss": 0.2326,
      "num_input_tokens_seen": 3404192,
      "step": 16135
    },
    {
      "epoch": 1.7755775577557755,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.02663201320132013,
      "loss": 0.232,
      "num_input_tokens_seen": 3405248,
      "step": 16140
    },
    {
      "epoch": 1.7761276127612762,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.02664026402640264,
      "loss": 0.2299,
      "num_input_tokens_seen": 3406304,
      "step": 16145
    },
    {
      "epoch": 1.7766776677667768,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.026648514851485148,
      "loss": 0.2273,
      "num_input_tokens_seen": 3407360,
      "step": 16150
    },
    {
      "epoch": 1.7772277227722773,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.026656765676567655,
      "loss": 0.2332,
      "num_input_tokens_seen": 3408480,
      "step": 16155
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 0.00830078125,
      "learning_rate": 0.026665016501650162,
      "loss": 0.2322,
      "num_input_tokens_seen": 3409504,
      "step": 16160
    },
    {
      "epoch": 1.7783278327832783,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.026673267326732673,
      "loss": 0.2281,
      "num_input_tokens_seen": 3410560,
      "step": 16165
    },
    {
      "epoch": 1.778877887788779,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.02668151815181518,
      "loss": 0.2328,
      "num_input_tokens_seen": 3411616,
      "step": 16170
    },
    {
      "epoch": 1.7794279427942794,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02668976897689769,
      "loss": 0.2292,
      "num_input_tokens_seen": 3412608,
      "step": 16175
    },
    {
      "epoch": 1.7799779977997798,
      "grad_norm": 0.002593994140625,
      "learning_rate": 0.026698019801980197,
      "loss": 0.2251,
      "num_input_tokens_seen": 3413664,
      "step": 16180
    },
    {
      "epoch": 1.7805280528052805,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.026706270627062705,
      "loss": 0.2262,
      "num_input_tokens_seen": 3414688,
      "step": 16185
    },
    {
      "epoch": 1.7810781078107811,
      "grad_norm": 0.016845703125,
      "learning_rate": 0.026714521452145215,
      "loss": 0.2348,
      "num_input_tokens_seen": 3415744,
      "step": 16190
    },
    {
      "epoch": 1.7816281628162818,
      "grad_norm": 0.016357421875,
      "learning_rate": 0.026722772277227722,
      "loss": 0.2337,
      "num_input_tokens_seen": 3416800,
      "step": 16195
    },
    {
      "epoch": 1.7821782178217822,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02673102310231023,
      "loss": 0.2317,
      "num_input_tokens_seen": 3417920,
      "step": 16200
    },
    {
      "epoch": 1.7827282728272826,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.026739273927392736,
      "loss": 0.2321,
      "num_input_tokens_seen": 3419008,
      "step": 16205
    },
    {
      "epoch": 1.7832783278327833,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.026747524752475247,
      "loss": 0.233,
      "num_input_tokens_seen": 3420064,
      "step": 16210
    },
    {
      "epoch": 1.783828382838284,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.026755775577557754,
      "loss": 0.234,
      "num_input_tokens_seen": 3421056,
      "step": 16215
    },
    {
      "epoch": 1.7843784378437844,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.026764026402640265,
      "loss": 0.2303,
      "num_input_tokens_seen": 3422080,
      "step": 16220
    },
    {
      "epoch": 1.7849284928492848,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.026772277227722772,
      "loss": 0.2251,
      "num_input_tokens_seen": 3423232,
      "step": 16225
    },
    {
      "epoch": 1.7854785478547854,
      "grad_norm": 0.003326416015625,
      "learning_rate": 0.02678052805280528,
      "loss": 0.2381,
      "num_input_tokens_seen": 3424320,
      "step": 16230
    },
    {
      "epoch": 1.786028602860286,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.02678877887788779,
      "loss": 0.2291,
      "num_input_tokens_seen": 3425344,
      "step": 16235
    },
    {
      "epoch": 1.7865786578657867,
      "grad_norm": 0.007080078125,
      "learning_rate": 0.026797029702970297,
      "loss": 0.2358,
      "num_input_tokens_seen": 3426400,
      "step": 16240
    },
    {
      "epoch": 1.7871287128712872,
      "grad_norm": 0.00830078125,
      "learning_rate": 0.026805280528052804,
      "loss": 0.2341,
      "num_input_tokens_seen": 3427488,
      "step": 16245
    },
    {
      "epoch": 1.7876787678767876,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.02681353135313531,
      "loss": 0.234,
      "num_input_tokens_seen": 3428544,
      "step": 16250
    },
    {
      "epoch": 1.7882288228822882,
      "grad_norm": 0.016357421875,
      "learning_rate": 0.02682178217821782,
      "loss": 0.2298,
      "num_input_tokens_seen": 3429600,
      "step": 16255
    },
    {
      "epoch": 1.7887788778877889,
      "grad_norm": 0.0029449462890625,
      "learning_rate": 0.02683003300330033,
      "loss": 0.2284,
      "num_input_tokens_seen": 3430656,
      "step": 16260
    },
    {
      "epoch": 1.7893289328932893,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02683828382838284,
      "loss": 0.2237,
      "num_input_tokens_seen": 3431680,
      "step": 16265
    },
    {
      "epoch": 1.7898789878987897,
      "grad_norm": 0.01495361328125,
      "learning_rate": 0.026846534653465346,
      "loss": 0.2409,
      "num_input_tokens_seen": 3432736,
      "step": 16270
    },
    {
      "epoch": 1.7904290429042904,
      "grad_norm": 0.0137939453125,
      "learning_rate": 0.026854785478547853,
      "loss": 0.2346,
      "num_input_tokens_seen": 3433824,
      "step": 16275
    },
    {
      "epoch": 1.790979097909791,
      "grad_norm": 0.0159912109375,
      "learning_rate": 0.026863036303630364,
      "loss": 0.2333,
      "num_input_tokens_seen": 3434848,
      "step": 16280
    },
    {
      "epoch": 1.7915291529152917,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.02687128712871287,
      "loss": 0.2273,
      "num_input_tokens_seen": 3435904,
      "step": 16285
    },
    {
      "epoch": 1.7920792079207921,
      "grad_norm": 0.0074462890625,
      "learning_rate": 0.026879537953795378,
      "loss": 0.2439,
      "num_input_tokens_seen": 3437056,
      "step": 16290
    },
    {
      "epoch": 1.7926292629262925,
      "grad_norm": 0.01385498046875,
      "learning_rate": 0.026887788778877885,
      "loss": 0.2311,
      "num_input_tokens_seen": 3438144,
      "step": 16295
    },
    {
      "epoch": 1.7931793179317932,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.026896039603960396,
      "loss": 0.2299,
      "num_input_tokens_seen": 3439168,
      "step": 16300
    },
    {
      "epoch": 1.7937293729372938,
      "grad_norm": 0.0137939453125,
      "learning_rate": 0.026904290429042903,
      "loss": 0.2315,
      "num_input_tokens_seen": 3440224,
      "step": 16305
    },
    {
      "epoch": 1.7942794279427943,
      "grad_norm": 0.0029754638671875,
      "learning_rate": 0.026912541254125413,
      "loss": 0.2305,
      "num_input_tokens_seen": 3441312,
      "step": 16310
    },
    {
      "epoch": 1.7948294829482947,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.026920792079207917,
      "loss": 0.231,
      "num_input_tokens_seen": 3442368,
      "step": 16315
    },
    {
      "epoch": 1.7953795379537953,
      "grad_norm": 0.007080078125,
      "learning_rate": 0.026929042904290428,
      "loss": 0.2299,
      "num_input_tokens_seen": 3443456,
      "step": 16320
    },
    {
      "epoch": 1.795929592959296,
      "grad_norm": 0.01422119140625,
      "learning_rate": 0.026937293729372938,
      "loss": 0.2304,
      "num_input_tokens_seen": 3444480,
      "step": 16325
    },
    {
      "epoch": 1.7964796479647966,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.026945544554455445,
      "loss": 0.2325,
      "num_input_tokens_seen": 3445536,
      "step": 16330
    },
    {
      "epoch": 1.797029702970297,
      "grad_norm": 0.0078125,
      "learning_rate": 0.026953795379537952,
      "loss": 0.2331,
      "num_input_tokens_seen": 3446592,
      "step": 16335
    },
    {
      "epoch": 1.7975797579757975,
      "grad_norm": 0.0140380859375,
      "learning_rate": 0.02696204620462046,
      "loss": 0.2264,
      "num_input_tokens_seen": 3447680,
      "step": 16340
    },
    {
      "epoch": 1.7981298129812981,
      "grad_norm": 0.0040283203125,
      "learning_rate": 0.02697029702970297,
      "loss": 0.2285,
      "num_input_tokens_seen": 3448704,
      "step": 16345
    },
    {
      "epoch": 1.7986798679867988,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.026978547854785477,
      "loss": 0.2326,
      "num_input_tokens_seen": 3449760,
      "step": 16350
    },
    {
      "epoch": 1.7992299229922992,
      "grad_norm": 0.007476806640625,
      "learning_rate": 0.026986798679867988,
      "loss": 0.2326,
      "num_input_tokens_seen": 3450816,
      "step": 16355
    },
    {
      "epoch": 1.7997799779977997,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.02699504950495049,
      "loss": 0.2347,
      "num_input_tokens_seen": 3451936,
      "step": 16360
    },
    {
      "epoch": 1.8003300330033003,
      "grad_norm": 0.003997802734375,
      "learning_rate": 0.027003300330033002,
      "loss": 0.234,
      "num_input_tokens_seen": 3452960,
      "step": 16365
    },
    {
      "epoch": 1.800880088008801,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.027011551155115512,
      "loss": 0.2341,
      "num_input_tokens_seen": 3453952,
      "step": 16370
    },
    {
      "epoch": 1.8014301430143014,
      "grad_norm": 0.0026397705078125,
      "learning_rate": 0.02701980198019802,
      "loss": 0.231,
      "num_input_tokens_seen": 3455008,
      "step": 16375
    },
    {
      "epoch": 1.801980198019802,
      "grad_norm": 0.007080078125,
      "learning_rate": 0.027028052805280527,
      "loss": 0.2341,
      "num_input_tokens_seen": 3456064,
      "step": 16380
    },
    {
      "epoch": 1.8025302530253025,
      "grad_norm": 0.00762939453125,
      "learning_rate": 0.027036303630363034,
      "loss": 0.2284,
      "num_input_tokens_seen": 3457120,
      "step": 16385
    },
    {
      "epoch": 1.803080308030803,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.027044554455445544,
      "loss": 0.2325,
      "num_input_tokens_seen": 3458176,
      "step": 16390
    },
    {
      "epoch": 1.8036303630363038,
      "grad_norm": 0.00750732421875,
      "learning_rate": 0.02705280528052805,
      "loss": 0.2357,
      "num_input_tokens_seen": 3459232,
      "step": 16395
    },
    {
      "epoch": 1.8041804180418042,
      "grad_norm": 0.01397705078125,
      "learning_rate": 0.027061056105610562,
      "loss": 0.233,
      "num_input_tokens_seen": 3460256,
      "step": 16400
    },
    {
      "epoch": 1.8047304730473046,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.027069306930693066,
      "loss": 0.2293,
      "num_input_tokens_seen": 3461280,
      "step": 16405
    },
    {
      "epoch": 1.8052805280528053,
      "grad_norm": 0.002960205078125,
      "learning_rate": 0.027077557755775576,
      "loss": 0.2313,
      "num_input_tokens_seen": 3462368,
      "step": 16410
    },
    {
      "epoch": 1.805830583058306,
      "grad_norm": 0.003082275390625,
      "learning_rate": 0.027085808580858087,
      "loss": 0.2324,
      "num_input_tokens_seen": 3463392,
      "step": 16415
    },
    {
      "epoch": 1.8063806380638063,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.027094059405940594,
      "loss": 0.2313,
      "num_input_tokens_seen": 3464448,
      "step": 16420
    },
    {
      "epoch": 1.806930693069307,
      "grad_norm": 0.00347900390625,
      "learning_rate": 0.0271023102310231,
      "loss": 0.2313,
      "num_input_tokens_seen": 3465536,
      "step": 16425
    },
    {
      "epoch": 1.8074807480748074,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.027110561056105608,
      "loss": 0.2313,
      "num_input_tokens_seen": 3466560,
      "step": 16430
    },
    {
      "epoch": 1.808030803080308,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.02711881188118812,
      "loss": 0.2313,
      "num_input_tokens_seen": 3467584,
      "step": 16435
    },
    {
      "epoch": 1.8085808580858087,
      "grad_norm": 0.0072021484375,
      "learning_rate": 0.027127062706270626,
      "loss": 0.2313,
      "num_input_tokens_seen": 3468640,
      "step": 16440
    },
    {
      "epoch": 1.8091309130913091,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.027135313531353136,
      "loss": 0.2313,
      "num_input_tokens_seen": 3469728,
      "step": 16445
    },
    {
      "epoch": 1.8096809680968096,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.02714356435643564,
      "loss": 0.2308,
      "num_input_tokens_seen": 3470784,
      "step": 16450
    },
    {
      "epoch": 1.8102310231023102,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.02715181518151815,
      "loss": 0.2303,
      "num_input_tokens_seen": 3471744,
      "step": 16455
    },
    {
      "epoch": 1.8107810781078109,
      "grad_norm": 0.0140380859375,
      "learning_rate": 0.02716006600660066,
      "loss": 0.235,
      "num_input_tokens_seen": 3472800,
      "step": 16460
    },
    {
      "epoch": 1.8113311331133113,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.027168316831683168,
      "loss": 0.2319,
      "num_input_tokens_seen": 3473824,
      "step": 16465
    },
    {
      "epoch": 1.811881188118812,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.027176567656765675,
      "loss": 0.2319,
      "num_input_tokens_seen": 3474944,
      "step": 16470
    },
    {
      "epoch": 1.8124312431243124,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.027184818481848182,
      "loss": 0.2288,
      "num_input_tokens_seen": 3476000,
      "step": 16475
    },
    {
      "epoch": 1.812981298129813,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.027193069306930693,
      "loss": 0.2329,
      "num_input_tokens_seen": 3477088,
      "step": 16480
    },
    {
      "epoch": 1.8135313531353137,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.0272013201320132,
      "loss": 0.2288,
      "num_input_tokens_seen": 3478144,
      "step": 16485
    },
    {
      "epoch": 1.814081408140814,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.02720957095709571,
      "loss": 0.2303,
      "num_input_tokens_seen": 3479200,
      "step": 16490
    },
    {
      "epoch": 1.8146314631463145,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.027217821782178214,
      "loss": 0.2335,
      "num_input_tokens_seen": 3480224,
      "step": 16495
    },
    {
      "epoch": 1.8151815181518152,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.027226072607260725,
      "loss": 0.2293,
      "num_input_tokens_seen": 3481216,
      "step": 16500
    },
    {
      "epoch": 1.8157315731573158,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.027234323432343235,
      "loss": 0.2325,
      "num_input_tokens_seen": 3482304,
      "step": 16505
    },
    {
      "epoch": 1.8162816281628162,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.027242574257425742,
      "loss": 0.2309,
      "num_input_tokens_seen": 3483328,
      "step": 16510
    },
    {
      "epoch": 1.8168316831683167,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.02725082508250825,
      "loss": 0.233,
      "num_input_tokens_seen": 3484352,
      "step": 16515
    },
    {
      "epoch": 1.8173817381738173,
      "grad_norm": 0.013671875,
      "learning_rate": 0.027259075907590757,
      "loss": 0.2314,
      "num_input_tokens_seen": 3485504,
      "step": 16520
    },
    {
      "epoch": 1.817931793179318,
      "grad_norm": 0.0076904296875,
      "learning_rate": 0.027267326732673267,
      "loss": 0.234,
      "num_input_tokens_seen": 3486528,
      "step": 16525
    },
    {
      "epoch": 1.8184818481848186,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.027275577557755774,
      "loss": 0.2335,
      "num_input_tokens_seen": 3487552,
      "step": 16530
    },
    {
      "epoch": 1.819031903190319,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.027283828382838285,
      "loss": 0.2314,
      "num_input_tokens_seen": 3488672,
      "step": 16535
    },
    {
      "epoch": 1.8195819581958195,
      "grad_norm": 0.0146484375,
      "learning_rate": 0.02729207920792079,
      "loss": 0.232,
      "num_input_tokens_seen": 3489728,
      "step": 16540
    },
    {
      "epoch": 1.8201320132013201,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0273003300330033,
      "loss": 0.2325,
      "num_input_tokens_seen": 3490720,
      "step": 16545
    },
    {
      "epoch": 1.8206820682068208,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.027308580858085806,
      "loss": 0.2304,
      "num_input_tokens_seen": 3491744,
      "step": 16550
    },
    {
      "epoch": 1.8212321232123212,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.027316831683168317,
      "loss": 0.232,
      "num_input_tokens_seen": 3492832,
      "step": 16555
    },
    {
      "epoch": 1.8217821782178216,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.027325082508250824,
      "loss": 0.2314,
      "num_input_tokens_seen": 3493920,
      "step": 16560
    },
    {
      "epoch": 1.8223322332233223,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.02733333333333333,
      "loss": 0.2299,
      "num_input_tokens_seen": 3494944,
      "step": 16565
    },
    {
      "epoch": 1.822882288228823,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.02734158415841584,
      "loss": 0.2315,
      "num_input_tokens_seen": 3495968,
      "step": 16570
    },
    {
      "epoch": 1.8234323432343236,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.02734983498349835,
      "loss": 0.2304,
      "num_input_tokens_seen": 3497024,
      "step": 16575
    },
    {
      "epoch": 1.823982398239824,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.02735808580858086,
      "loss": 0.233,
      "num_input_tokens_seen": 3498048,
      "step": 16580
    },
    {
      "epoch": 1.8245324532453244,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.027366336633663363,
      "loss": 0.2313,
      "num_input_tokens_seen": 3499040,
      "step": 16585
    },
    {
      "epoch": 1.825082508250825,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.027374587458745873,
      "loss": 0.2329,
      "num_input_tokens_seen": 3500128,
      "step": 16590
    },
    {
      "epoch": 1.8256325632563257,
      "grad_norm": 0.0032501220703125,
      "learning_rate": 0.02738283828382838,
      "loss": 0.2319,
      "num_input_tokens_seen": 3501152,
      "step": 16595
    },
    {
      "epoch": 1.8261826182618262,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.02739108910891089,
      "loss": 0.2318,
      "num_input_tokens_seen": 3502176,
      "step": 16600
    },
    {
      "epoch": 1.8267326732673266,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.027399339933993398,
      "loss": 0.2334,
      "num_input_tokens_seen": 3503232,
      "step": 16605
    },
    {
      "epoch": 1.8272827282728272,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.027407590759075905,
      "loss": 0.2298,
      "num_input_tokens_seen": 3504320,
      "step": 16610
    },
    {
      "epoch": 1.8278327832783279,
      "grad_norm": 0.007568359375,
      "learning_rate": 0.027415841584158416,
      "loss": 0.2293,
      "num_input_tokens_seen": 3505376,
      "step": 16615
    },
    {
      "epoch": 1.8283828382838285,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.027424092409240923,
      "loss": 0.2284,
      "num_input_tokens_seen": 3506464,
      "step": 16620
    },
    {
      "epoch": 1.828932893289329,
      "grad_norm": 0.003021240234375,
      "learning_rate": 0.027432343234323434,
      "loss": 0.2301,
      "num_input_tokens_seen": 3507520,
      "step": 16625
    },
    {
      "epoch": 1.8294829482948294,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.027440594059405937,
      "loss": 0.236,
      "num_input_tokens_seen": 3508576,
      "step": 16630
    },
    {
      "epoch": 1.83003300330033,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.027448844884488448,
      "loss": 0.238,
      "num_input_tokens_seen": 3509600,
      "step": 16635
    },
    {
      "epoch": 1.8305830583058307,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.027457095709570955,
      "loss": 0.2353,
      "num_input_tokens_seen": 3510592,
      "step": 16640
    },
    {
      "epoch": 1.831133113311331,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.027465346534653465,
      "loss": 0.2274,
      "num_input_tokens_seen": 3511648,
      "step": 16645
    },
    {
      "epoch": 1.8316831683168315,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.027473597359735973,
      "loss": 0.2335,
      "num_input_tokens_seen": 3512704,
      "step": 16650
    },
    {
      "epoch": 1.8322332233223322,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.02748184818481848,
      "loss": 0.2315,
      "num_input_tokens_seen": 3513792,
      "step": 16655
    },
    {
      "epoch": 1.8327832783278328,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02749009900990099,
      "loss": 0.2335,
      "num_input_tokens_seen": 3514816,
      "step": 16660
    },
    {
      "epoch": 1.8333333333333335,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.027498349834983497,
      "loss": 0.2314,
      "num_input_tokens_seen": 3515840,
      "step": 16665
    },
    {
      "epoch": 1.833883388338834,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.027506600660066008,
      "loss": 0.232,
      "num_input_tokens_seen": 3516832,
      "step": 16670
    },
    {
      "epoch": 1.8344334433443343,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02751485148514851,
      "loss": 0.2314,
      "num_input_tokens_seen": 3517856,
      "step": 16675
    },
    {
      "epoch": 1.834983498349835,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.027523102310231022,
      "loss": 0.2309,
      "num_input_tokens_seen": 3518944,
      "step": 16680
    },
    {
      "epoch": 1.8355335533553356,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.02753135313531353,
      "loss": 0.2299,
      "num_input_tokens_seen": 3520064,
      "step": 16685
    },
    {
      "epoch": 1.836083608360836,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.02753960396039604,
      "loss": 0.2294,
      "num_input_tokens_seen": 3521184,
      "step": 16690
    },
    {
      "epoch": 1.8366336633663365,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.027547854785478547,
      "loss": 0.232,
      "num_input_tokens_seen": 3522304,
      "step": 16695
    },
    {
      "epoch": 1.8371837183718371,
      "grad_norm": 0.007598876953125,
      "learning_rate": 0.027556105610561054,
      "loss": 0.233,
      "num_input_tokens_seen": 3523360,
      "step": 16700
    },
    {
      "epoch": 1.8377337733773378,
      "grad_norm": 0.0135498046875,
      "learning_rate": 0.027564356435643565,
      "loss": 0.2325,
      "num_input_tokens_seen": 3524416,
      "step": 16705
    },
    {
      "epoch": 1.8382838283828384,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.02757260726072607,
      "loss": 0.2288,
      "num_input_tokens_seen": 3525472,
      "step": 16710
    },
    {
      "epoch": 1.8388338833883389,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.027580858085808582,
      "loss": 0.2284,
      "num_input_tokens_seen": 3526528,
      "step": 16715
    },
    {
      "epoch": 1.8393839383938393,
      "grad_norm": 0.01397705078125,
      "learning_rate": 0.027589108910891086,
      "loss": 0.2372,
      "num_input_tokens_seen": 3527584,
      "step": 16720
    },
    {
      "epoch": 1.83993399339934,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.027597359735973596,
      "loss": 0.2304,
      "num_input_tokens_seen": 3528672,
      "step": 16725
    },
    {
      "epoch": 1.8404840484048406,
      "grad_norm": 0.00762939453125,
      "learning_rate": 0.027605610561056104,
      "loss": 0.2346,
      "num_input_tokens_seen": 3529760,
      "step": 16730
    },
    {
      "epoch": 1.841034103410341,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.027613861386138614,
      "loss": 0.232,
      "num_input_tokens_seen": 3530848,
      "step": 16735
    },
    {
      "epoch": 1.8415841584158414,
      "grad_norm": 0.00762939453125,
      "learning_rate": 0.02762211221122112,
      "loss": 0.2325,
      "num_input_tokens_seen": 3531872,
      "step": 16740
    },
    {
      "epoch": 1.842134213421342,
      "grad_norm": 0.007171630859375,
      "learning_rate": 0.02763036303630363,
      "loss": 0.2294,
      "num_input_tokens_seen": 3532864,
      "step": 16745
    },
    {
      "epoch": 1.8426842684268427,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.02763861386138614,
      "loss": 0.2293,
      "num_input_tokens_seen": 3533856,
      "step": 16750
    },
    {
      "epoch": 1.8432343234323434,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.027646864686468646,
      "loss": 0.2304,
      "num_input_tokens_seen": 3534944,
      "step": 16755
    },
    {
      "epoch": 1.8437843784378438,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.027655115511551157,
      "loss": 0.2325,
      "num_input_tokens_seen": 3536000,
      "step": 16760
    },
    {
      "epoch": 1.8443344334433442,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.02766336633663366,
      "loss": 0.2289,
      "num_input_tokens_seen": 3537088,
      "step": 16765
    },
    {
      "epoch": 1.844884488448845,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.02767161716171617,
      "loss": 0.2341,
      "num_input_tokens_seen": 3538144,
      "step": 16770
    },
    {
      "epoch": 1.8454345434543455,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.027679867986798678,
      "loss": 0.2326,
      "num_input_tokens_seen": 3539232,
      "step": 16775
    },
    {
      "epoch": 1.845984598459846,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.02768811881188119,
      "loss": 0.2279,
      "num_input_tokens_seen": 3540384,
      "step": 16780
    },
    {
      "epoch": 1.8465346534653464,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.027696369636963696,
      "loss": 0.2325,
      "num_input_tokens_seen": 3541472,
      "step": 16785
    },
    {
      "epoch": 1.847084708470847,
      "grad_norm": 0.00762939453125,
      "learning_rate": 0.027704620462046203,
      "loss": 0.2331,
      "num_input_tokens_seen": 3542528,
      "step": 16790
    },
    {
      "epoch": 1.8476347634763477,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.027712871287128713,
      "loss": 0.2336,
      "num_input_tokens_seen": 3543552,
      "step": 16795
    },
    {
      "epoch": 1.8481848184818483,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.02772112211221122,
      "loss": 0.2346,
      "num_input_tokens_seen": 3544608,
      "step": 16800
    },
    {
      "epoch": 1.8487348734873488,
      "grad_norm": 0.0030975341796875,
      "learning_rate": 0.02772937293729373,
      "loss": 0.234,
      "num_input_tokens_seen": 3545632,
      "step": 16805
    },
    {
      "epoch": 1.8492849284928492,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.027737623762376235,
      "loss": 0.2309,
      "num_input_tokens_seen": 3546720,
      "step": 16810
    },
    {
      "epoch": 1.8498349834983498,
      "grad_norm": 0.00762939453125,
      "learning_rate": 0.027745874587458745,
      "loss": 0.2314,
      "num_input_tokens_seen": 3547808,
      "step": 16815
    },
    {
      "epoch": 1.8503850385038505,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.027754125412541252,
      "loss": 0.2293,
      "num_input_tokens_seen": 3548832,
      "step": 16820
    },
    {
      "epoch": 1.850935093509351,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.027762376237623763,
      "loss": 0.2299,
      "num_input_tokens_seen": 3549888,
      "step": 16825
    },
    {
      "epoch": 1.8514851485148514,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.027770627062706266,
      "loss": 0.2346,
      "num_input_tokens_seen": 3550944,
      "step": 16830
    },
    {
      "epoch": 1.852035203520352,
      "grad_norm": 0.0074462890625,
      "learning_rate": 0.027778877887788777,
      "loss": 0.2324,
      "num_input_tokens_seen": 3551968,
      "step": 16835
    },
    {
      "epoch": 1.8525852585258527,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.027787128712871288,
      "loss": 0.2303,
      "num_input_tokens_seen": 3552960,
      "step": 16840
    },
    {
      "epoch": 1.8531353135313533,
      "grad_norm": 0.0135498046875,
      "learning_rate": 0.027795379537953795,
      "loss": 0.2308,
      "num_input_tokens_seen": 3554016,
      "step": 16845
    },
    {
      "epoch": 1.8536853685368537,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.027803630363036305,
      "loss": 0.2314,
      "num_input_tokens_seen": 3555072,
      "step": 16850
    },
    {
      "epoch": 1.8542354235423542,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.02781188118811881,
      "loss": 0.2304,
      "num_input_tokens_seen": 3556160,
      "step": 16855
    },
    {
      "epoch": 1.8547854785478548,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.02782013201320132,
      "loss": 0.231,
      "num_input_tokens_seen": 3557152,
      "step": 16860
    },
    {
      "epoch": 1.8553355335533555,
      "grad_norm": 0.002777099609375,
      "learning_rate": 0.027828382838283827,
      "loss": 0.2297,
      "num_input_tokens_seen": 3558208,
      "step": 16865
    },
    {
      "epoch": 1.8558855885588559,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.027836633663366337,
      "loss": 0.2347,
      "num_input_tokens_seen": 3559232,
      "step": 16870
    },
    {
      "epoch": 1.8564356435643563,
      "grad_norm": 0.0029296875,
      "learning_rate": 0.02784488448844884,
      "loss": 0.229,
      "num_input_tokens_seen": 3560320,
      "step": 16875
    },
    {
      "epoch": 1.856985698569857,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.02785313531353135,
      "loss": 0.2311,
      "num_input_tokens_seen": 3561440,
      "step": 16880
    },
    {
      "epoch": 1.8575357535753576,
      "grad_norm": 0.007568359375,
      "learning_rate": 0.027861386138613862,
      "loss": 0.2341,
      "num_input_tokens_seen": 3562496,
      "step": 16885
    },
    {
      "epoch": 1.858085808580858,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.02786963696369637,
      "loss": 0.2325,
      "num_input_tokens_seen": 3563648,
      "step": 16890
    },
    {
      "epoch": 1.8586358635863587,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02787788778877888,
      "loss": 0.2335,
      "num_input_tokens_seen": 3564640,
      "step": 16895
    },
    {
      "epoch": 1.859185918591859,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.027886138613861383,
      "loss": 0.2304,
      "num_input_tokens_seen": 3565696,
      "step": 16900
    },
    {
      "epoch": 1.8597359735973598,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.027894389438943894,
      "loss": 0.2324,
      "num_input_tokens_seen": 3566720,
      "step": 16905
    },
    {
      "epoch": 1.8602860286028604,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.0279026402640264,
      "loss": 0.2293,
      "num_input_tokens_seen": 3567776,
      "step": 16910
    },
    {
      "epoch": 1.8608360836083608,
      "grad_norm": 0.01324462890625,
      "learning_rate": 0.02791089108910891,
      "loss": 0.2329,
      "num_input_tokens_seen": 3568864,
      "step": 16915
    },
    {
      "epoch": 1.8613861386138613,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.027919141914191415,
      "loss": 0.2308,
      "num_input_tokens_seen": 3569952,
      "step": 16920
    },
    {
      "epoch": 1.861936193619362,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.027927392739273926,
      "loss": 0.2308,
      "num_input_tokens_seen": 3571072,
      "step": 16925
    },
    {
      "epoch": 1.8624862486248626,
      "grad_norm": 0.007080078125,
      "learning_rate": 0.027935643564356436,
      "loss": 0.2314,
      "num_input_tokens_seen": 3572224,
      "step": 16930
    },
    {
      "epoch": 1.863036303630363,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.027943894389438943,
      "loss": 0.2314,
      "num_input_tokens_seen": 3573248,
      "step": 16935
    },
    {
      "epoch": 1.8635863586358636,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.027952145214521454,
      "loss": 0.2319,
      "num_input_tokens_seen": 3574272,
      "step": 16940
    },
    {
      "epoch": 1.864136413641364,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.027960396039603957,
      "loss": 0.2303,
      "num_input_tokens_seen": 3575296,
      "step": 16945
    },
    {
      "epoch": 1.8646864686468647,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.027968646864686468,
      "loss": 0.2329,
      "num_input_tokens_seen": 3576384,
      "step": 16950
    },
    {
      "epoch": 1.8652365236523654,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.027976897689768975,
      "loss": 0.2288,
      "num_input_tokens_seen": 3577408,
      "step": 16955
    },
    {
      "epoch": 1.8657865786578658,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.027985148514851486,
      "loss": 0.2314,
      "num_input_tokens_seen": 3578400,
      "step": 16960
    },
    {
      "epoch": 1.8663366336633662,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02799339933993399,
      "loss": 0.2303,
      "num_input_tokens_seen": 3579424,
      "step": 16965
    },
    {
      "epoch": 1.8668866886688669,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.0280016501650165,
      "loss": 0.2319,
      "num_input_tokens_seen": 3580480,
      "step": 16970
    },
    {
      "epoch": 1.8674367436743675,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.02800990099009901,
      "loss": 0.2324,
      "num_input_tokens_seen": 3581536,
      "step": 16975
    },
    {
      "epoch": 1.867986798679868,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.028018151815181518,
      "loss": 0.2308,
      "num_input_tokens_seen": 3582656,
      "step": 16980
    },
    {
      "epoch": 1.8685368536853684,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.028026402640264028,
      "loss": 0.2313,
      "num_input_tokens_seen": 3583712,
      "step": 16985
    },
    {
      "epoch": 1.869086908690869,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.028034653465346532,
      "loss": 0.2319,
      "num_input_tokens_seen": 3584800,
      "step": 16990
    },
    {
      "epoch": 1.8696369636963697,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.028042904290429042,
      "loss": 0.2308,
      "num_input_tokens_seen": 3585856,
      "step": 16995
    },
    {
      "epoch": 1.8701870187018703,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.02805115511551155,
      "loss": 0.2324,
      "num_input_tokens_seen": 3586880,
      "step": 17000
    },
    {
      "epoch": 1.8707370737073707,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.02805940594059406,
      "loss": 0.2308,
      "num_input_tokens_seen": 3587936,
      "step": 17005
    },
    {
      "epoch": 1.8712871287128712,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.028067656765676564,
      "loss": 0.2298,
      "num_input_tokens_seen": 3588928,
      "step": 17010
    },
    {
      "epoch": 1.8718371837183718,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.028075907590759074,
      "loss": 0.2298,
      "num_input_tokens_seen": 3589984,
      "step": 17015
    },
    {
      "epoch": 1.8723872387238725,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.028084158415841585,
      "loss": 0.2308,
      "num_input_tokens_seen": 3591040,
      "step": 17020
    },
    {
      "epoch": 1.872937293729373,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.028092409240924092,
      "loss": 0.2298,
      "num_input_tokens_seen": 3592032,
      "step": 17025
    },
    {
      "epoch": 1.8734873487348733,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.028100660066006602,
      "loss": 0.2319,
      "num_input_tokens_seen": 3593024,
      "step": 17030
    },
    {
      "epoch": 1.874037403740374,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.028108910891089106,
      "loss": 0.2341,
      "num_input_tokens_seen": 3594016,
      "step": 17035
    },
    {
      "epoch": 1.8745874587458746,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.028117161716171617,
      "loss": 0.2326,
      "num_input_tokens_seen": 3595104,
      "step": 17040
    },
    {
      "epoch": 1.8751375137513753,
      "grad_norm": 0.01397705078125,
      "learning_rate": 0.028125412541254124,
      "loss": 0.23,
      "num_input_tokens_seen": 3596224,
      "step": 17045
    },
    {
      "epoch": 1.8756875687568757,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.028133663366336634,
      "loss": 0.2347,
      "num_input_tokens_seen": 3597248,
      "step": 17050
    },
    {
      "epoch": 1.8762376237623761,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.028141914191419138,
      "loss": 0.233,
      "num_input_tokens_seen": 3598208,
      "step": 17055
    },
    {
      "epoch": 1.8767876787678768,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.02815016501650165,
      "loss": 0.233,
      "num_input_tokens_seen": 3599232,
      "step": 17060
    },
    {
      "epoch": 1.8773377337733774,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.028158415841584156,
      "loss": 0.2314,
      "num_input_tokens_seen": 3600320,
      "step": 17065
    },
    {
      "epoch": 1.8778877887788779,
      "grad_norm": 0.007171630859375,
      "learning_rate": 0.028166666666666666,
      "loss": 0.2314,
      "num_input_tokens_seen": 3601408,
      "step": 17070
    },
    {
      "epoch": 1.8784378437843783,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.028174917491749177,
      "loss": 0.2304,
      "num_input_tokens_seen": 3602464,
      "step": 17075
    },
    {
      "epoch": 1.878987898789879,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.02818316831683168,
      "loss": 0.2288,
      "num_input_tokens_seen": 3603520,
      "step": 17080
    },
    {
      "epoch": 1.8795379537953796,
      "grad_norm": 0.007171630859375,
      "learning_rate": 0.02819141914191419,
      "loss": 0.2329,
      "num_input_tokens_seen": 3604608,
      "step": 17085
    },
    {
      "epoch": 1.8800880088008802,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.028199669966996698,
      "loss": 0.2324,
      "num_input_tokens_seen": 3605664,
      "step": 17090
    },
    {
      "epoch": 1.8806380638063807,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02820792079207921,
      "loss": 0.2319,
      "num_input_tokens_seen": 3606688,
      "step": 17095
    },
    {
      "epoch": 1.881188118811881,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.028216171617161712,
      "loss": 0.234,
      "num_input_tokens_seen": 3607744,
      "step": 17100
    },
    {
      "epoch": 1.8817381738173817,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.028224422442244223,
      "loss": 0.2314,
      "num_input_tokens_seen": 3608832,
      "step": 17105
    },
    {
      "epoch": 1.8822882288228824,
      "grad_norm": 0.000732421875,
      "learning_rate": 0.02823267326732673,
      "loss": 0.2314,
      "num_input_tokens_seen": 3609824,
      "step": 17110
    },
    {
      "epoch": 1.8828382838283828,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.02824092409240924,
      "loss": 0.2313,
      "num_input_tokens_seen": 3610880,
      "step": 17115
    },
    {
      "epoch": 1.8833883388338832,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.02824917491749175,
      "loss": 0.2335,
      "num_input_tokens_seen": 3611904,
      "step": 17120
    },
    {
      "epoch": 1.8839383938393839,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.028257425742574255,
      "loss": 0.2314,
      "num_input_tokens_seen": 3612896,
      "step": 17125
    },
    {
      "epoch": 1.8844884488448845,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.028265676567656765,
      "loss": 0.2303,
      "num_input_tokens_seen": 3613856,
      "step": 17130
    },
    {
      "epoch": 1.8850385038503852,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.028273927392739272,
      "loss": 0.232,
      "num_input_tokens_seen": 3614944,
      "step": 17135
    },
    {
      "epoch": 1.8855885588558856,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.028282178217821783,
      "loss": 0.2325,
      "num_input_tokens_seen": 3616032,
      "step": 17140
    },
    {
      "epoch": 1.886138613861386,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.028290429042904287,
      "loss": 0.2278,
      "num_input_tokens_seen": 3617056,
      "step": 17145
    },
    {
      "epoch": 1.8866886688668867,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.028298679867986797,
      "loss": 0.2289,
      "num_input_tokens_seen": 3618176,
      "step": 17150
    },
    {
      "epoch": 1.8872387238723873,
      "grad_norm": 0.0076904296875,
      "learning_rate": 0.028306930693069304,
      "loss": 0.2341,
      "num_input_tokens_seen": 3619232,
      "step": 17155
    },
    {
      "epoch": 1.8877887788778878,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.028315181518151815,
      "loss": 0.2278,
      "num_input_tokens_seen": 3620320,
      "step": 17160
    },
    {
      "epoch": 1.8883388338833882,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.028323432343234325,
      "loss": 0.2274,
      "num_input_tokens_seen": 3621376,
      "step": 17165
    },
    {
      "epoch": 1.8888888888888888,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.02833168316831683,
      "loss": 0.2306,
      "num_input_tokens_seen": 3622368,
      "step": 17170
    },
    {
      "epoch": 1.8894389438943895,
      "grad_norm": 0.007171630859375,
      "learning_rate": 0.02833993399339934,
      "loss": 0.2312,
      "num_input_tokens_seen": 3623392,
      "step": 17175
    },
    {
      "epoch": 1.8899889988998901,
      "grad_norm": 0.0081787109375,
      "learning_rate": 0.028348184818481847,
      "loss": 0.237,
      "num_input_tokens_seen": 3624416,
      "step": 17180
    },
    {
      "epoch": 1.8905390539053906,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.028356435643564357,
      "loss": 0.2337,
      "num_input_tokens_seen": 3625440,
      "step": 17185
    },
    {
      "epoch": 1.891089108910891,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.02836468646864686,
      "loss": 0.2351,
      "num_input_tokens_seen": 3626496,
      "step": 17190
    },
    {
      "epoch": 1.8916391639163916,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.02837293729372937,
      "loss": 0.233,
      "num_input_tokens_seen": 3627456,
      "step": 17195
    },
    {
      "epoch": 1.8921892189218923,
      "grad_norm": 0.01336669921875,
      "learning_rate": 0.02838118811881188,
      "loss": 0.2329,
      "num_input_tokens_seen": 3628480,
      "step": 17200
    },
    {
      "epoch": 1.8927392739273927,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.02838943894389439,
      "loss": 0.2324,
      "num_input_tokens_seen": 3629504,
      "step": 17205
    },
    {
      "epoch": 1.8932893289328931,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.0283976897689769,
      "loss": 0.2314,
      "num_input_tokens_seen": 3630592,
      "step": 17210
    },
    {
      "epoch": 1.8938393839383938,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.028405940594059403,
      "loss": 0.2303,
      "num_input_tokens_seen": 3631712,
      "step": 17215
    },
    {
      "epoch": 1.8943894389438944,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.028414191419141914,
      "loss": 0.2293,
      "num_input_tokens_seen": 3632768,
      "step": 17220
    },
    {
      "epoch": 1.894939493949395,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.02842244224422442,
      "loss": 0.2324,
      "num_input_tokens_seen": 3633856,
      "step": 17225
    },
    {
      "epoch": 1.8954895489548955,
      "grad_norm": 0.0135498046875,
      "learning_rate": 0.02843069306930693,
      "loss": 0.2319,
      "num_input_tokens_seen": 3634944,
      "step": 17230
    },
    {
      "epoch": 1.896039603960396,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.028438943894389435,
      "loss": 0.232,
      "num_input_tokens_seen": 3635968,
      "step": 17235
    },
    {
      "epoch": 1.8965896589658966,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.028447194719471946,
      "loss": 0.2324,
      "num_input_tokens_seen": 3637056,
      "step": 17240
    },
    {
      "epoch": 1.8971397139713972,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.028455445544554453,
      "loss": 0.2314,
      "num_input_tokens_seen": 3638080,
      "step": 17245
    },
    {
      "epoch": 1.8976897689768977,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.028463696369636964,
      "loss": 0.2319,
      "num_input_tokens_seen": 3639104,
      "step": 17250
    },
    {
      "epoch": 1.898239823982398,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.028471947194719474,
      "loss": 0.2335,
      "num_input_tokens_seen": 3640256,
      "step": 17255
    },
    {
      "epoch": 1.8987898789878987,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.028480198019801978,
      "loss": 0.2319,
      "num_input_tokens_seen": 3641280,
      "step": 17260
    },
    {
      "epoch": 1.8993399339933994,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.02848844884488449,
      "loss": 0.2319,
      "num_input_tokens_seen": 3642336,
      "step": 17265
    },
    {
      "epoch": 1.8998899889989,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.028496699669966995,
      "loss": 0.2309,
      "num_input_tokens_seen": 3643392,
      "step": 17270
    },
    {
      "epoch": 1.9004400440044005,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.028504950495049506,
      "loss": 0.2314,
      "num_input_tokens_seen": 3644480,
      "step": 17275
    },
    {
      "epoch": 1.900990099009901,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.02851320132013201,
      "loss": 0.2294,
      "num_input_tokens_seen": 3645568,
      "step": 17280
    },
    {
      "epoch": 1.9015401540154016,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.02852145214521452,
      "loss": 0.233,
      "num_input_tokens_seen": 3646592,
      "step": 17285
    },
    {
      "epoch": 1.9020902090209022,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.028529702970297027,
      "loss": 0.2324,
      "num_input_tokens_seen": 3647616,
      "step": 17290
    },
    {
      "epoch": 1.9026402640264026,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.028537953795379538,
      "loss": 0.2314,
      "num_input_tokens_seen": 3648704,
      "step": 17295
    },
    {
      "epoch": 1.903190319031903,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.028546204620462045,
      "loss": 0.2308,
      "num_input_tokens_seen": 3649792,
      "step": 17300
    },
    {
      "epoch": 1.9037403740374037,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.028554455445544552,
      "loss": 0.2314,
      "num_input_tokens_seen": 3650880,
      "step": 17305
    },
    {
      "epoch": 1.9042904290429044,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.028562706270627063,
      "loss": 0.2308,
      "num_input_tokens_seen": 3652000,
      "step": 17310
    },
    {
      "epoch": 1.904840484048405,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02857095709570957,
      "loss": 0.234,
      "num_input_tokens_seen": 3653024,
      "step": 17315
    },
    {
      "epoch": 1.9053905390539054,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.02857920792079208,
      "loss": 0.2329,
      "num_input_tokens_seen": 3654144,
      "step": 17320
    },
    {
      "epoch": 1.9059405940594059,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.028587458745874584,
      "loss": 0.2319,
      "num_input_tokens_seen": 3655168,
      "step": 17325
    },
    {
      "epoch": 1.9064906490649065,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.028595709570957095,
      "loss": 0.2319,
      "num_input_tokens_seen": 3656160,
      "step": 17330
    },
    {
      "epoch": 1.9070407040704072,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.0286039603960396,
      "loss": 0.2314,
      "num_input_tokens_seen": 3657312,
      "step": 17335
    },
    {
      "epoch": 1.9075907590759076,
      "grad_norm": 0.000728607177734375,
      "learning_rate": 0.028612211221122112,
      "loss": 0.2314,
      "num_input_tokens_seen": 3658400,
      "step": 17340
    },
    {
      "epoch": 1.908140814081408,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.02862046204620462,
      "loss": 0.233,
      "num_input_tokens_seen": 3659424,
      "step": 17345
    },
    {
      "epoch": 1.9086908690869087,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.028628712871287126,
      "loss": 0.2289,
      "num_input_tokens_seen": 3660544,
      "step": 17350
    },
    {
      "epoch": 1.9092409240924093,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.028636963696369637,
      "loss": 0.2314,
      "num_input_tokens_seen": 3661600,
      "step": 17355
    },
    {
      "epoch": 1.9097909790979097,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.028645214521452144,
      "loss": 0.2232,
      "num_input_tokens_seen": 3662656,
      "step": 17360
    },
    {
      "epoch": 1.9103410341034104,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.028653465346534655,
      "loss": 0.2327,
      "num_input_tokens_seen": 3663712,
      "step": 17365
    },
    {
      "epoch": 1.9108910891089108,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.028661716171617158,
      "loss": 0.2344,
      "num_input_tokens_seen": 3664736,
      "step": 17370
    },
    {
      "epoch": 1.9114411441144115,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.02866996699669967,
      "loss": 0.2291,
      "num_input_tokens_seen": 3665888,
      "step": 17375
    },
    {
      "epoch": 1.911991199119912,
      "grad_norm": 0.007720947265625,
      "learning_rate": 0.028678217821782176,
      "loss": 0.2343,
      "num_input_tokens_seen": 3666912,
      "step": 17380
    },
    {
      "epoch": 1.9125412541254125,
      "grad_norm": 0.007568359375,
      "learning_rate": 0.028686468646864687,
      "loss": 0.2416,
      "num_input_tokens_seen": 3667936,
      "step": 17385
    },
    {
      "epoch": 1.913091309130913,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.028694719471947194,
      "loss": 0.23,
      "num_input_tokens_seen": 3669056,
      "step": 17390
    },
    {
      "epoch": 1.9136413641364136,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.0287029702970297,
      "loss": 0.231,
      "num_input_tokens_seen": 3670112,
      "step": 17395
    },
    {
      "epoch": 1.9141914191419143,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.02871122112211221,
      "loss": 0.2362,
      "num_input_tokens_seen": 3671104,
      "step": 17400
    },
    {
      "epoch": 1.9147414741474147,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.02871947194719472,
      "loss": 0.232,
      "num_input_tokens_seen": 3672128,
      "step": 17405
    },
    {
      "epoch": 1.9152915291529153,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.02872772277227723,
      "loss": 0.2308,
      "num_input_tokens_seen": 3673248,
      "step": 17410
    },
    {
      "epoch": 1.9158415841584158,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.028735973597359733,
      "loss": 0.2319,
      "num_input_tokens_seen": 3674240,
      "step": 17415
    },
    {
      "epoch": 1.9163916391639164,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.028744224422442243,
      "loss": 0.2313,
      "num_input_tokens_seen": 3675392,
      "step": 17420
    },
    {
      "epoch": 1.916941694169417,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.02875247524752475,
      "loss": 0.2329,
      "num_input_tokens_seen": 3676544,
      "step": 17425
    },
    {
      "epoch": 1.9174917491749175,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.02876072607260726,
      "loss": 0.2329,
      "num_input_tokens_seen": 3677568,
      "step": 17430
    },
    {
      "epoch": 1.918041804180418,
      "grad_norm": 0.007080078125,
      "learning_rate": 0.028768976897689768,
      "loss": 0.2303,
      "num_input_tokens_seen": 3678592,
      "step": 17435
    },
    {
      "epoch": 1.9185918591859186,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.028777227722772275,
      "loss": 0.2319,
      "num_input_tokens_seen": 3679616,
      "step": 17440
    },
    {
      "epoch": 1.9191419141914192,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.028785478547854786,
      "loss": 0.2298,
      "num_input_tokens_seen": 3680576,
      "step": 17445
    },
    {
      "epoch": 1.9196919691969196,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.028793729372937293,
      "loss": 0.2314,
      "num_input_tokens_seen": 3681632,
      "step": 17450
    },
    {
      "epoch": 1.9202420242024203,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.028801980198019803,
      "loss": 0.2309,
      "num_input_tokens_seen": 3682656,
      "step": 17455
    },
    {
      "epoch": 1.9207920792079207,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.028810231023102307,
      "loss": 0.2327,
      "num_input_tokens_seen": 3683680,
      "step": 17460
    },
    {
      "epoch": 1.9213421342134214,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.028818481848184817,
      "loss": 0.2274,
      "num_input_tokens_seen": 3684704,
      "step": 17465
    },
    {
      "epoch": 1.921892189218922,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.028826732673267325,
      "loss": 0.2321,
      "num_input_tokens_seen": 3685728,
      "step": 17470
    },
    {
      "epoch": 1.9224422442244224,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.028834983498349835,
      "loss": 0.2312,
      "num_input_tokens_seen": 3686784,
      "step": 17475
    },
    {
      "epoch": 1.9229922992299229,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.028843234323432342,
      "loss": 0.2343,
      "num_input_tokens_seen": 3687808,
      "step": 17480
    },
    {
      "epoch": 1.9235423542354235,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.02885148514851485,
      "loss": 0.2305,
      "num_input_tokens_seen": 3688832,
      "step": 17485
    },
    {
      "epoch": 1.9240924092409242,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.02885973597359736,
      "loss": 0.2305,
      "num_input_tokens_seen": 3689920,
      "step": 17490
    },
    {
      "epoch": 1.9246424642464246,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.028867986798679867,
      "loss": 0.23,
      "num_input_tokens_seen": 3691008,
      "step": 17495
    },
    {
      "epoch": 1.925192519251925,
      "grad_norm": 0.007598876953125,
      "learning_rate": 0.028876237623762378,
      "loss": 0.2327,
      "num_input_tokens_seen": 3692096,
      "step": 17500
    },
    {
      "epoch": 1.9257425742574257,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.02888448844884488,
      "loss": 0.2316,
      "num_input_tokens_seen": 3693184,
      "step": 17505
    },
    {
      "epoch": 1.9262926292629263,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.028892739273927392,
      "loss": 0.2306,
      "num_input_tokens_seen": 3694240,
      "step": 17510
    },
    {
      "epoch": 1.926842684268427,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.0289009900990099,
      "loss": 0.2358,
      "num_input_tokens_seen": 3695328,
      "step": 17515
    },
    {
      "epoch": 1.9273927392739274,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.02890924092409241,
      "loss": 0.2285,
      "num_input_tokens_seen": 3696448,
      "step": 17520
    },
    {
      "epoch": 1.9279427942794278,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.028917491749174917,
      "loss": 0.2307,
      "num_input_tokens_seen": 3697536,
      "step": 17525
    },
    {
      "epoch": 1.9284928492849285,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.028925742574257424,
      "loss": 0.2306,
      "num_input_tokens_seen": 3698624,
      "step": 17530
    },
    {
      "epoch": 1.9290429042904291,
      "grad_norm": 0.014404296875,
      "learning_rate": 0.028933993399339934,
      "loss": 0.2317,
      "num_input_tokens_seen": 3699648,
      "step": 17535
    },
    {
      "epoch": 1.9295929592959296,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.02894224422442244,
      "loss": 0.2301,
      "num_input_tokens_seen": 3700768,
      "step": 17540
    },
    {
      "epoch": 1.93014301430143,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.028950495049504952,
      "loss": 0.2234,
      "num_input_tokens_seen": 3701824,
      "step": 17545
    },
    {
      "epoch": 1.9306930693069306,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.028958745874587456,
      "loss": 0.2297,
      "num_input_tokens_seen": 3702848,
      "step": 17550
    },
    {
      "epoch": 1.9312431243124313,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.028966996699669966,
      "loss": 0.2325,
      "num_input_tokens_seen": 3703968,
      "step": 17555
    },
    {
      "epoch": 1.931793179317932,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.028975247524752473,
      "loss": 0.2212,
      "num_input_tokens_seen": 3704992,
      "step": 17560
    },
    {
      "epoch": 1.9323432343234324,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.028983498349834984,
      "loss": 0.2312,
      "num_input_tokens_seen": 3706080,
      "step": 17565
    },
    {
      "epoch": 1.9328932893289328,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.02899174917491749,
      "loss": 0.2326,
      "num_input_tokens_seen": 3707168,
      "step": 17570
    },
    {
      "epoch": 1.9334433443344334,
      "grad_norm": 0.0087890625,
      "learning_rate": 0.028999999999999998,
      "loss": 0.2345,
      "num_input_tokens_seen": 3708224,
      "step": 17575
    },
    {
      "epoch": 1.933993399339934,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.029008250825082505,
      "loss": 0.2356,
      "num_input_tokens_seen": 3709312,
      "step": 17580
    },
    {
      "epoch": 1.9345434543454345,
      "grad_norm": 0.0150146484375,
      "learning_rate": 0.029016501650165016,
      "loss": 0.2355,
      "num_input_tokens_seen": 3710336,
      "step": 17585
    },
    {
      "epoch": 1.935093509350935,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.029024752475247526,
      "loss": 0.2382,
      "num_input_tokens_seen": 3711392,
      "step": 17590
    },
    {
      "epoch": 1.9356435643564356,
      "grad_norm": 0.01275634765625,
      "learning_rate": 0.02903300330033003,
      "loss": 0.2239,
      "num_input_tokens_seen": 3712448,
      "step": 17595
    },
    {
      "epoch": 1.9361936193619362,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02904125412541254,
      "loss": 0.2254,
      "num_input_tokens_seen": 3713504,
      "step": 17600
    },
    {
      "epoch": 1.9367436743674369,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.029049504950495048,
      "loss": 0.2298,
      "num_input_tokens_seen": 3714496,
      "step": 17605
    },
    {
      "epoch": 1.9372937293729373,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029057755775577558,
      "loss": 0.2272,
      "num_input_tokens_seen": 3715520,
      "step": 17610
    },
    {
      "epoch": 1.9378437843784377,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029066006600660065,
      "loss": 0.2368,
      "num_input_tokens_seen": 3716544,
      "step": 17615
    },
    {
      "epoch": 1.9383938393839384,
      "grad_norm": 0.014404296875,
      "learning_rate": 0.029074257425742572,
      "loss": 0.2346,
      "num_input_tokens_seen": 3717568,
      "step": 17620
    },
    {
      "epoch": 1.938943894389439,
      "grad_norm": 0.01446533203125,
      "learning_rate": 0.02908250825082508,
      "loss": 0.234,
      "num_input_tokens_seen": 3718528,
      "step": 17625
    },
    {
      "epoch": 1.9394939493949395,
      "grad_norm": 0.01275634765625,
      "learning_rate": 0.02909075907590759,
      "loss": 0.2252,
      "num_input_tokens_seen": 3719552,
      "step": 17630
    },
    {
      "epoch": 1.94004400440044,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0290990099009901,
      "loss": 0.235,
      "num_input_tokens_seen": 3720640,
      "step": 17635
    },
    {
      "epoch": 1.9405940594059405,
      "grad_norm": 0.007781982421875,
      "learning_rate": 0.029107260726072604,
      "loss": 0.2303,
      "num_input_tokens_seen": 3721760,
      "step": 17640
    },
    {
      "epoch": 1.9411441144114412,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.029115511551155115,
      "loss": 0.2272,
      "num_input_tokens_seen": 3722784,
      "step": 17645
    },
    {
      "epoch": 1.9416941694169418,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.029123762376237622,
      "loss": 0.2382,
      "num_input_tokens_seen": 3723840,
      "step": 17650
    },
    {
      "epoch": 1.9422442244224423,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.029132013201320132,
      "loss": 0.2313,
      "num_input_tokens_seen": 3724896,
      "step": 17655
    },
    {
      "epoch": 1.9427942794279427,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.02914026402640264,
      "loss": 0.2327,
      "num_input_tokens_seen": 3726048,
      "step": 17660
    },
    {
      "epoch": 1.9433443344334433,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.029148514851485147,
      "loss": 0.2295,
      "num_input_tokens_seen": 3727104,
      "step": 17665
    },
    {
      "epoch": 1.943894389438944,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.029156765676567654,
      "loss": 0.2353,
      "num_input_tokens_seen": 3728128,
      "step": 17670
    },
    {
      "epoch": 1.9444444444444444,
      "grad_norm": 0.00762939453125,
      "learning_rate": 0.029165016501650164,
      "loss": 0.2264,
      "num_input_tokens_seen": 3729184,
      "step": 17675
    },
    {
      "epoch": 1.9449944994499448,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.029173267326732675,
      "loss": 0.2301,
      "num_input_tokens_seen": 3730208,
      "step": 17680
    },
    {
      "epoch": 1.9455445544554455,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.02918151815181518,
      "loss": 0.2327,
      "num_input_tokens_seen": 3731296,
      "step": 17685
    },
    {
      "epoch": 1.9460946094609461,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.02918976897689769,
      "loss": 0.2364,
      "num_input_tokens_seen": 3732416,
      "step": 17690
    },
    {
      "epoch": 1.9466446644664468,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.029198019801980196,
      "loss": 0.2317,
      "num_input_tokens_seen": 3733472,
      "step": 17695
    },
    {
      "epoch": 1.9471947194719472,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.029206270627062707,
      "loss": 0.229,
      "num_input_tokens_seen": 3734560,
      "step": 17700
    },
    {
      "epoch": 1.9477447744774476,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029214521452145214,
      "loss": 0.2265,
      "num_input_tokens_seen": 3735616,
      "step": 17705
    },
    {
      "epoch": 1.9482948294829483,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02922277227722772,
      "loss": 0.226,
      "num_input_tokens_seen": 3736672,
      "step": 17710
    },
    {
      "epoch": 1.948844884488449,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.029231023102310228,
      "loss": 0.2371,
      "num_input_tokens_seen": 3737696,
      "step": 17715
    },
    {
      "epoch": 1.9493949394939494,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.02923927392739274,
      "loss": 0.2343,
      "num_input_tokens_seen": 3738688,
      "step": 17720
    },
    {
      "epoch": 1.9499449944994498,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02924752475247525,
      "loss": 0.2292,
      "num_input_tokens_seen": 3739744,
      "step": 17725
    },
    {
      "epoch": 1.9504950495049505,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.029255775577557753,
      "loss": 0.238,
      "num_input_tokens_seen": 3740768,
      "step": 17730
    },
    {
      "epoch": 1.951045104510451,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.029264026402640263,
      "loss": 0.2318,
      "num_input_tokens_seen": 3741824,
      "step": 17735
    },
    {
      "epoch": 1.9515951595159517,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.02927227722772277,
      "loss": 0.2332,
      "num_input_tokens_seen": 3742880,
      "step": 17740
    },
    {
      "epoch": 1.9521452145214522,
      "grad_norm": 0.007415771484375,
      "learning_rate": 0.02928052805280528,
      "loss": 0.2327,
      "num_input_tokens_seen": 3744032,
      "step": 17745
    },
    {
      "epoch": 1.9526952695269526,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029288778877887788,
      "loss": 0.23,
      "num_input_tokens_seen": 3745120,
      "step": 17750
    },
    {
      "epoch": 1.9532453245324533,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.029297029702970295,
      "loss": 0.2379,
      "num_input_tokens_seen": 3746144,
      "step": 17755
    },
    {
      "epoch": 1.953795379537954,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.029305280528052802,
      "loss": 0.2357,
      "num_input_tokens_seen": 3747168,
      "step": 17760
    },
    {
      "epoch": 1.9543454345434543,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029313531353135313,
      "loss": 0.232,
      "num_input_tokens_seen": 3748192,
      "step": 17765
    },
    {
      "epoch": 1.9548954895489548,
      "grad_norm": 0.0027313232421875,
      "learning_rate": 0.029321782178217824,
      "loss": 0.2304,
      "num_input_tokens_seen": 3749312,
      "step": 17770
    },
    {
      "epoch": 1.9554455445544554,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.029330033003300327,
      "loss": 0.2309,
      "num_input_tokens_seen": 3750336,
      "step": 17775
    },
    {
      "epoch": 1.955995599559956,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.029338283828382838,
      "loss": 0.2335,
      "num_input_tokens_seen": 3751360,
      "step": 17780
    },
    {
      "epoch": 1.9565456545654567,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.029346534653465345,
      "loss": 0.2319,
      "num_input_tokens_seen": 3752448,
      "step": 17785
    },
    {
      "epoch": 1.9570957095709571,
      "grad_norm": 0.0072021484375,
      "learning_rate": 0.029354785478547855,
      "loss": 0.2324,
      "num_input_tokens_seen": 3753504,
      "step": 17790
    },
    {
      "epoch": 1.9576457645764576,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.029363036303630363,
      "loss": 0.2309,
      "num_input_tokens_seen": 3754592,
      "step": 17795
    },
    {
      "epoch": 1.9581958195819582,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.02937128712871287,
      "loss": 0.232,
      "num_input_tokens_seen": 3755680,
      "step": 17800
    },
    {
      "epoch": 1.9587458745874589,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.029379537953795377,
      "loss": 0.2311,
      "num_input_tokens_seen": 3756672,
      "step": 17805
    },
    {
      "epoch": 1.9592959295929593,
      "grad_norm": 0.0074462890625,
      "learning_rate": 0.029387788778877887,
      "loss": 0.2322,
      "num_input_tokens_seen": 3757696,
      "step": 17810
    },
    {
      "epoch": 1.9598459845984597,
      "grad_norm": 0.0150146484375,
      "learning_rate": 0.029396039603960394,
      "loss": 0.2311,
      "num_input_tokens_seen": 3758720,
      "step": 17815
    },
    {
      "epoch": 1.9603960396039604,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.0294042904290429,
      "loss": 0.2364,
      "num_input_tokens_seen": 3759808,
      "step": 17820
    },
    {
      "epoch": 1.960946094609461,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.029412541254125412,
      "loss": 0.2357,
      "num_input_tokens_seen": 3760864,
      "step": 17825
    },
    {
      "epoch": 1.9614961496149617,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.02942079207920792,
      "loss": 0.233,
      "num_input_tokens_seen": 3761856,
      "step": 17830
    },
    {
      "epoch": 1.962046204620462,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.02942904290429043,
      "loss": 0.2335,
      "num_input_tokens_seen": 3762912,
      "step": 17835
    },
    {
      "epoch": 1.9625962596259625,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029437293729372937,
      "loss": 0.2309,
      "num_input_tokens_seen": 3764000,
      "step": 17840
    },
    {
      "epoch": 1.9631463146314632,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.029445544554455444,
      "loss": 0.2309,
      "num_input_tokens_seen": 3765024,
      "step": 17845
    },
    {
      "epoch": 1.9636963696369638,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.02945379537953795,
      "loss": 0.232,
      "num_input_tokens_seen": 3766048,
      "step": 17850
    },
    {
      "epoch": 1.9642464246424642,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.02946204620462046,
      "loss": 0.231,
      "num_input_tokens_seen": 3767072,
      "step": 17855
    },
    {
      "epoch": 1.9647964796479647,
      "grad_norm": 0.0072021484375,
      "learning_rate": 0.02947029702970297,
      "loss": 0.2295,
      "num_input_tokens_seen": 3768160,
      "step": 17860
    },
    {
      "epoch": 1.9653465346534653,
      "grad_norm": 0.013671875,
      "learning_rate": 0.029478547854785476,
      "loss": 0.2384,
      "num_input_tokens_seen": 3769216,
      "step": 17865
    },
    {
      "epoch": 1.965896589658966,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.029486798679867986,
      "loss": 0.2306,
      "num_input_tokens_seen": 3770304,
      "step": 17870
    },
    {
      "epoch": 1.9664466446644664,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.029495049504950493,
      "loss": 0.2351,
      "num_input_tokens_seen": 3771392,
      "step": 17875
    },
    {
      "epoch": 1.966996699669967,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.029503300330033004,
      "loss": 0.2319,
      "num_input_tokens_seen": 3772384,
      "step": 17880
    },
    {
      "epoch": 1.9675467546754675,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.02951155115511551,
      "loss": 0.2325,
      "num_input_tokens_seen": 3773472,
      "step": 17885
    },
    {
      "epoch": 1.9680968096809681,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.029519801980198018,
      "loss": 0.2314,
      "num_input_tokens_seen": 3774528,
      "step": 17890
    },
    {
      "epoch": 1.9686468646864688,
      "grad_norm": 0.01251220703125,
      "learning_rate": 0.029528052805280525,
      "loss": 0.2319,
      "num_input_tokens_seen": 3775584,
      "step": 17895
    },
    {
      "epoch": 1.9691969196919692,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.029536303630363036,
      "loss": 0.2319,
      "num_input_tokens_seen": 3776672,
      "step": 17900
    },
    {
      "epoch": 1.9697469746974696,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.029544554455445543,
      "loss": 0.2329,
      "num_input_tokens_seen": 3777696,
      "step": 17905
    },
    {
      "epoch": 1.9702970297029703,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.02955280528052805,
      "loss": 0.2324,
      "num_input_tokens_seen": 3778752,
      "step": 17910
    },
    {
      "epoch": 1.970847084708471,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.02956105610561056,
      "loss": 0.2319,
      "num_input_tokens_seen": 3779808,
      "step": 17915
    },
    {
      "epoch": 1.9713971397139713,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.029569306930693068,
      "loss": 0.2314,
      "num_input_tokens_seen": 3780832,
      "step": 17920
    },
    {
      "epoch": 1.971947194719472,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.02957755775577558,
      "loss": 0.2293,
      "num_input_tokens_seen": 3781856,
      "step": 17925
    },
    {
      "epoch": 1.9724972497249724,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.029585808580858085,
      "loss": 0.2283,
      "num_input_tokens_seen": 3782976,
      "step": 17930
    },
    {
      "epoch": 1.973047304730473,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029594059405940593,
      "loss": 0.229,
      "num_input_tokens_seen": 3784000,
      "step": 17935
    },
    {
      "epoch": 1.9735973597359737,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.0296023102310231,
      "loss": 0.2295,
      "num_input_tokens_seen": 3785056,
      "step": 17940
    },
    {
      "epoch": 1.9741474147414741,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02961056105610561,
      "loss": 0.2303,
      "num_input_tokens_seen": 3786080,
      "step": 17945
    },
    {
      "epoch": 1.9746974697469746,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.029618811881188117,
      "loss": 0.2372,
      "num_input_tokens_seen": 3787136,
      "step": 17950
    },
    {
      "epoch": 1.9752475247524752,
      "grad_norm": 0.007598876953125,
      "learning_rate": 0.029627062706270624,
      "loss": 0.2315,
      "num_input_tokens_seen": 3788224,
      "step": 17955
    },
    {
      "epoch": 1.9757975797579759,
      "grad_norm": 0.0078125,
      "learning_rate": 0.029635313531353135,
      "loss": 0.2355,
      "num_input_tokens_seen": 3789376,
      "step": 17960
    },
    {
      "epoch": 1.9763476347634763,
      "grad_norm": 0.0074462890625,
      "learning_rate": 0.029643564356435642,
      "loss": 0.2354,
      "num_input_tokens_seen": 3790432,
      "step": 17965
    },
    {
      "epoch": 1.976897689768977,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.029651815181518153,
      "loss": 0.2285,
      "num_input_tokens_seen": 3791520,
      "step": 17970
    },
    {
      "epoch": 1.9774477447744774,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.02966006600660066,
      "loss": 0.2338,
      "num_input_tokens_seen": 3792608,
      "step": 17975
    },
    {
      "epoch": 1.977997799779978,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.029668316831683167,
      "loss": 0.2312,
      "num_input_tokens_seen": 3793600,
      "step": 17980
    },
    {
      "epoch": 1.9785478547854787,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.029676567656765674,
      "loss": 0.2321,
      "num_input_tokens_seen": 3794592,
      "step": 17985
    },
    {
      "epoch": 1.979097909790979,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.029684818481848185,
      "loss": 0.2336,
      "num_input_tokens_seen": 3795584,
      "step": 17990
    },
    {
      "epoch": 1.9796479647964795,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.02969306930693069,
      "loss": 0.2295,
      "num_input_tokens_seen": 3796640,
      "step": 17995
    },
    {
      "epoch": 1.9801980198019802,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.0297013201320132,
      "loss": 0.2286,
      "num_input_tokens_seen": 3797728,
      "step": 18000
    },
    {
      "epoch": 1.9807480748074808,
      "grad_norm": 0.00750732421875,
      "learning_rate": 0.02970957095709571,
      "loss": 0.2306,
      "num_input_tokens_seen": 3798784,
      "step": 18005
    },
    {
      "epoch": 1.9812981298129813,
      "grad_norm": 0.00726318359375,
      "learning_rate": 0.029717821782178216,
      "loss": 0.2337,
      "num_input_tokens_seen": 3799808,
      "step": 18010
    },
    {
      "epoch": 1.9818481848184817,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.029726072607260727,
      "loss": 0.2331,
      "num_input_tokens_seen": 3800928,
      "step": 18015
    },
    {
      "epoch": 1.9823982398239823,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.029734323432343234,
      "loss": 0.2305,
      "num_input_tokens_seen": 3802016,
      "step": 18020
    },
    {
      "epoch": 1.982948294829483,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.02974257425742574,
      "loss": 0.2331,
      "num_input_tokens_seen": 3803008,
      "step": 18025
    },
    {
      "epoch": 1.9834983498349836,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.02975082508250825,
      "loss": 0.2305,
      "num_input_tokens_seen": 3804096,
      "step": 18030
    },
    {
      "epoch": 1.984048404840484,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.02975907590759076,
      "loss": 0.2284,
      "num_input_tokens_seen": 3805120,
      "step": 18035
    },
    {
      "epoch": 1.9845984598459845,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.029767326732673266,
      "loss": 0.2284,
      "num_input_tokens_seen": 3806176,
      "step": 18040
    },
    {
      "epoch": 1.9851485148514851,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029775577557755773,
      "loss": 0.231,
      "num_input_tokens_seen": 3807264,
      "step": 18045
    },
    {
      "epoch": 1.9856985698569858,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.029783828382838284,
      "loss": 0.2249,
      "num_input_tokens_seen": 3808320,
      "step": 18050
    },
    {
      "epoch": 1.9862486248624862,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02979207920792079,
      "loss": 0.2359,
      "num_input_tokens_seen": 3809408,
      "step": 18055
    },
    {
      "epoch": 1.9867986798679866,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.0298003300330033,
      "loss": 0.229,
      "num_input_tokens_seen": 3810528,
      "step": 18060
    },
    {
      "epoch": 1.9873487348734873,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02980858085808581,
      "loss": 0.2291,
      "num_input_tokens_seen": 3811584,
      "step": 18065
    },
    {
      "epoch": 1.987898789878988,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029816831683168316,
      "loss": 0.2251,
      "num_input_tokens_seen": 3812704,
      "step": 18070
    },
    {
      "epoch": 1.9884488448844886,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.029825082508250823,
      "loss": 0.2274,
      "num_input_tokens_seen": 3813760,
      "step": 18075
    },
    {
      "epoch": 1.988998899889989,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.029833333333333333,
      "loss": 0.2321,
      "num_input_tokens_seen": 3814784,
      "step": 18080
    },
    {
      "epoch": 1.9895489548954894,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.02984158415841584,
      "loss": 0.2259,
      "num_input_tokens_seen": 3815872,
      "step": 18085
    },
    {
      "epoch": 1.99009900990099,
      "grad_norm": 0.0081787109375,
      "learning_rate": 0.029849834983498347,
      "loss": 0.2416,
      "num_input_tokens_seen": 3816992,
      "step": 18090
    },
    {
      "epoch": 1.9906490649064907,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.029858085808580855,
      "loss": 0.2357,
      "num_input_tokens_seen": 3818080,
      "step": 18095
    },
    {
      "epoch": 1.9911991199119912,
      "grad_norm": 0.00775146484375,
      "learning_rate": 0.029866336633663365,
      "loss": 0.2372,
      "num_input_tokens_seen": 3819136,
      "step": 18100
    },
    {
      "epoch": 1.9917491749174916,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029874587458745876,
      "loss": 0.2306,
      "num_input_tokens_seen": 3820128,
      "step": 18105
    },
    {
      "epoch": 1.9922992299229922,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029882838283828383,
      "loss": 0.2315,
      "num_input_tokens_seen": 3821152,
      "step": 18110
    },
    {
      "epoch": 1.992849284928493,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.02989108910891089,
      "loss": 0.2294,
      "num_input_tokens_seen": 3822176,
      "step": 18115
    },
    {
      "epoch": 1.9933993399339935,
      "grad_norm": 0.007537841796875,
      "learning_rate": 0.029899339933993397,
      "loss": 0.2305,
      "num_input_tokens_seen": 3823232,
      "step": 18120
    },
    {
      "epoch": 1.993949394939494,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.029907590759075908,
      "loss": 0.2325,
      "num_input_tokens_seen": 3824288,
      "step": 18125
    },
    {
      "epoch": 1.9944994499449944,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.029915841584158415,
      "loss": 0.2351,
      "num_input_tokens_seen": 3825280,
      "step": 18130
    },
    {
      "epoch": 1.995049504950495,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.029924092409240922,
      "loss": 0.2319,
      "num_input_tokens_seen": 3826336,
      "step": 18135
    },
    {
      "epoch": 1.9955995599559957,
      "grad_norm": 0.007080078125,
      "learning_rate": 0.02993234323432343,
      "loss": 0.2314,
      "num_input_tokens_seen": 3827360,
      "step": 18140
    },
    {
      "epoch": 1.9961496149614961,
      "grad_norm": 0.0078125,
      "learning_rate": 0.02994059405940594,
      "loss": 0.2309,
      "num_input_tokens_seen": 3828448,
      "step": 18145
    },
    {
      "epoch": 1.9966996699669965,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.02994884488448845,
      "loss": 0.2309,
      "num_input_tokens_seen": 3829536,
      "step": 18150
    },
    {
      "epoch": 1.9972497249724972,
      "grad_norm": 0.01446533203125,
      "learning_rate": 0.029957095709570957,
      "loss": 0.233,
      "num_input_tokens_seen": 3830560,
      "step": 18155
    },
    {
      "epoch": 1.9977997799779978,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.029965346534653464,
      "loss": 0.232,
      "num_input_tokens_seen": 3831616,
      "step": 18160
    },
    {
      "epoch": 1.9983498349834985,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.02997359735973597,
      "loss": 0.2278,
      "num_input_tokens_seen": 3832672,
      "step": 18165
    },
    {
      "epoch": 1.998899889988999,
      "grad_norm": 0.0033416748046875,
      "learning_rate": 0.029981848184818482,
      "loss": 0.2325,
      "num_input_tokens_seen": 3833792,
      "step": 18170
    },
    {
      "epoch": 1.9994499449944994,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.02999009900990099,
      "loss": 0.2278,
      "num_input_tokens_seen": 3834912,
      "step": 18175
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.0177001953125,
      "learning_rate": 0.029998349834983496,
      "loss": 0.2347,
      "num_input_tokens_seen": 3835840,
      "step": 18180
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.23172347247600555,
      "eval_runtime": 60.558,
      "eval_samples_per_second": 66.713,
      "eval_steps_per_second": 16.678,
      "num_input_tokens_seen": 3835840,
      "step": 18180
    },
    {
      "epoch": 2.0005500550055006,
      "grad_norm": 0.016845703125,
      "learning_rate": 0.0299999999557607,
      "loss": 0.2326,
      "num_input_tokens_seen": 3836896,
      "step": 18185
    },
    {
      "epoch": 2.0011001100110013,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.02999999977603856,
      "loss": 0.2269,
      "num_input_tokens_seen": 3837984,
      "step": 18190
    },
    {
      "epoch": 2.0016501650165015,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.02999999945806862,
      "loss": 0.2374,
      "num_input_tokens_seen": 3839072,
      "step": 18195
    },
    {
      "epoch": 2.002200220022002,
      "grad_norm": 0.0150146484375,
      "learning_rate": 0.029999999001850877,
      "loss": 0.2346,
      "num_input_tokens_seen": 3840096,
      "step": 18200
    },
    {
      "epoch": 2.002750275027503,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.029999998407385344,
      "loss": 0.2309,
      "num_input_tokens_seen": 3841056,
      "step": 18205
    },
    {
      "epoch": 2.0033003300330035,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.029999997674672024,
      "loss": 0.2319,
      "num_input_tokens_seen": 3842144,
      "step": 18210
    },
    {
      "epoch": 2.0038503850385037,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.029999996803710918,
      "loss": 0.2309,
      "num_input_tokens_seen": 3843200,
      "step": 18215
    },
    {
      "epoch": 2.0044004400440043,
      "grad_norm": 0.0128173828125,
      "learning_rate": 0.029999995794502046,
      "loss": 0.2303,
      "num_input_tokens_seen": 3844256,
      "step": 18220
    },
    {
      "epoch": 2.004950495049505,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.029999994647045405,
      "loss": 0.2304,
      "num_input_tokens_seen": 3845344,
      "step": 18225
    },
    {
      "epoch": 2.0055005500550056,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.029999993361341015,
      "loss": 0.2319,
      "num_input_tokens_seen": 3846336,
      "step": 18230
    },
    {
      "epoch": 2.0060506050605063,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029999991937388885,
      "loss": 0.2303,
      "num_input_tokens_seen": 3847392,
      "step": 18235
    },
    {
      "epoch": 2.0066006600660065,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.029999990375189023,
      "loss": 0.2313,
      "num_input_tokens_seen": 3848480,
      "step": 18240
    },
    {
      "epoch": 2.007150715071507,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.029999988674741452,
      "loss": 0.2304,
      "num_input_tokens_seen": 3849536,
      "step": 18245
    },
    {
      "epoch": 2.0077007700770078,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.02999998683604618,
      "loss": 0.234,
      "num_input_tokens_seen": 3850560,
      "step": 18250
    },
    {
      "epoch": 2.0082508250825084,
      "grad_norm": 0.01275634765625,
      "learning_rate": 0.029999984859103226,
      "loss": 0.2298,
      "num_input_tokens_seen": 3851680,
      "step": 18255
    },
    {
      "epoch": 2.0088008800880086,
      "grad_norm": 0.00750732421875,
      "learning_rate": 0.029999982743912613,
      "loss": 0.233,
      "num_input_tokens_seen": 3852672,
      "step": 18260
    },
    {
      "epoch": 2.0093509350935093,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.02999998049047435,
      "loss": 0.2324,
      "num_input_tokens_seen": 3853824,
      "step": 18265
    },
    {
      "epoch": 2.00990099009901,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.02999997809878847,
      "loss": 0.2299,
      "num_input_tokens_seen": 3854848,
      "step": 18270
    },
    {
      "epoch": 2.0104510451045106,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.029999975568854993,
      "loss": 0.2345,
      "num_input_tokens_seen": 3855904,
      "step": 18275
    },
    {
      "epoch": 2.011001100110011,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.029999972900673928,
      "loss": 0.2329,
      "num_input_tokens_seen": 3856928,
      "step": 18280
    },
    {
      "epoch": 2.0115511551155114,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02999997009424532,
      "loss": 0.233,
      "num_input_tokens_seen": 3857952,
      "step": 18285
    },
    {
      "epoch": 2.012101210121012,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.029999967149569182,
      "loss": 0.2278,
      "num_input_tokens_seen": 3859072,
      "step": 18290
    },
    {
      "epoch": 2.0126512651265127,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.029999964066645544,
      "loss": 0.2288,
      "num_input_tokens_seen": 3860160,
      "step": 18295
    },
    {
      "epoch": 2.0132013201320134,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.029999960845474436,
      "loss": 0.2284,
      "num_input_tokens_seen": 3861216,
      "step": 18300
    },
    {
      "epoch": 2.0137513751375136,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.029999957486055887,
      "loss": 0.233,
      "num_input_tokens_seen": 3862272,
      "step": 18305
    },
    {
      "epoch": 2.014301430143014,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.029999953988389925,
      "loss": 0.2304,
      "num_input_tokens_seen": 3863360,
      "step": 18310
    },
    {
      "epoch": 2.014851485148515,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.029999950352476587,
      "loss": 0.2248,
      "num_input_tokens_seen": 3864416,
      "step": 18315
    },
    {
      "epoch": 2.0154015401540155,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.0299999465783159,
      "loss": 0.2326,
      "num_input_tokens_seen": 3865440,
      "step": 18320
    },
    {
      "epoch": 2.015951595159516,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.02999994266590791,
      "loss": 0.2328,
      "num_input_tokens_seen": 3866496,
      "step": 18325
    },
    {
      "epoch": 2.0165016501650164,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.02999993861525264,
      "loss": 0.2327,
      "num_input_tokens_seen": 3867520,
      "step": 18330
    },
    {
      "epoch": 2.017051705170517,
      "grad_norm": 0.0145263671875,
      "learning_rate": 0.029999934426350138,
      "loss": 0.229,
      "num_input_tokens_seen": 3868576,
      "step": 18335
    },
    {
      "epoch": 2.0176017601760177,
      "grad_norm": 0.0078125,
      "learning_rate": 0.029999930099200437,
      "loss": 0.2348,
      "num_input_tokens_seen": 3869632,
      "step": 18340
    },
    {
      "epoch": 2.0181518151815183,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.02999992563380358,
      "loss": 0.2352,
      "num_input_tokens_seen": 3870592,
      "step": 18345
    },
    {
      "epoch": 2.0187018701870185,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.0299999210301596,
      "loss": 0.232,
      "num_input_tokens_seen": 3871680,
      "step": 18350
    },
    {
      "epoch": 2.019251925192519,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.02999991628826855,
      "loss": 0.2341,
      "num_input_tokens_seen": 3872704,
      "step": 18355
    },
    {
      "epoch": 2.01980198019802,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.029999911408130464,
      "loss": 0.232,
      "num_input_tokens_seen": 3873728,
      "step": 18360
    },
    {
      "epoch": 2.0203520352035205,
      "grad_norm": 0.01324462890625,
      "learning_rate": 0.0299999063897454,
      "loss": 0.234,
      "num_input_tokens_seen": 3874752,
      "step": 18365
    },
    {
      "epoch": 2.020902090209021,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.02999990123311339,
      "loss": 0.2319,
      "num_input_tokens_seen": 3875808,
      "step": 18370
    },
    {
      "epoch": 2.0214521452145213,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.029999895938234496,
      "loss": 0.2319,
      "num_input_tokens_seen": 3876864,
      "step": 18375
    },
    {
      "epoch": 2.022002200220022,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.02999989050510875,
      "loss": 0.2309,
      "num_input_tokens_seen": 3877920,
      "step": 18380
    },
    {
      "epoch": 2.0225522552255226,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.029999884933736214,
      "loss": 0.2294,
      "num_input_tokens_seen": 3878912,
      "step": 18385
    },
    {
      "epoch": 2.0231023102310233,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.029999879224116937,
      "loss": 0.2362,
      "num_input_tokens_seen": 3879968,
      "step": 18390
    },
    {
      "epoch": 2.0236523652365235,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.02999987337625097,
      "loss": 0.2294,
      "num_input_tokens_seen": 3881024,
      "step": 18395
    },
    {
      "epoch": 2.024202420242024,
      "grad_norm": 0.0137939453125,
      "learning_rate": 0.029999867390138366,
      "loss": 0.234,
      "num_input_tokens_seen": 3882144,
      "step": 18400
    },
    {
      "epoch": 2.0247524752475248,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.029999861265779186,
      "loss": 0.2319,
      "num_input_tokens_seen": 3883232,
      "step": 18405
    },
    {
      "epoch": 2.0253025302530254,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.02999985500317348,
      "loss": 0.2319,
      "num_input_tokens_seen": 3884256,
      "step": 18410
    },
    {
      "epoch": 2.0258525852585256,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.02999984860232131,
      "loss": 0.2324,
      "num_input_tokens_seen": 3885312,
      "step": 18415
    },
    {
      "epoch": 2.0264026402640263,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.029999842063222728,
      "loss": 0.2335,
      "num_input_tokens_seen": 3886368,
      "step": 18420
    },
    {
      "epoch": 2.026952695269527,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029999835385877804,
      "loss": 0.2314,
      "num_input_tokens_seen": 3887488,
      "step": 18425
    },
    {
      "epoch": 2.0275027502750276,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.029999828570286595,
      "loss": 0.2319,
      "num_input_tokens_seen": 3888544,
      "step": 18430
    },
    {
      "epoch": 2.0280528052805282,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.029999821616449162,
      "loss": 0.2314,
      "num_input_tokens_seen": 3889632,
      "step": 18435
    },
    {
      "epoch": 2.0286028602860284,
      "grad_norm": 0.01324462890625,
      "learning_rate": 0.02999981452436557,
      "loss": 0.2314,
      "num_input_tokens_seen": 3890624,
      "step": 18440
    },
    {
      "epoch": 2.029152915291529,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.029999807294035886,
      "loss": 0.2314,
      "num_input_tokens_seen": 3891744,
      "step": 18445
    },
    {
      "epoch": 2.0297029702970297,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.029999799925460178,
      "loss": 0.233,
      "num_input_tokens_seen": 3892864,
      "step": 18450
    },
    {
      "epoch": 2.0302530253025304,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.02999979241863851,
      "loss": 0.2303,
      "num_input_tokens_seen": 3893920,
      "step": 18455
    },
    {
      "epoch": 2.0308030803080306,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.029999784773570952,
      "loss": 0.2319,
      "num_input_tokens_seen": 3895008,
      "step": 18460
    },
    {
      "epoch": 2.0313531353135312,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.029999776990257576,
      "loss": 0.2319,
      "num_input_tokens_seen": 3896064,
      "step": 18465
    },
    {
      "epoch": 2.031903190319032,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.029999769068698452,
      "loss": 0.2324,
      "num_input_tokens_seen": 3897120,
      "step": 18470
    },
    {
      "epoch": 2.0324532453245325,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.029999761008893656,
      "loss": 0.2314,
      "num_input_tokens_seen": 3898112,
      "step": 18475
    },
    {
      "epoch": 2.033003300330033,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02999975281084326,
      "loss": 0.2313,
      "num_input_tokens_seen": 3899136,
      "step": 18480
    },
    {
      "epoch": 2.0335533553355334,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.02999974447454734,
      "loss": 0.2319,
      "num_input_tokens_seen": 3900128,
      "step": 18485
    },
    {
      "epoch": 2.034103410341034,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.029999736000005973,
      "loss": 0.2319,
      "num_input_tokens_seen": 3901152,
      "step": 18490
    },
    {
      "epoch": 2.0346534653465347,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.029999727387219236,
      "loss": 0.2308,
      "num_input_tokens_seen": 3902240,
      "step": 18495
    },
    {
      "epoch": 2.0352035203520353,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.02999971863618721,
      "loss": 0.2319,
      "num_input_tokens_seen": 3903328,
      "step": 18500
    },
    {
      "epoch": 2.0357535753575355,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.029999709746909974,
      "loss": 0.2313,
      "num_input_tokens_seen": 3904352,
      "step": 18505
    },
    {
      "epoch": 2.036303630363036,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.02999970071938761,
      "loss": 0.2299,
      "num_input_tokens_seen": 3905408,
      "step": 18510
    },
    {
      "epoch": 2.036853685368537,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0299996915536202,
      "loss": 0.2299,
      "num_input_tokens_seen": 3906464,
      "step": 18515
    },
    {
      "epoch": 2.0374037403740375,
      "grad_norm": 0.007659912109375,
      "learning_rate": 0.029999682249607834,
      "loss": 0.2295,
      "num_input_tokens_seen": 3907584,
      "step": 18520
    },
    {
      "epoch": 2.037953795379538,
      "grad_norm": 0.007659912109375,
      "learning_rate": 0.029999672807350594,
      "loss": 0.2333,
      "num_input_tokens_seen": 3908672,
      "step": 18525
    },
    {
      "epoch": 2.0385038503850383,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.029999663226848567,
      "loss": 0.224,
      "num_input_tokens_seen": 3909760,
      "step": 18530
    },
    {
      "epoch": 2.039053905390539,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.029999653508101843,
      "loss": 0.238,
      "num_input_tokens_seen": 3910816,
      "step": 18535
    },
    {
      "epoch": 2.0396039603960396,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.02999964365111051,
      "loss": 0.2297,
      "num_input_tokens_seen": 3911840,
      "step": 18540
    },
    {
      "epoch": 2.0401540154015403,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.029999633655874658,
      "loss": 0.223,
      "num_input_tokens_seen": 3912864,
      "step": 18545
    },
    {
      "epoch": 2.0407040704070405,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029999623522394377,
      "loss": 0.236,
      "num_input_tokens_seen": 3913920,
      "step": 18550
    },
    {
      "epoch": 2.041254125412541,
      "grad_norm": 0.00775146484375,
      "learning_rate": 0.029999613250669768,
      "loss": 0.2328,
      "num_input_tokens_seen": 3915008,
      "step": 18555
    },
    {
      "epoch": 2.041804180418042,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.02999960284070092,
      "loss": 0.2261,
      "num_input_tokens_seen": 3916096,
      "step": 18560
    },
    {
      "epoch": 2.0423542354235424,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.029999592292487928,
      "loss": 0.2215,
      "num_input_tokens_seen": 3917056,
      "step": 18565
    },
    {
      "epoch": 2.042904290429043,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.029999581606030896,
      "loss": 0.2382,
      "num_input_tokens_seen": 3918144,
      "step": 18570
    },
    {
      "epoch": 2.0434543454345433,
      "grad_norm": 0.007568359375,
      "learning_rate": 0.029999570781329914,
      "loss": 0.2439,
      "num_input_tokens_seen": 3919232,
      "step": 18575
    },
    {
      "epoch": 2.044004400440044,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.029999559818385087,
      "loss": 0.2375,
      "num_input_tokens_seen": 3920256,
      "step": 18580
    },
    {
      "epoch": 2.0445544554455446,
      "grad_norm": 0.0027923583984375,
      "learning_rate": 0.02999954871719651,
      "loss": 0.2305,
      "num_input_tokens_seen": 3921376,
      "step": 18585
    },
    {
      "epoch": 2.0451045104510452,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.029999537477764297,
      "loss": 0.2309,
      "num_input_tokens_seen": 3922432,
      "step": 18590
    },
    {
      "epoch": 2.0456545654565454,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.02999952610008854,
      "loss": 0.2346,
      "num_input_tokens_seen": 3923488,
      "step": 18595
    },
    {
      "epoch": 2.046204620462046,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.02999951458416935,
      "loss": 0.2293,
      "num_input_tokens_seen": 3924512,
      "step": 18600
    },
    {
      "epoch": 2.0467546754675467,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.029999502930006833,
      "loss": 0.2319,
      "num_input_tokens_seen": 3925600,
      "step": 18605
    },
    {
      "epoch": 2.0473047304730474,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.029999491137601093,
      "loss": 0.2303,
      "num_input_tokens_seen": 3926656,
      "step": 18610
    },
    {
      "epoch": 2.047854785478548,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029999479206952243,
      "loss": 0.2298,
      "num_input_tokens_seen": 3927744,
      "step": 18615
    },
    {
      "epoch": 2.0484048404840483,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.029999467138060384,
      "loss": 0.2314,
      "num_input_tokens_seen": 3928736,
      "step": 18620
    },
    {
      "epoch": 2.048954895489549,
      "grad_norm": 0.01239013671875,
      "learning_rate": 0.029999454930925644,
      "loss": 0.2315,
      "num_input_tokens_seen": 3929760,
      "step": 18625
    },
    {
      "epoch": 2.0495049504950495,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029999442585548115,
      "loss": 0.2315,
      "num_input_tokens_seen": 3930816,
      "step": 18630
    },
    {
      "epoch": 2.05005500550055,
      "grad_norm": 0.001953125,
      "learning_rate": 0.02999943010192793,
      "loss": 0.2248,
      "num_input_tokens_seen": 3931904,
      "step": 18635
    },
    {
      "epoch": 2.0506050605060504,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.02999941748006519,
      "loss": 0.228,
      "num_input_tokens_seen": 3932896,
      "step": 18640
    },
    {
      "epoch": 2.051155115511551,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.02999940471996002,
      "loss": 0.2302,
      "num_input_tokens_seen": 3933984,
      "step": 18645
    },
    {
      "epoch": 2.0517051705170517,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.029999391821612528,
      "loss": 0.229,
      "num_input_tokens_seen": 3935072,
      "step": 18650
    },
    {
      "epoch": 2.0522552255225524,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029999378785022845,
      "loss": 0.231,
      "num_input_tokens_seen": 3936160,
      "step": 18655
    },
    {
      "epoch": 2.052805280528053,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.029999365610191082,
      "loss": 0.2274,
      "num_input_tokens_seen": 3937216,
      "step": 18660
    },
    {
      "epoch": 2.053355335533553,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.029999352297117365,
      "loss": 0.2198,
      "num_input_tokens_seen": 3938208,
      "step": 18665
    },
    {
      "epoch": 2.053905390539054,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.029999338845801818,
      "loss": 0.222,
      "num_input_tokens_seen": 3939232,
      "step": 18670
    },
    {
      "epoch": 2.0544554455445545,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.029999325256244557,
      "loss": 0.242,
      "num_input_tokens_seen": 3940288,
      "step": 18675
    },
    {
      "epoch": 2.055005500550055,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.02999931152844571,
      "loss": 0.2362,
      "num_input_tokens_seen": 3941376,
      "step": 18680
    },
    {
      "epoch": 2.0555555555555554,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.029999297662405416,
      "loss": 0.2314,
      "num_input_tokens_seen": 3942432,
      "step": 18685
    },
    {
      "epoch": 2.056105610561056,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.029999283658123786,
      "loss": 0.2226,
      "num_input_tokens_seen": 3943424,
      "step": 18690
    },
    {
      "epoch": 2.0566556655665567,
      "grad_norm": 0.002716064453125,
      "learning_rate": 0.029999269515600956,
      "loss": 0.2339,
      "num_input_tokens_seen": 3944448,
      "step": 18695
    },
    {
      "epoch": 2.0572057205720573,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.02999925523483706,
      "loss": 0.233,
      "num_input_tokens_seen": 3945504,
      "step": 18700
    },
    {
      "epoch": 2.057755775577558,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.02999924081583222,
      "loss": 0.2332,
      "num_input_tokens_seen": 3946656,
      "step": 18705
    },
    {
      "epoch": 2.058305830583058,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.02999922625858658,
      "loss": 0.2337,
      "num_input_tokens_seen": 3947616,
      "step": 18710
    },
    {
      "epoch": 2.058855885588559,
      "grad_norm": 0.0030517578125,
      "learning_rate": 0.02999921156310027,
      "loss": 0.2385,
      "num_input_tokens_seen": 3948672,
      "step": 18715
    },
    {
      "epoch": 2.0594059405940595,
      "grad_norm": 0.0078125,
      "learning_rate": 0.029999196729373418,
      "loss": 0.2345,
      "num_input_tokens_seen": 3949760,
      "step": 18720
    },
    {
      "epoch": 2.05995599559956,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.029999181757406172,
      "loss": 0.2259,
      "num_input_tokens_seen": 3950752,
      "step": 18725
    },
    {
      "epoch": 2.0605060506050603,
      "grad_norm": 0.0028076171875,
      "learning_rate": 0.029999166647198663,
      "loss": 0.2296,
      "num_input_tokens_seen": 3951776,
      "step": 18730
    },
    {
      "epoch": 2.061056105610561,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02999915139875103,
      "loss": 0.2299,
      "num_input_tokens_seen": 3952832,
      "step": 18735
    },
    {
      "epoch": 2.0616061606160616,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029999136012063422,
      "loss": 0.23,
      "num_input_tokens_seen": 3953920,
      "step": 18740
    },
    {
      "epoch": 2.0621562156215623,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.02999912048713597,
      "loss": 0.2382,
      "num_input_tokens_seen": 3955008,
      "step": 18745
    },
    {
      "epoch": 2.062706270627063,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029999104823968822,
      "loss": 0.2386,
      "num_input_tokens_seen": 3956096,
      "step": 18750
    },
    {
      "epoch": 2.063256325632563,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.029999089022562125,
      "loss": 0.2276,
      "num_input_tokens_seen": 3957152,
      "step": 18755
    },
    {
      "epoch": 2.0638063806380638,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.02999907308291602,
      "loss": 0.2394,
      "num_input_tokens_seen": 3958208,
      "step": 18760
    },
    {
      "epoch": 2.0643564356435644,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.029999057005030656,
      "loss": 0.2284,
      "num_input_tokens_seen": 3959264,
      "step": 18765
    },
    {
      "epoch": 2.064906490649065,
      "grad_norm": 0.007080078125,
      "learning_rate": 0.029999040788906178,
      "loss": 0.2304,
      "num_input_tokens_seen": 3960352,
      "step": 18770
    },
    {
      "epoch": 2.0654565456545653,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.02999902443454274,
      "loss": 0.2304,
      "num_input_tokens_seen": 3961440,
      "step": 18775
    },
    {
      "epoch": 2.066006600660066,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029999007941940492,
      "loss": 0.2341,
      "num_input_tokens_seen": 3962464,
      "step": 18780
    },
    {
      "epoch": 2.0665566556655666,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029998991311099586,
      "loss": 0.2299,
      "num_input_tokens_seen": 3963552,
      "step": 18785
    },
    {
      "epoch": 2.067106710671067,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.029998974542020176,
      "loss": 0.2335,
      "num_input_tokens_seen": 3964704,
      "step": 18790
    },
    {
      "epoch": 2.067656765676568,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.02999895763470241,
      "loss": 0.2283,
      "num_input_tokens_seen": 3965696,
      "step": 18795
    },
    {
      "epoch": 2.068206820682068,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.02999894058914645,
      "loss": 0.2289,
      "num_input_tokens_seen": 3966688,
      "step": 18800
    },
    {
      "epoch": 2.0687568756875687,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.02999892340535245,
      "loss": 0.228,
      "num_input_tokens_seen": 3967712,
      "step": 18805
    },
    {
      "epoch": 2.0693069306930694,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02999890608332057,
      "loss": 0.2266,
      "num_input_tokens_seen": 3968704,
      "step": 18810
    },
    {
      "epoch": 2.06985698569857,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.029998888623050977,
      "loss": 0.234,
      "num_input_tokens_seen": 3969760,
      "step": 18815
    },
    {
      "epoch": 2.0704070407040702,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.029998871024543818,
      "loss": 0.233,
      "num_input_tokens_seen": 3970816,
      "step": 18820
    },
    {
      "epoch": 2.070957095709571,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.029998853287799263,
      "loss": 0.234,
      "num_input_tokens_seen": 3971872,
      "step": 18825
    },
    {
      "epoch": 2.0715071507150715,
      "grad_norm": 0.00750732421875,
      "learning_rate": 0.029998835412817476,
      "loss": 0.2344,
      "num_input_tokens_seen": 3972864,
      "step": 18830
    },
    {
      "epoch": 2.072057205720572,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.02999881739959862,
      "loss": 0.2297,
      "num_input_tokens_seen": 3973888,
      "step": 18835
    },
    {
      "epoch": 2.072607260726073,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.02999879924814286,
      "loss": 0.2349,
      "num_input_tokens_seen": 3974880,
      "step": 18840
    },
    {
      "epoch": 2.073157315731573,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029998780958450367,
      "loss": 0.2303,
      "num_input_tokens_seen": 3975904,
      "step": 18845
    },
    {
      "epoch": 2.0737073707370737,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.029998762530521306,
      "loss": 0.2333,
      "num_input_tokens_seen": 3976896,
      "step": 18850
    },
    {
      "epoch": 2.0742574257425743,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.029998743964355847,
      "loss": 0.239,
      "num_input_tokens_seen": 3977888,
      "step": 18855
    },
    {
      "epoch": 2.074807480748075,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02999872525995416,
      "loss": 0.232,
      "num_input_tokens_seen": 3978944,
      "step": 18860
    },
    {
      "epoch": 2.075357535753575,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.02999870641731642,
      "loss": 0.2319,
      "num_input_tokens_seen": 3980064,
      "step": 18865
    },
    {
      "epoch": 2.075907590759076,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0299986874364428,
      "loss": 0.2319,
      "num_input_tokens_seen": 3981056,
      "step": 18870
    },
    {
      "epoch": 2.0764576457645765,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.02999866831733348,
      "loss": 0.2314,
      "num_input_tokens_seen": 3982080,
      "step": 18875
    },
    {
      "epoch": 2.077007700770077,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.029998649059988627,
      "loss": 0.2293,
      "num_input_tokens_seen": 3983168,
      "step": 18880
    },
    {
      "epoch": 2.0775577557755778,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.029998629664408423,
      "loss": 0.2309,
      "num_input_tokens_seen": 3984160,
      "step": 18885
    },
    {
      "epoch": 2.078107810781078,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.02999861013059305,
      "loss": 0.233,
      "num_input_tokens_seen": 3985248,
      "step": 18890
    },
    {
      "epoch": 2.0786578657865786,
      "grad_norm": 0.00286865234375,
      "learning_rate": 0.029998590458542678,
      "loss": 0.2325,
      "num_input_tokens_seen": 3986304,
      "step": 18895
    },
    {
      "epoch": 2.0792079207920793,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.029998570648257495,
      "loss": 0.2324,
      "num_input_tokens_seen": 3987360,
      "step": 18900
    },
    {
      "epoch": 2.07975797579758,
      "grad_norm": 0.01239013671875,
      "learning_rate": 0.029998550699737687,
      "loss": 0.2308,
      "num_input_tokens_seen": 3988352,
      "step": 18905
    },
    {
      "epoch": 2.08030803080308,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.029998530612983434,
      "loss": 0.233,
      "num_input_tokens_seen": 3989408,
      "step": 18910
    },
    {
      "epoch": 2.080858085808581,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.02999851038799492,
      "loss": 0.2324,
      "num_input_tokens_seen": 3990464,
      "step": 18915
    },
    {
      "epoch": 2.0814081408140814,
      "grad_norm": 0.0123291015625,
      "learning_rate": 0.029998490024772336,
      "loss": 0.2303,
      "num_input_tokens_seen": 3991520,
      "step": 18920
    },
    {
      "epoch": 2.081958195819582,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029998469523315863,
      "loss": 0.2314,
      "num_input_tokens_seen": 3992576,
      "step": 18925
    },
    {
      "epoch": 2.0825082508250823,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029998448883625697,
      "loss": 0.2308,
      "num_input_tokens_seen": 3993600,
      "step": 18930
    },
    {
      "epoch": 2.083058305830583,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029998428105702017,
      "loss": 0.2298,
      "num_input_tokens_seen": 3994592,
      "step": 18935
    },
    {
      "epoch": 2.0836083608360836,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029998407189545027,
      "loss": 0.234,
      "num_input_tokens_seen": 3995648,
      "step": 18940
    },
    {
      "epoch": 2.0841584158415842,
      "grad_norm": 0.01263427734375,
      "learning_rate": 0.029998386135154917,
      "loss": 0.2319,
      "num_input_tokens_seen": 3996672,
      "step": 18945
    },
    {
      "epoch": 2.084708470847085,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.02999836494253187,
      "loss": 0.2314,
      "num_input_tokens_seen": 3997696,
      "step": 18950
    },
    {
      "epoch": 2.085258525852585,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.029998343611676106,
      "loss": 0.2298,
      "num_input_tokens_seen": 3998784,
      "step": 18955
    },
    {
      "epoch": 2.0858085808580857,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.02999832214258779,
      "loss": 0.2293,
      "num_input_tokens_seen": 3999840,
      "step": 18960
    },
    {
      "epoch": 2.0863586358635864,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029998300535267145,
      "loss": 0.2324,
      "num_input_tokens_seen": 4000864,
      "step": 18965
    },
    {
      "epoch": 2.086908690869087,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.02999827878971436,
      "loss": 0.2298,
      "num_input_tokens_seen": 4001952,
      "step": 18970
    },
    {
      "epoch": 2.0874587458745877,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.029998256905929634,
      "loss": 0.2314,
      "num_input_tokens_seen": 4003040,
      "step": 18975
    },
    {
      "epoch": 2.088008800880088,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.02999823488391317,
      "loss": 0.2325,
      "num_input_tokens_seen": 4004160,
      "step": 18980
    },
    {
      "epoch": 2.0885588558855885,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.029998212723665176,
      "loss": 0.2331,
      "num_input_tokens_seen": 4005248,
      "step": 18985
    },
    {
      "epoch": 2.089108910891089,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.02999819042518585,
      "loss": 0.2314,
      "num_input_tokens_seen": 4006272,
      "step": 18990
    },
    {
      "epoch": 2.08965896589659,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029998167988475394,
      "loss": 0.2294,
      "num_input_tokens_seen": 4007296,
      "step": 18995
    },
    {
      "epoch": 2.09020902090209,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.02999814541353403,
      "loss": 0.2335,
      "num_input_tokens_seen": 4008320,
      "step": 19000
    },
    {
      "epoch": 2.0907590759075907,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.02999812270036195,
      "loss": 0.2341,
      "num_input_tokens_seen": 4009376,
      "step": 19005
    },
    {
      "epoch": 2.0913091309130913,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029998099848959372,
      "loss": 0.2324,
      "num_input_tokens_seen": 4010496,
      "step": 19010
    },
    {
      "epoch": 2.091859185918592,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.029998076859326504,
      "loss": 0.2314,
      "num_input_tokens_seen": 4011584,
      "step": 19015
    },
    {
      "epoch": 2.092409240924092,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02999805373146356,
      "loss": 0.2309,
      "num_input_tokens_seen": 4012672,
      "step": 19020
    },
    {
      "epoch": 2.092959295929593,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.02999803046537075,
      "loss": 0.2319,
      "num_input_tokens_seen": 4013696,
      "step": 19025
    },
    {
      "epoch": 2.0935093509350935,
      "grad_norm": 0.000675201416015625,
      "learning_rate": 0.02999800706104829,
      "loss": 0.2303,
      "num_input_tokens_seen": 4014752,
      "step": 19030
    },
    {
      "epoch": 2.094059405940594,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.029997983518496397,
      "loss": 0.2319,
      "num_input_tokens_seen": 4015872,
      "step": 19035
    },
    {
      "epoch": 2.094609460946095,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.029997959837715287,
      "loss": 0.233,
      "num_input_tokens_seen": 4016928,
      "step": 19040
    },
    {
      "epoch": 2.095159515951595,
      "grad_norm": 0.0029296875,
      "learning_rate": 0.029997936018705176,
      "loss": 0.2309,
      "num_input_tokens_seen": 4017984,
      "step": 19045
    },
    {
      "epoch": 2.0957095709570956,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.02999791206146629,
      "loss": 0.2303,
      "num_input_tokens_seen": 4019040,
      "step": 19050
    },
    {
      "epoch": 2.0962596259625963,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.029997887965998837,
      "loss": 0.2304,
      "num_input_tokens_seen": 4020064,
      "step": 19055
    },
    {
      "epoch": 2.096809680968097,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.02999786373230305,
      "loss": 0.2325,
      "num_input_tokens_seen": 4021088,
      "step": 19060
    },
    {
      "epoch": 2.097359735973597,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029997839360379155,
      "loss": 0.2304,
      "num_input_tokens_seen": 4022144,
      "step": 19065
    },
    {
      "epoch": 2.097909790979098,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.029997814850227365,
      "loss": 0.2293,
      "num_input_tokens_seen": 4023232,
      "step": 19070
    },
    {
      "epoch": 2.0984598459845984,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.029997790201847913,
      "loss": 0.2309,
      "num_input_tokens_seen": 4024320,
      "step": 19075
    },
    {
      "epoch": 2.099009900990099,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02999776541524103,
      "loss": 0.2314,
      "num_input_tokens_seen": 4025344,
      "step": 19080
    },
    {
      "epoch": 2.0995599559955997,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.02999774049040694,
      "loss": 0.2303,
      "num_input_tokens_seen": 4026336,
      "step": 19085
    },
    {
      "epoch": 2.1001100110011,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.029997715427345868,
      "loss": 0.2303,
      "num_input_tokens_seen": 4027424,
      "step": 19090
    },
    {
      "epoch": 2.1006600660066006,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.029997690226058048,
      "loss": 0.231,
      "num_input_tokens_seen": 4028480,
      "step": 19095
    },
    {
      "epoch": 2.1012101210121013,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.02999766488654372,
      "loss": 0.2336,
      "num_input_tokens_seen": 4029568,
      "step": 19100
    },
    {
      "epoch": 2.101760176017602,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.02999763940880311,
      "loss": 0.2279,
      "num_input_tokens_seen": 4030624,
      "step": 19105
    },
    {
      "epoch": 2.102310231023102,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.02999761379283645,
      "loss": 0.229,
      "num_input_tokens_seen": 4031648,
      "step": 19110
    },
    {
      "epoch": 2.1028602860286028,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029997588038643982,
      "loss": 0.2323,
      "num_input_tokens_seen": 4032640,
      "step": 19115
    },
    {
      "epoch": 2.1034103410341034,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.02999756214622594,
      "loss": 0.2338,
      "num_input_tokens_seen": 4033728,
      "step": 19120
    },
    {
      "epoch": 2.103960396039604,
      "grad_norm": 0.0128173828125,
      "learning_rate": 0.029997536115582566,
      "loss": 0.2301,
      "num_input_tokens_seen": 4034752,
      "step": 19125
    },
    {
      "epoch": 2.1045104510451047,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.0299975099467141,
      "loss": 0.2337,
      "num_input_tokens_seen": 4035776,
      "step": 19130
    },
    {
      "epoch": 2.105060506050605,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02999748363962078,
      "loss": 0.2341,
      "num_input_tokens_seen": 4036832,
      "step": 19135
    },
    {
      "epoch": 2.1056105610561056,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.02999745719430285,
      "loss": 0.232,
      "num_input_tokens_seen": 4037824,
      "step": 19140
    },
    {
      "epoch": 2.106160616061606,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.029997430610760555,
      "loss": 0.2335,
      "num_input_tokens_seen": 4038848,
      "step": 19145
    },
    {
      "epoch": 2.106710671067107,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.029997403888994134,
      "loss": 0.2288,
      "num_input_tokens_seen": 4039872,
      "step": 19150
    },
    {
      "epoch": 2.107260726072607,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.02999737702900384,
      "loss": 0.2319,
      "num_input_tokens_seen": 4040928,
      "step": 19155
    },
    {
      "epoch": 2.1078107810781077,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029997350030789918,
      "loss": 0.231,
      "num_input_tokens_seen": 4041984,
      "step": 19160
    },
    {
      "epoch": 2.1083608360836084,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.02999732289435262,
      "loss": 0.2319,
      "num_input_tokens_seen": 4043104,
      "step": 19165
    },
    {
      "epoch": 2.108910891089109,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02999729561969219,
      "loss": 0.232,
      "num_input_tokens_seen": 4044224,
      "step": 19170
    },
    {
      "epoch": 2.1094609460946097,
      "grad_norm": 0.00360107421875,
      "learning_rate": 0.029997268206808883,
      "loss": 0.2325,
      "num_input_tokens_seen": 4045312,
      "step": 19175
    },
    {
      "epoch": 2.11001100110011,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02999724065570295,
      "loss": 0.2335,
      "num_input_tokens_seen": 4046304,
      "step": 19180
    },
    {
      "epoch": 2.1105610561056105,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.029997212966374653,
      "loss": 0.2319,
      "num_input_tokens_seen": 4047360,
      "step": 19185
    },
    {
      "epoch": 2.111111111111111,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029997185138824232,
      "loss": 0.2324,
      "num_input_tokens_seen": 4048416,
      "step": 19190
    },
    {
      "epoch": 2.111661166116612,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.029997157173051958,
      "loss": 0.2314,
      "num_input_tokens_seen": 4049440,
      "step": 19195
    },
    {
      "epoch": 2.112211221122112,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.02999712906905808,
      "loss": 0.2314,
      "num_input_tokens_seen": 4050464,
      "step": 19200
    },
    {
      "epoch": 2.1127612761276127,
      "grad_norm": 0.01239013671875,
      "learning_rate": 0.02999710082684286,
      "loss": 0.2324,
      "num_input_tokens_seen": 4051424,
      "step": 19205
    },
    {
      "epoch": 2.1133113311331133,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02999707244640656,
      "loss": 0.2319,
      "num_input_tokens_seen": 4052480,
      "step": 19210
    },
    {
      "epoch": 2.113861386138614,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.029997043927749437,
      "loss": 0.2314,
      "num_input_tokens_seen": 4053504,
      "step": 19215
    },
    {
      "epoch": 2.1144114411441146,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.029997015270871757,
      "loss": 0.2303,
      "num_input_tokens_seen": 4054496,
      "step": 19220
    },
    {
      "epoch": 2.114961496149615,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029996986475773785,
      "loss": 0.2288,
      "num_input_tokens_seen": 4055520,
      "step": 19225
    },
    {
      "epoch": 2.1155115511551155,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029996957542455783,
      "loss": 0.2319,
      "num_input_tokens_seen": 4056544,
      "step": 19230
    },
    {
      "epoch": 2.116061606160616,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.029996928470918022,
      "loss": 0.234,
      "num_input_tokens_seen": 4057600,
      "step": 19235
    },
    {
      "epoch": 2.1166116611661168,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.029996899261160764,
      "loss": 0.2303,
      "num_input_tokens_seen": 4058656,
      "step": 19240
    },
    {
      "epoch": 2.117161716171617,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.029996869913184284,
      "loss": 0.2299,
      "num_input_tokens_seen": 4059712,
      "step": 19245
    },
    {
      "epoch": 2.1177117711771176,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.029996840426988847,
      "loss": 0.2309,
      "num_input_tokens_seen": 4060704,
      "step": 19250
    },
    {
      "epoch": 2.1182618261826183,
      "grad_norm": 0.007415771484375,
      "learning_rate": 0.02999681080257473,
      "loss": 0.233,
      "num_input_tokens_seen": 4061792,
      "step": 19255
    },
    {
      "epoch": 2.118811881188119,
      "grad_norm": 0.00311279296875,
      "learning_rate": 0.029996781039942206,
      "loss": 0.2324,
      "num_input_tokens_seen": 4062944,
      "step": 19260
    },
    {
      "epoch": 2.1193619361936196,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029996751139091546,
      "loss": 0.2324,
      "num_input_tokens_seen": 4063968,
      "step": 19265
    },
    {
      "epoch": 2.1199119911991198,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.029996721100023025,
      "loss": 0.2314,
      "num_input_tokens_seen": 4065056,
      "step": 19270
    },
    {
      "epoch": 2.1204620462046204,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.02999669092273692,
      "loss": 0.2308,
      "num_input_tokens_seen": 4066144,
      "step": 19275
    },
    {
      "epoch": 2.121012101210121,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.029996660607233522,
      "loss": 0.2309,
      "num_input_tokens_seen": 4067264,
      "step": 19280
    },
    {
      "epoch": 2.1215621562156217,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.02999663015351309,
      "loss": 0.2335,
      "num_input_tokens_seen": 4068320,
      "step": 19285
    },
    {
      "epoch": 2.122112211221122,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029996599561575913,
      "loss": 0.2324,
      "num_input_tokens_seen": 4069408,
      "step": 19290
    },
    {
      "epoch": 2.1226622662266226,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.02999656883142228,
      "loss": 0.2309,
      "num_input_tokens_seen": 4070464,
      "step": 19295
    },
    {
      "epoch": 2.1232123212321232,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029996537963052464,
      "loss": 0.2319,
      "num_input_tokens_seen": 4071488,
      "step": 19300
    },
    {
      "epoch": 2.123762376237624,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029996506956466752,
      "loss": 0.2324,
      "num_input_tokens_seen": 4072576,
      "step": 19305
    },
    {
      "epoch": 2.1243124312431245,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029996475811665434,
      "loss": 0.2308,
      "num_input_tokens_seen": 4073632,
      "step": 19310
    },
    {
      "epoch": 2.1248624862486247,
      "grad_norm": 0.0030517578125,
      "learning_rate": 0.029996444528648795,
      "loss": 0.2324,
      "num_input_tokens_seen": 4074720,
      "step": 19315
    },
    {
      "epoch": 2.1254125412541254,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02999641310741712,
      "loss": 0.2293,
      "num_input_tokens_seen": 4075776,
      "step": 19320
    },
    {
      "epoch": 2.125962596259626,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.029996381547970703,
      "loss": 0.2329,
      "num_input_tokens_seen": 4076864,
      "step": 19325
    },
    {
      "epoch": 2.1265126512651267,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.029996349850309835,
      "loss": 0.2304,
      "num_input_tokens_seen": 4077952,
      "step": 19330
    },
    {
      "epoch": 2.127062706270627,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029996318014434805,
      "loss": 0.234,
      "num_input_tokens_seen": 4079008,
      "step": 19335
    },
    {
      "epoch": 2.1276127612761275,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.029996286040345908,
      "loss": 0.232,
      "num_input_tokens_seen": 4080096,
      "step": 19340
    },
    {
      "epoch": 2.128162816281628,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.029996253928043438,
      "loss": 0.2315,
      "num_input_tokens_seen": 4081184,
      "step": 19345
    },
    {
      "epoch": 2.128712871287129,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.029996221677527694,
      "loss": 0.2335,
      "num_input_tokens_seen": 4082240,
      "step": 19350
    },
    {
      "epoch": 2.129262926292629,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.029996189288798967,
      "loss": 0.2315,
      "num_input_tokens_seen": 4083328,
      "step": 19355
    },
    {
      "epoch": 2.1298129812981297,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02999615676185756,
      "loss": 0.2284,
      "num_input_tokens_seen": 4084320,
      "step": 19360
    },
    {
      "epoch": 2.1303630363036303,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.029996124096703772,
      "loss": 0.2362,
      "num_input_tokens_seen": 4085312,
      "step": 19365
    },
    {
      "epoch": 2.130913091309131,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.029996091293337905,
      "loss": 0.2346,
      "num_input_tokens_seen": 4086464,
      "step": 19370
    },
    {
      "epoch": 2.1314631463146316,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.02999605835176026,
      "loss": 0.2299,
      "num_input_tokens_seen": 4087584,
      "step": 19375
    },
    {
      "epoch": 2.132013201320132,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.029996025271971143,
      "loss": 0.2299,
      "num_input_tokens_seen": 4088576,
      "step": 19380
    },
    {
      "epoch": 2.1325632563256325,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.02999599205397085,
      "loss": 0.2319,
      "num_input_tokens_seen": 4089632,
      "step": 19385
    },
    {
      "epoch": 2.133113311331133,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0299959586977597,
      "loss": 0.2304,
      "num_input_tokens_seen": 4090688,
      "step": 19390
    },
    {
      "epoch": 2.133663366336634,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.029995925203337995,
      "loss": 0.2304,
      "num_input_tokens_seen": 4091808,
      "step": 19395
    },
    {
      "epoch": 2.1342134213421344,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02999589157070604,
      "loss": 0.233,
      "num_input_tokens_seen": 4092832,
      "step": 19400
    },
    {
      "epoch": 2.1347634763476346,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.029995857799864148,
      "loss": 0.2341,
      "num_input_tokens_seen": 4093920,
      "step": 19405
    },
    {
      "epoch": 2.1353135313531353,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029995823890812633,
      "loss": 0.2309,
      "num_input_tokens_seen": 4094976,
      "step": 19410
    },
    {
      "epoch": 2.135863586358636,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.029995789843551807,
      "loss": 0.2335,
      "num_input_tokens_seen": 4096064,
      "step": 19415
    },
    {
      "epoch": 2.1364136413641366,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.029995755658081973,
      "loss": 0.2314,
      "num_input_tokens_seen": 4097056,
      "step": 19420
    },
    {
      "epoch": 2.136963696369637,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.029995721334403463,
      "loss": 0.233,
      "num_input_tokens_seen": 4098144,
      "step": 19425
    },
    {
      "epoch": 2.1375137513751374,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.029995686872516584,
      "loss": 0.2304,
      "num_input_tokens_seen": 4099200,
      "step": 19430
    },
    {
      "epoch": 2.138063806380638,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.029995652272421652,
      "loss": 0.2324,
      "num_input_tokens_seen": 4100288,
      "step": 19435
    },
    {
      "epoch": 2.1386138613861387,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.02999561753411899,
      "loss": 0.2309,
      "num_input_tokens_seen": 4101376,
      "step": 19440
    },
    {
      "epoch": 2.139163916391639,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029995582657608915,
      "loss": 0.2319,
      "num_input_tokens_seen": 4102432,
      "step": 19445
    },
    {
      "epoch": 2.1397139713971396,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029995547642891755,
      "loss": 0.2293,
      "num_input_tokens_seen": 4103488,
      "step": 19450
    },
    {
      "epoch": 2.1402640264026402,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029995512489967822,
      "loss": 0.2309,
      "num_input_tokens_seen": 4104544,
      "step": 19455
    },
    {
      "epoch": 2.140814081408141,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.02999547719883745,
      "loss": 0.2335,
      "num_input_tokens_seen": 4105600,
      "step": 19460
    },
    {
      "epoch": 2.1413641364136415,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.029995441769500957,
      "loss": 0.2278,
      "num_input_tokens_seen": 4106688,
      "step": 19465
    },
    {
      "epoch": 2.1419141914191417,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.029995406201958674,
      "loss": 0.2309,
      "num_input_tokens_seen": 4107776,
      "step": 19470
    },
    {
      "epoch": 2.1424642464246424,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.02999537049621093,
      "loss": 0.2315,
      "num_input_tokens_seen": 4108896,
      "step": 19475
    },
    {
      "epoch": 2.143014301430143,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.029995334652258048,
      "loss": 0.2326,
      "num_input_tokens_seen": 4109952,
      "step": 19480
    },
    {
      "epoch": 2.1435643564356437,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.029995298670100364,
      "loss": 0.233,
      "num_input_tokens_seen": 4110976,
      "step": 19485
    },
    {
      "epoch": 2.1441144114411443,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029995262549738208,
      "loss": 0.2288,
      "num_input_tokens_seen": 4112000,
      "step": 19490
    },
    {
      "epoch": 2.1446644664466445,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.02999522629117191,
      "loss": 0.2341,
      "num_input_tokens_seen": 4113024,
      "step": 19495
    },
    {
      "epoch": 2.145214521452145,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.029995189894401808,
      "loss": 0.2315,
      "num_input_tokens_seen": 4114016,
      "step": 19500
    },
    {
      "epoch": 2.145764576457646,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.029995153359428237,
      "loss": 0.2325,
      "num_input_tokens_seen": 4115040,
      "step": 19505
    },
    {
      "epoch": 2.1463146314631465,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029995116686251534,
      "loss": 0.2294,
      "num_input_tokens_seen": 4116096,
      "step": 19510
    },
    {
      "epoch": 2.1468646864686467,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.029995079874872033,
      "loss": 0.2319,
      "num_input_tokens_seen": 4117152,
      "step": 19515
    },
    {
      "epoch": 2.1474147414741473,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.029995042925290077,
      "loss": 0.2309,
      "num_input_tokens_seen": 4118208,
      "step": 19520
    },
    {
      "epoch": 2.147964796479648,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.029995005837506004,
      "loss": 0.2299,
      "num_input_tokens_seen": 4119232,
      "step": 19525
    },
    {
      "epoch": 2.1485148514851486,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029994968611520165,
      "loss": 0.2346,
      "num_input_tokens_seen": 4120288,
      "step": 19530
    },
    {
      "epoch": 2.149064906490649,
      "grad_norm": 0.01239013671875,
      "learning_rate": 0.029994931247332886,
      "loss": 0.2325,
      "num_input_tokens_seen": 4121248,
      "step": 19535
    },
    {
      "epoch": 2.1496149614961495,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.029994893744944524,
      "loss": 0.231,
      "num_input_tokens_seen": 4122240,
      "step": 19540
    },
    {
      "epoch": 2.15016501650165,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.029994856104355427,
      "loss": 0.2325,
      "num_input_tokens_seen": 4123264,
      "step": 19545
    },
    {
      "epoch": 2.150715071507151,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.029994818325565928,
      "loss": 0.2289,
      "num_input_tokens_seen": 4124320,
      "step": 19550
    },
    {
      "epoch": 2.1512651265126514,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.029994780408576387,
      "loss": 0.2304,
      "num_input_tokens_seen": 4125344,
      "step": 19555
    },
    {
      "epoch": 2.1518151815181517,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029994742353387154,
      "loss": 0.23,
      "num_input_tokens_seen": 4126432,
      "step": 19560
    },
    {
      "epoch": 2.1523652365236523,
      "grad_norm": 0.007171630859375,
      "learning_rate": 0.029994704159998574,
      "loss": 0.2342,
      "num_input_tokens_seen": 4127488,
      "step": 19565
    },
    {
      "epoch": 2.152915291529153,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029994665828411,
      "loss": 0.2294,
      "num_input_tokens_seen": 4128576,
      "step": 19570
    },
    {
      "epoch": 2.1534653465346536,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.029994627358624782,
      "loss": 0.2327,
      "num_input_tokens_seen": 4129664,
      "step": 19575
    },
    {
      "epoch": 2.1540154015401543,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029994588750640285,
      "loss": 0.2274,
      "num_input_tokens_seen": 4130688,
      "step": 19580
    },
    {
      "epoch": 2.1545654565456545,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029994550004457856,
      "loss": 0.2276,
      "num_input_tokens_seen": 4131680,
      "step": 19585
    },
    {
      "epoch": 2.155115511551155,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.029994511120077855,
      "loss": 0.2272,
      "num_input_tokens_seen": 4132736,
      "step": 19590
    },
    {
      "epoch": 2.1556655665566558,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.02999447209750064,
      "loss": 0.2342,
      "num_input_tokens_seen": 4133824,
      "step": 19595
    },
    {
      "epoch": 2.1562156215621564,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.029994432936726573,
      "loss": 0.2392,
      "num_input_tokens_seen": 4134880,
      "step": 19600
    },
    {
      "epoch": 2.1567656765676566,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.02999439363775601,
      "loss": 0.2246,
      "num_input_tokens_seen": 4135904,
      "step": 19605
    },
    {
      "epoch": 2.1573157315731573,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.029994354200589317,
      "loss": 0.2361,
      "num_input_tokens_seen": 4136960,
      "step": 19610
    },
    {
      "epoch": 2.157865786578658,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.029994314625226854,
      "loss": 0.2256,
      "num_input_tokens_seen": 4137952,
      "step": 19615
    },
    {
      "epoch": 2.1584158415841586,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.029994274911668988,
      "loss": 0.2263,
      "num_input_tokens_seen": 4139040,
      "step": 19620
    },
    {
      "epoch": 2.1589658965896588,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.02999423505991609,
      "loss": 0.2374,
      "num_input_tokens_seen": 4140064,
      "step": 19625
    },
    {
      "epoch": 2.1595159515951594,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02999419506996852,
      "loss": 0.2296,
      "num_input_tokens_seen": 4141120,
      "step": 19630
    },
    {
      "epoch": 2.16006600660066,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.02999415494182665,
      "loss": 0.2388,
      "num_input_tokens_seen": 4142176,
      "step": 19635
    },
    {
      "epoch": 2.1606160616061607,
      "grad_norm": 0.007537841796875,
      "learning_rate": 0.029994114675490843,
      "loss": 0.2334,
      "num_input_tokens_seen": 4143200,
      "step": 19640
    },
    {
      "epoch": 2.1611661166116614,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.02999407427096148,
      "loss": 0.2302,
      "num_input_tokens_seen": 4144224,
      "step": 19645
    },
    {
      "epoch": 2.1617161716171616,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.029994033728238927,
      "loss": 0.2412,
      "num_input_tokens_seen": 4145216,
      "step": 19650
    },
    {
      "epoch": 2.162266226622662,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.029993993047323565,
      "loss": 0.2322,
      "num_input_tokens_seen": 4146304,
      "step": 19655
    },
    {
      "epoch": 2.162816281628163,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.02999395222821576,
      "loss": 0.2295,
      "num_input_tokens_seen": 4147392,
      "step": 19660
    },
    {
      "epoch": 2.1633663366336635,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.029993911270915892,
      "loss": 0.2294,
      "num_input_tokens_seen": 4148480,
      "step": 19665
    },
    {
      "epoch": 2.1639163916391637,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.029993870175424344,
      "loss": 0.232,
      "num_input_tokens_seen": 4149472,
      "step": 19670
    },
    {
      "epoch": 2.1644664466446644,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.02999382894174148,
      "loss": 0.231,
      "num_input_tokens_seen": 4150464,
      "step": 19675
    },
    {
      "epoch": 2.165016501650165,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.0299937875698677,
      "loss": 0.2346,
      "num_input_tokens_seen": 4151552,
      "step": 19680
    },
    {
      "epoch": 2.1655665566556657,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02999374605980337,
      "loss": 0.2331,
      "num_input_tokens_seen": 4152576,
      "step": 19685
    },
    {
      "epoch": 2.1661166116611663,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.029993704411548874,
      "loss": 0.2305,
      "num_input_tokens_seen": 4153664,
      "step": 19690
    },
    {
      "epoch": 2.1666666666666665,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0299936626251046,
      "loss": 0.2309,
      "num_input_tokens_seen": 4154688,
      "step": 19695
    },
    {
      "epoch": 2.167216721672167,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.029993620700470933,
      "loss": 0.2335,
      "num_input_tokens_seen": 4155712,
      "step": 19700
    },
    {
      "epoch": 2.167766776677668,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.029993578637648264,
      "loss": 0.2314,
      "num_input_tokens_seen": 4156832,
      "step": 19705
    },
    {
      "epoch": 2.1683168316831685,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.02999353643663697,
      "loss": 0.2325,
      "num_input_tokens_seen": 4157824,
      "step": 19710
    },
    {
      "epoch": 2.1688668866886687,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.029993494097437443,
      "loss": 0.2283,
      "num_input_tokens_seen": 4158944,
      "step": 19715
    },
    {
      "epoch": 2.1694169416941693,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.02999345162005008,
      "loss": 0.2319,
      "num_input_tokens_seen": 4160032,
      "step": 19720
    },
    {
      "epoch": 2.16996699669967,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.029993409004475266,
      "loss": 0.2324,
      "num_input_tokens_seen": 4161088,
      "step": 19725
    },
    {
      "epoch": 2.1705170517051706,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.0299933662507134,
      "loss": 0.2303,
      "num_input_tokens_seen": 4162080,
      "step": 19730
    },
    {
      "epoch": 2.1710671067106713,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029993323358764864,
      "loss": 0.2304,
      "num_input_tokens_seen": 4163168,
      "step": 19735
    },
    {
      "epoch": 2.1716171617161715,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.029993280328630068,
      "loss": 0.2288,
      "num_input_tokens_seen": 4164224,
      "step": 19740
    },
    {
      "epoch": 2.172167216721672,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.029993237160309397,
      "loss": 0.2298,
      "num_input_tokens_seen": 4165312,
      "step": 19745
    },
    {
      "epoch": 2.1727172717271728,
      "grad_norm": 0.007171630859375,
      "learning_rate": 0.029993193853803255,
      "loss": 0.2309,
      "num_input_tokens_seen": 4166304,
      "step": 19750
    },
    {
      "epoch": 2.1732673267326734,
      "grad_norm": 0.007476806640625,
      "learning_rate": 0.029993150409112043,
      "loss": 0.2335,
      "num_input_tokens_seen": 4167296,
      "step": 19755
    },
    {
      "epoch": 2.1738173817381736,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.029993106826236154,
      "loss": 0.2289,
      "num_input_tokens_seen": 4168288,
      "step": 19760
    },
    {
      "epoch": 2.1743674367436743,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.029993063105175993,
      "loss": 0.2336,
      "num_input_tokens_seen": 4169312,
      "step": 19765
    },
    {
      "epoch": 2.174917491749175,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.02999301924593197,
      "loss": 0.231,
      "num_input_tokens_seen": 4170400,
      "step": 19770
    },
    {
      "epoch": 2.1754675467546756,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.029992975248504472,
      "loss": 0.2326,
      "num_input_tokens_seen": 4171488,
      "step": 19775
    },
    {
      "epoch": 2.1760176017601762,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.029992931112893927,
      "loss": 0.2331,
      "num_input_tokens_seen": 4172608,
      "step": 19780
    },
    {
      "epoch": 2.1765676567656764,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.029992886839100723,
      "loss": 0.231,
      "num_input_tokens_seen": 4173664,
      "step": 19785
    },
    {
      "epoch": 2.177117711771177,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.02999284242712528,
      "loss": 0.2315,
      "num_input_tokens_seen": 4174720,
      "step": 19790
    },
    {
      "epoch": 2.1776677667766777,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.029992797876968,
      "loss": 0.2315,
      "num_input_tokens_seen": 4175712,
      "step": 19795
    },
    {
      "epoch": 2.1782178217821784,
      "grad_norm": 0.0074462890625,
      "learning_rate": 0.029992753188629297,
      "loss": 0.2336,
      "num_input_tokens_seen": 4176832,
      "step": 19800
    },
    {
      "epoch": 2.1787678767876786,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.02999270836210958,
      "loss": 0.2346,
      "num_input_tokens_seen": 4177952,
      "step": 19805
    },
    {
      "epoch": 2.1793179317931792,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029992663397409264,
      "loss": 0.2278,
      "num_input_tokens_seen": 4179008,
      "step": 19810
    },
    {
      "epoch": 2.17986798679868,
      "grad_norm": 0.002655029296875,
      "learning_rate": 0.029992618294528768,
      "loss": 0.232,
      "num_input_tokens_seen": 4180096,
      "step": 19815
    },
    {
      "epoch": 2.1804180418041805,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.0299925730534685,
      "loss": 0.232,
      "num_input_tokens_seen": 4181120,
      "step": 19820
    },
    {
      "epoch": 2.180968096809681,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.029992527674228883,
      "loss": 0.2325,
      "num_input_tokens_seen": 4182176,
      "step": 19825
    },
    {
      "epoch": 2.1815181518151814,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02999248215681033,
      "loss": 0.2283,
      "num_input_tokens_seen": 4183296,
      "step": 19830
    },
    {
      "epoch": 2.182068206820682,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.029992436501213263,
      "loss": 0.2305,
      "num_input_tokens_seen": 4184320,
      "step": 19835
    },
    {
      "epoch": 2.1826182618261827,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.029992390707438102,
      "loss": 0.2304,
      "num_input_tokens_seen": 4185280,
      "step": 19840
    },
    {
      "epoch": 2.1831683168316833,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.02999234477548527,
      "loss": 0.2284,
      "num_input_tokens_seen": 4186368,
      "step": 19845
    },
    {
      "epoch": 2.1837183718371835,
      "grad_norm": 0.007537841796875,
      "learning_rate": 0.029992298705355195,
      "loss": 0.2363,
      "num_input_tokens_seen": 4187424,
      "step": 19850
    },
    {
      "epoch": 2.184268426842684,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.02999225249704829,
      "loss": 0.232,
      "num_input_tokens_seen": 4188480,
      "step": 19855
    },
    {
      "epoch": 2.184818481848185,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.02999220615056499,
      "loss": 0.2325,
      "num_input_tokens_seen": 4189536,
      "step": 19860
    },
    {
      "epoch": 2.1853685368536855,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.029992159665905724,
      "loss": 0.233,
      "num_input_tokens_seen": 4190592,
      "step": 19865
    },
    {
      "epoch": 2.1859185918591857,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029992113043070912,
      "loss": 0.2325,
      "num_input_tokens_seen": 4191648,
      "step": 19870
    },
    {
      "epoch": 2.1864686468646863,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029992066282060994,
      "loss": 0.2319,
      "num_input_tokens_seen": 4192672,
      "step": 19875
    },
    {
      "epoch": 2.187018701870187,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.02999201938287639,
      "loss": 0.2319,
      "num_input_tokens_seen": 4193664,
      "step": 19880
    },
    {
      "epoch": 2.1875687568756876,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.029991972345517537,
      "loss": 0.2314,
      "num_input_tokens_seen": 4194688,
      "step": 19885
    },
    {
      "epoch": 2.1881188118811883,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02999192516998487,
      "loss": 0.2298,
      "num_input_tokens_seen": 4195776,
      "step": 19890
    },
    {
      "epoch": 2.1886688668866885,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.02999187785627882,
      "loss": 0.2335,
      "num_input_tokens_seen": 4196832,
      "step": 19895
    },
    {
      "epoch": 2.189218921892189,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.029991830404399827,
      "loss": 0.2324,
      "num_input_tokens_seen": 4197984,
      "step": 19900
    },
    {
      "epoch": 2.18976897689769,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029991782814348326,
      "loss": 0.2309,
      "num_input_tokens_seen": 4199040,
      "step": 19905
    },
    {
      "epoch": 2.1903190319031904,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029991735086124755,
      "loss": 0.2314,
      "num_input_tokens_seen": 4200128,
      "step": 19910
    },
    {
      "epoch": 2.190869086908691,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.029991687219729564,
      "loss": 0.2303,
      "num_input_tokens_seen": 4201184,
      "step": 19915
    },
    {
      "epoch": 2.1914191419141913,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029991639215163176,
      "loss": 0.2303,
      "num_input_tokens_seen": 4202240,
      "step": 19920
    },
    {
      "epoch": 2.191969196919692,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.02999159107242605,
      "loss": 0.2308,
      "num_input_tokens_seen": 4203296,
      "step": 19925
    },
    {
      "epoch": 2.1925192519251926,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.029991542791518618,
      "loss": 0.2319,
      "num_input_tokens_seen": 4204352,
      "step": 19930
    },
    {
      "epoch": 2.1930693069306932,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.02999149437244133,
      "loss": 0.2313,
      "num_input_tokens_seen": 4205472,
      "step": 19935
    },
    {
      "epoch": 2.1936193619361934,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.029991445815194633,
      "loss": 0.2308,
      "num_input_tokens_seen": 4206528,
      "step": 19940
    },
    {
      "epoch": 2.194169416941694,
      "grad_norm": 0.01251220703125,
      "learning_rate": 0.029991397119778974,
      "loss": 0.2319,
      "num_input_tokens_seen": 4207584,
      "step": 19945
    },
    {
      "epoch": 2.1947194719471947,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029991348286194805,
      "loss": 0.2319,
      "num_input_tokens_seen": 4208640,
      "step": 19950
    },
    {
      "epoch": 2.1952695269526954,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.029991299314442568,
      "loss": 0.234,
      "num_input_tokens_seen": 4209632,
      "step": 19955
    },
    {
      "epoch": 2.1958195819581956,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.02999125020452272,
      "loss": 0.2293,
      "num_input_tokens_seen": 4210624,
      "step": 19960
    },
    {
      "epoch": 2.1963696369636962,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.02999120095643571,
      "loss": 0.2314,
      "num_input_tokens_seen": 4211680,
      "step": 19965
    },
    {
      "epoch": 2.196919691969197,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.029991151570182,
      "loss": 0.2278,
      "num_input_tokens_seen": 4212800,
      "step": 19970
    },
    {
      "epoch": 2.1974697469746975,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.029991102045762035,
      "loss": 0.2336,
      "num_input_tokens_seen": 4213824,
      "step": 19975
    },
    {
      "epoch": 2.198019801980198,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029991052383176273,
      "loss": 0.2315,
      "num_input_tokens_seen": 4214912,
      "step": 19980
    },
    {
      "epoch": 2.1985698569856984,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02999100258242518,
      "loss": 0.2305,
      "num_input_tokens_seen": 4216032,
      "step": 19985
    },
    {
      "epoch": 2.199119911991199,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.029990952643509206,
      "loss": 0.231,
      "num_input_tokens_seen": 4217056,
      "step": 19990
    },
    {
      "epoch": 2.1996699669966997,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029990902566428817,
      "loss": 0.2299,
      "num_input_tokens_seen": 4218080,
      "step": 19995
    },
    {
      "epoch": 2.2002200220022003,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.029990852351184472,
      "loss": 0.2357,
      "num_input_tokens_seen": 4219104,
      "step": 20000
    },
    {
      "epoch": 2.200770077007701,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.029990801997776633,
      "loss": 0.2315,
      "num_input_tokens_seen": 4220224,
      "step": 20005
    },
    {
      "epoch": 2.201320132013201,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.029990751506205764,
      "loss": 0.233,
      "num_input_tokens_seen": 4221312,
      "step": 20010
    },
    {
      "epoch": 2.201870187018702,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.029990700876472334,
      "loss": 0.233,
      "num_input_tokens_seen": 4222368,
      "step": 20015
    },
    {
      "epoch": 2.2024202420242025,
      "grad_norm": 0.01251220703125,
      "learning_rate": 0.02999065010857681,
      "loss": 0.2325,
      "num_input_tokens_seen": 4223392,
      "step": 20020
    },
    {
      "epoch": 2.202970297029703,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.029990599202519657,
      "loss": 0.2293,
      "num_input_tokens_seen": 4224416,
      "step": 20025
    },
    {
      "epoch": 2.2035203520352034,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.029990548158301336,
      "loss": 0.2324,
      "num_input_tokens_seen": 4225472,
      "step": 20030
    },
    {
      "epoch": 2.204070407040704,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.029990496975922334,
      "loss": 0.2308,
      "num_input_tokens_seen": 4226560,
      "step": 20035
    },
    {
      "epoch": 2.2046204620462047,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.02999044565538311,
      "loss": 0.2314,
      "num_input_tokens_seen": 4227616,
      "step": 20040
    },
    {
      "epoch": 2.2051705170517053,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029990394196684145,
      "loss": 0.2289,
      "num_input_tokens_seen": 4228640,
      "step": 20045
    },
    {
      "epoch": 2.2057205720572055,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.029990342599825907,
      "loss": 0.2281,
      "num_input_tokens_seen": 4229696,
      "step": 20050
    },
    {
      "epoch": 2.206270627062706,
      "grad_norm": 0.0159912109375,
      "learning_rate": 0.02999029086480888,
      "loss": 0.2318,
      "num_input_tokens_seen": 4230720,
      "step": 20055
    },
    {
      "epoch": 2.206820682068207,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.02999023899163353,
      "loss": 0.2305,
      "num_input_tokens_seen": 4231744,
      "step": 20060
    },
    {
      "epoch": 2.2073707370737075,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.02999018698030034,
      "loss": 0.2222,
      "num_input_tokens_seen": 4232832,
      "step": 20065
    },
    {
      "epoch": 2.207920792079208,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.02999013483080979,
      "loss": 0.2267,
      "num_input_tokens_seen": 4233856,
      "step": 20070
    },
    {
      "epoch": 2.2084708470847083,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.02999008254316236,
      "loss": 0.2264,
      "num_input_tokens_seen": 4234880,
      "step": 20075
    },
    {
      "epoch": 2.209020902090209,
      "grad_norm": 0.003082275390625,
      "learning_rate": 0.029990030117358533,
      "loss": 0.229,
      "num_input_tokens_seen": 4235968,
      "step": 20080
    },
    {
      "epoch": 2.2095709570957096,
      "grad_norm": 0.003204345703125,
      "learning_rate": 0.02998997755339879,
      "loss": 0.2383,
      "num_input_tokens_seen": 4237024,
      "step": 20085
    },
    {
      "epoch": 2.2101210121012103,
      "grad_norm": 0.016845703125,
      "learning_rate": 0.029989924851283618,
      "loss": 0.2345,
      "num_input_tokens_seen": 4238048,
      "step": 20090
    },
    {
      "epoch": 2.2106710671067105,
      "grad_norm": 0.01544189453125,
      "learning_rate": 0.029989872011013503,
      "loss": 0.2324,
      "num_input_tokens_seen": 4239136,
      "step": 20095
    },
    {
      "epoch": 2.211221122112211,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.02998981903258893,
      "loss": 0.2284,
      "num_input_tokens_seen": 4240128,
      "step": 20100
    },
    {
      "epoch": 2.2117711771177118,
      "grad_norm": 0.0172119140625,
      "learning_rate": 0.029989765916010384,
      "loss": 0.2424,
      "num_input_tokens_seen": 4241184,
      "step": 20105
    },
    {
      "epoch": 2.2123212321232124,
      "grad_norm": 0.0026702880859375,
      "learning_rate": 0.029989712661278364,
      "loss": 0.2234,
      "num_input_tokens_seen": 4242272,
      "step": 20110
    },
    {
      "epoch": 2.212871287128713,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.02998965926839335,
      "loss": 0.2271,
      "num_input_tokens_seen": 4243328,
      "step": 20115
    },
    {
      "epoch": 2.2134213421342133,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.029989605737355842,
      "loss": 0.2358,
      "num_input_tokens_seen": 4244384,
      "step": 20120
    },
    {
      "epoch": 2.213971397139714,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.02998955206816633,
      "loss": 0.2307,
      "num_input_tokens_seen": 4245440,
      "step": 20125
    },
    {
      "epoch": 2.2145214521452146,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.029989498260825313,
      "loss": 0.229,
      "num_input_tokens_seen": 4246464,
      "step": 20130
    },
    {
      "epoch": 2.215071507150715,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.029989444315333277,
      "loss": 0.241,
      "num_input_tokens_seen": 4247520,
      "step": 20135
    },
    {
      "epoch": 2.2156215621562154,
      "grad_norm": 0.007537841796875,
      "learning_rate": 0.029989390231690727,
      "loss": 0.234,
      "num_input_tokens_seen": 4248576,
      "step": 20140
    },
    {
      "epoch": 2.216171617161716,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.02998933600989816,
      "loss": 0.2291,
      "num_input_tokens_seen": 4249632,
      "step": 20145
    },
    {
      "epoch": 2.2167216721672167,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.029989281649956078,
      "loss": 0.2332,
      "num_input_tokens_seen": 4250752,
      "step": 20150
    },
    {
      "epoch": 2.2172717271727174,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.02998922715186498,
      "loss": 0.2259,
      "num_input_tokens_seen": 4251776,
      "step": 20155
    },
    {
      "epoch": 2.217821782178218,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.029989172515625367,
      "loss": 0.2319,
      "num_input_tokens_seen": 4252832,
      "step": 20160
    },
    {
      "epoch": 2.218371837183718,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.02998911774123774,
      "loss": 0.2328,
      "num_input_tokens_seen": 4253824,
      "step": 20165
    },
    {
      "epoch": 2.218921892189219,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.029989062828702607,
      "loss": 0.2291,
      "num_input_tokens_seen": 4254848,
      "step": 20170
    },
    {
      "epoch": 2.2194719471947195,
      "grad_norm": 0.0072021484375,
      "learning_rate": 0.02998900777802048,
      "loss": 0.2323,
      "num_input_tokens_seen": 4255904,
      "step": 20175
    },
    {
      "epoch": 2.22002200220022,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.029988952589191858,
      "loss": 0.2374,
      "num_input_tokens_seen": 4256960,
      "step": 20180
    },
    {
      "epoch": 2.2205720572057204,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029988897262217252,
      "loss": 0.2259,
      "num_input_tokens_seen": 4258112,
      "step": 20185
    },
    {
      "epoch": 2.221122112211221,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.02998884179709717,
      "loss": 0.2332,
      "num_input_tokens_seen": 4259200,
      "step": 20190
    },
    {
      "epoch": 2.2216721672167217,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.02998878619383213,
      "loss": 0.2363,
      "num_input_tokens_seen": 4260320,
      "step": 20195
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029988730452422636,
      "loss": 0.2295,
      "num_input_tokens_seen": 4261344,
      "step": 20200
    },
    {
      "epoch": 2.222772277227723,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.02998867457286921,
      "loss": 0.2347,
      "num_input_tokens_seen": 4262400,
      "step": 20205
    },
    {
      "epoch": 2.223322332233223,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.02998861855517236,
      "loss": 0.2269,
      "num_input_tokens_seen": 4263424,
      "step": 20210
    },
    {
      "epoch": 2.223872387238724,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029988562399332605,
      "loss": 0.231,
      "num_input_tokens_seen": 4264480,
      "step": 20215
    },
    {
      "epoch": 2.2244224422442245,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.029988506105350463,
      "loss": 0.2326,
      "num_input_tokens_seen": 4265504,
      "step": 20220
    },
    {
      "epoch": 2.224972497249725,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.02998844967322645,
      "loss": 0.2327,
      "num_input_tokens_seen": 4266528,
      "step": 20225
    },
    {
      "epoch": 2.2255225522552253,
      "grad_norm": 0.01190185546875,
      "learning_rate": 0.029988393102961093,
      "loss": 0.2321,
      "num_input_tokens_seen": 4267680,
      "step": 20230
    },
    {
      "epoch": 2.226072607260726,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029988336394554907,
      "loss": 0.2358,
      "num_input_tokens_seen": 4268736,
      "step": 20235
    },
    {
      "epoch": 2.2266226622662266,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029988279548008416,
      "loss": 0.2295,
      "num_input_tokens_seen": 4269760,
      "step": 20240
    },
    {
      "epoch": 2.2271727172717273,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029988222563322144,
      "loss": 0.2326,
      "num_input_tokens_seen": 4270912,
      "step": 20245
    },
    {
      "epoch": 2.227722772277228,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.029988165440496616,
      "loss": 0.2367,
      "num_input_tokens_seen": 4272096,
      "step": 20250
    },
    {
      "epoch": 2.228272827282728,
      "grad_norm": 0.01239013671875,
      "learning_rate": 0.02998810817953236,
      "loss": 0.2336,
      "num_input_tokens_seen": 4273152,
      "step": 20255
    },
    {
      "epoch": 2.228822882288229,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029988050780429907,
      "loss": 0.2309,
      "num_input_tokens_seen": 4274176,
      "step": 20260
    },
    {
      "epoch": 2.2293729372937294,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.029987993243189775,
      "loss": 0.2319,
      "num_input_tokens_seen": 4275264,
      "step": 20265
    },
    {
      "epoch": 2.22992299229923,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.02998793556781251,
      "loss": 0.2329,
      "num_input_tokens_seen": 4276352,
      "step": 20270
    },
    {
      "epoch": 2.2304730473047303,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029987877754298627,
      "loss": 0.2319,
      "num_input_tokens_seen": 4277408,
      "step": 20275
    },
    {
      "epoch": 2.231023102310231,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.029987819802648665,
      "loss": 0.2314,
      "num_input_tokens_seen": 4278496,
      "step": 20280
    },
    {
      "epoch": 2.2315731573157316,
      "grad_norm": 0.00064849853515625,
      "learning_rate": 0.02998776171286317,
      "loss": 0.2308,
      "num_input_tokens_seen": 4279520,
      "step": 20285
    },
    {
      "epoch": 2.2321232123212322,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.02998770348494266,
      "loss": 0.2309,
      "num_input_tokens_seen": 4280576,
      "step": 20290
    },
    {
      "epoch": 2.232673267326733,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.029987645118887678,
      "loss": 0.2325,
      "num_input_tokens_seen": 4281664,
      "step": 20295
    },
    {
      "epoch": 2.233223322332233,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.029987586614698764,
      "loss": 0.2298,
      "num_input_tokens_seen": 4282688,
      "step": 20300
    },
    {
      "epoch": 2.2337733773377337,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029987527972376457,
      "loss": 0.2335,
      "num_input_tokens_seen": 4283712,
      "step": 20305
    },
    {
      "epoch": 2.2343234323432344,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029987469191921294,
      "loss": 0.2314,
      "num_input_tokens_seen": 4284736,
      "step": 20310
    },
    {
      "epoch": 2.234873487348735,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.02998741027333382,
      "loss": 0.2319,
      "num_input_tokens_seen": 4285792,
      "step": 20315
    },
    {
      "epoch": 2.2354235423542352,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.02998735121661458,
      "loss": 0.2314,
      "num_input_tokens_seen": 4286848,
      "step": 20320
    },
    {
      "epoch": 2.235973597359736,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.029987292021764114,
      "loss": 0.2324,
      "num_input_tokens_seen": 4287840,
      "step": 20325
    },
    {
      "epoch": 2.2365236523652365,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.029987232688782967,
      "loss": 0.2319,
      "num_input_tokens_seen": 4288800,
      "step": 20330
    },
    {
      "epoch": 2.237073707370737,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.02998717321767169,
      "loss": 0.2303,
      "num_input_tokens_seen": 4289888,
      "step": 20335
    },
    {
      "epoch": 2.237623762376238,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02998711360843083,
      "loss": 0.2309,
      "num_input_tokens_seen": 4290976,
      "step": 20340
    },
    {
      "epoch": 2.238173817381738,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.02998705386106093,
      "loss": 0.2314,
      "num_input_tokens_seen": 4292032,
      "step": 20345
    },
    {
      "epoch": 2.2387238723872387,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.02998699397556255,
      "loss": 0.2319,
      "num_input_tokens_seen": 4293088,
      "step": 20350
    },
    {
      "epoch": 2.2392739273927393,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.029986933951936242,
      "loss": 0.234,
      "num_input_tokens_seen": 4294112,
      "step": 20355
    },
    {
      "epoch": 2.23982398239824,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.02998687379018255,
      "loss": 0.2314,
      "num_input_tokens_seen": 4295200,
      "step": 20360
    },
    {
      "epoch": 2.24037403740374,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.029986813490302036,
      "loss": 0.2351,
      "num_input_tokens_seen": 4296224,
      "step": 20365
    },
    {
      "epoch": 2.240924092409241,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02998675305229525,
      "loss": 0.2288,
      "num_input_tokens_seen": 4297280,
      "step": 20370
    },
    {
      "epoch": 2.2414741474147415,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.029986692476162754,
      "loss": 0.2309,
      "num_input_tokens_seen": 4298336,
      "step": 20375
    },
    {
      "epoch": 2.242024202420242,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029986631761905108,
      "loss": 0.2335,
      "num_input_tokens_seen": 4299456,
      "step": 20380
    },
    {
      "epoch": 2.2425742574257423,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.029986570909522864,
      "loss": 0.2283,
      "num_input_tokens_seen": 4300480,
      "step": 20385
    },
    {
      "epoch": 2.243124312431243,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.02998650991901659,
      "loss": 0.2335,
      "num_input_tokens_seen": 4301472,
      "step": 20390
    },
    {
      "epoch": 2.2436743674367436,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.029986448790386845,
      "loss": 0.2314,
      "num_input_tokens_seen": 4302496,
      "step": 20395
    },
    {
      "epoch": 2.2442244224422443,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.029986387523634188,
      "loss": 0.232,
      "num_input_tokens_seen": 4303584,
      "step": 20400
    },
    {
      "epoch": 2.244774477447745,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.029986326118759194,
      "loss": 0.2325,
      "num_input_tokens_seen": 4304704,
      "step": 20405
    },
    {
      "epoch": 2.245324532453245,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029986264575762422,
      "loss": 0.2303,
      "num_input_tokens_seen": 4305728,
      "step": 20410
    },
    {
      "epoch": 2.245874587458746,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.02998620289464444,
      "loss": 0.233,
      "num_input_tokens_seen": 4306752,
      "step": 20415
    },
    {
      "epoch": 2.2464246424642464,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.029986141075405818,
      "loss": 0.2325,
      "num_input_tokens_seen": 4307776,
      "step": 20420
    },
    {
      "epoch": 2.246974697469747,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02998607911804712,
      "loss": 0.2304,
      "num_input_tokens_seen": 4308832,
      "step": 20425
    },
    {
      "epoch": 2.2475247524752477,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02998601702256893,
      "loss": 0.2324,
      "num_input_tokens_seen": 4309888,
      "step": 20430
    },
    {
      "epoch": 2.248074807480748,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.029985954788971807,
      "loss": 0.2319,
      "num_input_tokens_seen": 4310944,
      "step": 20435
    },
    {
      "epoch": 2.2486248624862486,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.02998589241725633,
      "loss": 0.2298,
      "num_input_tokens_seen": 4312032,
      "step": 20440
    },
    {
      "epoch": 2.2491749174917492,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.029985829907423073,
      "loss": 0.2324,
      "num_input_tokens_seen": 4313088,
      "step": 20445
    },
    {
      "epoch": 2.24972497249725,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.029985767259472614,
      "loss": 0.2325,
      "num_input_tokens_seen": 4314112,
      "step": 20450
    },
    {
      "epoch": 2.25027502750275,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.02998570447340553,
      "loss": 0.2319,
      "num_input_tokens_seen": 4315136,
      "step": 20455
    },
    {
      "epoch": 2.2508250825082508,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.029985641549222394,
      "loss": 0.2309,
      "num_input_tokens_seen": 4316224,
      "step": 20460
    },
    {
      "epoch": 2.2513751375137514,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.0299855784869238,
      "loss": 0.2324,
      "num_input_tokens_seen": 4317248,
      "step": 20465
    },
    {
      "epoch": 2.251925192519252,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.02998551528651031,
      "loss": 0.2324,
      "num_input_tokens_seen": 4318304,
      "step": 20470
    },
    {
      "epoch": 2.2524752475247523,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02998545194798252,
      "loss": 0.2309,
      "num_input_tokens_seen": 4319392,
      "step": 20475
    },
    {
      "epoch": 2.253025302530253,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.02998538847134101,
      "loss": 0.2288,
      "num_input_tokens_seen": 4320384,
      "step": 20480
    },
    {
      "epoch": 2.2535753575357536,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.029985324856586365,
      "loss": 0.2304,
      "num_input_tokens_seen": 4321472,
      "step": 20485
    },
    {
      "epoch": 2.254125412541254,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.02998526110371917,
      "loss": 0.2315,
      "num_input_tokens_seen": 4322592,
      "step": 20490
    },
    {
      "epoch": 2.254675467546755,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.029985197212740018,
      "loss": 0.233,
      "num_input_tokens_seen": 4323680,
      "step": 20495
    },
    {
      "epoch": 2.255225522552255,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.02998513318364949,
      "loss": 0.233,
      "num_input_tokens_seen": 4324672,
      "step": 20500
    },
    {
      "epoch": 2.2557755775577557,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.029985069016448183,
      "loss": 0.2309,
      "num_input_tokens_seen": 4325664,
      "step": 20505
    },
    {
      "epoch": 2.2563256325632564,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.029985004711136683,
      "loss": 0.2341,
      "num_input_tokens_seen": 4326752,
      "step": 20510
    },
    {
      "epoch": 2.256875687568757,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.029984940267715587,
      "loss": 0.2304,
      "num_input_tokens_seen": 4327712,
      "step": 20515
    },
    {
      "epoch": 2.2574257425742577,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029984875686185488,
      "loss": 0.233,
      "num_input_tokens_seen": 4328704,
      "step": 20520
    },
    {
      "epoch": 2.257975797579758,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.029984810966546975,
      "loss": 0.2293,
      "num_input_tokens_seen": 4329728,
      "step": 20525
    },
    {
      "epoch": 2.2585258525852585,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.029984746108800656,
      "loss": 0.2299,
      "num_input_tokens_seen": 4330784,
      "step": 20530
    },
    {
      "epoch": 2.259075907590759,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.02998468111294712,
      "loss": 0.2324,
      "num_input_tokens_seen": 4331872,
      "step": 20535
    },
    {
      "epoch": 2.25962596259626,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029984615978986967,
      "loss": 0.2304,
      "num_input_tokens_seen": 4332960,
      "step": 20540
    },
    {
      "epoch": 2.26017601760176,
      "grad_norm": 0.01190185546875,
      "learning_rate": 0.0299845507069208,
      "loss": 0.2314,
      "num_input_tokens_seen": 4333952,
      "step": 20545
    },
    {
      "epoch": 2.2607260726072607,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.02998448529674922,
      "loss": 0.2309,
      "num_input_tokens_seen": 4334944,
      "step": 20550
    },
    {
      "epoch": 2.2612761276127613,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02998441974847283,
      "loss": 0.2309,
      "num_input_tokens_seen": 4336032,
      "step": 20555
    },
    {
      "epoch": 2.261826182618262,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.029984354062092232,
      "loss": 0.2314,
      "num_input_tokens_seen": 4337056,
      "step": 20560
    },
    {
      "epoch": 2.262376237623762,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.029984288237608035,
      "loss": 0.2314,
      "num_input_tokens_seen": 4338112,
      "step": 20565
    },
    {
      "epoch": 2.262926292629263,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.02998422227502084,
      "loss": 0.2325,
      "num_input_tokens_seen": 4339136,
      "step": 20570
    },
    {
      "epoch": 2.2634763476347635,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029984156174331263,
      "loss": 0.2325,
      "num_input_tokens_seen": 4340192,
      "step": 20575
    },
    {
      "epoch": 2.264026402640264,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.0299840899355399,
      "loss": 0.2309,
      "num_input_tokens_seen": 4341248,
      "step": 20580
    },
    {
      "epoch": 2.2645764576457648,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.02998402355864738,
      "loss": 0.2298,
      "num_input_tokens_seen": 4342272,
      "step": 20585
    },
    {
      "epoch": 2.265126512651265,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0299839570436543,
      "loss": 0.2268,
      "num_input_tokens_seen": 4343328,
      "step": 20590
    },
    {
      "epoch": 2.2656765676567656,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029983890390561277,
      "loss": 0.2319,
      "num_input_tokens_seen": 4344352,
      "step": 20595
    },
    {
      "epoch": 2.2662266226622663,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02998382359936893,
      "loss": 0.2289,
      "num_input_tokens_seen": 4345440,
      "step": 20600
    },
    {
      "epoch": 2.266776677667767,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.02998375667007787,
      "loss": 0.2331,
      "num_input_tokens_seen": 4346560,
      "step": 20605
    },
    {
      "epoch": 2.2673267326732676,
      "grad_norm": 0.0128173828125,
      "learning_rate": 0.02998368960268871,
      "loss": 0.232,
      "num_input_tokens_seen": 4347616,
      "step": 20610
    },
    {
      "epoch": 2.2678767876787678,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.029983622397202078,
      "loss": 0.2316,
      "num_input_tokens_seen": 4348672,
      "step": 20615
    },
    {
      "epoch": 2.2684268426842684,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.029983555053618584,
      "loss": 0.2315,
      "num_input_tokens_seen": 4349696,
      "step": 20620
    },
    {
      "epoch": 2.268976897689769,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.029983487571938857,
      "loss": 0.2289,
      "num_input_tokens_seen": 4350688,
      "step": 20625
    },
    {
      "epoch": 2.2695269526952697,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029983419952163512,
      "loss": 0.2325,
      "num_input_tokens_seen": 4351744,
      "step": 20630
    },
    {
      "epoch": 2.27007700770077,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.029983352194293177,
      "loss": 0.233,
      "num_input_tokens_seen": 4352800,
      "step": 20635
    },
    {
      "epoch": 2.2706270627062706,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.029983284298328472,
      "loss": 0.2315,
      "num_input_tokens_seen": 4353888,
      "step": 20640
    },
    {
      "epoch": 2.271177117711771,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.029983216264270026,
      "loss": 0.233,
      "num_input_tokens_seen": 4354944,
      "step": 20645
    },
    {
      "epoch": 2.271727172717272,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.02998314809211847,
      "loss": 0.2325,
      "num_input_tokens_seen": 4355968,
      "step": 20650
    },
    {
      "epoch": 2.272277227722772,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029983079781874422,
      "loss": 0.2289,
      "num_input_tokens_seen": 4356960,
      "step": 20655
    },
    {
      "epoch": 2.2728272827282727,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.02998301133353852,
      "loss": 0.2315,
      "num_input_tokens_seen": 4357984,
      "step": 20660
    },
    {
      "epoch": 2.2733773377337734,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029982942747111392,
      "loss": 0.2315,
      "num_input_tokens_seen": 4358976,
      "step": 20665
    },
    {
      "epoch": 2.273927392739274,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.029982874022593666,
      "loss": 0.2284,
      "num_input_tokens_seen": 4360000,
      "step": 20670
    },
    {
      "epoch": 2.2744774477447747,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.029982805159985984,
      "loss": 0.2304,
      "num_input_tokens_seen": 4361056,
      "step": 20675
    },
    {
      "epoch": 2.275027502750275,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.029982736159288974,
      "loss": 0.2341,
      "num_input_tokens_seen": 4362144,
      "step": 20680
    },
    {
      "epoch": 2.2755775577557755,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.029982667020503277,
      "loss": 0.2299,
      "num_input_tokens_seen": 4363264,
      "step": 20685
    },
    {
      "epoch": 2.276127612761276,
      "grad_norm": 0.01263427734375,
      "learning_rate": 0.02998259774362953,
      "loss": 0.2315,
      "num_input_tokens_seen": 4364352,
      "step": 20690
    },
    {
      "epoch": 2.276677667766777,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.02998252832866836,
      "loss": 0.232,
      "num_input_tokens_seen": 4365472,
      "step": 20695
    },
    {
      "epoch": 2.2772277227722775,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029982458775620422,
      "loss": 0.2351,
      "num_input_tokens_seen": 4366464,
      "step": 20700
    },
    {
      "epoch": 2.2777777777777777,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.02998238908448635,
      "loss": 0.2346,
      "num_input_tokens_seen": 4367520,
      "step": 20705
    },
    {
      "epoch": 2.2783278327832783,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.029982319255266784,
      "loss": 0.2309,
      "num_input_tokens_seen": 4368544,
      "step": 20710
    },
    {
      "epoch": 2.278877887788779,
      "grad_norm": 0.01171875,
      "learning_rate": 0.029982249287962375,
      "loss": 0.2319,
      "num_input_tokens_seen": 4369600,
      "step": 20715
    },
    {
      "epoch": 2.279427942794279,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02998217918257376,
      "loss": 0.2314,
      "num_input_tokens_seen": 4370656,
      "step": 20720
    },
    {
      "epoch": 2.27997799779978,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029982108939101595,
      "loss": 0.2309,
      "num_input_tokens_seen": 4371744,
      "step": 20725
    },
    {
      "epoch": 2.2805280528052805,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.02998203855754651,
      "loss": 0.2309,
      "num_input_tokens_seen": 4372800,
      "step": 20730
    },
    {
      "epoch": 2.281078107810781,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.029981968037909174,
      "loss": 0.2319,
      "num_input_tokens_seen": 4373792,
      "step": 20735
    },
    {
      "epoch": 2.281628162816282,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.029981897380190223,
      "loss": 0.2335,
      "num_input_tokens_seen": 4374816,
      "step": 20740
    },
    {
      "epoch": 2.282178217821782,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.029981826584390317,
      "loss": 0.2324,
      "num_input_tokens_seen": 4375872,
      "step": 20745
    },
    {
      "epoch": 2.2827282728272826,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.0299817556505101,
      "loss": 0.2319,
      "num_input_tokens_seen": 4376928,
      "step": 20750
    },
    {
      "epoch": 2.2832783278327833,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02998168457855023,
      "loss": 0.2314,
      "num_input_tokens_seen": 4377920,
      "step": 20755
    },
    {
      "epoch": 2.283828382838284,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.029981613368511364,
      "loss": 0.2298,
      "num_input_tokens_seen": 4378976,
      "step": 20760
    },
    {
      "epoch": 2.2843784378437846,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.02998154202039416,
      "loss": 0.2345,
      "num_input_tokens_seen": 4380128,
      "step": 20765
    },
    {
      "epoch": 2.284928492849285,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.029981470534199268,
      "loss": 0.2309,
      "num_input_tokens_seen": 4381184,
      "step": 20770
    },
    {
      "epoch": 2.2854785478547854,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.02998139890992735,
      "loss": 0.2319,
      "num_input_tokens_seen": 4382272,
      "step": 20775
    },
    {
      "epoch": 2.286028602860286,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02998132714757907,
      "loss": 0.2335,
      "num_input_tokens_seen": 4383328,
      "step": 20780
    },
    {
      "epoch": 2.2865786578657867,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.029981255247155086,
      "loss": 0.233,
      "num_input_tokens_seen": 4384384,
      "step": 20785
    },
    {
      "epoch": 2.287128712871287,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.02998118320865606,
      "loss": 0.234,
      "num_input_tokens_seen": 4385472,
      "step": 20790
    },
    {
      "epoch": 2.2876787678767876,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.029981111032082657,
      "loss": 0.2319,
      "num_input_tokens_seen": 4386560,
      "step": 20795
    },
    {
      "epoch": 2.2882288228822882,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.02998103871743554,
      "loss": 0.2298,
      "num_input_tokens_seen": 4387648,
      "step": 20800
    },
    {
      "epoch": 2.288778877887789,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.029980966264715383,
      "loss": 0.233,
      "num_input_tokens_seen": 4388736,
      "step": 20805
    },
    {
      "epoch": 2.289328932893289,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.029980893673922846,
      "loss": 0.2325,
      "num_input_tokens_seen": 4389856,
      "step": 20810
    },
    {
      "epoch": 2.2898789878987897,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.029980820945058596,
      "loss": 0.233,
      "num_input_tokens_seen": 4390944,
      "step": 20815
    },
    {
      "epoch": 2.2904290429042904,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029980748078123315,
      "loss": 0.2304,
      "num_input_tokens_seen": 4392000,
      "step": 20820
    },
    {
      "epoch": 2.290979097909791,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02998067507311766,
      "loss": 0.2319,
      "num_input_tokens_seen": 4392960,
      "step": 20825
    },
    {
      "epoch": 2.2915291529152917,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.029980601930042317,
      "loss": 0.2293,
      "num_input_tokens_seen": 4393952,
      "step": 20830
    },
    {
      "epoch": 2.292079207920792,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.029980528648897956,
      "loss": 0.2356,
      "num_input_tokens_seen": 4395040,
      "step": 20835
    },
    {
      "epoch": 2.2926292629262925,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.029980455229685246,
      "loss": 0.233,
      "num_input_tokens_seen": 4396064,
      "step": 20840
    },
    {
      "epoch": 2.293179317931793,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02998038167240487,
      "loss": 0.2299,
      "num_input_tokens_seen": 4397184,
      "step": 20845
    },
    {
      "epoch": 2.293729372937294,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029980307977057506,
      "loss": 0.233,
      "num_input_tokens_seen": 4398208,
      "step": 20850
    },
    {
      "epoch": 2.2942794279427945,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.029980234143643827,
      "loss": 0.2304,
      "num_input_tokens_seen": 4399264,
      "step": 20855
    },
    {
      "epoch": 2.2948294829482947,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029980160172164524,
      "loss": 0.2309,
      "num_input_tokens_seen": 4400320,
      "step": 20860
    },
    {
      "epoch": 2.2953795379537953,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029980086062620273,
      "loss": 0.2324,
      "num_input_tokens_seen": 4401344,
      "step": 20865
    },
    {
      "epoch": 2.295929592959296,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.02998001181501175,
      "loss": 0.2298,
      "num_input_tokens_seen": 4402368,
      "step": 20870
    },
    {
      "epoch": 2.2964796479647966,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.029979937429339652,
      "loss": 0.2319,
      "num_input_tokens_seen": 4403424,
      "step": 20875
    },
    {
      "epoch": 2.297029702970297,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.029979862905604657,
      "loss": 0.2319,
      "num_input_tokens_seen": 4404480,
      "step": 20880
    },
    {
      "epoch": 2.2975797579757975,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.029979788243807454,
      "loss": 0.2319,
      "num_input_tokens_seen": 4405536,
      "step": 20885
    },
    {
      "epoch": 2.298129812981298,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02997971344394873,
      "loss": 0.2309,
      "num_input_tokens_seen": 4406592,
      "step": 20890
    },
    {
      "epoch": 2.298679867986799,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.029979638506029176,
      "loss": 0.2288,
      "num_input_tokens_seen": 4407680,
      "step": 20895
    },
    {
      "epoch": 2.299229922992299,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029979563430049483,
      "loss": 0.2309,
      "num_input_tokens_seen": 4408736,
      "step": 20900
    },
    {
      "epoch": 2.2997799779977997,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.02997948821601034,
      "loss": 0.2325,
      "num_input_tokens_seen": 4409760,
      "step": 20905
    },
    {
      "epoch": 2.3003300330033003,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029979412863912442,
      "loss": 0.2304,
      "num_input_tokens_seen": 4410784,
      "step": 20910
    },
    {
      "epoch": 2.300880088008801,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029979337373756482,
      "loss": 0.2314,
      "num_input_tokens_seen": 4411872,
      "step": 20915
    },
    {
      "epoch": 2.3014301430143016,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.029979261745543162,
      "loss": 0.2315,
      "num_input_tokens_seen": 4412864,
      "step": 20920
    },
    {
      "epoch": 2.301980198019802,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.029979185979273168,
      "loss": 0.2341,
      "num_input_tokens_seen": 4413856,
      "step": 20925
    },
    {
      "epoch": 2.3025302530253025,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02997911007494721,
      "loss": 0.2314,
      "num_input_tokens_seen": 4414880,
      "step": 20930
    },
    {
      "epoch": 2.303080308030803,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02997903403256598,
      "loss": 0.2308,
      "num_input_tokens_seen": 4415904,
      "step": 20935
    },
    {
      "epoch": 2.3036303630363038,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.02997895785213018,
      "loss": 0.2314,
      "num_input_tokens_seen": 4416960,
      "step": 20940
    },
    {
      "epoch": 2.3041804180418044,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.029978881533640513,
      "loss": 0.233,
      "num_input_tokens_seen": 4417952,
      "step": 20945
    },
    {
      "epoch": 2.3047304730473046,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.029978805077097683,
      "loss": 0.2319,
      "num_input_tokens_seen": 4419104,
      "step": 20950
    },
    {
      "epoch": 2.3052805280528053,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029978728482502395,
      "loss": 0.2314,
      "num_input_tokens_seen": 4420160,
      "step": 20955
    },
    {
      "epoch": 2.305830583058306,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029978651749855353,
      "loss": 0.2319,
      "num_input_tokens_seen": 4421184,
      "step": 20960
    },
    {
      "epoch": 2.3063806380638066,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029978574879157265,
      "loss": 0.2319,
      "num_input_tokens_seen": 4422304,
      "step": 20965
    },
    {
      "epoch": 2.3069306930693068,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.029978497870408842,
      "loss": 0.2318,
      "num_input_tokens_seen": 4423392,
      "step": 20970
    },
    {
      "epoch": 2.3074807480748074,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02997842072361079,
      "loss": 0.2319,
      "num_input_tokens_seen": 4424448,
      "step": 20975
    },
    {
      "epoch": 2.308030803080308,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.02997834343876382,
      "loss": 0.2313,
      "num_input_tokens_seen": 4425504,
      "step": 20980
    },
    {
      "epoch": 2.3085808580858087,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.02997826601586865,
      "loss": 0.2324,
      "num_input_tokens_seen": 4426528,
      "step": 20985
    },
    {
      "epoch": 2.309130913091309,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029978188454925985,
      "loss": 0.2308,
      "num_input_tokens_seen": 4427584,
      "step": 20990
    },
    {
      "epoch": 2.3096809680968096,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02997811075593655,
      "loss": 0.2303,
      "num_input_tokens_seen": 4428608,
      "step": 20995
    },
    {
      "epoch": 2.31023102310231,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.02997803291890105,
      "loss": 0.2319,
      "num_input_tokens_seen": 4429664,
      "step": 21000
    },
    {
      "epoch": 2.310781078107811,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.029977954943820207,
      "loss": 0.2313,
      "num_input_tokens_seen": 4430752,
      "step": 21005
    },
    {
      "epoch": 2.3113311331133115,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029977876830694746,
      "loss": 0.2319,
      "num_input_tokens_seen": 4431808,
      "step": 21010
    },
    {
      "epoch": 2.3118811881188117,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029977798579525376,
      "loss": 0.2319,
      "num_input_tokens_seen": 4432800,
      "step": 21015
    },
    {
      "epoch": 2.3124312431243124,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029977720190312827,
      "loss": 0.2278,
      "num_input_tokens_seen": 4433856,
      "step": 21020
    },
    {
      "epoch": 2.312981298129813,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.029977641663057816,
      "loss": 0.2299,
      "num_input_tokens_seen": 4434944,
      "step": 21025
    },
    {
      "epoch": 2.3135313531353137,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.029977562997761067,
      "loss": 0.2309,
      "num_input_tokens_seen": 4436000,
      "step": 21030
    },
    {
      "epoch": 2.3140814081408143,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.02997748419442331,
      "loss": 0.2337,
      "num_input_tokens_seen": 4437056,
      "step": 21035
    },
    {
      "epoch": 2.3146314631463145,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.02997740525304527,
      "loss": 0.2281,
      "num_input_tokens_seen": 4438176,
      "step": 21040
    },
    {
      "epoch": 2.315181518151815,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.02997732617362767,
      "loss": 0.2373,
      "num_input_tokens_seen": 4439264,
      "step": 21045
    },
    {
      "epoch": 2.315731573157316,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029977246956171242,
      "loss": 0.2332,
      "num_input_tokens_seen": 4440288,
      "step": 21050
    },
    {
      "epoch": 2.3162816281628165,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.029977167600676712,
      "loss": 0.232,
      "num_input_tokens_seen": 4441376,
      "step": 21055
    },
    {
      "epoch": 2.3168316831683167,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.02997708810714482,
      "loss": 0.2279,
      "num_input_tokens_seen": 4442496,
      "step": 21060
    },
    {
      "epoch": 2.3173817381738173,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02997700847557629,
      "loss": 0.2285,
      "num_input_tokens_seen": 4443616,
      "step": 21065
    },
    {
      "epoch": 2.317931793179318,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02997692870597186,
      "loss": 0.2289,
      "num_input_tokens_seen": 4444608,
      "step": 21070
    },
    {
      "epoch": 2.3184818481848186,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.02997684879833227,
      "loss": 0.2327,
      "num_input_tokens_seen": 4445664,
      "step": 21075
    },
    {
      "epoch": 2.319031903190319,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029976768752658246,
      "loss": 0.2297,
      "num_input_tokens_seen": 4446656,
      "step": 21080
    },
    {
      "epoch": 2.3195819581958195,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02997668856895053,
      "loss": 0.2307,
      "num_input_tokens_seen": 4447776,
      "step": 21085
    },
    {
      "epoch": 2.32013201320132,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.029976608247209865,
      "loss": 0.2339,
      "num_input_tokens_seen": 4448864,
      "step": 21090
    },
    {
      "epoch": 2.3206820682068208,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.029976527787436986,
      "loss": 0.2297,
      "num_input_tokens_seen": 4449920,
      "step": 21095
    },
    {
      "epoch": 2.3212321232123214,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.02997644718963264,
      "loss": 0.2318,
      "num_input_tokens_seen": 4451008,
      "step": 21100
    },
    {
      "epoch": 2.3217821782178216,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029976366453797563,
      "loss": 0.2281,
      "num_input_tokens_seen": 4452000,
      "step": 21105
    },
    {
      "epoch": 2.3223322332233223,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.029976285579932503,
      "loss": 0.2411,
      "num_input_tokens_seen": 4453056,
      "step": 21110
    },
    {
      "epoch": 2.322882288228823,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.029976204568038206,
      "loss": 0.23,
      "num_input_tokens_seen": 4454112,
      "step": 21115
    },
    {
      "epoch": 2.3234323432343236,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.02997612341811542,
      "loss": 0.2264,
      "num_input_tokens_seen": 4455168,
      "step": 21120
    },
    {
      "epoch": 2.323982398239824,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.029976042130164887,
      "loss": 0.2343,
      "num_input_tokens_seen": 4456256,
      "step": 21125
    },
    {
      "epoch": 2.3245324532453244,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.029975960704187364,
      "loss": 0.2316,
      "num_input_tokens_seen": 4457312,
      "step": 21130
    },
    {
      "epoch": 2.325082508250825,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.029975879140183594,
      "loss": 0.2275,
      "num_input_tokens_seen": 4458368,
      "step": 21135
    },
    {
      "epoch": 2.3256325632563257,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.029975797438154334,
      "loss": 0.2342,
      "num_input_tokens_seen": 4459392,
      "step": 21140
    },
    {
      "epoch": 2.3261826182618264,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.029975715598100333,
      "loss": 0.2336,
      "num_input_tokens_seen": 4460448,
      "step": 21145
    },
    {
      "epoch": 2.3267326732673266,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.02997563362002235,
      "loss": 0.2336,
      "num_input_tokens_seen": 4461504,
      "step": 21150
    },
    {
      "epoch": 2.3272827282728272,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.02997555150392114,
      "loss": 0.2336,
      "num_input_tokens_seen": 4462528,
      "step": 21155
    },
    {
      "epoch": 2.327832783278328,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.029975469249797455,
      "loss": 0.233,
      "num_input_tokens_seen": 4463616,
      "step": 21160
    },
    {
      "epoch": 2.3283828382838285,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.029975386857652053,
      "loss": 0.2335,
      "num_input_tokens_seen": 4464640,
      "step": 21165
    },
    {
      "epoch": 2.3289328932893287,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.0299753043274857,
      "loss": 0.2309,
      "num_input_tokens_seen": 4465696,
      "step": 21170
    },
    {
      "epoch": 2.3294829482948294,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.02997522165929915,
      "loss": 0.2298,
      "num_input_tokens_seen": 4466656,
      "step": 21175
    },
    {
      "epoch": 2.33003300330033,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.029975138853093167,
      "loss": 0.2319,
      "num_input_tokens_seen": 4467648,
      "step": 21180
    },
    {
      "epoch": 2.3305830583058307,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.029975055908868516,
      "loss": 0.2335,
      "num_input_tokens_seen": 4468640,
      "step": 21185
    },
    {
      "epoch": 2.3311331133113313,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029974972826625965,
      "loss": 0.2319,
      "num_input_tokens_seen": 4469664,
      "step": 21190
    },
    {
      "epoch": 2.3316831683168315,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02997488960636627,
      "loss": 0.2293,
      "num_input_tokens_seen": 4470784,
      "step": 21195
    },
    {
      "epoch": 2.332233223322332,
      "grad_norm": 0.01202392578125,
      "learning_rate": 0.029974806248090204,
      "loss": 0.2324,
      "num_input_tokens_seen": 4471936,
      "step": 21200
    },
    {
      "epoch": 2.332783278327833,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.02997472275179854,
      "loss": 0.2314,
      "num_input_tokens_seen": 4472928,
      "step": 21205
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029974639117492034,
      "loss": 0.2346,
      "num_input_tokens_seen": 4473984,
      "step": 21210
    },
    {
      "epoch": 2.333883388338834,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02997455534517147,
      "loss": 0.2304,
      "num_input_tokens_seen": 4474976,
      "step": 21215
    },
    {
      "epoch": 2.3344334433443343,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.02997447143483761,
      "loss": 0.233,
      "num_input_tokens_seen": 4476000,
      "step": 21220
    },
    {
      "epoch": 2.334983498349835,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.029974387386491236,
      "loss": 0.2325,
      "num_input_tokens_seen": 4477056,
      "step": 21225
    },
    {
      "epoch": 2.3355335533553356,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02997430320013312,
      "loss": 0.2324,
      "num_input_tokens_seen": 4478112,
      "step": 21230
    },
    {
      "epoch": 2.336083608360836,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.029974218875764035,
      "loss": 0.2345,
      "num_input_tokens_seen": 4479200,
      "step": 21235
    },
    {
      "epoch": 2.3366336633663365,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.029974134413384756,
      "loss": 0.2288,
      "num_input_tokens_seen": 4480160,
      "step": 21240
    },
    {
      "epoch": 2.337183718371837,
      "grad_norm": 0.01171875,
      "learning_rate": 0.02997404981299607,
      "loss": 0.2304,
      "num_input_tokens_seen": 4481152,
      "step": 21245
    },
    {
      "epoch": 2.337733773377338,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.029973965074598753,
      "loss": 0.2325,
      "num_input_tokens_seen": 4482208,
      "step": 21250
    },
    {
      "epoch": 2.3382838283828384,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.029973880198193584,
      "loss": 0.2304,
      "num_input_tokens_seen": 4483232,
      "step": 21255
    },
    {
      "epoch": 2.3388338833883386,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029973795183781342,
      "loss": 0.2335,
      "num_input_tokens_seen": 4484288,
      "step": 21260
    },
    {
      "epoch": 2.3393839383938393,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.02997371003136282,
      "loss": 0.233,
      "num_input_tokens_seen": 4485376,
      "step": 21265
    },
    {
      "epoch": 2.33993399339934,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029973624740938797,
      "loss": 0.2319,
      "num_input_tokens_seen": 4486528,
      "step": 21270
    },
    {
      "epoch": 2.3404840484048406,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02997353931251006,
      "loss": 0.2298,
      "num_input_tokens_seen": 4487616,
      "step": 21275
    },
    {
      "epoch": 2.3410341034103412,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029973453746077395,
      "loss": 0.2309,
      "num_input_tokens_seen": 4488704,
      "step": 21280
    },
    {
      "epoch": 2.3415841584158414,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029973368041641593,
      "loss": 0.2308,
      "num_input_tokens_seen": 4489824,
      "step": 21285
    },
    {
      "epoch": 2.342134213421342,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.02997328219920344,
      "loss": 0.2324,
      "num_input_tokens_seen": 4490848,
      "step": 21290
    },
    {
      "epoch": 2.3426842684268427,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.029973196218763733,
      "loss": 0.2319,
      "num_input_tokens_seen": 4491840,
      "step": 21295
    },
    {
      "epoch": 2.3432343234323434,
      "grad_norm": 0.0128173828125,
      "learning_rate": 0.02997311010032326,
      "loss": 0.2314,
      "num_input_tokens_seen": 4492960,
      "step": 21300
    },
    {
      "epoch": 2.3437843784378436,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029973023843882816,
      "loss": 0.2308,
      "num_input_tokens_seen": 4494048,
      "step": 21305
    },
    {
      "epoch": 2.3443344334433442,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.029972937449443196,
      "loss": 0.2313,
      "num_input_tokens_seen": 4495168,
      "step": 21310
    },
    {
      "epoch": 2.344884488448845,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.0299728509170052,
      "loss": 0.233,
      "num_input_tokens_seen": 4496192,
      "step": 21315
    },
    {
      "epoch": 2.3454345434543455,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.029972764246569618,
      "loss": 0.2319,
      "num_input_tokens_seen": 4497248,
      "step": 21320
    },
    {
      "epoch": 2.3459845984598457,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.02997267743813725,
      "loss": 0.2308,
      "num_input_tokens_seen": 4498304,
      "step": 21325
    },
    {
      "epoch": 2.3465346534653464,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029972590491708903,
      "loss": 0.2314,
      "num_input_tokens_seen": 4499328,
      "step": 21330
    },
    {
      "epoch": 2.347084708470847,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029972503407285373,
      "loss": 0.2324,
      "num_input_tokens_seen": 4500384,
      "step": 21335
    },
    {
      "epoch": 2.3476347634763477,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.02997241618486746,
      "loss": 0.2313,
      "num_input_tokens_seen": 4501536,
      "step": 21340
    },
    {
      "epoch": 2.3481848184818483,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02997232882445597,
      "loss": 0.2319,
      "num_input_tokens_seen": 4502624,
      "step": 21345
    },
    {
      "epoch": 2.3487348734873486,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.029972241326051717,
      "loss": 0.2314,
      "num_input_tokens_seen": 4503648,
      "step": 21350
    },
    {
      "epoch": 2.349284928492849,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.029972153689655497,
      "loss": 0.233,
      "num_input_tokens_seen": 4504672,
      "step": 21355
    },
    {
      "epoch": 2.34983498349835,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.02997206591526812,
      "loss": 0.2309,
      "num_input_tokens_seen": 4505792,
      "step": 21360
    },
    {
      "epoch": 2.3503850385038505,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.029971978002890403,
      "loss": 0.2288,
      "num_input_tokens_seen": 4506880,
      "step": 21365
    },
    {
      "epoch": 2.350935093509351,
      "grad_norm": 0.002593994140625,
      "learning_rate": 0.02997188995252314,
      "loss": 0.2309,
      "num_input_tokens_seen": 4507904,
      "step": 21370
    },
    {
      "epoch": 2.3514851485148514,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.029971801764167152,
      "loss": 0.2336,
      "num_input_tokens_seen": 4508896,
      "step": 21375
    },
    {
      "epoch": 2.352035203520352,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.029971713437823254,
      "loss": 0.233,
      "num_input_tokens_seen": 4509920,
      "step": 21380
    },
    {
      "epoch": 2.3525852585258527,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.029971624973492258,
      "loss": 0.2314,
      "num_input_tokens_seen": 4510944,
      "step": 21385
    },
    {
      "epoch": 2.3531353135313533,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.029971536371174976,
      "loss": 0.2319,
      "num_input_tokens_seen": 4512032,
      "step": 21390
    },
    {
      "epoch": 2.3536853685368535,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02997144763087223,
      "loss": 0.2319,
      "num_input_tokens_seen": 4513088,
      "step": 21395
    },
    {
      "epoch": 2.354235423542354,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029971358752584835,
      "loss": 0.2319,
      "num_input_tokens_seen": 4514176,
      "step": 21400
    },
    {
      "epoch": 2.354785478547855,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.029971269736313606,
      "loss": 0.2308,
      "num_input_tokens_seen": 4515264,
      "step": 21405
    },
    {
      "epoch": 2.3553355335533555,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.029971180582059373,
      "loss": 0.2313,
      "num_input_tokens_seen": 4516288,
      "step": 21410
    },
    {
      "epoch": 2.3558855885588557,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.02997109128982295,
      "loss": 0.2313,
      "num_input_tokens_seen": 4517376,
      "step": 21415
    },
    {
      "epoch": 2.3564356435643563,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.029971001859605165,
      "loss": 0.2298,
      "num_input_tokens_seen": 4518400,
      "step": 21420
    },
    {
      "epoch": 2.356985698569857,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.029970912291406835,
      "loss": 0.2319,
      "num_input_tokens_seen": 4519424,
      "step": 21425
    },
    {
      "epoch": 2.3575357535753576,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029970822585228794,
      "loss": 0.2351,
      "num_input_tokens_seen": 4520512,
      "step": 21430
    },
    {
      "epoch": 2.3580858085808583,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.02997073274107186,
      "loss": 0.2314,
      "num_input_tokens_seen": 4521568,
      "step": 21435
    },
    {
      "epoch": 2.3586358635863585,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02997064275893687,
      "loss": 0.2324,
      "num_input_tokens_seen": 4522592,
      "step": 21440
    },
    {
      "epoch": 2.359185918591859,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.02997055263882465,
      "loss": 0.2308,
      "num_input_tokens_seen": 4523712,
      "step": 21445
    },
    {
      "epoch": 2.3597359735973598,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02997046238073603,
      "loss": 0.2313,
      "num_input_tokens_seen": 4524832,
      "step": 21450
    },
    {
      "epoch": 2.3602860286028604,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.02997037198467184,
      "loss": 0.2319,
      "num_input_tokens_seen": 4525856,
      "step": 21455
    },
    {
      "epoch": 2.360836083608361,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029970281450632914,
      "loss": 0.2308,
      "num_input_tokens_seen": 4526912,
      "step": 21460
    },
    {
      "epoch": 2.3613861386138613,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02997019077862009,
      "loss": 0.2308,
      "num_input_tokens_seen": 4527968,
      "step": 21465
    },
    {
      "epoch": 2.361936193619362,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0299700999686342,
      "loss": 0.2314,
      "num_input_tokens_seen": 4529024,
      "step": 21470
    },
    {
      "epoch": 2.3624862486248626,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02997000902067608,
      "loss": 0.2319,
      "num_input_tokens_seen": 4530048,
      "step": 21475
    },
    {
      "epoch": 2.363036303630363,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029969917934746577,
      "loss": 0.2325,
      "num_input_tokens_seen": 4531136,
      "step": 21480
    },
    {
      "epoch": 2.3635863586358634,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029969826710846515,
      "loss": 0.2299,
      "num_input_tokens_seen": 4532160,
      "step": 21485
    },
    {
      "epoch": 2.364136413641364,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029969735348976748,
      "loss": 0.2351,
      "num_input_tokens_seen": 4533216,
      "step": 21490
    },
    {
      "epoch": 2.3646864686468647,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.029969643849138115,
      "loss": 0.2308,
      "num_input_tokens_seen": 4534304,
      "step": 21495
    },
    {
      "epoch": 2.3652365236523654,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.02996955221133145,
      "loss": 0.2303,
      "num_input_tokens_seen": 4535328,
      "step": 21500
    },
    {
      "epoch": 2.3657865786578656,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029969460435557614,
      "loss": 0.2308,
      "num_input_tokens_seen": 4536448,
      "step": 21505
    },
    {
      "epoch": 2.366336633663366,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.029969368521817436,
      "loss": 0.2313,
      "num_input_tokens_seen": 4537568,
      "step": 21510
    },
    {
      "epoch": 2.366886688668867,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.02996927647011178,
      "loss": 0.2303,
      "num_input_tokens_seen": 4538624,
      "step": 21515
    },
    {
      "epoch": 2.3674367436743675,
      "grad_norm": 0.01190185546875,
      "learning_rate": 0.029969184280441482,
      "loss": 0.2313,
      "num_input_tokens_seen": 4539744,
      "step": 21520
    },
    {
      "epoch": 2.367986798679868,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0299690919528074,
      "loss": 0.2314,
      "num_input_tokens_seen": 4540864,
      "step": 21525
    },
    {
      "epoch": 2.3685368536853684,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.029968999487210372,
      "loss": 0.2308,
      "num_input_tokens_seen": 4541888,
      "step": 21530
    },
    {
      "epoch": 2.369086908690869,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.02996890688365126,
      "loss": 0.2313,
      "num_input_tokens_seen": 4542944,
      "step": 21535
    },
    {
      "epoch": 2.3696369636963697,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.02996881414213092,
      "loss": 0.2308,
      "num_input_tokens_seen": 4543968,
      "step": 21540
    },
    {
      "epoch": 2.3701870187018703,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0299687212626502,
      "loss": 0.2308,
      "num_input_tokens_seen": 4544992,
      "step": 21545
    },
    {
      "epoch": 2.370737073707371,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.02996862824520996,
      "loss": 0.2319,
      "num_input_tokens_seen": 4546016,
      "step": 21550
    },
    {
      "epoch": 2.371287128712871,
      "grad_norm": 0.01171875,
      "learning_rate": 0.029968535089811055,
      "loss": 0.2324,
      "num_input_tokens_seen": 4547104,
      "step": 21555
    },
    {
      "epoch": 2.371837183718372,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.029968441796454345,
      "loss": 0.2319,
      "num_input_tokens_seen": 4548160,
      "step": 21560
    },
    {
      "epoch": 2.3723872387238725,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.02996834836514069,
      "loss": 0.2299,
      "num_input_tokens_seen": 4549184,
      "step": 21565
    },
    {
      "epoch": 2.372937293729373,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.029968254795870952,
      "loss": 0.2284,
      "num_input_tokens_seen": 4550208,
      "step": 21570
    },
    {
      "epoch": 2.3734873487348733,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.02996816108864599,
      "loss": 0.227,
      "num_input_tokens_seen": 4551296,
      "step": 21575
    },
    {
      "epoch": 2.374037403740374,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.02996806724346667,
      "loss": 0.2344,
      "num_input_tokens_seen": 4552320,
      "step": 21580
    },
    {
      "epoch": 2.3745874587458746,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.029967973260333857,
      "loss": 0.2277,
      "num_input_tokens_seen": 4553312,
      "step": 21585
    },
    {
      "epoch": 2.3751375137513753,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.029967879139248414,
      "loss": 0.2376,
      "num_input_tokens_seen": 4554432,
      "step": 21590
    },
    {
      "epoch": 2.3756875687568755,
      "grad_norm": 0.01171875,
      "learning_rate": 0.02996778488021122,
      "loss": 0.2262,
      "num_input_tokens_seen": 4555488,
      "step": 21595
    },
    {
      "epoch": 2.376237623762376,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.029967690483223126,
      "loss": 0.2371,
      "num_input_tokens_seen": 4556544,
      "step": 21600
    },
    {
      "epoch": 2.3767876787678768,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.029967595948285016,
      "loss": 0.2334,
      "num_input_tokens_seen": 4557568,
      "step": 21605
    },
    {
      "epoch": 2.3773377337733774,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.029967501275397754,
      "loss": 0.2364,
      "num_input_tokens_seen": 4558656,
      "step": 21610
    },
    {
      "epoch": 2.377887788778878,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.029967406464562214,
      "loss": 0.2321,
      "num_input_tokens_seen": 4559744,
      "step": 21615
    },
    {
      "epoch": 2.3784378437843783,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.029967311515779273,
      "loss": 0.2316,
      "num_input_tokens_seen": 4560800,
      "step": 21620
    },
    {
      "epoch": 2.378987898789879,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029967216429049804,
      "loss": 0.2342,
      "num_input_tokens_seen": 4561824,
      "step": 21625
    },
    {
      "epoch": 2.3795379537953796,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02996712120437468,
      "loss": 0.232,
      "num_input_tokens_seen": 4562848,
      "step": 21630
    },
    {
      "epoch": 2.3800880088008802,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.029967025841754785,
      "loss": 0.23,
      "num_input_tokens_seen": 4563904,
      "step": 21635
    },
    {
      "epoch": 2.380638063806381,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02996693034119099,
      "loss": 0.233,
      "num_input_tokens_seen": 4564928,
      "step": 21640
    },
    {
      "epoch": 2.381188118811881,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029966834702684186,
      "loss": 0.2325,
      "num_input_tokens_seen": 4566016,
      "step": 21645
    },
    {
      "epoch": 2.3817381738173817,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02996673892623524,
      "loss": 0.2299,
      "num_input_tokens_seen": 4567104,
      "step": 21650
    },
    {
      "epoch": 2.3822882288228824,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.029966643011845054,
      "loss": 0.2283,
      "num_input_tokens_seen": 4568160,
      "step": 21655
    },
    {
      "epoch": 2.382838283828383,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029966546959514496,
      "loss": 0.2294,
      "num_input_tokens_seen": 4569184,
      "step": 21660
    },
    {
      "epoch": 2.3833883388338832,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.029966450769244455,
      "loss": 0.2299,
      "num_input_tokens_seen": 4570272,
      "step": 21665
    },
    {
      "epoch": 2.383938393839384,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.029966354441035816,
      "loss": 0.2279,
      "num_input_tokens_seen": 4571296,
      "step": 21670
    },
    {
      "epoch": 2.3844884488448845,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.029966257974889478,
      "loss": 0.2322,
      "num_input_tokens_seen": 4572288,
      "step": 21675
    },
    {
      "epoch": 2.385038503850385,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029966161370806318,
      "loss": 0.2349,
      "num_input_tokens_seen": 4573312,
      "step": 21680
    },
    {
      "epoch": 2.3855885588558854,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.029966064628787228,
      "loss": 0.2329,
      "num_input_tokens_seen": 4574368,
      "step": 21685
    },
    {
      "epoch": 2.386138613861386,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.029965967748833103,
      "loss": 0.2374,
      "num_input_tokens_seen": 4575424,
      "step": 21690
    },
    {
      "epoch": 2.3866886688668867,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.029965870730944835,
      "loss": 0.2275,
      "num_input_tokens_seen": 4576448,
      "step": 21695
    },
    {
      "epoch": 2.3872387238723873,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.029965773575123315,
      "loss": 0.2328,
      "num_input_tokens_seen": 4577504,
      "step": 21700
    },
    {
      "epoch": 2.387788778877888,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.02996567628136945,
      "loss": 0.2338,
      "num_input_tokens_seen": 4578528,
      "step": 21705
    },
    {
      "epoch": 2.388338833883388,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029965578849684114,
      "loss": 0.2302,
      "num_input_tokens_seen": 4579552,
      "step": 21710
    },
    {
      "epoch": 2.388888888888889,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.029965481280068228,
      "loss": 0.2306,
      "num_input_tokens_seen": 4580576,
      "step": 21715
    },
    {
      "epoch": 2.3894389438943895,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.02996538357252268,
      "loss": 0.2399,
      "num_input_tokens_seen": 4581664,
      "step": 21720
    },
    {
      "epoch": 2.38998899889989,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.029965285727048378,
      "loss": 0.231,
      "num_input_tokens_seen": 4582752,
      "step": 21725
    },
    {
      "epoch": 2.390539053905391,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.02996518774364621,
      "loss": 0.2326,
      "num_input_tokens_seen": 4583808,
      "step": 21730
    },
    {
      "epoch": 2.391089108910891,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029965089622317094,
      "loss": 0.2273,
      "num_input_tokens_seen": 4584864,
      "step": 21735
    },
    {
      "epoch": 2.3916391639163916,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.029964991363061924,
      "loss": 0.2304,
      "num_input_tokens_seen": 4585920,
      "step": 21740
    },
    {
      "epoch": 2.3921892189218923,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02996489296588161,
      "loss": 0.233,
      "num_input_tokens_seen": 4586944,
      "step": 21745
    },
    {
      "epoch": 2.3927392739273925,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02996479443077706,
      "loss": 0.2299,
      "num_input_tokens_seen": 4588000,
      "step": 21750
    },
    {
      "epoch": 2.393289328932893,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02996469575774918,
      "loss": 0.2314,
      "num_input_tokens_seen": 4589024,
      "step": 21755
    },
    {
      "epoch": 2.393839383938394,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.029964596946798882,
      "loss": 0.2325,
      "num_input_tokens_seen": 4590048,
      "step": 21760
    },
    {
      "epoch": 2.3943894389438944,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.02996449799792707,
      "loss": 0.2304,
      "num_input_tokens_seen": 4591104,
      "step": 21765
    },
    {
      "epoch": 2.394939493949395,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.029964398911134664,
      "loss": 0.2305,
      "num_input_tokens_seen": 4592128,
      "step": 21770
    },
    {
      "epoch": 2.3954895489548953,
      "grad_norm": 0.00286865234375,
      "learning_rate": 0.029964299686422575,
      "loss": 0.2325,
      "num_input_tokens_seen": 4593120,
      "step": 21775
    },
    {
      "epoch": 2.396039603960396,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.029964200323791712,
      "loss": 0.2326,
      "num_input_tokens_seen": 4594144,
      "step": 21780
    },
    {
      "epoch": 2.3965896589658966,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.029964100823243,
      "loss": 0.2346,
      "num_input_tokens_seen": 4595264,
      "step": 21785
    },
    {
      "epoch": 2.3971397139713972,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.029964001184777347,
      "loss": 0.2335,
      "num_input_tokens_seen": 4596352,
      "step": 21790
    },
    {
      "epoch": 2.397689768976898,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029963901408395676,
      "loss": 0.232,
      "num_input_tokens_seen": 4597408,
      "step": 21795
    },
    {
      "epoch": 2.398239823982398,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029963801494098906,
      "loss": 0.232,
      "num_input_tokens_seen": 4598464,
      "step": 21800
    },
    {
      "epoch": 2.3987898789878987,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02996370144188796,
      "loss": 0.2294,
      "num_input_tokens_seen": 4599488,
      "step": 21805
    },
    {
      "epoch": 2.3993399339933994,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.02996360125176376,
      "loss": 0.2325,
      "num_input_tokens_seen": 4600544,
      "step": 21810
    },
    {
      "epoch": 2.3998899889989,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02996350092372722,
      "loss": 0.2324,
      "num_input_tokens_seen": 4601568,
      "step": 21815
    },
    {
      "epoch": 2.4004400440044003,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02996340045777928,
      "loss": 0.2314,
      "num_input_tokens_seen": 4602624,
      "step": 21820
    },
    {
      "epoch": 2.400990099009901,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.029963299853920856,
      "loss": 0.2309,
      "num_input_tokens_seen": 4603680,
      "step": 21825
    },
    {
      "epoch": 2.4015401540154016,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.029963199112152877,
      "loss": 0.2314,
      "num_input_tokens_seen": 4604800,
      "step": 21830
    },
    {
      "epoch": 2.402090209020902,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.029963098232476268,
      "loss": 0.2298,
      "num_input_tokens_seen": 4605920,
      "step": 21835
    },
    {
      "epoch": 2.4026402640264024,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029962997214891967,
      "loss": 0.2335,
      "num_input_tokens_seen": 4607008,
      "step": 21840
    },
    {
      "epoch": 2.403190319031903,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029962896059400904,
      "loss": 0.2314,
      "num_input_tokens_seen": 4608032,
      "step": 21845
    },
    {
      "epoch": 2.4037403740374037,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.029962794766003997,
      "loss": 0.2304,
      "num_input_tokens_seen": 4609024,
      "step": 21850
    },
    {
      "epoch": 2.4042904290429044,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.029962693334702202,
      "loss": 0.2319,
      "num_input_tokens_seen": 4610112,
      "step": 21855
    },
    {
      "epoch": 2.404840484048405,
      "grad_norm": 0.01202392578125,
      "learning_rate": 0.029962591765496434,
      "loss": 0.2325,
      "num_input_tokens_seen": 4611232,
      "step": 21860
    },
    {
      "epoch": 2.405390539053905,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.02996249005838764,
      "loss": 0.2293,
      "num_input_tokens_seen": 4612256,
      "step": 21865
    },
    {
      "epoch": 2.405940594059406,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.029962388213376753,
      "loss": 0.2298,
      "num_input_tokens_seen": 4613312,
      "step": 21870
    },
    {
      "epoch": 2.4064906490649065,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02996228623046472,
      "loss": 0.2293,
      "num_input_tokens_seen": 4614400,
      "step": 21875
    },
    {
      "epoch": 2.407040704070407,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02996218410965247,
      "loss": 0.232,
      "num_input_tokens_seen": 4615424,
      "step": 21880
    },
    {
      "epoch": 2.407590759075908,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029962081850940948,
      "loss": 0.2242,
      "num_input_tokens_seen": 4616416,
      "step": 21885
    },
    {
      "epoch": 2.408140814081408,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.029961979454331097,
      "loss": 0.2347,
      "num_input_tokens_seen": 4617504,
      "step": 21890
    },
    {
      "epoch": 2.4086908690869087,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02996187691982386,
      "loss": 0.2296,
      "num_input_tokens_seen": 4618528,
      "step": 21895
    },
    {
      "epoch": 2.4092409240924093,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029961774247420185,
      "loss": 0.2292,
      "num_input_tokens_seen": 4619552,
      "step": 21900
    },
    {
      "epoch": 2.40979097909791,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02996167143712102,
      "loss": 0.2318,
      "num_input_tokens_seen": 4620640,
      "step": 21905
    },
    {
      "epoch": 2.41034103410341,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029961568488927302,
      "loss": 0.2354,
      "num_input_tokens_seen": 4621760,
      "step": 21910
    },
    {
      "epoch": 2.410891089108911,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.02996146540283999,
      "loss": 0.2308,
      "num_input_tokens_seen": 4622784,
      "step": 21915
    },
    {
      "epoch": 2.4114411441144115,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.02996136217886003,
      "loss": 0.2281,
      "num_input_tokens_seen": 4623808,
      "step": 21920
    },
    {
      "epoch": 2.411991199119912,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.02996125881698837,
      "loss": 0.2322,
      "num_input_tokens_seen": 4624800,
      "step": 21925
    },
    {
      "epoch": 2.4125412541254123,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.029961155317225974,
      "loss": 0.2317,
      "num_input_tokens_seen": 4625856,
      "step": 21930
    },
    {
      "epoch": 2.413091309130913,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.029961051679573785,
      "loss": 0.2395,
      "num_input_tokens_seen": 4626944,
      "step": 21935
    },
    {
      "epoch": 2.4136413641364136,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.02996094790403276,
      "loss": 0.2374,
      "num_input_tokens_seen": 4627968,
      "step": 21940
    },
    {
      "epoch": 2.4141914191419143,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.029960843990603857,
      "loss": 0.232,
      "num_input_tokens_seen": 4629088,
      "step": 21945
    },
    {
      "epoch": 2.414741474147415,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.029960739939288036,
      "loss": 0.2309,
      "num_input_tokens_seen": 4630208,
      "step": 21950
    },
    {
      "epoch": 2.415291529152915,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029960635750086253,
      "loss": 0.2341,
      "num_input_tokens_seen": 4631200,
      "step": 21955
    },
    {
      "epoch": 2.4158415841584158,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.029960531422999467,
      "loss": 0.2315,
      "num_input_tokens_seen": 4632256,
      "step": 21960
    },
    {
      "epoch": 2.4163916391639164,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029960426958028648,
      "loss": 0.233,
      "num_input_tokens_seen": 4633376,
      "step": 21965
    },
    {
      "epoch": 2.416941694169417,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.02996032235517475,
      "loss": 0.2314,
      "num_input_tokens_seen": 4634432,
      "step": 21970
    },
    {
      "epoch": 2.4174917491749177,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.029960217614438737,
      "loss": 0.2314,
      "num_input_tokens_seen": 4635456,
      "step": 21975
    },
    {
      "epoch": 2.418041804180418,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029960112735821574,
      "loss": 0.2319,
      "num_input_tokens_seen": 4636544,
      "step": 21980
    },
    {
      "epoch": 2.4185918591859186,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.029960007719324236,
      "loss": 0.2303,
      "num_input_tokens_seen": 4637664,
      "step": 21985
    },
    {
      "epoch": 2.419141914191419,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029959902564947682,
      "loss": 0.2308,
      "num_input_tokens_seen": 4638656,
      "step": 21990
    },
    {
      "epoch": 2.41969196919692,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02995979727269289,
      "loss": 0.2313,
      "num_input_tokens_seen": 4639712,
      "step": 21995
    },
    {
      "epoch": 2.42024202420242,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.029959691842560816,
      "loss": 0.2314,
      "num_input_tokens_seen": 4640704,
      "step": 22000
    },
    {
      "epoch": 2.4207920792079207,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.02995958627455245,
      "loss": 0.2314,
      "num_input_tokens_seen": 4641760,
      "step": 22005
    },
    {
      "epoch": 2.4213421342134214,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.029959480568668748,
      "loss": 0.2314,
      "num_input_tokens_seen": 4642752,
      "step": 22010
    },
    {
      "epoch": 2.421892189218922,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029959374724910698,
      "loss": 0.2308,
      "num_input_tokens_seen": 4643808,
      "step": 22015
    },
    {
      "epoch": 2.4224422442244222,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02995926874327927,
      "loss": 0.2313,
      "num_input_tokens_seen": 4644896,
      "step": 22020
    },
    {
      "epoch": 2.422992299229923,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029959162623775428,
      "loss": 0.2314,
      "num_input_tokens_seen": 4645952,
      "step": 22025
    },
    {
      "epoch": 2.4235423542354235,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029959056366400175,
      "loss": 0.2313,
      "num_input_tokens_seen": 4647072,
      "step": 22030
    },
    {
      "epoch": 2.424092409240924,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.02995894997115447,
      "loss": 0.2319,
      "num_input_tokens_seen": 4648064,
      "step": 22035
    },
    {
      "epoch": 2.424642464246425,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02995884343803931,
      "loss": 0.2314,
      "num_input_tokens_seen": 4649056,
      "step": 22040
    },
    {
      "epoch": 2.425192519251925,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.029958736767055662,
      "loss": 0.2324,
      "num_input_tokens_seen": 4650144,
      "step": 22045
    },
    {
      "epoch": 2.4257425742574257,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.029958629958204513,
      "loss": 0.2303,
      "num_input_tokens_seen": 4651136,
      "step": 22050
    },
    {
      "epoch": 2.4262926292629263,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029958523011486854,
      "loss": 0.2314,
      "num_input_tokens_seen": 4652160,
      "step": 22055
    },
    {
      "epoch": 2.426842684268427,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029958415926903664,
      "loss": 0.2314,
      "num_input_tokens_seen": 4653312,
      "step": 22060
    },
    {
      "epoch": 2.4273927392739276,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.02995830870445593,
      "loss": 0.2309,
      "num_input_tokens_seen": 4654368,
      "step": 22065
    },
    {
      "epoch": 2.427942794279428,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029958201344144647,
      "loss": 0.2273,
      "num_input_tokens_seen": 4655424,
      "step": 22070
    },
    {
      "epoch": 2.4284928492849285,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.0299580938459708,
      "loss": 0.2325,
      "num_input_tokens_seen": 4656512,
      "step": 22075
    },
    {
      "epoch": 2.429042904290429,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029957986209935374,
      "loss": 0.23,
      "num_input_tokens_seen": 4657632,
      "step": 22080
    },
    {
      "epoch": 2.4295929592959298,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.029957878436039368,
      "loss": 0.2373,
      "num_input_tokens_seen": 4658656,
      "step": 22085
    },
    {
      "epoch": 2.43014301430143,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029957770524283778,
      "loss": 0.2299,
      "num_input_tokens_seen": 4659648,
      "step": 22090
    },
    {
      "epoch": 2.4306930693069306,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02995766247466959,
      "loss": 0.2304,
      "num_input_tokens_seen": 4660736,
      "step": 22095
    },
    {
      "epoch": 2.4312431243124313,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02995755428719781,
      "loss": 0.2278,
      "num_input_tokens_seen": 4661824,
      "step": 22100
    },
    {
      "epoch": 2.431793179317932,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.029957445961869424,
      "loss": 0.2294,
      "num_input_tokens_seen": 4662912,
      "step": 22105
    },
    {
      "epoch": 2.432343234323432,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.02995733749868544,
      "loss": 0.232,
      "num_input_tokens_seen": 4663968,
      "step": 22110
    },
    {
      "epoch": 2.432893289328933,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.02995722889764685,
      "loss": 0.2342,
      "num_input_tokens_seen": 4665024,
      "step": 22115
    },
    {
      "epoch": 2.4334433443344334,
      "grad_norm": 0.007537841796875,
      "learning_rate": 0.02995712015875466,
      "loss": 0.2337,
      "num_input_tokens_seen": 4666176,
      "step": 22120
    },
    {
      "epoch": 2.433993399339934,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.02995701128200987,
      "loss": 0.23,
      "num_input_tokens_seen": 4667232,
      "step": 22125
    },
    {
      "epoch": 2.4345434543454347,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.029956902267413488,
      "loss": 0.2279,
      "num_input_tokens_seen": 4668288,
      "step": 22130
    },
    {
      "epoch": 2.435093509350935,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029956793114966507,
      "loss": 0.2316,
      "num_input_tokens_seen": 4669376,
      "step": 22135
    },
    {
      "epoch": 2.4356435643564356,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.029956683824669945,
      "loss": 0.2351,
      "num_input_tokens_seen": 4670400,
      "step": 22140
    },
    {
      "epoch": 2.4361936193619362,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02995657439652481,
      "loss": 0.2315,
      "num_input_tokens_seen": 4671456,
      "step": 22145
    },
    {
      "epoch": 2.436743674367437,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0299564648305321,
      "loss": 0.232,
      "num_input_tokens_seen": 4672512,
      "step": 22150
    },
    {
      "epoch": 2.4372937293729375,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02995635512669283,
      "loss": 0.2298,
      "num_input_tokens_seen": 4673600,
      "step": 22155
    },
    {
      "epoch": 2.4378437843784377,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.029956245285008017,
      "loss": 0.2304,
      "num_input_tokens_seen": 4674624,
      "step": 22160
    },
    {
      "epoch": 2.4383938393839384,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029956135305478658,
      "loss": 0.232,
      "num_input_tokens_seen": 4675712,
      "step": 22165
    },
    {
      "epoch": 2.438943894389439,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029956025188105785,
      "loss": 0.2345,
      "num_input_tokens_seen": 4676800,
      "step": 22170
    },
    {
      "epoch": 2.4394939493949397,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.0299559149328904,
      "loss": 0.233,
      "num_input_tokens_seen": 4677792,
      "step": 22175
    },
    {
      "epoch": 2.44004400440044,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.029955804539833528,
      "loss": 0.2319,
      "num_input_tokens_seen": 4678816,
      "step": 22180
    },
    {
      "epoch": 2.4405940594059405,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.029955694008936176,
      "loss": 0.2298,
      "num_input_tokens_seen": 4679872,
      "step": 22185
    },
    {
      "epoch": 2.441144114411441,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.029955583340199377,
      "loss": 0.2313,
      "num_input_tokens_seen": 4680896,
      "step": 22190
    },
    {
      "epoch": 2.441694169416942,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02995547253362414,
      "loss": 0.2319,
      "num_input_tokens_seen": 4681952,
      "step": 22195
    },
    {
      "epoch": 2.442244224422442,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029955361589211485,
      "loss": 0.2303,
      "num_input_tokens_seen": 4683008,
      "step": 22200
    },
    {
      "epoch": 2.4427942794279427,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029955250506962437,
      "loss": 0.2314,
      "num_input_tokens_seen": 4684128,
      "step": 22205
    },
    {
      "epoch": 2.4433443344334433,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.029955139286878027,
      "loss": 0.2303,
      "num_input_tokens_seen": 4685184,
      "step": 22210
    },
    {
      "epoch": 2.443894389438944,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.029955027928959273,
      "loss": 0.2319,
      "num_input_tokens_seen": 4686240,
      "step": 22215
    },
    {
      "epoch": 2.4444444444444446,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0299549164332072,
      "loss": 0.2308,
      "num_input_tokens_seen": 4687232,
      "step": 22220
    },
    {
      "epoch": 2.444994499449945,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02995480479962284,
      "loss": 0.2288,
      "num_input_tokens_seen": 4688288,
      "step": 22225
    },
    {
      "epoch": 2.4455445544554455,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.029954693028207222,
      "loss": 0.233,
      "num_input_tokens_seen": 4689280,
      "step": 22230
    },
    {
      "epoch": 2.446094609460946,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.029954581118961376,
      "loss": 0.2309,
      "num_input_tokens_seen": 4690336,
      "step": 22235
    },
    {
      "epoch": 2.446644664466447,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.029954469071886324,
      "loss": 0.2345,
      "num_input_tokens_seen": 4691360,
      "step": 22240
    },
    {
      "epoch": 2.4471947194719474,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.029954356886983113,
      "loss": 0.2304,
      "num_input_tokens_seen": 4692448,
      "step": 22245
    },
    {
      "epoch": 2.4477447744774476,
      "grad_norm": 0.01171875,
      "learning_rate": 0.029954244564252767,
      "loss": 0.2314,
      "num_input_tokens_seen": 4693472,
      "step": 22250
    },
    {
      "epoch": 2.4482948294829483,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029954132103696327,
      "loss": 0.2299,
      "num_input_tokens_seen": 4694592,
      "step": 22255
    },
    {
      "epoch": 2.448844884488449,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.029954019505314827,
      "loss": 0.2325,
      "num_input_tokens_seen": 4695648,
      "step": 22260
    },
    {
      "epoch": 2.449394939493949,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029953906769109303,
      "loss": 0.2346,
      "num_input_tokens_seen": 4696736,
      "step": 22265
    },
    {
      "epoch": 2.44994499449945,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.029953793895080797,
      "loss": 0.233,
      "num_input_tokens_seen": 4697792,
      "step": 22270
    },
    {
      "epoch": 2.4504950495049505,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.029953680883230346,
      "loss": 0.2335,
      "num_input_tokens_seen": 4698944,
      "step": 22275
    },
    {
      "epoch": 2.451045104510451,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029953567733558995,
      "loss": 0.2319,
      "num_input_tokens_seen": 4700032,
      "step": 22280
    },
    {
      "epoch": 2.4515951595159517,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029953454446067788,
      "loss": 0.2335,
      "num_input_tokens_seen": 4701088,
      "step": 22285
    },
    {
      "epoch": 2.452145214521452,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.029953341020757765,
      "loss": 0.2314,
      "num_input_tokens_seen": 4702144,
      "step": 22290
    },
    {
      "epoch": 2.4526952695269526,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029953227457629975,
      "loss": 0.233,
      "num_input_tokens_seen": 4703168,
      "step": 22295
    },
    {
      "epoch": 2.4532453245324533,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02995311375668546,
      "loss": 0.2319,
      "num_input_tokens_seen": 4704256,
      "step": 22300
    },
    {
      "epoch": 2.453795379537954,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.02995299991792527,
      "loss": 0.2314,
      "num_input_tokens_seen": 4705344,
      "step": 22305
    },
    {
      "epoch": 2.4543454345434546,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029952885941350458,
      "loss": 0.2329,
      "num_input_tokens_seen": 4706432,
      "step": 22310
    },
    {
      "epoch": 2.4548954895489548,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.02995277182696207,
      "loss": 0.2319,
      "num_input_tokens_seen": 4707552,
      "step": 22315
    },
    {
      "epoch": 2.4554455445544554,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02995265757476116,
      "loss": 0.2303,
      "num_input_tokens_seen": 4708608,
      "step": 22320
    },
    {
      "epoch": 2.455995599559956,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.029952543184748777,
      "loss": 0.2308,
      "num_input_tokens_seen": 4709632,
      "step": 22325
    },
    {
      "epoch": 2.4565456545654567,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029952428656925982,
      "loss": 0.2324,
      "num_input_tokens_seen": 4710624,
      "step": 22330
    },
    {
      "epoch": 2.457095709570957,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.029952313991293826,
      "loss": 0.2324,
      "num_input_tokens_seen": 4711712,
      "step": 22335
    },
    {
      "epoch": 2.4576457645764576,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029952199187853367,
      "loss": 0.2319,
      "num_input_tokens_seen": 4712832,
      "step": 22340
    },
    {
      "epoch": 2.458195819581958,
      "grad_norm": 0.001953125,
      "learning_rate": 0.029952084246605663,
      "loss": 0.2318,
      "num_input_tokens_seen": 4713952,
      "step": 22345
    },
    {
      "epoch": 2.458745874587459,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.02995196916755177,
      "loss": 0.2319,
      "num_input_tokens_seen": 4715008,
      "step": 22350
    },
    {
      "epoch": 2.459295929592959,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029951853950692758,
      "loss": 0.2303,
      "num_input_tokens_seen": 4716096,
      "step": 22355
    },
    {
      "epoch": 2.4598459845984597,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029951738596029676,
      "loss": 0.2303,
      "num_input_tokens_seen": 4717152,
      "step": 22360
    },
    {
      "epoch": 2.4603960396039604,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0299516231035636,
      "loss": 0.2314,
      "num_input_tokens_seen": 4718208,
      "step": 22365
    },
    {
      "epoch": 2.460946094609461,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029951507473295582,
      "loss": 0.2308,
      "num_input_tokens_seen": 4719232,
      "step": 22370
    },
    {
      "epoch": 2.4614961496149617,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.029951391705226697,
      "loss": 0.2298,
      "num_input_tokens_seen": 4720320,
      "step": 22375
    },
    {
      "epoch": 2.462046204620462,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.02995127579935801,
      "loss": 0.2303,
      "num_input_tokens_seen": 4721408,
      "step": 22380
    },
    {
      "epoch": 2.4625962596259625,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029951159755690588,
      "loss": 0.2298,
      "num_input_tokens_seen": 4722432,
      "step": 22385
    },
    {
      "epoch": 2.463146314631463,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.029951043574225497,
      "loss": 0.2303,
      "num_input_tokens_seen": 4723488,
      "step": 22390
    },
    {
      "epoch": 2.463696369636964,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.029950927254963816,
      "loss": 0.2298,
      "num_input_tokens_seen": 4724512,
      "step": 22395
    },
    {
      "epoch": 2.4642464246424645,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.02995081079790661,
      "loss": 0.2293,
      "num_input_tokens_seen": 4725632,
      "step": 22400
    },
    {
      "epoch": 2.4647964796479647,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029950694203054958,
      "loss": 0.2329,
      "num_input_tokens_seen": 4726720,
      "step": 22405
    },
    {
      "epoch": 2.4653465346534653,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.029950577470409932,
      "loss": 0.2324,
      "num_input_tokens_seen": 4727808,
      "step": 22410
    },
    {
      "epoch": 2.465896589658966,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.029950460599972605,
      "loss": 0.2324,
      "num_input_tokens_seen": 4728832,
      "step": 22415
    },
    {
      "epoch": 2.4664466446644666,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029950343591744054,
      "loss": 0.2329,
      "num_input_tokens_seen": 4729856,
      "step": 22420
    },
    {
      "epoch": 2.466996699669967,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029950226445725366,
      "loss": 0.2314,
      "num_input_tokens_seen": 4730912,
      "step": 22425
    },
    {
      "epoch": 2.4675467546754675,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02995010916191761,
      "loss": 0.2303,
      "num_input_tokens_seen": 4731904,
      "step": 22430
    },
    {
      "epoch": 2.468096809680968,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029949991740321875,
      "loss": 0.2308,
      "num_input_tokens_seen": 4732928,
      "step": 22435
    },
    {
      "epoch": 2.4686468646864688,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029949874180939237,
      "loss": 0.2308,
      "num_input_tokens_seen": 4733984,
      "step": 22440
    },
    {
      "epoch": 2.469196919691969,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.02994975648377078,
      "loss": 0.2303,
      "num_input_tokens_seen": 4734976,
      "step": 22445
    },
    {
      "epoch": 2.4697469746974696,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029949638648817598,
      "loss": 0.2313,
      "num_input_tokens_seen": 4735968,
      "step": 22450
    },
    {
      "epoch": 2.4702970297029703,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.029949520676080768,
      "loss": 0.2324,
      "num_input_tokens_seen": 4737056,
      "step": 22455
    },
    {
      "epoch": 2.470847084708471,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029949402565561375,
      "loss": 0.2319,
      "num_input_tokens_seen": 4738144,
      "step": 22460
    },
    {
      "epoch": 2.4713971397139716,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.02994928431726052,
      "loss": 0.2298,
      "num_input_tokens_seen": 4739232,
      "step": 22465
    },
    {
      "epoch": 2.4719471947194718,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02994916593117928,
      "loss": 0.2309,
      "num_input_tokens_seen": 4740352,
      "step": 22470
    },
    {
      "epoch": 2.4724972497249724,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.029949047407318748,
      "loss": 0.2324,
      "num_input_tokens_seen": 4741408,
      "step": 22475
    },
    {
      "epoch": 2.473047304730473,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.029948928745680024,
      "loss": 0.2288,
      "num_input_tokens_seen": 4742464,
      "step": 22480
    },
    {
      "epoch": 2.4735973597359737,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029948809946264195,
      "loss": 0.2329,
      "num_input_tokens_seen": 4743584,
      "step": 22485
    },
    {
      "epoch": 2.4741474147414744,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029948691009072357,
      "loss": 0.2308,
      "num_input_tokens_seen": 4744704,
      "step": 22490
    },
    {
      "epoch": 2.4746974697469746,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.029948571934105606,
      "loss": 0.2324,
      "num_input_tokens_seen": 4745760,
      "step": 22495
    },
    {
      "epoch": 2.4752475247524752,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.029948452721365044,
      "loss": 0.2324,
      "num_input_tokens_seen": 4746816,
      "step": 22500
    },
    {
      "epoch": 2.475797579757976,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.029948333370851767,
      "loss": 0.2303,
      "num_input_tokens_seen": 4747968,
      "step": 22505
    },
    {
      "epoch": 2.4763476347634765,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029948213882566868,
      "loss": 0.2314,
      "num_input_tokens_seen": 4749056,
      "step": 22510
    },
    {
      "epoch": 2.4768976897689767,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.02994809425651146,
      "loss": 0.2329,
      "num_input_tokens_seen": 4750176,
      "step": 22515
    },
    {
      "epoch": 2.4774477447744774,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.029947974492686637,
      "loss": 0.2303,
      "num_input_tokens_seen": 4751200,
      "step": 22520
    },
    {
      "epoch": 2.477997799779978,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.02994785459109351,
      "loss": 0.2314,
      "num_input_tokens_seen": 4752352,
      "step": 22525
    },
    {
      "epoch": 2.4785478547854787,
      "grad_norm": 0.0026702880859375,
      "learning_rate": 0.029947734551733177,
      "loss": 0.2335,
      "num_input_tokens_seen": 4753408,
      "step": 22530
    },
    {
      "epoch": 2.479097909790979,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029947614374606746,
      "loss": 0.2319,
      "num_input_tokens_seen": 4754432,
      "step": 22535
    },
    {
      "epoch": 2.4796479647964795,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.02994749405971533,
      "loss": 0.2293,
      "num_input_tokens_seen": 4755456,
      "step": 22540
    },
    {
      "epoch": 2.48019801980198,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029947373607060027,
      "loss": 0.2309,
      "num_input_tokens_seen": 4756512,
      "step": 22545
    },
    {
      "epoch": 2.480748074807481,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.029947253016641962,
      "loss": 0.2309,
      "num_input_tokens_seen": 4757600,
      "step": 22550
    },
    {
      "epoch": 2.4812981298129815,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.02994713228846223,
      "loss": 0.2335,
      "num_input_tokens_seen": 4758720,
      "step": 22555
    },
    {
      "epoch": 2.4818481848184817,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.02994701142252196,
      "loss": 0.2309,
      "num_input_tokens_seen": 4759776,
      "step": 22560
    },
    {
      "epoch": 2.4823982398239823,
      "grad_norm": 0.001953125,
      "learning_rate": 0.02994689041882225,
      "loss": 0.234,
      "num_input_tokens_seen": 4760832,
      "step": 22565
    },
    {
      "epoch": 2.482948294829483,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02994676927736423,
      "loss": 0.2289,
      "num_input_tokens_seen": 4761856,
      "step": 22570
    },
    {
      "epoch": 2.4834983498349836,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.029946647998149008,
      "loss": 0.2304,
      "num_input_tokens_seen": 4762848,
      "step": 22575
    },
    {
      "epoch": 2.4840484048404843,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029946526581177704,
      "loss": 0.232,
      "num_input_tokens_seen": 4763904,
      "step": 22580
    },
    {
      "epoch": 2.4845984598459845,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.029946405026451437,
      "loss": 0.233,
      "num_input_tokens_seen": 4764928,
      "step": 22585
    },
    {
      "epoch": 2.485148514851485,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029946283333971323,
      "loss": 0.2314,
      "num_input_tokens_seen": 4765984,
      "step": 22590
    },
    {
      "epoch": 2.485698569856986,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029946161503738487,
      "loss": 0.2325,
      "num_input_tokens_seen": 4767040,
      "step": 22595
    },
    {
      "epoch": 2.4862486248624864,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.029946039535754056,
      "loss": 0.2314,
      "num_input_tokens_seen": 4768128,
      "step": 22600
    },
    {
      "epoch": 2.4867986798679866,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02994591743001915,
      "loss": 0.2315,
      "num_input_tokens_seen": 4769152,
      "step": 22605
    },
    {
      "epoch": 2.4873487348734873,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.029945795186534896,
      "loss": 0.232,
      "num_input_tokens_seen": 4770240,
      "step": 22610
    },
    {
      "epoch": 2.487898789878988,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.029945672805302415,
      "loss": 0.2283,
      "num_input_tokens_seen": 4771296,
      "step": 22615
    },
    {
      "epoch": 2.4884488448844886,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029945550286322845,
      "loss": 0.2325,
      "num_input_tokens_seen": 4772288,
      "step": 22620
    },
    {
      "epoch": 2.488998899889989,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029945427629597305,
      "loss": 0.233,
      "num_input_tokens_seen": 4773312,
      "step": 22625
    },
    {
      "epoch": 2.4895489548954894,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029945304835126932,
      "loss": 0.232,
      "num_input_tokens_seen": 4774432,
      "step": 22630
    },
    {
      "epoch": 2.49009900990099,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.029945181902912853,
      "loss": 0.2346,
      "num_input_tokens_seen": 4775488,
      "step": 22635
    },
    {
      "epoch": 2.4906490649064907,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.029945058832956208,
      "loss": 0.2304,
      "num_input_tokens_seen": 4776480,
      "step": 22640
    },
    {
      "epoch": 2.4911991199119914,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02994493562525813,
      "loss": 0.2341,
      "num_input_tokens_seen": 4777536,
      "step": 22645
    },
    {
      "epoch": 2.4917491749174916,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029944812279819745,
      "loss": 0.2299,
      "num_input_tokens_seen": 4778592,
      "step": 22650
    },
    {
      "epoch": 2.4922992299229922,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.029944688796642194,
      "loss": 0.2314,
      "num_input_tokens_seen": 4779680,
      "step": 22655
    },
    {
      "epoch": 2.492849284928493,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029944565175726624,
      "loss": 0.2303,
      "num_input_tokens_seen": 4780864,
      "step": 22660
    },
    {
      "epoch": 2.4933993399339935,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.029944441417074166,
      "loss": 0.2309,
      "num_input_tokens_seen": 4781952,
      "step": 22665
    },
    {
      "epoch": 2.493949394939494,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02994431752068596,
      "loss": 0.2329,
      "num_input_tokens_seen": 4783008,
      "step": 22670
    },
    {
      "epoch": 2.4944994499449944,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029944193486563152,
      "loss": 0.2314,
      "num_input_tokens_seen": 4784064,
      "step": 22675
    },
    {
      "epoch": 2.495049504950495,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029944069314706886,
      "loss": 0.2319,
      "num_input_tokens_seen": 4785120,
      "step": 22680
    },
    {
      "epoch": 2.4955995599559957,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029943945005118305,
      "loss": 0.2315,
      "num_input_tokens_seen": 4786208,
      "step": 22685
    },
    {
      "epoch": 2.4961496149614963,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.029943820557798546,
      "loss": 0.2321,
      "num_input_tokens_seen": 4787200,
      "step": 22690
    },
    {
      "epoch": 2.4966996699669965,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02994369597274877,
      "loss": 0.2306,
      "num_input_tokens_seen": 4788192,
      "step": 22695
    },
    {
      "epoch": 2.497249724972497,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02994357124997012,
      "loss": 0.2266,
      "num_input_tokens_seen": 4789280,
      "step": 22700
    },
    {
      "epoch": 2.497799779977998,
      "grad_norm": 0.007080078125,
      "learning_rate": 0.029943446389463738,
      "loss": 0.2369,
      "num_input_tokens_seen": 4790400,
      "step": 22705
    },
    {
      "epoch": 2.4983498349834985,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.029943321391230786,
      "loss": 0.2311,
      "num_input_tokens_seen": 4791424,
      "step": 22710
    },
    {
      "epoch": 2.4988998899889987,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.029943196255272413,
      "loss": 0.2285,
      "num_input_tokens_seen": 4792544,
      "step": 22715
    },
    {
      "epoch": 2.4994499449944994,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.029943070981589763,
      "loss": 0.2363,
      "num_input_tokens_seen": 4793600,
      "step": 22720
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.029942945570184003,
      "loss": 0.2378,
      "num_input_tokens_seen": 4794624,
      "step": 22725
    },
    {
      "epoch": 2.5005500550055006,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.029942820021056287,
      "loss": 0.234,
      "num_input_tokens_seen": 4795648,
      "step": 22730
    },
    {
      "epoch": 2.5011001100110013,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.029942694334207768,
      "loss": 0.2324,
      "num_input_tokens_seen": 4796768,
      "step": 22735
    },
    {
      "epoch": 2.5016501650165015,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.029942568509639604,
      "loss": 0.2329,
      "num_input_tokens_seen": 4797824,
      "step": 22740
    },
    {
      "epoch": 2.502200220022002,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.029942442547352954,
      "loss": 0.232,
      "num_input_tokens_seen": 4798912,
      "step": 22745
    },
    {
      "epoch": 2.502750275027503,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.029942316447348984,
      "loss": 0.2314,
      "num_input_tokens_seen": 4799936,
      "step": 22750
    },
    {
      "epoch": 2.5033003300330035,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.029942190209628853,
      "loss": 0.2314,
      "num_input_tokens_seen": 4800992,
      "step": 22755
    },
    {
      "epoch": 2.503850385038504,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.029942063834193722,
      "loss": 0.2303,
      "num_input_tokens_seen": 4802112,
      "step": 22760
    },
    {
      "epoch": 2.5044004400440043,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.02994193732104476,
      "loss": 0.2308,
      "num_input_tokens_seen": 4803136,
      "step": 22765
    },
    {
      "epoch": 2.504950495049505,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.029941810670183134,
      "loss": 0.2294,
      "num_input_tokens_seen": 4804192,
      "step": 22770
    },
    {
      "epoch": 2.5055005500550056,
      "grad_norm": 0.01263427734375,
      "learning_rate": 0.029941683881610007,
      "loss": 0.233,
      "num_input_tokens_seen": 4805216,
      "step": 22775
    },
    {
      "epoch": 2.506050605060506,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029941556955326548,
      "loss": 0.2341,
      "num_input_tokens_seen": 4806304,
      "step": 22780
    },
    {
      "epoch": 2.5066006600660065,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02994142989133393,
      "loss": 0.234,
      "num_input_tokens_seen": 4807424,
      "step": 22785
    },
    {
      "epoch": 2.507150715071507,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.029941302689633322,
      "loss": 0.232,
      "num_input_tokens_seen": 4808480,
      "step": 22790
    },
    {
      "epoch": 2.5077007700770078,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029941175350225894,
      "loss": 0.2314,
      "num_input_tokens_seen": 4809504,
      "step": 22795
    },
    {
      "epoch": 2.5082508250825084,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029941047873112827,
      "loss": 0.2293,
      "num_input_tokens_seen": 4810592,
      "step": 22800
    },
    {
      "epoch": 2.5088008800880086,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029940920258295287,
      "loss": 0.2299,
      "num_input_tokens_seen": 4811648,
      "step": 22805
    },
    {
      "epoch": 2.5093509350935093,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029940792505774458,
      "loss": 0.2304,
      "num_input_tokens_seen": 4812736,
      "step": 22810
    },
    {
      "epoch": 2.50990099009901,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02994066461555151,
      "loss": 0.2299,
      "num_input_tokens_seen": 4813728,
      "step": 22815
    },
    {
      "epoch": 2.5104510451045106,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.02994053658762763,
      "loss": 0.2326,
      "num_input_tokens_seen": 4814816,
      "step": 22820
    },
    {
      "epoch": 2.511001100110011,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02994040842200399,
      "loss": 0.2321,
      "num_input_tokens_seen": 4815808,
      "step": 22825
    },
    {
      "epoch": 2.5115511551155114,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.029940280118681776,
      "loss": 0.2352,
      "num_input_tokens_seen": 4816864,
      "step": 22830
    },
    {
      "epoch": 2.512101210121012,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.029940151677662168,
      "loss": 0.2336,
      "num_input_tokens_seen": 4817888,
      "step": 22835
    },
    {
      "epoch": 2.5126512651265127,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029940023098946354,
      "loss": 0.2283,
      "num_input_tokens_seen": 4818976,
      "step": 22840
    },
    {
      "epoch": 2.5132013201320134,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.029939894382535512,
      "loss": 0.2325,
      "num_input_tokens_seen": 4820032,
      "step": 22845
    },
    {
      "epoch": 2.513751375137514,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02993976552843083,
      "loss": 0.2314,
      "num_input_tokens_seen": 4821056,
      "step": 22850
    },
    {
      "epoch": 2.514301430143014,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029939636536633507,
      "loss": 0.233,
      "num_input_tokens_seen": 4822176,
      "step": 22855
    },
    {
      "epoch": 2.514851485148515,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029939507407144718,
      "loss": 0.2319,
      "num_input_tokens_seen": 4823232,
      "step": 22860
    },
    {
      "epoch": 2.5154015401540155,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02993937813996566,
      "loss": 0.233,
      "num_input_tokens_seen": 4824288,
      "step": 22865
    },
    {
      "epoch": 2.5159515951595157,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02993924873509752,
      "loss": 0.233,
      "num_input_tokens_seen": 4825312,
      "step": 22870
    },
    {
      "epoch": 2.5165016501650164,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.02993911919254149,
      "loss": 0.2284,
      "num_input_tokens_seen": 4826304,
      "step": 22875
    },
    {
      "epoch": 2.517051705170517,
      "grad_norm": 0.0123291015625,
      "learning_rate": 0.029938989512298772,
      "loss": 0.2351,
      "num_input_tokens_seen": 4827360,
      "step": 22880
    },
    {
      "epoch": 2.5176017601760177,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.029938859694370552,
      "loss": 0.232,
      "num_input_tokens_seen": 4828448,
      "step": 22885
    },
    {
      "epoch": 2.5181518151815183,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.029938729738758035,
      "loss": 0.2273,
      "num_input_tokens_seen": 4829504,
      "step": 22890
    },
    {
      "epoch": 2.5187018701870185,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02993859964546241,
      "loss": 0.2295,
      "num_input_tokens_seen": 4830560,
      "step": 22895
    },
    {
      "epoch": 2.519251925192519,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.029938469414484883,
      "loss": 0.231,
      "num_input_tokens_seen": 4831616,
      "step": 22900
    },
    {
      "epoch": 2.51980198019802,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029938339045826646,
      "loss": 0.2259,
      "num_input_tokens_seen": 4832672,
      "step": 22905
    },
    {
      "epoch": 2.5203520352035205,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.029938208539488913,
      "loss": 0.2296,
      "num_input_tokens_seen": 4833728,
      "step": 22910
    },
    {
      "epoch": 2.520902090209021,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.029938077895472877,
      "loss": 0.2282,
      "num_input_tokens_seen": 4834816,
      "step": 22915
    },
    {
      "epoch": 2.5214521452145213,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.029937947113779746,
      "loss": 0.2345,
      "num_input_tokens_seen": 4835808,
      "step": 22920
    },
    {
      "epoch": 2.522002200220022,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.029937816194410723,
      "loss": 0.2376,
      "num_input_tokens_seen": 4836896,
      "step": 22925
    },
    {
      "epoch": 2.5225522552255226,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02993768513736702,
      "loss": 0.2292,
      "num_input_tokens_seen": 4837984,
      "step": 22930
    },
    {
      "epoch": 2.523102310231023,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.029937553942649835,
      "loss": 0.235,
      "num_input_tokens_seen": 4839008,
      "step": 22935
    },
    {
      "epoch": 2.523652365236524,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.029937422610260386,
      "loss": 0.2376,
      "num_input_tokens_seen": 4840064,
      "step": 22940
    },
    {
      "epoch": 2.524202420242024,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.02993729114019988,
      "loss": 0.2336,
      "num_input_tokens_seen": 4841184,
      "step": 22945
    },
    {
      "epoch": 2.5247524752475248,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.029937159532469535,
      "loss": 0.232,
      "num_input_tokens_seen": 4842208,
      "step": 22950
    },
    {
      "epoch": 2.5253025302530254,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.02993702778707055,
      "loss": 0.2309,
      "num_input_tokens_seen": 4843264,
      "step": 22955
    },
    {
      "epoch": 2.5258525852585256,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.02993689590400415,
      "loss": 0.2324,
      "num_input_tokens_seen": 4844288,
      "step": 22960
    },
    {
      "epoch": 2.5264026402640263,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.02993676388327155,
      "loss": 0.2314,
      "num_input_tokens_seen": 4845344,
      "step": 22965
    },
    {
      "epoch": 2.526952695269527,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.02993663172487396,
      "loss": 0.2319,
      "num_input_tokens_seen": 4846336,
      "step": 22970
    },
    {
      "epoch": 2.5275027502750276,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.029936499428812605,
      "loss": 0.2308,
      "num_input_tokens_seen": 4847424,
      "step": 22975
    },
    {
      "epoch": 2.5280528052805282,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.029936366995088705,
      "loss": 0.2308,
      "num_input_tokens_seen": 4848544,
      "step": 22980
    },
    {
      "epoch": 2.5286028602860284,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029936234423703474,
      "loss": 0.2303,
      "num_input_tokens_seen": 4849600,
      "step": 22985
    },
    {
      "epoch": 2.529152915291529,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029936101714658138,
      "loss": 0.2319,
      "num_input_tokens_seen": 4850592,
      "step": 22990
    },
    {
      "epoch": 2.5297029702970297,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.029935968867953917,
      "loss": 0.2314,
      "num_input_tokens_seen": 4851680,
      "step": 22995
    },
    {
      "epoch": 2.5302530253025304,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.02993583588359204,
      "loss": 0.2308,
      "num_input_tokens_seen": 4852704,
      "step": 23000
    },
    {
      "epoch": 2.530803080308031,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029935702761573733,
      "loss": 0.2319,
      "num_input_tokens_seen": 4853792,
      "step": 23005
    },
    {
      "epoch": 2.5313531353135312,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02993556950190022,
      "loss": 0.2303,
      "num_input_tokens_seen": 4854848,
      "step": 23010
    },
    {
      "epoch": 2.531903190319032,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02993543610457273,
      "loss": 0.2319,
      "num_input_tokens_seen": 4855904,
      "step": 23015
    },
    {
      "epoch": 2.5324532453245325,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.029935302569592495,
      "loss": 0.2303,
      "num_input_tokens_seen": 4856992,
      "step": 23020
    },
    {
      "epoch": 2.5330033003300327,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.029935168896960736,
      "loss": 0.2319,
      "num_input_tokens_seen": 4858080,
      "step": 23025
    },
    {
      "epoch": 2.533553355335534,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029935035086678694,
      "loss": 0.2308,
      "num_input_tokens_seen": 4859040,
      "step": 23030
    },
    {
      "epoch": 2.534103410341034,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.0299349011387476,
      "loss": 0.2298,
      "num_input_tokens_seen": 4860096,
      "step": 23035
    },
    {
      "epoch": 2.5346534653465347,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.02993476705316869,
      "loss": 0.2351,
      "num_input_tokens_seen": 4861152,
      "step": 23040
    },
    {
      "epoch": 2.5352035203520353,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.029934632829943197,
      "loss": 0.2319,
      "num_input_tokens_seen": 4862272,
      "step": 23045
    },
    {
      "epoch": 2.5357535753575355,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.029934498469072358,
      "loss": 0.2288,
      "num_input_tokens_seen": 4863360,
      "step": 23050
    },
    {
      "epoch": 2.536303630363036,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.029934363970557418,
      "loss": 0.2319,
      "num_input_tokens_seen": 4864384,
      "step": 23055
    },
    {
      "epoch": 2.536853685368537,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029934229334399612,
      "loss": 0.234,
      "num_input_tokens_seen": 4865440,
      "step": 23060
    },
    {
      "epoch": 2.5374037403740375,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.029934094560600172,
      "loss": 0.2319,
      "num_input_tokens_seen": 4866496,
      "step": 23065
    },
    {
      "epoch": 2.537953795379538,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.029933959649160357,
      "loss": 0.2319,
      "num_input_tokens_seen": 4867616,
      "step": 23070
    },
    {
      "epoch": 2.5385038503850383,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.029933824600081396,
      "loss": 0.2319,
      "num_input_tokens_seen": 4868704,
      "step": 23075
    },
    {
      "epoch": 2.539053905390539,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.029933689413364543,
      "loss": 0.2309,
      "num_input_tokens_seen": 4869856,
      "step": 23080
    },
    {
      "epoch": 2.5396039603960396,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.02993355408901104,
      "loss": 0.2324,
      "num_input_tokens_seen": 4870976,
      "step": 23085
    },
    {
      "epoch": 2.5401540154015403,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.029933418627022132,
      "loss": 0.2313,
      "num_input_tokens_seen": 4872064,
      "step": 23090
    },
    {
      "epoch": 2.540704070407041,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.029933283027399074,
      "loss": 0.2319,
      "num_input_tokens_seen": 4873120,
      "step": 23095
    },
    {
      "epoch": 2.541254125412541,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.029933147290143108,
      "loss": 0.2303,
      "num_input_tokens_seen": 4874176,
      "step": 23100
    },
    {
      "epoch": 2.541804180418042,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029933011415255492,
      "loss": 0.2314,
      "num_input_tokens_seen": 4875232,
      "step": 23105
    },
    {
      "epoch": 2.5423542354235424,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.02993287540273748,
      "loss": 0.2299,
      "num_input_tokens_seen": 4876224,
      "step": 23110
    },
    {
      "epoch": 2.5429042904290426,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029932739252590317,
      "loss": 0.2299,
      "num_input_tokens_seen": 4877280,
      "step": 23115
    },
    {
      "epoch": 2.5434543454345433,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029932602964815258,
      "loss": 0.2304,
      "num_input_tokens_seen": 4878272,
      "step": 23120
    },
    {
      "epoch": 2.544004400440044,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029932466539413564,
      "loss": 0.2309,
      "num_input_tokens_seen": 4879360,
      "step": 23125
    },
    {
      "epoch": 2.5445544554455446,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.029932329976386493,
      "loss": 0.2299,
      "num_input_tokens_seen": 4880384,
      "step": 23130
    },
    {
      "epoch": 2.5451045104510452,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.029932193275735302,
      "loss": 0.2304,
      "num_input_tokens_seen": 4881472,
      "step": 23135
    },
    {
      "epoch": 2.5456545654565454,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02993205643746125,
      "loss": 0.2304,
      "num_input_tokens_seen": 4882528,
      "step": 23140
    },
    {
      "epoch": 2.546204620462046,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.0299319194615656,
      "loss": 0.2314,
      "num_input_tokens_seen": 4883488,
      "step": 23145
    },
    {
      "epoch": 2.5467546754675467,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.02993178234804961,
      "loss": 0.2314,
      "num_input_tokens_seen": 4884512,
      "step": 23150
    },
    {
      "epoch": 2.5473047304730474,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.029931645096914553,
      "loss": 0.2304,
      "num_input_tokens_seen": 4885600,
      "step": 23155
    },
    {
      "epoch": 2.547854785478548,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.029931507708161685,
      "loss": 0.2309,
      "num_input_tokens_seen": 4886720,
      "step": 23160
    },
    {
      "epoch": 2.5484048404840483,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.029931370181792274,
      "loss": 0.2319,
      "num_input_tokens_seen": 4887808,
      "step": 23165
    },
    {
      "epoch": 2.548954895489549,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.029931232517807587,
      "loss": 0.2319,
      "num_input_tokens_seen": 4888864,
      "step": 23170
    },
    {
      "epoch": 2.5495049504950495,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.0299310947162089,
      "loss": 0.2314,
      "num_input_tokens_seen": 4889984,
      "step": 23175
    },
    {
      "epoch": 2.55005500550055,
      "grad_norm": 0.002655029296875,
      "learning_rate": 0.029930956776997475,
      "loss": 0.2319,
      "num_input_tokens_seen": 4891008,
      "step": 23180
    },
    {
      "epoch": 2.550605060506051,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029930818700174586,
      "loss": 0.2298,
      "num_input_tokens_seen": 4892032,
      "step": 23185
    },
    {
      "epoch": 2.551155115511551,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.029930680485741504,
      "loss": 0.2319,
      "num_input_tokens_seen": 4893152,
      "step": 23190
    },
    {
      "epoch": 2.5517051705170517,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.029930542133699507,
      "loss": 0.2335,
      "num_input_tokens_seen": 4894304,
      "step": 23195
    },
    {
      "epoch": 2.5522552255225524,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029930403644049863,
      "loss": 0.2325,
      "num_input_tokens_seen": 4895328,
      "step": 23200
    },
    {
      "epoch": 2.5528052805280526,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02993026501679386,
      "loss": 0.2298,
      "num_input_tokens_seen": 4896384,
      "step": 23205
    },
    {
      "epoch": 2.553355335533553,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.029930126251932764,
      "loss": 0.2319,
      "num_input_tokens_seen": 4897408,
      "step": 23210
    },
    {
      "epoch": 2.553905390539054,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.029929987349467863,
      "loss": 0.232,
      "num_input_tokens_seen": 4898400,
      "step": 23215
    },
    {
      "epoch": 2.5544554455445545,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029929848309400424,
      "loss": 0.2304,
      "num_input_tokens_seen": 4899424,
      "step": 23220
    },
    {
      "epoch": 2.555005500550055,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.029929709131731746,
      "loss": 0.2325,
      "num_input_tokens_seen": 4900512,
      "step": 23225
    },
    {
      "epoch": 2.5555555555555554,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0299295698164631,
      "loss": 0.233,
      "num_input_tokens_seen": 4901632,
      "step": 23230
    },
    {
      "epoch": 2.556105610561056,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02992943036359577,
      "loss": 0.2278,
      "num_input_tokens_seen": 4902624,
      "step": 23235
    },
    {
      "epoch": 2.5566556655665567,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029929290773131046,
      "loss": 0.2288,
      "num_input_tokens_seen": 4903648,
      "step": 23240
    },
    {
      "epoch": 2.5572057205720573,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029929151045070213,
      "loss": 0.2319,
      "num_input_tokens_seen": 4904768,
      "step": 23245
    },
    {
      "epoch": 2.557755775577558,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.02992901117941456,
      "loss": 0.232,
      "num_input_tokens_seen": 4905856,
      "step": 23250
    },
    {
      "epoch": 2.558305830583058,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029928871176165373,
      "loss": 0.2341,
      "num_input_tokens_seen": 4906944,
      "step": 23255
    },
    {
      "epoch": 2.558855885588559,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.029928731035323942,
      "loss": 0.2314,
      "num_input_tokens_seen": 4908064,
      "step": 23260
    },
    {
      "epoch": 2.5594059405940595,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029928590756891565,
      "loss": 0.2299,
      "num_input_tokens_seen": 4909120,
      "step": 23265
    },
    {
      "epoch": 2.55995599559956,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02992845034086953,
      "loss": 0.2324,
      "num_input_tokens_seen": 4910112,
      "step": 23270
    },
    {
      "epoch": 2.5605060506050608,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029928309787259126,
      "loss": 0.2293,
      "num_input_tokens_seen": 4911168,
      "step": 23275
    },
    {
      "epoch": 2.561056105610561,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.02992816909606166,
      "loss": 0.2273,
      "num_input_tokens_seen": 4912320,
      "step": 23280
    },
    {
      "epoch": 2.5616061606160616,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.029928028267278417,
      "loss": 0.232,
      "num_input_tokens_seen": 4913344,
      "step": 23285
    },
    {
      "epoch": 2.5621562156215623,
      "grad_norm": 0.01171875,
      "learning_rate": 0.029927887300910706,
      "loss": 0.2346,
      "num_input_tokens_seen": 4914432,
      "step": 23290
    },
    {
      "epoch": 2.5627062706270625,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029927746196959818,
      "loss": 0.2341,
      "num_input_tokens_seen": 4915488,
      "step": 23295
    },
    {
      "epoch": 2.563256325632563,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.02992760495542706,
      "loss": 0.232,
      "num_input_tokens_seen": 4916512,
      "step": 23300
    },
    {
      "epoch": 2.5638063806380638,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029927463576313724,
      "loss": 0.2351,
      "num_input_tokens_seen": 4917568,
      "step": 23305
    },
    {
      "epoch": 2.5643564356435644,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029927322059621127,
      "loss": 0.2324,
      "num_input_tokens_seen": 4918624,
      "step": 23310
    },
    {
      "epoch": 2.564906490649065,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.02992718040535056,
      "loss": 0.2293,
      "num_input_tokens_seen": 4919648,
      "step": 23315
    },
    {
      "epoch": 2.5654565456545653,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02992703861350333,
      "loss": 0.2309,
      "num_input_tokens_seen": 4920672,
      "step": 23320
    },
    {
      "epoch": 2.566006600660066,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.029926896684080755,
      "loss": 0.232,
      "num_input_tokens_seen": 4921760,
      "step": 23325
    },
    {
      "epoch": 2.5665566556655666,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.029926754617084134,
      "loss": 0.2335,
      "num_input_tokens_seen": 4922848,
      "step": 23330
    },
    {
      "epoch": 2.567106710671067,
      "grad_norm": 0.000888824462890625,
      "learning_rate": 0.029926612412514774,
      "loss": 0.2309,
      "num_input_tokens_seen": 4923904,
      "step": 23335
    },
    {
      "epoch": 2.567656765676568,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.029926470070373995,
      "loss": 0.2314,
      "num_input_tokens_seen": 4924960,
      "step": 23340
    },
    {
      "epoch": 2.568206820682068,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.029926327590663104,
      "loss": 0.2351,
      "num_input_tokens_seen": 4926016,
      "step": 23345
    },
    {
      "epoch": 2.5687568756875687,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02992618497338341,
      "loss": 0.2288,
      "num_input_tokens_seen": 4927040,
      "step": 23350
    },
    {
      "epoch": 2.5693069306930694,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.029926042218536234,
      "loss": 0.2309,
      "num_input_tokens_seen": 4928128,
      "step": 23355
    },
    {
      "epoch": 2.56985698569857,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029925899326122888,
      "loss": 0.2309,
      "num_input_tokens_seen": 4929184,
      "step": 23360
    },
    {
      "epoch": 2.5704070407040707,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.02992575629614469,
      "loss": 0.2304,
      "num_input_tokens_seen": 4930144,
      "step": 23365
    },
    {
      "epoch": 2.570957095709571,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029925613128602962,
      "loss": 0.2319,
      "num_input_tokens_seen": 4931104,
      "step": 23370
    },
    {
      "epoch": 2.5715071507150715,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.029925469823499015,
      "loss": 0.2319,
      "num_input_tokens_seen": 4932160,
      "step": 23375
    },
    {
      "epoch": 2.572057205720572,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.029925326380834175,
      "loss": 0.2288,
      "num_input_tokens_seen": 4933184,
      "step": 23380
    },
    {
      "epoch": 2.5726072607260724,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.029925182800609768,
      "loss": 0.233,
      "num_input_tokens_seen": 4934240,
      "step": 23385
    },
    {
      "epoch": 2.573157315731573,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029925039082827107,
      "loss": 0.2299,
      "num_input_tokens_seen": 4935296,
      "step": 23390
    },
    {
      "epoch": 2.5737073707370737,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02992489522748753,
      "loss": 0.2336,
      "num_input_tokens_seen": 4936320,
      "step": 23395
    },
    {
      "epoch": 2.5742574257425743,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02992475123459235,
      "loss": 0.2315,
      "num_input_tokens_seen": 4937312,
      "step": 23400
    },
    {
      "epoch": 2.574807480748075,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029924607104142898,
      "loss": 0.232,
      "num_input_tokens_seen": 4938368,
      "step": 23405
    },
    {
      "epoch": 2.575357535753575,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.029924462836140506,
      "loss": 0.2294,
      "num_input_tokens_seen": 4939424,
      "step": 23410
    },
    {
      "epoch": 2.575907590759076,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.0299243184305865,
      "loss": 0.2326,
      "num_input_tokens_seen": 4940480,
      "step": 23415
    },
    {
      "epoch": 2.5764576457645765,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.029924173887482215,
      "loss": 0.233,
      "num_input_tokens_seen": 4941504,
      "step": 23420
    },
    {
      "epoch": 2.577007700770077,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029924029206828982,
      "loss": 0.231,
      "num_input_tokens_seen": 4942592,
      "step": 23425
    },
    {
      "epoch": 2.5775577557755778,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.02992388438862813,
      "loss": 0.2372,
      "num_input_tokens_seen": 4943680,
      "step": 23430
    },
    {
      "epoch": 2.578107810781078,
      "grad_norm": 0.01190185546875,
      "learning_rate": 0.029923739432880996,
      "loss": 0.232,
      "num_input_tokens_seen": 4944768,
      "step": 23435
    },
    {
      "epoch": 2.5786578657865786,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.029923594339588917,
      "loss": 0.2315,
      "num_input_tokens_seen": 4945888,
      "step": 23440
    },
    {
      "epoch": 2.5792079207920793,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.029923449108753233,
      "loss": 0.2288,
      "num_input_tokens_seen": 4946912,
      "step": 23445
    },
    {
      "epoch": 2.5797579757975795,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02992330374037528,
      "loss": 0.233,
      "num_input_tokens_seen": 4948032,
      "step": 23450
    },
    {
      "epoch": 2.5803080308030806,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.02992315823445639,
      "loss": 0.234,
      "num_input_tokens_seen": 4949056,
      "step": 23455
    },
    {
      "epoch": 2.580858085808581,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.02992301259099792,
      "loss": 0.2309,
      "num_input_tokens_seen": 4950080,
      "step": 23460
    },
    {
      "epoch": 2.5814081408140814,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029922866810001197,
      "loss": 0.2319,
      "num_input_tokens_seen": 4951136,
      "step": 23465
    },
    {
      "epoch": 2.581958195819582,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.029922720891467573,
      "loss": 0.2319,
      "num_input_tokens_seen": 4952160,
      "step": 23470
    },
    {
      "epoch": 2.5825082508250823,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029922574835398397,
      "loss": 0.2278,
      "num_input_tokens_seen": 4953216,
      "step": 23475
    },
    {
      "epoch": 2.583058305830583,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029922428641795007,
      "loss": 0.2324,
      "num_input_tokens_seen": 4954240,
      "step": 23480
    },
    {
      "epoch": 2.5836083608360836,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.029922282310658747,
      "loss": 0.2283,
      "num_input_tokens_seen": 4955328,
      "step": 23485
    },
    {
      "epoch": 2.5841584158415842,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029922135841990976,
      "loss": 0.2329,
      "num_input_tokens_seen": 4956384,
      "step": 23490
    },
    {
      "epoch": 2.584708470847085,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029921989235793034,
      "loss": 0.2319,
      "num_input_tokens_seen": 4957440,
      "step": 23495
    },
    {
      "epoch": 2.585258525852585,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029921842492066284,
      "loss": 0.2319,
      "num_input_tokens_seen": 4958496,
      "step": 23500
    },
    {
      "epoch": 2.5858085808580857,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.029921695610812073,
      "loss": 0.2304,
      "num_input_tokens_seen": 4959552,
      "step": 23505
    },
    {
      "epoch": 2.5863586358635864,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02992154859203175,
      "loss": 0.2299,
      "num_input_tokens_seen": 4960640,
      "step": 23510
    },
    {
      "epoch": 2.586908690869087,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02992140143572667,
      "loss": 0.2278,
      "num_input_tokens_seen": 4961792,
      "step": 23515
    },
    {
      "epoch": 2.5874587458745877,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.0299212541418982,
      "loss": 0.2299,
      "num_input_tokens_seen": 4962816,
      "step": 23520
    },
    {
      "epoch": 2.588008800880088,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.029921106710547686,
      "loss": 0.2296,
      "num_input_tokens_seen": 4963872,
      "step": 23525
    },
    {
      "epoch": 2.5885588558855885,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.029920959141676494,
      "loss": 0.2364,
      "num_input_tokens_seen": 4964928,
      "step": 23530
    },
    {
      "epoch": 2.589108910891089,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02992081143528598,
      "loss": 0.2316,
      "num_input_tokens_seen": 4965952,
      "step": 23535
    },
    {
      "epoch": 2.5896589658965894,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029920663591377506,
      "loss": 0.2301,
      "num_input_tokens_seen": 4967008,
      "step": 23540
    },
    {
      "epoch": 2.5902090209020905,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029920515609952434,
      "loss": 0.2321,
      "num_input_tokens_seen": 4968128,
      "step": 23545
    },
    {
      "epoch": 2.5907590759075907,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029920367491012134,
      "loss": 0.2389,
      "num_input_tokens_seen": 4969152,
      "step": 23550
    },
    {
      "epoch": 2.5913091309130913,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.02992021923455796,
      "loss": 0.2315,
      "num_input_tokens_seen": 4970208,
      "step": 23555
    },
    {
      "epoch": 2.591859185918592,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02992007084059129,
      "loss": 0.2315,
      "num_input_tokens_seen": 4971232,
      "step": 23560
    },
    {
      "epoch": 2.592409240924092,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029919922309113483,
      "loss": 0.2346,
      "num_input_tokens_seen": 4972256,
      "step": 23565
    },
    {
      "epoch": 2.592959295929593,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029919773640125918,
      "loss": 0.231,
      "num_input_tokens_seen": 4973312,
      "step": 23570
    },
    {
      "epoch": 2.5935093509350935,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029919624833629952,
      "loss": 0.2268,
      "num_input_tokens_seen": 4974304,
      "step": 23575
    },
    {
      "epoch": 2.594059405940594,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029919475889626963,
      "loss": 0.2304,
      "num_input_tokens_seen": 4975360,
      "step": 23580
    },
    {
      "epoch": 2.594609460946095,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02991932680811833,
      "loss": 0.234,
      "num_input_tokens_seen": 4976416,
      "step": 23585
    },
    {
      "epoch": 2.595159515951595,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.029919177589105412,
      "loss": 0.2341,
      "num_input_tokens_seen": 4977408,
      "step": 23590
    },
    {
      "epoch": 2.5957095709570956,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029919028232589597,
      "loss": 0.2319,
      "num_input_tokens_seen": 4978528,
      "step": 23595
    },
    {
      "epoch": 2.5962596259625963,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.02991887873857226,
      "loss": 0.2314,
      "num_input_tokens_seen": 4979584,
      "step": 23600
    },
    {
      "epoch": 2.596809680968097,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029918729107054773,
      "loss": 0.2299,
      "num_input_tokens_seen": 4980672,
      "step": 23605
    },
    {
      "epoch": 2.5973597359735976,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02991857933803852,
      "loss": 0.2319,
      "num_input_tokens_seen": 4981696,
      "step": 23610
    },
    {
      "epoch": 2.597909790979098,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02991842943152488,
      "loss": 0.2315,
      "num_input_tokens_seen": 4982720,
      "step": 23615
    },
    {
      "epoch": 2.5984598459845984,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.029918279387515234,
      "loss": 0.2336,
      "num_input_tokens_seen": 4983744,
      "step": 23620
    },
    {
      "epoch": 2.599009900990099,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029918129206010963,
      "loss": 0.23,
      "num_input_tokens_seen": 4984864,
      "step": 23625
    },
    {
      "epoch": 2.5995599559955993,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.02991797888701346,
      "loss": 0.2341,
      "num_input_tokens_seen": 4985984,
      "step": 23630
    },
    {
      "epoch": 2.6001100110011,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.0299178284305241,
      "loss": 0.2362,
      "num_input_tokens_seen": 4987104,
      "step": 23635
    },
    {
      "epoch": 2.6006600660066006,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.02991767783654427,
      "loss": 0.2314,
      "num_input_tokens_seen": 4988160,
      "step": 23640
    },
    {
      "epoch": 2.6012101210121013,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029917527105075368,
      "loss": 0.2314,
      "num_input_tokens_seen": 4989280,
      "step": 23645
    },
    {
      "epoch": 2.601760176017602,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029917376236118773,
      "loss": 0.2319,
      "num_input_tokens_seen": 4990368,
      "step": 23650
    },
    {
      "epoch": 2.602310231023102,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02991722522967588,
      "loss": 0.2314,
      "num_input_tokens_seen": 4991456,
      "step": 23655
    },
    {
      "epoch": 2.6028602860286028,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.029917074085748084,
      "loss": 0.2319,
      "num_input_tokens_seen": 4992544,
      "step": 23660
    },
    {
      "epoch": 2.6034103410341034,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029916922804336766,
      "loss": 0.233,
      "num_input_tokens_seen": 4993600,
      "step": 23665
    },
    {
      "epoch": 2.603960396039604,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.029916771385443335,
      "loss": 0.2309,
      "num_input_tokens_seen": 4994656,
      "step": 23670
    },
    {
      "epoch": 2.6045104510451047,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.02991661982906918,
      "loss": 0.2303,
      "num_input_tokens_seen": 4995776,
      "step": 23675
    },
    {
      "epoch": 2.605060506050605,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.029916468135215694,
      "loss": 0.2304,
      "num_input_tokens_seen": 4996800,
      "step": 23680
    },
    {
      "epoch": 2.6056105610561056,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.02991631630388428,
      "loss": 0.2294,
      "num_input_tokens_seen": 4997824,
      "step": 23685
    },
    {
      "epoch": 2.606160616061606,
      "grad_norm": 0.01251220703125,
      "learning_rate": 0.029916164335076337,
      "loss": 0.2299,
      "num_input_tokens_seen": 4998880,
      "step": 23690
    },
    {
      "epoch": 2.606710671067107,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.029916012228793264,
      "loss": 0.2237,
      "num_input_tokens_seen": 4999904,
      "step": 23695
    },
    {
      "epoch": 2.6072607260726075,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.029915859985036465,
      "loss": 0.2345,
      "num_input_tokens_seen": 5000928,
      "step": 23700
    },
    {
      "epoch": 2.6078107810781077,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.02991570760380734,
      "loss": 0.2325,
      "num_input_tokens_seen": 5001952,
      "step": 23705
    },
    {
      "epoch": 2.6083608360836084,
      "grad_norm": 0.007080078125,
      "learning_rate": 0.029915555085107293,
      "loss": 0.2274,
      "num_input_tokens_seen": 5003008,
      "step": 23710
    },
    {
      "epoch": 2.608910891089109,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.029915402428937737,
      "loss": 0.2289,
      "num_input_tokens_seen": 5004064,
      "step": 23715
    },
    {
      "epoch": 2.609460946094609,
      "grad_norm": 0.00323486328125,
      "learning_rate": 0.029915249635300074,
      "loss": 0.224,
      "num_input_tokens_seen": 5005088,
      "step": 23720
    },
    {
      "epoch": 2.61001100110011,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.029915096704195707,
      "loss": 0.2337,
      "num_input_tokens_seen": 5006144,
      "step": 23725
    },
    {
      "epoch": 2.6105610561056105,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.029914943635626055,
      "loss": 0.2358,
      "num_input_tokens_seen": 5007200,
      "step": 23730
    },
    {
      "epoch": 2.611111111111111,
      "grad_norm": 0.00323486328125,
      "learning_rate": 0.029914790429592525,
      "loss": 0.2377,
      "num_input_tokens_seen": 5008288,
      "step": 23735
    },
    {
      "epoch": 2.611661166116612,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.029914637086096527,
      "loss": 0.2354,
      "num_input_tokens_seen": 5009344,
      "step": 23740
    },
    {
      "epoch": 2.612211221122112,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.029914483605139472,
      "loss": 0.2287,
      "num_input_tokens_seen": 5010400,
      "step": 23745
    },
    {
      "epoch": 2.6127612761276127,
      "grad_norm": 0.0142822265625,
      "learning_rate": 0.029914329986722783,
      "loss": 0.2365,
      "num_input_tokens_seen": 5011520,
      "step": 23750
    },
    {
      "epoch": 2.6133113311331133,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02991417623084787,
      "loss": 0.2274,
      "num_input_tokens_seen": 5012576,
      "step": 23755
    },
    {
      "epoch": 2.613861386138614,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.029914022337516152,
      "loss": 0.231,
      "num_input_tokens_seen": 5013664,
      "step": 23760
    },
    {
      "epoch": 2.6144114411441146,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.029913868306729045,
      "loss": 0.2352,
      "num_input_tokens_seen": 5014688,
      "step": 23765
    },
    {
      "epoch": 2.614961496149615,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.029913714138487965,
      "loss": 0.2284,
      "num_input_tokens_seen": 5015776,
      "step": 23770
    },
    {
      "epoch": 2.6155115511551155,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.029913559832794347,
      "loss": 0.232,
      "num_input_tokens_seen": 5016864,
      "step": 23775
    },
    {
      "epoch": 2.616061606160616,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0299134053896496,
      "loss": 0.2331,
      "num_input_tokens_seen": 5017888,
      "step": 23780
    },
    {
      "epoch": 2.6166116611661168,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02991325080905515,
      "loss": 0.232,
      "num_input_tokens_seen": 5018944,
      "step": 23785
    },
    {
      "epoch": 2.6171617161716174,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.02991309609101243,
      "loss": 0.2283,
      "num_input_tokens_seen": 5020000,
      "step": 23790
    },
    {
      "epoch": 2.6177117711771176,
      "grad_norm": 0.007415771484375,
      "learning_rate": 0.029912941235522857,
      "loss": 0.2345,
      "num_input_tokens_seen": 5021088,
      "step": 23795
    },
    {
      "epoch": 2.6182618261826183,
      "grad_norm": 0.0172119140625,
      "learning_rate": 0.029912786242587856,
      "loss": 0.2314,
      "num_input_tokens_seen": 5022112,
      "step": 23800
    },
    {
      "epoch": 2.618811881188119,
      "grad_norm": 0.01806640625,
      "learning_rate": 0.029912631112208862,
      "loss": 0.2324,
      "num_input_tokens_seen": 5023168,
      "step": 23805
    },
    {
      "epoch": 2.619361936193619,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.029912475844387304,
      "loss": 0.2324,
      "num_input_tokens_seen": 5024224,
      "step": 23810
    },
    {
      "epoch": 2.6199119911991198,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.029912320439124614,
      "loss": 0.2319,
      "num_input_tokens_seen": 5025312,
      "step": 23815
    },
    {
      "epoch": 2.6204620462046204,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.02991216489642222,
      "loss": 0.2314,
      "num_input_tokens_seen": 5026336,
      "step": 23820
    },
    {
      "epoch": 2.621012101210121,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.029912009216281555,
      "loss": 0.2325,
      "num_input_tokens_seen": 5027424,
      "step": 23825
    },
    {
      "epoch": 2.6215621562156217,
      "grad_norm": 0.017333984375,
      "learning_rate": 0.029911853398704058,
      "loss": 0.2301,
      "num_input_tokens_seen": 5028448,
      "step": 23830
    },
    {
      "epoch": 2.622112211221122,
      "grad_norm": 0.004241943359375,
      "learning_rate": 0.029911697443691168,
      "loss": 0.2219,
      "num_input_tokens_seen": 5029504,
      "step": 23835
    },
    {
      "epoch": 2.6226622662266226,
      "grad_norm": 0.036376953125,
      "learning_rate": 0.029911541351244315,
      "loss": 0.2357,
      "num_input_tokens_seen": 5030528,
      "step": 23840
    },
    {
      "epoch": 2.6232123212321232,
      "grad_norm": 0.0291748046875,
      "learning_rate": 0.02991138512136494,
      "loss": 0.2436,
      "num_input_tokens_seen": 5031584,
      "step": 23845
    },
    {
      "epoch": 2.623762376237624,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.02991122875405448,
      "loss": 0.2322,
      "num_input_tokens_seen": 5032608,
      "step": 23850
    },
    {
      "epoch": 2.6243124312431245,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.029911072249314385,
      "loss": 0.2319,
      "num_input_tokens_seen": 5033792,
      "step": 23855
    },
    {
      "epoch": 2.6248624862486247,
      "grad_norm": 0.003448486328125,
      "learning_rate": 0.02991091560714609,
      "loss": 0.232,
      "num_input_tokens_seen": 5034816,
      "step": 23860
    },
    {
      "epoch": 2.6254125412541254,
      "grad_norm": 0.0205078125,
      "learning_rate": 0.029910758827551044,
      "loss": 0.2309,
      "num_input_tokens_seen": 5035872,
      "step": 23865
    },
    {
      "epoch": 2.625962596259626,
      "grad_norm": 0.0196533203125,
      "learning_rate": 0.029910601910530684,
      "loss": 0.232,
      "num_input_tokens_seen": 5036960,
      "step": 23870
    },
    {
      "epoch": 2.6265126512651267,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.029910444856086463,
      "loss": 0.231,
      "num_input_tokens_seen": 5037952,
      "step": 23875
    },
    {
      "epoch": 2.6270627062706273,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.029910287664219826,
      "loss": 0.2363,
      "num_input_tokens_seen": 5039040,
      "step": 23880
    },
    {
      "epoch": 2.6276127612761275,
      "grad_norm": 0.0031585693359375,
      "learning_rate": 0.02991013033493222,
      "loss": 0.231,
      "num_input_tokens_seen": 5040032,
      "step": 23885
    },
    {
      "epoch": 2.628162816281628,
      "grad_norm": 0.0028076171875,
      "learning_rate": 0.029909972868225103,
      "loss": 0.2294,
      "num_input_tokens_seen": 5041056,
      "step": 23890
    },
    {
      "epoch": 2.628712871287129,
      "grad_norm": 0.003204345703125,
      "learning_rate": 0.02990981526409992,
      "loss": 0.2299,
      "num_input_tokens_seen": 5042112,
      "step": 23895
    },
    {
      "epoch": 2.629262926292629,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.02990965752255812,
      "loss": 0.2301,
      "num_input_tokens_seen": 5043136,
      "step": 23900
    },
    {
      "epoch": 2.6298129812981297,
      "grad_norm": 0.0135498046875,
      "learning_rate": 0.02990949964360116,
      "loss": 0.2358,
      "num_input_tokens_seen": 5044224,
      "step": 23905
    },
    {
      "epoch": 2.6303630363036303,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.0299093416272305,
      "loss": 0.2351,
      "num_input_tokens_seen": 5045344,
      "step": 23910
    },
    {
      "epoch": 2.630913091309131,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.02990918347344759,
      "loss": 0.2314,
      "num_input_tokens_seen": 5046368,
      "step": 23915
    },
    {
      "epoch": 2.6314631463146316,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.02990902518225389,
      "loss": 0.2329,
      "num_input_tokens_seen": 5047488,
      "step": 23920
    },
    {
      "epoch": 2.632013201320132,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.02990886675365086,
      "loss": 0.2319,
      "num_input_tokens_seen": 5048512,
      "step": 23925
    },
    {
      "epoch": 2.6325632563256325,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.02990870818763996,
      "loss": 0.2319,
      "num_input_tokens_seen": 5049600,
      "step": 23930
    },
    {
      "epoch": 2.633113311331133,
      "grad_norm": 0.001953125,
      "learning_rate": 0.02990854948422265,
      "loss": 0.2314,
      "num_input_tokens_seen": 5050624,
      "step": 23935
    },
    {
      "epoch": 2.633663366336634,
      "grad_norm": 0.0076904296875,
      "learning_rate": 0.029908390643400395,
      "loss": 0.2319,
      "num_input_tokens_seen": 5051776,
      "step": 23940
    },
    {
      "epoch": 2.6342134213421344,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.029908231665174655,
      "loss": 0.2335,
      "num_input_tokens_seen": 5052800,
      "step": 23945
    },
    {
      "epoch": 2.6347634763476346,
      "grad_norm": 0.00762939453125,
      "learning_rate": 0.0299080725495469,
      "loss": 0.2361,
      "num_input_tokens_seen": 5053856,
      "step": 23950
    },
    {
      "epoch": 2.6353135313531353,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02990791329651859,
      "loss": 0.2303,
      "num_input_tokens_seen": 5054944,
      "step": 23955
    },
    {
      "epoch": 2.635863586358636,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.029907753906091197,
      "loss": 0.2324,
      "num_input_tokens_seen": 5056000,
      "step": 23960
    },
    {
      "epoch": 2.636413641364136,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.029907594378266192,
      "loss": 0.2319,
      "num_input_tokens_seen": 5057088,
      "step": 23965
    },
    {
      "epoch": 2.6369636963696372,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.029907434713045043,
      "loss": 0.2293,
      "num_input_tokens_seen": 5058144,
      "step": 23970
    },
    {
      "epoch": 2.6375137513751374,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.029907274910429216,
      "loss": 0.2335,
      "num_input_tokens_seen": 5059168,
      "step": 23975
    },
    {
      "epoch": 2.638063806380638,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.029907114970420198,
      "loss": 0.2309,
      "num_input_tokens_seen": 5060160,
      "step": 23980
    },
    {
      "epoch": 2.6386138613861387,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02990695489301945,
      "loss": 0.2309,
      "num_input_tokens_seen": 5061184,
      "step": 23985
    },
    {
      "epoch": 2.639163916391639,
      "grad_norm": 0.013671875,
      "learning_rate": 0.02990679467822845,
      "loss": 0.2319,
      "num_input_tokens_seen": 5062272,
      "step": 23990
    },
    {
      "epoch": 2.6397139713971396,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.029906634326048676,
      "loss": 0.2293,
      "num_input_tokens_seen": 5063424,
      "step": 23995
    },
    {
      "epoch": 2.6402640264026402,
      "grad_norm": 0.0026397705078125,
      "learning_rate": 0.029906473836481613,
      "loss": 0.2299,
      "num_input_tokens_seen": 5064480,
      "step": 24000
    },
    {
      "epoch": 2.640814081408141,
      "grad_norm": 0.01544189453125,
      "learning_rate": 0.029906313209528728,
      "loss": 0.233,
      "num_input_tokens_seen": 5065536,
      "step": 24005
    },
    {
      "epoch": 2.6413641364136415,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.029906152445191507,
      "loss": 0.2305,
      "num_input_tokens_seen": 5066592,
      "step": 24010
    },
    {
      "epoch": 2.6419141914191417,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.02990599154347143,
      "loss": 0.2378,
      "num_input_tokens_seen": 5067584,
      "step": 24015
    },
    {
      "epoch": 2.6424642464246424,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029905830504369988,
      "loss": 0.2315,
      "num_input_tokens_seen": 5068608,
      "step": 24020
    },
    {
      "epoch": 2.643014301430143,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.029905669327888657,
      "loss": 0.2314,
      "num_input_tokens_seen": 5069632,
      "step": 24025
    },
    {
      "epoch": 2.6435643564356437,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.02990550801402892,
      "loss": 0.2293,
      "num_input_tokens_seen": 5070720,
      "step": 24030
    },
    {
      "epoch": 2.6441144114411443,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.02990534656279227,
      "loss": 0.2304,
      "num_input_tokens_seen": 5071776,
      "step": 24035
    },
    {
      "epoch": 2.6446644664466445,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.029905184974180196,
      "loss": 0.2319,
      "num_input_tokens_seen": 5072832,
      "step": 24040
    },
    {
      "epoch": 2.645214521452145,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02990502324819418,
      "loss": 0.2308,
      "num_input_tokens_seen": 5073952,
      "step": 24045
    },
    {
      "epoch": 2.645764576457646,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.029904861384835717,
      "loss": 0.2314,
      "num_input_tokens_seen": 5075104,
      "step": 24050
    },
    {
      "epoch": 2.646314631463146,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.0299046993841063,
      "loss": 0.2303,
      "num_input_tokens_seen": 5076224,
      "step": 24055
    },
    {
      "epoch": 2.6468646864686467,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.02990453724600742,
      "loss": 0.2298,
      "num_input_tokens_seen": 5077280,
      "step": 24060
    },
    {
      "epoch": 2.6474147414741473,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029904374970540572,
      "loss": 0.2314,
      "num_input_tokens_seen": 5078336,
      "step": 24065
    },
    {
      "epoch": 2.647964796479648,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.029904212557707255,
      "loss": 0.2309,
      "num_input_tokens_seen": 5079392,
      "step": 24070
    },
    {
      "epoch": 2.6485148514851486,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.02990405000750896,
      "loss": 0.2351,
      "num_input_tokens_seen": 5080480,
      "step": 24075
    },
    {
      "epoch": 2.649064906490649,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.029903887319947187,
      "loss": 0.2324,
      "num_input_tokens_seen": 5081472,
      "step": 24080
    },
    {
      "epoch": 2.6496149614961495,
      "grad_norm": 0.01263427734375,
      "learning_rate": 0.02990372449502344,
      "loss": 0.2319,
      "num_input_tokens_seen": 5082528,
      "step": 24085
    },
    {
      "epoch": 2.65016501650165,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.02990356153273921,
      "loss": 0.2308,
      "num_input_tokens_seen": 5083584,
      "step": 24090
    },
    {
      "epoch": 2.650715071507151,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.029903398433096005,
      "loss": 0.2319,
      "num_input_tokens_seen": 5084608,
      "step": 24095
    },
    {
      "epoch": 2.6512651265126514,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.02990323519609533,
      "loss": 0.2345,
      "num_input_tokens_seen": 5085664,
      "step": 24100
    },
    {
      "epoch": 2.6518151815181517,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029903071821738694,
      "loss": 0.2329,
      "num_input_tokens_seen": 5086720,
      "step": 24105
    },
    {
      "epoch": 2.6523652365236523,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.02990290831002759,
      "loss": 0.2303,
      "num_input_tokens_seen": 5087744,
      "step": 24110
    },
    {
      "epoch": 2.652915291529153,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.02990274466096353,
      "loss": 0.2298,
      "num_input_tokens_seen": 5088768,
      "step": 24115
    },
    {
      "epoch": 2.6534653465346536,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.029902580874548026,
      "loss": 0.2319,
      "num_input_tokens_seen": 5089824,
      "step": 24120
    },
    {
      "epoch": 2.6540154015401543,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.029902416950782582,
      "loss": 0.2324,
      "num_input_tokens_seen": 5090880,
      "step": 24125
    },
    {
      "epoch": 2.6545654565456545,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.029902252889668714,
      "loss": 0.2309,
      "num_input_tokens_seen": 5091904,
      "step": 24130
    },
    {
      "epoch": 2.655115511551155,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.029902088691207935,
      "loss": 0.2319,
      "num_input_tokens_seen": 5092928,
      "step": 24135
    },
    {
      "epoch": 2.6556655665566558,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.02990192435540175,
      "loss": 0.2308,
      "num_input_tokens_seen": 5093952,
      "step": 24140
    },
    {
      "epoch": 2.656215621562156,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.029901759882251678,
      "loss": 0.2309,
      "num_input_tokens_seen": 5094912,
      "step": 24145
    },
    {
      "epoch": 2.6567656765676566,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029901595271759243,
      "loss": 0.2335,
      "num_input_tokens_seen": 5095904,
      "step": 24150
    },
    {
      "epoch": 2.6573157315731573,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.029901430523925947,
      "loss": 0.2314,
      "num_input_tokens_seen": 5096928,
      "step": 24155
    },
    {
      "epoch": 2.657865786578658,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.02990126563875332,
      "loss": 0.2319,
      "num_input_tokens_seen": 5098048,
      "step": 24160
    },
    {
      "epoch": 2.6584158415841586,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02990110061624288,
      "loss": 0.2335,
      "num_input_tokens_seen": 5099136,
      "step": 24165
    },
    {
      "epoch": 2.6589658965896588,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.029900935456396143,
      "loss": 0.2324,
      "num_input_tokens_seen": 5100192,
      "step": 24170
    },
    {
      "epoch": 2.6595159515951594,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.02990077015921463,
      "loss": 0.2293,
      "num_input_tokens_seen": 5101312,
      "step": 24175
    },
    {
      "epoch": 2.66006600660066,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.029900604724699877,
      "loss": 0.2314,
      "num_input_tokens_seen": 5102336,
      "step": 24180
    },
    {
      "epoch": 2.6606160616061607,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029900439152853394,
      "loss": 0.2293,
      "num_input_tokens_seen": 5103424,
      "step": 24185
    },
    {
      "epoch": 2.6611661166116614,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02990027344367672,
      "loss": 0.2319,
      "num_input_tokens_seen": 5104512,
      "step": 24190
    },
    {
      "epoch": 2.6617161716171616,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.02990010759717137,
      "loss": 0.2351,
      "num_input_tokens_seen": 5105536,
      "step": 24195
    },
    {
      "epoch": 2.662266226622662,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.02989994161333888,
      "loss": 0.2315,
      "num_input_tokens_seen": 5106560,
      "step": 24200
    },
    {
      "epoch": 2.662816281628163,
      "grad_norm": 0.01300048828125,
      "learning_rate": 0.029899775492180775,
      "loss": 0.2309,
      "num_input_tokens_seen": 5107616,
      "step": 24205
    },
    {
      "epoch": 2.6633663366336635,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.029899609233698592,
      "loss": 0.232,
      "num_input_tokens_seen": 5108672,
      "step": 24210
    },
    {
      "epoch": 2.663916391639164,
      "grad_norm": 0.007080078125,
      "learning_rate": 0.029899442837893857,
      "loss": 0.2294,
      "num_input_tokens_seen": 5109760,
      "step": 24215
    },
    {
      "epoch": 2.6644664466446644,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.02989927630476811,
      "loss": 0.2305,
      "num_input_tokens_seen": 5110880,
      "step": 24220
    },
    {
      "epoch": 2.665016501650165,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.029899109634322883,
      "loss": 0.232,
      "num_input_tokens_seen": 5111904,
      "step": 24225
    },
    {
      "epoch": 2.6655665566556657,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029898942826559707,
      "loss": 0.2268,
      "num_input_tokens_seen": 5112960,
      "step": 24230
    },
    {
      "epoch": 2.666116611661166,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.029898775881480127,
      "loss": 0.2316,
      "num_input_tokens_seen": 5114112,
      "step": 24235
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.02989860879908568,
      "loss": 0.2373,
      "num_input_tokens_seen": 5115168,
      "step": 24240
    },
    {
      "epoch": 2.667216721672167,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029898441579377908,
      "loss": 0.2315,
      "num_input_tokens_seen": 5116288,
      "step": 24245
    },
    {
      "epoch": 2.667766776677668,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.029898274222358343,
      "loss": 0.2305,
      "num_input_tokens_seen": 5117312,
      "step": 24250
    },
    {
      "epoch": 2.6683168316831685,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.029898106728028534,
      "loss": 0.231,
      "num_input_tokens_seen": 5118368,
      "step": 24255
    },
    {
      "epoch": 2.6688668866886687,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.02989793909639003,
      "loss": 0.2332,
      "num_input_tokens_seen": 5119488,
      "step": 24260
    },
    {
      "epoch": 2.6694169416941693,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.029897771327444363,
      "loss": 0.2299,
      "num_input_tokens_seen": 5120608,
      "step": 24265
    },
    {
      "epoch": 2.66996699669967,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02989760342119309,
      "loss": 0.2336,
      "num_input_tokens_seen": 5121664,
      "step": 24270
    },
    {
      "epoch": 2.6705170517051706,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.02989743537763775,
      "loss": 0.233,
      "num_input_tokens_seen": 5122656,
      "step": 24275
    },
    {
      "epoch": 2.6710671067106713,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029897267196779907,
      "loss": 0.231,
      "num_input_tokens_seen": 5123712,
      "step": 24280
    },
    {
      "epoch": 2.6716171617161715,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02989709887862109,
      "loss": 0.2346,
      "num_input_tokens_seen": 5124768,
      "step": 24285
    },
    {
      "epoch": 2.672167216721672,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029896930423162865,
      "loss": 0.2299,
      "num_input_tokens_seen": 5125856,
      "step": 24290
    },
    {
      "epoch": 2.6727172717271728,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029896761830406782,
      "loss": 0.2319,
      "num_input_tokens_seen": 5126912,
      "step": 24295
    },
    {
      "epoch": 2.6732673267326734,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.029896593100354394,
      "loss": 0.2314,
      "num_input_tokens_seen": 5127936,
      "step": 24300
    },
    {
      "epoch": 2.673817381738174,
      "grad_norm": 0.00347900390625,
      "learning_rate": 0.02989642423300725,
      "loss": 0.2324,
      "num_input_tokens_seen": 5128992,
      "step": 24305
    },
    {
      "epoch": 2.6743674367436743,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.029896255228366916,
      "loss": 0.2298,
      "num_input_tokens_seen": 5130048,
      "step": 24310
    },
    {
      "epoch": 2.674917491749175,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029896086086434945,
      "loss": 0.2319,
      "num_input_tokens_seen": 5131104,
      "step": 24315
    },
    {
      "epoch": 2.6754675467546756,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.029895916807212895,
      "loss": 0.2319,
      "num_input_tokens_seen": 5132224,
      "step": 24320
    },
    {
      "epoch": 2.676017601760176,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.029895747390702332,
      "loss": 0.2294,
      "num_input_tokens_seen": 5133280,
      "step": 24325
    },
    {
      "epoch": 2.6765676567656764,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.02989557783690481,
      "loss": 0.2304,
      "num_input_tokens_seen": 5134368,
      "step": 24330
    },
    {
      "epoch": 2.677117711771177,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02989540814582189,
      "loss": 0.2351,
      "num_input_tokens_seen": 5135360,
      "step": 24335
    },
    {
      "epoch": 2.6776677667766777,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.029895238317455147,
      "loss": 0.2304,
      "num_input_tokens_seen": 5136480,
      "step": 24340
    },
    {
      "epoch": 2.6782178217821784,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029895068351806137,
      "loss": 0.2293,
      "num_input_tokens_seen": 5137472,
      "step": 24345
    },
    {
      "epoch": 2.6787678767876786,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02989489824887643,
      "loss": 0.2278,
      "num_input_tokens_seen": 5138464,
      "step": 24350
    },
    {
      "epoch": 2.6793179317931792,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.029894728008667593,
      "loss": 0.2309,
      "num_input_tokens_seen": 5139488,
      "step": 24355
    },
    {
      "epoch": 2.67986798679868,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029894557631181193,
      "loss": 0.2278,
      "num_input_tokens_seen": 5140512,
      "step": 24360
    },
    {
      "epoch": 2.6804180418041805,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.029894387116418803,
      "loss": 0.2342,
      "num_input_tokens_seen": 5141568,
      "step": 24365
    },
    {
      "epoch": 2.680968096809681,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029894216464382,
      "loss": 0.2284,
      "num_input_tokens_seen": 5142624,
      "step": 24370
    },
    {
      "epoch": 2.6815181518151814,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.02989404567507235,
      "loss": 0.2336,
      "num_input_tokens_seen": 5143680,
      "step": 24375
    },
    {
      "epoch": 2.682068206820682,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.02989387474849142,
      "loss": 0.23,
      "num_input_tokens_seen": 5144736,
      "step": 24380
    },
    {
      "epoch": 2.6826182618261827,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029893703684640796,
      "loss": 0.2372,
      "num_input_tokens_seen": 5145792,
      "step": 24385
    },
    {
      "epoch": 2.6831683168316833,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029893532483522056,
      "loss": 0.2335,
      "num_input_tokens_seen": 5146816,
      "step": 24390
    },
    {
      "epoch": 2.683718371837184,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029893361145136774,
      "loss": 0.2298,
      "num_input_tokens_seen": 5147840,
      "step": 24395
    },
    {
      "epoch": 2.684268426842684,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029893189669486522,
      "loss": 0.2304,
      "num_input_tokens_seen": 5148896,
      "step": 24400
    },
    {
      "epoch": 2.684818481848185,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.029893018056572893,
      "loss": 0.2329,
      "num_input_tokens_seen": 5149984,
      "step": 24405
    },
    {
      "epoch": 2.6853685368536855,
      "grad_norm": 0.0123291015625,
      "learning_rate": 0.02989284630639746,
      "loss": 0.2324,
      "num_input_tokens_seen": 5150976,
      "step": 24410
    },
    {
      "epoch": 2.6859185918591857,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02989267441896181,
      "loss": 0.2303,
      "num_input_tokens_seen": 5152032,
      "step": 24415
    },
    {
      "epoch": 2.6864686468646863,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.029892502394267526,
      "loss": 0.2329,
      "num_input_tokens_seen": 5153088,
      "step": 24420
    },
    {
      "epoch": 2.687018701870187,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029892330232316196,
      "loss": 0.233,
      "num_input_tokens_seen": 5154144,
      "step": 24425
    },
    {
      "epoch": 2.6875687568756876,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0298921579331094,
      "loss": 0.2324,
      "num_input_tokens_seen": 5155200,
      "step": 24430
    },
    {
      "epoch": 2.6881188118811883,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029891985496648732,
      "loss": 0.2319,
      "num_input_tokens_seen": 5156256,
      "step": 24435
    },
    {
      "epoch": 2.6886688668866885,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.02989181292293578,
      "loss": 0.2335,
      "num_input_tokens_seen": 5157312,
      "step": 24440
    },
    {
      "epoch": 2.689218921892189,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029891640211972135,
      "loss": 0.2319,
      "num_input_tokens_seen": 5158368,
      "step": 24445
    },
    {
      "epoch": 2.68976897689769,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.029891467363759385,
      "loss": 0.2314,
      "num_input_tokens_seen": 5159392,
      "step": 24450
    },
    {
      "epoch": 2.6903190319031904,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029891294378299127,
      "loss": 0.2329,
      "num_input_tokens_seen": 5160448,
      "step": 24455
    },
    {
      "epoch": 2.690869086908691,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.029891121255592956,
      "loss": 0.2293,
      "num_input_tokens_seen": 5161600,
      "step": 24460
    },
    {
      "epoch": 2.6914191419141913,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029890947995642465,
      "loss": 0.2298,
      "num_input_tokens_seen": 5162656,
      "step": 24465
    },
    {
      "epoch": 2.691969196919692,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029890774598449254,
      "loss": 0.2314,
      "num_input_tokens_seen": 5163648,
      "step": 24470
    },
    {
      "epoch": 2.6925192519251926,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029890601064014914,
      "loss": 0.2319,
      "num_input_tokens_seen": 5164704,
      "step": 24475
    },
    {
      "epoch": 2.693069306930693,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.029890427392341053,
      "loss": 0.233,
      "num_input_tokens_seen": 5165728,
      "step": 24480
    },
    {
      "epoch": 2.693619361936194,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.029890253583429265,
      "loss": 0.2324,
      "num_input_tokens_seen": 5166752,
      "step": 24485
    },
    {
      "epoch": 2.694169416941694,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.02989007963728116,
      "loss": 0.2298,
      "num_input_tokens_seen": 5167776,
      "step": 24490
    },
    {
      "epoch": 2.6947194719471947,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.029889905553898328,
      "loss": 0.2298,
      "num_input_tokens_seen": 5168832,
      "step": 24495
    },
    {
      "epoch": 2.6952695269526954,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.029889731333282384,
      "loss": 0.234,
      "num_input_tokens_seen": 5169824,
      "step": 24500
    },
    {
      "epoch": 2.6958195819581956,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029889556975434935,
      "loss": 0.2308,
      "num_input_tokens_seen": 5170912,
      "step": 24505
    },
    {
      "epoch": 2.6963696369636962,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029889382480357583,
      "loss": 0.2324,
      "num_input_tokens_seen": 5171936,
      "step": 24510
    },
    {
      "epoch": 2.696919691969197,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029889207848051935,
      "loss": 0.2324,
      "num_input_tokens_seen": 5173024,
      "step": 24515
    },
    {
      "epoch": 2.6974697469746975,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.029889033078519603,
      "loss": 0.2313,
      "num_input_tokens_seen": 5174080,
      "step": 24520
    },
    {
      "epoch": 2.698019801980198,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.029888858171762196,
      "loss": 0.2319,
      "num_input_tokens_seen": 5175072,
      "step": 24525
    },
    {
      "epoch": 2.6985698569856984,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029888683127781333,
      "loss": 0.2324,
      "num_input_tokens_seen": 5176128,
      "step": 24530
    },
    {
      "epoch": 2.699119911991199,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.029888507946578616,
      "loss": 0.2334,
      "num_input_tokens_seen": 5177184,
      "step": 24535
    },
    {
      "epoch": 2.6996699669966997,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.02988833262815567,
      "loss": 0.2313,
      "num_input_tokens_seen": 5178240,
      "step": 24540
    },
    {
      "epoch": 2.7002200220022003,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.029888157172514106,
      "loss": 0.2319,
      "num_input_tokens_seen": 5179296,
      "step": 24545
    },
    {
      "epoch": 2.700770077007701,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.02988798157965554,
      "loss": 0.2319,
      "num_input_tokens_seen": 5180352,
      "step": 24550
    },
    {
      "epoch": 2.701320132013201,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.02988780584958159,
      "loss": 0.2304,
      "num_input_tokens_seen": 5181344,
      "step": 24555
    },
    {
      "epoch": 2.701870187018702,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.02988762998229388,
      "loss": 0.2315,
      "num_input_tokens_seen": 5182432,
      "step": 24560
    },
    {
      "epoch": 2.7024202420242025,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029887453977794028,
      "loss": 0.233,
      "num_input_tokens_seen": 5183456,
      "step": 24565
    },
    {
      "epoch": 2.7029702970297027,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029887277836083654,
      "loss": 0.2341,
      "num_input_tokens_seen": 5184512,
      "step": 24570
    },
    {
      "epoch": 2.7035203520352034,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.029887101557164383,
      "loss": 0.233,
      "num_input_tokens_seen": 5185664,
      "step": 24575
    },
    {
      "epoch": 2.704070407040704,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029886925141037846,
      "loss": 0.2308,
      "num_input_tokens_seen": 5186752,
      "step": 24580
    },
    {
      "epoch": 2.7046204620462047,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.029886748587705663,
      "loss": 0.2319,
      "num_input_tokens_seen": 5187776,
      "step": 24585
    },
    {
      "epoch": 2.7051705170517053,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029886571897169454,
      "loss": 0.2283,
      "num_input_tokens_seen": 5188864,
      "step": 24590
    },
    {
      "epoch": 2.7057205720572055,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.02988639506943086,
      "loss": 0.2289,
      "num_input_tokens_seen": 5189952,
      "step": 24595
    },
    {
      "epoch": 2.706270627062706,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.029886218104491505,
      "loss": 0.2312,
      "num_input_tokens_seen": 5191008,
      "step": 24600
    },
    {
      "epoch": 2.706820682068207,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.029886041002353023,
      "loss": 0.2261,
      "num_input_tokens_seen": 5192064,
      "step": 24605
    },
    {
      "epoch": 2.7073707370737075,
      "grad_norm": 0.007659912109375,
      "learning_rate": 0.029885863763017044,
      "loss": 0.2367,
      "num_input_tokens_seen": 5193152,
      "step": 24610
    },
    {
      "epoch": 2.707920792079208,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0298856863864852,
      "loss": 0.2263,
      "num_input_tokens_seen": 5194272,
      "step": 24615
    },
    {
      "epoch": 2.7084708470847083,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.02988550887275913,
      "loss": 0.2352,
      "num_input_tokens_seen": 5195328,
      "step": 24620
    },
    {
      "epoch": 2.709020902090209,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029885331221840464,
      "loss": 0.2283,
      "num_input_tokens_seen": 5196320,
      "step": 24625
    },
    {
      "epoch": 2.7095709570957096,
      "grad_norm": 0.00775146484375,
      "learning_rate": 0.02988515343373084,
      "loss": 0.2388,
      "num_input_tokens_seen": 5197344,
      "step": 24630
    },
    {
      "epoch": 2.7101210121012103,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.029884975508431905,
      "loss": 0.234,
      "num_input_tokens_seen": 5198336,
      "step": 24635
    },
    {
      "epoch": 2.710671067106711,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.029884797445945292,
      "loss": 0.2309,
      "num_input_tokens_seen": 5199424,
      "step": 24640
    },
    {
      "epoch": 2.711221122112211,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029884619246272646,
      "loss": 0.2339,
      "num_input_tokens_seen": 5200512,
      "step": 24645
    },
    {
      "epoch": 2.7117711771177118,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.029884440909415604,
      "loss": 0.2423,
      "num_input_tokens_seen": 5201536,
      "step": 24650
    },
    {
      "epoch": 2.7123212321232124,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.02988426243537581,
      "loss": 0.2269,
      "num_input_tokens_seen": 5202560,
      "step": 24655
    },
    {
      "epoch": 2.7128712871287126,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029884083824154917,
      "loss": 0.231,
      "num_input_tokens_seen": 5203552,
      "step": 24660
    },
    {
      "epoch": 2.7134213421342133,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02988390507575456,
      "loss": 0.2341,
      "num_input_tokens_seen": 5204608,
      "step": 24665
    },
    {
      "epoch": 2.713971397139714,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.02988372619017639,
      "loss": 0.2336,
      "num_input_tokens_seen": 5205632,
      "step": 24670
    },
    {
      "epoch": 2.7145214521452146,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029883547167422062,
      "loss": 0.2319,
      "num_input_tokens_seen": 5206688,
      "step": 24675
    },
    {
      "epoch": 2.715071507150715,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.029883368007493225,
      "loss": 0.2309,
      "num_input_tokens_seen": 5207680,
      "step": 24680
    },
    {
      "epoch": 2.7156215621562154,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.02988318871039152,
      "loss": 0.2304,
      "num_input_tokens_seen": 5208640,
      "step": 24685
    },
    {
      "epoch": 2.716171617161716,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029883009276118613,
      "loss": 0.2293,
      "num_input_tokens_seen": 5209696,
      "step": 24690
    },
    {
      "epoch": 2.7167216721672167,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029882829704676143,
      "loss": 0.2314,
      "num_input_tokens_seen": 5210816,
      "step": 24695
    },
    {
      "epoch": 2.7172717271727174,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.029882649996065776,
      "loss": 0.2324,
      "num_input_tokens_seen": 5211904,
      "step": 24700
    },
    {
      "epoch": 2.717821782178218,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.029882470150289168,
      "loss": 0.233,
      "num_input_tokens_seen": 5212960,
      "step": 24705
    },
    {
      "epoch": 2.718371837183718,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029882290167347975,
      "loss": 0.2304,
      "num_input_tokens_seen": 5214112,
      "step": 24710
    },
    {
      "epoch": 2.718921892189219,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02988211004724385,
      "loss": 0.231,
      "num_input_tokens_seen": 5215168,
      "step": 24715
    },
    {
      "epoch": 2.7194719471947195,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.029881929789978463,
      "loss": 0.2248,
      "num_input_tokens_seen": 5216288,
      "step": 24720
    },
    {
      "epoch": 2.72002200220022,
      "grad_norm": 0.01263427734375,
      "learning_rate": 0.029881749395553464,
      "loss": 0.2316,
      "num_input_tokens_seen": 5217344,
      "step": 24725
    },
    {
      "epoch": 2.720572057205721,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029881568863970528,
      "loss": 0.2371,
      "num_input_tokens_seen": 5218368,
      "step": 24730
    },
    {
      "epoch": 2.721122112211221,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.02988138819523131,
      "loss": 0.2369,
      "num_input_tokens_seen": 5219520,
      "step": 24735
    },
    {
      "epoch": 2.7216721672167217,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029881207389337475,
      "loss": 0.2354,
      "num_input_tokens_seen": 5220544,
      "step": 24740
    },
    {
      "epoch": 2.7222222222222223,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029881026446290697,
      "loss": 0.2342,
      "num_input_tokens_seen": 5221632,
      "step": 24745
    },
    {
      "epoch": 2.7227722772277225,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029880845366092635,
      "loss": 0.2268,
      "num_input_tokens_seen": 5222688,
      "step": 24750
    },
    {
      "epoch": 2.723322332233223,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029880664148744965,
      "loss": 0.2306,
      "num_input_tokens_seen": 5223680,
      "step": 24755
    },
    {
      "epoch": 2.723872387238724,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02988048279424935,
      "loss": 0.2289,
      "num_input_tokens_seen": 5224736,
      "step": 24760
    },
    {
      "epoch": 2.7244224422442245,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.02988030130260747,
      "loss": 0.2321,
      "num_input_tokens_seen": 5225792,
      "step": 24765
    },
    {
      "epoch": 2.724972497249725,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029880119673820992,
      "loss": 0.2312,
      "num_input_tokens_seen": 5226816,
      "step": 24770
    },
    {
      "epoch": 2.7255225522552253,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02987993790789159,
      "loss": 0.2342,
      "num_input_tokens_seen": 5227872,
      "step": 24775
    },
    {
      "epoch": 2.726072607260726,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.02987975600482094,
      "loss": 0.2322,
      "num_input_tokens_seen": 5228928,
      "step": 24780
    },
    {
      "epoch": 2.7266226622662266,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029879573964610717,
      "loss": 0.23,
      "num_input_tokens_seen": 5229984,
      "step": 24785
    },
    {
      "epoch": 2.7271727172717273,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029879391787262604,
      "loss": 0.2337,
      "num_input_tokens_seen": 5231040,
      "step": 24790
    },
    {
      "epoch": 2.727722772277228,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.029879209472778275,
      "loss": 0.2321,
      "num_input_tokens_seen": 5232096,
      "step": 24795
    },
    {
      "epoch": 2.728272827282728,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.029879027021159417,
      "loss": 0.2336,
      "num_input_tokens_seen": 5233152,
      "step": 24800
    },
    {
      "epoch": 2.728822882288229,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.0298788444324077,
      "loss": 0.2325,
      "num_input_tokens_seen": 5234176,
      "step": 24805
    },
    {
      "epoch": 2.7293729372937294,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029878661706524815,
      "loss": 0.2299,
      "num_input_tokens_seen": 5235232,
      "step": 24810
    },
    {
      "epoch": 2.72992299229923,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.029878478843512447,
      "loss": 0.2315,
      "num_input_tokens_seen": 5236256,
      "step": 24815
    },
    {
      "epoch": 2.7304730473047307,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.029878295843372277,
      "loss": 0.2335,
      "num_input_tokens_seen": 5237248,
      "step": 24820
    },
    {
      "epoch": 2.731023102310231,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029878112706105995,
      "loss": 0.2325,
      "num_input_tokens_seen": 5238336,
      "step": 24825
    },
    {
      "epoch": 2.7315731573157316,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029877929431715283,
      "loss": 0.2324,
      "num_input_tokens_seen": 5239328,
      "step": 24830
    },
    {
      "epoch": 2.7321232123212322,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02987774602020184,
      "loss": 0.2278,
      "num_input_tokens_seen": 5240384,
      "step": 24835
    },
    {
      "epoch": 2.7326732673267324,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.029877562471567348,
      "loss": 0.2335,
      "num_input_tokens_seen": 5241440,
      "step": 24840
    },
    {
      "epoch": 2.733223322332233,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.0298773787858135,
      "loss": 0.2315,
      "num_input_tokens_seen": 5242496,
      "step": 24845
    },
    {
      "epoch": 2.7337733773377337,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.029877194962941996,
      "loss": 0.2314,
      "num_input_tokens_seen": 5243616,
      "step": 24850
    },
    {
      "epoch": 2.7343234323432344,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.02987701100295452,
      "loss": 0.2335,
      "num_input_tokens_seen": 5244704,
      "step": 24855
    },
    {
      "epoch": 2.734873487348735,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.029876826905852777,
      "loss": 0.2325,
      "num_input_tokens_seen": 5245760,
      "step": 24860
    },
    {
      "epoch": 2.7354235423542352,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02987664267163845,
      "loss": 0.2288,
      "num_input_tokens_seen": 5246816,
      "step": 24865
    },
    {
      "epoch": 2.735973597359736,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02987645830031326,
      "loss": 0.2288,
      "num_input_tokens_seen": 5247840,
      "step": 24870
    },
    {
      "epoch": 2.7365236523652365,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.02987627379187888,
      "loss": 0.232,
      "num_input_tokens_seen": 5248928,
      "step": 24875
    },
    {
      "epoch": 2.737073707370737,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.029876089146337027,
      "loss": 0.234,
      "num_input_tokens_seen": 5249984,
      "step": 24880
    },
    {
      "epoch": 2.737623762376238,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029875904363689403,
      "loss": 0.2294,
      "num_input_tokens_seen": 5251072,
      "step": 24885
    },
    {
      "epoch": 2.738173817381738,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029875719443937698,
      "loss": 0.2304,
      "num_input_tokens_seen": 5252096,
      "step": 24890
    },
    {
      "epoch": 2.7387238723872387,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.02987553438708363,
      "loss": 0.2309,
      "num_input_tokens_seen": 5253120,
      "step": 24895
    },
    {
      "epoch": 2.7392739273927393,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029875349193128897,
      "loss": 0.2309,
      "num_input_tokens_seen": 5254144,
      "step": 24900
    },
    {
      "epoch": 2.73982398239824,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.029875163862075213,
      "loss": 0.2346,
      "num_input_tokens_seen": 5255200,
      "step": 24905
    },
    {
      "epoch": 2.7403740374037406,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.02987497839392428,
      "loss": 0.234,
      "num_input_tokens_seen": 5256192,
      "step": 24910
    },
    {
      "epoch": 2.740924092409241,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.029874792788677804,
      "loss": 0.2308,
      "num_input_tokens_seen": 5257280,
      "step": 24915
    },
    {
      "epoch": 2.7414741474147415,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.029874607046337504,
      "loss": 0.2335,
      "num_input_tokens_seen": 5258272,
      "step": 24920
    },
    {
      "epoch": 2.742024202420242,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.029874421166905092,
      "loss": 0.2319,
      "num_input_tokens_seen": 5259328,
      "step": 24925
    },
    {
      "epoch": 2.7425742574257423,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02987423515038227,
      "loss": 0.2314,
      "num_input_tokens_seen": 5260352,
      "step": 24930
    },
    {
      "epoch": 2.743124312431243,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029874048996770767,
      "loss": 0.2299,
      "num_input_tokens_seen": 5261408,
      "step": 24935
    },
    {
      "epoch": 2.7436743674367436,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.029873862706072285,
      "loss": 0.2315,
      "num_input_tokens_seen": 5262400,
      "step": 24940
    },
    {
      "epoch": 2.7442244224422443,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.029873676278288557,
      "loss": 0.232,
      "num_input_tokens_seen": 5263488,
      "step": 24945
    },
    {
      "epoch": 2.744774477447745,
      "grad_norm": 0.007598876953125,
      "learning_rate": 0.02987348971342128,
      "loss": 0.2325,
      "num_input_tokens_seen": 5264512,
      "step": 24950
    },
    {
      "epoch": 2.745324532453245,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.02987330301147219,
      "loss": 0.2315,
      "num_input_tokens_seen": 5265536,
      "step": 24955
    },
    {
      "epoch": 2.745874587458746,
      "grad_norm": 0.01434326171875,
      "learning_rate": 0.02987311617244301,
      "loss": 0.2336,
      "num_input_tokens_seen": 5266624,
      "step": 24960
    },
    {
      "epoch": 2.7464246424642464,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.029872929196335447,
      "loss": 0.232,
      "num_input_tokens_seen": 5267680,
      "step": 24965
    },
    {
      "epoch": 2.746974697469747,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.029872742083151233,
      "loss": 0.234,
      "num_input_tokens_seen": 5268704,
      "step": 24970
    },
    {
      "epoch": 2.7475247524752477,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.029872554832892098,
      "loss": 0.2298,
      "num_input_tokens_seen": 5269728,
      "step": 24975
    },
    {
      "epoch": 2.748074807480748,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029872367445559756,
      "loss": 0.2309,
      "num_input_tokens_seen": 5270784,
      "step": 24980
    },
    {
      "epoch": 2.7486248624862486,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029872179921155942,
      "loss": 0.2309,
      "num_input_tokens_seen": 5271840,
      "step": 24985
    },
    {
      "epoch": 2.7491749174917492,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.02987199225968238,
      "loss": 0.2293,
      "num_input_tokens_seen": 5272928,
      "step": 24990
    },
    {
      "epoch": 2.7497249724972495,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0298718044611408,
      "loss": 0.2304,
      "num_input_tokens_seen": 5273984,
      "step": 24995
    },
    {
      "epoch": 2.7502750275027505,
      "grad_norm": 0.01409912109375,
      "learning_rate": 0.02987161652553294,
      "loss": 0.2315,
      "num_input_tokens_seen": 5274976,
      "step": 25000
    },
    {
      "epoch": 2.7508250825082508,
      "grad_norm": 0.007476806640625,
      "learning_rate": 0.029871428452860522,
      "loss": 0.2315,
      "num_input_tokens_seen": 5276000,
      "step": 25005
    },
    {
      "epoch": 2.7513751375137514,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.029871240243125288,
      "loss": 0.2315,
      "num_input_tokens_seen": 5277088,
      "step": 25010
    },
    {
      "epoch": 2.751925192519252,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.029871051896328965,
      "loss": 0.2325,
      "num_input_tokens_seen": 5278112,
      "step": 25015
    },
    {
      "epoch": 2.7524752475247523,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029870863412473293,
      "loss": 0.232,
      "num_input_tokens_seen": 5279104,
      "step": 25020
    },
    {
      "epoch": 2.753025302530253,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.02987067479156001,
      "loss": 0.2283,
      "num_input_tokens_seen": 5280224,
      "step": 25025
    },
    {
      "epoch": 2.7535753575357536,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.029870486033590854,
      "loss": 0.2294,
      "num_input_tokens_seen": 5281248,
      "step": 25030
    },
    {
      "epoch": 2.754125412541254,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.02987029713856756,
      "loss": 0.2315,
      "num_input_tokens_seen": 5282336,
      "step": 25035
    },
    {
      "epoch": 2.754675467546755,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029870108106491874,
      "loss": 0.232,
      "num_input_tokens_seen": 5283392,
      "step": 25040
    },
    {
      "epoch": 2.755225522552255,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.02986991893736554,
      "loss": 0.2325,
      "num_input_tokens_seen": 5284384,
      "step": 25045
    },
    {
      "epoch": 2.7557755775577557,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.029869729631190296,
      "loss": 0.2346,
      "num_input_tokens_seen": 5285440,
      "step": 25050
    },
    {
      "epoch": 2.7563256325632564,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.02986954018796789,
      "loss": 0.2309,
      "num_input_tokens_seen": 5286560,
      "step": 25055
    },
    {
      "epoch": 2.756875687568757,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.029869350607700067,
      "loss": 0.2356,
      "num_input_tokens_seen": 5287648,
      "step": 25060
    },
    {
      "epoch": 2.7574257425742577,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.02986916089038857,
      "loss": 0.2325,
      "num_input_tokens_seen": 5288704,
      "step": 25065
    },
    {
      "epoch": 2.757975797579758,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02986897103603516,
      "loss": 0.2314,
      "num_input_tokens_seen": 5289728,
      "step": 25070
    },
    {
      "epoch": 2.7585258525852585,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.029868781044641576,
      "loss": 0.2309,
      "num_input_tokens_seen": 5290816,
      "step": 25075
    },
    {
      "epoch": 2.759075907590759,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.02986859091620957,
      "loss": 0.2308,
      "num_input_tokens_seen": 5291936,
      "step": 25080
    },
    {
      "epoch": 2.7596259625962594,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0298684006507409,
      "loss": 0.233,
      "num_input_tokens_seen": 5293024,
      "step": 25085
    },
    {
      "epoch": 2.76017601760176,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02986821024823731,
      "loss": 0.2314,
      "num_input_tokens_seen": 5294048,
      "step": 25090
    },
    {
      "epoch": 2.7607260726072607,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029868019708700563,
      "loss": 0.2314,
      "num_input_tokens_seen": 5295072,
      "step": 25095
    },
    {
      "epoch": 2.7612761276127613,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02986782903213241,
      "loss": 0.2319,
      "num_input_tokens_seen": 5296096,
      "step": 25100
    },
    {
      "epoch": 2.761826182618262,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.029867638218534616,
      "loss": 0.2303,
      "num_input_tokens_seen": 5297248,
      "step": 25105
    },
    {
      "epoch": 2.762376237623762,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.029867447267908933,
      "loss": 0.2335,
      "num_input_tokens_seen": 5298240,
      "step": 25110
    },
    {
      "epoch": 2.762926292629263,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02986725618025712,
      "loss": 0.233,
      "num_input_tokens_seen": 5299296,
      "step": 25115
    },
    {
      "epoch": 2.7634763476347635,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029867064955580946,
      "loss": 0.2308,
      "num_input_tokens_seen": 5300288,
      "step": 25120
    },
    {
      "epoch": 2.764026402640264,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.029866873593882168,
      "loss": 0.2319,
      "num_input_tokens_seen": 5301312,
      "step": 25125
    },
    {
      "epoch": 2.7645764576457648,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.029866682095162547,
      "loss": 0.2314,
      "num_input_tokens_seen": 5302336,
      "step": 25130
    },
    {
      "epoch": 2.765126512651265,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.02986649045942385,
      "loss": 0.2314,
      "num_input_tokens_seen": 5303424,
      "step": 25135
    },
    {
      "epoch": 2.7656765676567656,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029866298686667847,
      "loss": 0.2309,
      "num_input_tokens_seen": 5304448,
      "step": 25140
    },
    {
      "epoch": 2.7662266226622663,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.029866106776896297,
      "loss": 0.2324,
      "num_input_tokens_seen": 5305408,
      "step": 25145
    },
    {
      "epoch": 2.766776677667767,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02986591473011098,
      "loss": 0.2324,
      "num_input_tokens_seen": 5306464,
      "step": 25150
    },
    {
      "epoch": 2.7673267326732676,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.02986572254631366,
      "loss": 0.233,
      "num_input_tokens_seen": 5307552,
      "step": 25155
    },
    {
      "epoch": 2.7678767876787678,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.029865530225506105,
      "loss": 0.2319,
      "num_input_tokens_seen": 5308608,
      "step": 25160
    },
    {
      "epoch": 2.7684268426842684,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02986533776769009,
      "loss": 0.2308,
      "num_input_tokens_seen": 5309600,
      "step": 25165
    },
    {
      "epoch": 2.768976897689769,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.029865145172867393,
      "loss": 0.2314,
      "num_input_tokens_seen": 5310656,
      "step": 25170
    },
    {
      "epoch": 2.7695269526952693,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02986495244103979,
      "loss": 0.2308,
      "num_input_tokens_seen": 5311680,
      "step": 25175
    },
    {
      "epoch": 2.77007700770077,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02986475957220905,
      "loss": 0.2314,
      "num_input_tokens_seen": 5312768,
      "step": 25180
    },
    {
      "epoch": 2.7706270627062706,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02986456656637695,
      "loss": 0.2335,
      "num_input_tokens_seen": 5313760,
      "step": 25185
    },
    {
      "epoch": 2.771177117711771,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.029864373423545275,
      "loss": 0.2314,
      "num_input_tokens_seen": 5314848,
      "step": 25190
    },
    {
      "epoch": 2.771727172717272,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029864180143715804,
      "loss": 0.2325,
      "num_input_tokens_seen": 5315872,
      "step": 25195
    },
    {
      "epoch": 2.772277227722772,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029863986726890317,
      "loss": 0.2304,
      "num_input_tokens_seen": 5316896,
      "step": 25200
    },
    {
      "epoch": 2.7728272827282727,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.0298637931730706,
      "loss": 0.2336,
      "num_input_tokens_seen": 5317984,
      "step": 25205
    },
    {
      "epoch": 2.7733773377337734,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.02986359948225843,
      "loss": 0.2309,
      "num_input_tokens_seen": 5318976,
      "step": 25210
    },
    {
      "epoch": 2.773927392739274,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.029863405654455593,
      "loss": 0.2304,
      "num_input_tokens_seen": 5320064,
      "step": 25215
    },
    {
      "epoch": 2.7744774477447747,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.02986321168966388,
      "loss": 0.2335,
      "num_input_tokens_seen": 5321088,
      "step": 25220
    },
    {
      "epoch": 2.775027502750275,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.02986301758788508,
      "loss": 0.232,
      "num_input_tokens_seen": 5322112,
      "step": 25225
    },
    {
      "epoch": 2.7755775577557755,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.029862823349120975,
      "loss": 0.233,
      "num_input_tokens_seen": 5323104,
      "step": 25230
    },
    {
      "epoch": 2.776127612761276,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.029862628973373363,
      "loss": 0.2325,
      "num_input_tokens_seen": 5324128,
      "step": 25235
    },
    {
      "epoch": 2.776677667766777,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02986243446064403,
      "loss": 0.2303,
      "num_input_tokens_seen": 5325216,
      "step": 25240
    },
    {
      "epoch": 2.7772277227722775,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.029862239810934766,
      "loss": 0.2288,
      "num_input_tokens_seen": 5326304,
      "step": 25245
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029862045024247375,
      "loss": 0.2335,
      "num_input_tokens_seen": 5327456,
      "step": 25250
    },
    {
      "epoch": 2.7783278327832783,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029861850100583644,
      "loss": 0.2309,
      "num_input_tokens_seen": 5328544,
      "step": 25255
    },
    {
      "epoch": 2.778877887788779,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.029861655039945374,
      "loss": 0.2351,
      "num_input_tokens_seen": 5329504,
      "step": 25260
    },
    {
      "epoch": 2.779427942794279,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02986145984233436,
      "loss": 0.2304,
      "num_input_tokens_seen": 5330592,
      "step": 25265
    },
    {
      "epoch": 2.77997799779978,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.0298612645077524,
      "loss": 0.2289,
      "num_input_tokens_seen": 5331712,
      "step": 25270
    },
    {
      "epoch": 2.7805280528052805,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.029861069036201296,
      "loss": 0.2325,
      "num_input_tokens_seen": 5332736,
      "step": 25275
    },
    {
      "epoch": 2.781078107810781,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029860873427682854,
      "loss": 0.2325,
      "num_input_tokens_seen": 5333792,
      "step": 25280
    },
    {
      "epoch": 2.781628162816282,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.02986067768219887,
      "loss": 0.2304,
      "num_input_tokens_seen": 5334912,
      "step": 25285
    },
    {
      "epoch": 2.782178217821782,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.02986048179975115,
      "loss": 0.2298,
      "num_input_tokens_seen": 5336000,
      "step": 25290
    },
    {
      "epoch": 2.7827282728272826,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0298602857803415,
      "loss": 0.2335,
      "num_input_tokens_seen": 5337088,
      "step": 25295
    },
    {
      "epoch": 2.7832783278327833,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029860089623971724,
      "loss": 0.2325,
      "num_input_tokens_seen": 5338080,
      "step": 25300
    },
    {
      "epoch": 2.783828382838284,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.029859893330643636,
      "loss": 0.2309,
      "num_input_tokens_seen": 5339200,
      "step": 25305
    },
    {
      "epoch": 2.7843784378437846,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029859696900359042,
      "loss": 0.2314,
      "num_input_tokens_seen": 5340288,
      "step": 25310
    },
    {
      "epoch": 2.784928492849285,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02985950033311975,
      "loss": 0.2278,
      "num_input_tokens_seen": 5341312,
      "step": 25315
    },
    {
      "epoch": 2.7854785478547854,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.029859303628927573,
      "loss": 0.233,
      "num_input_tokens_seen": 5342368,
      "step": 25320
    },
    {
      "epoch": 2.786028602860286,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02985910678778433,
      "loss": 0.2346,
      "num_input_tokens_seen": 5343456,
      "step": 25325
    },
    {
      "epoch": 2.7865786578657867,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.029858909809691827,
      "loss": 0.2299,
      "num_input_tokens_seen": 5344512,
      "step": 25330
    },
    {
      "epoch": 2.7871287128712874,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029858712694651884,
      "loss": 0.2304,
      "num_input_tokens_seen": 5345600,
      "step": 25335
    },
    {
      "epoch": 2.7876787678767876,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02985851544266631,
      "loss": 0.2351,
      "num_input_tokens_seen": 5346592,
      "step": 25340
    },
    {
      "epoch": 2.7882288228822882,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.029858318053736933,
      "loss": 0.2294,
      "num_input_tokens_seen": 5347680,
      "step": 25345
    },
    {
      "epoch": 2.788778877887789,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02985812052786557,
      "loss": 0.2309,
      "num_input_tokens_seen": 5348832,
      "step": 25350
    },
    {
      "epoch": 2.789328932893289,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029857922865054034,
      "loss": 0.2304,
      "num_input_tokens_seen": 5349888,
      "step": 25355
    },
    {
      "epoch": 2.7898789878987897,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.02985772506530416,
      "loss": 0.2325,
      "num_input_tokens_seen": 5350912,
      "step": 25360
    },
    {
      "epoch": 2.7904290429042904,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029857527128617757,
      "loss": 0.2288,
      "num_input_tokens_seen": 5351968,
      "step": 25365
    },
    {
      "epoch": 2.790979097909791,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.02985732905499666,
      "loss": 0.2283,
      "num_input_tokens_seen": 5353088,
      "step": 25370
    },
    {
      "epoch": 2.7915291529152917,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029857130844442685,
      "loss": 0.2335,
      "num_input_tokens_seen": 5354176,
      "step": 25375
    },
    {
      "epoch": 2.792079207920792,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02985693249695767,
      "loss": 0.2325,
      "num_input_tokens_seen": 5355264,
      "step": 25380
    },
    {
      "epoch": 2.7926292629262925,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029856734012543434,
      "loss": 0.2293,
      "num_input_tokens_seen": 5356288,
      "step": 25385
    },
    {
      "epoch": 2.793179317931793,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.02985653539120181,
      "loss": 0.2346,
      "num_input_tokens_seen": 5357312,
      "step": 25390
    },
    {
      "epoch": 2.793729372937294,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.02985633663293463,
      "loss": 0.2351,
      "num_input_tokens_seen": 5358368,
      "step": 25395
    },
    {
      "epoch": 2.7942794279427945,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.029856137737743718,
      "loss": 0.2309,
      "num_input_tokens_seen": 5359392,
      "step": 25400
    },
    {
      "epoch": 2.7948294829482947,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02985593870563092,
      "loss": 0.2304,
      "num_input_tokens_seen": 5360416,
      "step": 25405
    },
    {
      "epoch": 2.7953795379537953,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.02985573953659806,
      "loss": 0.2314,
      "num_input_tokens_seen": 5361472,
      "step": 25410
    },
    {
      "epoch": 2.795929592959296,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029855540230646977,
      "loss": 0.2309,
      "num_input_tokens_seen": 5362528,
      "step": 25415
    },
    {
      "epoch": 2.7964796479647966,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029855340787779513,
      "loss": 0.2319,
      "num_input_tokens_seen": 5363584,
      "step": 25420
    },
    {
      "epoch": 2.7970297029702973,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.029855141207997496,
      "loss": 0.2314,
      "num_input_tokens_seen": 5364672,
      "step": 25425
    },
    {
      "epoch": 2.7975797579757975,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029854941491302776,
      "loss": 0.2325,
      "num_input_tokens_seen": 5365792,
      "step": 25430
    },
    {
      "epoch": 2.798129812981298,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.029854741637697184,
      "loss": 0.2314,
      "num_input_tokens_seen": 5366848,
      "step": 25435
    },
    {
      "epoch": 2.798679867986799,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.029854541647182566,
      "loss": 0.2299,
      "num_input_tokens_seen": 5367936,
      "step": 25440
    },
    {
      "epoch": 2.799229922992299,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029854341519760765,
      "loss": 0.232,
      "num_input_tokens_seen": 5369056,
      "step": 25445
    },
    {
      "epoch": 2.7997799779977997,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029854141255433633,
      "loss": 0.2289,
      "num_input_tokens_seen": 5370144,
      "step": 25450
    },
    {
      "epoch": 2.8003300330033003,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029853940854203002,
      "loss": 0.2289,
      "num_input_tokens_seen": 5371264,
      "step": 25455
    },
    {
      "epoch": 2.800880088008801,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02985374031607073,
      "loss": 0.2306,
      "num_input_tokens_seen": 5372352,
      "step": 25460
    },
    {
      "epoch": 2.8014301430143016,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029853539641038658,
      "loss": 0.2301,
      "num_input_tokens_seen": 5373408,
      "step": 25465
    },
    {
      "epoch": 2.801980198019802,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.029853338829108636,
      "loss": 0.2344,
      "num_input_tokens_seen": 5374496,
      "step": 25470
    },
    {
      "epoch": 2.8025302530253025,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.02985313788028252,
      "loss": 0.2322,
      "num_input_tokens_seen": 5375552,
      "step": 25475
    },
    {
      "epoch": 2.803080308030803,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.02985293679456216,
      "loss": 0.2264,
      "num_input_tokens_seen": 5376576,
      "step": 25480
    },
    {
      "epoch": 2.8036303630363038,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.029852735571949406,
      "loss": 0.237,
      "num_input_tokens_seen": 5377600,
      "step": 25485
    },
    {
      "epoch": 2.8041804180418044,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02985253421244612,
      "loss": 0.2297,
      "num_input_tokens_seen": 5378688,
      "step": 25490
    },
    {
      "epoch": 2.8047304730473046,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.02985233271605415,
      "loss": 0.2332,
      "num_input_tokens_seen": 5379776,
      "step": 25495
    },
    {
      "epoch": 2.8052805280528053,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.029852131082775357,
      "loss": 0.2316,
      "num_input_tokens_seen": 5380800,
      "step": 25500
    },
    {
      "epoch": 2.805830583058306,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.0298519293126116,
      "loss": 0.2274,
      "num_input_tokens_seen": 5381856,
      "step": 25505
    },
    {
      "epoch": 2.806380638063806,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.029851727405564733,
      "loss": 0.2369,
      "num_input_tokens_seen": 5382880,
      "step": 25510
    },
    {
      "epoch": 2.806930693069307,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.029851525361636624,
      "loss": 0.2328,
      "num_input_tokens_seen": 5384000,
      "step": 25515
    },
    {
      "epoch": 2.8074807480748074,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029851323180829132,
      "loss": 0.23,
      "num_input_tokens_seen": 5385024,
      "step": 25520
    },
    {
      "epoch": 2.808030803080308,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.02985112086314412,
      "loss": 0.2311,
      "num_input_tokens_seen": 5386112,
      "step": 25525
    },
    {
      "epoch": 2.8085808580858087,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029850918408583452,
      "loss": 0.2315,
      "num_input_tokens_seen": 5387200,
      "step": 25530
    },
    {
      "epoch": 2.809130913091309,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.029850715817148998,
      "loss": 0.2326,
      "num_input_tokens_seen": 5388256,
      "step": 25535
    },
    {
      "epoch": 2.8096809680968096,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.029850513088842623,
      "loss": 0.2331,
      "num_input_tokens_seen": 5389248,
      "step": 25540
    },
    {
      "epoch": 2.81023102310231,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02985031022366619,
      "loss": 0.2289,
      "num_input_tokens_seen": 5390272,
      "step": 25545
    },
    {
      "epoch": 2.810781078107811,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.029850107221621576,
      "loss": 0.2336,
      "num_input_tokens_seen": 5391360,
      "step": 25550
    },
    {
      "epoch": 2.8113311331133115,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02984990408271065,
      "loss": 0.2289,
      "num_input_tokens_seen": 5392384,
      "step": 25555
    },
    {
      "epoch": 2.8118811881188117,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.029849700806935286,
      "loss": 0.2299,
      "num_input_tokens_seen": 5393376,
      "step": 25560
    },
    {
      "epoch": 2.8124312431243124,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.02984949739429735,
      "loss": 0.231,
      "num_input_tokens_seen": 5394400,
      "step": 25565
    },
    {
      "epoch": 2.812981298129813,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.02984929384479873,
      "loss": 0.2336,
      "num_input_tokens_seen": 5395424,
      "step": 25570
    },
    {
      "epoch": 2.8135313531353137,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02984909015844129,
      "loss": 0.2342,
      "num_input_tokens_seen": 5396416,
      "step": 25575
    },
    {
      "epoch": 2.8140814081408143,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029848886335226907,
      "loss": 0.232,
      "num_input_tokens_seen": 5397536,
      "step": 25580
    },
    {
      "epoch": 2.8146314631463145,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02984868237515747,
      "loss": 0.2315,
      "num_input_tokens_seen": 5398592,
      "step": 25585
    },
    {
      "epoch": 2.815181518151815,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02984847827823485,
      "loss": 0.2294,
      "num_input_tokens_seen": 5399616,
      "step": 25590
    },
    {
      "epoch": 2.815731573157316,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02984827404446093,
      "loss": 0.232,
      "num_input_tokens_seen": 5400672,
      "step": 25595
    },
    {
      "epoch": 2.816281628162816,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.029848069673837595,
      "loss": 0.2341,
      "num_input_tokens_seen": 5401632,
      "step": 25600
    },
    {
      "epoch": 2.8168316831683167,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.029847865166366724,
      "loss": 0.2273,
      "num_input_tokens_seen": 5402688,
      "step": 25605
    },
    {
      "epoch": 2.8173817381738173,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029847660522050203,
      "loss": 0.2289,
      "num_input_tokens_seen": 5403712,
      "step": 25610
    },
    {
      "epoch": 2.817931793179318,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029847455740889924,
      "loss": 0.2331,
      "num_input_tokens_seen": 5404768,
      "step": 25615
    },
    {
      "epoch": 2.8184818481848186,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.029847250822887767,
      "loss": 0.2336,
      "num_input_tokens_seen": 5405792,
      "step": 25620
    },
    {
      "epoch": 2.819031903190319,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02984704576804563,
      "loss": 0.2315,
      "num_input_tokens_seen": 5406848,
      "step": 25625
    },
    {
      "epoch": 2.8195819581958195,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02984684057636539,
      "loss": 0.2331,
      "num_input_tokens_seen": 5407936,
      "step": 25630
    },
    {
      "epoch": 2.82013201320132,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029846635247848946,
      "loss": 0.2335,
      "num_input_tokens_seen": 5409024,
      "step": 25635
    },
    {
      "epoch": 2.8206820682068208,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029846429782498185,
      "loss": 0.232,
      "num_input_tokens_seen": 5410048,
      "step": 25640
    },
    {
      "epoch": 2.8212321232123214,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.02984622418031501,
      "loss": 0.2335,
      "num_input_tokens_seen": 5411072,
      "step": 25645
    },
    {
      "epoch": 2.8217821782178216,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.02984601844130131,
      "loss": 0.2309,
      "num_input_tokens_seen": 5412064,
      "step": 25650
    },
    {
      "epoch": 2.8223322332233223,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.02984581256545898,
      "loss": 0.2303,
      "num_input_tokens_seen": 5413056,
      "step": 25655
    },
    {
      "epoch": 2.822882288228823,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02984560655278992,
      "loss": 0.2319,
      "num_input_tokens_seen": 5414208,
      "step": 25660
    },
    {
      "epoch": 2.8234323432343236,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.02984540040329603,
      "loss": 0.2293,
      "num_input_tokens_seen": 5415296,
      "step": 25665
    },
    {
      "epoch": 2.823982398239824,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.029845194116979206,
      "loss": 0.2329,
      "num_input_tokens_seen": 5416352,
      "step": 25670
    },
    {
      "epoch": 2.8245324532453244,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02984498769384135,
      "loss": 0.2314,
      "num_input_tokens_seen": 5417408,
      "step": 25675
    },
    {
      "epoch": 2.825082508250825,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029844781133884373,
      "loss": 0.2335,
      "num_input_tokens_seen": 5418496,
      "step": 25680
    },
    {
      "epoch": 2.8256325632563257,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.029844574437110163,
      "loss": 0.2314,
      "num_input_tokens_seen": 5419520,
      "step": 25685
    },
    {
      "epoch": 2.826182618261826,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029844367603520636,
      "loss": 0.2293,
      "num_input_tokens_seen": 5420576,
      "step": 25690
    },
    {
      "epoch": 2.8267326732673266,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029844160633117695,
      "loss": 0.2293,
      "num_input_tokens_seen": 5421632,
      "step": 25695
    },
    {
      "epoch": 2.8272827282728272,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02984395352590325,
      "loss": 0.2335,
      "num_input_tokens_seen": 5422688,
      "step": 25700
    },
    {
      "epoch": 2.827832783278328,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.02984374628187921,
      "loss": 0.2319,
      "num_input_tokens_seen": 5423808,
      "step": 25705
    },
    {
      "epoch": 2.8283828382838285,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.02984353890104748,
      "loss": 0.234,
      "num_input_tokens_seen": 5424832,
      "step": 25710
    },
    {
      "epoch": 2.8289328932893287,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.029843331383409975,
      "loss": 0.2309,
      "num_input_tokens_seen": 5425856,
      "step": 25715
    },
    {
      "epoch": 2.8294829482948294,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.029843123728968607,
      "loss": 0.233,
      "num_input_tokens_seen": 5426912,
      "step": 25720
    },
    {
      "epoch": 2.83003300330033,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02984291593772529,
      "loss": 0.2324,
      "num_input_tokens_seen": 5428000,
      "step": 25725
    },
    {
      "epoch": 2.8305830583058307,
      "grad_norm": 0.000652313232421875,
      "learning_rate": 0.029842708009681942,
      "loss": 0.2303,
      "num_input_tokens_seen": 5428992,
      "step": 25730
    },
    {
      "epoch": 2.8311331133113313,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.029842499944840475,
      "loss": 0.2308,
      "num_input_tokens_seen": 5430048,
      "step": 25735
    },
    {
      "epoch": 2.8316831683168315,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029842291743202805,
      "loss": 0.2308,
      "num_input_tokens_seen": 5431136,
      "step": 25740
    },
    {
      "epoch": 2.832233223322332,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02984208340477086,
      "loss": 0.2335,
      "num_input_tokens_seen": 5432192,
      "step": 25745
    },
    {
      "epoch": 2.832783278327833,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.029841874929546548,
      "loss": 0.2303,
      "num_input_tokens_seen": 5433248,
      "step": 25750
    },
    {
      "epoch": 2.8333333333333335,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0298416663175318,
      "loss": 0.2303,
      "num_input_tokens_seen": 5434304,
      "step": 25755
    },
    {
      "epoch": 2.833883388338834,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02984145756872854,
      "loss": 0.2319,
      "num_input_tokens_seen": 5435360,
      "step": 25760
    },
    {
      "epoch": 2.8344334433443343,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.02984124868313868,
      "loss": 0.2324,
      "num_input_tokens_seen": 5436416,
      "step": 25765
    },
    {
      "epoch": 2.834983498349835,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.029841039660764148,
      "loss": 0.2309,
      "num_input_tokens_seen": 5437408,
      "step": 25770
    },
    {
      "epoch": 2.8355335533553356,
      "grad_norm": 0.01202392578125,
      "learning_rate": 0.029840830501606883,
      "loss": 0.2334,
      "num_input_tokens_seen": 5438496,
      "step": 25775
    },
    {
      "epoch": 2.836083608360836,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.029840621205668803,
      "loss": 0.2313,
      "num_input_tokens_seen": 5439488,
      "step": 25780
    },
    {
      "epoch": 2.8366336633663365,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02984041177295184,
      "loss": 0.2319,
      "num_input_tokens_seen": 5440512,
      "step": 25785
    },
    {
      "epoch": 2.837183718371837,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02984020220345792,
      "loss": 0.2309,
      "num_input_tokens_seen": 5441536,
      "step": 25790
    },
    {
      "epoch": 2.837733773377338,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.029839992497188977,
      "loss": 0.2319,
      "num_input_tokens_seen": 5442560,
      "step": 25795
    },
    {
      "epoch": 2.8382838283828384,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029839782654146947,
      "loss": 0.2319,
      "num_input_tokens_seen": 5443616,
      "step": 25800
    },
    {
      "epoch": 2.8388338833883386,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.029839572674333757,
      "loss": 0.2309,
      "num_input_tokens_seen": 5444672,
      "step": 25805
    },
    {
      "epoch": 2.8393839383938393,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02983936255775135,
      "loss": 0.2303,
      "num_input_tokens_seen": 5445792,
      "step": 25810
    },
    {
      "epoch": 2.83993399339934,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02983915230440166,
      "loss": 0.2303,
      "num_input_tokens_seen": 5446880,
      "step": 25815
    },
    {
      "epoch": 2.8404840484048406,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.02983894191428662,
      "loss": 0.2314,
      "num_input_tokens_seen": 5447936,
      "step": 25820
    },
    {
      "epoch": 2.8410341034103412,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029838731387408175,
      "loss": 0.2309,
      "num_input_tokens_seen": 5448960,
      "step": 25825
    },
    {
      "epoch": 2.8415841584158414,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02983852072376826,
      "loss": 0.2309,
      "num_input_tokens_seen": 5450016,
      "step": 25830
    },
    {
      "epoch": 2.842134213421342,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.029838309923368826,
      "loss": 0.2351,
      "num_input_tokens_seen": 5451104,
      "step": 25835
    },
    {
      "epoch": 2.8426842684268427,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.029838098986211804,
      "loss": 0.2324,
      "num_input_tokens_seen": 5452192,
      "step": 25840
    },
    {
      "epoch": 2.8432343234323434,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029837887912299146,
      "loss": 0.2293,
      "num_input_tokens_seen": 5453216,
      "step": 25845
    },
    {
      "epoch": 2.843784378437844,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.029837676701632794,
      "loss": 0.234,
      "num_input_tokens_seen": 5454240,
      "step": 25850
    },
    {
      "epoch": 2.8443344334433442,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.029837465354214696,
      "loss": 0.2314,
      "num_input_tokens_seen": 5455296,
      "step": 25855
    },
    {
      "epoch": 2.844884488448845,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.029837253870046803,
      "loss": 0.2293,
      "num_input_tokens_seen": 5456352,
      "step": 25860
    },
    {
      "epoch": 2.8454345434543455,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029837042249131054,
      "loss": 0.2329,
      "num_input_tokens_seen": 5457472,
      "step": 25865
    },
    {
      "epoch": 2.8459845984598457,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.029836830491469412,
      "loss": 0.2308,
      "num_input_tokens_seen": 5458496,
      "step": 25870
    },
    {
      "epoch": 2.8465346534653464,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.029836618597063817,
      "loss": 0.2314,
      "num_input_tokens_seen": 5459520,
      "step": 25875
    },
    {
      "epoch": 2.847084708470847,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.02983640656591623,
      "loss": 0.2304,
      "num_input_tokens_seen": 5460608,
      "step": 25880
    },
    {
      "epoch": 2.8476347634763477,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.029836194398028607,
      "loss": 0.234,
      "num_input_tokens_seen": 5461728,
      "step": 25885
    },
    {
      "epoch": 2.8481848184818483,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.029835982093402894,
      "loss": 0.2288,
      "num_input_tokens_seen": 5462720,
      "step": 25890
    },
    {
      "epoch": 2.8487348734873486,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.029835769652041058,
      "loss": 0.2299,
      "num_input_tokens_seen": 5463808,
      "step": 25895
    },
    {
      "epoch": 2.849284928492849,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.029835557073945047,
      "loss": 0.2309,
      "num_input_tokens_seen": 5464832,
      "step": 25900
    },
    {
      "epoch": 2.84983498349835,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029835344359116826,
      "loss": 0.2289,
      "num_input_tokens_seen": 5465888,
      "step": 25905
    },
    {
      "epoch": 2.8503850385038505,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02983513150755836,
      "loss": 0.2341,
      "num_input_tokens_seen": 5466912,
      "step": 25910
    },
    {
      "epoch": 2.850935093509351,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.0298349185192716,
      "loss": 0.2336,
      "num_input_tokens_seen": 5467904,
      "step": 25915
    },
    {
      "epoch": 2.8514851485148514,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.029834705394258514,
      "loss": 0.2357,
      "num_input_tokens_seen": 5468928,
      "step": 25920
    },
    {
      "epoch": 2.852035203520352,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02983449213252107,
      "loss": 0.2283,
      "num_input_tokens_seen": 5469952,
      "step": 25925
    },
    {
      "epoch": 2.8525852585258527,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.02983427873406123,
      "loss": 0.2324,
      "num_input_tokens_seen": 5471008,
      "step": 25930
    },
    {
      "epoch": 2.8531353135313533,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029834065198880957,
      "loss": 0.2283,
      "num_input_tokens_seen": 5472096,
      "step": 25935
    },
    {
      "epoch": 2.853685368536854,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.02983385152698223,
      "loss": 0.2308,
      "num_input_tokens_seen": 5473152,
      "step": 25940
    },
    {
      "epoch": 2.854235423542354,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.029833637718367008,
      "loss": 0.2293,
      "num_input_tokens_seen": 5474240,
      "step": 25945
    },
    {
      "epoch": 2.854785478547855,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029833423773037264,
      "loss": 0.2268,
      "num_input_tokens_seen": 5475360,
      "step": 25950
    },
    {
      "epoch": 2.8553355335533555,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.029833209690994972,
      "loss": 0.2305,
      "num_input_tokens_seen": 5476416,
      "step": 25955
    },
    {
      "epoch": 2.8558855885588557,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0298329954722421,
      "loss": 0.2332,
      "num_input_tokens_seen": 5477472,
      "step": 25960
    },
    {
      "epoch": 2.8564356435643563,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029832781116780633,
      "loss": 0.2249,
      "num_input_tokens_seen": 5478592,
      "step": 25965
    },
    {
      "epoch": 2.856985698569857,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029832566624612537,
      "loss": 0.2355,
      "num_input_tokens_seen": 5479680,
      "step": 25970
    },
    {
      "epoch": 2.8575357535753576,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029832351995739793,
      "loss": 0.2287,
      "num_input_tokens_seen": 5480736,
      "step": 25975
    },
    {
      "epoch": 2.8580858085808583,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.029832137230164375,
      "loss": 0.2313,
      "num_input_tokens_seen": 5481760,
      "step": 25980
    },
    {
      "epoch": 2.8586358635863585,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.02983192232788827,
      "loss": 0.2354,
      "num_input_tokens_seen": 5482784,
      "step": 25985
    },
    {
      "epoch": 2.859185918591859,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.029831707288913448,
      "loss": 0.2313,
      "num_input_tokens_seen": 5483872,
      "step": 25990
    },
    {
      "epoch": 2.8597359735973598,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0298314921132419,
      "loss": 0.2281,
      "num_input_tokens_seen": 5484960,
      "step": 25995
    },
    {
      "epoch": 2.8602860286028604,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029831276800875606,
      "loss": 0.2338,
      "num_input_tokens_seen": 5486016,
      "step": 26000
    },
    {
      "epoch": 2.860836083608361,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.02983106135181655,
      "loss": 0.238,
      "num_input_tokens_seen": 5487072,
      "step": 26005
    },
    {
      "epoch": 2.8613861386138613,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02983084576606672,
      "loss": 0.2326,
      "num_input_tokens_seen": 5488192,
      "step": 26010
    },
    {
      "epoch": 2.861936193619362,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.029830630043628098,
      "loss": 0.23,
      "num_input_tokens_seen": 5489248,
      "step": 26015
    },
    {
      "epoch": 2.8624862486248626,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029830414184502677,
      "loss": 0.2315,
      "num_input_tokens_seen": 5490304,
      "step": 26020
    },
    {
      "epoch": 2.8630363036303628,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.029830198188692442,
      "loss": 0.232,
      "num_input_tokens_seen": 5491360,
      "step": 26025
    },
    {
      "epoch": 2.863586358635864,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.029829982056199394,
      "loss": 0.233,
      "num_input_tokens_seen": 5492416,
      "step": 26030
    },
    {
      "epoch": 2.864136413641364,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.02982976578702551,
      "loss": 0.2294,
      "num_input_tokens_seen": 5493504,
      "step": 26035
    },
    {
      "epoch": 2.8646864686468647,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02982954938117279,
      "loss": 0.2304,
      "num_input_tokens_seen": 5494560,
      "step": 26040
    },
    {
      "epoch": 2.8652365236523654,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.029829332838643235,
      "loss": 0.2351,
      "num_input_tokens_seen": 5495584,
      "step": 26045
    },
    {
      "epoch": 2.8657865786578656,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029829116159438833,
      "loss": 0.2325,
      "num_input_tokens_seen": 5496608,
      "step": 26050
    },
    {
      "epoch": 2.866336633663366,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.02982889934356158,
      "loss": 0.2319,
      "num_input_tokens_seen": 5497632,
      "step": 26055
    },
    {
      "epoch": 2.866886688668867,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.02982868239101348,
      "loss": 0.2324,
      "num_input_tokens_seen": 5498624,
      "step": 26060
    },
    {
      "epoch": 2.8674367436743675,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02982846530179653,
      "loss": 0.2319,
      "num_input_tokens_seen": 5499584,
      "step": 26065
    },
    {
      "epoch": 2.867986798679868,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.02982824807591273,
      "loss": 0.2335,
      "num_input_tokens_seen": 5500704,
      "step": 26070
    },
    {
      "epoch": 2.8685368536853684,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029828030713364085,
      "loss": 0.2319,
      "num_input_tokens_seen": 5501856,
      "step": 26075
    },
    {
      "epoch": 2.869086908690869,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.029827813214152597,
      "loss": 0.234,
      "num_input_tokens_seen": 5502880,
      "step": 26080
    },
    {
      "epoch": 2.8696369636963697,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.029827595578280266,
      "loss": 0.2319,
      "num_input_tokens_seen": 5503936,
      "step": 26085
    },
    {
      "epoch": 2.8701870187018703,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.029827377805749105,
      "loss": 0.2293,
      "num_input_tokens_seen": 5504992,
      "step": 26090
    },
    {
      "epoch": 2.870737073707371,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029827159896561112,
      "loss": 0.2293,
      "num_input_tokens_seen": 5506080,
      "step": 26095
    },
    {
      "epoch": 2.871287128712871,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.029826941850718304,
      "loss": 0.2314,
      "num_input_tokens_seen": 5507168,
      "step": 26100
    },
    {
      "epoch": 2.871837183718372,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029826723668222692,
      "loss": 0.2314,
      "num_input_tokens_seen": 5508224,
      "step": 26105
    },
    {
      "epoch": 2.8723872387238725,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.029826505349076278,
      "loss": 0.2309,
      "num_input_tokens_seen": 5509248,
      "step": 26110
    },
    {
      "epoch": 2.8729372937293727,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029826286893281078,
      "loss": 0.2324,
      "num_input_tokens_seen": 5510304,
      "step": 26115
    },
    {
      "epoch": 2.8734873487348733,
      "grad_norm": 0.001953125,
      "learning_rate": 0.029826068300839104,
      "loss": 0.2314,
      "num_input_tokens_seen": 5511296,
      "step": 26120
    },
    {
      "epoch": 2.874037403740374,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.029825849571752376,
      "loss": 0.2308,
      "num_input_tokens_seen": 5512352,
      "step": 26125
    },
    {
      "epoch": 2.8745874587458746,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.029825630706022906,
      "loss": 0.2288,
      "num_input_tokens_seen": 5513408,
      "step": 26130
    },
    {
      "epoch": 2.8751375137513753,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.029825411703652716,
      "loss": 0.2324,
      "num_input_tokens_seen": 5514464,
      "step": 26135
    },
    {
      "epoch": 2.8756875687568755,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.029825192564643815,
      "loss": 0.2304,
      "num_input_tokens_seen": 5515520,
      "step": 26140
    },
    {
      "epoch": 2.876237623762376,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.02982497328899823,
      "loss": 0.2325,
      "num_input_tokens_seen": 5516608,
      "step": 26145
    },
    {
      "epoch": 2.8767876787678768,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.029824753876717975,
      "loss": 0.234,
      "num_input_tokens_seen": 5517696,
      "step": 26150
    },
    {
      "epoch": 2.8773377337733774,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02982453432780509,
      "loss": 0.2324,
      "num_input_tokens_seen": 5518752,
      "step": 26155
    },
    {
      "epoch": 2.877887788778878,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02982431464226157,
      "loss": 0.2298,
      "num_input_tokens_seen": 5519776,
      "step": 26160
    },
    {
      "epoch": 2.8784378437843783,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029824094820089464,
      "loss": 0.2314,
      "num_input_tokens_seen": 5520832,
      "step": 26165
    },
    {
      "epoch": 2.878987898789879,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02982387486129079,
      "loss": 0.2319,
      "num_input_tokens_seen": 5521888,
      "step": 26170
    },
    {
      "epoch": 2.8795379537953796,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.02982365476586757,
      "loss": 0.2314,
      "num_input_tokens_seen": 5522912,
      "step": 26175
    },
    {
      "epoch": 2.8800880088008802,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02982343453382184,
      "loss": 0.2314,
      "num_input_tokens_seen": 5523904,
      "step": 26180
    },
    {
      "epoch": 2.880638063806381,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.02982321416515563,
      "loss": 0.2314,
      "num_input_tokens_seen": 5524960,
      "step": 26185
    },
    {
      "epoch": 2.881188118811881,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02982299365987096,
      "loss": 0.2319,
      "num_input_tokens_seen": 5525984,
      "step": 26190
    },
    {
      "epoch": 2.8817381738173817,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.029822773017969872,
      "loss": 0.2309,
      "num_input_tokens_seen": 5527072,
      "step": 26195
    },
    {
      "epoch": 2.8822882288228824,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029822552239454403,
      "loss": 0.233,
      "num_input_tokens_seen": 5528064,
      "step": 26200
    },
    {
      "epoch": 2.8828382838283826,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.02982233132432658,
      "loss": 0.2304,
      "num_input_tokens_seen": 5529120,
      "step": 26205
    },
    {
      "epoch": 2.8833883388338832,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029822110272588444,
      "loss": 0.2288,
      "num_input_tokens_seen": 5530208,
      "step": 26210
    },
    {
      "epoch": 2.883938393839384,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.029821889084242027,
      "loss": 0.2345,
      "num_input_tokens_seen": 5531232,
      "step": 26215
    },
    {
      "epoch": 2.8844884488448845,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.029821667759289367,
      "loss": 0.2293,
      "num_input_tokens_seen": 5532352,
      "step": 26220
    },
    {
      "epoch": 2.885038503850385,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029821446297732517,
      "loss": 0.2319,
      "num_input_tokens_seen": 5533376,
      "step": 26225
    },
    {
      "epoch": 2.8855885588558854,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.0298212246995735,
      "loss": 0.2309,
      "num_input_tokens_seen": 5534464,
      "step": 26230
    },
    {
      "epoch": 2.886138613861386,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.02982100296481437,
      "loss": 0.2324,
      "num_input_tokens_seen": 5535552,
      "step": 26235
    },
    {
      "epoch": 2.8866886688668867,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.029820781093457167,
      "loss": 0.2314,
      "num_input_tokens_seen": 5536576,
      "step": 26240
    },
    {
      "epoch": 2.8872387238723873,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.029820559085503934,
      "loss": 0.234,
      "num_input_tokens_seen": 5537632,
      "step": 26245
    },
    {
      "epoch": 2.887788778877888,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02982033694095672,
      "loss": 0.2288,
      "num_input_tokens_seen": 5538688,
      "step": 26250
    },
    {
      "epoch": 2.888338833883388,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.02982011465981758,
      "loss": 0.2289,
      "num_input_tokens_seen": 5539712,
      "step": 26255
    },
    {
      "epoch": 2.888888888888889,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.029819892242088544,
      "loss": 0.2331,
      "num_input_tokens_seen": 5540800,
      "step": 26260
    },
    {
      "epoch": 2.8894389438943895,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.02981966968777168,
      "loss": 0.2305,
      "num_input_tokens_seen": 5541888,
      "step": 26265
    },
    {
      "epoch": 2.88998899889989,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029819446996869028,
      "loss": 0.232,
      "num_input_tokens_seen": 5542976,
      "step": 26270
    },
    {
      "epoch": 2.890539053905391,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02981922416938264,
      "loss": 0.2367,
      "num_input_tokens_seen": 5544032,
      "step": 26275
    },
    {
      "epoch": 2.891089108910891,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02981900120531458,
      "loss": 0.2315,
      "num_input_tokens_seen": 5545056,
      "step": 26280
    },
    {
      "epoch": 2.8916391639163916,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.029818778104666897,
      "loss": 0.2273,
      "num_input_tokens_seen": 5546112,
      "step": 26285
    },
    {
      "epoch": 2.8921892189218923,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.029818554867441647,
      "loss": 0.232,
      "num_input_tokens_seen": 5547200,
      "step": 26290
    },
    {
      "epoch": 2.8927392739273925,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.02981833149364088,
      "loss": 0.2309,
      "num_input_tokens_seen": 5548256,
      "step": 26295
    },
    {
      "epoch": 2.893289328932893,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029818107983266673,
      "loss": 0.231,
      "num_input_tokens_seen": 5549344,
      "step": 26300
    },
    {
      "epoch": 2.893839383938394,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.02981788433632107,
      "loss": 0.2325,
      "num_input_tokens_seen": 5550368,
      "step": 26305
    },
    {
      "epoch": 2.8943894389438944,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02981766055280614,
      "loss": 0.2314,
      "num_input_tokens_seen": 5551360,
      "step": 26310
    },
    {
      "epoch": 2.894939493949395,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.029817436632723935,
      "loss": 0.2335,
      "num_input_tokens_seen": 5552416,
      "step": 26315
    },
    {
      "epoch": 2.8954895489548953,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.029817212576076533,
      "loss": 0.2314,
      "num_input_tokens_seen": 5553472,
      "step": 26320
    },
    {
      "epoch": 2.896039603960396,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02981698838286599,
      "loss": 0.2309,
      "num_input_tokens_seen": 5554464,
      "step": 26325
    },
    {
      "epoch": 2.8965896589658966,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.02981676405309438,
      "loss": 0.2325,
      "num_input_tokens_seen": 5555456,
      "step": 26330
    },
    {
      "epoch": 2.8971397139713972,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.029816539586763762,
      "loss": 0.2346,
      "num_input_tokens_seen": 5556448,
      "step": 26335
    },
    {
      "epoch": 2.897689768976898,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.029816314983876206,
      "loss": 0.2304,
      "num_input_tokens_seen": 5557536,
      "step": 26340
    },
    {
      "epoch": 2.898239823982398,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.029816090244433786,
      "loss": 0.2283,
      "num_input_tokens_seen": 5558592,
      "step": 26345
    },
    {
      "epoch": 2.8987898789878987,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029815865368438572,
      "loss": 0.2304,
      "num_input_tokens_seen": 5559680,
      "step": 26350
    },
    {
      "epoch": 2.8993399339933994,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02981564035589264,
      "loss": 0.2278,
      "num_input_tokens_seen": 5560800,
      "step": 26355
    },
    {
      "epoch": 2.8998899889989,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029815415206798056,
      "loss": 0.2309,
      "num_input_tokens_seen": 5561856,
      "step": 26360
    },
    {
      "epoch": 2.9004400440044007,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0298151899211569,
      "loss": 0.233,
      "num_input_tokens_seen": 5562912,
      "step": 26365
    },
    {
      "epoch": 2.900990099009901,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029814964498971246,
      "loss": 0.2283,
      "num_input_tokens_seen": 5564000,
      "step": 26370
    },
    {
      "epoch": 2.9015401540154016,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.029814738940243175,
      "loss": 0.233,
      "num_input_tokens_seen": 5564992,
      "step": 26375
    },
    {
      "epoch": 2.902090209020902,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029814513244974764,
      "loss": 0.2278,
      "num_input_tokens_seen": 5566048,
      "step": 26380
    },
    {
      "epoch": 2.9026402640264024,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.029814287413168092,
      "loss": 0.231,
      "num_input_tokens_seen": 5567040,
      "step": 26385
    },
    {
      "epoch": 2.903190319031903,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029814061444825242,
      "loss": 0.231,
      "num_input_tokens_seen": 5568160,
      "step": 26390
    },
    {
      "epoch": 2.9037403740374037,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029813835339948296,
      "loss": 0.2278,
      "num_input_tokens_seen": 5569216,
      "step": 26395
    },
    {
      "epoch": 2.9042904290429044,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029813609098539338,
      "loss": 0.2326,
      "num_input_tokens_seen": 5570304,
      "step": 26400
    },
    {
      "epoch": 2.904840484048405,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.02981338272060045,
      "loss": 0.2326,
      "num_input_tokens_seen": 5571392,
      "step": 26405
    },
    {
      "epoch": 2.905390539053905,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02981315620613373,
      "loss": 0.2347,
      "num_input_tokens_seen": 5572416,
      "step": 26410
    },
    {
      "epoch": 2.905940594059406,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.02981292955514125,
      "loss": 0.2336,
      "num_input_tokens_seen": 5573632,
      "step": 26415
    },
    {
      "epoch": 2.9064906490649065,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.029812702767625106,
      "loss": 0.2284,
      "num_input_tokens_seen": 5574688,
      "step": 26420
    },
    {
      "epoch": 2.907040704070407,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029812475843587392,
      "loss": 0.2331,
      "num_input_tokens_seen": 5575744,
      "step": 26425
    },
    {
      "epoch": 2.907590759075908,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029812248783030196,
      "loss": 0.2279,
      "num_input_tokens_seen": 5576832,
      "step": 26430
    },
    {
      "epoch": 2.908140814081408,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.029812021585955608,
      "loss": 0.232,
      "num_input_tokens_seen": 5577888,
      "step": 26435
    },
    {
      "epoch": 2.9086908690869087,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029811794252365728,
      "loss": 0.23,
      "num_input_tokens_seen": 5578944,
      "step": 26440
    },
    {
      "epoch": 2.9092409240924093,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.029811566782262645,
      "loss": 0.2279,
      "num_input_tokens_seen": 5580032,
      "step": 26445
    },
    {
      "epoch": 2.9097909790979095,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02981133917564846,
      "loss": 0.2306,
      "num_input_tokens_seen": 5581056,
      "step": 26450
    },
    {
      "epoch": 2.9103410341034106,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.029811111432525267,
      "loss": 0.2311,
      "num_input_tokens_seen": 5582208,
      "step": 26455
    },
    {
      "epoch": 2.910891089108911,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.029810883552895167,
      "loss": 0.2406,
      "num_input_tokens_seen": 5583264,
      "step": 26460
    },
    {
      "epoch": 2.9114411441144115,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02981065553676026,
      "loss": 0.226,
      "num_input_tokens_seen": 5584288,
      "step": 26465
    },
    {
      "epoch": 2.911991199119912,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02981042738412265,
      "loss": 0.2306,
      "num_input_tokens_seen": 5585376,
      "step": 26470
    },
    {
      "epoch": 2.9125412541254123,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.02981019909498444,
      "loss": 0.2269,
      "num_input_tokens_seen": 5586528,
      "step": 26475
    },
    {
      "epoch": 2.913091309130913,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.029809970669347725,
      "loss": 0.2332,
      "num_input_tokens_seen": 5587616,
      "step": 26480
    },
    {
      "epoch": 2.9136413641364136,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02980974210721462,
      "loss": 0.2322,
      "num_input_tokens_seen": 5588640,
      "step": 26485
    },
    {
      "epoch": 2.9141914191419143,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.02980951340858723,
      "loss": 0.2286,
      "num_input_tokens_seen": 5589600,
      "step": 26490
    },
    {
      "epoch": 2.914741474147415,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029809284573467657,
      "loss": 0.2327,
      "num_input_tokens_seen": 5590624,
      "step": 26495
    },
    {
      "epoch": 2.915291529152915,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02980905560185802,
      "loss": 0.2292,
      "num_input_tokens_seen": 5591712,
      "step": 26500
    },
    {
      "epoch": 2.9158415841584158,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.029808826493760418,
      "loss": 0.2292,
      "num_input_tokens_seen": 5592768,
      "step": 26505
    },
    {
      "epoch": 2.9163916391639164,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.02980859724917697,
      "loss": 0.2329,
      "num_input_tokens_seen": 5593856,
      "step": 26510
    },
    {
      "epoch": 2.916941694169417,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029808367868109788,
      "loss": 0.2307,
      "num_input_tokens_seen": 5594880,
      "step": 26515
    },
    {
      "epoch": 2.9174917491749177,
      "grad_norm": 0.01275634765625,
      "learning_rate": 0.029808138350560988,
      "loss": 0.2358,
      "num_input_tokens_seen": 5595904,
      "step": 26520
    },
    {
      "epoch": 2.918041804180418,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029807908696532676,
      "loss": 0.2322,
      "num_input_tokens_seen": 5596960,
      "step": 26525
    },
    {
      "epoch": 2.9185918591859186,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029807678906026976,
      "loss": 0.2295,
      "num_input_tokens_seen": 5597984,
      "step": 26530
    },
    {
      "epoch": 2.919141914191419,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.029807448979046008,
      "loss": 0.2295,
      "num_input_tokens_seen": 5599072,
      "step": 26535
    },
    {
      "epoch": 2.9196919691969194,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029807218915591887,
      "loss": 0.2317,
      "num_input_tokens_seen": 5600128,
      "step": 26540
    },
    {
      "epoch": 2.9202420242024205,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02980698871566674,
      "loss": 0.2338,
      "num_input_tokens_seen": 5601152,
      "step": 26545
    },
    {
      "epoch": 2.9207920792079207,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.029806758379272675,
      "loss": 0.2244,
      "num_input_tokens_seen": 5602208,
      "step": 26550
    },
    {
      "epoch": 2.9213421342134214,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029806527906411823,
      "loss": 0.2321,
      "num_input_tokens_seen": 5603296,
      "step": 26555
    },
    {
      "epoch": 2.921892189218922,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029806297297086313,
      "loss": 0.225,
      "num_input_tokens_seen": 5604320,
      "step": 26560
    },
    {
      "epoch": 2.9224422442244222,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029806066551298263,
      "loss": 0.2276,
      "num_input_tokens_seen": 5605408,
      "step": 26565
    },
    {
      "epoch": 2.922992299229923,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029805835669049804,
      "loss": 0.2333,
      "num_input_tokens_seen": 5606464,
      "step": 26570
    },
    {
      "epoch": 2.9235423542354235,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.02980560465034306,
      "loss": 0.2328,
      "num_input_tokens_seen": 5607520,
      "step": 26575
    },
    {
      "epoch": 2.924092409240924,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.029805373495180164,
      "loss": 0.2323,
      "num_input_tokens_seen": 5608576,
      "step": 26580
    },
    {
      "epoch": 2.924642464246425,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.029805142203563246,
      "loss": 0.2291,
      "num_input_tokens_seen": 5609632,
      "step": 26585
    },
    {
      "epoch": 2.925192519251925,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.02980491077549443,
      "loss": 0.2358,
      "num_input_tokens_seen": 5610720,
      "step": 26590
    },
    {
      "epoch": 2.9257425742574257,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029804679210975862,
      "loss": 0.2302,
      "num_input_tokens_seen": 5611776,
      "step": 26595
    },
    {
      "epoch": 2.9262926292629263,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.02980444751000967,
      "loss": 0.2312,
      "num_input_tokens_seen": 5612768,
      "step": 26600
    },
    {
      "epoch": 2.926842684268427,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.029804215672597986,
      "loss": 0.2338,
      "num_input_tokens_seen": 5613792,
      "step": 26605
    },
    {
      "epoch": 2.9273927392739276,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029803983698742948,
      "loss": 0.2275,
      "num_input_tokens_seen": 5614816,
      "step": 26610
    },
    {
      "epoch": 2.927942794279428,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0298037515884467,
      "loss": 0.2265,
      "num_input_tokens_seen": 5615872,
      "step": 26615
    },
    {
      "epoch": 2.9284928492849285,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.029803519341711375,
      "loss": 0.2312,
      "num_input_tokens_seen": 5616928,
      "step": 26620
    },
    {
      "epoch": 2.929042904290429,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.029803286958539118,
      "loss": 0.2333,
      "num_input_tokens_seen": 5617952,
      "step": 26625
    },
    {
      "epoch": 2.9295929592959293,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.029803054438932063,
      "loss": 0.2359,
      "num_input_tokens_seen": 5619040,
      "step": 26630
    },
    {
      "epoch": 2.93014301430143,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.029802821782892362,
      "loss": 0.227,
      "num_input_tokens_seen": 5620128,
      "step": 26635
    },
    {
      "epoch": 2.9306930693069306,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.02980258899042216,
      "loss": 0.2343,
      "num_input_tokens_seen": 5621184,
      "step": 26640
    },
    {
      "epoch": 2.9312431243124313,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.029802356061523594,
      "loss": 0.2296,
      "num_input_tokens_seen": 5622336,
      "step": 26645
    },
    {
      "epoch": 2.931793179317932,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.029802122996198815,
      "loss": 0.2347,
      "num_input_tokens_seen": 5623360,
      "step": 26650
    },
    {
      "epoch": 2.932343234323432,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.02980188979444997,
      "loss": 0.2311,
      "num_input_tokens_seen": 5624448,
      "step": 26655
    },
    {
      "epoch": 2.932893289328933,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.029801656456279208,
      "loss": 0.2346,
      "num_input_tokens_seen": 5625536,
      "step": 26660
    },
    {
      "epoch": 2.9334433443344334,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.02980142298168869,
      "loss": 0.2336,
      "num_input_tokens_seen": 5626560,
      "step": 26665
    },
    {
      "epoch": 2.933993399339934,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.02980118937068055,
      "loss": 0.2341,
      "num_input_tokens_seen": 5627520,
      "step": 26670
    },
    {
      "epoch": 2.9345434543454347,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029800955623256953,
      "loss": 0.2298,
      "num_input_tokens_seen": 5628640,
      "step": 26675
    },
    {
      "epoch": 2.935093509350935,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.029800721739420048,
      "loss": 0.2309,
      "num_input_tokens_seen": 5629696,
      "step": 26680
    },
    {
      "epoch": 2.9356435643564356,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.029800487719171997,
      "loss": 0.2293,
      "num_input_tokens_seen": 5630720,
      "step": 26685
    },
    {
      "epoch": 2.9361936193619362,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.029800253562514947,
      "loss": 0.2314,
      "num_input_tokens_seen": 5631872,
      "step": 26690
    },
    {
      "epoch": 2.936743674367437,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029800019269451066,
      "loss": 0.2304,
      "num_input_tokens_seen": 5632864,
      "step": 26695
    },
    {
      "epoch": 2.9372937293729375,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029799784839982506,
      "loss": 0.2294,
      "num_input_tokens_seen": 5633984,
      "step": 26700
    },
    {
      "epoch": 2.9378437843784377,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.029799550274111436,
      "loss": 0.233,
      "num_input_tokens_seen": 5635008,
      "step": 26705
    },
    {
      "epoch": 2.9383938393839384,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029799315571840006,
      "loss": 0.2299,
      "num_input_tokens_seen": 5636064,
      "step": 26710
    },
    {
      "epoch": 2.938943894389439,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.02979908073317039,
      "loss": 0.2299,
      "num_input_tokens_seen": 5637120,
      "step": 26715
    },
    {
      "epoch": 2.9394939493949392,
      "grad_norm": 0.002593994140625,
      "learning_rate": 0.02979884575810475,
      "loss": 0.2324,
      "num_input_tokens_seen": 5638176,
      "step": 26720
    },
    {
      "epoch": 2.94004400440044,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029798610646645244,
      "loss": 0.2314,
      "num_input_tokens_seen": 5639232,
      "step": 26725
    },
    {
      "epoch": 2.9405940594059405,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.02979837539879405,
      "loss": 0.234,
      "num_input_tokens_seen": 5640288,
      "step": 26730
    },
    {
      "epoch": 2.941144114411441,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.029798140014553325,
      "loss": 0.2283,
      "num_input_tokens_seen": 5641344,
      "step": 26735
    },
    {
      "epoch": 2.941694169416942,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.02979790449392525,
      "loss": 0.2303,
      "num_input_tokens_seen": 5642400,
      "step": 26740
    },
    {
      "epoch": 2.942244224422442,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.02979766883691199,
      "loss": 0.2319,
      "num_input_tokens_seen": 5643456,
      "step": 26745
    },
    {
      "epoch": 2.9427942794279427,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029797433043515714,
      "loss": 0.2288,
      "num_input_tokens_seen": 5644448,
      "step": 26750
    },
    {
      "epoch": 2.9433443344334433,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.0297971971137386,
      "loss": 0.2298,
      "num_input_tokens_seen": 5645536,
      "step": 26755
    },
    {
      "epoch": 2.943894389438944,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02979696104758282,
      "loss": 0.2299,
      "num_input_tokens_seen": 5646560,
      "step": 26760
    },
    {
      "epoch": 2.9444444444444446,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.02979672484505055,
      "loss": 0.2294,
      "num_input_tokens_seen": 5647520,
      "step": 26765
    },
    {
      "epoch": 2.944994499449945,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.029796488506143963,
      "loss": 0.2352,
      "num_input_tokens_seen": 5648608,
      "step": 26770
    },
    {
      "epoch": 2.9455445544554455,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02979625203086525,
      "loss": 0.231,
      "num_input_tokens_seen": 5649664,
      "step": 26775
    },
    {
      "epoch": 2.946094609460946,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.029796015419216577,
      "loss": 0.23,
      "num_input_tokens_seen": 5650720,
      "step": 26780
    },
    {
      "epoch": 2.946644664466447,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029795778671200132,
      "loss": 0.231,
      "num_input_tokens_seen": 5651776,
      "step": 26785
    },
    {
      "epoch": 2.9471947194719474,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02979554178681809,
      "loss": 0.2341,
      "num_input_tokens_seen": 5652768,
      "step": 26790
    },
    {
      "epoch": 2.9477447744774476,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029795304766072644,
      "loss": 0.2325,
      "num_input_tokens_seen": 5653856,
      "step": 26795
    },
    {
      "epoch": 2.9482948294829483,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02979506760896597,
      "loss": 0.23,
      "num_input_tokens_seen": 5654944,
      "step": 26800
    },
    {
      "epoch": 2.948844884488449,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.02979483031550026,
      "loss": 0.231,
      "num_input_tokens_seen": 5655968,
      "step": 26805
    },
    {
      "epoch": 2.949394939493949,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.029794592885677693,
      "loss": 0.2341,
      "num_input_tokens_seen": 5657056,
      "step": 26810
    },
    {
      "epoch": 2.94994499449945,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02979435531950047,
      "loss": 0.2304,
      "num_input_tokens_seen": 5658144,
      "step": 26815
    },
    {
      "epoch": 2.9504950495049505,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.02979411761697077,
      "loss": 0.2309,
      "num_input_tokens_seen": 5659136,
      "step": 26820
    },
    {
      "epoch": 2.951045104510451,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.02979387977809079,
      "loss": 0.2325,
      "num_input_tokens_seen": 5660160,
      "step": 26825
    },
    {
      "epoch": 2.9515951595159517,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.02979364180286271,
      "loss": 0.2325,
      "num_input_tokens_seen": 5661184,
      "step": 26830
    },
    {
      "epoch": 2.952145214521452,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.029793403691288738,
      "loss": 0.2351,
      "num_input_tokens_seen": 5662240,
      "step": 26835
    },
    {
      "epoch": 2.9526952695269526,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.029793165443371065,
      "loss": 0.2319,
      "num_input_tokens_seen": 5663328,
      "step": 26840
    },
    {
      "epoch": 2.9532453245324533,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.02979292705911188,
      "loss": 0.2314,
      "num_input_tokens_seen": 5664352,
      "step": 26845
    },
    {
      "epoch": 2.953795379537954,
      "grad_norm": 0.0135498046875,
      "learning_rate": 0.029792688538513388,
      "loss": 0.2304,
      "num_input_tokens_seen": 5665376,
      "step": 26850
    },
    {
      "epoch": 2.9543454345434546,
      "grad_norm": 0.00070953369140625,
      "learning_rate": 0.029792449881577778,
      "loss": 0.2309,
      "num_input_tokens_seen": 5666368,
      "step": 26855
    },
    {
      "epoch": 2.9548954895489548,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.029792211088307263,
      "loss": 0.2346,
      "num_input_tokens_seen": 5667424,
      "step": 26860
    },
    {
      "epoch": 2.9554455445544554,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.029791972158704032,
      "loss": 0.2336,
      "num_input_tokens_seen": 5668480,
      "step": 26865
    },
    {
      "epoch": 2.955995599559956,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.029791733092770292,
      "loss": 0.233,
      "num_input_tokens_seen": 5669536,
      "step": 26870
    },
    {
      "epoch": 2.9565456545654567,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.029791493890508246,
      "loss": 0.2314,
      "num_input_tokens_seen": 5670528,
      "step": 26875
    },
    {
      "epoch": 2.9570957095709574,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.0297912545519201,
      "loss": 0.2314,
      "num_input_tokens_seen": 5671616,
      "step": 26880
    },
    {
      "epoch": 2.9576457645764576,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.029791015077008057,
      "loss": 0.2304,
      "num_input_tokens_seen": 5672736,
      "step": 26885
    },
    {
      "epoch": 2.958195819581958,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029790775465774327,
      "loss": 0.2298,
      "num_input_tokens_seen": 5673760,
      "step": 26890
    },
    {
      "epoch": 2.958745874587459,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.029790535718221114,
      "loss": 0.2283,
      "num_input_tokens_seen": 5674784,
      "step": 26895
    },
    {
      "epoch": 2.959295929592959,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.02979029583435063,
      "loss": 0.2309,
      "num_input_tokens_seen": 5675808,
      "step": 26900
    },
    {
      "epoch": 2.9598459845984597,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.029790055814165092,
      "loss": 0.2315,
      "num_input_tokens_seen": 5676864,
      "step": 26905
    },
    {
      "epoch": 2.9603960396039604,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.029789815657666703,
      "loss": 0.2326,
      "num_input_tokens_seen": 5677952,
      "step": 26910
    },
    {
      "epoch": 2.960946094609461,
      "grad_norm": 0.0078125,
      "learning_rate": 0.02978957536485768,
      "loss": 0.2362,
      "num_input_tokens_seen": 5679008,
      "step": 26915
    },
    {
      "epoch": 2.9614961496149617,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.02978933493574024,
      "loss": 0.2319,
      "num_input_tokens_seen": 5680064,
      "step": 26920
    },
    {
      "epoch": 2.962046204620462,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02978909437031659,
      "loss": 0.2329,
      "num_input_tokens_seen": 5681088,
      "step": 26925
    },
    {
      "epoch": 2.9625962596259625,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02978885366858896,
      "loss": 0.2298,
      "num_input_tokens_seen": 5682080,
      "step": 26930
    },
    {
      "epoch": 2.963146314631463,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02978861283055956,
      "loss": 0.2329,
      "num_input_tokens_seen": 5683200,
      "step": 26935
    },
    {
      "epoch": 2.963696369636964,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029788371856230608,
      "loss": 0.2319,
      "num_input_tokens_seen": 5684288,
      "step": 26940
    },
    {
      "epoch": 2.9642464246424645,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029788130745604335,
      "loss": 0.2324,
      "num_input_tokens_seen": 5685344,
      "step": 26945
    },
    {
      "epoch": 2.9647964796479647,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.029787889498682954,
      "loss": 0.2319,
      "num_input_tokens_seen": 5686432,
      "step": 26950
    },
    {
      "epoch": 2.9653465346534653,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029787648115468694,
      "loss": 0.2293,
      "num_input_tokens_seen": 5687488,
      "step": 26955
    },
    {
      "epoch": 2.965896589658966,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029787406595963774,
      "loss": 0.2303,
      "num_input_tokens_seen": 5688512,
      "step": 26960
    },
    {
      "epoch": 2.966446644664466,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029787164940170426,
      "loss": 0.2308,
      "num_input_tokens_seen": 5689568,
      "step": 26965
    },
    {
      "epoch": 2.9669966996699673,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.029786923148090873,
      "loss": 0.2314,
      "num_input_tokens_seen": 5690624,
      "step": 26970
    },
    {
      "epoch": 2.9675467546754675,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.029786681219727343,
      "loss": 0.2319,
      "num_input_tokens_seen": 5691680,
      "step": 26975
    },
    {
      "epoch": 2.968096809680968,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.02978643915508207,
      "loss": 0.2309,
      "num_input_tokens_seen": 5692800,
      "step": 26980
    },
    {
      "epoch": 2.9686468646864688,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02978619695415728,
      "loss": 0.2314,
      "num_input_tokens_seen": 5693792,
      "step": 26985
    },
    {
      "epoch": 2.969196919691969,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029785954616955206,
      "loss": 0.2319,
      "num_input_tokens_seen": 5694816,
      "step": 26990
    },
    {
      "epoch": 2.9697469746974696,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.029785712143478086,
      "loss": 0.2293,
      "num_input_tokens_seen": 5695840,
      "step": 26995
    },
    {
      "epoch": 2.9702970297029703,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.029785469533728154,
      "loss": 0.2304,
      "num_input_tokens_seen": 5696896,
      "step": 27000
    },
    {
      "epoch": 2.970847084708471,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.029785226787707644,
      "loss": 0.2309,
      "num_input_tokens_seen": 5697984,
      "step": 27005
    },
    {
      "epoch": 2.9713971397139716,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.02978498390541879,
      "loss": 0.2324,
      "num_input_tokens_seen": 5698944,
      "step": 27010
    },
    {
      "epoch": 2.9719471947194718,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.029784740886863836,
      "loss": 0.2314,
      "num_input_tokens_seen": 5699968,
      "step": 27015
    },
    {
      "epoch": 2.9724972497249724,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02978449773204502,
      "loss": 0.2288,
      "num_input_tokens_seen": 5700960,
      "step": 27020
    },
    {
      "epoch": 2.973047304730473,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.02978425444096458,
      "loss": 0.2335,
      "num_input_tokens_seen": 5701984,
      "step": 27025
    },
    {
      "epoch": 2.9735973597359737,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02978401101362476,
      "loss": 0.2346,
      "num_input_tokens_seen": 5703072,
      "step": 27030
    },
    {
      "epoch": 2.9741474147414744,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.02978376745002781,
      "loss": 0.2324,
      "num_input_tokens_seen": 5704064,
      "step": 27035
    },
    {
      "epoch": 2.9746974697469746,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.02978352375017597,
      "loss": 0.2303,
      "num_input_tokens_seen": 5705088,
      "step": 27040
    },
    {
      "epoch": 2.9752475247524752,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02978327991407148,
      "loss": 0.2293,
      "num_input_tokens_seen": 5706176,
      "step": 27045
    },
    {
      "epoch": 2.975797579757976,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.029783035941716596,
      "loss": 0.2319,
      "num_input_tokens_seen": 5707264,
      "step": 27050
    },
    {
      "epoch": 2.976347634763476,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02978279183311356,
      "loss": 0.2283,
      "num_input_tokens_seen": 5708256,
      "step": 27055
    },
    {
      "epoch": 2.976897689768977,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.029782547588264627,
      "loss": 0.2295,
      "num_input_tokens_seen": 5709344,
      "step": 27060
    },
    {
      "epoch": 2.9774477447744774,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.02978230320717205,
      "loss": 0.2296,
      "num_input_tokens_seen": 5710336,
      "step": 27065
    },
    {
      "epoch": 2.977997799779978,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029782058689838072,
      "loss": 0.2287,
      "num_input_tokens_seen": 5711360,
      "step": 27070
    },
    {
      "epoch": 2.9785478547854787,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.029781814036264957,
      "loss": 0.2272,
      "num_input_tokens_seen": 5712416,
      "step": 27075
    },
    {
      "epoch": 2.979097909790979,
      "grad_norm": 0.0177001953125,
      "learning_rate": 0.02978156924645495,
      "loss": 0.2286,
      "num_input_tokens_seen": 5713408,
      "step": 27080
    },
    {
      "epoch": 2.9796479647964795,
      "grad_norm": 0.00775146484375,
      "learning_rate": 0.029781324320410318,
      "loss": 0.2196,
      "num_input_tokens_seen": 5714400,
      "step": 27085
    },
    {
      "epoch": 2.98019801980198,
      "grad_norm": 0.002899169921875,
      "learning_rate": 0.029781079258133305,
      "loss": 0.2329,
      "num_input_tokens_seen": 5715424,
      "step": 27090
    },
    {
      "epoch": 2.980748074807481,
      "grad_norm": 0.0135498046875,
      "learning_rate": 0.02978083405962618,
      "loss": 0.2341,
      "num_input_tokens_seen": 5716512,
      "step": 27095
    },
    {
      "epoch": 2.9812981298129815,
      "grad_norm": 0.003326416015625,
      "learning_rate": 0.029780588724891203,
      "loss": 0.2261,
      "num_input_tokens_seen": 5717536,
      "step": 27100
    },
    {
      "epoch": 2.9818481848184817,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.02978034325393063,
      "loss": 0.2208,
      "num_input_tokens_seen": 5718656,
      "step": 27105
    },
    {
      "epoch": 2.9823982398239823,
      "grad_norm": 0.004425048828125,
      "learning_rate": 0.029780097646746728,
      "loss": 0.2337,
      "num_input_tokens_seen": 5719744,
      "step": 27110
    },
    {
      "epoch": 2.982948294829483,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.029779851903341755,
      "loss": 0.2459,
      "num_input_tokens_seen": 5720800,
      "step": 27115
    },
    {
      "epoch": 2.9834983498349836,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.029779606023717983,
      "loss": 0.2295,
      "num_input_tokens_seen": 5721824,
      "step": 27120
    },
    {
      "epoch": 2.9840484048404843,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.02977936000787767,
      "loss": 0.2244,
      "num_input_tokens_seen": 5722944,
      "step": 27125
    },
    {
      "epoch": 2.9845984598459845,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.029779113855823085,
      "loss": 0.234,
      "num_input_tokens_seen": 5724032,
      "step": 27130
    },
    {
      "epoch": 2.985148514851485,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.029778867567556506,
      "loss": 0.2363,
      "num_input_tokens_seen": 5725056,
      "step": 27135
    },
    {
      "epoch": 2.985698569856986,
      "grad_norm": 0.014892578125,
      "learning_rate": 0.029778621143080195,
      "loss": 0.2387,
      "num_input_tokens_seen": 5726080,
      "step": 27140
    },
    {
      "epoch": 2.986248624862486,
      "grad_norm": 0.017578125,
      "learning_rate": 0.029778374582396423,
      "loss": 0.2253,
      "num_input_tokens_seen": 5727136,
      "step": 27145
    },
    {
      "epoch": 2.9867986798679866,
      "grad_norm": 0.0247802734375,
      "learning_rate": 0.029778127885507464,
      "loss": 0.2292,
      "num_input_tokens_seen": 5728160,
      "step": 27150
    },
    {
      "epoch": 2.9873487348734873,
      "grad_norm": 0.00323486328125,
      "learning_rate": 0.02977788105241559,
      "loss": 0.2302,
      "num_input_tokens_seen": 5729184,
      "step": 27155
    },
    {
      "epoch": 2.987898789878988,
      "grad_norm": 0.0152587890625,
      "learning_rate": 0.02977763408312308,
      "loss": 0.233,
      "num_input_tokens_seen": 5730208,
      "step": 27160
    },
    {
      "epoch": 2.9884488448844886,
      "grad_norm": 0.0380859375,
      "learning_rate": 0.029777386977632204,
      "loss": 0.2262,
      "num_input_tokens_seen": 5731232,
      "step": 27165
    },
    {
      "epoch": 2.988998899889989,
      "grad_norm": 0.0220947265625,
      "learning_rate": 0.029777139735945243,
      "loss": 0.2355,
      "num_input_tokens_seen": 5732288,
      "step": 27170
    },
    {
      "epoch": 2.9895489548954894,
      "grad_norm": 0.0322265625,
      "learning_rate": 0.02977689235806448,
      "loss": 0.2302,
      "num_input_tokens_seen": 5733344,
      "step": 27175
    },
    {
      "epoch": 2.99009900990099,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.029776644843992194,
      "loss": 0.2314,
      "num_input_tokens_seen": 5734336,
      "step": 27180
    },
    {
      "epoch": 2.9906490649064907,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.029776397193730656,
      "loss": 0.2301,
      "num_input_tokens_seen": 5735424,
      "step": 27185
    },
    {
      "epoch": 2.9911991199119914,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.029776149407282158,
      "loss": 0.2316,
      "num_input_tokens_seen": 5736512,
      "step": 27190
    },
    {
      "epoch": 2.9917491749174916,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.02977590148464898,
      "loss": 0.2332,
      "num_input_tokens_seen": 5737536,
      "step": 27195
    },
    {
      "epoch": 2.9922992299229922,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.02977565342583341,
      "loss": 0.2265,
      "num_input_tokens_seen": 5738592,
      "step": 27200
    },
    {
      "epoch": 2.992849284928493,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02977540523083773,
      "loss": 0.2337,
      "num_input_tokens_seen": 5739648,
      "step": 27205
    },
    {
      "epoch": 2.9933993399339935,
      "grad_norm": 0.01324462890625,
      "learning_rate": 0.029775156899664233,
      "loss": 0.2384,
      "num_input_tokens_seen": 5740640,
      "step": 27210
    },
    {
      "epoch": 2.993949394939494,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.029774908432315204,
      "loss": 0.2285,
      "num_input_tokens_seen": 5741728,
      "step": 27215
    },
    {
      "epoch": 2.9944994499449944,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.02977465982879293,
      "loss": 0.2357,
      "num_input_tokens_seen": 5742784,
      "step": 27220
    },
    {
      "epoch": 2.995049504950495,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.029774411089099707,
      "loss": 0.2336,
      "num_input_tokens_seen": 5743872,
      "step": 27225
    },
    {
      "epoch": 2.9955995599559957,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.02977416221323783,
      "loss": 0.2298,
      "num_input_tokens_seen": 5744928,
      "step": 27230
    },
    {
      "epoch": 2.996149614961496,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.02977391320120959,
      "loss": 0.2288,
      "num_input_tokens_seen": 5745984,
      "step": 27235
    },
    {
      "epoch": 2.9966996699669965,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.029773664053017276,
      "loss": 0.2324,
      "num_input_tokens_seen": 5747040,
      "step": 27240
    },
    {
      "epoch": 2.997249724972497,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.029773414768663194,
      "loss": 0.2326,
      "num_input_tokens_seen": 5748064,
      "step": 27245
    },
    {
      "epoch": 2.997799779977998,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02977316534814963,
      "loss": 0.2304,
      "num_input_tokens_seen": 5749120,
      "step": 27250
    },
    {
      "epoch": 2.9983498349834985,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.029772915791478897,
      "loss": 0.2319,
      "num_input_tokens_seen": 5750176,
      "step": 27255
    },
    {
      "epoch": 2.9988998899889987,
      "grad_norm": 0.01251220703125,
      "learning_rate": 0.029772666098653287,
      "loss": 0.2309,
      "num_input_tokens_seen": 5751232,
      "step": 27260
    },
    {
      "epoch": 2.9994499449944994,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.0297724162696751,
      "loss": 0.232,
      "num_input_tokens_seen": 5752256,
      "step": 27265
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.02977216630454664,
      "loss": 0.2335,
      "num_input_tokens_seen": 5753152,
      "step": 27270
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.23131345212459564,
      "eval_runtime": 60.5407,
      "eval_samples_per_second": 66.732,
      "eval_steps_per_second": 16.683,
      "num_input_tokens_seen": 5753152,
      "step": 27270
    },
    {
      "epoch": 3.0005500550055006,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.029771916203270216,
      "loss": 0.232,
      "num_input_tokens_seen": 5754176,
      "step": 27275
    },
    {
      "epoch": 3.0011001100110013,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.029771665965848126,
      "loss": 0.2335,
      "num_input_tokens_seen": 5755232,
      "step": 27280
    },
    {
      "epoch": 3.0016501650165015,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.029771415592282677,
      "loss": 0.2288,
      "num_input_tokens_seen": 5756384,
      "step": 27285
    },
    {
      "epoch": 3.002200220022002,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.02977116508257618,
      "loss": 0.2293,
      "num_input_tokens_seen": 5757504,
      "step": 27290
    },
    {
      "epoch": 3.002750275027503,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.029770914436730942,
      "loss": 0.2356,
      "num_input_tokens_seen": 5758624,
      "step": 27295
    },
    {
      "epoch": 3.0033003300330035,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.02977066365474927,
      "loss": 0.2288,
      "num_input_tokens_seen": 5759712,
      "step": 27300
    },
    {
      "epoch": 3.0038503850385037,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.029770412736633486,
      "loss": 0.2329,
      "num_input_tokens_seen": 5760768,
      "step": 27305
    },
    {
      "epoch": 3.0044004400440043,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.029770161682385894,
      "loss": 0.2303,
      "num_input_tokens_seen": 5761920,
      "step": 27310
    },
    {
      "epoch": 3.004950495049505,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.029769910492008805,
      "loss": 0.2319,
      "num_input_tokens_seen": 5763008,
      "step": 27315
    },
    {
      "epoch": 3.0055005500550056,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.029769659165504544,
      "loss": 0.2313,
      "num_input_tokens_seen": 5764096,
      "step": 27320
    },
    {
      "epoch": 3.0060506050605063,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.02976940770287542,
      "loss": 0.2335,
      "num_input_tokens_seen": 5765216,
      "step": 27325
    },
    {
      "epoch": 3.0066006600660065,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.029769156104123746,
      "loss": 0.235,
      "num_input_tokens_seen": 5766336,
      "step": 27330
    },
    {
      "epoch": 3.007150715071507,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.02976890436925185,
      "loss": 0.2308,
      "num_input_tokens_seen": 5767360,
      "step": 27335
    },
    {
      "epoch": 3.0077007700770078,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.029768652498262053,
      "loss": 0.2292,
      "num_input_tokens_seen": 5768480,
      "step": 27340
    },
    {
      "epoch": 3.0082508250825084,
      "grad_norm": 0.01409912109375,
      "learning_rate": 0.029768400491156668,
      "loss": 0.2308,
      "num_input_tokens_seen": 5769504,
      "step": 27345
    },
    {
      "epoch": 3.0088008800880086,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.029768148347938027,
      "loss": 0.2313,
      "num_input_tokens_seen": 5770624,
      "step": 27350
    },
    {
      "epoch": 3.0093509350935093,
      "grad_norm": 0.0072021484375,
      "learning_rate": 0.029767896068608447,
      "loss": 0.2297,
      "num_input_tokens_seen": 5771648,
      "step": 27355
    },
    {
      "epoch": 3.00990099009901,
      "grad_norm": 0.0150146484375,
      "learning_rate": 0.029767643653170254,
      "loss": 0.2323,
      "num_input_tokens_seen": 5772672,
      "step": 27360
    },
    {
      "epoch": 3.0104510451045106,
      "grad_norm": 0.0084228515625,
      "learning_rate": 0.029767391101625778,
      "loss": 0.2318,
      "num_input_tokens_seen": 5773696,
      "step": 27365
    },
    {
      "epoch": 3.011001100110011,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.029767138413977344,
      "loss": 0.2323,
      "num_input_tokens_seen": 5774848,
      "step": 27370
    },
    {
      "epoch": 3.0115511551155114,
      "grad_norm": 0.007659912109375,
      "learning_rate": 0.02976688559022728,
      "loss": 0.2323,
      "num_input_tokens_seen": 5775936,
      "step": 27375
    },
    {
      "epoch": 3.012101210121012,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.029766632630377914,
      "loss": 0.2318,
      "num_input_tokens_seen": 5776960,
      "step": 27380
    },
    {
      "epoch": 3.0126512651265127,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02976637953443159,
      "loss": 0.2288,
      "num_input_tokens_seen": 5777984,
      "step": 27385
    },
    {
      "epoch": 3.0132013201320134,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029766126302390623,
      "loss": 0.2341,
      "num_input_tokens_seen": 5779040,
      "step": 27390
    },
    {
      "epoch": 3.0137513751375136,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.029765872934257358,
      "loss": 0.2314,
      "num_input_tokens_seen": 5780128,
      "step": 27395
    },
    {
      "epoch": 3.014301430143014,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029765619430034126,
      "loss": 0.2269,
      "num_input_tokens_seen": 5781216,
      "step": 27400
    },
    {
      "epoch": 3.014851485148515,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.029765365789723268,
      "loss": 0.2294,
      "num_input_tokens_seen": 5782336,
      "step": 27405
    },
    {
      "epoch": 3.0154015401540155,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029765112013327117,
      "loss": 0.2306,
      "num_input_tokens_seen": 5783360,
      "step": 27410
    },
    {
      "epoch": 3.015951595159516,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.029764858100848016,
      "loss": 0.2327,
      "num_input_tokens_seen": 5784352,
      "step": 27415
    },
    {
      "epoch": 3.0165016501650164,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0297646040522883,
      "loss": 0.2341,
      "num_input_tokens_seen": 5785440,
      "step": 27420
    },
    {
      "epoch": 3.017051705170517,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.02976434986765031,
      "loss": 0.2321,
      "num_input_tokens_seen": 5786528,
      "step": 27425
    },
    {
      "epoch": 3.0176017601760177,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.029764095546936396,
      "loss": 0.23,
      "num_input_tokens_seen": 5787584,
      "step": 27430
    },
    {
      "epoch": 3.0181518151815183,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0297638410901489,
      "loss": 0.2336,
      "num_input_tokens_seen": 5788704,
      "step": 27435
    },
    {
      "epoch": 3.0187018701870185,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029763586497290162,
      "loss": 0.2315,
      "num_input_tokens_seen": 5789760,
      "step": 27440
    },
    {
      "epoch": 3.019251925192519,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.029763331768362534,
      "loss": 0.2346,
      "num_input_tokens_seen": 5790784,
      "step": 27445
    },
    {
      "epoch": 3.01980198019802,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.02976307690336836,
      "loss": 0.2335,
      "num_input_tokens_seen": 5791808,
      "step": 27450
    },
    {
      "epoch": 3.0203520352035205,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.029762821902309985,
      "loss": 0.234,
      "num_input_tokens_seen": 5792896,
      "step": 27455
    },
    {
      "epoch": 3.020902090209021,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029762566765189773,
      "loss": 0.2308,
      "num_input_tokens_seen": 5793888,
      "step": 27460
    },
    {
      "epoch": 3.0214521452145213,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02976231149201006,
      "loss": 0.2303,
      "num_input_tokens_seen": 5794976,
      "step": 27465
    },
    {
      "epoch": 3.022002200220022,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.02976205608277321,
      "loss": 0.2324,
      "num_input_tokens_seen": 5795968,
      "step": 27470
    },
    {
      "epoch": 3.0225522552255226,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.02976180053748157,
      "loss": 0.2314,
      "num_input_tokens_seen": 5797056,
      "step": 27475
    },
    {
      "epoch": 3.0231023102310233,
      "grad_norm": 0.01190185546875,
      "learning_rate": 0.0297615448561375,
      "loss": 0.2324,
      "num_input_tokens_seen": 5798080,
      "step": 27480
    },
    {
      "epoch": 3.0236523652365235,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.029761289038743352,
      "loss": 0.2313,
      "num_input_tokens_seen": 5799104,
      "step": 27485
    },
    {
      "epoch": 3.024202420242024,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029761033085301487,
      "loss": 0.2313,
      "num_input_tokens_seen": 5800160,
      "step": 27490
    },
    {
      "epoch": 3.0247524752475248,
      "grad_norm": 0.00775146484375,
      "learning_rate": 0.02976077699581426,
      "loss": 0.2293,
      "num_input_tokens_seen": 5801280,
      "step": 27495
    },
    {
      "epoch": 3.0253025302530254,
      "grad_norm": 0.00848388671875,
      "learning_rate": 0.02976052077028404,
      "loss": 0.2296,
      "num_input_tokens_seen": 5802336,
      "step": 27500
    },
    {
      "epoch": 3.0258525852585256,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.029760264408713183,
      "loss": 0.237,
      "num_input_tokens_seen": 5803456,
      "step": 27505
    },
    {
      "epoch": 3.0264026402640263,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.029760007911104048,
      "loss": 0.2333,
      "num_input_tokens_seen": 5804480,
      "step": 27510
    },
    {
      "epoch": 3.026952695269527,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.029759751277459,
      "loss": 0.2271,
      "num_input_tokens_seen": 5805536,
      "step": 27515
    },
    {
      "epoch": 3.0275027502750276,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.029759494507780412,
      "loss": 0.2327,
      "num_input_tokens_seen": 5806560,
      "step": 27520
    },
    {
      "epoch": 3.0280528052805282,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029759237602070644,
      "loss": 0.2337,
      "num_input_tokens_seen": 5807584,
      "step": 27525
    },
    {
      "epoch": 3.0286028602860284,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.029758980560332066,
      "loss": 0.2306,
      "num_input_tokens_seen": 5808672,
      "step": 27530
    },
    {
      "epoch": 3.029152915291529,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.029758723382567045,
      "loss": 0.2316,
      "num_input_tokens_seen": 5809728,
      "step": 27535
    },
    {
      "epoch": 3.0297029702970297,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029758466068777953,
      "loss": 0.2306,
      "num_input_tokens_seen": 5810720,
      "step": 27540
    },
    {
      "epoch": 3.0302530253025304,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02975820861896716,
      "loss": 0.2311,
      "num_input_tokens_seen": 5811808,
      "step": 27545
    },
    {
      "epoch": 3.0308030803080306,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.02975795103313704,
      "loss": 0.2384,
      "num_input_tokens_seen": 5812864,
      "step": 27550
    },
    {
      "epoch": 3.0313531353135312,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.02975769331128997,
      "loss": 0.2305,
      "num_input_tokens_seen": 5813856,
      "step": 27555
    },
    {
      "epoch": 3.031903190319032,
      "grad_norm": 0.0123291015625,
      "learning_rate": 0.02975743545342832,
      "loss": 0.2273,
      "num_input_tokens_seen": 5815008,
      "step": 27560
    },
    {
      "epoch": 3.0324532453245325,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02975717745955447,
      "loss": 0.2336,
      "num_input_tokens_seen": 5816160,
      "step": 27565
    },
    {
      "epoch": 3.033003300330033,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029756919329670797,
      "loss": 0.2341,
      "num_input_tokens_seen": 5817280,
      "step": 27570
    },
    {
      "epoch": 3.0335533553355334,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.029756661063779676,
      "loss": 0.2325,
      "num_input_tokens_seen": 5818304,
      "step": 27575
    },
    {
      "epoch": 3.034103410341034,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029756402661883494,
      "loss": 0.2309,
      "num_input_tokens_seen": 5819328,
      "step": 27580
    },
    {
      "epoch": 3.0346534653465347,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.029756144123984626,
      "loss": 0.2319,
      "num_input_tokens_seen": 5820384,
      "step": 27585
    },
    {
      "epoch": 3.0352035203520353,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029755885450085458,
      "loss": 0.2329,
      "num_input_tokens_seen": 5821440,
      "step": 27590
    },
    {
      "epoch": 3.0357535753575355,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02975562664018838,
      "loss": 0.2309,
      "num_input_tokens_seen": 5822560,
      "step": 27595
    },
    {
      "epoch": 3.036303630363036,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.02975536769429577,
      "loss": 0.233,
      "num_input_tokens_seen": 5823680,
      "step": 27600
    },
    {
      "epoch": 3.036853685368537,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.02975510861241001,
      "loss": 0.2319,
      "num_input_tokens_seen": 5824736,
      "step": 27605
    },
    {
      "epoch": 3.0374037403740375,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029754849394533503,
      "loss": 0.2309,
      "num_input_tokens_seen": 5825792,
      "step": 27610
    },
    {
      "epoch": 3.037953795379538,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029754590040668622,
      "loss": 0.2309,
      "num_input_tokens_seen": 5826816,
      "step": 27615
    },
    {
      "epoch": 3.0385038503850383,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02975433055081777,
      "loss": 0.2319,
      "num_input_tokens_seen": 5827904,
      "step": 27620
    },
    {
      "epoch": 3.039053905390539,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029754070924983334,
      "loss": 0.2319,
      "num_input_tokens_seen": 5828928,
      "step": 27625
    },
    {
      "epoch": 3.0396039603960396,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.0297538111631677,
      "loss": 0.2319,
      "num_input_tokens_seen": 5829952,
      "step": 27630
    },
    {
      "epoch": 3.0401540154015403,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.029753551265373268,
      "loss": 0.2314,
      "num_input_tokens_seen": 5831072,
      "step": 27635
    },
    {
      "epoch": 3.0407040704070405,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02975329123160244,
      "loss": 0.2308,
      "num_input_tokens_seen": 5832128,
      "step": 27640
    },
    {
      "epoch": 3.041254125412541,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029753031061857598,
      "loss": 0.2298,
      "num_input_tokens_seen": 5833248,
      "step": 27645
    },
    {
      "epoch": 3.041804180418042,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.029752770756141154,
      "loss": 0.2298,
      "num_input_tokens_seen": 5834304,
      "step": 27650
    },
    {
      "epoch": 3.0423542354235424,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.0297525103144555,
      "loss": 0.234,
      "num_input_tokens_seen": 5835328,
      "step": 27655
    },
    {
      "epoch": 3.042904290429043,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.029752249736803037,
      "loss": 0.2309,
      "num_input_tokens_seen": 5836352,
      "step": 27660
    },
    {
      "epoch": 3.0434543454345433,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.029751989023186164,
      "loss": 0.2308,
      "num_input_tokens_seen": 5837440,
      "step": 27665
    },
    {
      "epoch": 3.044004400440044,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.02975172817360729,
      "loss": 0.2309,
      "num_input_tokens_seen": 5838432,
      "step": 27670
    },
    {
      "epoch": 3.0445544554455446,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.029751467188068818,
      "loss": 0.2319,
      "num_input_tokens_seen": 5839520,
      "step": 27675
    },
    {
      "epoch": 3.0451045104510452,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029751206066573145,
      "loss": 0.2319,
      "num_input_tokens_seen": 5840576,
      "step": 27680
    },
    {
      "epoch": 3.0456545654565454,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.02975094480912269,
      "loss": 0.2319,
      "num_input_tokens_seen": 5841568,
      "step": 27685
    },
    {
      "epoch": 3.046204620462046,
      "grad_norm": 0.00080108642578125,
      "learning_rate": 0.029750683415719854,
      "loss": 0.2298,
      "num_input_tokens_seen": 5842624,
      "step": 27690
    },
    {
      "epoch": 3.0467546754675467,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.029750421886367045,
      "loss": 0.2304,
      "num_input_tokens_seen": 5843712,
      "step": 27695
    },
    {
      "epoch": 3.0473047304730474,
      "grad_norm": 0.007171630859375,
      "learning_rate": 0.029750160221066673,
      "loss": 0.23,
      "num_input_tokens_seen": 5844832,
      "step": 27700
    },
    {
      "epoch": 3.047854785478548,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.029749898419821157,
      "loss": 0.2348,
      "num_input_tokens_seen": 5845952,
      "step": 27705
    },
    {
      "epoch": 3.0484048404840483,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.029749636482632905,
      "loss": 0.2296,
      "num_input_tokens_seen": 5846976,
      "step": 27710
    },
    {
      "epoch": 3.048954895489549,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.02974937440950433,
      "loss": 0.2363,
      "num_input_tokens_seen": 5848032,
      "step": 27715
    },
    {
      "epoch": 3.0495049504950495,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.029749112200437848,
      "loss": 0.2316,
      "num_input_tokens_seen": 5849120,
      "step": 27720
    },
    {
      "epoch": 3.05005500550055,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.029748849855435872,
      "loss": 0.228,
      "num_input_tokens_seen": 5850112,
      "step": 27725
    },
    {
      "epoch": 3.0506050605060504,
      "grad_norm": 0.00726318359375,
      "learning_rate": 0.02974858737450083,
      "loss": 0.228,
      "num_input_tokens_seen": 5851200,
      "step": 27730
    },
    {
      "epoch": 3.051155115511551,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02974832475763513,
      "loss": 0.2292,
      "num_input_tokens_seen": 5852224,
      "step": 27735
    },
    {
      "epoch": 3.0517051705170517,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029748062004841203,
      "loss": 0.2328,
      "num_input_tokens_seen": 5853280,
      "step": 27740
    },
    {
      "epoch": 3.0522552255225524,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.02974779911612146,
      "loss": 0.2245,
      "num_input_tokens_seen": 5854368,
      "step": 27745
    },
    {
      "epoch": 3.052805280528053,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.029747536091478337,
      "loss": 0.2398,
      "num_input_tokens_seen": 5855392,
      "step": 27750
    },
    {
      "epoch": 3.053355335533553,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029747272930914247,
      "loss": 0.2392,
      "num_input_tokens_seen": 5856384,
      "step": 27755
    },
    {
      "epoch": 3.053905390539054,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.029747009634431616,
      "loss": 0.2239,
      "num_input_tokens_seen": 5857408,
      "step": 27760
    },
    {
      "epoch": 3.0544554455445545,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.029746746202032874,
      "loss": 0.2332,
      "num_input_tokens_seen": 5858464,
      "step": 27765
    },
    {
      "epoch": 3.055005500550055,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02974648263372045,
      "loss": 0.2353,
      "num_input_tokens_seen": 5859520,
      "step": 27770
    },
    {
      "epoch": 3.0555555555555554,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.029746218929496773,
      "loss": 0.2259,
      "num_input_tokens_seen": 5860576,
      "step": 27775
    },
    {
      "epoch": 3.056105610561056,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.02974595508936427,
      "loss": 0.2223,
      "num_input_tokens_seen": 5861632,
      "step": 27780
    },
    {
      "epoch": 3.0566556655665567,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.029745691113325375,
      "loss": 0.2332,
      "num_input_tokens_seen": 5862656,
      "step": 27785
    },
    {
      "epoch": 3.0572057205720573,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02974542700138252,
      "loss": 0.2287,
      "num_input_tokens_seen": 5863712,
      "step": 27790
    },
    {
      "epoch": 3.057755775577558,
      "grad_norm": 0.01422119140625,
      "learning_rate": 0.02974516275353814,
      "loss": 0.2381,
      "num_input_tokens_seen": 5864704,
      "step": 27795
    },
    {
      "epoch": 3.058305830583058,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02974489836979467,
      "loss": 0.2303,
      "num_input_tokens_seen": 5865728,
      "step": 27800
    },
    {
      "epoch": 3.058855885588559,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.02974463385015455,
      "loss": 0.2323,
      "num_input_tokens_seen": 5866752,
      "step": 27805
    },
    {
      "epoch": 3.0594059405940595,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.02974436919462021,
      "loss": 0.2364,
      "num_input_tokens_seen": 5867808,
      "step": 27810
    },
    {
      "epoch": 3.05995599559956,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029744104403194097,
      "loss": 0.2332,
      "num_input_tokens_seen": 5868832,
      "step": 27815
    },
    {
      "epoch": 3.0605060506050603,
      "grad_norm": 0.01190185546875,
      "learning_rate": 0.02974383947587865,
      "loss": 0.2362,
      "num_input_tokens_seen": 5869856,
      "step": 27820
    },
    {
      "epoch": 3.061056105610561,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.029743574412676308,
      "loss": 0.2326,
      "num_input_tokens_seen": 5870944,
      "step": 27825
    },
    {
      "epoch": 3.0616061606160616,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029743309213589513,
      "loss": 0.2315,
      "num_input_tokens_seen": 5872000,
      "step": 27830
    },
    {
      "epoch": 3.0621562156215623,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.029743043878620718,
      "loss": 0.232,
      "num_input_tokens_seen": 5873120,
      "step": 27835
    },
    {
      "epoch": 3.062706270627063,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.029742778407772355,
      "loss": 0.2294,
      "num_input_tokens_seen": 5874144,
      "step": 27840
    },
    {
      "epoch": 3.063256325632563,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.02974251280104688,
      "loss": 0.2314,
      "num_input_tokens_seen": 5875232,
      "step": 27845
    },
    {
      "epoch": 3.0638063806380638,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02974224705844674,
      "loss": 0.2324,
      "num_input_tokens_seen": 5876256,
      "step": 27850
    },
    {
      "epoch": 3.0643564356435644,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.029741981179974383,
      "loss": 0.2329,
      "num_input_tokens_seen": 5877312,
      "step": 27855
    },
    {
      "epoch": 3.064906490649065,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029741715165632256,
      "loss": 0.2308,
      "num_input_tokens_seen": 5878336,
      "step": 27860
    },
    {
      "epoch": 3.0654565456545653,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.02974144901542282,
      "loss": 0.2308,
      "num_input_tokens_seen": 5879424,
      "step": 27865
    },
    {
      "epoch": 3.066006600660066,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029741182729348514,
      "loss": 0.2298,
      "num_input_tokens_seen": 5880512,
      "step": 27870
    },
    {
      "epoch": 3.0665566556655666,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.029740916307411807,
      "loss": 0.2304,
      "num_input_tokens_seen": 5881504,
      "step": 27875
    },
    {
      "epoch": 3.067106710671067,
      "grad_norm": 0.01171875,
      "learning_rate": 0.029740649749615145,
      "loss": 0.2299,
      "num_input_tokens_seen": 5882496,
      "step": 27880
    },
    {
      "epoch": 3.067656765676568,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.029740383055960988,
      "loss": 0.2315,
      "num_input_tokens_seen": 5883584,
      "step": 27885
    },
    {
      "epoch": 3.068206820682068,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.029740116226451795,
      "loss": 0.2304,
      "num_input_tokens_seen": 5884672,
      "step": 27890
    },
    {
      "epoch": 3.0687568756875687,
      "grad_norm": 0.0081787109375,
      "learning_rate": 0.02973984926109002,
      "loss": 0.2259,
      "num_input_tokens_seen": 5885696,
      "step": 27895
    },
    {
      "epoch": 3.0693069306930694,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.029739582159878132,
      "loss": 0.2264,
      "num_input_tokens_seen": 5886784,
      "step": 27900
    },
    {
      "epoch": 3.06985698569857,
      "grad_norm": 0.0206298828125,
      "learning_rate": 0.029739314922818585,
      "loss": 0.2344,
      "num_input_tokens_seen": 5887872,
      "step": 27905
    },
    {
      "epoch": 3.0704070407040702,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.029739047549913845,
      "loss": 0.2334,
      "num_input_tokens_seen": 5888896,
      "step": 27910
    },
    {
      "epoch": 3.070957095709571,
      "grad_norm": 0.007720947265625,
      "learning_rate": 0.029738780041166375,
      "loss": 0.2261,
      "num_input_tokens_seen": 5889888,
      "step": 27915
    },
    {
      "epoch": 3.0715071507150715,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.02973851239657864,
      "loss": 0.2292,
      "num_input_tokens_seen": 5890912,
      "step": 27920
    },
    {
      "epoch": 3.072057205720572,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.029738244616153115,
      "loss": 0.2202,
      "num_input_tokens_seen": 5892000,
      "step": 27925
    },
    {
      "epoch": 3.072607260726073,
      "grad_norm": 0.02294921875,
      "learning_rate": 0.029737976699892258,
      "loss": 0.2346,
      "num_input_tokens_seen": 5893024,
      "step": 27930
    },
    {
      "epoch": 3.073157315731573,
      "grad_norm": 0.0033721923828125,
      "learning_rate": 0.029737708647798542,
      "loss": 0.2327,
      "num_input_tokens_seen": 5894048,
      "step": 27935
    },
    {
      "epoch": 3.0737073707370737,
      "grad_norm": 0.0159912109375,
      "learning_rate": 0.029737440459874438,
      "loss": 0.2348,
      "num_input_tokens_seen": 5895104,
      "step": 27940
    },
    {
      "epoch": 3.0742574257425743,
      "grad_norm": 0.01251220703125,
      "learning_rate": 0.029737172136122418,
      "loss": 0.2303,
      "num_input_tokens_seen": 5896096,
      "step": 27945
    },
    {
      "epoch": 3.074807480748075,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.02973690367654495,
      "loss": 0.2346,
      "num_input_tokens_seen": 5897088,
      "step": 27950
    },
    {
      "epoch": 3.075357535753575,
      "grad_norm": 0.0150146484375,
      "learning_rate": 0.029736635081144516,
      "loss": 0.2315,
      "num_input_tokens_seen": 5898208,
      "step": 27955
    },
    {
      "epoch": 3.075907590759076,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.029736366349923587,
      "loss": 0.2382,
      "num_input_tokens_seen": 5899264,
      "step": 27960
    },
    {
      "epoch": 3.0764576457645765,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029736097482884638,
      "loss": 0.2301,
      "num_input_tokens_seen": 5900352,
      "step": 27965
    },
    {
      "epoch": 3.077007700770077,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02973582848003015,
      "loss": 0.2296,
      "num_input_tokens_seen": 5901376,
      "step": 27970
    },
    {
      "epoch": 3.0775577557755778,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.029735559341362607,
      "loss": 0.2327,
      "num_input_tokens_seen": 5902400,
      "step": 27975
    },
    {
      "epoch": 3.078107810781078,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.02973529006688448,
      "loss": 0.2317,
      "num_input_tokens_seen": 5903488,
      "step": 27980
    },
    {
      "epoch": 3.0786578657865786,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029735020656598252,
      "loss": 0.2296,
      "num_input_tokens_seen": 5904576,
      "step": 27985
    },
    {
      "epoch": 3.0792079207920793,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.029734751110506413,
      "loss": 0.2333,
      "num_input_tokens_seen": 5905600,
      "step": 27990
    },
    {
      "epoch": 3.07975797579758,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.029734481428611445,
      "loss": 0.2342,
      "num_input_tokens_seen": 5906560,
      "step": 27995
    },
    {
      "epoch": 3.08030803080308,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029734211610915832,
      "loss": 0.233,
      "num_input_tokens_seen": 5907552,
      "step": 28000
    },
    {
      "epoch": 3.080858085808581,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.02973394165742206,
      "loss": 0.2278,
      "num_input_tokens_seen": 5908640,
      "step": 28005
    },
    {
      "epoch": 3.0814081408140814,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.029733671568132614,
      "loss": 0.2336,
      "num_input_tokens_seen": 5909760,
      "step": 28010
    },
    {
      "epoch": 3.081958195819582,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.029733401343049987,
      "loss": 0.232,
      "num_input_tokens_seen": 5910880,
      "step": 28015
    },
    {
      "epoch": 3.0825082508250823,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.029733130982176673,
      "loss": 0.233,
      "num_input_tokens_seen": 5911904,
      "step": 28020
    },
    {
      "epoch": 3.083058305830583,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.029732860485515163,
      "loss": 0.2325,
      "num_input_tokens_seen": 5912960,
      "step": 28025
    },
    {
      "epoch": 3.0836083608360836,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.029732589853067943,
      "loss": 0.2309,
      "num_input_tokens_seen": 5914016,
      "step": 28030
    },
    {
      "epoch": 3.0841584158415842,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.029732319084837516,
      "loss": 0.2314,
      "num_input_tokens_seen": 5915136,
      "step": 28035
    },
    {
      "epoch": 3.084708470847085,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.02973204818082637,
      "loss": 0.2319,
      "num_input_tokens_seen": 5916192,
      "step": 28040
    },
    {
      "epoch": 3.085258525852585,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.029731777141037006,
      "loss": 0.2309,
      "num_input_tokens_seen": 5917280,
      "step": 28045
    },
    {
      "epoch": 3.0858085808580857,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.029731505965471916,
      "loss": 0.2319,
      "num_input_tokens_seen": 5918400,
      "step": 28050
    },
    {
      "epoch": 3.0863586358635864,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.02973123465413361,
      "loss": 0.2324,
      "num_input_tokens_seen": 5919424,
      "step": 28055
    },
    {
      "epoch": 3.086908690869087,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.02973096320702458,
      "loss": 0.2309,
      "num_input_tokens_seen": 5920480,
      "step": 28060
    },
    {
      "epoch": 3.0874587458745877,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.029730691624147335,
      "loss": 0.2325,
      "num_input_tokens_seen": 5921536,
      "step": 28065
    },
    {
      "epoch": 3.088008800880088,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.02973041990550437,
      "loss": 0.231,
      "num_input_tokens_seen": 5922656,
      "step": 28070
    },
    {
      "epoch": 3.0885588558855885,
      "grad_norm": 0.003387451171875,
      "learning_rate": 0.029730148051098196,
      "loss": 0.2327,
      "num_input_tokens_seen": 5923712,
      "step": 28075
    },
    {
      "epoch": 3.089108910891089,
      "grad_norm": 0.008544921875,
      "learning_rate": 0.029729876060931314,
      "loss": 0.2275,
      "num_input_tokens_seen": 5924768,
      "step": 28080
    },
    {
      "epoch": 3.08965896589659,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.02972960393500623,
      "loss": 0.2329,
      "num_input_tokens_seen": 5925792,
      "step": 28085
    },
    {
      "epoch": 3.09020902090209,
      "grad_norm": 0.01202392578125,
      "learning_rate": 0.029729331673325456,
      "loss": 0.238,
      "num_input_tokens_seen": 5926816,
      "step": 28090
    },
    {
      "epoch": 3.0907590759075907,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.0297290592758915,
      "loss": 0.228,
      "num_input_tokens_seen": 5927808,
      "step": 28095
    },
    {
      "epoch": 3.0913091309130913,
      "grad_norm": 0.0189208984375,
      "learning_rate": 0.029728786742706875,
      "loss": 0.2374,
      "num_input_tokens_seen": 5928928,
      "step": 28100
    },
    {
      "epoch": 3.091859185918592,
      "grad_norm": 0.00323486328125,
      "learning_rate": 0.029728514073774085,
      "loss": 0.2294,
      "num_input_tokens_seen": 5930016,
      "step": 28105
    },
    {
      "epoch": 3.092409240924092,
      "grad_norm": 0.00897216796875,
      "learning_rate": 0.02972824126909565,
      "loss": 0.2335,
      "num_input_tokens_seen": 5931040,
      "step": 28110
    },
    {
      "epoch": 3.092959295929593,
      "grad_norm": 0.0147705078125,
      "learning_rate": 0.02972796832867409,
      "loss": 0.2294,
      "num_input_tokens_seen": 5932096,
      "step": 28115
    },
    {
      "epoch": 3.0935093509350935,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.029727695252511906,
      "loss": 0.232,
      "num_input_tokens_seen": 5933120,
      "step": 28120
    },
    {
      "epoch": 3.094059405940594,
      "grad_norm": 0.00830078125,
      "learning_rate": 0.02972742204061162,
      "loss": 0.2309,
      "num_input_tokens_seen": 5934208,
      "step": 28125
    },
    {
      "epoch": 3.094609460946095,
      "grad_norm": 0.007537841796875,
      "learning_rate": 0.029727148692975755,
      "loss": 0.2341,
      "num_input_tokens_seen": 5935232,
      "step": 28130
    },
    {
      "epoch": 3.095159515951595,
      "grad_norm": 0.00750732421875,
      "learning_rate": 0.029726875209606828,
      "loss": 0.2278,
      "num_input_tokens_seen": 5936288,
      "step": 28135
    },
    {
      "epoch": 3.0957095709570956,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.02972660159050736,
      "loss": 0.2289,
      "num_input_tokens_seen": 5937312,
      "step": 28140
    },
    {
      "epoch": 3.0962596259625963,
      "grad_norm": 0.0084228515625,
      "learning_rate": 0.029726327835679866,
      "loss": 0.2367,
      "num_input_tokens_seen": 5938336,
      "step": 28145
    },
    {
      "epoch": 3.096809680968097,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.029726053945126882,
      "loss": 0.2341,
      "num_input_tokens_seen": 5939392,
      "step": 28150
    },
    {
      "epoch": 3.097359735973597,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.029725779918850916,
      "loss": 0.2293,
      "num_input_tokens_seen": 5940448,
      "step": 28155
    },
    {
      "epoch": 3.097909790979098,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.029725505756854512,
      "loss": 0.2319,
      "num_input_tokens_seen": 5941440,
      "step": 28160
    },
    {
      "epoch": 3.0984598459845984,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.029725231459140182,
      "loss": 0.2335,
      "num_input_tokens_seen": 5942496,
      "step": 28165
    },
    {
      "epoch": 3.099009900990099,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029724957025710458,
      "loss": 0.2308,
      "num_input_tokens_seen": 5943616,
      "step": 28170
    },
    {
      "epoch": 3.0995599559955997,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.029724682456567873,
      "loss": 0.2319,
      "num_input_tokens_seen": 5944736,
      "step": 28175
    },
    {
      "epoch": 3.1001100110011,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.02972440775171496,
      "loss": 0.2314,
      "num_input_tokens_seen": 5945824,
      "step": 28180
    },
    {
      "epoch": 3.1006600660066006,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02972413291115424,
      "loss": 0.2308,
      "num_input_tokens_seen": 5946880,
      "step": 28185
    },
    {
      "epoch": 3.1012101210121013,
      "grad_norm": 0.01300048828125,
      "learning_rate": 0.029723857934888256,
      "loss": 0.2335,
      "num_input_tokens_seen": 5947936,
      "step": 28190
    },
    {
      "epoch": 3.101760176017602,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.029723582822919536,
      "loss": 0.2288,
      "num_input_tokens_seen": 5948896,
      "step": 28195
    },
    {
      "epoch": 3.102310231023102,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.02972330757525062,
      "loss": 0.234,
      "num_input_tokens_seen": 5949984,
      "step": 28200
    },
    {
      "epoch": 3.1028602860286028,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.02972303219188404,
      "loss": 0.2324,
      "num_input_tokens_seen": 5951104,
      "step": 28205
    },
    {
      "epoch": 3.1034103410341034,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.029722756672822342,
      "loss": 0.2308,
      "num_input_tokens_seen": 5952128,
      "step": 28210
    },
    {
      "epoch": 3.103960396039604,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.029722481018068056,
      "loss": 0.2324,
      "num_input_tokens_seen": 5953184,
      "step": 28215
    },
    {
      "epoch": 3.1045104510451047,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029722205227623727,
      "loss": 0.2319,
      "num_input_tokens_seen": 5954240,
      "step": 28220
    },
    {
      "epoch": 3.105060506050605,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029721929301491902,
      "loss": 0.2314,
      "num_input_tokens_seen": 5955264,
      "step": 28225
    },
    {
      "epoch": 3.1056105610561056,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029721653239675114,
      "loss": 0.2329,
      "num_input_tokens_seen": 5956384,
      "step": 28230
    },
    {
      "epoch": 3.106160616061606,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.029721377042175912,
      "loss": 0.2298,
      "num_input_tokens_seen": 5957408,
      "step": 28235
    },
    {
      "epoch": 3.106710671067107,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029721100708996844,
      "loss": 0.2324,
      "num_input_tokens_seen": 5958432,
      "step": 28240
    },
    {
      "epoch": 3.107260726072607,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029720824240140456,
      "loss": 0.2303,
      "num_input_tokens_seen": 5959456,
      "step": 28245
    },
    {
      "epoch": 3.1078107810781077,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.02972054763560929,
      "loss": 0.2298,
      "num_input_tokens_seen": 5960576,
      "step": 28250
    },
    {
      "epoch": 3.1083608360836084,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029720270895405902,
      "loss": 0.2309,
      "num_input_tokens_seen": 5961632,
      "step": 28255
    },
    {
      "epoch": 3.108910891089109,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.029719994019532844,
      "loss": 0.2295,
      "num_input_tokens_seen": 5962688,
      "step": 28260
    },
    {
      "epoch": 3.1094609460946097,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.029719717007992662,
      "loss": 0.231,
      "num_input_tokens_seen": 5963776,
      "step": 28265
    },
    {
      "epoch": 3.11001100110011,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.02971943986078791,
      "loss": 0.2363,
      "num_input_tokens_seen": 5964800,
      "step": 28270
    },
    {
      "epoch": 3.1105610561056105,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.02971916257792115,
      "loss": 0.2347,
      "num_input_tokens_seen": 5965824,
      "step": 28275
    },
    {
      "epoch": 3.111111111111111,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.029718885159394925,
      "loss": 0.2304,
      "num_input_tokens_seen": 5966912,
      "step": 28280
    },
    {
      "epoch": 3.111661166116612,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.029718607605211802,
      "loss": 0.2289,
      "num_input_tokens_seen": 5967904,
      "step": 28285
    },
    {
      "epoch": 3.112211221122112,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02971832991537433,
      "loss": 0.2306,
      "num_input_tokens_seen": 5968992,
      "step": 28290
    },
    {
      "epoch": 3.1127612761276127,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029718052089885077,
      "loss": 0.2362,
      "num_input_tokens_seen": 5970048,
      "step": 28295
    },
    {
      "epoch": 3.1133113311331133,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.029717774128746603,
      "loss": 0.2268,
      "num_input_tokens_seen": 5971104,
      "step": 28300
    },
    {
      "epoch": 3.113861386138614,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.02971749603196146,
      "loss": 0.2284,
      "num_input_tokens_seen": 5972160,
      "step": 28305
    },
    {
      "epoch": 3.1144114411441146,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.029717217799532226,
      "loss": 0.2321,
      "num_input_tokens_seen": 5973280,
      "step": 28310
    },
    {
      "epoch": 3.114961496149615,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.02971693943146145,
      "loss": 0.2333,
      "num_input_tokens_seen": 5974400,
      "step": 28315
    },
    {
      "epoch": 3.1155115511551155,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02971666092775171,
      "loss": 0.2306,
      "num_input_tokens_seen": 5975424,
      "step": 28320
    },
    {
      "epoch": 3.116061606160616,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029716382288405573,
      "loss": 0.2325,
      "num_input_tokens_seen": 5976480,
      "step": 28325
    },
    {
      "epoch": 3.1166116611661168,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.029716103513425594,
      "loss": 0.2248,
      "num_input_tokens_seen": 5977472,
      "step": 28330
    },
    {
      "epoch": 3.117161716171617,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.029715824602814356,
      "loss": 0.2312,
      "num_input_tokens_seen": 5978624,
      "step": 28335
    },
    {
      "epoch": 3.1177117711771176,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02971554555657442,
      "loss": 0.2343,
      "num_input_tokens_seen": 5979648,
      "step": 28340
    },
    {
      "epoch": 3.1182618261826183,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029715266374708362,
      "loss": 0.229,
      "num_input_tokens_seen": 5980704,
      "step": 28345
    },
    {
      "epoch": 3.118811881188119,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.029714987057218756,
      "loss": 0.2306,
      "num_input_tokens_seen": 5981728,
      "step": 28350
    },
    {
      "epoch": 3.1193619361936196,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.029714707604108173,
      "loss": 0.227,
      "num_input_tokens_seen": 5982752,
      "step": 28355
    },
    {
      "epoch": 3.1199119911991198,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.029714428015379192,
      "loss": 0.2297,
      "num_input_tokens_seen": 5983776,
      "step": 28360
    },
    {
      "epoch": 3.1204620462046204,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.02971414829103439,
      "loss": 0.2348,
      "num_input_tokens_seen": 5984864,
      "step": 28365
    },
    {
      "epoch": 3.121012101210121,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029713868431076342,
      "loss": 0.2274,
      "num_input_tokens_seen": 5985920,
      "step": 28370
    },
    {
      "epoch": 3.1215621562156217,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.02971358843550763,
      "loss": 0.2332,
      "num_input_tokens_seen": 5987072,
      "step": 28375
    },
    {
      "epoch": 3.122112211221122,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.029713308304330836,
      "loss": 0.2337,
      "num_input_tokens_seen": 5988128,
      "step": 28380
    },
    {
      "epoch": 3.1226622662266226,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029713028037548536,
      "loss": 0.2284,
      "num_input_tokens_seen": 5989184,
      "step": 28385
    },
    {
      "epoch": 3.1232123212321232,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02971274763516332,
      "loss": 0.2281,
      "num_input_tokens_seen": 5990272,
      "step": 28390
    },
    {
      "epoch": 3.123762376237624,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.02971246709717777,
      "loss": 0.2343,
      "num_input_tokens_seen": 5991328,
      "step": 28395
    },
    {
      "epoch": 3.1243124312431245,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.029712186423594463,
      "loss": 0.227,
      "num_input_tokens_seen": 5992352,
      "step": 28400
    },
    {
      "epoch": 3.1248624862486247,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.029711905614415994,
      "loss": 0.227,
      "num_input_tokens_seen": 5993344,
      "step": 28405
    },
    {
      "epoch": 3.1254125412541254,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.029711624669644957,
      "loss": 0.2312,
      "num_input_tokens_seen": 5994400,
      "step": 28410
    },
    {
      "epoch": 3.125962596259626,
      "grad_norm": 0.007171630859375,
      "learning_rate": 0.029711343589283932,
      "loss": 0.2328,
      "num_input_tokens_seen": 5995520,
      "step": 28415
    },
    {
      "epoch": 3.1265126512651267,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029711062373335515,
      "loss": 0.2256,
      "num_input_tokens_seen": 5996608,
      "step": 28420
    },
    {
      "epoch": 3.127062706270627,
      "grad_norm": 0.0142822265625,
      "learning_rate": 0.029710781021802288,
      "loss": 0.2313,
      "num_input_tokens_seen": 5997600,
      "step": 28425
    },
    {
      "epoch": 3.1276127612761275,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.029710499534686855,
      "loss": 0.2346,
      "num_input_tokens_seen": 5998688,
      "step": 28430
    },
    {
      "epoch": 3.128162816281628,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.02971021791199181,
      "loss": 0.2335,
      "num_input_tokens_seen": 5999776,
      "step": 28435
    },
    {
      "epoch": 3.128712871287129,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.02970993615371974,
      "loss": 0.2286,
      "num_input_tokens_seen": 6000832,
      "step": 28440
    },
    {
      "epoch": 3.129262926292629,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02970965425987325,
      "loss": 0.2359,
      "num_input_tokens_seen": 6001824,
      "step": 28445
    },
    {
      "epoch": 3.1298129812981297,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.029709372230454938,
      "loss": 0.2282,
      "num_input_tokens_seen": 6002912,
      "step": 28450
    },
    {
      "epoch": 3.1303630363036303,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.029709090065467396,
      "loss": 0.2329,
      "num_input_tokens_seen": 6004000,
      "step": 28455
    },
    {
      "epoch": 3.130913091309131,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.02970880776491323,
      "loss": 0.2334,
      "num_input_tokens_seen": 6005024,
      "step": 28460
    },
    {
      "epoch": 3.1314631463146316,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.029708525328795036,
      "loss": 0.2281,
      "num_input_tokens_seen": 6006080,
      "step": 28465
    },
    {
      "epoch": 3.132013201320132,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02970824275711543,
      "loss": 0.2338,
      "num_input_tokens_seen": 6007104,
      "step": 28470
    },
    {
      "epoch": 3.1325632563256325,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.029707960049877,
      "loss": 0.2318,
      "num_input_tokens_seen": 6008160,
      "step": 28475
    },
    {
      "epoch": 3.133113311331133,
      "grad_norm": 0.01171875,
      "learning_rate": 0.029707677207082367,
      "loss": 0.2245,
      "num_input_tokens_seen": 6009216,
      "step": 28480
    },
    {
      "epoch": 3.133663366336634,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029707394228734127,
      "loss": 0.236,
      "num_input_tokens_seen": 6010336,
      "step": 28485
    },
    {
      "epoch": 3.1342134213421344,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.02970711111483489,
      "loss": 0.2307,
      "num_input_tokens_seen": 6011424,
      "step": 28490
    },
    {
      "epoch": 3.1347634763476346,
      "grad_norm": 0.01202392578125,
      "learning_rate": 0.029706827865387274,
      "loss": 0.2254,
      "num_input_tokens_seen": 6012448,
      "step": 28495
    },
    {
      "epoch": 3.1353135313531353,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.029706544480393875,
      "loss": 0.2324,
      "num_input_tokens_seen": 6013472,
      "step": 28500
    },
    {
      "epoch": 3.135863586358636,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029706260959857317,
      "loss": 0.2272,
      "num_input_tokens_seen": 6014528,
      "step": 28505
    },
    {
      "epoch": 3.1364136413641366,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.029705977303780207,
      "loss": 0.2366,
      "num_input_tokens_seen": 6015520,
      "step": 28510
    },
    {
      "epoch": 3.136963696369637,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.02970569351216516,
      "loss": 0.2318,
      "num_input_tokens_seen": 6016544,
      "step": 28515
    },
    {
      "epoch": 3.1375137513751374,
      "grad_norm": 0.007568359375,
      "learning_rate": 0.029705409585014796,
      "loss": 0.2355,
      "num_input_tokens_seen": 6017632,
      "step": 28520
    },
    {
      "epoch": 3.138063806380638,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.029705125522331722,
      "loss": 0.2277,
      "num_input_tokens_seen": 6018752,
      "step": 28525
    },
    {
      "epoch": 3.1386138613861387,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.029704841324118568,
      "loss": 0.2261,
      "num_input_tokens_seen": 6019776,
      "step": 28530
    },
    {
      "epoch": 3.139163916391639,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.029704556990377946,
      "loss": 0.2292,
      "num_input_tokens_seen": 6020896,
      "step": 28535
    },
    {
      "epoch": 3.1397139713971396,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.029704272521112476,
      "loss": 0.2293,
      "num_input_tokens_seen": 6021952,
      "step": 28540
    },
    {
      "epoch": 3.1402640264026402,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.029703987916324785,
      "loss": 0.2298,
      "num_input_tokens_seen": 6023104,
      "step": 28545
    },
    {
      "epoch": 3.140814081408141,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.02970370317601749,
      "loss": 0.2256,
      "num_input_tokens_seen": 6024128,
      "step": 28550
    },
    {
      "epoch": 3.1413641364136415,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029703418300193223,
      "loss": 0.2356,
      "num_input_tokens_seen": 6025152,
      "step": 28555
    },
    {
      "epoch": 3.1419141914191417,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.0297031332888546,
      "loss": 0.2371,
      "num_input_tokens_seen": 6026240,
      "step": 28560
    },
    {
      "epoch": 3.1424642464246424,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.029702848142004253,
      "loss": 0.224,
      "num_input_tokens_seen": 6027328,
      "step": 28565
    },
    {
      "epoch": 3.143014301430143,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029702562859644812,
      "loss": 0.225,
      "num_input_tokens_seen": 6028384,
      "step": 28570
    },
    {
      "epoch": 3.1435643564356437,
      "grad_norm": 0.007781982421875,
      "learning_rate": 0.029702277441778904,
      "loss": 0.2315,
      "num_input_tokens_seen": 6029440,
      "step": 28575
    },
    {
      "epoch": 3.1441144114411443,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02970199188840916,
      "loss": 0.2351,
      "num_input_tokens_seen": 6030496,
      "step": 28580
    },
    {
      "epoch": 3.1446644664466445,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.02970170619953821,
      "loss": 0.236,
      "num_input_tokens_seen": 6031584,
      "step": 28585
    },
    {
      "epoch": 3.145214521452145,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.029701420375168687,
      "loss": 0.2387,
      "num_input_tokens_seen": 6032672,
      "step": 28590
    },
    {
      "epoch": 3.145764576457646,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.02970113441530323,
      "loss": 0.2301,
      "num_input_tokens_seen": 6033664,
      "step": 28595
    },
    {
      "epoch": 3.1463146314631465,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02970084831994447,
      "loss": 0.227,
      "num_input_tokens_seen": 6034752,
      "step": 28600
    },
    {
      "epoch": 3.1468646864686467,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.029700562089095047,
      "loss": 0.2322,
      "num_input_tokens_seen": 6035776,
      "step": 28605
    },
    {
      "epoch": 3.1474147414741473,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029700275722757595,
      "loss": 0.2285,
      "num_input_tokens_seen": 6036832,
      "step": 28610
    },
    {
      "epoch": 3.147964796479648,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.029699989220934756,
      "loss": 0.2364,
      "num_input_tokens_seen": 6037952,
      "step": 28615
    },
    {
      "epoch": 3.1485148514851486,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.029699702583629167,
      "loss": 0.2333,
      "num_input_tokens_seen": 6039040,
      "step": 28620
    },
    {
      "epoch": 3.149064906490649,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.029699415810843476,
      "loss": 0.2332,
      "num_input_tokens_seen": 6040128,
      "step": 28625
    },
    {
      "epoch": 3.1496149614961495,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029699128902580325,
      "loss": 0.2352,
      "num_input_tokens_seen": 6041248,
      "step": 28630
    },
    {
      "epoch": 3.15016501650165,
      "grad_norm": 0.01171875,
      "learning_rate": 0.029698841858842355,
      "loss": 0.2336,
      "num_input_tokens_seen": 6042336,
      "step": 28635
    },
    {
      "epoch": 3.150715071507151,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02969855467963221,
      "loss": 0.2315,
      "num_input_tokens_seen": 6043424,
      "step": 28640
    },
    {
      "epoch": 3.1512651265126514,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029698267364952544,
      "loss": 0.2304,
      "num_input_tokens_seen": 6044448,
      "step": 28645
    },
    {
      "epoch": 3.1518151815181517,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.029697979914805998,
      "loss": 0.232,
      "num_input_tokens_seen": 6045504,
      "step": 28650
    },
    {
      "epoch": 3.1523652365236523,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.029697692329195224,
      "loss": 0.233,
      "num_input_tokens_seen": 6046592,
      "step": 28655
    },
    {
      "epoch": 3.152915291529153,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02969740460812287,
      "loss": 0.2314,
      "num_input_tokens_seen": 6047680,
      "step": 28660
    },
    {
      "epoch": 3.1534653465346536,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.029697116751591594,
      "loss": 0.2298,
      "num_input_tokens_seen": 6048768,
      "step": 28665
    },
    {
      "epoch": 3.1540154015401543,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.029696828759604044,
      "loss": 0.2324,
      "num_input_tokens_seen": 6049824,
      "step": 28670
    },
    {
      "epoch": 3.1545654565456545,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029696540632162878,
      "loss": 0.233,
      "num_input_tokens_seen": 6050848,
      "step": 28675
    },
    {
      "epoch": 3.155115511551155,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.029696252369270745,
      "loss": 0.2325,
      "num_input_tokens_seen": 6051936,
      "step": 28680
    },
    {
      "epoch": 3.1556655665566558,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029695963970930307,
      "loss": 0.2314,
      "num_input_tokens_seen": 6052928,
      "step": 28685
    },
    {
      "epoch": 3.1562156215621564,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02969567543714422,
      "loss": 0.2324,
      "num_input_tokens_seen": 6054016,
      "step": 28690
    },
    {
      "epoch": 3.1567656765676566,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.02969538676791515,
      "loss": 0.2324,
      "num_input_tokens_seen": 6055040,
      "step": 28695
    },
    {
      "epoch": 3.1573157315731573,
      "grad_norm": 0.001953125,
      "learning_rate": 0.029695097963245747,
      "loss": 0.2303,
      "num_input_tokens_seen": 6056096,
      "step": 28700
    },
    {
      "epoch": 3.157865786578658,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029694809023138678,
      "loss": 0.2314,
      "num_input_tokens_seen": 6057152,
      "step": 28705
    },
    {
      "epoch": 3.1584158415841586,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.029694519947596604,
      "loss": 0.2324,
      "num_input_tokens_seen": 6058240,
      "step": 28710
    },
    {
      "epoch": 3.1589658965896588,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.029694230736622196,
      "loss": 0.2324,
      "num_input_tokens_seen": 6059264,
      "step": 28715
    },
    {
      "epoch": 3.1595159515951594,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.029693941390218106,
      "loss": 0.2324,
      "num_input_tokens_seen": 6060256,
      "step": 28720
    },
    {
      "epoch": 3.16006600660066,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.029693651908387013,
      "loss": 0.2329,
      "num_input_tokens_seen": 6061280,
      "step": 28725
    },
    {
      "epoch": 3.1606160616061607,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02969336229113158,
      "loss": 0.2319,
      "num_input_tokens_seen": 6062336,
      "step": 28730
    },
    {
      "epoch": 3.1611661166116614,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029693072538454483,
      "loss": 0.2308,
      "num_input_tokens_seen": 6063328,
      "step": 28735
    },
    {
      "epoch": 3.1617161716171616,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029692782650358383,
      "loss": 0.2303,
      "num_input_tokens_seen": 6064352,
      "step": 28740
    },
    {
      "epoch": 3.162266226622662,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.029692492626845955,
      "loss": 0.2319,
      "num_input_tokens_seen": 6065408,
      "step": 28745
    },
    {
      "epoch": 3.162816281628163,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.02969220246791987,
      "loss": 0.2319,
      "num_input_tokens_seen": 6066496,
      "step": 28750
    },
    {
      "epoch": 3.1633663366336635,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.02969191217358281,
      "loss": 0.2304,
      "num_input_tokens_seen": 6067616,
      "step": 28755
    },
    {
      "epoch": 3.1639163916391637,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.02969162174383744,
      "loss": 0.232,
      "num_input_tokens_seen": 6068672,
      "step": 28760
    },
    {
      "epoch": 3.1644664466446644,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.02969133117868645,
      "loss": 0.2336,
      "num_input_tokens_seen": 6069760,
      "step": 28765
    },
    {
      "epoch": 3.165016501650165,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.0296910404781325,
      "loss": 0.2336,
      "num_input_tokens_seen": 6070816,
      "step": 28770
    },
    {
      "epoch": 3.1655665566556657,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.029690749642178285,
      "loss": 0.2305,
      "num_input_tokens_seen": 6071936,
      "step": 28775
    },
    {
      "epoch": 3.1661166116611663,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.02969045867082648,
      "loss": 0.231,
      "num_input_tokens_seen": 6073024,
      "step": 28780
    },
    {
      "epoch": 3.1666666666666665,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.029690167564079765,
      "loss": 0.2283,
      "num_input_tokens_seen": 6074048,
      "step": 28785
    },
    {
      "epoch": 3.167216721672167,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.029689876321940825,
      "loss": 0.2336,
      "num_input_tokens_seen": 6075072,
      "step": 28790
    },
    {
      "epoch": 3.167766776677668,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.02968958494441235,
      "loss": 0.2294,
      "num_input_tokens_seen": 6076064,
      "step": 28795
    },
    {
      "epoch": 3.1683168316831685,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029689293431497014,
      "loss": 0.2273,
      "num_input_tokens_seen": 6077120,
      "step": 28800
    },
    {
      "epoch": 3.1688668866886687,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.029689001783197506,
      "loss": 0.2306,
      "num_input_tokens_seen": 6078144,
      "step": 28805
    },
    {
      "epoch": 3.1694169416941693,
      "grad_norm": 0.007537841796875,
      "learning_rate": 0.02968870999951652,
      "loss": 0.2321,
      "num_input_tokens_seen": 6079296,
      "step": 28810
    },
    {
      "epoch": 3.16996699669967,
      "grad_norm": 0.0140380859375,
      "learning_rate": 0.02968841808045674,
      "loss": 0.2296,
      "num_input_tokens_seen": 6080352,
      "step": 28815
    },
    {
      "epoch": 3.1705170517051706,
      "grad_norm": 0.007537841796875,
      "learning_rate": 0.02968812602602086,
      "loss": 0.2343,
      "num_input_tokens_seen": 6081408,
      "step": 28820
    },
    {
      "epoch": 3.1710671067106713,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.029687833836211576,
      "loss": 0.2321,
      "num_input_tokens_seen": 6082496,
      "step": 28825
    },
    {
      "epoch": 3.1716171617161715,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02968754151103157,
      "loss": 0.2325,
      "num_input_tokens_seen": 6083584,
      "step": 28830
    },
    {
      "epoch": 3.172167216721672,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.029687249050483543,
      "loss": 0.2288,
      "num_input_tokens_seen": 6084608,
      "step": 28835
    },
    {
      "epoch": 3.1727172717271728,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.029686956454570184,
      "loss": 0.233,
      "num_input_tokens_seen": 6085696,
      "step": 28840
    },
    {
      "epoch": 3.1732673267326734,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0296866637232942,
      "loss": 0.2304,
      "num_input_tokens_seen": 6086784,
      "step": 28845
    },
    {
      "epoch": 3.1738173817381736,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029686370856658286,
      "loss": 0.2299,
      "num_input_tokens_seen": 6087872,
      "step": 28850
    },
    {
      "epoch": 3.1743674367436743,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.029686077854665135,
      "loss": 0.232,
      "num_input_tokens_seen": 6088928,
      "step": 28855
    },
    {
      "epoch": 3.174917491749175,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.029685784717317452,
      "loss": 0.2322,
      "num_input_tokens_seen": 6089952,
      "step": 28860
    },
    {
      "epoch": 3.1754675467546756,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02968549144461794,
      "loss": 0.2306,
      "num_input_tokens_seen": 6091008,
      "step": 28865
    },
    {
      "epoch": 3.1760176017601762,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.0296851980365693,
      "loss": 0.2284,
      "num_input_tokens_seen": 6092096,
      "step": 28870
    },
    {
      "epoch": 3.1765676567656764,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.029684904493174236,
      "loss": 0.23,
      "num_input_tokens_seen": 6093152,
      "step": 28875
    },
    {
      "epoch": 3.177117711771177,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.02968461081443545,
      "loss": 0.2291,
      "num_input_tokens_seen": 6094176,
      "step": 28880
    },
    {
      "epoch": 3.1776677667766777,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029684317000355658,
      "loss": 0.2313,
      "num_input_tokens_seen": 6095296,
      "step": 28885
    },
    {
      "epoch": 3.1782178217821784,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.02968402305093756,
      "loss": 0.2286,
      "num_input_tokens_seen": 6096352,
      "step": 28890
    },
    {
      "epoch": 3.1787678767876786,
      "grad_norm": 0.01416015625,
      "learning_rate": 0.02968372896618387,
      "loss": 0.2225,
      "num_input_tokens_seen": 6097408,
      "step": 28895
    },
    {
      "epoch": 3.1793179317931792,
      "grad_norm": 0.007568359375,
      "learning_rate": 0.029683434746097293,
      "loss": 0.2297,
      "num_input_tokens_seen": 6098400,
      "step": 28900
    },
    {
      "epoch": 3.17986798679868,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.02968314039068054,
      "loss": 0.244,
      "num_input_tokens_seen": 6099456,
      "step": 28905
    },
    {
      "epoch": 3.1804180418041805,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.029682845899936337,
      "loss": 0.2369,
      "num_input_tokens_seen": 6100544,
      "step": 28910
    },
    {
      "epoch": 3.180968096809681,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.02968255127386738,
      "loss": 0.2288,
      "num_input_tokens_seen": 6101568,
      "step": 28915
    },
    {
      "epoch": 3.1815181518151814,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.0296822565124764,
      "loss": 0.2302,
      "num_input_tokens_seen": 6102656,
      "step": 28920
    },
    {
      "epoch": 3.182068206820682,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.0296819616157661,
      "loss": 0.2303,
      "num_input_tokens_seen": 6103680,
      "step": 28925
    },
    {
      "epoch": 3.1826182618261827,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.029681666583739212,
      "loss": 0.229,
      "num_input_tokens_seen": 6104704,
      "step": 28930
    },
    {
      "epoch": 3.1831683168316833,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029681371416398445,
      "loss": 0.2276,
      "num_input_tokens_seen": 6105728,
      "step": 28935
    },
    {
      "epoch": 3.1837183718371835,
      "grad_norm": 0.01519775390625,
      "learning_rate": 0.029681076113746527,
      "loss": 0.2387,
      "num_input_tokens_seen": 6106816,
      "step": 28940
    },
    {
      "epoch": 3.184268426842684,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.02968078067578617,
      "loss": 0.2376,
      "num_input_tokens_seen": 6107808,
      "step": 28945
    },
    {
      "epoch": 3.184818481848185,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.0296804851025201,
      "loss": 0.2321,
      "num_input_tokens_seen": 6108896,
      "step": 28950
    },
    {
      "epoch": 3.1853685368536855,
      "grad_norm": 0.01171875,
      "learning_rate": 0.029680189393951048,
      "loss": 0.2263,
      "num_input_tokens_seen": 6109920,
      "step": 28955
    },
    {
      "epoch": 3.1859185918591857,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.02967989355008173,
      "loss": 0.2326,
      "num_input_tokens_seen": 6110880,
      "step": 28960
    },
    {
      "epoch": 3.1864686468646863,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.029679597570914883,
      "loss": 0.232,
      "num_input_tokens_seen": 6111936,
      "step": 28965
    },
    {
      "epoch": 3.187018701870187,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029679301456453224,
      "loss": 0.2294,
      "num_input_tokens_seen": 6112992,
      "step": 28970
    },
    {
      "epoch": 3.1875687568756876,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.02967900520669949,
      "loss": 0.2336,
      "num_input_tokens_seen": 6114080,
      "step": 28975
    },
    {
      "epoch": 3.1881188118811883,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029678708821656408,
      "loss": 0.2388,
      "num_input_tokens_seen": 6115200,
      "step": 28980
    },
    {
      "epoch": 3.1886688668866885,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.029678412301326708,
      "loss": 0.2298,
      "num_input_tokens_seen": 6116256,
      "step": 28985
    },
    {
      "epoch": 3.189218921892189,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02967811564571313,
      "loss": 0.2303,
      "num_input_tokens_seen": 6117280,
      "step": 28990
    },
    {
      "epoch": 3.18976897689769,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.029677818854818398,
      "loss": 0.2308,
      "num_input_tokens_seen": 6118336,
      "step": 28995
    },
    {
      "epoch": 3.1903190319031904,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.029677521928645256,
      "loss": 0.2303,
      "num_input_tokens_seen": 6119456,
      "step": 29000
    },
    {
      "epoch": 3.190869086908691,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.029677224867196438,
      "loss": 0.2319,
      "num_input_tokens_seen": 6120448,
      "step": 29005
    },
    {
      "epoch": 3.1914191419141913,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.029676927670474678,
      "loss": 0.2345,
      "num_input_tokens_seen": 6121536,
      "step": 29010
    },
    {
      "epoch": 3.191969196919692,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.029676630338482723,
      "loss": 0.234,
      "num_input_tokens_seen": 6122496,
      "step": 29015
    },
    {
      "epoch": 3.1925192519251926,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.029676332871223305,
      "loss": 0.2308,
      "num_input_tokens_seen": 6123584,
      "step": 29020
    },
    {
      "epoch": 3.1930693069306932,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029676035268699174,
      "loss": 0.2287,
      "num_input_tokens_seen": 6124672,
      "step": 29025
    },
    {
      "epoch": 3.1936193619361934,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029675737530913064,
      "loss": 0.2308,
      "num_input_tokens_seen": 6125728,
      "step": 29030
    },
    {
      "epoch": 3.194169416941694,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.02967543965786772,
      "loss": 0.233,
      "num_input_tokens_seen": 6126752,
      "step": 29035
    },
    {
      "epoch": 3.1947194719471947,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0296751416495659,
      "loss": 0.2293,
      "num_input_tokens_seen": 6127936,
      "step": 29040
    },
    {
      "epoch": 3.1952695269526954,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029674843506010334,
      "loss": 0.2319,
      "num_input_tokens_seen": 6129024,
      "step": 29045
    },
    {
      "epoch": 3.1958195819581956,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.02967454522720378,
      "loss": 0.2319,
      "num_input_tokens_seen": 6130080,
      "step": 29050
    },
    {
      "epoch": 3.1963696369636962,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02967424681314898,
      "loss": 0.2319,
      "num_input_tokens_seen": 6131072,
      "step": 29055
    },
    {
      "epoch": 3.196919691969197,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.02967394826384869,
      "loss": 0.2329,
      "num_input_tokens_seen": 6132128,
      "step": 29060
    },
    {
      "epoch": 3.1974697469746975,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029673649579305666,
      "loss": 0.2303,
      "num_input_tokens_seen": 6133216,
      "step": 29065
    },
    {
      "epoch": 3.198019801980198,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.029673350759522647,
      "loss": 0.2303,
      "num_input_tokens_seen": 6134240,
      "step": 29070
    },
    {
      "epoch": 3.1985698569856984,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.0296730518045024,
      "loss": 0.2308,
      "num_input_tokens_seen": 6135296,
      "step": 29075
    },
    {
      "epoch": 3.199119911991199,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.02967275271424767,
      "loss": 0.2319,
      "num_input_tokens_seen": 6136416,
      "step": 29080
    },
    {
      "epoch": 3.1996699669966997,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.029672453488761227,
      "loss": 0.2329,
      "num_input_tokens_seen": 6137408,
      "step": 29085
    },
    {
      "epoch": 3.2002200220022003,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029672154128045816,
      "loss": 0.2313,
      "num_input_tokens_seen": 6138496,
      "step": 29090
    },
    {
      "epoch": 3.200770077007701,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.0296718546321042,
      "loss": 0.2334,
      "num_input_tokens_seen": 6139584,
      "step": 29095
    },
    {
      "epoch": 3.201320132013201,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.02967155500093914,
      "loss": 0.2313,
      "num_input_tokens_seen": 6140608,
      "step": 29100
    },
    {
      "epoch": 3.201870187018702,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0296712552345534,
      "loss": 0.2318,
      "num_input_tokens_seen": 6141632,
      "step": 29105
    },
    {
      "epoch": 3.2024202420242025,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029670955332949743,
      "loss": 0.2313,
      "num_input_tokens_seen": 6142656,
      "step": 29110
    },
    {
      "epoch": 3.202970297029703,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029670655296130926,
      "loss": 0.2313,
      "num_input_tokens_seen": 6143712,
      "step": 29115
    },
    {
      "epoch": 3.2035203520352034,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.029670355124099723,
      "loss": 0.2318,
      "num_input_tokens_seen": 6144800,
      "step": 29120
    },
    {
      "epoch": 3.204070407040704,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.02967005481685889,
      "loss": 0.2318,
      "num_input_tokens_seen": 6145792,
      "step": 29125
    },
    {
      "epoch": 3.2046204620462047,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029669754374411206,
      "loss": 0.2334,
      "num_input_tokens_seen": 6146912,
      "step": 29130
    },
    {
      "epoch": 3.2051705170517053,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02966945379675943,
      "loss": 0.2329,
      "num_input_tokens_seen": 6147904,
      "step": 29135
    },
    {
      "epoch": 3.2057205720572055,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.029669153083906344,
      "loss": 0.2334,
      "num_input_tokens_seen": 6148992,
      "step": 29140
    },
    {
      "epoch": 3.206270627062706,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.02966885223585471,
      "loss": 0.2297,
      "num_input_tokens_seen": 6149984,
      "step": 29145
    },
    {
      "epoch": 3.206820682068207,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029668551252607304,
      "loss": 0.2313,
      "num_input_tokens_seen": 6151072,
      "step": 29150
    },
    {
      "epoch": 3.2073707370737075,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029668250134166897,
      "loss": 0.2297,
      "num_input_tokens_seen": 6152096,
      "step": 29155
    },
    {
      "epoch": 3.207920792079208,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02966794888053627,
      "loss": 0.2308,
      "num_input_tokens_seen": 6153216,
      "step": 29160
    },
    {
      "epoch": 3.2084708470847083,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.0296676474917182,
      "loss": 0.2308,
      "num_input_tokens_seen": 6154240,
      "step": 29165
    },
    {
      "epoch": 3.209020902090209,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02966734596771545,
      "loss": 0.2313,
      "num_input_tokens_seen": 6155232,
      "step": 29170
    },
    {
      "epoch": 3.2095709570957096,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02966704430853082,
      "loss": 0.2313,
      "num_input_tokens_seen": 6156288,
      "step": 29175
    },
    {
      "epoch": 3.2101210121012103,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.02966674251416708,
      "loss": 0.2318,
      "num_input_tokens_seen": 6157344,
      "step": 29180
    },
    {
      "epoch": 3.2106710671067105,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029666440584627012,
      "loss": 0.2303,
      "num_input_tokens_seen": 6158304,
      "step": 29185
    },
    {
      "epoch": 3.211221122112211,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.029666138519913395,
      "loss": 0.2323,
      "num_input_tokens_seen": 6159296,
      "step": 29190
    },
    {
      "epoch": 3.2117711771177118,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029665836320029018,
      "loss": 0.2328,
      "num_input_tokens_seen": 6160352,
      "step": 29195
    },
    {
      "epoch": 3.2123212321232124,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029665533984976664,
      "loss": 0.2323,
      "num_input_tokens_seen": 6161440,
      "step": 29200
    },
    {
      "epoch": 3.212871287128713,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029665231514759124,
      "loss": 0.2287,
      "num_input_tokens_seen": 6162528,
      "step": 29205
    },
    {
      "epoch": 3.2134213421342133,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.02966492890937918,
      "loss": 0.2307,
      "num_input_tokens_seen": 6163520,
      "step": 29210
    },
    {
      "epoch": 3.213971397139714,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029664626168839622,
      "loss": 0.2297,
      "num_input_tokens_seen": 6164512,
      "step": 29215
    },
    {
      "epoch": 3.2145214521452146,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.029664323293143245,
      "loss": 0.2313,
      "num_input_tokens_seen": 6165536,
      "step": 29220
    },
    {
      "epoch": 3.215071507150715,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029664020282292834,
      "loss": 0.2298,
      "num_input_tokens_seen": 6166592,
      "step": 29225
    },
    {
      "epoch": 3.2156215621562154,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029663717136291186,
      "loss": 0.2273,
      "num_input_tokens_seen": 6167616,
      "step": 29230
    },
    {
      "epoch": 3.216171617161716,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.029663413855141093,
      "loss": 0.2291,
      "num_input_tokens_seen": 6168672,
      "step": 29235
    },
    {
      "epoch": 3.2167216721672167,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.029663110438845352,
      "loss": 0.2328,
      "num_input_tokens_seen": 6169696,
      "step": 29240
    },
    {
      "epoch": 3.2172717271727174,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.029662806887406752,
      "loss": 0.2275,
      "num_input_tokens_seen": 6170720,
      "step": 29245
    },
    {
      "epoch": 3.217821782178218,
      "grad_norm": 0.009033203125,
      "learning_rate": 0.0296625032008281,
      "loss": 0.2349,
      "num_input_tokens_seen": 6171776,
      "step": 29250
    },
    {
      "epoch": 3.218371837183718,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.029662199379112197,
      "loss": 0.2293,
      "num_input_tokens_seen": 6172864,
      "step": 29255
    },
    {
      "epoch": 3.218921892189219,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.029661895422261834,
      "loss": 0.2375,
      "num_input_tokens_seen": 6173888,
      "step": 29260
    },
    {
      "epoch": 3.2194719471947195,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.02966159133027982,
      "loss": 0.228,
      "num_input_tokens_seen": 6174944,
      "step": 29265
    },
    {
      "epoch": 3.22002200220022,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.029661287103168947,
      "loss": 0.2347,
      "num_input_tokens_seen": 6175968,
      "step": 29270
    },
    {
      "epoch": 3.2205720572057204,
      "grad_norm": 0.01336669921875,
      "learning_rate": 0.029660982740932026,
      "loss": 0.2378,
      "num_input_tokens_seen": 6176992,
      "step": 29275
    },
    {
      "epoch": 3.221122112211221,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.029660678243571867,
      "loss": 0.233,
      "num_input_tokens_seen": 6178080,
      "step": 29280
    },
    {
      "epoch": 3.2216721672167217,
      "grad_norm": 0.001953125,
      "learning_rate": 0.02966037361109127,
      "loss": 0.2308,
      "num_input_tokens_seen": 6179104,
      "step": 29285
    },
    {
      "epoch": 3.2222222222222223,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029660068843493047,
      "loss": 0.2308,
      "num_input_tokens_seen": 6180224,
      "step": 29290
    },
    {
      "epoch": 3.222772277227723,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029659763940779998,
      "loss": 0.2334,
      "num_input_tokens_seen": 6181216,
      "step": 29295
    },
    {
      "epoch": 3.223322332233223,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.029659458902954947,
      "loss": 0.2335,
      "num_input_tokens_seen": 6182272,
      "step": 29300
    },
    {
      "epoch": 3.223872387238724,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.02965915373002069,
      "loss": 0.2335,
      "num_input_tokens_seen": 6183296,
      "step": 29305
    },
    {
      "epoch": 3.2244224422442245,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02965884842198005,
      "loss": 0.2314,
      "num_input_tokens_seen": 6184384,
      "step": 29310
    },
    {
      "epoch": 3.224972497249725,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02965854297883584,
      "loss": 0.2324,
      "num_input_tokens_seen": 6185472,
      "step": 29315
    },
    {
      "epoch": 3.2255225522552253,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02965823740059087,
      "loss": 0.2308,
      "num_input_tokens_seen": 6186528,
      "step": 29320
    },
    {
      "epoch": 3.226072607260726,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029657931687247962,
      "loss": 0.2309,
      "num_input_tokens_seen": 6187616,
      "step": 29325
    },
    {
      "epoch": 3.2266226622662266,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029657625838809928,
      "loss": 0.2325,
      "num_input_tokens_seen": 6188704,
      "step": 29330
    },
    {
      "epoch": 3.2271727172717273,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02965731985527959,
      "loss": 0.2298,
      "num_input_tokens_seen": 6189728,
      "step": 29335
    },
    {
      "epoch": 3.227722772277228,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029657013736659774,
      "loss": 0.233,
      "num_input_tokens_seen": 6190752,
      "step": 29340
    },
    {
      "epoch": 3.228272827282728,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.029656707482953293,
      "loss": 0.232,
      "num_input_tokens_seen": 6191808,
      "step": 29345
    },
    {
      "epoch": 3.228822882288229,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.029656401094162968,
      "loss": 0.2324,
      "num_input_tokens_seen": 6192864,
      "step": 29350
    },
    {
      "epoch": 3.2293729372937294,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.02965609457029163,
      "loss": 0.2314,
      "num_input_tokens_seen": 6193888,
      "step": 29355
    },
    {
      "epoch": 3.22992299229923,
      "grad_norm": 0.000522613525390625,
      "learning_rate": 0.029655787911342102,
      "loss": 0.2324,
      "num_input_tokens_seen": 6194880,
      "step": 29360
    },
    {
      "epoch": 3.2304730473047303,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.02965548111731721,
      "loss": 0.2313,
      "num_input_tokens_seen": 6195936,
      "step": 29365
    },
    {
      "epoch": 3.231023102310231,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02965517418821978,
      "loss": 0.2314,
      "num_input_tokens_seen": 6197024,
      "step": 29370
    },
    {
      "epoch": 3.2315731573157316,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.029654867124052638,
      "loss": 0.2303,
      "num_input_tokens_seen": 6198048,
      "step": 29375
    },
    {
      "epoch": 3.2321232123212322,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02965455992481862,
      "loss": 0.2324,
      "num_input_tokens_seen": 6199136,
      "step": 29380
    },
    {
      "epoch": 3.232673267326733,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.029654252590520555,
      "loss": 0.2293,
      "num_input_tokens_seen": 6200192,
      "step": 29385
    },
    {
      "epoch": 3.233223322332233,
      "grad_norm": 0.0006866455078125,
      "learning_rate": 0.02965394512116128,
      "loss": 0.2303,
      "num_input_tokens_seen": 6201248,
      "step": 29390
    },
    {
      "epoch": 3.2337733773377337,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.02965363751674362,
      "loss": 0.2308,
      "num_input_tokens_seen": 6202304,
      "step": 29395
    },
    {
      "epoch": 3.2343234323432344,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.029653329777270414,
      "loss": 0.2319,
      "num_input_tokens_seen": 6203360,
      "step": 29400
    },
    {
      "epoch": 3.234873487348735,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.029653021902744502,
      "loss": 0.2308,
      "num_input_tokens_seen": 6204352,
      "step": 29405
    },
    {
      "epoch": 3.2354235423542352,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029652713893168717,
      "loss": 0.2303,
      "num_input_tokens_seen": 6205312,
      "step": 29410
    },
    {
      "epoch": 3.235973597359736,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0296524057485459,
      "loss": 0.2324,
      "num_input_tokens_seen": 6206400,
      "step": 29415
    },
    {
      "epoch": 3.2365236523652365,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.02965209746887889,
      "loss": 0.2329,
      "num_input_tokens_seen": 6207488,
      "step": 29420
    },
    {
      "epoch": 3.237073707370737,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.029651789054170527,
      "loss": 0.2334,
      "num_input_tokens_seen": 6208512,
      "step": 29425
    },
    {
      "epoch": 3.237623762376238,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029651480504423653,
      "loss": 0.2298,
      "num_input_tokens_seen": 6209632,
      "step": 29430
    },
    {
      "epoch": 3.238173817381738,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02965117181964112,
      "loss": 0.2318,
      "num_input_tokens_seen": 6210656,
      "step": 29435
    },
    {
      "epoch": 3.2387238723872387,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.02965086299982576,
      "loss": 0.2319,
      "num_input_tokens_seen": 6211648,
      "step": 29440
    },
    {
      "epoch": 3.2392739273927393,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.029650554044980428,
      "loss": 0.2308,
      "num_input_tokens_seen": 6212768,
      "step": 29445
    },
    {
      "epoch": 3.23982398239824,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029650244955107968,
      "loss": 0.2319,
      "num_input_tokens_seen": 6213824,
      "step": 29450
    },
    {
      "epoch": 3.24037403740374,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029649935730211234,
      "loss": 0.2324,
      "num_input_tokens_seen": 6214912,
      "step": 29455
    },
    {
      "epoch": 3.240924092409241,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.029649626370293072,
      "loss": 0.2304,
      "num_input_tokens_seen": 6215968,
      "step": 29460
    },
    {
      "epoch": 3.2414741474147415,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02964931687535633,
      "loss": 0.2335,
      "num_input_tokens_seen": 6216992,
      "step": 29465
    },
    {
      "epoch": 3.242024202420242,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.029649007245403865,
      "loss": 0.2303,
      "num_input_tokens_seen": 6218080,
      "step": 29470
    },
    {
      "epoch": 3.2425742574257423,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02964869748043853,
      "loss": 0.2298,
      "num_input_tokens_seen": 6219136,
      "step": 29475
    },
    {
      "epoch": 3.243124312431243,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.029648387580463183,
      "loss": 0.2303,
      "num_input_tokens_seen": 6220256,
      "step": 29480
    },
    {
      "epoch": 3.2436743674367436,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02964807754548067,
      "loss": 0.2308,
      "num_input_tokens_seen": 6221376,
      "step": 29485
    },
    {
      "epoch": 3.2442244224422443,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029647767375493857,
      "loss": 0.2319,
      "num_input_tokens_seen": 6222432,
      "step": 29490
    },
    {
      "epoch": 3.244774477447745,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029647457070505603,
      "loss": 0.2303,
      "num_input_tokens_seen": 6223520,
      "step": 29495
    },
    {
      "epoch": 3.245324532453245,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.029647146630518765,
      "loss": 0.2314,
      "num_input_tokens_seen": 6224576,
      "step": 29500
    },
    {
      "epoch": 3.245874587458746,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.0296468360555362,
      "loss": 0.2303,
      "num_input_tokens_seen": 6225664,
      "step": 29505
    },
    {
      "epoch": 3.2464246424642464,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02964652534556078,
      "loss": 0.2303,
      "num_input_tokens_seen": 6226752,
      "step": 29510
    },
    {
      "epoch": 3.246974697469747,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029646214500595364,
      "loss": 0.2308,
      "num_input_tokens_seen": 6227744,
      "step": 29515
    },
    {
      "epoch": 3.2475247524752477,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.02964590352064282,
      "loss": 0.2319,
      "num_input_tokens_seen": 6228768,
      "step": 29520
    },
    {
      "epoch": 3.248074807480748,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029645592405706005,
      "loss": 0.2314,
      "num_input_tokens_seen": 6229824,
      "step": 29525
    },
    {
      "epoch": 3.2486248624862486,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029645281155787795,
      "loss": 0.2324,
      "num_input_tokens_seen": 6230816,
      "step": 29530
    },
    {
      "epoch": 3.2491749174917492,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029644969770891056,
      "loss": 0.2308,
      "num_input_tokens_seen": 6231904,
      "step": 29535
    },
    {
      "epoch": 3.24972497249725,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.029644658251018658,
      "loss": 0.2298,
      "num_input_tokens_seen": 6232928,
      "step": 29540
    },
    {
      "epoch": 3.25027502750275,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.029644346596173474,
      "loss": 0.2308,
      "num_input_tokens_seen": 6233952,
      "step": 29545
    },
    {
      "epoch": 3.2508250825082508,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.02964403480635837,
      "loss": 0.2298,
      "num_input_tokens_seen": 6234944,
      "step": 29550
    },
    {
      "epoch": 3.2513751375137514,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.029643722881576226,
      "loss": 0.2324,
      "num_input_tokens_seen": 6235936,
      "step": 29555
    },
    {
      "epoch": 3.251925192519252,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.029643410821829915,
      "loss": 0.2298,
      "num_input_tokens_seen": 6237088,
      "step": 29560
    },
    {
      "epoch": 3.2524752475247523,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.029643098627122316,
      "loss": 0.2298,
      "num_input_tokens_seen": 6238176,
      "step": 29565
    },
    {
      "epoch": 3.253025302530253,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0296427862974563,
      "loss": 0.2319,
      "num_input_tokens_seen": 6239200,
      "step": 29570
    },
    {
      "epoch": 3.2535753575357536,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02964247383283475,
      "loss": 0.2308,
      "num_input_tokens_seen": 6240288,
      "step": 29575
    },
    {
      "epoch": 3.254125412541254,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029642161233260543,
      "loss": 0.2319,
      "num_input_tokens_seen": 6241344,
      "step": 29580
    },
    {
      "epoch": 3.254675467546755,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029641848498736564,
      "loss": 0.2308,
      "num_input_tokens_seen": 6242368,
      "step": 29585
    },
    {
      "epoch": 3.255225522552255,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.029641535629265692,
      "loss": 0.2313,
      "num_input_tokens_seen": 6243392,
      "step": 29590
    },
    {
      "epoch": 3.2557755775577557,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029641222624850812,
      "loss": 0.2308,
      "num_input_tokens_seen": 6244480,
      "step": 29595
    },
    {
      "epoch": 3.2563256325632564,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029640909485494814,
      "loss": 0.2308,
      "num_input_tokens_seen": 6245600,
      "step": 29600
    },
    {
      "epoch": 3.256875687568757,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029640596211200573,
      "loss": 0.2298,
      "num_input_tokens_seen": 6246656,
      "step": 29605
    },
    {
      "epoch": 3.2574257425742577,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029640282801970983,
      "loss": 0.2319,
      "num_input_tokens_seen": 6247744,
      "step": 29610
    },
    {
      "epoch": 3.257975797579758,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029639969257808933,
      "loss": 0.2313,
      "num_input_tokens_seen": 6248832,
      "step": 29615
    },
    {
      "epoch": 3.2585258525852585,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029639655578717308,
      "loss": 0.2319,
      "num_input_tokens_seen": 6249888,
      "step": 29620
    },
    {
      "epoch": 3.259075907590759,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029639341764699,
      "loss": 0.2324,
      "num_input_tokens_seen": 6250912,
      "step": 29625
    },
    {
      "epoch": 3.25962596259626,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02963902781575691,
      "loss": 0.2308,
      "num_input_tokens_seen": 6251936,
      "step": 29630
    },
    {
      "epoch": 3.26017601760176,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029638713731893924,
      "loss": 0.2303,
      "num_input_tokens_seen": 6253056,
      "step": 29635
    },
    {
      "epoch": 3.2607260726072607,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029638399513112937,
      "loss": 0.2319,
      "num_input_tokens_seen": 6254144,
      "step": 29640
    },
    {
      "epoch": 3.2612761276127613,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029638085159416842,
      "loss": 0.2303,
      "num_input_tokens_seen": 6255200,
      "step": 29645
    },
    {
      "epoch": 3.261826182618262,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02963777067080855,
      "loss": 0.2329,
      "num_input_tokens_seen": 6256288,
      "step": 29650
    },
    {
      "epoch": 3.262376237623762,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02963745604729094,
      "loss": 0.2319,
      "num_input_tokens_seen": 6257312,
      "step": 29655
    },
    {
      "epoch": 3.262926292629263,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.029637141288866926,
      "loss": 0.2324,
      "num_input_tokens_seen": 6258368,
      "step": 29660
    },
    {
      "epoch": 3.2634763476347635,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.0296368263955394,
      "loss": 0.2329,
      "num_input_tokens_seen": 6259392,
      "step": 29665
    },
    {
      "epoch": 3.264026402640264,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029636511367311276,
      "loss": 0.2308,
      "num_input_tokens_seen": 6260384,
      "step": 29670
    },
    {
      "epoch": 3.2645764576457648,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.029636196204185443,
      "loss": 0.2288,
      "num_input_tokens_seen": 6261376,
      "step": 29675
    },
    {
      "epoch": 3.265126512651265,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029635880906164817,
      "loss": 0.2313,
      "num_input_tokens_seen": 6262400,
      "step": 29680
    },
    {
      "epoch": 3.2656765676567656,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029635565473252294,
      "loss": 0.2293,
      "num_input_tokens_seen": 6263488,
      "step": 29685
    },
    {
      "epoch": 3.2662266226622663,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029635249905450793,
      "loss": 0.2345,
      "num_input_tokens_seen": 6264544,
      "step": 29690
    },
    {
      "epoch": 3.266776677667767,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.029634934202763214,
      "loss": 0.233,
      "num_input_tokens_seen": 6265568,
      "step": 29695
    },
    {
      "epoch": 3.2673267326732676,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.02963461836519247,
      "loss": 0.2282,
      "num_input_tokens_seen": 6266560,
      "step": 29700
    },
    {
      "epoch": 3.2678767876787678,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029634302392741466,
      "loss": 0.2314,
      "num_input_tokens_seen": 6267584,
      "step": 29705
    },
    {
      "epoch": 3.2684268426842684,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029633986285413125,
      "loss": 0.234,
      "num_input_tokens_seen": 6268608,
      "step": 29710
    },
    {
      "epoch": 3.268976897689769,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.029633670043210356,
      "loss": 0.2267,
      "num_input_tokens_seen": 6269632,
      "step": 29715
    },
    {
      "epoch": 3.2695269526952697,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029633353666136064,
      "loss": 0.2314,
      "num_input_tokens_seen": 6270752,
      "step": 29720
    },
    {
      "epoch": 3.27007700770077,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.029633037154193177,
      "loss": 0.235,
      "num_input_tokens_seen": 6271808,
      "step": 29725
    },
    {
      "epoch": 3.2706270627062706,
      "grad_norm": 0.000530242919921875,
      "learning_rate": 0.02963272050738461,
      "loss": 0.2319,
      "num_input_tokens_seen": 6272864,
      "step": 29730
    },
    {
      "epoch": 3.271177117711771,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.02963240372571328,
      "loss": 0.2319,
      "num_input_tokens_seen": 6273920,
      "step": 29735
    },
    {
      "epoch": 3.271727172717272,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0296320868091821,
      "loss": 0.2319,
      "num_input_tokens_seen": 6274976,
      "step": 29740
    },
    {
      "epoch": 3.272277227722772,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029631769757794007,
      "loss": 0.2308,
      "num_input_tokens_seen": 6276096,
      "step": 29745
    },
    {
      "epoch": 3.2728272827282727,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029631452571551906,
      "loss": 0.2319,
      "num_input_tokens_seen": 6277088,
      "step": 29750
    },
    {
      "epoch": 3.2733773377337734,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02963113525045873,
      "loss": 0.2313,
      "num_input_tokens_seen": 6278176,
      "step": 29755
    },
    {
      "epoch": 3.273927392739274,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.029630817794517397,
      "loss": 0.2329,
      "num_input_tokens_seen": 6279296,
      "step": 29760
    },
    {
      "epoch": 3.2744774477447747,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029630500203730842,
      "loss": 0.2324,
      "num_input_tokens_seen": 6280352,
      "step": 29765
    },
    {
      "epoch": 3.275027502750275,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029630182478101987,
      "loss": 0.2309,
      "num_input_tokens_seen": 6281440,
      "step": 29770
    },
    {
      "epoch": 3.2755775577557755,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029629864617633755,
      "loss": 0.2313,
      "num_input_tokens_seen": 6282496,
      "step": 29775
    },
    {
      "epoch": 3.276127612761276,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.029629546622329087,
      "loss": 0.233,
      "num_input_tokens_seen": 6283520,
      "step": 29780
    },
    {
      "epoch": 3.276677667766777,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029629228492190903,
      "loss": 0.2325,
      "num_input_tokens_seen": 6284672,
      "step": 29785
    },
    {
      "epoch": 3.2772277227722775,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.029628910227222142,
      "loss": 0.233,
      "num_input_tokens_seen": 6285664,
      "step": 29790
    },
    {
      "epoch": 3.2777777777777777,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029628591827425735,
      "loss": 0.2335,
      "num_input_tokens_seen": 6286688,
      "step": 29795
    },
    {
      "epoch": 3.2783278327832783,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02962827329280462,
      "loss": 0.2303,
      "num_input_tokens_seen": 6287712,
      "step": 29800
    },
    {
      "epoch": 3.278877887788779,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.029627954623361725,
      "loss": 0.2303,
      "num_input_tokens_seen": 6288800,
      "step": 29805
    },
    {
      "epoch": 3.279427942794279,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.02962763581909999,
      "loss": 0.2324,
      "num_input_tokens_seen": 6289888,
      "step": 29810
    },
    {
      "epoch": 3.27997799779978,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.029627316880022358,
      "loss": 0.2319,
      "num_input_tokens_seen": 6290880,
      "step": 29815
    },
    {
      "epoch": 3.2805280528052805,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.029626997806131768,
      "loss": 0.2303,
      "num_input_tokens_seen": 6291872,
      "step": 29820
    },
    {
      "epoch": 3.281078107810781,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.02962667859743115,
      "loss": 0.2308,
      "num_input_tokens_seen": 6292928,
      "step": 29825
    },
    {
      "epoch": 3.281628162816282,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.02962635925392346,
      "loss": 0.2303,
      "num_input_tokens_seen": 6294016,
      "step": 29830
    },
    {
      "epoch": 3.282178217821782,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.02962603977561163,
      "loss": 0.2324,
      "num_input_tokens_seen": 6295040,
      "step": 29835
    },
    {
      "epoch": 3.2827282728272826,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029625720162498616,
      "loss": 0.2314,
      "num_input_tokens_seen": 6296096,
      "step": 29840
    },
    {
      "epoch": 3.2832783278327833,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029625400414587354,
      "loss": 0.2303,
      "num_input_tokens_seen": 6297152,
      "step": 29845
    },
    {
      "epoch": 3.283828382838284,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.029625080531880794,
      "loss": 0.2283,
      "num_input_tokens_seen": 6298144,
      "step": 29850
    },
    {
      "epoch": 3.2843784378437846,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029624760514381884,
      "loss": 0.2325,
      "num_input_tokens_seen": 6299200,
      "step": 29855
    },
    {
      "epoch": 3.284928492849285,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029624440362093576,
      "loss": 0.2309,
      "num_input_tokens_seen": 6300256,
      "step": 29860
    },
    {
      "epoch": 3.2854785478547854,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029624120075018818,
      "loss": 0.2288,
      "num_input_tokens_seen": 6301248,
      "step": 29865
    },
    {
      "epoch": 3.286028602860286,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.029623799653160563,
      "loss": 0.2335,
      "num_input_tokens_seen": 6302336,
      "step": 29870
    },
    {
      "epoch": 3.2865786578657867,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.02962347909652176,
      "loss": 0.2335,
      "num_input_tokens_seen": 6303392,
      "step": 29875
    },
    {
      "epoch": 3.287128712871287,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.029623158405105373,
      "loss": 0.2319,
      "num_input_tokens_seen": 6304448,
      "step": 29880
    },
    {
      "epoch": 3.2876787678767876,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029622837578914353,
      "loss": 0.2319,
      "num_input_tokens_seen": 6305568,
      "step": 29885
    },
    {
      "epoch": 3.2882288228822882,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029622516617951653,
      "loss": 0.2294,
      "num_input_tokens_seen": 6306592,
      "step": 29890
    },
    {
      "epoch": 3.288778877887789,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02962219552222023,
      "loss": 0.2309,
      "num_input_tokens_seen": 6307648,
      "step": 29895
    },
    {
      "epoch": 3.289328932893289,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.029621874291723048,
      "loss": 0.2304,
      "num_input_tokens_seen": 6308736,
      "step": 29900
    },
    {
      "epoch": 3.2898789878987897,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.02962155292646307,
      "loss": 0.2319,
      "num_input_tokens_seen": 6309792,
      "step": 29905
    },
    {
      "epoch": 3.2904290429042904,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029621231426443257,
      "loss": 0.2294,
      "num_input_tokens_seen": 6310816,
      "step": 29910
    },
    {
      "epoch": 3.290979097909791,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.029620909791666563,
      "loss": 0.232,
      "num_input_tokens_seen": 6311872,
      "step": 29915
    },
    {
      "epoch": 3.2915291529152917,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.029620588022135963,
      "loss": 0.2268,
      "num_input_tokens_seen": 6312928,
      "step": 29920
    },
    {
      "epoch": 3.292079207920792,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.02962026611785442,
      "loss": 0.2326,
      "num_input_tokens_seen": 6313952,
      "step": 29925
    },
    {
      "epoch": 3.2926292629262925,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.029619944078824897,
      "loss": 0.2332,
      "num_input_tokens_seen": 6314944,
      "step": 29930
    },
    {
      "epoch": 3.293179317931793,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.029619621905050367,
      "loss": 0.2331,
      "num_input_tokens_seen": 6315968,
      "step": 29935
    },
    {
      "epoch": 3.293729372937294,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.029619299596533796,
      "loss": 0.2305,
      "num_input_tokens_seen": 6316928,
      "step": 29940
    },
    {
      "epoch": 3.2942794279427945,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.02961897715327816,
      "loss": 0.2341,
      "num_input_tokens_seen": 6317984,
      "step": 29945
    },
    {
      "epoch": 3.2948294829482947,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.02961865457528642,
      "loss": 0.232,
      "num_input_tokens_seen": 6319040,
      "step": 29950
    },
    {
      "epoch": 3.2953795379537953,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029618331862561558,
      "loss": 0.2346,
      "num_input_tokens_seen": 6320064,
      "step": 29955
    },
    {
      "epoch": 3.295929592959296,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029618009015106545,
      "loss": 0.2309,
      "num_input_tokens_seen": 6321088,
      "step": 29960
    },
    {
      "epoch": 3.2964796479647966,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.02961768603292436,
      "loss": 0.2314,
      "num_input_tokens_seen": 6322112,
      "step": 29965
    },
    {
      "epoch": 3.297029702970297,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.029617362916017975,
      "loss": 0.2315,
      "num_input_tokens_seen": 6323136,
      "step": 29970
    },
    {
      "epoch": 3.2975797579757975,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.02961703966439037,
      "loss": 0.2325,
      "num_input_tokens_seen": 6324160,
      "step": 29975
    },
    {
      "epoch": 3.298129812981298,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.029616716278044526,
      "loss": 0.2293,
      "num_input_tokens_seen": 6325248,
      "step": 29980
    },
    {
      "epoch": 3.298679867986799,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.02961639275698342,
      "loss": 0.232,
      "num_input_tokens_seen": 6326240,
      "step": 29985
    },
    {
      "epoch": 3.299229922992299,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029616069101210035,
      "loss": 0.2304,
      "num_input_tokens_seen": 6327360,
      "step": 29990
    },
    {
      "epoch": 3.2997799779977997,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.029615745310727356,
      "loss": 0.2304,
      "num_input_tokens_seen": 6328448,
      "step": 29995
    },
    {
      "epoch": 3.3003300330033003,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029615421385538365,
      "loss": 0.2336,
      "num_input_tokens_seen": 6329504,
      "step": 30000
    },
    {
      "epoch": 3.300880088008801,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029615097325646052,
      "loss": 0.2299,
      "num_input_tokens_seen": 6330560,
      "step": 30005
    },
    {
      "epoch": 3.3014301430143016,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.02961477313105339,
      "loss": 0.232,
      "num_input_tokens_seen": 6331648,
      "step": 30010
    },
    {
      "epoch": 3.301980198019802,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029614448801763386,
      "loss": 0.2325,
      "num_input_tokens_seen": 6332736,
      "step": 30015
    },
    {
      "epoch": 3.3025302530253025,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.02961412433777902,
      "loss": 0.231,
      "num_input_tokens_seen": 6333824,
      "step": 30020
    },
    {
      "epoch": 3.303080308030803,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029613799739103275,
      "loss": 0.2284,
      "num_input_tokens_seen": 6334816,
      "step": 30025
    },
    {
      "epoch": 3.3036303630363038,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.029613475005739156,
      "loss": 0.2279,
      "num_input_tokens_seen": 6335936,
      "step": 30030
    },
    {
      "epoch": 3.3041804180418044,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.029613150137689653,
      "loss": 0.2341,
      "num_input_tokens_seen": 6337024,
      "step": 30035
    },
    {
      "epoch": 3.3047304730473046,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029612825134957752,
      "loss": 0.2321,
      "num_input_tokens_seen": 6338112,
      "step": 30040
    },
    {
      "epoch": 3.3052805280528053,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.029612499997546456,
      "loss": 0.2326,
      "num_input_tokens_seen": 6339168,
      "step": 30045
    },
    {
      "epoch": 3.305830583058306,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.029612174725458757,
      "loss": 0.231,
      "num_input_tokens_seen": 6340192,
      "step": 30050
    },
    {
      "epoch": 3.3063806380638066,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02961184931869766,
      "loss": 0.2357,
      "num_input_tokens_seen": 6341216,
      "step": 30055
    },
    {
      "epoch": 3.3069306930693068,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029611523777266155,
      "loss": 0.232,
      "num_input_tokens_seen": 6342272,
      "step": 30060
    },
    {
      "epoch": 3.3074807480748074,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029611198101167246,
      "loss": 0.2304,
      "num_input_tokens_seen": 6343328,
      "step": 30065
    },
    {
      "epoch": 3.308030803080308,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.029610872290403937,
      "loss": 0.2356,
      "num_input_tokens_seen": 6344352,
      "step": 30070
    },
    {
      "epoch": 3.3085808580858087,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02961054634497923,
      "loss": 0.2319,
      "num_input_tokens_seen": 6345344,
      "step": 30075
    },
    {
      "epoch": 3.309130913091309,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.029610220264896125,
      "loss": 0.2308,
      "num_input_tokens_seen": 6346304,
      "step": 30080
    },
    {
      "epoch": 3.3096809680968096,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.029609894050157636,
      "loss": 0.2303,
      "num_input_tokens_seen": 6347360,
      "step": 30085
    },
    {
      "epoch": 3.31023102310231,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.029609567700766757,
      "loss": 0.2293,
      "num_input_tokens_seen": 6348448,
      "step": 30090
    },
    {
      "epoch": 3.310781078107811,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02960924121672651,
      "loss": 0.2334,
      "num_input_tokens_seen": 6349536,
      "step": 30095
    },
    {
      "epoch": 3.3113311331133115,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029608914598039892,
      "loss": 0.2293,
      "num_input_tokens_seen": 6350560,
      "step": 30100
    },
    {
      "epoch": 3.3118811881188117,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.02960858784470992,
      "loss": 0.2314,
      "num_input_tokens_seen": 6351584,
      "step": 30105
    },
    {
      "epoch": 3.3124312431243124,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.0296082609567396,
      "loss": 0.2345,
      "num_input_tokens_seen": 6352640,
      "step": 30110
    },
    {
      "epoch": 3.312981298129813,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.029607933934131952,
      "loss": 0.2324,
      "num_input_tokens_seen": 6353792,
      "step": 30115
    },
    {
      "epoch": 3.3135313531353137,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029607606776889988,
      "loss": 0.234,
      "num_input_tokens_seen": 6354816,
      "step": 30120
    },
    {
      "epoch": 3.3140814081408143,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02960727948501672,
      "loss": 0.2329,
      "num_input_tokens_seen": 6355872,
      "step": 30125
    },
    {
      "epoch": 3.3146314631463145,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.029606952058515167,
      "loss": 0.2308,
      "num_input_tokens_seen": 6356896,
      "step": 30130
    },
    {
      "epoch": 3.315181518151815,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029606624497388344,
      "loss": 0.2329,
      "num_input_tokens_seen": 6357952,
      "step": 30135
    },
    {
      "epoch": 3.315731573157316,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.029606296801639273,
      "loss": 0.2313,
      "num_input_tokens_seen": 6359072,
      "step": 30140
    },
    {
      "epoch": 3.3162816281628165,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.029605968971270977,
      "loss": 0.2313,
      "num_input_tokens_seen": 6360128,
      "step": 30145
    },
    {
      "epoch": 3.3168316831683167,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.029605641006286466,
      "loss": 0.2308,
      "num_input_tokens_seen": 6361152,
      "step": 30150
    },
    {
      "epoch": 3.3173817381738173,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.029605312906688774,
      "loss": 0.2308,
      "num_input_tokens_seen": 6362208,
      "step": 30155
    },
    {
      "epoch": 3.317931793179318,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.029604984672480922,
      "loss": 0.2308,
      "num_input_tokens_seen": 6363232,
      "step": 30160
    },
    {
      "epoch": 3.3184818481848186,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.029604656303665936,
      "loss": 0.2314,
      "num_input_tokens_seen": 6364288,
      "step": 30165
    },
    {
      "epoch": 3.319031903190319,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.02960432780024684,
      "loss": 0.2303,
      "num_input_tokens_seen": 6365280,
      "step": 30170
    },
    {
      "epoch": 3.3195819581958195,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029603999162226658,
      "loss": 0.2334,
      "num_input_tokens_seen": 6366336,
      "step": 30175
    },
    {
      "epoch": 3.32013201320132,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029603670389608427,
      "loss": 0.2304,
      "num_input_tokens_seen": 6367360,
      "step": 30180
    },
    {
      "epoch": 3.3206820682068208,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.029603341482395178,
      "loss": 0.2304,
      "num_input_tokens_seen": 6368320,
      "step": 30185
    },
    {
      "epoch": 3.3212321232123214,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02960301244058993,
      "loss": 0.233,
      "num_input_tokens_seen": 6369440,
      "step": 30190
    },
    {
      "epoch": 3.3217821782178216,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.029602683264195725,
      "loss": 0.2324,
      "num_input_tokens_seen": 6370560,
      "step": 30195
    },
    {
      "epoch": 3.3223322332233223,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0296023539532156,
      "loss": 0.232,
      "num_input_tokens_seen": 6371648,
      "step": 30200
    },
    {
      "epoch": 3.322882288228823,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029602024507652584,
      "loss": 0.2309,
      "num_input_tokens_seen": 6372736,
      "step": 30205
    },
    {
      "epoch": 3.3234323432343236,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.029601694927509713,
      "loss": 0.2314,
      "num_input_tokens_seen": 6373792,
      "step": 30210
    },
    {
      "epoch": 3.323982398239824,
      "grad_norm": 0.001953125,
      "learning_rate": 0.029601365212790026,
      "loss": 0.2314,
      "num_input_tokens_seen": 6374784,
      "step": 30215
    },
    {
      "epoch": 3.3245324532453244,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.02960103536349656,
      "loss": 0.2277,
      "num_input_tokens_seen": 6375872,
      "step": 30220
    },
    {
      "epoch": 3.325082508250825,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029600705379632365,
      "loss": 0.2309,
      "num_input_tokens_seen": 6376928,
      "step": 30225
    },
    {
      "epoch": 3.3256325632563257,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02960037526120047,
      "loss": 0.2309,
      "num_input_tokens_seen": 6377952,
      "step": 30230
    },
    {
      "epoch": 3.3261826182618264,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.02960004500820392,
      "loss": 0.2294,
      "num_input_tokens_seen": 6379008,
      "step": 30235
    },
    {
      "epoch": 3.3267326732673266,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.029599714620645763,
      "loss": 0.2325,
      "num_input_tokens_seen": 6380032,
      "step": 30240
    },
    {
      "epoch": 3.3272827282728272,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.029599384098529042,
      "loss": 0.2299,
      "num_input_tokens_seen": 6381120,
      "step": 30245
    },
    {
      "epoch": 3.327832783278328,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0295990534418568,
      "loss": 0.2289,
      "num_input_tokens_seen": 6382208,
      "step": 30250
    },
    {
      "epoch": 3.3283828382838285,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.029598722650632098,
      "loss": 0.2337,
      "num_input_tokens_seen": 6383232,
      "step": 30255
    },
    {
      "epoch": 3.3289328932893287,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.029598391724857964,
      "loss": 0.228,
      "num_input_tokens_seen": 6384256,
      "step": 30260
    },
    {
      "epoch": 3.3294829482948294,
      "grad_norm": 0.007598876953125,
      "learning_rate": 0.029598060664537462,
      "loss": 0.2312,
      "num_input_tokens_seen": 6385344,
      "step": 30265
    },
    {
      "epoch": 3.33003300330033,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.029597729469673643,
      "loss": 0.2292,
      "num_input_tokens_seen": 6386336,
      "step": 30270
    },
    {
      "epoch": 3.3305830583058307,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02959739814026955,
      "loss": 0.2348,
      "num_input_tokens_seen": 6387424,
      "step": 30275
    },
    {
      "epoch": 3.3311331133113313,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.02959706667632825,
      "loss": 0.2316,
      "num_input_tokens_seen": 6388416,
      "step": 30280
    },
    {
      "epoch": 3.3316831683168315,
      "grad_norm": 0.001953125,
      "learning_rate": 0.029596735077852782,
      "loss": 0.2307,
      "num_input_tokens_seen": 6389504,
      "step": 30285
    },
    {
      "epoch": 3.332233223322332,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.029596403344846214,
      "loss": 0.2341,
      "num_input_tokens_seen": 6390560,
      "step": 30290
    },
    {
      "epoch": 3.332783278327833,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.029596071477311605,
      "loss": 0.2305,
      "num_input_tokens_seen": 6391584,
      "step": 30295
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029595739475252004,
      "loss": 0.2289,
      "num_input_tokens_seen": 6392640,
      "step": 30300
    },
    {
      "epoch": 3.333883388338834,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.029595407338670478,
      "loss": 0.23,
      "num_input_tokens_seen": 6393696,
      "step": 30305
    },
    {
      "epoch": 3.3344334433443343,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.029595075067570087,
      "loss": 0.2311,
      "num_input_tokens_seen": 6394688,
      "step": 30310
    },
    {
      "epoch": 3.334983498349835,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02959474266195389,
      "loss": 0.2254,
      "num_input_tokens_seen": 6395808,
      "step": 30315
    },
    {
      "epoch": 3.3355335533553356,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029594410121824958,
      "loss": 0.2255,
      "num_input_tokens_seen": 6396832,
      "step": 30320
    },
    {
      "epoch": 3.336083608360836,
      "grad_norm": 0.0172119140625,
      "learning_rate": 0.029594077447186347,
      "loss": 0.2377,
      "num_input_tokens_seen": 6397888,
      "step": 30325
    },
    {
      "epoch": 3.3366336633663365,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.029593744638041125,
      "loss": 0.2372,
      "num_input_tokens_seen": 6398944,
      "step": 30330
    },
    {
      "epoch": 3.337183718371837,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.029593411694392367,
      "loss": 0.2334,
      "num_input_tokens_seen": 6400032,
      "step": 30335
    },
    {
      "epoch": 3.337733773377338,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.02959307861624313,
      "loss": 0.2332,
      "num_input_tokens_seen": 6401088,
      "step": 30340
    },
    {
      "epoch": 3.3382838283828384,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.029592745403596497,
      "loss": 0.231,
      "num_input_tokens_seen": 6402144,
      "step": 30345
    },
    {
      "epoch": 3.3388338833883386,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.029592412056455526,
      "loss": 0.2321,
      "num_input_tokens_seen": 6403200,
      "step": 30350
    },
    {
      "epoch": 3.3393839383938393,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.029592078574823297,
      "loss": 0.228,
      "num_input_tokens_seen": 6404288,
      "step": 30355
    },
    {
      "epoch": 3.33993399339934,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029591744958702884,
      "loss": 0.2312,
      "num_input_tokens_seen": 6405312,
      "step": 30360
    },
    {
      "epoch": 3.3404840484048406,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029591411208097357,
      "loss": 0.2285,
      "num_input_tokens_seen": 6406432,
      "step": 30365
    },
    {
      "epoch": 3.3410341034103412,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.029591077323009793,
      "loss": 0.2347,
      "num_input_tokens_seen": 6407520,
      "step": 30370
    },
    {
      "epoch": 3.3415841584158414,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.029590743303443273,
      "loss": 0.2297,
      "num_input_tokens_seen": 6408544,
      "step": 30375
    },
    {
      "epoch": 3.342134213421342,
      "grad_norm": 0.01251220703125,
      "learning_rate": 0.029590409149400875,
      "loss": 0.2328,
      "num_input_tokens_seen": 6409568,
      "step": 30380
    },
    {
      "epoch": 3.3426842684268427,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.029590074860885676,
      "loss": 0.241,
      "num_input_tokens_seen": 6410624,
      "step": 30385
    },
    {
      "epoch": 3.3432343234323434,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.029589740437900757,
      "loss": 0.2326,
      "num_input_tokens_seen": 6411680,
      "step": 30390
    },
    {
      "epoch": 3.3437843784378436,
      "grad_norm": 0.000522613525390625,
      "learning_rate": 0.029589405880449202,
      "loss": 0.2325,
      "num_input_tokens_seen": 6412672,
      "step": 30395
    },
    {
      "epoch": 3.3443344334433442,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.029589071188534093,
      "loss": 0.2309,
      "num_input_tokens_seen": 6413792,
      "step": 30400
    },
    {
      "epoch": 3.344884488448845,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.029588736362158512,
      "loss": 0.232,
      "num_input_tokens_seen": 6414880,
      "step": 30405
    },
    {
      "epoch": 3.3454345434543455,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.029588401401325553,
      "loss": 0.2309,
      "num_input_tokens_seen": 6415936,
      "step": 30410
    },
    {
      "epoch": 3.3459845984598457,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0295880663060383,
      "loss": 0.233,
      "num_input_tokens_seen": 6417024,
      "step": 30415
    },
    {
      "epoch": 3.3465346534653464,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.029587731076299837,
      "loss": 0.2304,
      "num_input_tokens_seen": 6418080,
      "step": 30420
    },
    {
      "epoch": 3.347084708470847,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.029587395712113256,
      "loss": 0.2304,
      "num_input_tokens_seen": 6419168,
      "step": 30425
    },
    {
      "epoch": 3.3476347634763477,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.029587060213481644,
      "loss": 0.2319,
      "num_input_tokens_seen": 6420192,
      "step": 30430
    },
    {
      "epoch": 3.3481848184818483,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.029586724580408105,
      "loss": 0.2309,
      "num_input_tokens_seen": 6421216,
      "step": 30435
    },
    {
      "epoch": 3.3487348734873486,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.029586388812895725,
      "loss": 0.2319,
      "num_input_tokens_seen": 6422272,
      "step": 30440
    },
    {
      "epoch": 3.349284928492849,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.029586052910947593,
      "loss": 0.2309,
      "num_input_tokens_seen": 6423296,
      "step": 30445
    },
    {
      "epoch": 3.34983498349835,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.029585716874566813,
      "loss": 0.2309,
      "num_input_tokens_seen": 6424352,
      "step": 30450
    },
    {
      "epoch": 3.3503850385038505,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.02958538070375648,
      "loss": 0.2324,
      "num_input_tokens_seen": 6425344,
      "step": 30455
    },
    {
      "epoch": 3.350935093509351,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02958504439851969,
      "loss": 0.2314,
      "num_input_tokens_seen": 6426464,
      "step": 30460
    },
    {
      "epoch": 3.3514851485148514,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02958470795885955,
      "loss": 0.2299,
      "num_input_tokens_seen": 6427488,
      "step": 30465
    },
    {
      "epoch": 3.352035203520352,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.029584371384779148,
      "loss": 0.2324,
      "num_input_tokens_seen": 6428512,
      "step": 30470
    },
    {
      "epoch": 3.3525852585258527,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0295840346762816,
      "loss": 0.2288,
      "num_input_tokens_seen": 6429536,
      "step": 30475
    },
    {
      "epoch": 3.3531353135313533,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.029583697833369997,
      "loss": 0.2298,
      "num_input_tokens_seen": 6430656,
      "step": 30480
    },
    {
      "epoch": 3.3536853685368535,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02958336085604745,
      "loss": 0.2309,
      "num_input_tokens_seen": 6431712,
      "step": 30485
    },
    {
      "epoch": 3.354235423542354,
      "grad_norm": 0.001953125,
      "learning_rate": 0.029583023744317063,
      "loss": 0.2288,
      "num_input_tokens_seen": 6432768,
      "step": 30490
    },
    {
      "epoch": 3.354785478547855,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029582686498181947,
      "loss": 0.2314,
      "num_input_tokens_seen": 6433856,
      "step": 30495
    },
    {
      "epoch": 3.3553355335533555,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.029582349117645203,
      "loss": 0.2309,
      "num_input_tokens_seen": 6434976,
      "step": 30500
    },
    {
      "epoch": 3.3558855885588557,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02958201160270995,
      "loss": 0.2314,
      "num_input_tokens_seen": 6436000,
      "step": 30505
    },
    {
      "epoch": 3.3564356435643563,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.029581673953379288,
      "loss": 0.2319,
      "num_input_tokens_seen": 6437088,
      "step": 30510
    },
    {
      "epoch": 3.356985698569857,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029581336169656337,
      "loss": 0.2298,
      "num_input_tokens_seen": 6438080,
      "step": 30515
    },
    {
      "epoch": 3.3575357535753576,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.02958099825154421,
      "loss": 0.2319,
      "num_input_tokens_seen": 6439168,
      "step": 30520
    },
    {
      "epoch": 3.3580858085808583,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.029580660199046014,
      "loss": 0.2314,
      "num_input_tokens_seen": 6440160,
      "step": 30525
    },
    {
      "epoch": 3.3586358635863585,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.029580322012164872,
      "loss": 0.232,
      "num_input_tokens_seen": 6441280,
      "step": 30530
    },
    {
      "epoch": 3.359185918591859,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.029579983690903897,
      "loss": 0.232,
      "num_input_tokens_seen": 6442368,
      "step": 30535
    },
    {
      "epoch": 3.3597359735973598,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.029579645235266207,
      "loss": 0.2288,
      "num_input_tokens_seen": 6443456,
      "step": 30540
    },
    {
      "epoch": 3.3602860286028604,
      "grad_norm": 0.0006866455078125,
      "learning_rate": 0.029579306645254926,
      "loss": 0.2304,
      "num_input_tokens_seen": 6444480,
      "step": 30545
    },
    {
      "epoch": 3.360836083608361,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02957896792087317,
      "loss": 0.2257,
      "num_input_tokens_seen": 6445568,
      "step": 30550
    },
    {
      "epoch": 3.3613861386138613,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.029578629062124065,
      "loss": 0.2304,
      "num_input_tokens_seen": 6446656,
      "step": 30555
    },
    {
      "epoch": 3.361936193619362,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.029578290069010724,
      "loss": 0.2305,
      "num_input_tokens_seen": 6447712,
      "step": 30560
    },
    {
      "epoch": 3.3624862486248626,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029577950941536285,
      "loss": 0.2304,
      "num_input_tokens_seen": 6448736,
      "step": 30565
    },
    {
      "epoch": 3.363036303630363,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.029577611679703866,
      "loss": 0.2363,
      "num_input_tokens_seen": 6449760,
      "step": 30570
    },
    {
      "epoch": 3.3635863586358634,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.029577272283516594,
      "loss": 0.2341,
      "num_input_tokens_seen": 6450816,
      "step": 30575
    },
    {
      "epoch": 3.364136413641364,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0295769327529776,
      "loss": 0.2331,
      "num_input_tokens_seen": 6451904,
      "step": 30580
    },
    {
      "epoch": 3.3646864686468647,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.029576593088090013,
      "loss": 0.2362,
      "num_input_tokens_seen": 6452960,
      "step": 30585
    },
    {
      "epoch": 3.3652365236523654,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029576253288856956,
      "loss": 0.232,
      "num_input_tokens_seen": 6454080,
      "step": 30590
    },
    {
      "epoch": 3.3657865786578656,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029575913355281574,
      "loss": 0.2304,
      "num_input_tokens_seen": 6455168,
      "step": 30595
    },
    {
      "epoch": 3.366336633663366,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02957557328736699,
      "loss": 0.2309,
      "num_input_tokens_seen": 6456160,
      "step": 30600
    },
    {
      "epoch": 3.366886688668867,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.029575233085116337,
      "loss": 0.2304,
      "num_input_tokens_seen": 6457216,
      "step": 30605
    },
    {
      "epoch": 3.3674367436743675,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.02957489274853276,
      "loss": 0.2299,
      "num_input_tokens_seen": 6458240,
      "step": 30610
    },
    {
      "epoch": 3.367986798679868,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02957455227761939,
      "loss": 0.233,
      "num_input_tokens_seen": 6459296,
      "step": 30615
    },
    {
      "epoch": 3.3685368536853684,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.02957421167237936,
      "loss": 0.2325,
      "num_input_tokens_seen": 6460384,
      "step": 30620
    },
    {
      "epoch": 3.369086908690869,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02957387093281582,
      "loss": 0.234,
      "num_input_tokens_seen": 6461408,
      "step": 30625
    },
    {
      "epoch": 3.3696369636963697,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.029573530058931898,
      "loss": 0.2304,
      "num_input_tokens_seen": 6462432,
      "step": 30630
    },
    {
      "epoch": 3.3701870187018703,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.02957318905073075,
      "loss": 0.2324,
      "num_input_tokens_seen": 6463424,
      "step": 30635
    },
    {
      "epoch": 3.370737073707371,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02957284790821551,
      "loss": 0.233,
      "num_input_tokens_seen": 6464512,
      "step": 30640
    },
    {
      "epoch": 3.371287128712871,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.02957250663138932,
      "loss": 0.2293,
      "num_input_tokens_seen": 6465504,
      "step": 30645
    },
    {
      "epoch": 3.371837183718372,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02957216522025533,
      "loss": 0.2325,
      "num_input_tokens_seen": 6466592,
      "step": 30650
    },
    {
      "epoch": 3.3723872387238725,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02957182367481669,
      "loss": 0.2299,
      "num_input_tokens_seen": 6467648,
      "step": 30655
    },
    {
      "epoch": 3.372937293729373,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02957148199507654,
      "loss": 0.2309,
      "num_input_tokens_seen": 6468704,
      "step": 30660
    },
    {
      "epoch": 3.3734873487348733,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02957114018103803,
      "loss": 0.2304,
      "num_input_tokens_seen": 6469760,
      "step": 30665
    },
    {
      "epoch": 3.374037403740374,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.029570798232704316,
      "loss": 0.2314,
      "num_input_tokens_seen": 6470848,
      "step": 30670
    },
    {
      "epoch": 3.3745874587458746,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.02957045615007855,
      "loss": 0.2335,
      "num_input_tokens_seen": 6471936,
      "step": 30675
    },
    {
      "epoch": 3.3751375137513753,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.029570113933163875,
      "loss": 0.2288,
      "num_input_tokens_seen": 6472992,
      "step": 30680
    },
    {
      "epoch": 3.3756875687568755,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.029569771581963455,
      "loss": 0.2294,
      "num_input_tokens_seen": 6474048,
      "step": 30685
    },
    {
      "epoch": 3.376237623762376,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029569429096480436,
      "loss": 0.232,
      "num_input_tokens_seen": 6475200,
      "step": 30690
    },
    {
      "epoch": 3.3767876787678768,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.029569086476717986,
      "loss": 0.2304,
      "num_input_tokens_seen": 6476320,
      "step": 30695
    },
    {
      "epoch": 3.3773377337733774,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029568743722679255,
      "loss": 0.2314,
      "num_input_tokens_seen": 6477344,
      "step": 30700
    },
    {
      "epoch": 3.377887788778878,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.029568400834367403,
      "loss": 0.2283,
      "num_input_tokens_seen": 6478368,
      "step": 30705
    },
    {
      "epoch": 3.3784378437843783,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.029568057811785592,
      "loss": 0.232,
      "num_input_tokens_seen": 6479392,
      "step": 30710
    },
    {
      "epoch": 3.378987898789879,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.02956771465493698,
      "loss": 0.2331,
      "num_input_tokens_seen": 6480448,
      "step": 30715
    },
    {
      "epoch": 3.3795379537953796,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02956737136382474,
      "loss": 0.2283,
      "num_input_tokens_seen": 6481504,
      "step": 30720
    },
    {
      "epoch": 3.3800880088008802,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.02956702793845202,
      "loss": 0.2319,
      "num_input_tokens_seen": 6482624,
      "step": 30725
    },
    {
      "epoch": 3.380638063806381,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029566684378822,
      "loss": 0.2346,
      "num_input_tokens_seen": 6483712,
      "step": 30730
    },
    {
      "epoch": 3.381188118811881,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.029566340684937836,
      "loss": 0.2314,
      "num_input_tokens_seen": 6484832,
      "step": 30735
    },
    {
      "epoch": 3.3817381738173817,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0295659968568027,
      "loss": 0.2335,
      "num_input_tokens_seen": 6485920,
      "step": 30740
    },
    {
      "epoch": 3.3822882288228824,
      "grad_norm": 0.0006256103515625,
      "learning_rate": 0.02956565289441976,
      "loss": 0.2319,
      "num_input_tokens_seen": 6487008,
      "step": 30745
    },
    {
      "epoch": 3.382838283828383,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.029565308797792185,
      "loss": 0.2278,
      "num_input_tokens_seen": 6488032,
      "step": 30750
    },
    {
      "epoch": 3.3833883388338832,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02956496456692315,
      "loss": 0.2304,
      "num_input_tokens_seen": 6489120,
      "step": 30755
    },
    {
      "epoch": 3.383938393839384,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.029564620201815824,
      "loss": 0.2299,
      "num_input_tokens_seen": 6490240,
      "step": 30760
    },
    {
      "epoch": 3.3844884488448845,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029564275702473386,
      "loss": 0.2278,
      "num_input_tokens_seen": 6491328,
      "step": 30765
    },
    {
      "epoch": 3.385038503850385,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029563931068899006,
      "loss": 0.231,
      "num_input_tokens_seen": 6492320,
      "step": 30770
    },
    {
      "epoch": 3.3855885588558854,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02956358630109586,
      "loss": 0.2299,
      "num_input_tokens_seen": 6493408,
      "step": 30775
    },
    {
      "epoch": 3.386138613861386,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.02956324139906713,
      "loss": 0.2325,
      "num_input_tokens_seen": 6494464,
      "step": 30780
    },
    {
      "epoch": 3.3866886688668867,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02956289636281599,
      "loss": 0.233,
      "num_input_tokens_seen": 6495520,
      "step": 30785
    },
    {
      "epoch": 3.3872387238723873,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029562551192345618,
      "loss": 0.2299,
      "num_input_tokens_seen": 6496544,
      "step": 30790
    },
    {
      "epoch": 3.387788778877888,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.029562205887659208,
      "loss": 0.234,
      "num_input_tokens_seen": 6497568,
      "step": 30795
    },
    {
      "epoch": 3.388338833883388,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.02956186044875993,
      "loss": 0.2324,
      "num_input_tokens_seen": 6498592,
      "step": 30800
    },
    {
      "epoch": 3.388888888888889,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.029561514875650972,
      "loss": 0.2309,
      "num_input_tokens_seen": 6499584,
      "step": 30805
    },
    {
      "epoch": 3.3894389438943895,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.02956116916833552,
      "loss": 0.2304,
      "num_input_tokens_seen": 6500640,
      "step": 30810
    },
    {
      "epoch": 3.38998899889989,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.029560823326816758,
      "loss": 0.233,
      "num_input_tokens_seen": 6501728,
      "step": 30815
    },
    {
      "epoch": 3.390539053905391,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02956047735109787,
      "loss": 0.2356,
      "num_input_tokens_seen": 6502720,
      "step": 30820
    },
    {
      "epoch": 3.391089108910891,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.029560131241182057,
      "loss": 0.2298,
      "num_input_tokens_seen": 6503808,
      "step": 30825
    },
    {
      "epoch": 3.3916391639163916,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029559784997072493,
      "loss": 0.2314,
      "num_input_tokens_seen": 6504832,
      "step": 30830
    },
    {
      "epoch": 3.3921892189218923,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.029559438618772384,
      "loss": 0.2293,
      "num_input_tokens_seen": 6505856,
      "step": 30835
    },
    {
      "epoch": 3.3927392739273925,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02955909210628491,
      "loss": 0.2303,
      "num_input_tokens_seen": 6506912,
      "step": 30840
    },
    {
      "epoch": 3.393289328932893,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.029558745459613272,
      "loss": 0.2293,
      "num_input_tokens_seen": 6507936,
      "step": 30845
    },
    {
      "epoch": 3.393839383938394,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029558398678760665,
      "loss": 0.2314,
      "num_input_tokens_seen": 6509024,
      "step": 30850
    },
    {
      "epoch": 3.3943894389438944,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.029558051763730282,
      "loss": 0.2319,
      "num_input_tokens_seen": 6510080,
      "step": 30855
    },
    {
      "epoch": 3.394939493949395,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.02955770471452532,
      "loss": 0.2314,
      "num_input_tokens_seen": 6511104,
      "step": 30860
    },
    {
      "epoch": 3.3954895489548953,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.029557357531148982,
      "loss": 0.233,
      "num_input_tokens_seen": 6512064,
      "step": 30865
    },
    {
      "epoch": 3.396039603960396,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.029557010213604462,
      "loss": 0.2309,
      "num_input_tokens_seen": 6513152,
      "step": 30870
    },
    {
      "epoch": 3.3965896589658966,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.029556662761894965,
      "loss": 0.2325,
      "num_input_tokens_seen": 6514240,
      "step": 30875
    },
    {
      "epoch": 3.3971397139713972,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029556315176023695,
      "loss": 0.2309,
      "num_input_tokens_seen": 6515296,
      "step": 30880
    },
    {
      "epoch": 3.397689768976898,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.029555967455993853,
      "loss": 0.2319,
      "num_input_tokens_seen": 6516352,
      "step": 30885
    },
    {
      "epoch": 3.398239823982398,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.02955561960180864,
      "loss": 0.2294,
      "num_input_tokens_seen": 6517408,
      "step": 30890
    },
    {
      "epoch": 3.3987898789878987,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.029555271613471264,
      "loss": 0.233,
      "num_input_tokens_seen": 6518400,
      "step": 30895
    },
    {
      "epoch": 3.3993399339933994,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.029554923490984937,
      "loss": 0.2325,
      "num_input_tokens_seen": 6519488,
      "step": 30900
    },
    {
      "epoch": 3.3998899889989,
      "grad_norm": 0.000797271728515625,
      "learning_rate": 0.029554575234352867,
      "loss": 0.2325,
      "num_input_tokens_seen": 6520544,
      "step": 30905
    },
    {
      "epoch": 3.4004400440044003,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.029554226843578255,
      "loss": 0.234,
      "num_input_tokens_seen": 6521600,
      "step": 30910
    },
    {
      "epoch": 3.400990099009901,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.02955387831866432,
      "loss": 0.2303,
      "num_input_tokens_seen": 6522592,
      "step": 30915
    },
    {
      "epoch": 3.4015401540154016,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.029553529659614275,
      "loss": 0.2303,
      "num_input_tokens_seen": 6523584,
      "step": 30920
    },
    {
      "epoch": 3.402090209020902,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029553180866431328,
      "loss": 0.2304,
      "num_input_tokens_seen": 6524576,
      "step": 30925
    },
    {
      "epoch": 3.4026402640264024,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.029552831939118698,
      "loss": 0.2314,
      "num_input_tokens_seen": 6525696,
      "step": 30930
    },
    {
      "epoch": 3.403190319031903,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.029552482877679597,
      "loss": 0.2299,
      "num_input_tokens_seen": 6526784,
      "step": 30935
    },
    {
      "epoch": 3.4037403740374037,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.029552133682117246,
      "loss": 0.2298,
      "num_input_tokens_seen": 6527904,
      "step": 30940
    },
    {
      "epoch": 3.4042904290429044,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.029551784352434864,
      "loss": 0.234,
      "num_input_tokens_seen": 6528928,
      "step": 30945
    },
    {
      "epoch": 3.404840484048405,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.029551434888635666,
      "loss": 0.2304,
      "num_input_tokens_seen": 6529984,
      "step": 30950
    },
    {
      "epoch": 3.405390539053905,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029551085290722874,
      "loss": 0.2299,
      "num_input_tokens_seen": 6531072,
      "step": 30955
    },
    {
      "epoch": 3.405940594059406,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029550735558699706,
      "loss": 0.2293,
      "num_input_tokens_seen": 6532128,
      "step": 30960
    },
    {
      "epoch": 3.4064906490649065,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.029550385692569396,
      "loss": 0.2319,
      "num_input_tokens_seen": 6533152,
      "step": 30965
    },
    {
      "epoch": 3.407040704070407,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.029550035692335167,
      "loss": 0.233,
      "num_input_tokens_seen": 6534208,
      "step": 30970
    },
    {
      "epoch": 3.407590759075908,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029549685558000235,
      "loss": 0.2304,
      "num_input_tokens_seen": 6535232,
      "step": 30975
    },
    {
      "epoch": 3.408140814081408,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.029549335289567834,
      "loss": 0.2319,
      "num_input_tokens_seen": 6536288,
      "step": 30980
    },
    {
      "epoch": 3.4086908690869087,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.02954898488704119,
      "loss": 0.2314,
      "num_input_tokens_seen": 6537344,
      "step": 30985
    },
    {
      "epoch": 3.4092409240924093,
      "grad_norm": 0.00079345703125,
      "learning_rate": 0.029548634350423538,
      "loss": 0.233,
      "num_input_tokens_seen": 6538400,
      "step": 30990
    },
    {
      "epoch": 3.40979097909791,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.029548283679718103,
      "loss": 0.2319,
      "num_input_tokens_seen": 6539488,
      "step": 30995
    },
    {
      "epoch": 3.41034103410341,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.029547932874928115,
      "loss": 0.2319,
      "num_input_tokens_seen": 6540544,
      "step": 31000
    },
    {
      "epoch": 3.410891089108911,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029547581936056812,
      "loss": 0.2308,
      "num_input_tokens_seen": 6541632,
      "step": 31005
    },
    {
      "epoch": 3.4114411441144115,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.029547230863107427,
      "loss": 0.2313,
      "num_input_tokens_seen": 6542656,
      "step": 31010
    },
    {
      "epoch": 3.411991199119912,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029546879656083196,
      "loss": 0.2314,
      "num_input_tokens_seen": 6543744,
      "step": 31015
    },
    {
      "epoch": 3.4125412541254123,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029546528314987358,
      "loss": 0.2298,
      "num_input_tokens_seen": 6544736,
      "step": 31020
    },
    {
      "epoch": 3.413091309130913,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029546176839823145,
      "loss": 0.2314,
      "num_input_tokens_seen": 6545760,
      "step": 31025
    },
    {
      "epoch": 3.4136413641364136,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029545825230593802,
      "loss": 0.2319,
      "num_input_tokens_seen": 6546880,
      "step": 31030
    },
    {
      "epoch": 3.4141914191419143,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.029545473487302566,
      "loss": 0.2309,
      "num_input_tokens_seen": 6547968,
      "step": 31035
    },
    {
      "epoch": 3.414741474147415,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02954512160995268,
      "loss": 0.2329,
      "num_input_tokens_seen": 6549024,
      "step": 31040
    },
    {
      "epoch": 3.415291529152915,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.02954476959854739,
      "loss": 0.2324,
      "num_input_tokens_seen": 6550016,
      "step": 31045
    },
    {
      "epoch": 3.4158415841584158,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029544417453089934,
      "loss": 0.2313,
      "num_input_tokens_seen": 6551072,
      "step": 31050
    },
    {
      "epoch": 3.4163916391639164,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029544065173583565,
      "loss": 0.2304,
      "num_input_tokens_seen": 6552064,
      "step": 31055
    },
    {
      "epoch": 3.416941694169417,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029543712760031524,
      "loss": 0.2319,
      "num_input_tokens_seen": 6553120,
      "step": 31060
    },
    {
      "epoch": 3.4174917491749177,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.029543360212437066,
      "loss": 0.2319,
      "num_input_tokens_seen": 6554208,
      "step": 31065
    },
    {
      "epoch": 3.418041804180418,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02954300753080343,
      "loss": 0.2304,
      "num_input_tokens_seen": 6555264,
      "step": 31070
    },
    {
      "epoch": 3.4185918591859186,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029542654715133874,
      "loss": 0.2314,
      "num_input_tokens_seen": 6556320,
      "step": 31075
    },
    {
      "epoch": 3.419141914191419,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.029542301765431645,
      "loss": 0.2309,
      "num_input_tokens_seen": 6557344,
      "step": 31080
    },
    {
      "epoch": 3.41969196919692,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029541948681700004,
      "loss": 0.2303,
      "num_input_tokens_seen": 6558368,
      "step": 31085
    },
    {
      "epoch": 3.42024202420242,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029541595463942198,
      "loss": 0.2309,
      "num_input_tokens_seen": 6559456,
      "step": 31090
    },
    {
      "epoch": 3.4207920792079207,
      "grad_norm": 0.000766754150390625,
      "learning_rate": 0.029541242112161484,
      "loss": 0.2299,
      "num_input_tokens_seen": 6560544,
      "step": 31095
    },
    {
      "epoch": 3.4213421342134214,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.029540888626361117,
      "loss": 0.2337,
      "num_input_tokens_seen": 6561632,
      "step": 31100
    },
    {
      "epoch": 3.421892189218922,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029540535006544355,
      "loss": 0.2333,
      "num_input_tokens_seen": 6562720,
      "step": 31105
    },
    {
      "epoch": 3.4224422442244222,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.029540181252714466,
      "loss": 0.2295,
      "num_input_tokens_seen": 6563808,
      "step": 31110
    },
    {
      "epoch": 3.422992299229923,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0295398273648747,
      "loss": 0.2286,
      "num_input_tokens_seen": 6564928,
      "step": 31115
    },
    {
      "epoch": 3.4235423542354235,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.029539473343028325,
      "loss": 0.2321,
      "num_input_tokens_seen": 6565984,
      "step": 31120
    },
    {
      "epoch": 3.424092409240924,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029539119187178595,
      "loss": 0.228,
      "num_input_tokens_seen": 6567040,
      "step": 31125
    },
    {
      "epoch": 3.424642464246425,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.029538764897328786,
      "loss": 0.2322,
      "num_input_tokens_seen": 6568096,
      "step": 31130
    },
    {
      "epoch": 3.425192519251925,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.02953841047348215,
      "loss": 0.2307,
      "num_input_tokens_seen": 6569088,
      "step": 31135
    },
    {
      "epoch": 3.4257425742574257,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.029538055915641966,
      "loss": 0.2332,
      "num_input_tokens_seen": 6570208,
      "step": 31140
    },
    {
      "epoch": 3.4262926292629263,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.029537701223811497,
      "loss": 0.2363,
      "num_input_tokens_seen": 6571264,
      "step": 31145
    },
    {
      "epoch": 3.426842684268427,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.029537346397994012,
      "loss": 0.2351,
      "num_input_tokens_seen": 6572320,
      "step": 31150
    },
    {
      "epoch": 3.4273927392739276,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.029536991438192776,
      "loss": 0.2319,
      "num_input_tokens_seen": 6573344,
      "step": 31155
    },
    {
      "epoch": 3.427942794279428,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029536636344411067,
      "loss": 0.2309,
      "num_input_tokens_seen": 6574432,
      "step": 31160
    },
    {
      "epoch": 3.4284928492849285,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02953628111665216,
      "loss": 0.233,
      "num_input_tokens_seen": 6575488,
      "step": 31165
    },
    {
      "epoch": 3.429042904290429,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029535925754919326,
      "loss": 0.2314,
      "num_input_tokens_seen": 6576576,
      "step": 31170
    },
    {
      "epoch": 3.4295929592959298,
      "grad_norm": 0.00081634521484375,
      "learning_rate": 0.029535570259215837,
      "loss": 0.2309,
      "num_input_tokens_seen": 6577632,
      "step": 31175
    },
    {
      "epoch": 3.43014301430143,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02953521462954497,
      "loss": 0.2309,
      "num_input_tokens_seen": 6578720,
      "step": 31180
    },
    {
      "epoch": 3.4306930693069306,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029534858865910004,
      "loss": 0.2309,
      "num_input_tokens_seen": 6579776,
      "step": 31185
    },
    {
      "epoch": 3.4312431243124313,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02953450296831422,
      "loss": 0.2299,
      "num_input_tokens_seen": 6580832,
      "step": 31190
    },
    {
      "epoch": 3.431793179317932,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029534146936760895,
      "loss": 0.2325,
      "num_input_tokens_seen": 6581888,
      "step": 31195
    },
    {
      "epoch": 3.432343234323432,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.029533790771253315,
      "loss": 0.2283,
      "num_input_tokens_seen": 6582976,
      "step": 31200
    },
    {
      "epoch": 3.432893289328933,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.029533434471794757,
      "loss": 0.2289,
      "num_input_tokens_seen": 6584064,
      "step": 31205
    },
    {
      "epoch": 3.4334433443344334,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02953307803838851,
      "loss": 0.2373,
      "num_input_tokens_seen": 6585184,
      "step": 31210
    },
    {
      "epoch": 3.433993399339934,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.029532721471037856,
      "loss": 0.231,
      "num_input_tokens_seen": 6586304,
      "step": 31215
    },
    {
      "epoch": 3.4345434543454347,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.029532364769746078,
      "loss": 0.23,
      "num_input_tokens_seen": 6587360,
      "step": 31220
    },
    {
      "epoch": 3.435093509350935,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02953200793451647,
      "loss": 0.2357,
      "num_input_tokens_seen": 6588416,
      "step": 31225
    },
    {
      "epoch": 3.4356435643564356,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.02953165096535232,
      "loss": 0.2325,
      "num_input_tokens_seen": 6589504,
      "step": 31230
    },
    {
      "epoch": 3.4361936193619362,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.029531293862256912,
      "loss": 0.231,
      "num_input_tokens_seen": 6590656,
      "step": 31235
    },
    {
      "epoch": 3.436743674367437,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.029530936625233543,
      "loss": 0.233,
      "num_input_tokens_seen": 6591648,
      "step": 31240
    },
    {
      "epoch": 3.4372937293729375,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.0295305792542855,
      "loss": 0.2288,
      "num_input_tokens_seen": 6592736,
      "step": 31245
    },
    {
      "epoch": 3.4378437843784377,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029530221749416087,
      "loss": 0.2324,
      "num_input_tokens_seen": 6593760,
      "step": 31250
    },
    {
      "epoch": 3.4383938393839384,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.02952986411062859,
      "loss": 0.2299,
      "num_input_tokens_seen": 6594816,
      "step": 31255
    },
    {
      "epoch": 3.438943894389439,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.029529506337926305,
      "loss": 0.2351,
      "num_input_tokens_seen": 6595840,
      "step": 31260
    },
    {
      "epoch": 3.4394939493949397,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.029529148431312536,
      "loss": 0.2341,
      "num_input_tokens_seen": 6596864,
      "step": 31265
    },
    {
      "epoch": 3.44004400440044,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.029528790390790576,
      "loss": 0.2283,
      "num_input_tokens_seen": 6597920,
      "step": 31270
    },
    {
      "epoch": 3.4405940594059405,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.029528432216363722,
      "loss": 0.2325,
      "num_input_tokens_seen": 6599008,
      "step": 31275
    },
    {
      "epoch": 3.441144114411441,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029528073908035288,
      "loss": 0.2289,
      "num_input_tokens_seen": 6600064,
      "step": 31280
    },
    {
      "epoch": 3.441694169416942,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02952771546580856,
      "loss": 0.231,
      "num_input_tokens_seen": 6601120,
      "step": 31285
    },
    {
      "epoch": 3.442244224422442,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.029527356889686853,
      "loss": 0.2294,
      "num_input_tokens_seen": 6602208,
      "step": 31290
    },
    {
      "epoch": 3.4427942794279427,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.029526998179673473,
      "loss": 0.232,
      "num_input_tokens_seen": 6603264,
      "step": 31295
    },
    {
      "epoch": 3.4433443344334433,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029526639335771716,
      "loss": 0.2299,
      "num_input_tokens_seen": 6604320,
      "step": 31300
    },
    {
      "epoch": 3.443894389438944,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.029526280357984897,
      "loss": 0.2315,
      "num_input_tokens_seen": 6605344,
      "step": 31305
    },
    {
      "epoch": 3.4444444444444446,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.029525921246316325,
      "loss": 0.232,
      "num_input_tokens_seen": 6606464,
      "step": 31310
    },
    {
      "epoch": 3.444994499449945,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.0295255620007693,
      "loss": 0.2305,
      "num_input_tokens_seen": 6607552,
      "step": 31315
    },
    {
      "epoch": 3.4455445544554455,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.029525202621347144,
      "loss": 0.2336,
      "num_input_tokens_seen": 6608576,
      "step": 31320
    },
    {
      "epoch": 3.446094609460946,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029524843108053162,
      "loss": 0.2284,
      "num_input_tokens_seen": 6609632,
      "step": 31325
    },
    {
      "epoch": 3.446644664466447,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.02952448346089068,
      "loss": 0.2321,
      "num_input_tokens_seen": 6610752,
      "step": 31330
    },
    {
      "epoch": 3.4471947194719474,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.029524123679862995,
      "loss": 0.2289,
      "num_input_tokens_seen": 6611776,
      "step": 31335
    },
    {
      "epoch": 3.4477447744774476,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02952376376497343,
      "loss": 0.2357,
      "num_input_tokens_seen": 6612864,
      "step": 31340
    },
    {
      "epoch": 3.4482948294829483,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.02952340371622531,
      "loss": 0.2336,
      "num_input_tokens_seen": 6613952,
      "step": 31345
    },
    {
      "epoch": 3.448844884488449,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.029523043533621944,
      "loss": 0.2305,
      "num_input_tokens_seen": 6614976,
      "step": 31350
    },
    {
      "epoch": 3.449394939493949,
      "grad_norm": 0.000728607177734375,
      "learning_rate": 0.029522683217166657,
      "loss": 0.2331,
      "num_input_tokens_seen": 6616032,
      "step": 31355
    },
    {
      "epoch": 3.44994499449945,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02952232276686276,
      "loss": 0.2304,
      "num_input_tokens_seen": 6617152,
      "step": 31360
    },
    {
      "epoch": 3.4504950495049505,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02952196218271359,
      "loss": 0.2335,
      "num_input_tokens_seen": 6618240,
      "step": 31365
    },
    {
      "epoch": 3.451045104510451,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.029521601464722463,
      "loss": 0.2341,
      "num_input_tokens_seen": 6619296,
      "step": 31370
    },
    {
      "epoch": 3.4515951595159517,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.029521240612892703,
      "loss": 0.233,
      "num_input_tokens_seen": 6620352,
      "step": 31375
    },
    {
      "epoch": 3.452145214521452,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02952087962722763,
      "loss": 0.2299,
      "num_input_tokens_seen": 6621408,
      "step": 31380
    },
    {
      "epoch": 3.4526952695269526,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.029520518507730582,
      "loss": 0.2289,
      "num_input_tokens_seen": 6622432,
      "step": 31385
    },
    {
      "epoch": 3.4532453245324533,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.029520157254404884,
      "loss": 0.2341,
      "num_input_tokens_seen": 6623520,
      "step": 31390
    },
    {
      "epoch": 3.453795379537954,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02951979586725386,
      "loss": 0.2299,
      "num_input_tokens_seen": 6624512,
      "step": 31395
    },
    {
      "epoch": 3.4543454345434546,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02951943434628085,
      "loss": 0.2294,
      "num_input_tokens_seen": 6625504,
      "step": 31400
    },
    {
      "epoch": 3.4548954895489548,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029519072691489175,
      "loss": 0.2314,
      "num_input_tokens_seen": 6626560,
      "step": 31405
    },
    {
      "epoch": 3.4554455445544554,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.02951871090288218,
      "loss": 0.2299,
      "num_input_tokens_seen": 6627616,
      "step": 31410
    },
    {
      "epoch": 3.455995599559956,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.02951834898046319,
      "loss": 0.2299,
      "num_input_tokens_seen": 6628768,
      "step": 31415
    },
    {
      "epoch": 3.4565456545654567,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029517986924235542,
      "loss": 0.2377,
      "num_input_tokens_seen": 6629792,
      "step": 31420
    },
    {
      "epoch": 3.457095709570957,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.029517624734202576,
      "loss": 0.2331,
      "num_input_tokens_seen": 6630848,
      "step": 31425
    },
    {
      "epoch": 3.4576457645764576,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029517262410367635,
      "loss": 0.2315,
      "num_input_tokens_seen": 6631904,
      "step": 31430
    },
    {
      "epoch": 3.458195819581958,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029516899952734044,
      "loss": 0.232,
      "num_input_tokens_seen": 6632992,
      "step": 31435
    },
    {
      "epoch": 3.458745874587459,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.029516537361305155,
      "loss": 0.2288,
      "num_input_tokens_seen": 6634016,
      "step": 31440
    },
    {
      "epoch": 3.459295929592959,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.029516174636084306,
      "loss": 0.2289,
      "num_input_tokens_seen": 6635040,
      "step": 31445
    },
    {
      "epoch": 3.4598459845984597,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.029515811777074845,
      "loss": 0.2351,
      "num_input_tokens_seen": 6636096,
      "step": 31450
    },
    {
      "epoch": 3.4603960396039604,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.02951544878428011,
      "loss": 0.2299,
      "num_input_tokens_seen": 6637120,
      "step": 31455
    },
    {
      "epoch": 3.460946094609461,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02951508565770345,
      "loss": 0.2299,
      "num_input_tokens_seen": 6638144,
      "step": 31460
    },
    {
      "epoch": 3.4614961496149617,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.029514722397348207,
      "loss": 0.2325,
      "num_input_tokens_seen": 6639232,
      "step": 31465
    },
    {
      "epoch": 3.462046204620462,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029514359003217732,
      "loss": 0.2304,
      "num_input_tokens_seen": 6640224,
      "step": 31470
    },
    {
      "epoch": 3.4625962596259625,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02951399547531538,
      "loss": 0.2288,
      "num_input_tokens_seen": 6641312,
      "step": 31475
    },
    {
      "epoch": 3.463146314631463,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.029513631813644487,
      "loss": 0.2299,
      "num_input_tokens_seen": 6642336,
      "step": 31480
    },
    {
      "epoch": 3.463696369636964,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029513268018208427,
      "loss": 0.2357,
      "num_input_tokens_seen": 6643424,
      "step": 31485
    },
    {
      "epoch": 3.4642464246424645,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029512904089010528,
      "loss": 0.2325,
      "num_input_tokens_seen": 6644480,
      "step": 31490
    },
    {
      "epoch": 3.4647964796479647,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.02951254002605416,
      "loss": 0.232,
      "num_input_tokens_seen": 6645536,
      "step": 31495
    },
    {
      "epoch": 3.4653465346534653,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02951217582934268,
      "loss": 0.2288,
      "num_input_tokens_seen": 6646560,
      "step": 31500
    },
    {
      "epoch": 3.465896589658966,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029511811498879433,
      "loss": 0.2309,
      "num_input_tokens_seen": 6647584,
      "step": 31505
    },
    {
      "epoch": 3.4664466446644666,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029511447034667782,
      "loss": 0.2309,
      "num_input_tokens_seen": 6648608,
      "step": 31510
    },
    {
      "epoch": 3.466996699669967,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029511082436711092,
      "loss": 0.2314,
      "num_input_tokens_seen": 6649632,
      "step": 31515
    },
    {
      "epoch": 3.4675467546754675,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029510717705012714,
      "loss": 0.2314,
      "num_input_tokens_seen": 6650720,
      "step": 31520
    },
    {
      "epoch": 3.468096809680968,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.029510352839576016,
      "loss": 0.2283,
      "num_input_tokens_seen": 6651776,
      "step": 31525
    },
    {
      "epoch": 3.4686468646864688,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029509987840404355,
      "loss": 0.2315,
      "num_input_tokens_seen": 6652832,
      "step": 31530
    },
    {
      "epoch": 3.469196919691969,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.029509622707501102,
      "loss": 0.2294,
      "num_input_tokens_seen": 6653920,
      "step": 31535
    },
    {
      "epoch": 3.4697469746974696,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029509257440869623,
      "loss": 0.2341,
      "num_input_tokens_seen": 6654944,
      "step": 31540
    },
    {
      "epoch": 3.4702970297029703,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02950889204051327,
      "loss": 0.2289,
      "num_input_tokens_seen": 6655968,
      "step": 31545
    },
    {
      "epoch": 3.470847084708471,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02950852650643543,
      "loss": 0.2268,
      "num_input_tokens_seen": 6657024,
      "step": 31550
    },
    {
      "epoch": 3.4713971397139716,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.02950816083863946,
      "loss": 0.2357,
      "num_input_tokens_seen": 6658112,
      "step": 31555
    },
    {
      "epoch": 3.4719471947194718,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02950779503712873,
      "loss": 0.2357,
      "num_input_tokens_seen": 6659168,
      "step": 31560
    },
    {
      "epoch": 3.4724972497249724,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02950742910190662,
      "loss": 0.2284,
      "num_input_tokens_seen": 6660320,
      "step": 31565
    },
    {
      "epoch": 3.473047304730473,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.029507063032976497,
      "loss": 0.2325,
      "num_input_tokens_seen": 6661376,
      "step": 31570
    },
    {
      "epoch": 3.4735973597359737,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02950669683034173,
      "loss": 0.2325,
      "num_input_tokens_seen": 6662432,
      "step": 31575
    },
    {
      "epoch": 3.4741474147414744,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0295063304940057,
      "loss": 0.2299,
      "num_input_tokens_seen": 6663488,
      "step": 31580
    },
    {
      "epoch": 3.4746974697469746,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02950596402397179,
      "loss": 0.2294,
      "num_input_tokens_seen": 6664544,
      "step": 31585
    },
    {
      "epoch": 3.4752475247524752,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.029505597420243362,
      "loss": 0.2299,
      "num_input_tokens_seen": 6665664,
      "step": 31590
    },
    {
      "epoch": 3.475797579757976,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.029505230682823808,
      "loss": 0.2289,
      "num_input_tokens_seen": 6666720,
      "step": 31595
    },
    {
      "epoch": 3.4763476347634765,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.029504863811716502,
      "loss": 0.2315,
      "num_input_tokens_seen": 6667808,
      "step": 31600
    },
    {
      "epoch": 3.4768976897689767,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.029504496806924827,
      "loss": 0.231,
      "num_input_tokens_seen": 6668928,
      "step": 31605
    },
    {
      "epoch": 3.4774477447744774,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029504129668452162,
      "loss": 0.2268,
      "num_input_tokens_seen": 6669984,
      "step": 31610
    },
    {
      "epoch": 3.477997799779978,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.029503762396301897,
      "loss": 0.2341,
      "num_input_tokens_seen": 6671008,
      "step": 31615
    },
    {
      "epoch": 3.4785478547854787,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.029503394990477414,
      "loss": 0.2285,
      "num_input_tokens_seen": 6672096,
      "step": 31620
    },
    {
      "epoch": 3.479097909790979,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029503027450982093,
      "loss": 0.2363,
      "num_input_tokens_seen": 6673088,
      "step": 31625
    },
    {
      "epoch": 3.4796479647964795,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.029502659777819334,
      "loss": 0.2284,
      "num_input_tokens_seen": 6674112,
      "step": 31630
    },
    {
      "epoch": 3.48019801980198,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.029502291970992512,
      "loss": 0.2341,
      "num_input_tokens_seen": 6675200,
      "step": 31635
    },
    {
      "epoch": 3.480748074807481,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.029501924030505028,
      "loss": 0.2273,
      "num_input_tokens_seen": 6676256,
      "step": 31640
    },
    {
      "epoch": 3.4812981298129815,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.029501555956360268,
      "loss": 0.2346,
      "num_input_tokens_seen": 6677248,
      "step": 31645
    },
    {
      "epoch": 3.4818481848184817,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029501187748561628,
      "loss": 0.2335,
      "num_input_tokens_seen": 6678336,
      "step": 31650
    },
    {
      "epoch": 3.4823982398239823,
      "grad_norm": 0.000598907470703125,
      "learning_rate": 0.029500819407112498,
      "loss": 0.233,
      "num_input_tokens_seen": 6679392,
      "step": 31655
    },
    {
      "epoch": 3.482948294829483,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02950045093201627,
      "loss": 0.2319,
      "num_input_tokens_seen": 6680448,
      "step": 31660
    },
    {
      "epoch": 3.4834983498349836,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02950008232327635,
      "loss": 0.232,
      "num_input_tokens_seen": 6681568,
      "step": 31665
    },
    {
      "epoch": 3.4840484048404843,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.029499713580896123,
      "loss": 0.2309,
      "num_input_tokens_seen": 6682624,
      "step": 31670
    },
    {
      "epoch": 3.4845984598459845,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029499344704878996,
      "loss": 0.2361,
      "num_input_tokens_seen": 6683680,
      "step": 31675
    },
    {
      "epoch": 3.485148514851485,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029498975695228365,
      "loss": 0.234,
      "num_input_tokens_seen": 6684736,
      "step": 31680
    },
    {
      "epoch": 3.485698569856986,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.029498606551947637,
      "loss": 0.2309,
      "num_input_tokens_seen": 6685856,
      "step": 31685
    },
    {
      "epoch": 3.4862486248624864,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029498237275040202,
      "loss": 0.233,
      "num_input_tokens_seen": 6686976,
      "step": 31690
    },
    {
      "epoch": 3.4867986798679866,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029497867864509475,
      "loss": 0.2314,
      "num_input_tokens_seen": 6688000,
      "step": 31695
    },
    {
      "epoch": 3.4873487348734873,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029497498320358858,
      "loss": 0.2319,
      "num_input_tokens_seen": 6689152,
      "step": 31700
    },
    {
      "epoch": 3.487898789878988,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029497128642591755,
      "loss": 0.2303,
      "num_input_tokens_seen": 6690208,
      "step": 31705
    },
    {
      "epoch": 3.4884488448844886,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.02949675883121157,
      "loss": 0.2319,
      "num_input_tokens_seen": 6691296,
      "step": 31710
    },
    {
      "epoch": 3.488998899889989,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02949638888622172,
      "loss": 0.2324,
      "num_input_tokens_seen": 6692352,
      "step": 31715
    },
    {
      "epoch": 3.4895489548954894,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029496018807625604,
      "loss": 0.2324,
      "num_input_tokens_seen": 6693408,
      "step": 31720
    },
    {
      "epoch": 3.49009900990099,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.029495648595426645,
      "loss": 0.2329,
      "num_input_tokens_seen": 6694464,
      "step": 31725
    },
    {
      "epoch": 3.4906490649064907,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029495278249628245,
      "loss": 0.2288,
      "num_input_tokens_seen": 6695520,
      "step": 31730
    },
    {
      "epoch": 3.4911991199119914,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.029494907770233822,
      "loss": 0.232,
      "num_input_tokens_seen": 6696512,
      "step": 31735
    },
    {
      "epoch": 3.4917491749174916,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.029494537157246786,
      "loss": 0.2284,
      "num_input_tokens_seen": 6697568,
      "step": 31740
    },
    {
      "epoch": 3.4922992299229922,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.029494166410670562,
      "loss": 0.2342,
      "num_input_tokens_seen": 6698624,
      "step": 31745
    },
    {
      "epoch": 3.492849284928493,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.029493795530508557,
      "loss": 0.2422,
      "num_input_tokens_seen": 6699648,
      "step": 31750
    },
    {
      "epoch": 3.4933993399339935,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029493424516764195,
      "loss": 0.2337,
      "num_input_tokens_seen": 6700736,
      "step": 31755
    },
    {
      "epoch": 3.493949394939494,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.029493053369440896,
      "loss": 0.2304,
      "num_input_tokens_seen": 6701760,
      "step": 31760
    },
    {
      "epoch": 3.4944994499449944,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.029492682088542072,
      "loss": 0.2304,
      "num_input_tokens_seen": 6702848,
      "step": 31765
    },
    {
      "epoch": 3.495049504950495,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.029492310674071157,
      "loss": 0.2283,
      "num_input_tokens_seen": 6704032,
      "step": 31770
    },
    {
      "epoch": 3.4955995599559957,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.02949193912603157,
      "loss": 0.232,
      "num_input_tokens_seen": 6705152,
      "step": 31775
    },
    {
      "epoch": 3.4961496149614963,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029491567444426728,
      "loss": 0.2341,
      "num_input_tokens_seen": 6706240,
      "step": 31780
    },
    {
      "epoch": 3.4966996699669965,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.029491195629260065,
      "loss": 0.232,
      "num_input_tokens_seen": 6707296,
      "step": 31785
    },
    {
      "epoch": 3.497249724972497,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.029490823680535006,
      "loss": 0.2319,
      "num_input_tokens_seen": 6708384,
      "step": 31790
    },
    {
      "epoch": 3.497799779977998,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.02949045159825498,
      "loss": 0.2309,
      "num_input_tokens_seen": 6709408,
      "step": 31795
    },
    {
      "epoch": 3.4983498349834985,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.029490079382423415,
      "loss": 0.2319,
      "num_input_tokens_seen": 6710496,
      "step": 31800
    },
    {
      "epoch": 3.4988998899889987,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02948970703304374,
      "loss": 0.2283,
      "num_input_tokens_seen": 6711552,
      "step": 31805
    },
    {
      "epoch": 3.4994499449944994,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.029489334550119387,
      "loss": 0.2299,
      "num_input_tokens_seen": 6712608,
      "step": 31810
    },
    {
      "epoch": 3.5,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.02948896193365379,
      "loss": 0.2326,
      "num_input_tokens_seen": 6713664,
      "step": 31815
    },
    {
      "epoch": 3.5005500550055006,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.029488589183650382,
      "loss": 0.2321,
      "num_input_tokens_seen": 6714752,
      "step": 31820
    },
    {
      "epoch": 3.5011001100110013,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.029488216300112603,
      "loss": 0.229,
      "num_input_tokens_seen": 6715776,
      "step": 31825
    },
    {
      "epoch": 3.5016501650165015,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.029487843283043887,
      "loss": 0.2342,
      "num_input_tokens_seen": 6716800,
      "step": 31830
    },
    {
      "epoch": 3.502200220022002,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029487470132447666,
      "loss": 0.23,
      "num_input_tokens_seen": 6717888,
      "step": 31835
    },
    {
      "epoch": 3.502750275027503,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.029487096848327392,
      "loss": 0.2295,
      "num_input_tokens_seen": 6719008,
      "step": 31840
    },
    {
      "epoch": 3.5033003300330035,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.029486723430686493,
      "loss": 0.2342,
      "num_input_tokens_seen": 6720064,
      "step": 31845
    },
    {
      "epoch": 3.503850385038504,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029486349879528417,
      "loss": 0.2341,
      "num_input_tokens_seen": 6721216,
      "step": 31850
    },
    {
      "epoch": 3.5044004400440043,
      "grad_norm": 0.01171875,
      "learning_rate": 0.029485976194856607,
      "loss": 0.2363,
      "num_input_tokens_seen": 6722304,
      "step": 31855
    },
    {
      "epoch": 3.504950495049505,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029485602376674504,
      "loss": 0.2335,
      "num_input_tokens_seen": 6723328,
      "step": 31860
    },
    {
      "epoch": 3.5055005500550056,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.029485228424985556,
      "loss": 0.2309,
      "num_input_tokens_seen": 6724416,
      "step": 31865
    },
    {
      "epoch": 3.506050605060506,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.029484854339793205,
      "loss": 0.232,
      "num_input_tokens_seen": 6725472,
      "step": 31870
    },
    {
      "epoch": 3.5066006600660065,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029484480121100904,
      "loss": 0.2304,
      "num_input_tokens_seen": 6726560,
      "step": 31875
    },
    {
      "epoch": 3.507150715071507,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.0294841057689121,
      "loss": 0.2315,
      "num_input_tokens_seen": 6727584,
      "step": 31880
    },
    {
      "epoch": 3.5077007700770078,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.029483731283230244,
      "loss": 0.2314,
      "num_input_tokens_seen": 6728640,
      "step": 31885
    },
    {
      "epoch": 3.5082508250825084,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029483356664058786,
      "loss": 0.2335,
      "num_input_tokens_seen": 6729728,
      "step": 31890
    },
    {
      "epoch": 3.5088008800880086,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029482981911401182,
      "loss": 0.2288,
      "num_input_tokens_seen": 6730848,
      "step": 31895
    },
    {
      "epoch": 3.5093509350935093,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02948260702526088,
      "loss": 0.2309,
      "num_input_tokens_seen": 6731904,
      "step": 31900
    },
    {
      "epoch": 3.50990099009901,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.02948223200564134,
      "loss": 0.232,
      "num_input_tokens_seen": 6732928,
      "step": 31905
    },
    {
      "epoch": 3.5104510451045106,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.029481856852546014,
      "loss": 0.2319,
      "num_input_tokens_seen": 6733952,
      "step": 31910
    },
    {
      "epoch": 3.511001100110011,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.029481481565978365,
      "loss": 0.2293,
      "num_input_tokens_seen": 6735040,
      "step": 31915
    },
    {
      "epoch": 3.5115511551155114,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.02948110614594185,
      "loss": 0.2314,
      "num_input_tokens_seen": 6736096,
      "step": 31920
    },
    {
      "epoch": 3.512101210121012,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029480730592439927,
      "loss": 0.2293,
      "num_input_tokens_seen": 6737152,
      "step": 31925
    },
    {
      "epoch": 3.5126512651265127,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029480354905476056,
      "loss": 0.2319,
      "num_input_tokens_seen": 6738208,
      "step": 31930
    },
    {
      "epoch": 3.5132013201320134,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.029479979085053707,
      "loss": 0.2308,
      "num_input_tokens_seen": 6739232,
      "step": 31935
    },
    {
      "epoch": 3.513751375137514,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.029479603131176337,
      "loss": 0.2293,
      "num_input_tokens_seen": 6740320,
      "step": 31940
    },
    {
      "epoch": 3.514301430143014,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029479227043847412,
      "loss": 0.2319,
      "num_input_tokens_seen": 6741312,
      "step": 31945
    },
    {
      "epoch": 3.514851485148515,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.0294788508230704,
      "loss": 0.2299,
      "num_input_tokens_seen": 6742336,
      "step": 31950
    },
    {
      "epoch": 3.5154015401540155,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02947847446884877,
      "loss": 0.2309,
      "num_input_tokens_seen": 6743392,
      "step": 31955
    },
    {
      "epoch": 3.5159515951595157,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.029478097981185985,
      "loss": 0.2319,
      "num_input_tokens_seen": 6744352,
      "step": 31960
    },
    {
      "epoch": 3.5165016501650164,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029477721360085522,
      "loss": 0.2329,
      "num_input_tokens_seen": 6745440,
      "step": 31965
    },
    {
      "epoch": 3.517051705170517,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029477344605550845,
      "loss": 0.2325,
      "num_input_tokens_seen": 6746464,
      "step": 31970
    },
    {
      "epoch": 3.5176017601760177,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02947696771758543,
      "loss": 0.2324,
      "num_input_tokens_seen": 6747584,
      "step": 31975
    },
    {
      "epoch": 3.5181518151815183,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.02947659069619275,
      "loss": 0.2324,
      "num_input_tokens_seen": 6748608,
      "step": 31980
    },
    {
      "epoch": 3.5187018701870185,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029476213541376278,
      "loss": 0.2319,
      "num_input_tokens_seen": 6749664,
      "step": 31985
    },
    {
      "epoch": 3.519251925192519,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029475836253139497,
      "loss": 0.2303,
      "num_input_tokens_seen": 6750720,
      "step": 31990
    },
    {
      "epoch": 3.51980198019802,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.029475458831485876,
      "loss": 0.2293,
      "num_input_tokens_seen": 6751808,
      "step": 31995
    },
    {
      "epoch": 3.5203520352035205,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.0294750812764189,
      "loss": 0.2308,
      "num_input_tokens_seen": 6752864,
      "step": 32000
    },
    {
      "epoch": 3.520902090209021,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02947470358794204,
      "loss": 0.2319,
      "num_input_tokens_seen": 6753984,
      "step": 32005
    },
    {
      "epoch": 3.5214521452145213,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02947432576605879,
      "loss": 0.233,
      "num_input_tokens_seen": 6755008,
      "step": 32010
    },
    {
      "epoch": 3.522002200220022,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029473947810772616,
      "loss": 0.2325,
      "num_input_tokens_seen": 6756096,
      "step": 32015
    },
    {
      "epoch": 3.5225522552255226,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029473569722087017,
      "loss": 0.2314,
      "num_input_tokens_seen": 6757184,
      "step": 32020
    },
    {
      "epoch": 3.523102310231023,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.02947319150000547,
      "loss": 0.2314,
      "num_input_tokens_seen": 6758304,
      "step": 32025
    },
    {
      "epoch": 3.523652365236524,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029472813144531464,
      "loss": 0.2319,
      "num_input_tokens_seen": 6759360,
      "step": 32030
    },
    {
      "epoch": 3.524202420242024,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02947243465566848,
      "loss": 0.2309,
      "num_input_tokens_seen": 6760416,
      "step": 32035
    },
    {
      "epoch": 3.5247524752475248,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029472056033420013,
      "loss": 0.2341,
      "num_input_tokens_seen": 6761472,
      "step": 32040
    },
    {
      "epoch": 3.5253025302530254,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.02947167727778955,
      "loss": 0.2314,
      "num_input_tokens_seen": 6762432,
      "step": 32045
    },
    {
      "epoch": 3.5258525852585256,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.029471298388780577,
      "loss": 0.232,
      "num_input_tokens_seen": 6763456,
      "step": 32050
    },
    {
      "epoch": 3.5264026402640263,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.029470919366396595,
      "loss": 0.2325,
      "num_input_tokens_seen": 6764512,
      "step": 32055
    },
    {
      "epoch": 3.526952695269527,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.02947054021064109,
      "loss": 0.2314,
      "num_input_tokens_seen": 6765568,
      "step": 32060
    },
    {
      "epoch": 3.5275027502750276,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029470160921517562,
      "loss": 0.2335,
      "num_input_tokens_seen": 6766656,
      "step": 32065
    },
    {
      "epoch": 3.5280528052805282,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.029469781499029506,
      "loss": 0.2278,
      "num_input_tokens_seen": 6767712,
      "step": 32070
    },
    {
      "epoch": 3.5286028602860284,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.029469401943180415,
      "loss": 0.2283,
      "num_input_tokens_seen": 6768768,
      "step": 32075
    },
    {
      "epoch": 3.529152915291529,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029469022253973787,
      "loss": 0.2309,
      "num_input_tokens_seen": 6769760,
      "step": 32080
    },
    {
      "epoch": 3.5297029702970297,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.02946864243141313,
      "loss": 0.2309,
      "num_input_tokens_seen": 6770784,
      "step": 32085
    },
    {
      "epoch": 3.5302530253025304,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.029468262475501935,
      "loss": 0.2298,
      "num_input_tokens_seen": 6771904,
      "step": 32090
    },
    {
      "epoch": 3.530803080308031,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029467882386243704,
      "loss": 0.2324,
      "num_input_tokens_seen": 6772928,
      "step": 32095
    },
    {
      "epoch": 3.5313531353135312,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029467502163641947,
      "loss": 0.2304,
      "num_input_tokens_seen": 6774016,
      "step": 32100
    },
    {
      "epoch": 3.531903190319032,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.029467121807700163,
      "loss": 0.2293,
      "num_input_tokens_seen": 6775040,
      "step": 32105
    },
    {
      "epoch": 3.5324532453245325,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.029466741318421863,
      "loss": 0.2314,
      "num_input_tokens_seen": 6776096,
      "step": 32110
    },
    {
      "epoch": 3.5330033003300327,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.02946636069581055,
      "loss": 0.2303,
      "num_input_tokens_seen": 6777120,
      "step": 32115
    },
    {
      "epoch": 3.533553355335534,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.029465979939869727,
      "loss": 0.2314,
      "num_input_tokens_seen": 6778176,
      "step": 32120
    },
    {
      "epoch": 3.534103410341034,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.02946559905060291,
      "loss": 0.2319,
      "num_input_tokens_seen": 6779168,
      "step": 32125
    },
    {
      "epoch": 3.5346534653465347,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.029465218028013607,
      "loss": 0.234,
      "num_input_tokens_seen": 6780256,
      "step": 32130
    },
    {
      "epoch": 3.5352035203520353,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.029464836872105337,
      "loss": 0.2335,
      "num_input_tokens_seen": 6781344,
      "step": 32135
    },
    {
      "epoch": 3.5357535753575355,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029464455582881598,
      "loss": 0.2314,
      "num_input_tokens_seen": 6782432,
      "step": 32140
    },
    {
      "epoch": 3.536303630363036,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02946407416034592,
      "loss": 0.2303,
      "num_input_tokens_seen": 6783456,
      "step": 32145
    },
    {
      "epoch": 3.536853685368537,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029463692604501806,
      "loss": 0.2324,
      "num_input_tokens_seen": 6784480,
      "step": 32150
    },
    {
      "epoch": 3.5374037403740375,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02946331091535278,
      "loss": 0.2319,
      "num_input_tokens_seen": 6785600,
      "step": 32155
    },
    {
      "epoch": 3.537953795379538,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029462929092902355,
      "loss": 0.2319,
      "num_input_tokens_seen": 6786784,
      "step": 32160
    },
    {
      "epoch": 3.5385038503850383,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.029462547137154054,
      "loss": 0.2303,
      "num_input_tokens_seen": 6787904,
      "step": 32165
    },
    {
      "epoch": 3.539053905390539,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.029462165048111395,
      "loss": 0.2293,
      "num_input_tokens_seen": 6788928,
      "step": 32170
    },
    {
      "epoch": 3.5396039603960396,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029461782825777903,
      "loss": 0.232,
      "num_input_tokens_seen": 6789952,
      "step": 32175
    },
    {
      "epoch": 3.5401540154015403,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.029461400470157093,
      "loss": 0.2325,
      "num_input_tokens_seen": 6791040,
      "step": 32180
    },
    {
      "epoch": 3.540704070407041,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0294610179812525,
      "loss": 0.232,
      "num_input_tokens_seen": 6792128,
      "step": 32185
    },
    {
      "epoch": 3.541254125412541,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.029460635359067643,
      "loss": 0.2325,
      "num_input_tokens_seen": 6793152,
      "step": 32190
    },
    {
      "epoch": 3.541804180418042,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029460252603606043,
      "loss": 0.2335,
      "num_input_tokens_seen": 6794272,
      "step": 32195
    },
    {
      "epoch": 3.5423542354235424,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029459869714871238,
      "loss": 0.2314,
      "num_input_tokens_seen": 6795392,
      "step": 32200
    },
    {
      "epoch": 3.5429042904290426,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.029459486692866754,
      "loss": 0.2319,
      "num_input_tokens_seen": 6796448,
      "step": 32205
    },
    {
      "epoch": 3.5434543454345433,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.029459103537596114,
      "loss": 0.233,
      "num_input_tokens_seen": 6797440,
      "step": 32210
    },
    {
      "epoch": 3.544004400440044,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02945872024906286,
      "loss": 0.2298,
      "num_input_tokens_seen": 6798464,
      "step": 32215
    },
    {
      "epoch": 3.5445544554455446,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.029458336827270518,
      "loss": 0.2324,
      "num_input_tokens_seen": 6799520,
      "step": 32220
    },
    {
      "epoch": 3.5451045104510452,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02945795327222262,
      "loss": 0.2324,
      "num_input_tokens_seen": 6800608,
      "step": 32225
    },
    {
      "epoch": 3.5456545654565454,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.029457569583922705,
      "loss": 0.2308,
      "num_input_tokens_seen": 6801664,
      "step": 32230
    },
    {
      "epoch": 3.546204620462046,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.029457185762374313,
      "loss": 0.2319,
      "num_input_tokens_seen": 6802688,
      "step": 32235
    },
    {
      "epoch": 3.5467546754675467,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029456801807580976,
      "loss": 0.2298,
      "num_input_tokens_seen": 6803744,
      "step": 32240
    },
    {
      "epoch": 3.5473047304730474,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02945641771954623,
      "loss": 0.2314,
      "num_input_tokens_seen": 6804864,
      "step": 32245
    },
    {
      "epoch": 3.547854785478548,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.02945603349827362,
      "loss": 0.2277,
      "num_input_tokens_seen": 6805888,
      "step": 32250
    },
    {
      "epoch": 3.5484048404840483,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02945564914376669,
      "loss": 0.2308,
      "num_input_tokens_seen": 6807008,
      "step": 32255
    },
    {
      "epoch": 3.548954895489549,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.029455264656028976,
      "loss": 0.2329,
      "num_input_tokens_seen": 6808032,
      "step": 32260
    },
    {
      "epoch": 3.5495049504950495,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02945488003506402,
      "loss": 0.2308,
      "num_input_tokens_seen": 6809120,
      "step": 32265
    },
    {
      "epoch": 3.55005500550055,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.029454495280875376,
      "loss": 0.2324,
      "num_input_tokens_seen": 6810208,
      "step": 32270
    },
    {
      "epoch": 3.550605060506051,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.02945411039346658,
      "loss": 0.2303,
      "num_input_tokens_seen": 6811264,
      "step": 32275
    },
    {
      "epoch": 3.551155115511551,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02945372537284119,
      "loss": 0.233,
      "num_input_tokens_seen": 6812320,
      "step": 32280
    },
    {
      "epoch": 3.5517051705170517,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02945334021900274,
      "loss": 0.2314,
      "num_input_tokens_seen": 6813344,
      "step": 32285
    },
    {
      "epoch": 3.5522552255225524,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0294529549319548,
      "loss": 0.2325,
      "num_input_tokens_seen": 6814368,
      "step": 32290
    },
    {
      "epoch": 3.5528052805280526,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.029452569511700905,
      "loss": 0.2314,
      "num_input_tokens_seen": 6815424,
      "step": 32295
    },
    {
      "epoch": 3.553355335533553,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02945218395824461,
      "loss": 0.2335,
      "num_input_tokens_seen": 6816512,
      "step": 32300
    },
    {
      "epoch": 3.553905390539054,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02945179827158947,
      "loss": 0.2309,
      "num_input_tokens_seen": 6817632,
      "step": 32305
    },
    {
      "epoch": 3.5544554455445545,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.029451412451739042,
      "loss": 0.2329,
      "num_input_tokens_seen": 6818688,
      "step": 32310
    },
    {
      "epoch": 3.555005500550055,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02945102649869688,
      "loss": 0.2319,
      "num_input_tokens_seen": 6819744,
      "step": 32315
    },
    {
      "epoch": 3.5555555555555554,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.02945064041246654,
      "loss": 0.2319,
      "num_input_tokens_seen": 6820768,
      "step": 32320
    },
    {
      "epoch": 3.556105610561056,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02945025419305158,
      "loss": 0.2308,
      "num_input_tokens_seen": 6821824,
      "step": 32325
    },
    {
      "epoch": 3.5566556655665567,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029449867840455564,
      "loss": 0.2319,
      "num_input_tokens_seen": 6822976,
      "step": 32330
    },
    {
      "epoch": 3.5572057205720573,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029449481354682047,
      "loss": 0.2314,
      "num_input_tokens_seen": 6824064,
      "step": 32335
    },
    {
      "epoch": 3.557755775577558,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.029449094735734597,
      "loss": 0.2309,
      "num_input_tokens_seen": 6825056,
      "step": 32340
    },
    {
      "epoch": 3.558305830583058,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.029448707983616772,
      "loss": 0.2308,
      "num_input_tokens_seen": 6826112,
      "step": 32345
    },
    {
      "epoch": 3.558855885588559,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02944832109833214,
      "loss": 0.2319,
      "num_input_tokens_seen": 6827104,
      "step": 32350
    },
    {
      "epoch": 3.5594059405940595,
      "grad_norm": 0.000614166259765625,
      "learning_rate": 0.029447934079884268,
      "loss": 0.2303,
      "num_input_tokens_seen": 6828192,
      "step": 32355
    },
    {
      "epoch": 3.55995599559956,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.029447546928276714,
      "loss": 0.233,
      "num_input_tokens_seen": 6829184,
      "step": 32360
    },
    {
      "epoch": 3.5605060506050608,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.029447159643513056,
      "loss": 0.2334,
      "num_input_tokens_seen": 6830272,
      "step": 32365
    },
    {
      "epoch": 3.561056105610561,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02944677222559686,
      "loss": 0.2319,
      "num_input_tokens_seen": 6831296,
      "step": 32370
    },
    {
      "epoch": 3.5616061606160616,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.029446384674531693,
      "loss": 0.2304,
      "num_input_tokens_seen": 6832352,
      "step": 32375
    },
    {
      "epoch": 3.5621562156215623,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.029445996990321133,
      "loss": 0.2304,
      "num_input_tokens_seen": 6833440,
      "step": 32380
    },
    {
      "epoch": 3.5627062706270625,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029445609172968755,
      "loss": 0.233,
      "num_input_tokens_seen": 6834592,
      "step": 32385
    },
    {
      "epoch": 3.563256325632563,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.029445221222478123,
      "loss": 0.2309,
      "num_input_tokens_seen": 6835648,
      "step": 32390
    },
    {
      "epoch": 3.5638063806380638,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02944483313885282,
      "loss": 0.2304,
      "num_input_tokens_seen": 6836672,
      "step": 32395
    },
    {
      "epoch": 3.5643564356435644,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.029444444922096427,
      "loss": 0.2314,
      "num_input_tokens_seen": 6837696,
      "step": 32400
    },
    {
      "epoch": 3.564906490649065,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02944405657221251,
      "loss": 0.2304,
      "num_input_tokens_seen": 6838752,
      "step": 32405
    },
    {
      "epoch": 3.5654565456545653,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.029443668089204653,
      "loss": 0.2288,
      "num_input_tokens_seen": 6839840,
      "step": 32410
    },
    {
      "epoch": 3.566006600660066,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.029443279473076442,
      "loss": 0.233,
      "num_input_tokens_seen": 6840832,
      "step": 32415
    },
    {
      "epoch": 3.5665566556655666,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.029442890723831452,
      "loss": 0.2314,
      "num_input_tokens_seen": 6841888,
      "step": 32420
    },
    {
      "epoch": 3.567106710671067,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.029442501841473272,
      "loss": 0.2324,
      "num_input_tokens_seen": 6842944,
      "step": 32425
    },
    {
      "epoch": 3.567656765676568,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.02944211282600548,
      "loss": 0.2273,
      "num_input_tokens_seen": 6844032,
      "step": 32430
    },
    {
      "epoch": 3.568206820682068,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.029441723677431662,
      "loss": 0.2293,
      "num_input_tokens_seen": 6845120,
      "step": 32435
    },
    {
      "epoch": 3.5687568756875687,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.029441334395755414,
      "loss": 0.2304,
      "num_input_tokens_seen": 6846144,
      "step": 32440
    },
    {
      "epoch": 3.5693069306930694,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.02944094498098031,
      "loss": 0.2309,
      "num_input_tokens_seen": 6847232,
      "step": 32445
    },
    {
      "epoch": 3.56985698569857,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.029440555433109947,
      "loss": 0.2293,
      "num_input_tokens_seen": 6848288,
      "step": 32450
    },
    {
      "epoch": 3.5704070407040707,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029440165752147914,
      "loss": 0.2304,
      "num_input_tokens_seen": 6849280,
      "step": 32455
    },
    {
      "epoch": 3.570957095709571,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029439775938097802,
      "loss": 0.234,
      "num_input_tokens_seen": 6850240,
      "step": 32460
    },
    {
      "epoch": 3.5715071507150715,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029439385990963206,
      "loss": 0.232,
      "num_input_tokens_seen": 6851232,
      "step": 32465
    },
    {
      "epoch": 3.572057205720572,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029438995910747716,
      "loss": 0.2335,
      "num_input_tokens_seen": 6852288,
      "step": 32470
    },
    {
      "epoch": 3.5726072607260724,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.029438605697454934,
      "loss": 0.2288,
      "num_input_tokens_seen": 6853344,
      "step": 32475
    },
    {
      "epoch": 3.573157315731573,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.029438215351088443,
      "loss": 0.2293,
      "num_input_tokens_seen": 6854400,
      "step": 32480
    },
    {
      "epoch": 3.5737073707370737,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029437824871651853,
      "loss": 0.2309,
      "num_input_tokens_seen": 6855392,
      "step": 32485
    },
    {
      "epoch": 3.5742574257425743,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.02943743425914876,
      "loss": 0.2345,
      "num_input_tokens_seen": 6856416,
      "step": 32490
    },
    {
      "epoch": 3.574807480748075,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.029437043513582766,
      "loss": 0.2319,
      "num_input_tokens_seen": 6857504,
      "step": 32495
    },
    {
      "epoch": 3.575357535753575,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.029436652634957466,
      "loss": 0.2314,
      "num_input_tokens_seen": 6858496,
      "step": 32500
    },
    {
      "epoch": 3.575907590759076,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.029436261623276463,
      "loss": 0.2267,
      "num_input_tokens_seen": 6859552,
      "step": 32505
    },
    {
      "epoch": 3.5764576457645765,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.029435870478543368,
      "loss": 0.231,
      "num_input_tokens_seen": 6860608,
      "step": 32510
    },
    {
      "epoch": 3.577007700770077,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.029435479200761783,
      "loss": 0.231,
      "num_input_tokens_seen": 6861632,
      "step": 32515
    },
    {
      "epoch": 3.5775577557755778,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02943508778993531,
      "loss": 0.231,
      "num_input_tokens_seen": 6862720,
      "step": 32520
    },
    {
      "epoch": 3.578107810781078,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029434696246067562,
      "loss": 0.2283,
      "num_input_tokens_seen": 6863808,
      "step": 32525
    },
    {
      "epoch": 3.5786578657865786,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029434304569162146,
      "loss": 0.2321,
      "num_input_tokens_seen": 6864864,
      "step": 32530
    },
    {
      "epoch": 3.5792079207920793,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.029433912759222667,
      "loss": 0.2223,
      "num_input_tokens_seen": 6865920,
      "step": 32535
    },
    {
      "epoch": 3.5797579757975795,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.02943352081625274,
      "loss": 0.2343,
      "num_input_tokens_seen": 6866944,
      "step": 32540
    },
    {
      "epoch": 3.5803080308030806,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.029433128740255984,
      "loss": 0.2329,
      "num_input_tokens_seen": 6867968,
      "step": 32545
    },
    {
      "epoch": 3.580858085808581,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.029432736531236,
      "loss": 0.2364,
      "num_input_tokens_seen": 6869024,
      "step": 32550
    },
    {
      "epoch": 3.5814081408140814,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.02943234418919641,
      "loss": 0.2349,
      "num_input_tokens_seen": 6870112,
      "step": 32555
    },
    {
      "epoch": 3.581958195819582,
      "grad_norm": 0.00080108642578125,
      "learning_rate": 0.02943195171414083,
      "loss": 0.2269,
      "num_input_tokens_seen": 6871200,
      "step": 32560
    },
    {
      "epoch": 3.5825082508250823,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.029431559106072876,
      "loss": 0.2332,
      "num_input_tokens_seen": 6872288,
      "step": 32565
    },
    {
      "epoch": 3.583058305830583,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.029431166364996167,
      "loss": 0.231,
      "num_input_tokens_seen": 6873312,
      "step": 32570
    },
    {
      "epoch": 3.5836083608360836,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02943077349091432,
      "loss": 0.2289,
      "num_input_tokens_seen": 6874432,
      "step": 32575
    },
    {
      "epoch": 3.5841584158415842,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02943038048383096,
      "loss": 0.2325,
      "num_input_tokens_seen": 6875488,
      "step": 32580
    },
    {
      "epoch": 3.584708470847085,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029429987343749707,
      "loss": 0.2378,
      "num_input_tokens_seen": 6876512,
      "step": 32585
    },
    {
      "epoch": 3.585258525852585,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.029429594070674185,
      "loss": 0.2304,
      "num_input_tokens_seen": 6877568,
      "step": 32590
    },
    {
      "epoch": 3.5858085808580857,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.029429200664608017,
      "loss": 0.2293,
      "num_input_tokens_seen": 6878656,
      "step": 32595
    },
    {
      "epoch": 3.5863586358635864,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.02942880712555483,
      "loss": 0.2304,
      "num_input_tokens_seen": 6879808,
      "step": 32600
    },
    {
      "epoch": 3.586908690869087,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029428413453518258,
      "loss": 0.2299,
      "num_input_tokens_seen": 6880832,
      "step": 32605
    },
    {
      "epoch": 3.5874587458745877,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.02942801964850192,
      "loss": 0.2319,
      "num_input_tokens_seen": 6881856,
      "step": 32610
    },
    {
      "epoch": 3.588008800880088,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029427625710509442,
      "loss": 0.234,
      "num_input_tokens_seen": 6882912,
      "step": 32615
    },
    {
      "epoch": 3.5885588558855885,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.029427231639544468,
      "loss": 0.2288,
      "num_input_tokens_seen": 6884000,
      "step": 32620
    },
    {
      "epoch": 3.589108910891089,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.029426837435610618,
      "loss": 0.232,
      "num_input_tokens_seen": 6885056,
      "step": 32625
    },
    {
      "epoch": 3.5896589658965894,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.029426443098711532,
      "loss": 0.2309,
      "num_input_tokens_seen": 6886112,
      "step": 32630
    },
    {
      "epoch": 3.5902090209020905,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.029426048628850843,
      "loss": 0.232,
      "num_input_tokens_seen": 6887136,
      "step": 32635
    },
    {
      "epoch": 3.5907590759075907,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.029425654026032186,
      "loss": 0.2304,
      "num_input_tokens_seen": 6888160,
      "step": 32640
    },
    {
      "epoch": 3.5913091309130913,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.029425259290259194,
      "loss": 0.2315,
      "num_input_tokens_seen": 6889248,
      "step": 32645
    },
    {
      "epoch": 3.591859185918592,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.029424864421535513,
      "loss": 0.2283,
      "num_input_tokens_seen": 6890272,
      "step": 32650
    },
    {
      "epoch": 3.592409240924092,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029424469419864777,
      "loss": 0.2346,
      "num_input_tokens_seen": 6891392,
      "step": 32655
    },
    {
      "epoch": 3.592959295929593,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.02942407428525063,
      "loss": 0.2299,
      "num_input_tokens_seen": 6892384,
      "step": 32660
    },
    {
      "epoch": 3.5935093509350935,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02942367901769671,
      "loss": 0.2273,
      "num_input_tokens_seen": 6893472,
      "step": 32665
    },
    {
      "epoch": 3.594059405940594,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.029423283617206662,
      "loss": 0.2346,
      "num_input_tokens_seen": 6894496,
      "step": 32670
    },
    {
      "epoch": 3.594609460946095,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.029422888083784127,
      "loss": 0.2278,
      "num_input_tokens_seen": 6895552,
      "step": 32675
    },
    {
      "epoch": 3.595159515951595,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02942249241743276,
      "loss": 0.2346,
      "num_input_tokens_seen": 6896640,
      "step": 32680
    },
    {
      "epoch": 3.5957095709570956,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.029422096618156195,
      "loss": 0.2304,
      "num_input_tokens_seen": 6897696,
      "step": 32685
    },
    {
      "epoch": 3.5962596259625963,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029421700685958087,
      "loss": 0.232,
      "num_input_tokens_seen": 6898688,
      "step": 32690
    },
    {
      "epoch": 3.596809680968097,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02942130462084209,
      "loss": 0.2294,
      "num_input_tokens_seen": 6899744,
      "step": 32695
    },
    {
      "epoch": 3.5973597359735976,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029420908422811837,
      "loss": 0.2293,
      "num_input_tokens_seen": 6900768,
      "step": 32700
    },
    {
      "epoch": 3.597909790979098,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.029420512091870996,
      "loss": 0.232,
      "num_input_tokens_seen": 6901824,
      "step": 32705
    },
    {
      "epoch": 3.5984598459845984,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02942011562802322,
      "loss": 0.2346,
      "num_input_tokens_seen": 6902912,
      "step": 32710
    },
    {
      "epoch": 3.599009900990099,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029419719031272152,
      "loss": 0.2315,
      "num_input_tokens_seen": 6903968,
      "step": 32715
    },
    {
      "epoch": 3.5995599559955993,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.029419322301621455,
      "loss": 0.2351,
      "num_input_tokens_seen": 6905024,
      "step": 32720
    },
    {
      "epoch": 3.6001100110011,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.029418925439074782,
      "loss": 0.2304,
      "num_input_tokens_seen": 6906048,
      "step": 32725
    },
    {
      "epoch": 3.6006600660066006,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.029418528443635794,
      "loss": 0.2303,
      "num_input_tokens_seen": 6907072,
      "step": 32730
    },
    {
      "epoch": 3.6012101210121013,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.029418131315308144,
      "loss": 0.2324,
      "num_input_tokens_seen": 6908160,
      "step": 32735
    },
    {
      "epoch": 3.601760176017602,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0294177340540955,
      "loss": 0.232,
      "num_input_tokens_seen": 6909216,
      "step": 32740
    },
    {
      "epoch": 3.602310231023102,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02941733666000152,
      "loss": 0.2293,
      "num_input_tokens_seen": 6910272,
      "step": 32745
    },
    {
      "epoch": 3.6028602860286028,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029416939133029864,
      "loss": 0.2319,
      "num_input_tokens_seen": 6911264,
      "step": 32750
    },
    {
      "epoch": 3.6034103410341034,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0294165414731842,
      "loss": 0.2319,
      "num_input_tokens_seen": 6912384,
      "step": 32755
    },
    {
      "epoch": 3.603960396039604,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.029416143680468188,
      "loss": 0.2298,
      "num_input_tokens_seen": 6913440,
      "step": 32760
    },
    {
      "epoch": 3.6045104510451047,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029415745754885498,
      "loss": 0.2308,
      "num_input_tokens_seen": 6914432,
      "step": 32765
    },
    {
      "epoch": 3.605060506050605,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.029415347696439796,
      "loss": 0.2299,
      "num_input_tokens_seen": 6915456,
      "step": 32770
    },
    {
      "epoch": 3.6056105610561056,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029414949505134753,
      "loss": 0.2298,
      "num_input_tokens_seen": 6916544,
      "step": 32775
    },
    {
      "epoch": 3.606160616061606,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.02941455118097404,
      "loss": 0.2309,
      "num_input_tokens_seen": 6917600,
      "step": 32780
    },
    {
      "epoch": 3.606710671067107,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02941415272396132,
      "loss": 0.2293,
      "num_input_tokens_seen": 6918624,
      "step": 32785
    },
    {
      "epoch": 3.6072607260726075,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.029413754134100276,
      "loss": 0.2324,
      "num_input_tokens_seen": 6919744,
      "step": 32790
    },
    {
      "epoch": 3.6078107810781077,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.029413355411394573,
      "loss": 0.2314,
      "num_input_tokens_seen": 6920800,
      "step": 32795
    },
    {
      "epoch": 3.6083608360836084,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.029412956555847896,
      "loss": 0.2309,
      "num_input_tokens_seen": 6921952,
      "step": 32800
    },
    {
      "epoch": 3.608910891089109,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029412557567463905,
      "loss": 0.2324,
      "num_input_tokens_seen": 6923040,
      "step": 32805
    },
    {
      "epoch": 3.609460946094609,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.029412158446246294,
      "loss": 0.2298,
      "num_input_tokens_seen": 6924160,
      "step": 32810
    },
    {
      "epoch": 3.61001100110011,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.029411759192198732,
      "loss": 0.2314,
      "num_input_tokens_seen": 6925152,
      "step": 32815
    },
    {
      "epoch": 3.6105610561056105,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0294113598053249,
      "loss": 0.2335,
      "num_input_tokens_seen": 6926304,
      "step": 32820
    },
    {
      "epoch": 3.611111111111111,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.02941096028562848,
      "loss": 0.2329,
      "num_input_tokens_seen": 6927328,
      "step": 32825
    },
    {
      "epoch": 3.611661166116612,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.02941056063311315,
      "loss": 0.2298,
      "num_input_tokens_seen": 6928320,
      "step": 32830
    },
    {
      "epoch": 3.612211221122112,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.029410160847782602,
      "loss": 0.2303,
      "num_input_tokens_seen": 6929408,
      "step": 32835
    },
    {
      "epoch": 3.6127612761276127,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.029409760929640514,
      "loss": 0.2329,
      "num_input_tokens_seen": 6930464,
      "step": 32840
    },
    {
      "epoch": 3.6133113311331133,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029409360878690578,
      "loss": 0.2309,
      "num_input_tokens_seen": 6931552,
      "step": 32845
    },
    {
      "epoch": 3.613861386138614,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029408960694936474,
      "loss": 0.2299,
      "num_input_tokens_seen": 6932704,
      "step": 32850
    },
    {
      "epoch": 3.6144114411441146,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.029408560378381896,
      "loss": 0.2294,
      "num_input_tokens_seen": 6933728,
      "step": 32855
    },
    {
      "epoch": 3.614961496149615,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.02940815992903053,
      "loss": 0.2291,
      "num_input_tokens_seen": 6934752,
      "step": 32860
    },
    {
      "epoch": 3.6155115511551155,
      "grad_norm": 0.00830078125,
      "learning_rate": 0.029407759346886064,
      "loss": 0.2334,
      "num_input_tokens_seen": 6935840,
      "step": 32865
    },
    {
      "epoch": 3.616061606160616,
      "grad_norm": 0.0076904296875,
      "learning_rate": 0.029407358631952197,
      "loss": 0.2324,
      "num_input_tokens_seen": 6936864,
      "step": 32870
    },
    {
      "epoch": 3.6166116611661168,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.029406957784232617,
      "loss": 0.2312,
      "num_input_tokens_seen": 6937920,
      "step": 32875
    },
    {
      "epoch": 3.6171617161716174,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.02940655680373102,
      "loss": 0.2349,
      "num_input_tokens_seen": 6938976,
      "step": 32880
    },
    {
      "epoch": 3.6177117711771176,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029406155690451106,
      "loss": 0.2235,
      "num_input_tokens_seen": 6940000,
      "step": 32885
    },
    {
      "epoch": 3.6182618261826183,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.029405754444396564,
      "loss": 0.2327,
      "num_input_tokens_seen": 6941056,
      "step": 32890
    },
    {
      "epoch": 3.618811881188119,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.029405353065571096,
      "loss": 0.2344,
      "num_input_tokens_seen": 6942080,
      "step": 32895
    },
    {
      "epoch": 3.619361936193619,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.029404951553978403,
      "loss": 0.2306,
      "num_input_tokens_seen": 6943072,
      "step": 32900
    },
    {
      "epoch": 3.6199119911991198,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.02940454990962218,
      "loss": 0.2322,
      "num_input_tokens_seen": 6944096,
      "step": 32905
    },
    {
      "epoch": 3.6204620462046204,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.02940414813250614,
      "loss": 0.2352,
      "num_input_tokens_seen": 6945184,
      "step": 32910
    },
    {
      "epoch": 3.621012101210121,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.029403746222633968,
      "loss": 0.233,
      "num_input_tokens_seen": 6946208,
      "step": 32915
    },
    {
      "epoch": 3.6215621562156217,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029403344180009387,
      "loss": 0.2324,
      "num_input_tokens_seen": 6947264,
      "step": 32920
    },
    {
      "epoch": 3.622112211221122,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029402942004636095,
      "loss": 0.2324,
      "num_input_tokens_seen": 6948352,
      "step": 32925
    },
    {
      "epoch": 3.6226622662266226,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02940253969651779,
      "loss": 0.2319,
      "num_input_tokens_seen": 6949408,
      "step": 32930
    },
    {
      "epoch": 3.6232123212321232,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02940213725565819,
      "loss": 0.2308,
      "num_input_tokens_seen": 6950464,
      "step": 32935
    },
    {
      "epoch": 3.623762376237624,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029401734682061007,
      "loss": 0.2313,
      "num_input_tokens_seen": 6951552,
      "step": 32940
    },
    {
      "epoch": 3.6243124312431245,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.02940133197572994,
      "loss": 0.2308,
      "num_input_tokens_seen": 6952640,
      "step": 32945
    },
    {
      "epoch": 3.6248624862486247,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02940092913666871,
      "loss": 0.2313,
      "num_input_tokens_seen": 6953760,
      "step": 32950
    },
    {
      "epoch": 3.6254125412541254,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.029400526164881025,
      "loss": 0.2329,
      "num_input_tokens_seen": 6954848,
      "step": 32955
    },
    {
      "epoch": 3.625962596259626,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.0294001230603706,
      "loss": 0.2308,
      "num_input_tokens_seen": 6955968,
      "step": 32960
    },
    {
      "epoch": 3.6265126512651267,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02939971982314115,
      "loss": 0.2329,
      "num_input_tokens_seen": 6957024,
      "step": 32965
    },
    {
      "epoch": 3.6270627062706273,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0293993164531964,
      "loss": 0.2313,
      "num_input_tokens_seen": 6958016,
      "step": 32970
    },
    {
      "epoch": 3.6276127612761275,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.029398912950540052,
      "loss": 0.2324,
      "num_input_tokens_seen": 6959072,
      "step": 32975
    },
    {
      "epoch": 3.628162816281628,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029398509315175833,
      "loss": 0.2319,
      "num_input_tokens_seen": 6960096,
      "step": 32980
    },
    {
      "epoch": 3.628712871287129,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.029398105547107464,
      "loss": 0.2299,
      "num_input_tokens_seen": 6961120,
      "step": 32985
    },
    {
      "epoch": 3.629262926292629,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02939770164633866,
      "loss": 0.2319,
      "num_input_tokens_seen": 6962144,
      "step": 32990
    },
    {
      "epoch": 3.6298129812981297,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.029397297612873158,
      "loss": 0.2309,
      "num_input_tokens_seen": 6963264,
      "step": 32995
    },
    {
      "epoch": 3.6303630363036303,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029396893446714662,
      "loss": 0.2351,
      "num_input_tokens_seen": 6964320,
      "step": 33000
    },
    {
      "epoch": 3.630913091309131,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.029396489147866918,
      "loss": 0.2293,
      "num_input_tokens_seen": 6965376,
      "step": 33005
    },
    {
      "epoch": 3.6314631463146316,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.029396084716333634,
      "loss": 0.2293,
      "num_input_tokens_seen": 6966400,
      "step": 33010
    },
    {
      "epoch": 3.632013201320132,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.029395680152118547,
      "loss": 0.2314,
      "num_input_tokens_seen": 6967424,
      "step": 33015
    },
    {
      "epoch": 3.6325632563256325,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.029395275455225387,
      "loss": 0.2319,
      "num_input_tokens_seen": 6968416,
      "step": 33020
    },
    {
      "epoch": 3.633113311331133,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02939487062565788,
      "loss": 0.2324,
      "num_input_tokens_seen": 6969408,
      "step": 33025
    },
    {
      "epoch": 3.633663366336634,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029394465663419755,
      "loss": 0.2314,
      "num_input_tokens_seen": 6970464,
      "step": 33030
    },
    {
      "epoch": 3.6342134213421344,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.029394060568514745,
      "loss": 0.2288,
      "num_input_tokens_seen": 6971552,
      "step": 33035
    },
    {
      "epoch": 3.6347634763476346,
      "grad_norm": 0.001953125,
      "learning_rate": 0.029393655340946588,
      "loss": 0.234,
      "num_input_tokens_seen": 6972640,
      "step": 33040
    },
    {
      "epoch": 3.6353135313531353,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029393249980719016,
      "loss": 0.2329,
      "num_input_tokens_seen": 6973792,
      "step": 33045
    },
    {
      "epoch": 3.635863586358636,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02939284448783577,
      "loss": 0.2324,
      "num_input_tokens_seen": 6974848,
      "step": 33050
    },
    {
      "epoch": 3.636413641364136,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029392438862300578,
      "loss": 0.2324,
      "num_input_tokens_seen": 6975904,
      "step": 33055
    },
    {
      "epoch": 3.6369636963696372,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029392033104117185,
      "loss": 0.2319,
      "num_input_tokens_seen": 6977056,
      "step": 33060
    },
    {
      "epoch": 3.6375137513751374,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029391627213289327,
      "loss": 0.2324,
      "num_input_tokens_seen": 6978048,
      "step": 33065
    },
    {
      "epoch": 3.638063806380638,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.02939122118982075,
      "loss": 0.2298,
      "num_input_tokens_seen": 6979136,
      "step": 33070
    },
    {
      "epoch": 3.6386138613861387,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02939081503371519,
      "loss": 0.2319,
      "num_input_tokens_seen": 6980224,
      "step": 33075
    },
    {
      "epoch": 3.639163916391639,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029390408744976396,
      "loss": 0.2313,
      "num_input_tokens_seen": 6981312,
      "step": 33080
    },
    {
      "epoch": 3.6397139713971396,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029390002323608107,
      "loss": 0.2308,
      "num_input_tokens_seen": 6982368,
      "step": 33085
    },
    {
      "epoch": 3.6402640264026402,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029389595769614075,
      "loss": 0.2318,
      "num_input_tokens_seen": 6983456,
      "step": 33090
    },
    {
      "epoch": 3.640814081408141,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02938918908299804,
      "loss": 0.2329,
      "num_input_tokens_seen": 6984544,
      "step": 33095
    },
    {
      "epoch": 3.6413641364136415,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029388782263763756,
      "loss": 0.2303,
      "num_input_tokens_seen": 6985600,
      "step": 33100
    },
    {
      "epoch": 3.6419141914191417,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.029388375311914967,
      "loss": 0.2308,
      "num_input_tokens_seen": 6986624,
      "step": 33105
    },
    {
      "epoch": 3.6424642464246424,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02938796822745543,
      "loss": 0.2287,
      "num_input_tokens_seen": 6987648,
      "step": 33110
    },
    {
      "epoch": 3.643014301430143,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029387561010388894,
      "loss": 0.2318,
      "num_input_tokens_seen": 6988672,
      "step": 33115
    },
    {
      "epoch": 3.6435643564356437,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029387153660719107,
      "loss": 0.2314,
      "num_input_tokens_seen": 6989696,
      "step": 33120
    },
    {
      "epoch": 3.6441144114411443,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.029386746178449837,
      "loss": 0.2324,
      "num_input_tokens_seen": 6990720,
      "step": 33125
    },
    {
      "epoch": 3.6446644664466445,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029386338563584826,
      "loss": 0.2319,
      "num_input_tokens_seen": 6991744,
      "step": 33130
    },
    {
      "epoch": 3.645214521452145,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029385930816127837,
      "loss": 0.234,
      "num_input_tokens_seen": 6992768,
      "step": 33135
    },
    {
      "epoch": 3.645764576457646,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.029385522936082622,
      "loss": 0.2329,
      "num_input_tokens_seen": 6993760,
      "step": 33140
    },
    {
      "epoch": 3.646314631463146,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02938511492345295,
      "loss": 0.2314,
      "num_input_tokens_seen": 6994816,
      "step": 33145
    },
    {
      "epoch": 3.6468646864686467,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029384706778242575,
      "loss": 0.2308,
      "num_input_tokens_seen": 6995872,
      "step": 33150
    },
    {
      "epoch": 3.6474147414741473,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02938429850045526,
      "loss": 0.2313,
      "num_input_tokens_seen": 6996928,
      "step": 33155
    },
    {
      "epoch": 3.647964796479648,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.02938389009009477,
      "loss": 0.2329,
      "num_input_tokens_seen": 6998016,
      "step": 33160
    },
    {
      "epoch": 3.6485148514851486,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.029383481547164865,
      "loss": 0.2318,
      "num_input_tokens_seen": 6999104,
      "step": 33165
    },
    {
      "epoch": 3.649064906490649,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029383072871669313,
      "loss": 0.2329,
      "num_input_tokens_seen": 7000160,
      "step": 33170
    },
    {
      "epoch": 3.6496149614961495,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02938266406361188,
      "loss": 0.2298,
      "num_input_tokens_seen": 7001216,
      "step": 33175
    },
    {
      "epoch": 3.65016501650165,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.02938225512299633,
      "loss": 0.2313,
      "num_input_tokens_seen": 7002272,
      "step": 33180
    },
    {
      "epoch": 3.650715071507151,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02938184604982644,
      "loss": 0.2313,
      "num_input_tokens_seen": 7003328,
      "step": 33185
    },
    {
      "epoch": 3.6512651265126514,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.029381436844105976,
      "loss": 0.2309,
      "num_input_tokens_seen": 7004384,
      "step": 33190
    },
    {
      "epoch": 3.6518151815181517,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02938102750583871,
      "loss": 0.2313,
      "num_input_tokens_seen": 7005376,
      "step": 33195
    },
    {
      "epoch": 3.6523652365236523,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.029380618035028415,
      "loss": 0.2324,
      "num_input_tokens_seen": 7006432,
      "step": 33200
    },
    {
      "epoch": 3.652915291529153,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.02938020843167886,
      "loss": 0.2303,
      "num_input_tokens_seen": 7007456,
      "step": 33205
    },
    {
      "epoch": 3.6534653465346536,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.029379798695793827,
      "loss": 0.2288,
      "num_input_tokens_seen": 7008480,
      "step": 33210
    },
    {
      "epoch": 3.6540154015401543,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02937938882737709,
      "loss": 0.2288,
      "num_input_tokens_seen": 7009504,
      "step": 33215
    },
    {
      "epoch": 3.6545654565456545,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.029378978826432423,
      "loss": 0.2288,
      "num_input_tokens_seen": 7010560,
      "step": 33220
    },
    {
      "epoch": 3.655115511551155,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.029378568692963607,
      "loss": 0.2289,
      "num_input_tokens_seen": 7011648,
      "step": 33225
    },
    {
      "epoch": 3.6556655665566558,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.029378158426974426,
      "loss": 0.2337,
      "num_input_tokens_seen": 7012672,
      "step": 33230
    },
    {
      "epoch": 3.656215621562156,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.02937774802846866,
      "loss": 0.2364,
      "num_input_tokens_seen": 7013664,
      "step": 33235
    },
    {
      "epoch": 3.6567656765676566,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.029377337497450088,
      "loss": 0.2296,
      "num_input_tokens_seen": 7014720,
      "step": 33240
    },
    {
      "epoch": 3.6573157315731573,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.029376926833922495,
      "loss": 0.2338,
      "num_input_tokens_seen": 7015712,
      "step": 33245
    },
    {
      "epoch": 3.657865786578658,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.029376516037889665,
      "loss": 0.2374,
      "num_input_tokens_seen": 7016736,
      "step": 33250
    },
    {
      "epoch": 3.6584158415841586,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.029376105109355385,
      "loss": 0.2341,
      "num_input_tokens_seen": 7017728,
      "step": 33255
    },
    {
      "epoch": 3.6589658965896588,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.02937569404832344,
      "loss": 0.2361,
      "num_input_tokens_seen": 7018784,
      "step": 33260
    },
    {
      "epoch": 3.6595159515951594,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.029375282854797624,
      "loss": 0.2314,
      "num_input_tokens_seen": 7019840,
      "step": 33265
    },
    {
      "epoch": 3.66006600660066,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029374871528781727,
      "loss": 0.2319,
      "num_input_tokens_seen": 7020896,
      "step": 33270
    },
    {
      "epoch": 3.6606160616061607,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029374460070279534,
      "loss": 0.235,
      "num_input_tokens_seen": 7021984,
      "step": 33275
    },
    {
      "epoch": 3.6611661166116614,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.02937404847929484,
      "loss": 0.2324,
      "num_input_tokens_seen": 7023104,
      "step": 33280
    },
    {
      "epoch": 3.6617161716171616,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.029373636755831437,
      "loss": 0.233,
      "num_input_tokens_seen": 7024224,
      "step": 33285
    },
    {
      "epoch": 3.662266226622662,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02937322489989312,
      "loss": 0.2309,
      "num_input_tokens_seen": 7025280,
      "step": 33290
    },
    {
      "epoch": 3.662816281628163,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.029372812911483696,
      "loss": 0.2319,
      "num_input_tokens_seen": 7026304,
      "step": 33295
    },
    {
      "epoch": 3.6633663366336635,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.029372400790606944,
      "loss": 0.2298,
      "num_input_tokens_seen": 7027392,
      "step": 33300
    },
    {
      "epoch": 3.663916391639164,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.029371988537266674,
      "loss": 0.2298,
      "num_input_tokens_seen": 7028512,
      "step": 33305
    },
    {
      "epoch": 3.6644664466446644,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.029371576151466676,
      "loss": 0.2304,
      "num_input_tokens_seen": 7029568,
      "step": 33310
    },
    {
      "epoch": 3.665016501650165,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.02937116363321076,
      "loss": 0.2324,
      "num_input_tokens_seen": 7030592,
      "step": 33315
    },
    {
      "epoch": 3.6655665566556657,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.029370750982502726,
      "loss": 0.2309,
      "num_input_tokens_seen": 7031712,
      "step": 33320
    },
    {
      "epoch": 3.666116611661166,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.029370338199346372,
      "loss": 0.2303,
      "num_input_tokens_seen": 7032768,
      "step": 33325
    },
    {
      "epoch": 3.6666666666666665,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02936992528374551,
      "loss": 0.2283,
      "num_input_tokens_seen": 7033824,
      "step": 33330
    },
    {
      "epoch": 3.667216721672167,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029369512235703943,
      "loss": 0.2299,
      "num_input_tokens_seen": 7034880,
      "step": 33335
    },
    {
      "epoch": 3.667766776677668,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029369099055225476,
      "loss": 0.2293,
      "num_input_tokens_seen": 7035840,
      "step": 33340
    },
    {
      "epoch": 3.6683168316831685,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029368685742313914,
      "loss": 0.2304,
      "num_input_tokens_seen": 7036960,
      "step": 33345
    },
    {
      "epoch": 3.6688668866886687,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029368272296973075,
      "loss": 0.2304,
      "num_input_tokens_seen": 7038080,
      "step": 33350
    },
    {
      "epoch": 3.6694169416941693,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029367858719206764,
      "loss": 0.2351,
      "num_input_tokens_seen": 7039200,
      "step": 33355
    },
    {
      "epoch": 3.66996699669967,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02936744500901879,
      "loss": 0.2314,
      "num_input_tokens_seen": 7040288,
      "step": 33360
    },
    {
      "epoch": 3.6705170517051706,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.029367031166412972,
      "loss": 0.2304,
      "num_input_tokens_seen": 7041408,
      "step": 33365
    },
    {
      "epoch": 3.6710671067106713,
      "grad_norm": 0.000782012939453125,
      "learning_rate": 0.029366617191393125,
      "loss": 0.2309,
      "num_input_tokens_seen": 7042464,
      "step": 33370
    },
    {
      "epoch": 3.6716171617161715,
      "grad_norm": 0.00080108642578125,
      "learning_rate": 0.02936620308396306,
      "loss": 0.2314,
      "num_input_tokens_seen": 7043424,
      "step": 33375
    },
    {
      "epoch": 3.672167216721672,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02936578884412659,
      "loss": 0.2283,
      "num_input_tokens_seen": 7044480,
      "step": 33380
    },
    {
      "epoch": 3.6727172717271728,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.029365374471887543,
      "loss": 0.2304,
      "num_input_tokens_seen": 7045472,
      "step": 33385
    },
    {
      "epoch": 3.6732673267326734,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.029364959967249728,
      "loss": 0.2294,
      "num_input_tokens_seen": 7046496,
      "step": 33390
    },
    {
      "epoch": 3.673817381738174,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.02936454533021697,
      "loss": 0.2326,
      "num_input_tokens_seen": 7047584,
      "step": 33395
    },
    {
      "epoch": 3.6743674367436743,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.02936413056079309,
      "loss": 0.232,
      "num_input_tokens_seen": 7048672,
      "step": 33400
    },
    {
      "epoch": 3.674917491749175,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02936371565898192,
      "loss": 0.2304,
      "num_input_tokens_seen": 7049664,
      "step": 33405
    },
    {
      "epoch": 3.6754675467546756,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.029363300624787265,
      "loss": 0.2284,
      "num_input_tokens_seen": 7050688,
      "step": 33410
    },
    {
      "epoch": 3.676017601760176,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029362885458212966,
      "loss": 0.2268,
      "num_input_tokens_seen": 7051712,
      "step": 33415
    },
    {
      "epoch": 3.6765676567656764,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029362470159262844,
      "loss": 0.23,
      "num_input_tokens_seen": 7052736,
      "step": 33420
    },
    {
      "epoch": 3.677117711771177,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02936205472794072,
      "loss": 0.2337,
      "num_input_tokens_seen": 7053824,
      "step": 33425
    },
    {
      "epoch": 3.6776677667766777,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.029361639164250435,
      "loss": 0.2322,
      "num_input_tokens_seen": 7054880,
      "step": 33430
    },
    {
      "epoch": 3.6782178217821784,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.029361223468195814,
      "loss": 0.23,
      "num_input_tokens_seen": 7055872,
      "step": 33435
    },
    {
      "epoch": 3.6787678767876786,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.029360807639780682,
      "loss": 0.2326,
      "num_input_tokens_seen": 7056992,
      "step": 33440
    },
    {
      "epoch": 3.6793179317931792,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029360391679008883,
      "loss": 0.2279,
      "num_input_tokens_seen": 7057984,
      "step": 33445
    },
    {
      "epoch": 3.67986798679868,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02935997558588424,
      "loss": 0.23,
      "num_input_tokens_seen": 7059008,
      "step": 33450
    },
    {
      "epoch": 3.6804180418041805,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.029359559360410598,
      "loss": 0.2347,
      "num_input_tokens_seen": 7060064,
      "step": 33455
    },
    {
      "epoch": 3.680968096809681,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.029359143002591785,
      "loss": 0.2363,
      "num_input_tokens_seen": 7061152,
      "step": 33460
    },
    {
      "epoch": 3.6815181518151814,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.029358726512431637,
      "loss": 0.2325,
      "num_input_tokens_seen": 7062272,
      "step": 33465
    },
    {
      "epoch": 3.682068206820682,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.029358309889934,
      "loss": 0.2273,
      "num_input_tokens_seen": 7063360,
      "step": 33470
    },
    {
      "epoch": 3.6826182618261827,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.029357893135102713,
      "loss": 0.2288,
      "num_input_tokens_seen": 7064448,
      "step": 33475
    },
    {
      "epoch": 3.6831683168316833,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.02935747624794161,
      "loss": 0.2372,
      "num_input_tokens_seen": 7065408,
      "step": 33480
    },
    {
      "epoch": 3.683718371837184,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.02935705922845454,
      "loss": 0.2284,
      "num_input_tokens_seen": 7066400,
      "step": 33485
    },
    {
      "epoch": 3.684268426842684,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02935664207664534,
      "loss": 0.2351,
      "num_input_tokens_seen": 7067520,
      "step": 33490
    },
    {
      "epoch": 3.684818481848185,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02935622479251787,
      "loss": 0.2356,
      "num_input_tokens_seen": 7068512,
      "step": 33495
    },
    {
      "epoch": 3.6853685368536855,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.029355807376075956,
      "loss": 0.2309,
      "num_input_tokens_seen": 7069568,
      "step": 33500
    },
    {
      "epoch": 3.6859185918591857,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.029355389827323455,
      "loss": 0.2314,
      "num_input_tokens_seen": 7070528,
      "step": 33505
    },
    {
      "epoch": 3.6864686468646863,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.029354972146264216,
      "loss": 0.232,
      "num_input_tokens_seen": 7071520,
      "step": 33510
    },
    {
      "epoch": 3.687018701870187,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029354554332902086,
      "loss": 0.2309,
      "num_input_tokens_seen": 7072640,
      "step": 33515
    },
    {
      "epoch": 3.6875687568756876,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02935413638724092,
      "loss": 0.2304,
      "num_input_tokens_seen": 7073728,
      "step": 33520
    },
    {
      "epoch": 3.6881188118811883,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.029353718309284564,
      "loss": 0.2314,
      "num_input_tokens_seen": 7074752,
      "step": 33525
    },
    {
      "epoch": 3.6886688668866885,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02935330009903687,
      "loss": 0.2283,
      "num_input_tokens_seen": 7075776,
      "step": 33530
    },
    {
      "epoch": 3.689218921892189,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.029352881756501702,
      "loss": 0.2319,
      "num_input_tokens_seen": 7076896,
      "step": 33535
    },
    {
      "epoch": 3.68976897689769,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.02935246328168291,
      "loss": 0.2314,
      "num_input_tokens_seen": 7077920,
      "step": 33540
    },
    {
      "epoch": 3.6903190319031904,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029352044674584344,
      "loss": 0.233,
      "num_input_tokens_seen": 7078944,
      "step": 33545
    },
    {
      "epoch": 3.690869086908691,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.029351625935209874,
      "loss": 0.2294,
      "num_input_tokens_seen": 7080000,
      "step": 33550
    },
    {
      "epoch": 3.6914191419141913,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.029351207063563357,
      "loss": 0.232,
      "num_input_tokens_seen": 7080992,
      "step": 33555
    },
    {
      "epoch": 3.691969196919692,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.02935078805964865,
      "loss": 0.2325,
      "num_input_tokens_seen": 7082048,
      "step": 33560
    },
    {
      "epoch": 3.6925192519251926,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.02935036892346961,
      "loss": 0.2314,
      "num_input_tokens_seen": 7083136,
      "step": 33565
    },
    {
      "epoch": 3.693069306930693,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02934994965503011,
      "loss": 0.234,
      "num_input_tokens_seen": 7084160,
      "step": 33570
    },
    {
      "epoch": 3.693619361936194,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.029349530254334007,
      "loss": 0.2324,
      "num_input_tokens_seen": 7085184,
      "step": 33575
    },
    {
      "epoch": 3.694169416941694,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.029349110721385175,
      "loss": 0.2314,
      "num_input_tokens_seen": 7086240,
      "step": 33580
    },
    {
      "epoch": 3.6947194719471947,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.02934869105618747,
      "loss": 0.2319,
      "num_input_tokens_seen": 7087232,
      "step": 33585
    },
    {
      "epoch": 3.6952695269526954,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.029348271258744767,
      "loss": 0.2314,
      "num_input_tokens_seen": 7088352,
      "step": 33590
    },
    {
      "epoch": 3.6958195819581956,
      "grad_norm": 0.0008392333984375,
      "learning_rate": 0.029347851329060932,
      "loss": 0.2304,
      "num_input_tokens_seen": 7089312,
      "step": 33595
    },
    {
      "epoch": 3.6963696369636962,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.029347431267139838,
      "loss": 0.2304,
      "num_input_tokens_seen": 7090272,
      "step": 33600
    },
    {
      "epoch": 3.696919691969197,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02934701107298535,
      "loss": 0.2314,
      "num_input_tokens_seen": 7091424,
      "step": 33605
    },
    {
      "epoch": 3.6974697469746975,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02934659074660135,
      "loss": 0.2299,
      "num_input_tokens_seen": 7092512,
      "step": 33610
    },
    {
      "epoch": 3.698019801980198,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0293461702879917,
      "loss": 0.2289,
      "num_input_tokens_seen": 7093536,
      "step": 33615
    },
    {
      "epoch": 3.6985698569856984,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.02934574969716029,
      "loss": 0.2294,
      "num_input_tokens_seen": 7094560,
      "step": 33620
    },
    {
      "epoch": 3.699119911991199,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029345328974110983,
      "loss": 0.2358,
      "num_input_tokens_seen": 7095616,
      "step": 33625
    },
    {
      "epoch": 3.6996699669966997,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.029344908118847667,
      "loss": 0.2321,
      "num_input_tokens_seen": 7096704,
      "step": 33630
    },
    {
      "epoch": 3.7002200220022003,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029344487131374216,
      "loss": 0.2259,
      "num_input_tokens_seen": 7097856,
      "step": 33635
    },
    {
      "epoch": 3.700770077007701,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02934406601169451,
      "loss": 0.2274,
      "num_input_tokens_seen": 7098944,
      "step": 33640
    },
    {
      "epoch": 3.701320132013201,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.029343644759812427,
      "loss": 0.2246,
      "num_input_tokens_seen": 7100000,
      "step": 33645
    },
    {
      "epoch": 3.701870187018702,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.029343223375731857,
      "loss": 0.23,
      "num_input_tokens_seen": 7101024,
      "step": 33650
    },
    {
      "epoch": 3.7024202420242025,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.029342801859456678,
      "loss": 0.2234,
      "num_input_tokens_seen": 7102080,
      "step": 33655
    },
    {
      "epoch": 3.7029702970297027,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.029342380210990775,
      "loss": 0.2399,
      "num_input_tokens_seen": 7103104,
      "step": 33660
    },
    {
      "epoch": 3.7035203520352034,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.029341958430338037,
      "loss": 0.2267,
      "num_input_tokens_seen": 7104224,
      "step": 33665
    },
    {
      "epoch": 3.704070407040704,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.029341536517502355,
      "loss": 0.233,
      "num_input_tokens_seen": 7105216,
      "step": 33670
    },
    {
      "epoch": 3.7046204620462047,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.029341114472487606,
      "loss": 0.2297,
      "num_input_tokens_seen": 7106272,
      "step": 33675
    },
    {
      "epoch": 3.7051705170517053,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.02934069229529769,
      "loss": 0.2318,
      "num_input_tokens_seen": 7107264,
      "step": 33680
    },
    {
      "epoch": 3.7057205720572055,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.02934026998593649,
      "loss": 0.2261,
      "num_input_tokens_seen": 7108320,
      "step": 33685
    },
    {
      "epoch": 3.706270627062706,
      "grad_norm": 0.014892578125,
      "learning_rate": 0.029339847544407906,
      "loss": 0.2299,
      "num_input_tokens_seen": 7109376,
      "step": 33690
    },
    {
      "epoch": 3.706820682068207,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.02933942497071583,
      "loss": 0.2294,
      "num_input_tokens_seen": 7110464,
      "step": 33695
    },
    {
      "epoch": 3.7073707370737075,
      "grad_norm": 0.0028533935546875,
      "learning_rate": 0.029339002264864154,
      "loss": 0.2295,
      "num_input_tokens_seen": 7111520,
      "step": 33700
    },
    {
      "epoch": 3.707920792079208,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.029338579426856772,
      "loss": 0.2436,
      "num_input_tokens_seen": 7112544,
      "step": 33705
    },
    {
      "epoch": 3.7084708470847083,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.02933815645669759,
      "loss": 0.2378,
      "num_input_tokens_seen": 7113632,
      "step": 33710
    },
    {
      "epoch": 3.709020902090209,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.029337733354390494,
      "loss": 0.2316,
      "num_input_tokens_seen": 7114720,
      "step": 33715
    },
    {
      "epoch": 3.7095709570957096,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02933731011993939,
      "loss": 0.229,
      "num_input_tokens_seen": 7115744,
      "step": 33720
    },
    {
      "epoch": 3.7101210121012103,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.02933688675334818,
      "loss": 0.233,
      "num_input_tokens_seen": 7116800,
      "step": 33725
    },
    {
      "epoch": 3.710671067106711,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.029336463254620762,
      "loss": 0.233,
      "num_input_tokens_seen": 7117824,
      "step": 33730
    },
    {
      "epoch": 3.711221122112211,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.029336039623761044,
      "loss": 0.233,
      "num_input_tokens_seen": 7118880,
      "step": 33735
    },
    {
      "epoch": 3.7117711771177118,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.029335615860772924,
      "loss": 0.2319,
      "num_input_tokens_seen": 7119936,
      "step": 33740
    },
    {
      "epoch": 3.7123212321232124,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.029335191965660316,
      "loss": 0.2308,
      "num_input_tokens_seen": 7121056,
      "step": 33745
    },
    {
      "epoch": 3.7128712871287126,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029334767938427122,
      "loss": 0.2282,
      "num_input_tokens_seen": 7122080,
      "step": 33750
    },
    {
      "epoch": 3.7134213421342133,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.02933434377907725,
      "loss": 0.2277,
      "num_input_tokens_seen": 7123072,
      "step": 33755
    },
    {
      "epoch": 3.713971397139714,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.02933391948761461,
      "loss": 0.2294,
      "num_input_tokens_seen": 7124096,
      "step": 33760
    },
    {
      "epoch": 3.7145214521452146,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.02933349506404311,
      "loss": 0.2341,
      "num_input_tokens_seen": 7125184,
      "step": 33765
    },
    {
      "epoch": 3.715071507150715,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.029333070508366664,
      "loss": 0.2317,
      "num_input_tokens_seen": 7126240,
      "step": 33770
    },
    {
      "epoch": 3.7156215621562154,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.029332645820589184,
      "loss": 0.239,
      "num_input_tokens_seen": 7127296,
      "step": 33775
    },
    {
      "epoch": 3.716171617161716,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.029332221000714587,
      "loss": 0.2325,
      "num_input_tokens_seen": 7128352,
      "step": 33780
    },
    {
      "epoch": 3.7167216721672167,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029331796048746788,
      "loss": 0.2315,
      "num_input_tokens_seen": 7129376,
      "step": 33785
    },
    {
      "epoch": 3.7172717271727174,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.0293313709646897,
      "loss": 0.2304,
      "num_input_tokens_seen": 7130432,
      "step": 33790
    },
    {
      "epoch": 3.717821782178218,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.029330945748547242,
      "loss": 0.233,
      "num_input_tokens_seen": 7131520,
      "step": 33795
    },
    {
      "epoch": 3.718371837183718,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029330520400323332,
      "loss": 0.2345,
      "num_input_tokens_seen": 7132608,
      "step": 33800
    },
    {
      "epoch": 3.718921892189219,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.029330094920021896,
      "loss": 0.2314,
      "num_input_tokens_seen": 7133568,
      "step": 33805
    },
    {
      "epoch": 3.7194719471947195,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.029329669307646848,
      "loss": 0.2318,
      "num_input_tokens_seen": 7134592,
      "step": 33810
    },
    {
      "epoch": 3.72002200220022,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029329243563202115,
      "loss": 0.2313,
      "num_input_tokens_seen": 7135584,
      "step": 33815
    },
    {
      "epoch": 3.720572057205721,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029328817686691624,
      "loss": 0.2339,
      "num_input_tokens_seen": 7136640,
      "step": 33820
    },
    {
      "epoch": 3.721122112211221,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.029328391678119292,
      "loss": 0.2313,
      "num_input_tokens_seen": 7137696,
      "step": 33825
    },
    {
      "epoch": 3.7216721672167217,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.029327965537489047,
      "loss": 0.2307,
      "num_input_tokens_seen": 7138752,
      "step": 33830
    },
    {
      "epoch": 3.7222222222222223,
      "grad_norm": 0.0026702880859375,
      "learning_rate": 0.029327539264804824,
      "loss": 0.2308,
      "num_input_tokens_seen": 7139840,
      "step": 33835
    },
    {
      "epoch": 3.7227722772277225,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02932711286007054,
      "loss": 0.2308,
      "num_input_tokens_seen": 7140832,
      "step": 33840
    },
    {
      "epoch": 3.723322332233223,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02932668632329014,
      "loss": 0.2297,
      "num_input_tokens_seen": 7141824,
      "step": 33845
    },
    {
      "epoch": 3.723872387238724,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.02932625965446754,
      "loss": 0.2313,
      "num_input_tokens_seen": 7142880,
      "step": 33850
    },
    {
      "epoch": 3.7244224422442245,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02932583285360668,
      "loss": 0.2319,
      "num_input_tokens_seen": 7144000,
      "step": 33855
    },
    {
      "epoch": 3.724972497249725,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.029325405920711495,
      "loss": 0.2283,
      "num_input_tokens_seen": 7145152,
      "step": 33860
    },
    {
      "epoch": 3.7255225522552253,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02932497885578592,
      "loss": 0.2319,
      "num_input_tokens_seen": 7146176,
      "step": 33865
    },
    {
      "epoch": 3.726072607260726,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029324551658833888,
      "loss": 0.2303,
      "num_input_tokens_seen": 7147200,
      "step": 33870
    },
    {
      "epoch": 3.7266226622662266,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029324124329859334,
      "loss": 0.233,
      "num_input_tokens_seen": 7148256,
      "step": 33875
    },
    {
      "epoch": 3.7271727172717273,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.029323696868866198,
      "loss": 0.2304,
      "num_input_tokens_seen": 7149408,
      "step": 33880
    },
    {
      "epoch": 3.727722772277228,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.029323269275858425,
      "loss": 0.2288,
      "num_input_tokens_seen": 7150432,
      "step": 33885
    },
    {
      "epoch": 3.728272827282728,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02932284155083995,
      "loss": 0.2319,
      "num_input_tokens_seen": 7151456,
      "step": 33890
    },
    {
      "epoch": 3.728822882288229,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02932241369381472,
      "loss": 0.2293,
      "num_input_tokens_seen": 7152544,
      "step": 33895
    },
    {
      "epoch": 3.7293729372937294,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029321985704786672,
      "loss": 0.2304,
      "num_input_tokens_seen": 7153568,
      "step": 33900
    },
    {
      "epoch": 3.72992299229923,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029321557583759754,
      "loss": 0.2325,
      "num_input_tokens_seen": 7154560,
      "step": 33905
    },
    {
      "epoch": 3.7304730473047307,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.029321129330737914,
      "loss": 0.23,
      "num_input_tokens_seen": 7155584,
      "step": 33910
    },
    {
      "epoch": 3.731023102310231,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029320700945725097,
      "loss": 0.2356,
      "num_input_tokens_seen": 7156608,
      "step": 33915
    },
    {
      "epoch": 3.7315731573157316,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029320272428725244,
      "loss": 0.2298,
      "num_input_tokens_seen": 7157696,
      "step": 33920
    },
    {
      "epoch": 3.7321232123212322,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.029319843779742324,
      "loss": 0.233,
      "num_input_tokens_seen": 7158752,
      "step": 33925
    },
    {
      "epoch": 3.7326732673267324,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02931941499878027,
      "loss": 0.2314,
      "num_input_tokens_seen": 7159776,
      "step": 33930
    },
    {
      "epoch": 3.733223322332233,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.029318986085843032,
      "loss": 0.2319,
      "num_input_tokens_seen": 7160864,
      "step": 33935
    },
    {
      "epoch": 3.7337733773377337,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.02931855704093458,
      "loss": 0.2309,
      "num_input_tokens_seen": 7161984,
      "step": 33940
    },
    {
      "epoch": 3.7343234323432344,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.029318127864058855,
      "loss": 0.2303,
      "num_input_tokens_seen": 7163008,
      "step": 33945
    },
    {
      "epoch": 3.734873487348735,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029317698555219818,
      "loss": 0.2278,
      "num_input_tokens_seen": 7164000,
      "step": 33950
    },
    {
      "epoch": 3.7354235423542352,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.029317269114421417,
      "loss": 0.2299,
      "num_input_tokens_seen": 7165088,
      "step": 33955
    },
    {
      "epoch": 3.735973597359736,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02931683954166762,
      "loss": 0.2298,
      "num_input_tokens_seen": 7166112,
      "step": 33960
    },
    {
      "epoch": 3.7365236523652365,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02931640983696239,
      "loss": 0.2335,
      "num_input_tokens_seen": 7167136,
      "step": 33965
    },
    {
      "epoch": 3.737073707370737,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02931598000030967,
      "loss": 0.2309,
      "num_input_tokens_seen": 7168192,
      "step": 33970
    },
    {
      "epoch": 3.737623762376238,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.029315550031713437,
      "loss": 0.2319,
      "num_input_tokens_seen": 7169280,
      "step": 33975
    },
    {
      "epoch": 3.738173817381738,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029315119931177646,
      "loss": 0.2288,
      "num_input_tokens_seen": 7170304,
      "step": 33980
    },
    {
      "epoch": 3.7387238723872387,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029314689698706268,
      "loss": 0.2319,
      "num_input_tokens_seen": 7171296,
      "step": 33985
    },
    {
      "epoch": 3.7392739273927393,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.02931425933430326,
      "loss": 0.2293,
      "num_input_tokens_seen": 7172416,
      "step": 33990
    },
    {
      "epoch": 3.73982398239824,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029313828837972594,
      "loss": 0.2304,
      "num_input_tokens_seen": 7173440,
      "step": 33995
    },
    {
      "epoch": 3.7403740374037406,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.029313398209718233,
      "loss": 0.2304,
      "num_input_tokens_seen": 7174528,
      "step": 34000
    },
    {
      "epoch": 3.740924092409241,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02931296744954415,
      "loss": 0.231,
      "num_input_tokens_seen": 7175584,
      "step": 34005
    },
    {
      "epoch": 3.7414741474147415,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.02931253655745432,
      "loss": 0.2314,
      "num_input_tokens_seen": 7176736,
      "step": 34010
    },
    {
      "epoch": 3.742024202420242,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0293121055334527,
      "loss": 0.2341,
      "num_input_tokens_seen": 7177728,
      "step": 34015
    },
    {
      "epoch": 3.7425742574257423,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.029311674377543277,
      "loss": 0.2288,
      "num_input_tokens_seen": 7178752,
      "step": 34020
    },
    {
      "epoch": 3.743124312431243,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.029311243089730016,
      "loss": 0.2288,
      "num_input_tokens_seen": 7179840,
      "step": 34025
    },
    {
      "epoch": 3.7436743674367436,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029310811670016895,
      "loss": 0.2299,
      "num_input_tokens_seen": 7180928,
      "step": 34030
    },
    {
      "epoch": 3.7442244224422443,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.02931038011840789,
      "loss": 0.2362,
      "num_input_tokens_seen": 7181920,
      "step": 34035
    },
    {
      "epoch": 3.744774477447745,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.02930994843490698,
      "loss": 0.2268,
      "num_input_tokens_seen": 7182976,
      "step": 34040
    },
    {
      "epoch": 3.745324532453245,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.02930951661951814,
      "loss": 0.2321,
      "num_input_tokens_seen": 7184064,
      "step": 34045
    },
    {
      "epoch": 3.745874587458746,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02930908467224535,
      "loss": 0.229,
      "num_input_tokens_seen": 7185120,
      "step": 34050
    },
    {
      "epoch": 3.7464246424642464,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.029308652593092593,
      "loss": 0.2358,
      "num_input_tokens_seen": 7186176,
      "step": 34055
    },
    {
      "epoch": 3.746974697469747,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.029308220382063852,
      "loss": 0.2289,
      "num_input_tokens_seen": 7187200,
      "step": 34060
    },
    {
      "epoch": 3.7475247524752477,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.02930778803916311,
      "loss": 0.2279,
      "num_input_tokens_seen": 7188320,
      "step": 34065
    },
    {
      "epoch": 3.748074807480748,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029307355564394358,
      "loss": 0.23,
      "num_input_tokens_seen": 7189376,
      "step": 34070
    },
    {
      "epoch": 3.7486248624862486,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029306922957761564,
      "loss": 0.2373,
      "num_input_tokens_seen": 7190400,
      "step": 34075
    },
    {
      "epoch": 3.7491749174917492,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02930649021926873,
      "loss": 0.2269,
      "num_input_tokens_seen": 7191488,
      "step": 34080
    },
    {
      "epoch": 3.7497249724972495,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02930605734891984,
      "loss": 0.2311,
      "num_input_tokens_seen": 7192576,
      "step": 34085
    },
    {
      "epoch": 3.7502750275027505,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02930562434671889,
      "loss": 0.2317,
      "num_input_tokens_seen": 7193632,
      "step": 34090
    },
    {
      "epoch": 3.7508250825082508,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.029305191212669855,
      "loss": 0.2352,
      "num_input_tokens_seen": 7194656,
      "step": 34095
    },
    {
      "epoch": 3.7513751375137514,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.029304757946776745,
      "loss": 0.2284,
      "num_input_tokens_seen": 7195776,
      "step": 34100
    },
    {
      "epoch": 3.751925192519252,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029304324549043543,
      "loss": 0.2285,
      "num_input_tokens_seen": 7196864,
      "step": 34105
    },
    {
      "epoch": 3.7524752475247523,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.02930389101947424,
      "loss": 0.2279,
      "num_input_tokens_seen": 7197856,
      "step": 34110
    },
    {
      "epoch": 3.753025302530253,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.029303457358072844,
      "loss": 0.2322,
      "num_input_tokens_seen": 7198880,
      "step": 34115
    },
    {
      "epoch": 3.7535753575357536,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.029303023564843343,
      "loss": 0.227,
      "num_input_tokens_seen": 7200032,
      "step": 34120
    },
    {
      "epoch": 3.754125412541254,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.029302589639789735,
      "loss": 0.2327,
      "num_input_tokens_seen": 7200992,
      "step": 34125
    },
    {
      "epoch": 3.754675467546755,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.029302155582916028,
      "loss": 0.2303,
      "num_input_tokens_seen": 7202016,
      "step": 34130
    },
    {
      "epoch": 3.755225522552255,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.029301721394226207,
      "loss": 0.2266,
      "num_input_tokens_seen": 7203072,
      "step": 34135
    },
    {
      "epoch": 3.7557755775577557,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.02930128707372429,
      "loss": 0.2328,
      "num_input_tokens_seen": 7204160,
      "step": 34140
    },
    {
      "epoch": 3.7563256325632564,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.029300852621414266,
      "loss": 0.2359,
      "num_input_tokens_seen": 7205280,
      "step": 34145
    },
    {
      "epoch": 3.756875687568757,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.029300418037300147,
      "loss": 0.2281,
      "num_input_tokens_seen": 7206272,
      "step": 34150
    },
    {
      "epoch": 3.7574257425742577,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.02929998332138594,
      "loss": 0.2338,
      "num_input_tokens_seen": 7207360,
      "step": 34155
    },
    {
      "epoch": 3.757975797579758,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029299548473675645,
      "loss": 0.2286,
      "num_input_tokens_seen": 7208544,
      "step": 34160
    },
    {
      "epoch": 3.7585258525852585,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.02929911349417328,
      "loss": 0.2327,
      "num_input_tokens_seen": 7209696,
      "step": 34165
    },
    {
      "epoch": 3.759075907590759,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02929867838288284,
      "loss": 0.2317,
      "num_input_tokens_seen": 7210688,
      "step": 34170
    },
    {
      "epoch": 3.7596259625962594,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029298243139808347,
      "loss": 0.2328,
      "num_input_tokens_seen": 7211712,
      "step": 34175
    },
    {
      "epoch": 3.76017601760176,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.029297807764953804,
      "loss": 0.2286,
      "num_input_tokens_seen": 7212768,
      "step": 34180
    },
    {
      "epoch": 3.7607260726072607,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.02929737225832323,
      "loss": 0.2267,
      "num_input_tokens_seen": 7213760,
      "step": 34185
    },
    {
      "epoch": 3.7612761276127613,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.029296936619920638,
      "loss": 0.2317,
      "num_input_tokens_seen": 7214848,
      "step": 34190
    },
    {
      "epoch": 3.761826182618262,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.029296500849750037,
      "loss": 0.235,
      "num_input_tokens_seen": 7215872,
      "step": 34195
    },
    {
      "epoch": 3.762376237623762,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02929606494781545,
      "loss": 0.2303,
      "num_input_tokens_seen": 7216864,
      "step": 34200
    },
    {
      "epoch": 3.762926292629263,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.029295628914120894,
      "loss": 0.2354,
      "num_input_tokens_seen": 7217920,
      "step": 34205
    },
    {
      "epoch": 3.7634763476347635,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029295192748670387,
      "loss": 0.2297,
      "num_input_tokens_seen": 7218944,
      "step": 34210
    },
    {
      "epoch": 3.764026402640264,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029294756451467945,
      "loss": 0.2339,
      "num_input_tokens_seen": 7219936,
      "step": 34215
    },
    {
      "epoch": 3.7645764576457648,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.029294320022517594,
      "loss": 0.227,
      "num_input_tokens_seen": 7220960,
      "step": 34220
    },
    {
      "epoch": 3.765126512651265,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02929388346182335,
      "loss": 0.2301,
      "num_input_tokens_seen": 7221920,
      "step": 34225
    },
    {
      "epoch": 3.7656765676567656,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.029293446769389248,
      "loss": 0.2312,
      "num_input_tokens_seen": 7222944,
      "step": 34230
    },
    {
      "epoch": 3.7662266226622663,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.029293009945219297,
      "loss": 0.2317,
      "num_input_tokens_seen": 7224032,
      "step": 34235
    },
    {
      "epoch": 3.766776677667767,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02929257298931754,
      "loss": 0.2276,
      "num_input_tokens_seen": 7225088,
      "step": 34240
    },
    {
      "epoch": 3.7673267326732676,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.029292135901687992,
      "loss": 0.2379,
      "num_input_tokens_seen": 7226112,
      "step": 34245
    },
    {
      "epoch": 3.7678767876787678,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029291698682334683,
      "loss": 0.2327,
      "num_input_tokens_seen": 7227072,
      "step": 34250
    },
    {
      "epoch": 3.7684268426842684,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029291261331261656,
      "loss": 0.2326,
      "num_input_tokens_seen": 7228160,
      "step": 34255
    },
    {
      "epoch": 3.768976897689769,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.029290823848472917,
      "loss": 0.2341,
      "num_input_tokens_seen": 7229216,
      "step": 34260
    },
    {
      "epoch": 3.7695269526952693,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029290386233972523,
      "loss": 0.23,
      "num_input_tokens_seen": 7230240,
      "step": 34265
    },
    {
      "epoch": 3.77007700770077,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.02928994848776449,
      "loss": 0.2315,
      "num_input_tokens_seen": 7231360,
      "step": 34270
    },
    {
      "epoch": 3.7706270627062706,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.02928951060985286,
      "loss": 0.2284,
      "num_input_tokens_seen": 7232448,
      "step": 34275
    },
    {
      "epoch": 3.771177117711771,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.029289072600241668,
      "loss": 0.2326,
      "num_input_tokens_seen": 7233472,
      "step": 34280
    },
    {
      "epoch": 3.771727172717272,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.02928863445893495,
      "loss": 0.2331,
      "num_input_tokens_seen": 7234560,
      "step": 34285
    },
    {
      "epoch": 3.772277227722772,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029288196185936748,
      "loss": 0.2315,
      "num_input_tokens_seen": 7235616,
      "step": 34290
    },
    {
      "epoch": 3.7728272827282727,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.029287757781251093,
      "loss": 0.2315,
      "num_input_tokens_seen": 7236736,
      "step": 34295
    },
    {
      "epoch": 3.7733773377337734,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02928731924488203,
      "loss": 0.2315,
      "num_input_tokens_seen": 7237760,
      "step": 34300
    },
    {
      "epoch": 3.773927392739274,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029286880576833603,
      "loss": 0.232,
      "num_input_tokens_seen": 7238784,
      "step": 34305
    },
    {
      "epoch": 3.7744774477447747,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.029286441777109855,
      "loss": 0.232,
      "num_input_tokens_seen": 7239872,
      "step": 34310
    },
    {
      "epoch": 3.775027502750275,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.029286002845714827,
      "loss": 0.2331,
      "num_input_tokens_seen": 7240928,
      "step": 34315
    },
    {
      "epoch": 3.7755775577557755,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02928556378265257,
      "loss": 0.2274,
      "num_input_tokens_seen": 7241952,
      "step": 34320
    },
    {
      "epoch": 3.776127612761276,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.02928512458792712,
      "loss": 0.2299,
      "num_input_tokens_seen": 7243040,
      "step": 34325
    },
    {
      "epoch": 3.776677667766777,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029284685261542532,
      "loss": 0.2346,
      "num_input_tokens_seen": 7244064,
      "step": 34330
    },
    {
      "epoch": 3.7772277227722775,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.029284245803502855,
      "loss": 0.2273,
      "num_input_tokens_seen": 7245088,
      "step": 34335
    },
    {
      "epoch": 3.7777777777777777,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02928380621381214,
      "loss": 0.2279,
      "num_input_tokens_seen": 7246112,
      "step": 34340
    },
    {
      "epoch": 3.7783278327832783,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.029283366492474438,
      "loss": 0.226,
      "num_input_tokens_seen": 7247104,
      "step": 34345
    },
    {
      "epoch": 3.778877887788779,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.029282926639493802,
      "loss": 0.2343,
      "num_input_tokens_seen": 7248224,
      "step": 34350
    },
    {
      "epoch": 3.779427942794279,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02928248665487428,
      "loss": 0.2287,
      "num_input_tokens_seen": 7249248,
      "step": 34355
    },
    {
      "epoch": 3.77997799779978,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.029282046538619936,
      "loss": 0.2334,
      "num_input_tokens_seen": 7250400,
      "step": 34360
    },
    {
      "epoch": 3.7805280528052805,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02928160629073482,
      "loss": 0.2302,
      "num_input_tokens_seen": 7251456,
      "step": 34365
    },
    {
      "epoch": 3.781078107810781,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.02928116591122299,
      "loss": 0.2303,
      "num_input_tokens_seen": 7252640,
      "step": 34370
    },
    {
      "epoch": 3.781628162816282,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.029280725400088508,
      "loss": 0.2273,
      "num_input_tokens_seen": 7253728,
      "step": 34375
    },
    {
      "epoch": 3.782178217821782,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029280284757335438,
      "loss": 0.2354,
      "num_input_tokens_seen": 7254784,
      "step": 34380
    },
    {
      "epoch": 3.7827282728272826,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.029279843982967832,
      "loss": 0.2318,
      "num_input_tokens_seen": 7255776,
      "step": 34385
    },
    {
      "epoch": 3.7832783278327833,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.02927940307698975,
      "loss": 0.2307,
      "num_input_tokens_seen": 7256768,
      "step": 34390
    },
    {
      "epoch": 3.783828382838284,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.02927896203940527,
      "loss": 0.2287,
      "num_input_tokens_seen": 7257728,
      "step": 34395
    },
    {
      "epoch": 3.7843784378437846,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029278520870218448,
      "loss": 0.2282,
      "num_input_tokens_seen": 7258752,
      "step": 34400
    },
    {
      "epoch": 3.784928492849285,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.029278079569433346,
      "loss": 0.2292,
      "num_input_tokens_seen": 7259776,
      "step": 34405
    },
    {
      "epoch": 3.7854785478547854,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029277638137054036,
      "loss": 0.2288,
      "num_input_tokens_seen": 7260736,
      "step": 34410
    },
    {
      "epoch": 3.786028602860286,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029277196573084592,
      "loss": 0.2344,
      "num_input_tokens_seen": 7261792,
      "step": 34415
    },
    {
      "epoch": 3.7865786578657867,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.029276754877529076,
      "loss": 0.2328,
      "num_input_tokens_seen": 7262848,
      "step": 34420
    },
    {
      "epoch": 3.7871287128712874,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.029276313050391557,
      "loss": 0.226,
      "num_input_tokens_seen": 7263904,
      "step": 34425
    },
    {
      "epoch": 3.7876787678767876,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029275871091676117,
      "loss": 0.2324,
      "num_input_tokens_seen": 7264992,
      "step": 34430
    },
    {
      "epoch": 3.7882288228822882,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.02927542900138682,
      "loss": 0.2365,
      "num_input_tokens_seen": 7266016,
      "step": 34435
    },
    {
      "epoch": 3.788778877887789,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.02927498677952774,
      "loss": 0.2322,
      "num_input_tokens_seen": 7267040,
      "step": 34440
    },
    {
      "epoch": 3.789328932893289,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.029274544426102965,
      "loss": 0.2354,
      "num_input_tokens_seen": 7268064,
      "step": 34445
    },
    {
      "epoch": 3.7898789878987897,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02927410194111656,
      "loss": 0.2274,
      "num_input_tokens_seen": 7269120,
      "step": 34450
    },
    {
      "epoch": 3.7904290429042904,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.029273659324572605,
      "loss": 0.2296,
      "num_input_tokens_seen": 7270144,
      "step": 34455
    },
    {
      "epoch": 3.790979097909791,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.029273216576475178,
      "loss": 0.2291,
      "num_input_tokens_seen": 7271232,
      "step": 34460
    },
    {
      "epoch": 3.7915291529152917,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02927277369682837,
      "loss": 0.2322,
      "num_input_tokens_seen": 7272320,
      "step": 34465
    },
    {
      "epoch": 3.792079207920792,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.029272330685636252,
      "loss": 0.2333,
      "num_input_tokens_seen": 7273376,
      "step": 34470
    },
    {
      "epoch": 3.7926292629262925,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02927188754290291,
      "loss": 0.2353,
      "num_input_tokens_seen": 7274432,
      "step": 34475
    },
    {
      "epoch": 3.793179317931793,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.029271444268632436,
      "loss": 0.2311,
      "num_input_tokens_seen": 7275520,
      "step": 34480
    },
    {
      "epoch": 3.793729372937294,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0292710008628289,
      "loss": 0.2295,
      "num_input_tokens_seen": 7276608,
      "step": 34485
    },
    {
      "epoch": 3.7942794279427945,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.0292705573254964,
      "loss": 0.2285,
      "num_input_tokens_seen": 7277696,
      "step": 34490
    },
    {
      "epoch": 3.7948294829482947,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02927011365663902,
      "loss": 0.2321,
      "num_input_tokens_seen": 7278752,
      "step": 34495
    },
    {
      "epoch": 3.7953795379537953,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.02926966985626085,
      "loss": 0.2363,
      "num_input_tokens_seen": 7279808,
      "step": 34500
    },
    {
      "epoch": 3.795929592959296,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.029269225924365982,
      "loss": 0.23,
      "num_input_tokens_seen": 7280832,
      "step": 34505
    },
    {
      "epoch": 3.7964796479647966,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.029268781860958503,
      "loss": 0.2279,
      "num_input_tokens_seen": 7281888,
      "step": 34510
    },
    {
      "epoch": 3.7970297029702973,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.029268337666042508,
      "loss": 0.2332,
      "num_input_tokens_seen": 7282976,
      "step": 34515
    },
    {
      "epoch": 3.7975797579757975,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029267893339622094,
      "loss": 0.2353,
      "num_input_tokens_seen": 7284064,
      "step": 34520
    },
    {
      "epoch": 3.798129812981298,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.02926744888170135,
      "loss": 0.2353,
      "num_input_tokens_seen": 7285184,
      "step": 34525
    },
    {
      "epoch": 3.798679867986799,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02926700429228438,
      "loss": 0.2289,
      "num_input_tokens_seen": 7286304,
      "step": 34530
    },
    {
      "epoch": 3.799229922992299,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.029266559571375275,
      "loss": 0.2337,
      "num_input_tokens_seen": 7287392,
      "step": 34535
    },
    {
      "epoch": 3.7997799779977997,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029266114718978137,
      "loss": 0.2331,
      "num_input_tokens_seen": 7288352,
      "step": 34540
    },
    {
      "epoch": 3.8003300330033003,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.029265669735097064,
      "loss": 0.2351,
      "num_input_tokens_seen": 7289472,
      "step": 34545
    },
    {
      "epoch": 3.800880088008801,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.029265224619736158,
      "loss": 0.233,
      "num_input_tokens_seen": 7290464,
      "step": 34550
    },
    {
      "epoch": 3.8014301430143016,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029264779372899525,
      "loss": 0.2309,
      "num_input_tokens_seen": 7291520,
      "step": 34555
    },
    {
      "epoch": 3.801980198019802,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029264333994591264,
      "loss": 0.2284,
      "num_input_tokens_seen": 7292576,
      "step": 34560
    },
    {
      "epoch": 3.8025302530253025,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02926388848481548,
      "loss": 0.231,
      "num_input_tokens_seen": 7293568,
      "step": 34565
    },
    {
      "epoch": 3.803080308030803,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.029263442843576282,
      "loss": 0.2325,
      "num_input_tokens_seen": 7294656,
      "step": 34570
    },
    {
      "epoch": 3.8036303630363038,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029262997070877775,
      "loss": 0.2325,
      "num_input_tokens_seen": 7295744,
      "step": 34575
    },
    {
      "epoch": 3.8041804180418044,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029262551166724066,
      "loss": 0.2309,
      "num_input_tokens_seen": 7296864,
      "step": 34580
    },
    {
      "epoch": 3.8047304730473046,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029262105131119272,
      "loss": 0.232,
      "num_input_tokens_seen": 7297856,
      "step": 34585
    },
    {
      "epoch": 3.8052805280528053,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.029261658964067492,
      "loss": 0.232,
      "num_input_tokens_seen": 7298880,
      "step": 34590
    },
    {
      "epoch": 3.805830583058306,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.029261212665572845,
      "loss": 0.2325,
      "num_input_tokens_seen": 7299968,
      "step": 34595
    },
    {
      "epoch": 3.806380638063806,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.029260766235639453,
      "loss": 0.2335,
      "num_input_tokens_seen": 7300992,
      "step": 34600
    },
    {
      "epoch": 3.806930693069307,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.02926031967427141,
      "loss": 0.2309,
      "num_input_tokens_seen": 7302048,
      "step": 34605
    },
    {
      "epoch": 3.8074807480748074,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.02925987298147285,
      "loss": 0.2319,
      "num_input_tokens_seen": 7303168,
      "step": 34610
    },
    {
      "epoch": 3.808030803080308,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.029259426157247883,
      "loss": 0.2319,
      "num_input_tokens_seen": 7304224,
      "step": 34615
    },
    {
      "epoch": 3.8085808580858087,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.029258979201600626,
      "loss": 0.2303,
      "num_input_tokens_seen": 7305216,
      "step": 34620
    },
    {
      "epoch": 3.809130913091309,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.0292585321145352,
      "loss": 0.2319,
      "num_input_tokens_seen": 7306272,
      "step": 34625
    },
    {
      "epoch": 3.8096809680968096,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.029258084896055726,
      "loss": 0.2304,
      "num_input_tokens_seen": 7307328,
      "step": 34630
    },
    {
      "epoch": 3.81023102310231,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.029257637546166324,
      "loss": 0.2347,
      "num_input_tokens_seen": 7308384,
      "step": 34635
    },
    {
      "epoch": 3.810781078107811,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.029257190064871118,
      "loss": 0.2311,
      "num_input_tokens_seen": 7309472,
      "step": 34640
    },
    {
      "epoch": 3.8113311331133115,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.029256742452174234,
      "loss": 0.2307,
      "num_input_tokens_seen": 7310464,
      "step": 34645
    },
    {
      "epoch": 3.8118811881188117,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.0292562947080798,
      "loss": 0.2291,
      "num_input_tokens_seen": 7311488,
      "step": 34650
    },
    {
      "epoch": 3.8124312431243124,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.02925584683259193,
      "loss": 0.2357,
      "num_input_tokens_seen": 7312608,
      "step": 34655
    },
    {
      "epoch": 3.812981298129813,
      "grad_norm": 0.01239013671875,
      "learning_rate": 0.029255398825714766,
      "loss": 0.2304,
      "num_input_tokens_seen": 7313600,
      "step": 34660
    },
    {
      "epoch": 3.8135313531353137,
      "grad_norm": 0.0029296875,
      "learning_rate": 0.02925495068745243,
      "loss": 0.234,
      "num_input_tokens_seen": 7314592,
      "step": 34665
    },
    {
      "epoch": 3.8140814081408143,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.02925450241780905,
      "loss": 0.2272,
      "num_input_tokens_seen": 7315680,
      "step": 34670
    },
    {
      "epoch": 3.8146314631463145,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.029254054016788766,
      "loss": 0.229,
      "num_input_tokens_seen": 7316768,
      "step": 34675
    },
    {
      "epoch": 3.815181518151815,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.029253605484395706,
      "loss": 0.2318,
      "num_input_tokens_seen": 7317824,
      "step": 34680
    },
    {
      "epoch": 3.815731573157316,
      "grad_norm": 0.0078125,
      "learning_rate": 0.029253156820634,
      "loss": 0.2354,
      "num_input_tokens_seen": 7318944,
      "step": 34685
    },
    {
      "epoch": 3.816281628162816,
      "grad_norm": 0.007476806640625,
      "learning_rate": 0.029252708025507793,
      "loss": 0.2316,
      "num_input_tokens_seen": 7320000,
      "step": 34690
    },
    {
      "epoch": 3.8168316831683167,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.02925225909902121,
      "loss": 0.2342,
      "num_input_tokens_seen": 7321056,
      "step": 34695
    },
    {
      "epoch": 3.8173817381738173,
      "grad_norm": 0.01495361328125,
      "learning_rate": 0.029251810041178397,
      "loss": 0.233,
      "num_input_tokens_seen": 7322112,
      "step": 34700
    },
    {
      "epoch": 3.817931793179318,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.029251360851983484,
      "loss": 0.232,
      "num_input_tokens_seen": 7323168,
      "step": 34705
    },
    {
      "epoch": 3.8184818481848186,
      "grad_norm": 0.01251220703125,
      "learning_rate": 0.02925091153144062,
      "loss": 0.2309,
      "num_input_tokens_seen": 7324256,
      "step": 34710
    },
    {
      "epoch": 3.819031903190319,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029250462079553945,
      "loss": 0.2299,
      "num_input_tokens_seen": 7325312,
      "step": 34715
    },
    {
      "epoch": 3.8195819581958195,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029250012496327596,
      "loss": 0.2319,
      "num_input_tokens_seen": 7326368,
      "step": 34720
    },
    {
      "epoch": 3.82013201320132,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.02924956278176572,
      "loss": 0.2329,
      "num_input_tokens_seen": 7327328,
      "step": 34725
    },
    {
      "epoch": 3.8206820682068208,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.029249112935872464,
      "loss": 0.2314,
      "num_input_tokens_seen": 7328352,
      "step": 34730
    },
    {
      "epoch": 3.8212321232123214,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029248662958651972,
      "loss": 0.2314,
      "num_input_tokens_seen": 7329440,
      "step": 34735
    },
    {
      "epoch": 3.8217821782178216,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029248212850108386,
      "loss": 0.2309,
      "num_input_tokens_seen": 7330432,
      "step": 34740
    },
    {
      "epoch": 3.8223322332233223,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02924776261024586,
      "loss": 0.2324,
      "num_input_tokens_seen": 7331552,
      "step": 34745
    },
    {
      "epoch": 3.822882288228823,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029247312239068547,
      "loss": 0.2304,
      "num_input_tokens_seen": 7332608,
      "step": 34750
    },
    {
      "epoch": 3.8234323432343236,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.029246861736580587,
      "loss": 0.2308,
      "num_input_tokens_seen": 7333632,
      "step": 34755
    },
    {
      "epoch": 3.823982398239824,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.029246411102786144,
      "loss": 0.2324,
      "num_input_tokens_seen": 7334752,
      "step": 34760
    },
    {
      "epoch": 3.8245324532453244,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.029245960337689367,
      "loss": 0.2324,
      "num_input_tokens_seen": 7335808,
      "step": 34765
    },
    {
      "epoch": 3.825082508250825,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029245509441294407,
      "loss": 0.2319,
      "num_input_tokens_seen": 7336928,
      "step": 34770
    },
    {
      "epoch": 3.8256325632563257,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.02924505841360542,
      "loss": 0.2356,
      "num_input_tokens_seen": 7337888,
      "step": 34775
    },
    {
      "epoch": 3.826182618261826,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.029244607254626565,
      "loss": 0.234,
      "num_input_tokens_seen": 7338976,
      "step": 34780
    },
    {
      "epoch": 3.8267326732673266,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.029244155964362004,
      "loss": 0.2314,
      "num_input_tokens_seen": 7340032,
      "step": 34785
    },
    {
      "epoch": 3.8272827282728272,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.029243704542815892,
      "loss": 0.2319,
      "num_input_tokens_seen": 7341088,
      "step": 34790
    },
    {
      "epoch": 3.827832783278328,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029243252989992388,
      "loss": 0.2314,
      "num_input_tokens_seen": 7342112,
      "step": 34795
    },
    {
      "epoch": 3.8283828382838285,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029242801305895655,
      "loss": 0.2314,
      "num_input_tokens_seen": 7343200,
      "step": 34800
    },
    {
      "epoch": 3.8289328932893287,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.029242349490529857,
      "loss": 0.2304,
      "num_input_tokens_seen": 7344256,
      "step": 34805
    },
    {
      "epoch": 3.8294829482948294,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.029241897543899158,
      "loss": 0.2329,
      "num_input_tokens_seen": 7345280,
      "step": 34810
    },
    {
      "epoch": 3.83003300330033,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.029241445466007727,
      "loss": 0.2293,
      "num_input_tokens_seen": 7346368,
      "step": 34815
    },
    {
      "epoch": 3.8305830583058307,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.029240993256859724,
      "loss": 0.2314,
      "num_input_tokens_seen": 7347392,
      "step": 34820
    },
    {
      "epoch": 3.8311331133113313,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02924054091645932,
      "loss": 0.2319,
      "num_input_tokens_seen": 7348512,
      "step": 34825
    },
    {
      "epoch": 3.8316831683168315,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.029240088444810686,
      "loss": 0.2309,
      "num_input_tokens_seen": 7349600,
      "step": 34830
    },
    {
      "epoch": 3.832233223322332,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.029239635841917987,
      "loss": 0.2314,
      "num_input_tokens_seen": 7350720,
      "step": 34835
    },
    {
      "epoch": 3.832783278327833,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029239183107785397,
      "loss": 0.2325,
      "num_input_tokens_seen": 7351776,
      "step": 34840
    },
    {
      "epoch": 3.8333333333333335,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.029238730242417087,
      "loss": 0.2325,
      "num_input_tokens_seen": 7352864,
      "step": 34845
    },
    {
      "epoch": 3.833883388338834,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.029238277245817238,
      "loss": 0.2314,
      "num_input_tokens_seen": 7353888,
      "step": 34850
    },
    {
      "epoch": 3.8344334433443343,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.029237824117990013,
      "loss": 0.2335,
      "num_input_tokens_seen": 7354912,
      "step": 34855
    },
    {
      "epoch": 3.834983498349835,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.029237370858939602,
      "loss": 0.2309,
      "num_input_tokens_seen": 7355936,
      "step": 34860
    },
    {
      "epoch": 3.8355335533553356,
      "grad_norm": 0.00080108642578125,
      "learning_rate": 0.029236917468670173,
      "loss": 0.2319,
      "num_input_tokens_seen": 7357024,
      "step": 34865
    },
    {
      "epoch": 3.836083608360836,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029236463947185907,
      "loss": 0.2309,
      "num_input_tokens_seen": 7358112,
      "step": 34870
    },
    {
      "epoch": 3.8366336633663365,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.029236010294490983,
      "loss": 0.2303,
      "num_input_tokens_seen": 7359264,
      "step": 34875
    },
    {
      "epoch": 3.837183718371837,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.029235556510589584,
      "loss": 0.2309,
      "num_input_tokens_seen": 7360288,
      "step": 34880
    },
    {
      "epoch": 3.837733773377338,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.02923510259548589,
      "loss": 0.2319,
      "num_input_tokens_seen": 7361312,
      "step": 34885
    },
    {
      "epoch": 3.8382838283828384,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02923464854918409,
      "loss": 0.2293,
      "num_input_tokens_seen": 7362400,
      "step": 34890
    },
    {
      "epoch": 3.8388338833883386,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.029234194371688362,
      "loss": 0.233,
      "num_input_tokens_seen": 7363456,
      "step": 34895
    },
    {
      "epoch": 3.8393839383938393,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.029233740063002896,
      "loss": 0.2293,
      "num_input_tokens_seen": 7364480,
      "step": 34900
    },
    {
      "epoch": 3.83993399339934,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.029233285623131877,
      "loss": 0.2293,
      "num_input_tokens_seen": 7365536,
      "step": 34905
    },
    {
      "epoch": 3.8404840484048406,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.029232831052079498,
      "loss": 0.2299,
      "num_input_tokens_seen": 7366528,
      "step": 34910
    },
    {
      "epoch": 3.8410341034103412,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.029232376349849938,
      "loss": 0.233,
      "num_input_tokens_seen": 7367584,
      "step": 34915
    },
    {
      "epoch": 3.8415841584158414,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0292319215164474,
      "loss": 0.2329,
      "num_input_tokens_seen": 7368672,
      "step": 34920
    },
    {
      "epoch": 3.842134213421342,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.029231466551876067,
      "loss": 0.2293,
      "num_input_tokens_seen": 7369760,
      "step": 34925
    },
    {
      "epoch": 3.8426842684268427,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.029231011456140137,
      "loss": 0.2319,
      "num_input_tokens_seen": 7370816,
      "step": 34930
    },
    {
      "epoch": 3.8432343234323434,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.029230556229243802,
      "loss": 0.2278,
      "num_input_tokens_seen": 7371936,
      "step": 34935
    },
    {
      "epoch": 3.843784378437844,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02923010087119126,
      "loss": 0.231,
      "num_input_tokens_seen": 7373056,
      "step": 34940
    },
    {
      "epoch": 3.8443344334433442,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.029229645381986705,
      "loss": 0.2279,
      "num_input_tokens_seen": 7374112,
      "step": 34945
    },
    {
      "epoch": 3.844884488448845,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.02922918976163434,
      "loss": 0.2327,
      "num_input_tokens_seen": 7375104,
      "step": 34950
    },
    {
      "epoch": 3.8454345434543455,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.029228734010138358,
      "loss": 0.2342,
      "num_input_tokens_seen": 7376192,
      "step": 34955
    },
    {
      "epoch": 3.8459845984598457,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.029228278127502963,
      "loss": 0.227,
      "num_input_tokens_seen": 7377216,
      "step": 34960
    },
    {
      "epoch": 3.8465346534653464,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.029227822113732354,
      "loss": 0.2321,
      "num_input_tokens_seen": 7378272,
      "step": 34965
    },
    {
      "epoch": 3.847084708470847,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.029227365968830736,
      "loss": 0.2326,
      "num_input_tokens_seen": 7379232,
      "step": 34970
    },
    {
      "epoch": 3.8476347634763477,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.029226909692802318,
      "loss": 0.2296,
      "num_input_tokens_seen": 7380288,
      "step": 34975
    },
    {
      "epoch": 3.8481848184818483,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029226453285651297,
      "loss": 0.2317,
      "num_input_tokens_seen": 7381280,
      "step": 34980
    },
    {
      "epoch": 3.8487348734873486,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029225996747381883,
      "loss": 0.2347,
      "num_input_tokens_seen": 7382304,
      "step": 34985
    },
    {
      "epoch": 3.849284928492849,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.029225540077998283,
      "loss": 0.2327,
      "num_input_tokens_seen": 7383296,
      "step": 34990
    },
    {
      "epoch": 3.84983498349835,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.029225083277504702,
      "loss": 0.2258,
      "num_input_tokens_seen": 7384320,
      "step": 34995
    },
    {
      "epoch": 3.8503850385038505,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.02922462634590536,
      "loss": 0.2305,
      "num_input_tokens_seen": 7385376,
      "step": 35000
    },
    {
      "epoch": 3.850935093509351,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.029224169283204458,
      "loss": 0.2321,
      "num_input_tokens_seen": 7386464,
      "step": 35005
    },
    {
      "epoch": 3.8514851485148514,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.029223712089406217,
      "loss": 0.2341,
      "num_input_tokens_seen": 7387456,
      "step": 35010
    },
    {
      "epoch": 3.852035203520352,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.029223254764514844,
      "loss": 0.2315,
      "num_input_tokens_seen": 7388480,
      "step": 35015
    },
    {
      "epoch": 3.8525852585258527,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.02922279730853456,
      "loss": 0.2293,
      "num_input_tokens_seen": 7389536,
      "step": 35020
    },
    {
      "epoch": 3.8531353135313533,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.029222339721469575,
      "loss": 0.2298,
      "num_input_tokens_seen": 7390656,
      "step": 35025
    },
    {
      "epoch": 3.853685368536854,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.02922188200332411,
      "loss": 0.2314,
      "num_input_tokens_seen": 7391680,
      "step": 35030
    },
    {
      "epoch": 3.854235423542354,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029221424154102386,
      "loss": 0.2314,
      "num_input_tokens_seen": 7392704,
      "step": 35035
    },
    {
      "epoch": 3.854785478547855,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029220966173808618,
      "loss": 0.2303,
      "num_input_tokens_seen": 7393728,
      "step": 35040
    },
    {
      "epoch": 3.8553355335533555,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029220508062447026,
      "loss": 0.2278,
      "num_input_tokens_seen": 7394784,
      "step": 35045
    },
    {
      "epoch": 3.8558855885588557,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029220049820021837,
      "loss": 0.2309,
      "num_input_tokens_seen": 7395840,
      "step": 35050
    },
    {
      "epoch": 3.8564356435643563,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.029219591446537272,
      "loss": 0.2298,
      "num_input_tokens_seen": 7396896,
      "step": 35055
    },
    {
      "epoch": 3.856985698569857,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029219132941997555,
      "loss": 0.2314,
      "num_input_tokens_seen": 7397920,
      "step": 35060
    },
    {
      "epoch": 3.8575357535753576,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029218674306406914,
      "loss": 0.2309,
      "num_input_tokens_seen": 7398912,
      "step": 35065
    },
    {
      "epoch": 3.8580858085808583,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.029218215539769575,
      "loss": 0.2299,
      "num_input_tokens_seen": 7400064,
      "step": 35070
    },
    {
      "epoch": 3.8586358635863585,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02921775664208976,
      "loss": 0.2294,
      "num_input_tokens_seen": 7401088,
      "step": 35075
    },
    {
      "epoch": 3.859185918591859,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.02921729761337171,
      "loss": 0.2279,
      "num_input_tokens_seen": 7402144,
      "step": 35080
    },
    {
      "epoch": 3.8597359735973598,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.02921683845361965,
      "loss": 0.2336,
      "num_input_tokens_seen": 7403232,
      "step": 35085
    },
    {
      "epoch": 3.8602860286028604,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.029216379162837813,
      "loss": 0.2315,
      "num_input_tokens_seen": 7404192,
      "step": 35090
    },
    {
      "epoch": 3.860836083608361,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.029215919741030428,
      "loss": 0.2279,
      "num_input_tokens_seen": 7405248,
      "step": 35095
    },
    {
      "epoch": 3.8613861386138613,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.029215460188201738,
      "loss": 0.2342,
      "num_input_tokens_seen": 7406240,
      "step": 35100
    },
    {
      "epoch": 3.861936193619362,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.029215000504355965,
      "loss": 0.2222,
      "num_input_tokens_seen": 7407232,
      "step": 35105
    },
    {
      "epoch": 3.8624862486248626,
      "grad_norm": 0.00750732421875,
      "learning_rate": 0.029214540689497356,
      "loss": 0.2285,
      "num_input_tokens_seen": 7408320,
      "step": 35110
    },
    {
      "epoch": 3.8630363036303628,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.029214080743630147,
      "loss": 0.2294,
      "num_input_tokens_seen": 7409344,
      "step": 35115
    },
    {
      "epoch": 3.863586358635864,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.029213620666758578,
      "loss": 0.2321,
      "num_input_tokens_seen": 7410400,
      "step": 35120
    },
    {
      "epoch": 3.864136413641364,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.029213160458886882,
      "loss": 0.2256,
      "num_input_tokens_seen": 7411456,
      "step": 35125
    },
    {
      "epoch": 3.8646864686468647,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.02921270012001931,
      "loss": 0.2457,
      "num_input_tokens_seen": 7412480,
      "step": 35130
    },
    {
      "epoch": 3.8652365236523654,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029212239650160104,
      "loss": 0.2246,
      "num_input_tokens_seen": 7413568,
      "step": 35135
    },
    {
      "epoch": 3.8657865786578656,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0292117790493135,
      "loss": 0.2349,
      "num_input_tokens_seen": 7414560,
      "step": 35140
    },
    {
      "epoch": 3.866336633663366,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02921131831748375,
      "loss": 0.2292,
      "num_input_tokens_seen": 7415584,
      "step": 35145
    },
    {
      "epoch": 3.866886688668867,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.029210857454675103,
      "loss": 0.2365,
      "num_input_tokens_seen": 7416640,
      "step": 35150
    },
    {
      "epoch": 3.8674367436743675,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.029210396460891797,
      "loss": 0.2338,
      "num_input_tokens_seen": 7417728,
      "step": 35155
    },
    {
      "epoch": 3.867986798679868,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029209935336138082,
      "loss": 0.2296,
      "num_input_tokens_seen": 7418816,
      "step": 35160
    },
    {
      "epoch": 3.8685368536853684,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.029209474080418216,
      "loss": 0.2322,
      "num_input_tokens_seen": 7419872,
      "step": 35165
    },
    {
      "epoch": 3.869086908690869,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029209012693736446,
      "loss": 0.2337,
      "num_input_tokens_seen": 7420928,
      "step": 35170
    },
    {
      "epoch": 3.8696369636963697,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.02920855117609702,
      "loss": 0.2321,
      "num_input_tokens_seen": 7422016,
      "step": 35175
    },
    {
      "epoch": 3.8701870187018703,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0292080895275042,
      "loss": 0.2304,
      "num_input_tokens_seen": 7423008,
      "step": 35180
    },
    {
      "epoch": 3.870737073707371,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.029207627747962237,
      "loss": 0.2336,
      "num_input_tokens_seen": 7424064,
      "step": 35185
    },
    {
      "epoch": 3.871287128712871,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.029207165837475384,
      "loss": 0.2321,
      "num_input_tokens_seen": 7425152,
      "step": 35190
    },
    {
      "epoch": 3.871837183718372,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0292067037960479,
      "loss": 0.2284,
      "num_input_tokens_seen": 7426208,
      "step": 35195
    },
    {
      "epoch": 3.8723872387238725,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.029206241623684046,
      "loss": 0.2352,
      "num_input_tokens_seen": 7427328,
      "step": 35200
    },
    {
      "epoch": 3.8729372937293727,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.029205779320388082,
      "loss": 0.2321,
      "num_input_tokens_seen": 7428352,
      "step": 35205
    },
    {
      "epoch": 3.8734873487348733,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.029205316886164262,
      "loss": 0.2295,
      "num_input_tokens_seen": 7429440,
      "step": 35210
    },
    {
      "epoch": 3.874037403740374,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029204854321016858,
      "loss": 0.2249,
      "num_input_tokens_seen": 7430496,
      "step": 35215
    },
    {
      "epoch": 3.8745874587458746,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.029204391624950123,
      "loss": 0.2358,
      "num_input_tokens_seen": 7431552,
      "step": 35220
    },
    {
      "epoch": 3.8751375137513753,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02920392879796833,
      "loss": 0.2286,
      "num_input_tokens_seen": 7432544,
      "step": 35225
    },
    {
      "epoch": 3.8756875687568755,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.029203465840075733,
      "loss": 0.226,
      "num_input_tokens_seen": 7433536,
      "step": 35230
    },
    {
      "epoch": 3.876237623762376,
      "grad_norm": 0.007415771484375,
      "learning_rate": 0.029203002751276616,
      "loss": 0.2287,
      "num_input_tokens_seen": 7434624,
      "step": 35235
    },
    {
      "epoch": 3.8767876787678768,
      "grad_norm": 0.01483154296875,
      "learning_rate": 0.02920253953157524,
      "loss": 0.2356,
      "num_input_tokens_seen": 7435744,
      "step": 35240
    },
    {
      "epoch": 3.8773377337733774,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.02920207618097586,
      "loss": 0.2314,
      "num_input_tokens_seen": 7436768,
      "step": 35245
    },
    {
      "epoch": 3.877887788778878,
      "grad_norm": 0.0146484375,
      "learning_rate": 0.02920161269948277,
      "loss": 0.2356,
      "num_input_tokens_seen": 7437824,
      "step": 35250
    },
    {
      "epoch": 3.8784378437843783,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.029201149087100225,
      "loss": 0.2292,
      "num_input_tokens_seen": 7438880,
      "step": 35255
    },
    {
      "epoch": 3.878987898789879,
      "grad_norm": 0.007537841796875,
      "learning_rate": 0.0292006853438325,
      "loss": 0.2282,
      "num_input_tokens_seen": 7440000,
      "step": 35260
    },
    {
      "epoch": 3.8795379537953796,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.02920022146968388,
      "loss": 0.2293,
      "num_input_tokens_seen": 7441024,
      "step": 35265
    },
    {
      "epoch": 3.8800880088008802,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.02919975746465863,
      "loss": 0.2319,
      "num_input_tokens_seen": 7442144,
      "step": 35270
    },
    {
      "epoch": 3.880638063806381,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.029199293328761027,
      "loss": 0.236,
      "num_input_tokens_seen": 7443104,
      "step": 35275
    },
    {
      "epoch": 3.881188118811881,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02919882906199535,
      "loss": 0.236,
      "num_input_tokens_seen": 7444128,
      "step": 35280
    },
    {
      "epoch": 3.8817381738173817,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.029198364664365883,
      "loss": 0.2296,
      "num_input_tokens_seen": 7445184,
      "step": 35285
    },
    {
      "epoch": 3.8822882288228824,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.0291979001358769,
      "loss": 0.226,
      "num_input_tokens_seen": 7446208,
      "step": 35290
    },
    {
      "epoch": 3.8828382838283826,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.029197435476532682,
      "loss": 0.223,
      "num_input_tokens_seen": 7447232,
      "step": 35295
    },
    {
      "epoch": 3.8833883388338832,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.029196970686337517,
      "loss": 0.2287,
      "num_input_tokens_seen": 7448288,
      "step": 35300
    },
    {
      "epoch": 3.883938393839384,
      "grad_norm": 0.007415771484375,
      "learning_rate": 0.029196505765295686,
      "loss": 0.232,
      "num_input_tokens_seen": 7449408,
      "step": 35305
    },
    {
      "epoch": 3.8844884488448845,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.02919604071341147,
      "loss": 0.2336,
      "num_input_tokens_seen": 7450464,
      "step": 35310
    },
    {
      "epoch": 3.885038503850385,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.02919557553068916,
      "loss": 0.2362,
      "num_input_tokens_seen": 7451488,
      "step": 35315
    },
    {
      "epoch": 3.8855885588558854,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029195110217133044,
      "loss": 0.2298,
      "num_input_tokens_seen": 7452544,
      "step": 35320
    },
    {
      "epoch": 3.886138613861386,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.029194644772747406,
      "loss": 0.2386,
      "num_input_tokens_seen": 7453632,
      "step": 35325
    },
    {
      "epoch": 3.8866886688668867,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.02919417919753654,
      "loss": 0.2411,
      "num_input_tokens_seen": 7454688,
      "step": 35330
    },
    {
      "epoch": 3.8872387238723873,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.029193713491504736,
      "loss": 0.2326,
      "num_input_tokens_seen": 7455808,
      "step": 35335
    },
    {
      "epoch": 3.887788778877888,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029193247654656283,
      "loss": 0.2309,
      "num_input_tokens_seen": 7456928,
      "step": 35340
    },
    {
      "epoch": 3.888338833883388,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02919278168699548,
      "loss": 0.2294,
      "num_input_tokens_seen": 7457952,
      "step": 35345
    },
    {
      "epoch": 3.888888888888889,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02919231558852662,
      "loss": 0.2346,
      "num_input_tokens_seen": 7459072,
      "step": 35350
    },
    {
      "epoch": 3.8894389438943895,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.02919184935925399,
      "loss": 0.2304,
      "num_input_tokens_seen": 7460064,
      "step": 35355
    },
    {
      "epoch": 3.88998899889989,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.029191382999181902,
      "loss": 0.2314,
      "num_input_tokens_seen": 7461088,
      "step": 35360
    },
    {
      "epoch": 3.890539053905391,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02919091650831464,
      "loss": 0.2304,
      "num_input_tokens_seen": 7462176,
      "step": 35365
    },
    {
      "epoch": 3.891089108910891,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029190449886656516,
      "loss": 0.2289,
      "num_input_tokens_seen": 7463232,
      "step": 35370
    },
    {
      "epoch": 3.8916391639163916,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.02918998313421182,
      "loss": 0.234,
      "num_input_tokens_seen": 7464256,
      "step": 35375
    },
    {
      "epoch": 3.8921892189218923,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.029189516250984864,
      "loss": 0.2309,
      "num_input_tokens_seen": 7465344,
      "step": 35380
    },
    {
      "epoch": 3.8927392739273925,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.02918904923697994,
      "loss": 0.2293,
      "num_input_tokens_seen": 7466432,
      "step": 35385
    },
    {
      "epoch": 3.893289328932893,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.029188582092201362,
      "loss": 0.2309,
      "num_input_tokens_seen": 7467488,
      "step": 35390
    },
    {
      "epoch": 3.893839383938394,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029188114816653427,
      "loss": 0.2309,
      "num_input_tokens_seen": 7468544,
      "step": 35395
    },
    {
      "epoch": 3.8943894389438944,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.029187647410340446,
      "loss": 0.2335,
      "num_input_tokens_seen": 7469632,
      "step": 35400
    },
    {
      "epoch": 3.894939493949395,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.02918717987326673,
      "loss": 0.2304,
      "num_input_tokens_seen": 7470720,
      "step": 35405
    },
    {
      "epoch": 3.8954895489548953,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02918671220543658,
      "loss": 0.2325,
      "num_input_tokens_seen": 7471712,
      "step": 35410
    },
    {
      "epoch": 3.896039603960396,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.029186244406854318,
      "loss": 0.2314,
      "num_input_tokens_seen": 7472768,
      "step": 35415
    },
    {
      "epoch": 3.8965896589658966,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.029185776477524242,
      "loss": 0.2314,
      "num_input_tokens_seen": 7473888,
      "step": 35420
    },
    {
      "epoch": 3.8971397139713972,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.02918530841745068,
      "loss": 0.2308,
      "num_input_tokens_seen": 7474880,
      "step": 35425
    },
    {
      "epoch": 3.897689768976898,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029184840226637927,
      "loss": 0.2303,
      "num_input_tokens_seen": 7475936,
      "step": 35430
    },
    {
      "epoch": 3.898239823982398,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.029184371905090312,
      "loss": 0.2324,
      "num_input_tokens_seen": 7477024,
      "step": 35435
    },
    {
      "epoch": 3.8987898789878987,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02918390345281215,
      "loss": 0.2303,
      "num_input_tokens_seen": 7478048,
      "step": 35440
    },
    {
      "epoch": 3.8993399339933994,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02918343486980775,
      "loss": 0.2314,
      "num_input_tokens_seen": 7479008,
      "step": 35445
    },
    {
      "epoch": 3.8998899889989,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.02918296615608144,
      "loss": 0.2325,
      "num_input_tokens_seen": 7480064,
      "step": 35450
    },
    {
      "epoch": 3.9004400440044007,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.02918249731163754,
      "loss": 0.2304,
      "num_input_tokens_seen": 7481120,
      "step": 35455
    },
    {
      "epoch": 3.900990099009901,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.029182028336480363,
      "loss": 0.2299,
      "num_input_tokens_seen": 7482176,
      "step": 35460
    },
    {
      "epoch": 3.9015401540154016,
      "grad_norm": 0.0028533935546875,
      "learning_rate": 0.02918155923061424,
      "loss": 0.2309,
      "num_input_tokens_seen": 7483200,
      "step": 35465
    },
    {
      "epoch": 3.902090209020902,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.029181089994043487,
      "loss": 0.2314,
      "num_input_tokens_seen": 7484256,
      "step": 35470
    },
    {
      "epoch": 3.9026402640264024,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.029180620626772437,
      "loss": 0.2299,
      "num_input_tokens_seen": 7485408,
      "step": 35475
    },
    {
      "epoch": 3.903190319031903,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.029180151128805405,
      "loss": 0.233,
      "num_input_tokens_seen": 7486464,
      "step": 35480
    },
    {
      "epoch": 3.9037403740374037,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.02917968150014673,
      "loss": 0.2325,
      "num_input_tokens_seen": 7487520,
      "step": 35485
    },
    {
      "epoch": 3.9042904290429044,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.02917921174080073,
      "loss": 0.233,
      "num_input_tokens_seen": 7488544,
      "step": 35490
    },
    {
      "epoch": 3.904840484048405,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.029178741850771742,
      "loss": 0.2314,
      "num_input_tokens_seen": 7489568,
      "step": 35495
    },
    {
      "epoch": 3.905390539053905,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.029178271830064093,
      "loss": 0.2309,
      "num_input_tokens_seen": 7490592,
      "step": 35500
    },
    {
      "epoch": 3.905940594059406,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029177801678682114,
      "loss": 0.2272,
      "num_input_tokens_seen": 7491648,
      "step": 35505
    },
    {
      "epoch": 3.9064906490649065,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02917733139663014,
      "loss": 0.2299,
      "num_input_tokens_seen": 7492672,
      "step": 35510
    },
    {
      "epoch": 3.907040704070407,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.02917686098391251,
      "loss": 0.231,
      "num_input_tokens_seen": 7493728,
      "step": 35515
    },
    {
      "epoch": 3.907590759075908,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029176390440533554,
      "loss": 0.2284,
      "num_input_tokens_seen": 7494752,
      "step": 35520
    },
    {
      "epoch": 3.908140814081408,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029175919766497607,
      "loss": 0.2352,
      "num_input_tokens_seen": 7495776,
      "step": 35525
    },
    {
      "epoch": 3.9086908690869087,
      "grad_norm": 0.01263427734375,
      "learning_rate": 0.029175448961809014,
      "loss": 0.2342,
      "num_input_tokens_seen": 7496832,
      "step": 35530
    },
    {
      "epoch": 3.9092409240924093,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029174978026472106,
      "loss": 0.23,
      "num_input_tokens_seen": 7497856,
      "step": 35535
    },
    {
      "epoch": 3.9097909790979095,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029174506960491228,
      "loss": 0.2315,
      "num_input_tokens_seen": 7498944,
      "step": 35540
    },
    {
      "epoch": 3.9103410341034106,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.02917403576387072,
      "loss": 0.23,
      "num_input_tokens_seen": 7500000,
      "step": 35545
    },
    {
      "epoch": 3.910891089108911,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029173564436614927,
      "loss": 0.2341,
      "num_input_tokens_seen": 7501088,
      "step": 35550
    },
    {
      "epoch": 3.9114411441144115,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.029173092978728193,
      "loss": 0.2346,
      "num_input_tokens_seen": 7502176,
      "step": 35555
    },
    {
      "epoch": 3.911991199119912,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.02917262139021486,
      "loss": 0.2309,
      "num_input_tokens_seen": 7503264,
      "step": 35560
    },
    {
      "epoch": 3.9125412541254123,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.029172149671079277,
      "loss": 0.233,
      "num_input_tokens_seen": 7504320,
      "step": 35565
    },
    {
      "epoch": 3.913091309130913,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.029171677821325792,
      "loss": 0.2283,
      "num_input_tokens_seen": 7505344,
      "step": 35570
    },
    {
      "epoch": 3.9136413641364136,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.029171205840958753,
      "loss": 0.2303,
      "num_input_tokens_seen": 7506432,
      "step": 35575
    },
    {
      "epoch": 3.9141914191419143,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.02917073372998251,
      "loss": 0.2314,
      "num_input_tokens_seen": 7507488,
      "step": 35580
    },
    {
      "epoch": 3.914741474147415,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02917026148840141,
      "loss": 0.2346,
      "num_input_tokens_seen": 7508544,
      "step": 35585
    },
    {
      "epoch": 3.915291529152915,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029169789116219813,
      "loss": 0.2325,
      "num_input_tokens_seen": 7509536,
      "step": 35590
    },
    {
      "epoch": 3.9158415841584158,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.02916931661344207,
      "loss": 0.2304,
      "num_input_tokens_seen": 7510656,
      "step": 35595
    },
    {
      "epoch": 3.9163916391639164,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.02916884398007253,
      "loss": 0.2335,
      "num_input_tokens_seen": 7511680,
      "step": 35600
    },
    {
      "epoch": 3.916941694169417,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.029168371216115558,
      "loss": 0.2319,
      "num_input_tokens_seen": 7512704,
      "step": 35605
    },
    {
      "epoch": 3.9174917491749177,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.029167898321575503,
      "loss": 0.2293,
      "num_input_tokens_seen": 7513792,
      "step": 35610
    },
    {
      "epoch": 3.918041804180418,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02916742529645673,
      "loss": 0.234,
      "num_input_tokens_seen": 7514816,
      "step": 35615
    },
    {
      "epoch": 3.9185918591859186,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029166952140763598,
      "loss": 0.2314,
      "num_input_tokens_seen": 7515872,
      "step": 35620
    },
    {
      "epoch": 3.919141914191419,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.029166478854500466,
      "loss": 0.2298,
      "num_input_tokens_seen": 7516928,
      "step": 35625
    },
    {
      "epoch": 3.9196919691969194,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029166005437671696,
      "loss": 0.2293,
      "num_input_tokens_seen": 7517952,
      "step": 35630
    },
    {
      "epoch": 3.9202420242024205,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.02916553189028165,
      "loss": 0.234,
      "num_input_tokens_seen": 7518944,
      "step": 35635
    },
    {
      "epoch": 3.9207920792079207,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.029165058212334692,
      "loss": 0.2319,
      "num_input_tokens_seen": 7520032,
      "step": 35640
    },
    {
      "epoch": 3.9213421342134214,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029164584403835195,
      "loss": 0.2319,
      "num_input_tokens_seen": 7521088,
      "step": 35645
    },
    {
      "epoch": 3.921892189218922,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.029164110464787514,
      "loss": 0.2299,
      "num_input_tokens_seen": 7522112,
      "step": 35650
    },
    {
      "epoch": 3.9224422442244222,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02916363639519603,
      "loss": 0.2309,
      "num_input_tokens_seen": 7523168,
      "step": 35655
    },
    {
      "epoch": 3.922992299229923,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.0291631621950651,
      "loss": 0.232,
      "num_input_tokens_seen": 7524224,
      "step": 35660
    },
    {
      "epoch": 3.9235423542354235,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.029162687864399104,
      "loss": 0.232,
      "num_input_tokens_seen": 7525312,
      "step": 35665
    },
    {
      "epoch": 3.924092409240924,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02916221340320241,
      "loss": 0.2299,
      "num_input_tokens_seen": 7526400,
      "step": 35670
    },
    {
      "epoch": 3.924642464246425,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.029161738811479387,
      "loss": 0.2363,
      "num_input_tokens_seen": 7527424,
      "step": 35675
    },
    {
      "epoch": 3.925192519251925,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.02916126408923441,
      "loss": 0.2368,
      "num_input_tokens_seen": 7528512,
      "step": 35680
    },
    {
      "epoch": 3.9257425742574257,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02916078923647187,
      "loss": 0.2309,
      "num_input_tokens_seen": 7529536,
      "step": 35685
    },
    {
      "epoch": 3.9262926292629263,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029160314253196117,
      "loss": 0.2335,
      "num_input_tokens_seen": 7530592,
      "step": 35690
    },
    {
      "epoch": 3.926842684268427,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.029159839139411548,
      "loss": 0.232,
      "num_input_tokens_seen": 7531584,
      "step": 35695
    },
    {
      "epoch": 3.9273927392739276,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.02915936389512254,
      "loss": 0.2309,
      "num_input_tokens_seen": 7532576,
      "step": 35700
    },
    {
      "epoch": 3.927942794279428,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.029158888520333465,
      "loss": 0.2314,
      "num_input_tokens_seen": 7533664,
      "step": 35705
    },
    {
      "epoch": 3.9284928492849285,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029158413015048706,
      "loss": 0.2308,
      "num_input_tokens_seen": 7534752,
      "step": 35710
    },
    {
      "epoch": 3.929042904290429,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029157937379272653,
      "loss": 0.2298,
      "num_input_tokens_seen": 7535776,
      "step": 35715
    },
    {
      "epoch": 3.9295929592959293,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02915746161300968,
      "loss": 0.2303,
      "num_input_tokens_seen": 7536832,
      "step": 35720
    },
    {
      "epoch": 3.93014301430143,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.029156985716264183,
      "loss": 0.2325,
      "num_input_tokens_seen": 7537856,
      "step": 35725
    },
    {
      "epoch": 3.9306930693069306,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029156509689040537,
      "loss": 0.2319,
      "num_input_tokens_seen": 7538976,
      "step": 35730
    },
    {
      "epoch": 3.9312431243124313,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.029156033531343137,
      "loss": 0.2309,
      "num_input_tokens_seen": 7540096,
      "step": 35735
    },
    {
      "epoch": 3.931793179317932,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029155557243176367,
      "loss": 0.2303,
      "num_input_tokens_seen": 7541120,
      "step": 35740
    },
    {
      "epoch": 3.932343234323432,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02915508082454462,
      "loss": 0.2314,
      "num_input_tokens_seen": 7542208,
      "step": 35745
    },
    {
      "epoch": 3.932893289328933,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.029154604275452283,
      "loss": 0.2303,
      "num_input_tokens_seen": 7543232,
      "step": 35750
    },
    {
      "epoch": 3.9334433443344334,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.029154127595903752,
      "loss": 0.2293,
      "num_input_tokens_seen": 7544256,
      "step": 35755
    },
    {
      "epoch": 3.933993399339934,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02915365078590342,
      "loss": 0.2319,
      "num_input_tokens_seen": 7545376,
      "step": 35760
    },
    {
      "epoch": 3.9345434543454347,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.029153173845455675,
      "loss": 0.2267,
      "num_input_tokens_seen": 7546464,
      "step": 35765
    },
    {
      "epoch": 3.935093509350935,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029152696774564925,
      "loss": 0.2315,
      "num_input_tokens_seen": 7547520,
      "step": 35770
    },
    {
      "epoch": 3.9356435643564356,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.029152219573235556,
      "loss": 0.2351,
      "num_input_tokens_seen": 7548544,
      "step": 35775
    },
    {
      "epoch": 3.9361936193619362,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.029151742241471972,
      "loss": 0.2294,
      "num_input_tokens_seen": 7549632,
      "step": 35780
    },
    {
      "epoch": 3.936743674367437,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02915126477927857,
      "loss": 0.2294,
      "num_input_tokens_seen": 7550656,
      "step": 35785
    },
    {
      "epoch": 3.9372937293729375,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.029150787186659748,
      "loss": 0.2321,
      "num_input_tokens_seen": 7551712,
      "step": 35790
    },
    {
      "epoch": 3.9378437843784377,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029150309463619913,
      "loss": 0.2331,
      "num_input_tokens_seen": 7552768,
      "step": 35795
    },
    {
      "epoch": 3.9383938393839384,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029149831610163466,
      "loss": 0.2357,
      "num_input_tokens_seen": 7553792,
      "step": 35800
    },
    {
      "epoch": 3.938943894389439,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.02914935362629481,
      "loss": 0.2299,
      "num_input_tokens_seen": 7554880,
      "step": 35805
    },
    {
      "epoch": 3.9394939493949392,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.02914887551201835,
      "loss": 0.2304,
      "num_input_tokens_seen": 7555936,
      "step": 35810
    },
    {
      "epoch": 3.94004400440044,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029148397267338495,
      "loss": 0.231,
      "num_input_tokens_seen": 7556928,
      "step": 35815
    },
    {
      "epoch": 3.9405940594059405,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.029147918892259653,
      "loss": 0.2288,
      "num_input_tokens_seen": 7557984,
      "step": 35820
    },
    {
      "epoch": 3.941144114411441,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.02914744038678623,
      "loss": 0.23,
      "num_input_tokens_seen": 7559072,
      "step": 35825
    },
    {
      "epoch": 3.941694169416942,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029146961750922636,
      "loss": 0.2346,
      "num_input_tokens_seen": 7560160,
      "step": 35830
    },
    {
      "epoch": 3.942244224422442,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.029146482984673285,
      "loss": 0.233,
      "num_input_tokens_seen": 7561184,
      "step": 35835
    },
    {
      "epoch": 3.9427942794279427,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02914600408804259,
      "loss": 0.2279,
      "num_input_tokens_seen": 7562208,
      "step": 35840
    },
    {
      "epoch": 3.9433443344334433,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02914552506103496,
      "loss": 0.23,
      "num_input_tokens_seen": 7563200,
      "step": 35845
    },
    {
      "epoch": 3.943894389438944,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.029145045903654816,
      "loss": 0.2279,
      "num_input_tokens_seen": 7564256,
      "step": 35850
    },
    {
      "epoch": 3.9444444444444446,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.029144566615906572,
      "loss": 0.2373,
      "num_input_tokens_seen": 7565312,
      "step": 35855
    },
    {
      "epoch": 3.944994499449945,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.029144087197794644,
      "loss": 0.2306,
      "num_input_tokens_seen": 7566336,
      "step": 35860
    },
    {
      "epoch": 3.9455445544554455,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.029143607649323448,
      "loss": 0.2326,
      "num_input_tokens_seen": 7567360,
      "step": 35865
    },
    {
      "epoch": 3.946094609460946,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.029143127970497413,
      "loss": 0.2341,
      "num_input_tokens_seen": 7568480,
      "step": 35870
    },
    {
      "epoch": 3.946644664466447,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.02914264816132095,
      "loss": 0.2331,
      "num_input_tokens_seen": 7569536,
      "step": 35875
    },
    {
      "epoch": 3.9471947194719474,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029142168221798485,
      "loss": 0.2346,
      "num_input_tokens_seen": 7570560,
      "step": 35880
    },
    {
      "epoch": 3.9477447744774476,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.029141688151934443,
      "loss": 0.2325,
      "num_input_tokens_seen": 7571584,
      "step": 35885
    },
    {
      "epoch": 3.9482948294829483,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029141207951733247,
      "loss": 0.232,
      "num_input_tokens_seen": 7572672,
      "step": 35890
    },
    {
      "epoch": 3.948844884488449,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02914072762119932,
      "loss": 0.2309,
      "num_input_tokens_seen": 7573728,
      "step": 35895
    },
    {
      "epoch": 3.949394939493949,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0291402471603371,
      "loss": 0.2309,
      "num_input_tokens_seen": 7574752,
      "step": 35900
    },
    {
      "epoch": 3.94994499449945,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.029139766569150998,
      "loss": 0.2324,
      "num_input_tokens_seen": 7575840,
      "step": 35905
    },
    {
      "epoch": 3.9504950495049505,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.029139285847645455,
      "loss": 0.234,
      "num_input_tokens_seen": 7576896,
      "step": 35910
    },
    {
      "epoch": 3.951045104510451,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.0291388049958249,
      "loss": 0.2309,
      "num_input_tokens_seen": 7577952,
      "step": 35915
    },
    {
      "epoch": 3.9515951595159517,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02913832401369376,
      "loss": 0.2314,
      "num_input_tokens_seen": 7579008,
      "step": 35920
    },
    {
      "epoch": 3.952145214521452,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02913784290125648,
      "loss": 0.2319,
      "num_input_tokens_seen": 7580096,
      "step": 35925
    },
    {
      "epoch": 3.9526952695269526,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02913736165851748,
      "loss": 0.233,
      "num_input_tokens_seen": 7581120,
      "step": 35930
    },
    {
      "epoch": 3.9532453245324533,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.0291368802854812,
      "loss": 0.2309,
      "num_input_tokens_seen": 7582144,
      "step": 35935
    },
    {
      "epoch": 3.953795379537954,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029136398782152087,
      "loss": 0.2325,
      "num_input_tokens_seen": 7583168,
      "step": 35940
    },
    {
      "epoch": 3.9543454345434546,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.029135917148534564,
      "loss": 0.2294,
      "num_input_tokens_seen": 7584192,
      "step": 35945
    },
    {
      "epoch": 3.9548954895489548,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.029135435384633073,
      "loss": 0.2305,
      "num_input_tokens_seen": 7585248,
      "step": 35950
    },
    {
      "epoch": 3.9554455445544554,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029134953490452063,
      "loss": 0.23,
      "num_input_tokens_seen": 7586272,
      "step": 35955
    },
    {
      "epoch": 3.955995599559956,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.029134471465995966,
      "loss": 0.2336,
      "num_input_tokens_seen": 7587328,
      "step": 35960
    },
    {
      "epoch": 3.9565456545654567,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.029133989311269225,
      "loss": 0.231,
      "num_input_tokens_seen": 7588384,
      "step": 35965
    },
    {
      "epoch": 3.9570957095709574,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.02913350702627629,
      "loss": 0.2341,
      "num_input_tokens_seen": 7589440,
      "step": 35970
    },
    {
      "epoch": 3.9576457645764576,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029133024611021602,
      "loss": 0.2325,
      "num_input_tokens_seen": 7590496,
      "step": 35975
    },
    {
      "epoch": 3.958195819581958,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029132542065509606,
      "loss": 0.2299,
      "num_input_tokens_seen": 7591584,
      "step": 35980
    },
    {
      "epoch": 3.958745874587459,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02913205938974475,
      "loss": 0.2346,
      "num_input_tokens_seen": 7592608,
      "step": 35985
    },
    {
      "epoch": 3.959295929592959,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029131576583731486,
      "loss": 0.2299,
      "num_input_tokens_seen": 7593696,
      "step": 35990
    },
    {
      "epoch": 3.9598459845984597,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.029131093647474262,
      "loss": 0.2305,
      "num_input_tokens_seen": 7594816,
      "step": 35995
    },
    {
      "epoch": 3.9603960396039604,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02913061058097753,
      "loss": 0.2325,
      "num_input_tokens_seen": 7595808,
      "step": 36000
    },
    {
      "epoch": 3.960946094609461,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.029130127384245738,
      "loss": 0.2299,
      "num_input_tokens_seen": 7596864,
      "step": 36005
    },
    {
      "epoch": 3.9614961496149617,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029129644057283348,
      "loss": 0.232,
      "num_input_tokens_seen": 7597920,
      "step": 36010
    },
    {
      "epoch": 3.962046204620462,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.029129160600094805,
      "loss": 0.23,
      "num_input_tokens_seen": 7598912,
      "step": 36015
    },
    {
      "epoch": 3.9625962596259625,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029128677012684567,
      "loss": 0.229,
      "num_input_tokens_seen": 7599936,
      "step": 36020
    },
    {
      "epoch": 3.963146314631463,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.029128193295057097,
      "loss": 0.2331,
      "num_input_tokens_seen": 7600992,
      "step": 36025
    },
    {
      "epoch": 3.963696369636964,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02912770944721684,
      "loss": 0.2284,
      "num_input_tokens_seen": 7602016,
      "step": 36030
    },
    {
      "epoch": 3.9642464246424645,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02912722546916827,
      "loss": 0.2353,
      "num_input_tokens_seen": 7603072,
      "step": 36035
    },
    {
      "epoch": 3.9647964796479647,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029126741360915847,
      "loss": 0.2306,
      "num_input_tokens_seen": 7604032,
      "step": 36040
    },
    {
      "epoch": 3.9653465346534653,
      "grad_norm": 0.000766754150390625,
      "learning_rate": 0.029126257122464017,
      "loss": 0.228,
      "num_input_tokens_seen": 7605120,
      "step": 36045
    },
    {
      "epoch": 3.965896589658966,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.02912577275381726,
      "loss": 0.2347,
      "num_input_tokens_seen": 7606112,
      "step": 36050
    },
    {
      "epoch": 3.966446644664466,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.029125288254980033,
      "loss": 0.2332,
      "num_input_tokens_seen": 7607168,
      "step": 36055
    },
    {
      "epoch": 3.9669966996699673,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.0291248036259568,
      "loss": 0.2316,
      "num_input_tokens_seen": 7608256,
      "step": 36060
    },
    {
      "epoch": 3.9675467546754675,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02912431886675203,
      "loss": 0.2343,
      "num_input_tokens_seen": 7609280,
      "step": 36065
    },
    {
      "epoch": 3.968096809680968,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029123833977370193,
      "loss": 0.2315,
      "num_input_tokens_seen": 7610336,
      "step": 36070
    },
    {
      "epoch": 3.9686468646864688,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.029123348957815754,
      "loss": 0.2341,
      "num_input_tokens_seen": 7611456,
      "step": 36075
    },
    {
      "epoch": 3.969196919691969,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.02912286380809319,
      "loss": 0.232,
      "num_input_tokens_seen": 7612576,
      "step": 36080
    },
    {
      "epoch": 3.9697469746974696,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029122378528206958,
      "loss": 0.2309,
      "num_input_tokens_seen": 7613696,
      "step": 36085
    },
    {
      "epoch": 3.9702970297029703,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.029121893118161542,
      "loss": 0.2309,
      "num_input_tokens_seen": 7614720,
      "step": 36090
    },
    {
      "epoch": 3.970847084708471,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.029121407577961417,
      "loss": 0.2325,
      "num_input_tokens_seen": 7615776,
      "step": 36095
    },
    {
      "epoch": 3.9713971397139716,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.029120921907611053,
      "loss": 0.2304,
      "num_input_tokens_seen": 7616800,
      "step": 36100
    },
    {
      "epoch": 3.9719471947194718,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.029120436107114928,
      "loss": 0.2284,
      "num_input_tokens_seen": 7617920,
      "step": 36105
    },
    {
      "epoch": 3.9724972497249724,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.029119950176477515,
      "loss": 0.231,
      "num_input_tokens_seen": 7619008,
      "step": 36110
    },
    {
      "epoch": 3.973047304730473,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.029119464115703302,
      "loss": 0.2289,
      "num_input_tokens_seen": 7620064,
      "step": 36115
    },
    {
      "epoch": 3.9735973597359737,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.029118977924796764,
      "loss": 0.2316,
      "num_input_tokens_seen": 7621184,
      "step": 36120
    },
    {
      "epoch": 3.9741474147414744,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.029118491603762373,
      "loss": 0.2331,
      "num_input_tokens_seen": 7622336,
      "step": 36125
    },
    {
      "epoch": 3.9746974697469746,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02911800515260463,
      "loss": 0.2331,
      "num_input_tokens_seen": 7623456,
      "step": 36130
    },
    {
      "epoch": 3.9752475247524752,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.029117518571328,
      "loss": 0.2351,
      "num_input_tokens_seen": 7624512,
      "step": 36135
    },
    {
      "epoch": 3.975797579757976,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02911703185993698,
      "loss": 0.234,
      "num_input_tokens_seen": 7625536,
      "step": 36140
    },
    {
      "epoch": 3.976347634763476,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.029116545018436052,
      "loss": 0.2329,
      "num_input_tokens_seen": 7626592,
      "step": 36145
    },
    {
      "epoch": 3.976897689768977,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0291160580468297,
      "loss": 0.2314,
      "num_input_tokens_seen": 7627680,
      "step": 36150
    },
    {
      "epoch": 3.9774477447744774,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029115570945122418,
      "loss": 0.2308,
      "num_input_tokens_seen": 7628736,
      "step": 36155
    },
    {
      "epoch": 3.977997799779978,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.029115083713318685,
      "loss": 0.2309,
      "num_input_tokens_seen": 7629728,
      "step": 36160
    },
    {
      "epoch": 3.9785478547854787,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029114596351423005,
      "loss": 0.2319,
      "num_input_tokens_seen": 7630816,
      "step": 36165
    },
    {
      "epoch": 3.979097909790979,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.02911410885943986,
      "loss": 0.2298,
      "num_input_tokens_seen": 7631872,
      "step": 36170
    },
    {
      "epoch": 3.9796479647964795,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02911362123737375,
      "loss": 0.2309,
      "num_input_tokens_seen": 7632992,
      "step": 36175
    },
    {
      "epoch": 3.98019801980198,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.029113133485229162,
      "loss": 0.2319,
      "num_input_tokens_seen": 7634048,
      "step": 36180
    },
    {
      "epoch": 3.980748074807481,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029112645603010592,
      "loss": 0.2303,
      "num_input_tokens_seen": 7635168,
      "step": 36185
    },
    {
      "epoch": 3.9812981298129815,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02911215759072254,
      "loss": 0.2308,
      "num_input_tokens_seen": 7636288,
      "step": 36190
    },
    {
      "epoch": 3.9818481848184817,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029111669448369508,
      "loss": 0.2294,
      "num_input_tokens_seen": 7637312,
      "step": 36195
    },
    {
      "epoch": 3.9823982398239823,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02911118117595599,
      "loss": 0.2286,
      "num_input_tokens_seen": 7638400,
      "step": 36200
    },
    {
      "epoch": 3.982948294829483,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.02911069277348648,
      "loss": 0.237,
      "num_input_tokens_seen": 7639488,
      "step": 36205
    },
    {
      "epoch": 3.9834983498349836,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.029110204240965495,
      "loss": 0.2275,
      "num_input_tokens_seen": 7640512,
      "step": 36210
    },
    {
      "epoch": 3.9840484048404843,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02910971557839752,
      "loss": 0.2234,
      "num_input_tokens_seen": 7641568,
      "step": 36215
    },
    {
      "epoch": 3.9845984598459845,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.029109226785787073,
      "loss": 0.2339,
      "num_input_tokens_seen": 7642592,
      "step": 36220
    },
    {
      "epoch": 3.985148514851485,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.029108737863138646,
      "loss": 0.2329,
      "num_input_tokens_seen": 7643648,
      "step": 36225
    },
    {
      "epoch": 3.985698569856986,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.029108248810456757,
      "loss": 0.2231,
      "num_input_tokens_seen": 7644768,
      "step": 36230
    },
    {
      "epoch": 3.986248624862486,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.029107759627745908,
      "loss": 0.2283,
      "num_input_tokens_seen": 7645856,
      "step": 36235
    },
    {
      "epoch": 3.9867986798679866,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.02910727031501061,
      "loss": 0.227,
      "num_input_tokens_seen": 7646912,
      "step": 36240
    },
    {
      "epoch": 3.9873487348734873,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029106780872255367,
      "loss": 0.2282,
      "num_input_tokens_seen": 7647936,
      "step": 36245
    },
    {
      "epoch": 3.987898789878988,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.029106291299484697,
      "loss": 0.2432,
      "num_input_tokens_seen": 7648960,
      "step": 36250
    },
    {
      "epoch": 3.9884488448844886,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.029105801596703104,
      "loss": 0.2346,
      "num_input_tokens_seen": 7649984,
      "step": 36255
    },
    {
      "epoch": 3.988998899889989,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.029105311763915113,
      "loss": 0.2294,
      "num_input_tokens_seen": 7651040,
      "step": 36260
    },
    {
      "epoch": 3.9895489548954894,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.029104821801125228,
      "loss": 0.2396,
      "num_input_tokens_seen": 7652128,
      "step": 36265
    },
    {
      "epoch": 3.99009900990099,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.02910433170833797,
      "loss": 0.2307,
      "num_input_tokens_seen": 7653216,
      "step": 36270
    },
    {
      "epoch": 3.9906490649064907,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.029103841485557853,
      "loss": 0.2301,
      "num_input_tokens_seen": 7654272,
      "step": 36275
    },
    {
      "epoch": 3.9911991199119914,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.0291033511327894,
      "loss": 0.2337,
      "num_input_tokens_seen": 7655328,
      "step": 36280
    },
    {
      "epoch": 3.9917491749174916,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.029102860650037123,
      "loss": 0.2316,
      "num_input_tokens_seen": 7656416,
      "step": 36285
    },
    {
      "epoch": 3.9922992299229922,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.02910237003730555,
      "loss": 0.2305,
      "num_input_tokens_seen": 7657472,
      "step": 36290
    },
    {
      "epoch": 3.992849284928493,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.029101879294599196,
      "loss": 0.2352,
      "num_input_tokens_seen": 7658464,
      "step": 36295
    },
    {
      "epoch": 3.9933993399339935,
      "grad_norm": 0.000701904296875,
      "learning_rate": 0.02910138842192259,
      "loss": 0.2284,
      "num_input_tokens_seen": 7659488,
      "step": 36300
    },
    {
      "epoch": 3.993949394939494,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029100897419280253,
      "loss": 0.23,
      "num_input_tokens_seen": 7660608,
      "step": 36305
    },
    {
      "epoch": 3.9944994499449944,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.029100406286676705,
      "loss": 0.232,
      "num_input_tokens_seen": 7661664,
      "step": 36310
    },
    {
      "epoch": 3.995049504950495,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.029099915024116487,
      "loss": 0.2294,
      "num_input_tokens_seen": 7662688,
      "step": 36315
    },
    {
      "epoch": 3.9955995599559957,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.029099423631604115,
      "loss": 0.2289,
      "num_input_tokens_seen": 7663744,
      "step": 36320
    },
    {
      "epoch": 3.996149614961496,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02909893210914412,
      "loss": 0.2253,
      "num_input_tokens_seen": 7664832,
      "step": 36325
    },
    {
      "epoch": 3.9966996699669965,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02909844045674103,
      "loss": 0.2338,
      "num_input_tokens_seen": 7665856,
      "step": 36330
    },
    {
      "epoch": 3.997249724972497,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.029097948674399382,
      "loss": 0.2312,
      "num_input_tokens_seen": 7666912,
      "step": 36335
    },
    {
      "epoch": 3.997799779977998,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02909745676212371,
      "loss": 0.2395,
      "num_input_tokens_seen": 7667904,
      "step": 36340
    },
    {
      "epoch": 3.9983498349834985,
      "grad_norm": 0.0026397705078125,
      "learning_rate": 0.029096964719918537,
      "loss": 0.2297,
      "num_input_tokens_seen": 7669024,
      "step": 36345
    },
    {
      "epoch": 3.9988998899889987,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02909647254778841,
      "loss": 0.2321,
      "num_input_tokens_seen": 7670016,
      "step": 36350
    },
    {
      "epoch": 3.9994499449944994,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.029095980245737858,
      "loss": 0.2301,
      "num_input_tokens_seen": 7671072,
      "step": 36355
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.029095487813771418,
      "loss": 0.2317,
      "num_input_tokens_seen": 7672000,
      "step": 36360
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.23157790303230286,
      "eval_runtime": 60.533,
      "eval_samples_per_second": 66.74,
      "eval_steps_per_second": 16.685,
      "num_input_tokens_seen": 7672000,
      "step": 36360
    },
    {
      "epoch": 4.0005500550055,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.029094995251893633,
      "loss": 0.2327,
      "num_input_tokens_seen": 7673024,
      "step": 36365
    },
    {
      "epoch": 4.001100110011001,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.02909450256010904,
      "loss": 0.2352,
      "num_input_tokens_seen": 7674144,
      "step": 36370
    },
    {
      "epoch": 4.0016501650165015,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02909400973842218,
      "loss": 0.2296,
      "num_input_tokens_seen": 7675168,
      "step": 36375
    },
    {
      "epoch": 4.002200220022003,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.029093516786837595,
      "loss": 0.2342,
      "num_input_tokens_seen": 7676192,
      "step": 36380
    },
    {
      "epoch": 4.002750275027503,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.02909302370535983,
      "loss": 0.2284,
      "num_input_tokens_seen": 7677312,
      "step": 36385
    },
    {
      "epoch": 4.003300330033003,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02909253049399342,
      "loss": 0.2368,
      "num_input_tokens_seen": 7678368,
      "step": 36390
    },
    {
      "epoch": 4.003850385038504,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.029092037152742927,
      "loss": 0.2326,
      "num_input_tokens_seen": 7679488,
      "step": 36395
    },
    {
      "epoch": 4.004400440044004,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02909154368161289,
      "loss": 0.2325,
      "num_input_tokens_seen": 7680448,
      "step": 36400
    },
    {
      "epoch": 4.0049504950495045,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.029091050080607854,
      "loss": 0.2314,
      "num_input_tokens_seen": 7681472,
      "step": 36405
    },
    {
      "epoch": 4.005500550055006,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.029090556349732368,
      "loss": 0.2309,
      "num_input_tokens_seen": 7682592,
      "step": 36410
    },
    {
      "epoch": 4.006050605060506,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029090062488990986,
      "loss": 0.2335,
      "num_input_tokens_seen": 7683616,
      "step": 36415
    },
    {
      "epoch": 4.006600660066007,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.029089568498388264,
      "loss": 0.2335,
      "num_input_tokens_seen": 7684704,
      "step": 36420
    },
    {
      "epoch": 4.007150715071507,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.02908907437792875,
      "loss": 0.2319,
      "num_input_tokens_seen": 7685760,
      "step": 36425
    },
    {
      "epoch": 4.007700770077007,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02908858012761699,
      "loss": 0.2319,
      "num_input_tokens_seen": 7686912,
      "step": 36430
    },
    {
      "epoch": 4.008250825082508,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.029088085747457554,
      "loss": 0.2319,
      "num_input_tokens_seen": 7688096,
      "step": 36435
    },
    {
      "epoch": 4.008800880088009,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.029087591237454993,
      "loss": 0.2314,
      "num_input_tokens_seen": 7689152,
      "step": 36440
    },
    {
      "epoch": 4.00935093509351,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02908709659761386,
      "loss": 0.2319,
      "num_input_tokens_seen": 7690112,
      "step": 36445
    },
    {
      "epoch": 4.00990099009901,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.029086601827938718,
      "loss": 0.2314,
      "num_input_tokens_seen": 7691200,
      "step": 36450
    },
    {
      "epoch": 4.01045104510451,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029086106928434125,
      "loss": 0.2324,
      "num_input_tokens_seen": 7692160,
      "step": 36455
    },
    {
      "epoch": 4.011001100110011,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029085611899104646,
      "loss": 0.2324,
      "num_input_tokens_seen": 7693248,
      "step": 36460
    },
    {
      "epoch": 4.011551155115511,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02908511673995484,
      "loss": 0.2303,
      "num_input_tokens_seen": 7694240,
      "step": 36465
    },
    {
      "epoch": 4.0121012101210125,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.029084621450989275,
      "loss": 0.2324,
      "num_input_tokens_seen": 7695392,
      "step": 36470
    },
    {
      "epoch": 4.012651265126513,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.029084126032212505,
      "loss": 0.2324,
      "num_input_tokens_seen": 7696512,
      "step": 36475
    },
    {
      "epoch": 4.013201320132013,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02908363048362911,
      "loss": 0.2314,
      "num_input_tokens_seen": 7697536,
      "step": 36480
    },
    {
      "epoch": 4.013751375137514,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02908313480524365,
      "loss": 0.2329,
      "num_input_tokens_seen": 7698624,
      "step": 36485
    },
    {
      "epoch": 4.014301430143014,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029082638997060695,
      "loss": 0.234,
      "num_input_tokens_seen": 7699776,
      "step": 36490
    },
    {
      "epoch": 4.014851485148514,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.02908214305908481,
      "loss": 0.2313,
      "num_input_tokens_seen": 7700800,
      "step": 36495
    },
    {
      "epoch": 4.0154015401540155,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029081646991320573,
      "loss": 0.2314,
      "num_input_tokens_seen": 7701888,
      "step": 36500
    },
    {
      "epoch": 4.015951595159516,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.02908115079377255,
      "loss": 0.2319,
      "num_input_tokens_seen": 7703008,
      "step": 36505
    },
    {
      "epoch": 4.016501650165017,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029080654466445317,
      "loss": 0.2319,
      "num_input_tokens_seen": 7704000,
      "step": 36510
    },
    {
      "epoch": 4.017051705170517,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02908015800934345,
      "loss": 0.2303,
      "num_input_tokens_seen": 7705056,
      "step": 36515
    },
    {
      "epoch": 4.017601760176017,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.029079661422471526,
      "loss": 0.2308,
      "num_input_tokens_seen": 7706176,
      "step": 36520
    },
    {
      "epoch": 4.018151815181518,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.029079164705834114,
      "loss": 0.2314,
      "num_input_tokens_seen": 7707264,
      "step": 36525
    },
    {
      "epoch": 4.0187018701870185,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029078667859435803,
      "loss": 0.2309,
      "num_input_tokens_seen": 7708384,
      "step": 36530
    },
    {
      "epoch": 4.01925192519252,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.02907817088328116,
      "loss": 0.2314,
      "num_input_tokens_seen": 7709408,
      "step": 36535
    },
    {
      "epoch": 4.01980198019802,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.029077673777374767,
      "loss": 0.2314,
      "num_input_tokens_seen": 7710464,
      "step": 36540
    },
    {
      "epoch": 4.02035203520352,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.029077176541721216,
      "loss": 0.2294,
      "num_input_tokens_seen": 7711520,
      "step": 36545
    },
    {
      "epoch": 4.020902090209021,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029076679176325088,
      "loss": 0.2319,
      "num_input_tokens_seen": 7712608,
      "step": 36550
    },
    {
      "epoch": 4.021452145214521,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.029076181681190957,
      "loss": 0.2335,
      "num_input_tokens_seen": 7713664,
      "step": 36555
    },
    {
      "epoch": 4.022002200220022,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.029075684056323412,
      "loss": 0.2309,
      "num_input_tokens_seen": 7714720,
      "step": 36560
    },
    {
      "epoch": 4.022552255225523,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029075186301727046,
      "loss": 0.2299,
      "num_input_tokens_seen": 7715776,
      "step": 36565
    },
    {
      "epoch": 4.023102310231023,
      "grad_norm": 0.01202392578125,
      "learning_rate": 0.029074688417406437,
      "loss": 0.2356,
      "num_input_tokens_seen": 7716832,
      "step": 36570
    },
    {
      "epoch": 4.023652365236524,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.02907419040336618,
      "loss": 0.2314,
      "num_input_tokens_seen": 7717824,
      "step": 36575
    },
    {
      "epoch": 4.024202420242024,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.02907369225961087,
      "loss": 0.2314,
      "num_input_tokens_seen": 7718848,
      "step": 36580
    },
    {
      "epoch": 4.024752475247524,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029073193986145083,
      "loss": 0.2309,
      "num_input_tokens_seen": 7719936,
      "step": 36585
    },
    {
      "epoch": 4.025302530253025,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029072695582973426,
      "loss": 0.2272,
      "num_input_tokens_seen": 7721024,
      "step": 36590
    },
    {
      "epoch": 4.025852585258526,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029072197050100484,
      "loss": 0.2309,
      "num_input_tokens_seen": 7722080,
      "step": 36595
    },
    {
      "epoch": 4.026402640264027,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.029071698387530855,
      "loss": 0.2309,
      "num_input_tokens_seen": 7723168,
      "step": 36600
    },
    {
      "epoch": 4.026952695269527,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02907119959526913,
      "loss": 0.2351,
      "num_input_tokens_seen": 7724160,
      "step": 36605
    },
    {
      "epoch": 4.027502750275027,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.02907070067331991,
      "loss": 0.2319,
      "num_input_tokens_seen": 7725184,
      "step": 36610
    },
    {
      "epoch": 4.028052805280528,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0290702016216878,
      "loss": 0.2288,
      "num_input_tokens_seen": 7726240,
      "step": 36615
    },
    {
      "epoch": 4.028602860286028,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.029069702440377394,
      "loss": 0.2309,
      "num_input_tokens_seen": 7727360,
      "step": 36620
    },
    {
      "epoch": 4.0291529152915295,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.029069203129393286,
      "loss": 0.2294,
      "num_input_tokens_seen": 7728416,
      "step": 36625
    },
    {
      "epoch": 4.02970297029703,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.029068703688740087,
      "loss": 0.231,
      "num_input_tokens_seen": 7729408,
      "step": 36630
    },
    {
      "epoch": 4.03025302530253,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.029068204118422397,
      "loss": 0.2295,
      "num_input_tokens_seen": 7730432,
      "step": 36635
    },
    {
      "epoch": 4.030803080308031,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.02906770441844482,
      "loss": 0.2342,
      "num_input_tokens_seen": 7731456,
      "step": 36640
    },
    {
      "epoch": 4.031353135313531,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02906720458881196,
      "loss": 0.2269,
      "num_input_tokens_seen": 7732512,
      "step": 36645
    },
    {
      "epoch": 4.031903190319032,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.029066704629528428,
      "loss": 0.2357,
      "num_input_tokens_seen": 7733536,
      "step": 36650
    },
    {
      "epoch": 4.0324532453245325,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.029066204540598828,
      "loss": 0.23,
      "num_input_tokens_seen": 7734624,
      "step": 36655
    },
    {
      "epoch": 4.033003300330033,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.02906570432202777,
      "loss": 0.2346,
      "num_input_tokens_seen": 7735712,
      "step": 36660
    },
    {
      "epoch": 4.033553355335534,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.029065203973819863,
      "loss": 0.2309,
      "num_input_tokens_seen": 7736832,
      "step": 36665
    },
    {
      "epoch": 4.034103410341034,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.029064703495979727,
      "loss": 0.2315,
      "num_input_tokens_seen": 7737920,
      "step": 36670
    },
    {
      "epoch": 4.034653465346534,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02906420288851196,
      "loss": 0.2314,
      "num_input_tokens_seen": 7738976,
      "step": 36675
    },
    {
      "epoch": 4.035203520352035,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02906370215142119,
      "loss": 0.2294,
      "num_input_tokens_seen": 7740032,
      "step": 36680
    },
    {
      "epoch": 4.0357535753575355,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02906320128471202,
      "loss": 0.233,
      "num_input_tokens_seen": 7741056,
      "step": 36685
    },
    {
      "epoch": 4.036303630363037,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.029062700288389073,
      "loss": 0.2299,
      "num_input_tokens_seen": 7742080,
      "step": 36690
    },
    {
      "epoch": 4.036853685368537,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.029062199162456965,
      "loss": 0.2319,
      "num_input_tokens_seen": 7743136,
      "step": 36695
    },
    {
      "epoch": 4.037403740374037,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.029061697906920318,
      "loss": 0.2309,
      "num_input_tokens_seen": 7744256,
      "step": 36700
    },
    {
      "epoch": 4.037953795379538,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.02906119652178375,
      "loss": 0.2341,
      "num_input_tokens_seen": 7745344,
      "step": 36705
    },
    {
      "epoch": 4.038503850385038,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.02906069500705187,
      "loss": 0.2329,
      "num_input_tokens_seen": 7746336,
      "step": 36710
    },
    {
      "epoch": 4.039053905390539,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.029060193362729324,
      "loss": 0.233,
      "num_input_tokens_seen": 7747360,
      "step": 36715
    },
    {
      "epoch": 4.03960396039604,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.029059691588820716,
      "loss": 0.2314,
      "num_input_tokens_seen": 7748384,
      "step": 36720
    },
    {
      "epoch": 4.04015401540154,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.02905918968533068,
      "loss": 0.233,
      "num_input_tokens_seen": 7749408,
      "step": 36725
    },
    {
      "epoch": 4.040704070407041,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02905868765226384,
      "loss": 0.2304,
      "num_input_tokens_seen": 7750432,
      "step": 36730
    },
    {
      "epoch": 4.041254125412541,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.029058185489624817,
      "loss": 0.232,
      "num_input_tokens_seen": 7751424,
      "step": 36735
    },
    {
      "epoch": 4.041804180418042,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.02905768319741825,
      "loss": 0.2304,
      "num_input_tokens_seen": 7752480,
      "step": 36740
    },
    {
      "epoch": 4.042354235423542,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02905718077564876,
      "loss": 0.2252,
      "num_input_tokens_seen": 7753504,
      "step": 36745
    },
    {
      "epoch": 4.042904290429043,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.029056678224320977,
      "loss": 0.2331,
      "num_input_tokens_seen": 7754624,
      "step": 36750
    },
    {
      "epoch": 4.043454345434544,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02905617554343954,
      "loss": 0.2305,
      "num_input_tokens_seen": 7755680,
      "step": 36755
    },
    {
      "epoch": 4.044004400440044,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.029055672733009075,
      "loss": 0.2332,
      "num_input_tokens_seen": 7756736,
      "step": 36760
    },
    {
      "epoch": 4.044554455445544,
      "grad_norm": 0.005859375,
      "learning_rate": 0.029055169793034224,
      "loss": 0.2321,
      "num_input_tokens_seen": 7757760,
      "step": 36765
    },
    {
      "epoch": 4.045104510451045,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.02905466672351961,
      "loss": 0.2364,
      "num_input_tokens_seen": 7758816,
      "step": 36770
    },
    {
      "epoch": 4.0456545654565454,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02905416352446988,
      "loss": 0.2296,
      "num_input_tokens_seen": 7759904,
      "step": 36775
    },
    {
      "epoch": 4.0462046204620465,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02905366019588967,
      "loss": 0.2352,
      "num_input_tokens_seen": 7760960,
      "step": 36780
    },
    {
      "epoch": 4.046754675467547,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.029053156737783616,
      "loss": 0.232,
      "num_input_tokens_seen": 7761984,
      "step": 36785
    },
    {
      "epoch": 4.047304730473047,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02905265315015636,
      "loss": 0.2346,
      "num_input_tokens_seen": 7763072,
      "step": 36790
    },
    {
      "epoch": 4.047854785478548,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.029052149433012545,
      "loss": 0.2299,
      "num_input_tokens_seen": 7764160,
      "step": 36795
    },
    {
      "epoch": 4.048404840484048,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02905164558635681,
      "loss": 0.2314,
      "num_input_tokens_seen": 7765248,
      "step": 36800
    },
    {
      "epoch": 4.048954895489549,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029051141610193797,
      "loss": 0.2325,
      "num_input_tokens_seen": 7766304,
      "step": 36805
    },
    {
      "epoch": 4.0495049504950495,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.029050637504528156,
      "loss": 0.2309,
      "num_input_tokens_seen": 7767360,
      "step": 36810
    },
    {
      "epoch": 4.05005500550055,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029050133269364533,
      "loss": 0.234,
      "num_input_tokens_seen": 7768416,
      "step": 36815
    },
    {
      "epoch": 4.050605060506051,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.029049628904707574,
      "loss": 0.2314,
      "num_input_tokens_seen": 7769440,
      "step": 36820
    },
    {
      "epoch": 4.051155115511551,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02904912441056193,
      "loss": 0.2283,
      "num_input_tokens_seen": 7770528,
      "step": 36825
    },
    {
      "epoch": 4.051705170517051,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02904861978693224,
      "loss": 0.2319,
      "num_input_tokens_seen": 7771552,
      "step": 36830
    },
    {
      "epoch": 4.052255225522552,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02904811503382317,
      "loss": 0.2252,
      "num_input_tokens_seen": 7772608,
      "step": 36835
    },
    {
      "epoch": 4.052805280528053,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.02904761015123936,
      "loss": 0.231,
      "num_input_tokens_seen": 7773664,
      "step": 36840
    },
    {
      "epoch": 4.053355335533554,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.02904710513918547,
      "loss": 0.2337,
      "num_input_tokens_seen": 7774720,
      "step": 36845
    },
    {
      "epoch": 4.053905390539054,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02904659999766615,
      "loss": 0.2348,
      "num_input_tokens_seen": 7775680,
      "step": 36850
    },
    {
      "epoch": 4.054455445544554,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.02904609472668606,
      "loss": 0.2274,
      "num_input_tokens_seen": 7776768,
      "step": 36855
    },
    {
      "epoch": 4.055005500550055,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.029045589326249854,
      "loss": 0.2341,
      "num_input_tokens_seen": 7777824,
      "step": 36860
    },
    {
      "epoch": 4.055555555555555,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029045083796362194,
      "loss": 0.2321,
      "num_input_tokens_seen": 7778880,
      "step": 36865
    },
    {
      "epoch": 4.0561056105610565,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.02904457813702773,
      "loss": 0.2331,
      "num_input_tokens_seen": 7779936,
      "step": 36870
    },
    {
      "epoch": 4.056655665566557,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.029044072348251133,
      "loss": 0.2321,
      "num_input_tokens_seen": 7781024,
      "step": 36875
    },
    {
      "epoch": 4.057205720572057,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.029043566430037058,
      "loss": 0.2321,
      "num_input_tokens_seen": 7782080,
      "step": 36880
    },
    {
      "epoch": 4.057755775577558,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02904306038239017,
      "loss": 0.2299,
      "num_input_tokens_seen": 7783232,
      "step": 36885
    },
    {
      "epoch": 4.058305830583058,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.029042554205315138,
      "loss": 0.2326,
      "num_input_tokens_seen": 7784256,
      "step": 36890
    },
    {
      "epoch": 4.058855885588559,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.029042047898816616,
      "loss": 0.232,
      "num_input_tokens_seen": 7785376,
      "step": 36895
    },
    {
      "epoch": 4.0594059405940595,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.029041541462899282,
      "loss": 0.2325,
      "num_input_tokens_seen": 7786432,
      "step": 36900
    },
    {
      "epoch": 4.05995599559956,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.029041034897567795,
      "loss": 0.2335,
      "num_input_tokens_seen": 7787424,
      "step": 36905
    },
    {
      "epoch": 4.060506050605061,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.029040528202826828,
      "loss": 0.2288,
      "num_input_tokens_seen": 7788448,
      "step": 36910
    },
    {
      "epoch": 4.061056105610561,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.029040021378681048,
      "loss": 0.2314,
      "num_input_tokens_seen": 7789472,
      "step": 36915
    },
    {
      "epoch": 4.061606160616061,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.029039514425135127,
      "loss": 0.233,
      "num_input_tokens_seen": 7790560,
      "step": 36920
    },
    {
      "epoch": 4.062156215621562,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02903900734219374,
      "loss": 0.2325,
      "num_input_tokens_seen": 7791616,
      "step": 36925
    },
    {
      "epoch": 4.0627062706270625,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02903850012986156,
      "loss": 0.232,
      "num_input_tokens_seen": 7792608,
      "step": 36930
    },
    {
      "epoch": 4.063256325632564,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.02903799278814326,
      "loss": 0.2324,
      "num_input_tokens_seen": 7793664,
      "step": 36935
    },
    {
      "epoch": 4.063806380638064,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.029037485317043518,
      "loss": 0.2298,
      "num_input_tokens_seen": 7794656,
      "step": 36940
    },
    {
      "epoch": 4.064356435643564,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.029036977716567006,
      "loss": 0.2278,
      "num_input_tokens_seen": 7795680,
      "step": 36945
    },
    {
      "epoch": 4.064906490649065,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.029036469986718407,
      "loss": 0.2314,
      "num_input_tokens_seen": 7796736,
      "step": 36950
    },
    {
      "epoch": 4.065456545654565,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.0290359621275024,
      "loss": 0.2309,
      "num_input_tokens_seen": 7797728,
      "step": 36955
    },
    {
      "epoch": 4.066006600660066,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.029035454138923666,
      "loss": 0.2315,
      "num_input_tokens_seen": 7798720,
      "step": 36960
    },
    {
      "epoch": 4.066556655665567,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.029034946020986885,
      "loss": 0.233,
      "num_input_tokens_seen": 7799744,
      "step": 36965
    },
    {
      "epoch": 4.067106710671067,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.029034437773696745,
      "loss": 0.2309,
      "num_input_tokens_seen": 7800832,
      "step": 36970
    },
    {
      "epoch": 4.067656765676568,
      "grad_norm": 0.009765625,
      "learning_rate": 0.029033929397057918,
      "loss": 0.2309,
      "num_input_tokens_seen": 7801856,
      "step": 36975
    },
    {
      "epoch": 4.068206820682068,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.029033420891075103,
      "loss": 0.2293,
      "num_input_tokens_seen": 7802912,
      "step": 36980
    },
    {
      "epoch": 4.068756875687569,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.029032912255752982,
      "loss": 0.2304,
      "num_input_tokens_seen": 7803872,
      "step": 36985
    },
    {
      "epoch": 4.069306930693069,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.029032403491096245,
      "loss": 0.232,
      "num_input_tokens_seen": 7804864,
      "step": 36990
    },
    {
      "epoch": 4.06985698569857,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.029031894597109576,
      "loss": 0.2331,
      "num_input_tokens_seen": 7805920,
      "step": 36995
    },
    {
      "epoch": 4.070407040704071,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02903138557379767,
      "loss": 0.2316,
      "num_input_tokens_seen": 7806944,
      "step": 37000
    },
    {
      "epoch": 4.070957095709571,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02903087642116521,
      "loss": 0.23,
      "num_input_tokens_seen": 7808000,
      "step": 37005
    },
    {
      "epoch": 4.071507150715071,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.029030367139216896,
      "loss": 0.2279,
      "num_input_tokens_seen": 7809056,
      "step": 37010
    },
    {
      "epoch": 4.072057205720572,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.029029857727957426,
      "loss": 0.226,
      "num_input_tokens_seen": 7810176,
      "step": 37015
    },
    {
      "epoch": 4.072607260726072,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.029029348187391484,
      "loss": 0.2293,
      "num_input_tokens_seen": 7811200,
      "step": 37020
    },
    {
      "epoch": 4.0731573157315735,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.029028838517523775,
      "loss": 0.2287,
      "num_input_tokens_seen": 7812224,
      "step": 37025
    },
    {
      "epoch": 4.073707370737074,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.029028328718358993,
      "loss": 0.2314,
      "num_input_tokens_seen": 7813312,
      "step": 37030
    },
    {
      "epoch": 4.074257425742574,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.029027818789901832,
      "loss": 0.2314,
      "num_input_tokens_seen": 7814368,
      "step": 37035
    },
    {
      "epoch": 4.074807480748075,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.029027308732157,
      "loss": 0.2257,
      "num_input_tokens_seen": 7815424,
      "step": 37040
    },
    {
      "epoch": 4.075357535753575,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.029026798545129195,
      "loss": 0.2315,
      "num_input_tokens_seen": 7816544,
      "step": 37045
    },
    {
      "epoch": 4.075907590759076,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.029026288228823114,
      "loss": 0.2357,
      "num_input_tokens_seen": 7817536,
      "step": 37050
    },
    {
      "epoch": 4.0764576457645765,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02902577778324347,
      "loss": 0.2289,
      "num_input_tokens_seen": 7818560,
      "step": 37055
    },
    {
      "epoch": 4.077007700770077,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.029025267208394965,
      "loss": 0.2247,
      "num_input_tokens_seen": 7819712,
      "step": 37060
    },
    {
      "epoch": 4.077557755775578,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.0290247565042823,
      "loss": 0.229,
      "num_input_tokens_seen": 7820800,
      "step": 37065
    },
    {
      "epoch": 4.078107810781078,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.029024245670910175,
      "loss": 0.2352,
      "num_input_tokens_seen": 7821856,
      "step": 37070
    },
    {
      "epoch": 4.078657865786579,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.02902373470828332,
      "loss": 0.239,
      "num_input_tokens_seen": 7822944,
      "step": 37075
    },
    {
      "epoch": 4.079207920792079,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.029023223616406423,
      "loss": 0.2351,
      "num_input_tokens_seen": 7824000,
      "step": 37080
    },
    {
      "epoch": 4.0797579757975795,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.029022712395284204,
      "loss": 0.2272,
      "num_input_tokens_seen": 7825024,
      "step": 37085
    },
    {
      "epoch": 4.080308030803081,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.029022201044921377,
      "loss": 0.2376,
      "num_input_tokens_seen": 7826048,
      "step": 37090
    },
    {
      "epoch": 4.080858085808581,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02902168956532265,
      "loss": 0.2338,
      "num_input_tokens_seen": 7827168,
      "step": 37095
    },
    {
      "epoch": 4.081408140814081,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.029021177956492734,
      "loss": 0.2364,
      "num_input_tokens_seen": 7828160,
      "step": 37100
    },
    {
      "epoch": 4.081958195819582,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.029020666218436356,
      "loss": 0.2331,
      "num_input_tokens_seen": 7829216,
      "step": 37105
    },
    {
      "epoch": 4.082508250825082,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.02902015435115822,
      "loss": 0.2289,
      "num_input_tokens_seen": 7830272,
      "step": 37110
    },
    {
      "epoch": 4.083058305830583,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.029019642354663054,
      "loss": 0.2335,
      "num_input_tokens_seen": 7831328,
      "step": 37115
    },
    {
      "epoch": 4.083608360836084,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.029019130228955567,
      "loss": 0.232,
      "num_input_tokens_seen": 7832448,
      "step": 37120
    },
    {
      "epoch": 4.084158415841584,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029018617974040487,
      "loss": 0.2356,
      "num_input_tokens_seen": 7833536,
      "step": 37125
    },
    {
      "epoch": 4.084708470847085,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02901810558992253,
      "loss": 0.2319,
      "num_input_tokens_seen": 7834560,
      "step": 37130
    },
    {
      "epoch": 4.085258525852585,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.02901759307660642,
      "loss": 0.2308,
      "num_input_tokens_seen": 7835584,
      "step": 37135
    },
    {
      "epoch": 4.085808580858086,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.029017080434096884,
      "loss": 0.2314,
      "num_input_tokens_seen": 7836640,
      "step": 37140
    },
    {
      "epoch": 4.086358635863586,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02901656766239864,
      "loss": 0.2308,
      "num_input_tokens_seen": 7837664,
      "step": 37145
    },
    {
      "epoch": 4.086908690869087,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.029016054761516415,
      "loss": 0.2319,
      "num_input_tokens_seen": 7838720,
      "step": 37150
    },
    {
      "epoch": 4.087458745874588,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.029015541731454944,
      "loss": 0.2324,
      "num_input_tokens_seen": 7839776,
      "step": 37155
    },
    {
      "epoch": 4.088008800880088,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02901502857221895,
      "loss": 0.2324,
      "num_input_tokens_seen": 7840832,
      "step": 37160
    },
    {
      "epoch": 4.088558855885589,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.029014515283813158,
      "loss": 0.2324,
      "num_input_tokens_seen": 7841952,
      "step": 37165
    },
    {
      "epoch": 4.089108910891089,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02901400186624231,
      "loss": 0.2324,
      "num_input_tokens_seen": 7842944,
      "step": 37170
    },
    {
      "epoch": 4.089658965896589,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02901348831951113,
      "loss": 0.2282,
      "num_input_tokens_seen": 7844000,
      "step": 37175
    },
    {
      "epoch": 4.0902090209020905,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02901297464362435,
      "loss": 0.2319,
      "num_input_tokens_seen": 7845120,
      "step": 37180
    },
    {
      "epoch": 4.090759075907591,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.029012460838586707,
      "loss": 0.2329,
      "num_input_tokens_seen": 7846208,
      "step": 37185
    },
    {
      "epoch": 4.091309130913091,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.029011946904402938,
      "loss": 0.2303,
      "num_input_tokens_seen": 7847264,
      "step": 37190
    },
    {
      "epoch": 4.091859185918592,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.029011432841077777,
      "loss": 0.2313,
      "num_input_tokens_seen": 7848320,
      "step": 37195
    },
    {
      "epoch": 4.092409240924092,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.029010918648615967,
      "loss": 0.2303,
      "num_input_tokens_seen": 7849376,
      "step": 37200
    },
    {
      "epoch": 4.092959295929593,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.029010404327022236,
      "loss": 0.2318,
      "num_input_tokens_seen": 7850464,
      "step": 37205
    },
    {
      "epoch": 4.0935093509350935,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.029009889876301334,
      "loss": 0.2319,
      "num_input_tokens_seen": 7851456,
      "step": 37210
    },
    {
      "epoch": 4.094059405940594,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.029009375296457997,
      "loss": 0.2298,
      "num_input_tokens_seen": 7852544,
      "step": 37215
    },
    {
      "epoch": 4.094609460946095,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.029008860587496972,
      "loss": 0.234,
      "num_input_tokens_seen": 7853600,
      "step": 37220
    },
    {
      "epoch": 4.095159515951595,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.029008345749423004,
      "loss": 0.2298,
      "num_input_tokens_seen": 7854656,
      "step": 37225
    },
    {
      "epoch": 4.095709570957096,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02900783078224083,
      "loss": 0.2324,
      "num_input_tokens_seen": 7855744,
      "step": 37230
    },
    {
      "epoch": 4.096259625962596,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.029007315685955205,
      "loss": 0.2314,
      "num_input_tokens_seen": 7856768,
      "step": 37235
    },
    {
      "epoch": 4.0968096809680965,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029006800460570876,
      "loss": 0.2324,
      "num_input_tokens_seen": 7857824,
      "step": 37240
    },
    {
      "epoch": 4.097359735973598,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.029006285106092582,
      "loss": 0.2308,
      "num_input_tokens_seen": 7858848,
      "step": 37245
    },
    {
      "epoch": 4.097909790979098,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02900576962252508,
      "loss": 0.2329,
      "num_input_tokens_seen": 7859968,
      "step": 37250
    },
    {
      "epoch": 4.098459845984599,
      "grad_norm": 0.0029449462890625,
      "learning_rate": 0.029005254009873124,
      "loss": 0.2303,
      "num_input_tokens_seen": 7861024,
      "step": 37255
    },
    {
      "epoch": 4.099009900990099,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.029004738268141456,
      "loss": 0.2319,
      "num_input_tokens_seen": 7862080,
      "step": 37260
    },
    {
      "epoch": 4.099559955995599,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.029004222397334842,
      "loss": 0.2288,
      "num_input_tokens_seen": 7863168,
      "step": 37265
    },
    {
      "epoch": 4.1001100110011,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.029003706397458022,
      "loss": 0.2308,
      "num_input_tokens_seen": 7864192,
      "step": 37270
    },
    {
      "epoch": 4.100660066006601,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029003190268515767,
      "loss": 0.2335,
      "num_input_tokens_seen": 7865280,
      "step": 37275
    },
    {
      "epoch": 4.101210121012101,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.029002674010512826,
      "loss": 0.2335,
      "num_input_tokens_seen": 7866304,
      "step": 37280
    },
    {
      "epoch": 4.101760176017602,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.02900215762345396,
      "loss": 0.2319,
      "num_input_tokens_seen": 7867360,
      "step": 37285
    },
    {
      "epoch": 4.102310231023102,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.029001641107343917,
      "loss": 0.2308,
      "num_input_tokens_seen": 7868384,
      "step": 37290
    },
    {
      "epoch": 4.102860286028603,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.029001124462187472,
      "loss": 0.2324,
      "num_input_tokens_seen": 7869440,
      "step": 37295
    },
    {
      "epoch": 4.103410341034103,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.029000607687989382,
      "loss": 0.2324,
      "num_input_tokens_seen": 7870432,
      "step": 37300
    },
    {
      "epoch": 4.103960396039604,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02900009078475441,
      "loss": 0.2319,
      "num_input_tokens_seen": 7871488,
      "step": 37305
    },
    {
      "epoch": 4.104510451045105,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.028999573752487316,
      "loss": 0.2308,
      "num_input_tokens_seen": 7872448,
      "step": 37310
    },
    {
      "epoch": 4.105060506050605,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.028999056591192868,
      "loss": 0.2318,
      "num_input_tokens_seen": 7873472,
      "step": 37315
    },
    {
      "epoch": 4.105610561056106,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.02899853930087583,
      "loss": 0.2308,
      "num_input_tokens_seen": 7874528,
      "step": 37320
    },
    {
      "epoch": 4.106160616061606,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02899802188154098,
      "loss": 0.2313,
      "num_input_tokens_seen": 7875552,
      "step": 37325
    },
    {
      "epoch": 4.106710671067106,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.028997504333193078,
      "loss": 0.2298,
      "num_input_tokens_seen": 7876608,
      "step": 37330
    },
    {
      "epoch": 4.1072607260726075,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028996986655836894,
      "loss": 0.2313,
      "num_input_tokens_seen": 7877600,
      "step": 37335
    },
    {
      "epoch": 4.107810781078108,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028996468849477198,
      "loss": 0.2313,
      "num_input_tokens_seen": 7878688,
      "step": 37340
    },
    {
      "epoch": 4.108360836083609,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.028995950914118766,
      "loss": 0.2313,
      "num_input_tokens_seen": 7879744,
      "step": 37345
    },
    {
      "epoch": 4.108910891089109,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.028995432849766373,
      "loss": 0.2324,
      "num_input_tokens_seen": 7880736,
      "step": 37350
    },
    {
      "epoch": 4.109460946094609,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.02899491465642479,
      "loss": 0.2293,
      "num_input_tokens_seen": 7881760,
      "step": 37355
    },
    {
      "epoch": 4.11001100110011,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02899439633409879,
      "loss": 0.2308,
      "num_input_tokens_seen": 7882816,
      "step": 37360
    },
    {
      "epoch": 4.1105610561056105,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.028993877882793156,
      "loss": 0.2298,
      "num_input_tokens_seen": 7883872,
      "step": 37365
    },
    {
      "epoch": 4.111111111111111,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028993359302512664,
      "loss": 0.2319,
      "num_input_tokens_seen": 7884896,
      "step": 37370
    },
    {
      "epoch": 4.111661166116612,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0289928405932621,
      "loss": 0.2309,
      "num_input_tokens_seen": 7885984,
      "step": 37375
    },
    {
      "epoch": 4.112211221122112,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028992321755046232,
      "loss": 0.2303,
      "num_input_tokens_seen": 7887040,
      "step": 37380
    },
    {
      "epoch": 4.112761276127613,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.02899180278786985,
      "loss": 0.2298,
      "num_input_tokens_seen": 7888096,
      "step": 37385
    },
    {
      "epoch": 4.113311331133113,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.02899128369173774,
      "loss": 0.2298,
      "num_input_tokens_seen": 7889152,
      "step": 37390
    },
    {
      "epoch": 4.1138613861386135,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.028990764466654675,
      "loss": 0.235,
      "num_input_tokens_seen": 7890208,
      "step": 37395
    },
    {
      "epoch": 4.114411441144115,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.028990245112625452,
      "loss": 0.2288,
      "num_input_tokens_seen": 7891232,
      "step": 37400
    },
    {
      "epoch": 4.114961496149615,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02898972562965485,
      "loss": 0.235,
      "num_input_tokens_seen": 7892352,
      "step": 37405
    },
    {
      "epoch": 4.115511551155116,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02898920601774766,
      "loss": 0.2314,
      "num_input_tokens_seen": 7893440,
      "step": 37410
    },
    {
      "epoch": 4.116061606160616,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.028988686276908673,
      "loss": 0.2314,
      "num_input_tokens_seen": 7894528,
      "step": 37415
    },
    {
      "epoch": 4.116611661166116,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.028988166407142676,
      "loss": 0.2308,
      "num_input_tokens_seen": 7895552,
      "step": 37420
    },
    {
      "epoch": 4.117161716171617,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02898764640845446,
      "loss": 0.2324,
      "num_input_tokens_seen": 7896512,
      "step": 37425
    },
    {
      "epoch": 4.117711771177118,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028987126280848822,
      "loss": 0.2282,
      "num_input_tokens_seen": 7897504,
      "step": 37430
    },
    {
      "epoch": 4.118261826182618,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.028986606024330553,
      "loss": 0.2308,
      "num_input_tokens_seen": 7898592,
      "step": 37435
    },
    {
      "epoch": 4.118811881188119,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.028986085638904445,
      "loss": 0.233,
      "num_input_tokens_seen": 7899680,
      "step": 37440
    },
    {
      "epoch": 4.119361936193619,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.0289855651245753,
      "loss": 0.2288,
      "num_input_tokens_seen": 7900768,
      "step": 37445
    },
    {
      "epoch": 4.11991199119912,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.028985044481347907,
      "loss": 0.2283,
      "num_input_tokens_seen": 7901856,
      "step": 37450
    },
    {
      "epoch": 4.12046204620462,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.028984523709227075,
      "loss": 0.2319,
      "num_input_tokens_seen": 7903008,
      "step": 37455
    },
    {
      "epoch": 4.121012101210121,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028984002808217595,
      "loss": 0.2314,
      "num_input_tokens_seen": 7904096,
      "step": 37460
    },
    {
      "epoch": 4.121562156215622,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.02898348177832427,
      "loss": 0.233,
      "num_input_tokens_seen": 7905088,
      "step": 37465
    },
    {
      "epoch": 4.122112211221122,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.02898296061955191,
      "loss": 0.2319,
      "num_input_tokens_seen": 7906048,
      "step": 37470
    },
    {
      "epoch": 4.122662266226623,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.028982439331905308,
      "loss": 0.2283,
      "num_input_tokens_seen": 7907072,
      "step": 37475
    },
    {
      "epoch": 4.123212321232123,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02898191791538927,
      "loss": 0.2314,
      "num_input_tokens_seen": 7908160,
      "step": 37480
    },
    {
      "epoch": 4.123762376237623,
      "grad_norm": 0.005859375,
      "learning_rate": 0.028981396370008603,
      "loss": 0.2304,
      "num_input_tokens_seen": 7909152,
      "step": 37485
    },
    {
      "epoch": 4.1243124312431245,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.028980874695768117,
      "loss": 0.2304,
      "num_input_tokens_seen": 7910144,
      "step": 37490
    },
    {
      "epoch": 4.124862486248625,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.028980352892672617,
      "loss": 0.2304,
      "num_input_tokens_seen": 7911200,
      "step": 37495
    },
    {
      "epoch": 4.125412541254126,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.028979830960726913,
      "loss": 0.2336,
      "num_input_tokens_seen": 7912224,
      "step": 37500
    },
    {
      "epoch": 4.125962596259626,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028979308899935816,
      "loss": 0.2289,
      "num_input_tokens_seen": 7913280,
      "step": 37505
    },
    {
      "epoch": 4.126512651265126,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.028978786710304138,
      "loss": 0.2288,
      "num_input_tokens_seen": 7914368,
      "step": 37510
    },
    {
      "epoch": 4.127062706270627,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.028978264391836688,
      "loss": 0.2367,
      "num_input_tokens_seen": 7915424,
      "step": 37515
    },
    {
      "epoch": 4.1276127612761275,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.028977741944538282,
      "loss": 0.232,
      "num_input_tokens_seen": 7916512,
      "step": 37520
    },
    {
      "epoch": 4.128162816281628,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.02897721936841374,
      "loss": 0.2335,
      "num_input_tokens_seen": 7917568,
      "step": 37525
    },
    {
      "epoch": 4.128712871287129,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.02897669666346787,
      "loss": 0.2304,
      "num_input_tokens_seen": 7918656,
      "step": 37530
    },
    {
      "epoch": 4.129262926292629,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.028976173829705496,
      "loss": 0.2298,
      "num_input_tokens_seen": 7919680,
      "step": 37535
    },
    {
      "epoch": 4.12981298129813,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.028975650867131428,
      "loss": 0.2299,
      "num_input_tokens_seen": 7920768,
      "step": 37540
    },
    {
      "epoch": 4.13036303630363,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.0289751277757505,
      "loss": 0.2314,
      "num_input_tokens_seen": 7921888,
      "step": 37545
    },
    {
      "epoch": 4.1309130913091305,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02897460455556752,
      "loss": 0.232,
      "num_input_tokens_seen": 7923008,
      "step": 37550
    },
    {
      "epoch": 4.131463146314632,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02897408120658732,
      "loss": 0.233,
      "num_input_tokens_seen": 7924064,
      "step": 37555
    },
    {
      "epoch": 4.132013201320132,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.028973557728814717,
      "loss": 0.2325,
      "num_input_tokens_seen": 7925184,
      "step": 37560
    },
    {
      "epoch": 4.132563256325633,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.028973034122254535,
      "loss": 0.2314,
      "num_input_tokens_seen": 7926240,
      "step": 37565
    },
    {
      "epoch": 4.133113311331133,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028972510386911607,
      "loss": 0.2293,
      "num_input_tokens_seen": 7927296,
      "step": 37570
    },
    {
      "epoch": 4.133663366336633,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02897198652279075,
      "loss": 0.2319,
      "num_input_tokens_seen": 7928320,
      "step": 37575
    },
    {
      "epoch": 4.134213421342134,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.028971462529896803,
      "loss": 0.2314,
      "num_input_tokens_seen": 7929344,
      "step": 37580
    },
    {
      "epoch": 4.134763476347635,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.028970938408234583,
      "loss": 0.2324,
      "num_input_tokens_seen": 7930336,
      "step": 37585
    },
    {
      "epoch": 4.135313531353136,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.028970414157808932,
      "loss": 0.2325,
      "num_input_tokens_seen": 7931424,
      "step": 37590
    },
    {
      "epoch": 4.135863586358636,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.028969889778624677,
      "loss": 0.2303,
      "num_input_tokens_seen": 7932448,
      "step": 37595
    },
    {
      "epoch": 4.136413641364136,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.028969365270686658,
      "loss": 0.2335,
      "num_input_tokens_seen": 7933440,
      "step": 37600
    },
    {
      "epoch": 4.136963696369637,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.028968840633999696,
      "loss": 0.2329,
      "num_input_tokens_seen": 7934432,
      "step": 37605
    },
    {
      "epoch": 4.137513751375137,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02896831586856863,
      "loss": 0.2314,
      "num_input_tokens_seen": 7935456,
      "step": 37610
    },
    {
      "epoch": 4.138063806380638,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.028967790974398305,
      "loss": 0.2288,
      "num_input_tokens_seen": 7936544,
      "step": 37615
    },
    {
      "epoch": 4.138613861386139,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.028967265951493546,
      "loss": 0.2342,
      "num_input_tokens_seen": 7937568,
      "step": 37620
    },
    {
      "epoch": 4.139163916391639,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.028966740799859204,
      "loss": 0.2322,
      "num_input_tokens_seen": 7938624,
      "step": 37625
    },
    {
      "epoch": 4.13971397139714,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.028966215519500113,
      "loss": 0.2291,
      "num_input_tokens_seen": 7939648,
      "step": 37630
    },
    {
      "epoch": 4.14026402640264,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.028965690110421115,
      "loss": 0.2385,
      "num_input_tokens_seen": 7940640,
      "step": 37635
    },
    {
      "epoch": 4.1408140814081404,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.028965164572627055,
      "loss": 0.2301,
      "num_input_tokens_seen": 7941696,
      "step": 37640
    },
    {
      "epoch": 4.1413641364136415,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.028964638906122775,
      "loss": 0.2281,
      "num_input_tokens_seen": 7942720,
      "step": 37645
    },
    {
      "epoch": 4.141914191419142,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.028964113110913115,
      "loss": 0.227,
      "num_input_tokens_seen": 7943744,
      "step": 37650
    },
    {
      "epoch": 4.142464246424643,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.028963587187002926,
      "loss": 0.2301,
      "num_input_tokens_seen": 7944800,
      "step": 37655
    },
    {
      "epoch": 4.143014301430143,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.028963061134397054,
      "loss": 0.2281,
      "num_input_tokens_seen": 7945856,
      "step": 37660
    },
    {
      "epoch": 4.143564356435643,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.02896253495310035,
      "loss": 0.2312,
      "num_input_tokens_seen": 7946944,
      "step": 37665
    },
    {
      "epoch": 4.144114411441144,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.028962008643117657,
      "loss": 0.2296,
      "num_input_tokens_seen": 7948000,
      "step": 37670
    },
    {
      "epoch": 4.1446644664466445,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.028961482204453833,
      "loss": 0.2229,
      "num_input_tokens_seen": 7949056,
      "step": 37675
    },
    {
      "epoch": 4.145214521452146,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02896095563711373,
      "loss": 0.2319,
      "num_input_tokens_seen": 7950080,
      "step": 37680
    },
    {
      "epoch": 4.145764576457646,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.028960428941102193,
      "loss": 0.2309,
      "num_input_tokens_seen": 7951136,
      "step": 37685
    },
    {
      "epoch": 4.146314631463146,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.02895990211642408,
      "loss": 0.2319,
      "num_input_tokens_seen": 7952192,
      "step": 37690
    },
    {
      "epoch": 4.146864686468647,
      "grad_norm": 0.0028533935546875,
      "learning_rate": 0.028959375163084253,
      "loss": 0.2298,
      "num_input_tokens_seen": 7953216,
      "step": 37695
    },
    {
      "epoch": 4.147414741474147,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.02895884808108756,
      "loss": 0.2428,
      "num_input_tokens_seen": 7954272,
      "step": 37700
    },
    {
      "epoch": 4.1479647964796476,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.028958320870438863,
      "loss": 0.2323,
      "num_input_tokens_seen": 7955328,
      "step": 37705
    },
    {
      "epoch": 4.148514851485149,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.02895779353114302,
      "loss": 0.2306,
      "num_input_tokens_seen": 7956352,
      "step": 37710
    },
    {
      "epoch": 4.149064906490649,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.028957266063204895,
      "loss": 0.2321,
      "num_input_tokens_seen": 7957408,
      "step": 37715
    },
    {
      "epoch": 4.14961496149615,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.028956738466629342,
      "loss": 0.2359,
      "num_input_tokens_seen": 7958432,
      "step": 37720
    },
    {
      "epoch": 4.15016501650165,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.028956210741421226,
      "loss": 0.2285,
      "num_input_tokens_seen": 7959488,
      "step": 37725
    },
    {
      "epoch": 4.15071507150715,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02895568288758542,
      "loss": 0.2336,
      "num_input_tokens_seen": 7960544,
      "step": 37730
    },
    {
      "epoch": 4.1512651265126514,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.028955154905126772,
      "loss": 0.2305,
      "num_input_tokens_seen": 7961600,
      "step": 37735
    },
    {
      "epoch": 4.151815181518152,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.028954626794050167,
      "loss": 0.2316,
      "num_input_tokens_seen": 7962656,
      "step": 37740
    },
    {
      "epoch": 4.152365236523653,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02895409855436046,
      "loss": 0.2341,
      "num_input_tokens_seen": 7963712,
      "step": 37745
    },
    {
      "epoch": 4.152915291529153,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02895357018606252,
      "loss": 0.2284,
      "num_input_tokens_seen": 7964864,
      "step": 37750
    },
    {
      "epoch": 4.153465346534653,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.028953041689161216,
      "loss": 0.2268,
      "num_input_tokens_seen": 7965952,
      "step": 37755
    },
    {
      "epoch": 4.154015401540154,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02895251306366143,
      "loss": 0.2294,
      "num_input_tokens_seen": 7966976,
      "step": 37760
    },
    {
      "epoch": 4.1545654565456545,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02895198430956802,
      "loss": 0.2299,
      "num_input_tokens_seen": 7968032,
      "step": 37765
    },
    {
      "epoch": 4.1551155115511555,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.028951455426885868,
      "loss": 0.2321,
      "num_input_tokens_seen": 7969088,
      "step": 37770
    },
    {
      "epoch": 4.155665566556656,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.028950926415619846,
      "loss": 0.2296,
      "num_input_tokens_seen": 7970048,
      "step": 37775
    },
    {
      "epoch": 4.156215621562156,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.028950397275774833,
      "loss": 0.2311,
      "num_input_tokens_seen": 7971168,
      "step": 37780
    },
    {
      "epoch": 4.156765676567657,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.028949868007355693,
      "loss": 0.2354,
      "num_input_tokens_seen": 7972224,
      "step": 37785
    },
    {
      "epoch": 4.157315731573157,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02894933861036732,
      "loss": 0.2353,
      "num_input_tokens_seen": 7973312,
      "step": 37790
    },
    {
      "epoch": 4.1578657865786575,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.02894880908481459,
      "loss": 0.2274,
      "num_input_tokens_seen": 7974368,
      "step": 37795
    },
    {
      "epoch": 4.158415841584159,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02894827943070237,
      "loss": 0.2306,
      "num_input_tokens_seen": 7975424,
      "step": 37800
    },
    {
      "epoch": 4.158965896589659,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.028947749648035558,
      "loss": 0.2322,
      "num_input_tokens_seen": 7976448,
      "step": 37805
    },
    {
      "epoch": 4.15951595159516,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.02894721973681903,
      "loss": 0.2312,
      "num_input_tokens_seen": 7977440,
      "step": 37810
    },
    {
      "epoch": 4.16006600660066,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.028946689697057666,
      "loss": 0.2263,
      "num_input_tokens_seen": 7978464,
      "step": 37815
    },
    {
      "epoch": 4.16061606160616,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02894615952875636,
      "loss": 0.2312,
      "num_input_tokens_seen": 7979488,
      "step": 37820
    },
    {
      "epoch": 4.161166116611661,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.028945629231919994,
      "loss": 0.2269,
      "num_input_tokens_seen": 7980544,
      "step": 37825
    },
    {
      "epoch": 4.161716171617162,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.028945098806553455,
      "loss": 0.2267,
      "num_input_tokens_seen": 7981600,
      "step": 37830
    },
    {
      "epoch": 4.162266226622663,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.028944568252661628,
      "loss": 0.2351,
      "num_input_tokens_seen": 7982656,
      "step": 37835
    },
    {
      "epoch": 4.162816281628163,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.02894403757024941,
      "loss": 0.233,
      "num_input_tokens_seen": 7983744,
      "step": 37840
    },
    {
      "epoch": 4.163366336633663,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.028943506759321686,
      "loss": 0.2413,
      "num_input_tokens_seen": 7984864,
      "step": 37845
    },
    {
      "epoch": 4.163916391639164,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02894297581988335,
      "loss": 0.2324,
      "num_input_tokens_seen": 7985952,
      "step": 37850
    },
    {
      "epoch": 4.164466446644664,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.028942444751939307,
      "loss": 0.2338,
      "num_input_tokens_seen": 7987040,
      "step": 37855
    },
    {
      "epoch": 4.165016501650165,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.028941913555494428,
      "loss": 0.2281,
      "num_input_tokens_seen": 7988032,
      "step": 37860
    },
    {
      "epoch": 4.165566556655666,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02894138223055363,
      "loss": 0.2255,
      "num_input_tokens_seen": 7989120,
      "step": 37865
    },
    {
      "epoch": 4.166116611661166,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.028940850777121795,
      "loss": 0.2297,
      "num_input_tokens_seen": 7990176,
      "step": 37870
    },
    {
      "epoch": 4.166666666666667,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.028940319195203835,
      "loss": 0.2303,
      "num_input_tokens_seen": 7991200,
      "step": 37875
    },
    {
      "epoch": 4.167216721672167,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.028939787484804635,
      "loss": 0.2366,
      "num_input_tokens_seen": 7992288,
      "step": 37880
    },
    {
      "epoch": 4.167766776677667,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.02893925564592911,
      "loss": 0.2369,
      "num_input_tokens_seen": 7993312,
      "step": 37885
    },
    {
      "epoch": 4.1683168316831685,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02893872367858215,
      "loss": 0.2306,
      "num_input_tokens_seen": 7994368,
      "step": 37890
    },
    {
      "epoch": 4.168866886688669,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.02893819158276866,
      "loss": 0.2306,
      "num_input_tokens_seen": 7995424,
      "step": 37895
    },
    {
      "epoch": 4.16941694169417,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028937659358493552,
      "loss": 0.2305,
      "num_input_tokens_seen": 7996480,
      "step": 37900
    },
    {
      "epoch": 4.16996699669967,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.028937127005761724,
      "loss": 0.2337,
      "num_input_tokens_seen": 7997568,
      "step": 37905
    },
    {
      "epoch": 4.17051705170517,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.028936594524578082,
      "loss": 0.2336,
      "num_input_tokens_seen": 7998656,
      "step": 37910
    },
    {
      "epoch": 4.171067106710671,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.028936061914947533,
      "loss": 0.2362,
      "num_input_tokens_seen": 7999776,
      "step": 37915
    },
    {
      "epoch": 4.1716171617161715,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.028935529176874995,
      "loss": 0.2304,
      "num_input_tokens_seen": 8000832,
      "step": 37920
    },
    {
      "epoch": 4.172167216721673,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02893499631036537,
      "loss": 0.2345,
      "num_input_tokens_seen": 8001824,
      "step": 37925
    },
    {
      "epoch": 4.172717271727173,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02893446331542357,
      "loss": 0.2303,
      "num_input_tokens_seen": 8002848,
      "step": 37930
    },
    {
      "epoch": 4.173267326732673,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.028933930192054504,
      "loss": 0.2298,
      "num_input_tokens_seen": 8003936,
      "step": 37935
    },
    {
      "epoch": 4.173817381738174,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.028933396940263097,
      "loss": 0.2303,
      "num_input_tokens_seen": 8004960,
      "step": 37940
    },
    {
      "epoch": 4.174367436743674,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02893286356005425,
      "loss": 0.2303,
      "num_input_tokens_seen": 8005952,
      "step": 37945
    },
    {
      "epoch": 4.174917491749175,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02893233005143289,
      "loss": 0.2304,
      "num_input_tokens_seen": 8007072,
      "step": 37950
    },
    {
      "epoch": 4.175467546754676,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.028931796414403924,
      "loss": 0.2298,
      "num_input_tokens_seen": 8008096,
      "step": 37955
    },
    {
      "epoch": 4.176017601760176,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.028931262648972278,
      "loss": 0.2335,
      "num_input_tokens_seen": 8009216,
      "step": 37960
    },
    {
      "epoch": 4.176567656765677,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.028930728755142868,
      "loss": 0.2335,
      "num_input_tokens_seen": 8010208,
      "step": 37965
    },
    {
      "epoch": 4.177117711771177,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.02893019473292062,
      "loss": 0.2319,
      "num_input_tokens_seen": 8011264,
      "step": 37970
    },
    {
      "epoch": 4.177667766776677,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.02892966058231045,
      "loss": 0.2309,
      "num_input_tokens_seen": 8012384,
      "step": 37975
    },
    {
      "epoch": 4.178217821782178,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.028929126303317277,
      "loss": 0.2283,
      "num_input_tokens_seen": 8013472,
      "step": 37980
    },
    {
      "epoch": 4.178767876787679,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.028928591895946036,
      "loss": 0.2288,
      "num_input_tokens_seen": 8014496,
      "step": 37985
    },
    {
      "epoch": 4.17931793179318,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.028928057360201646,
      "loss": 0.2303,
      "num_input_tokens_seen": 8015616,
      "step": 37990
    },
    {
      "epoch": 4.17986798679868,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.028927522696089034,
      "loss": 0.233,
      "num_input_tokens_seen": 8016736,
      "step": 37995
    },
    {
      "epoch": 4.18041804180418,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02892698790361313,
      "loss": 0.2319,
      "num_input_tokens_seen": 8017728,
      "step": 38000
    },
    {
      "epoch": 4.180968096809681,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.028926452982778857,
      "loss": 0.2319,
      "num_input_tokens_seen": 8018784,
      "step": 38005
    },
    {
      "epoch": 4.181518151815181,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.028925917933591155,
      "loss": 0.2329,
      "num_input_tokens_seen": 8019840,
      "step": 38010
    },
    {
      "epoch": 4.1820682068206825,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.028925382756054947,
      "loss": 0.2308,
      "num_input_tokens_seen": 8020896,
      "step": 38015
    },
    {
      "epoch": 4.182618261826183,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.028924847450175168,
      "loss": 0.2324,
      "num_input_tokens_seen": 8021952,
      "step": 38020
    },
    {
      "epoch": 4.183168316831683,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028924312015956754,
      "loss": 0.2314,
      "num_input_tokens_seen": 8023008,
      "step": 38025
    },
    {
      "epoch": 4.183718371837184,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.028923776453404635,
      "loss": 0.2319,
      "num_input_tokens_seen": 8024032,
      "step": 38030
    },
    {
      "epoch": 4.184268426842684,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02892324076252375,
      "loss": 0.2313,
      "num_input_tokens_seen": 8025056,
      "step": 38035
    },
    {
      "epoch": 4.184818481848184,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.028922704943319038,
      "loss": 0.2319,
      "num_input_tokens_seen": 8026080,
      "step": 38040
    },
    {
      "epoch": 4.1853685368536855,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.028922168995795434,
      "loss": 0.2309,
      "num_input_tokens_seen": 8027136,
      "step": 38045
    },
    {
      "epoch": 4.185918591859186,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.028921632919957876,
      "loss": 0.2309,
      "num_input_tokens_seen": 8028160,
      "step": 38050
    },
    {
      "epoch": 4.186468646864687,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02892109671581131,
      "loss": 0.2314,
      "num_input_tokens_seen": 8029216,
      "step": 38055
    },
    {
      "epoch": 4.187018701870187,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.028920560383360678,
      "loss": 0.2298,
      "num_input_tokens_seen": 8030208,
      "step": 38060
    },
    {
      "epoch": 4.187568756875687,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.028920023922610918,
      "loss": 0.2299,
      "num_input_tokens_seen": 8031232,
      "step": 38065
    },
    {
      "epoch": 4.188118811881188,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02891948733356698,
      "loss": 0.2351,
      "num_input_tokens_seen": 8032256,
      "step": 38070
    },
    {
      "epoch": 4.1886688668866885,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.028918950616233802,
      "loss": 0.2315,
      "num_input_tokens_seen": 8033344,
      "step": 38075
    },
    {
      "epoch": 4.18921892189219,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02891841377061634,
      "loss": 0.2268,
      "num_input_tokens_seen": 8034400,
      "step": 38080
    },
    {
      "epoch": 4.18976897689769,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.028917876796719533,
      "loss": 0.2304,
      "num_input_tokens_seen": 8035456,
      "step": 38085
    },
    {
      "epoch": 4.19031903190319,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.028917339694548336,
      "loss": 0.2284,
      "num_input_tokens_seen": 8036480,
      "step": 38090
    },
    {
      "epoch": 4.190869086908691,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.028916802464107695,
      "loss": 0.2326,
      "num_input_tokens_seen": 8037600,
      "step": 38095
    },
    {
      "epoch": 4.191419141914191,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.028916265105402565,
      "loss": 0.2295,
      "num_input_tokens_seen": 8038624,
      "step": 38100
    },
    {
      "epoch": 4.191969196919692,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.0289157276184379,
      "loss": 0.2332,
      "num_input_tokens_seen": 8039712,
      "step": 38105
    },
    {
      "epoch": 4.192519251925193,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.028915190003218644,
      "loss": 0.2331,
      "num_input_tokens_seen": 8040768,
      "step": 38110
    },
    {
      "epoch": 4.193069306930693,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.028914652259749762,
      "loss": 0.2315,
      "num_input_tokens_seen": 8041792,
      "step": 38115
    },
    {
      "epoch": 4.193619361936194,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.028914114388036212,
      "loss": 0.2278,
      "num_input_tokens_seen": 8042944,
      "step": 38120
    },
    {
      "epoch": 4.194169416941694,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02891357638808294,
      "loss": 0.2299,
      "num_input_tokens_seen": 8044000,
      "step": 38125
    },
    {
      "epoch": 4.194719471947194,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.028913038259894917,
      "loss": 0.2315,
      "num_input_tokens_seen": 8045088,
      "step": 38130
    },
    {
      "epoch": 4.195269526952695,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.02891250000347709,
      "loss": 0.232,
      "num_input_tokens_seen": 8046144,
      "step": 38135
    },
    {
      "epoch": 4.195819581958196,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.028911961618834428,
      "loss": 0.233,
      "num_input_tokens_seen": 8047168,
      "step": 38140
    },
    {
      "epoch": 4.196369636963697,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.028911423105971897,
      "loss": 0.232,
      "num_input_tokens_seen": 8048288,
      "step": 38145
    },
    {
      "epoch": 4.196919691969197,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.02891088446489445,
      "loss": 0.2325,
      "num_input_tokens_seen": 8049344,
      "step": 38150
    },
    {
      "epoch": 4.197469746974697,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02891034569560705,
      "loss": 0.2341,
      "num_input_tokens_seen": 8050336,
      "step": 38155
    },
    {
      "epoch": 4.198019801980198,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.028909806798114678,
      "loss": 0.232,
      "num_input_tokens_seen": 8051392,
      "step": 38160
    },
    {
      "epoch": 4.198569856985698,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.028909267772422288,
      "loss": 0.2324,
      "num_input_tokens_seen": 8052480,
      "step": 38165
    },
    {
      "epoch": 4.1991199119911995,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028908728618534853,
      "loss": 0.233,
      "num_input_tokens_seen": 8053536,
      "step": 38170
    },
    {
      "epoch": 4.1996699669967,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.02890818933645734,
      "loss": 0.2325,
      "num_input_tokens_seen": 8054624,
      "step": 38175
    },
    {
      "epoch": 4.2002200220022,
      "grad_norm": 0.00067138671875,
      "learning_rate": 0.028907649926194715,
      "loss": 0.2319,
      "num_input_tokens_seen": 8055680,
      "step": 38180
    },
    {
      "epoch": 4.200770077007701,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02890711038775196,
      "loss": 0.2324,
      "num_input_tokens_seen": 8056736,
      "step": 38185
    },
    {
      "epoch": 4.201320132013201,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.028906570721134037,
      "loss": 0.2324,
      "num_input_tokens_seen": 8057792,
      "step": 38190
    },
    {
      "epoch": 4.201870187018702,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.028906030926345926,
      "loss": 0.2324,
      "num_input_tokens_seen": 8058848,
      "step": 38195
    },
    {
      "epoch": 4.2024202420242025,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.028905491003392604,
      "loss": 0.2298,
      "num_input_tokens_seen": 8059840,
      "step": 38200
    },
    {
      "epoch": 4.202970297029703,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.028904950952279043,
      "loss": 0.2273,
      "num_input_tokens_seen": 8060896,
      "step": 38205
    },
    {
      "epoch": 4.203520352035204,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.02890441077301022,
      "loss": 0.2337,
      "num_input_tokens_seen": 8061920,
      "step": 38210
    },
    {
      "epoch": 4.204070407040704,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.028903870465591112,
      "loss": 0.2329,
      "num_input_tokens_seen": 8062944,
      "step": 38215
    },
    {
      "epoch": 4.204620462046204,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.028903330030026708,
      "loss": 0.2355,
      "num_input_tokens_seen": 8064000,
      "step": 38220
    },
    {
      "epoch": 4.205170517051705,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.028902789466321976,
      "loss": 0.2297,
      "num_input_tokens_seen": 8065088,
      "step": 38225
    },
    {
      "epoch": 4.2057205720572055,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.028902248774481908,
      "loss": 0.2354,
      "num_input_tokens_seen": 8066112,
      "step": 38230
    },
    {
      "epoch": 4.206270627062707,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.028901707954511482,
      "loss": 0.2321,
      "num_input_tokens_seen": 8067168,
      "step": 38235
    },
    {
      "epoch": 4.206820682068207,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.028901167006415686,
      "loss": 0.2248,
      "num_input_tokens_seen": 8068256,
      "step": 38240
    },
    {
      "epoch": 4.207370737073707,
      "grad_norm": 0.01409912109375,
      "learning_rate": 0.028900625930199503,
      "loss": 0.231,
      "num_input_tokens_seen": 8069312,
      "step": 38245
    },
    {
      "epoch": 4.207920792079208,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.028900084725867924,
      "loss": 0.2296,
      "num_input_tokens_seen": 8070368,
      "step": 38250
    },
    {
      "epoch": 4.208470847084708,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.028899543393425934,
      "loss": 0.2321,
      "num_input_tokens_seen": 8071424,
      "step": 38255
    },
    {
      "epoch": 4.209020902090209,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.02889900193287852,
      "loss": 0.228,
      "num_input_tokens_seen": 8072480,
      "step": 38260
    },
    {
      "epoch": 4.20957095709571,
      "grad_norm": 0.0074462890625,
      "learning_rate": 0.028898460344230676,
      "loss": 0.2272,
      "num_input_tokens_seen": 8073536,
      "step": 38265
    },
    {
      "epoch": 4.21012101210121,
      "grad_norm": 0.0089111328125,
      "learning_rate": 0.028897918627487394,
      "loss": 0.2397,
      "num_input_tokens_seen": 8074560,
      "step": 38270
    },
    {
      "epoch": 4.210671067106711,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.028897376782653656,
      "loss": 0.236,
      "num_input_tokens_seen": 8075616,
      "step": 38275
    },
    {
      "epoch": 4.211221122112211,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.028896834809734474,
      "loss": 0.2311,
      "num_input_tokens_seen": 8076704,
      "step": 38280
    },
    {
      "epoch": 4.211771177117711,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.028896292708734832,
      "loss": 0.2295,
      "num_input_tokens_seen": 8077792,
      "step": 38285
    },
    {
      "epoch": 4.212321232123212,
      "grad_norm": 0.0072021484375,
      "learning_rate": 0.028895750479659723,
      "loss": 0.2353,
      "num_input_tokens_seen": 8078848,
      "step": 38290
    },
    {
      "epoch": 4.212871287128713,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.028895208122514157,
      "loss": 0.2301,
      "num_input_tokens_seen": 8079904,
      "step": 38295
    },
    {
      "epoch": 4.213421342134214,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.02889466563730312,
      "loss": 0.2328,
      "num_input_tokens_seen": 8080960,
      "step": 38300
    },
    {
      "epoch": 4.213971397139714,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.028894123024031618,
      "loss": 0.2343,
      "num_input_tokens_seen": 8081984,
      "step": 38305
    },
    {
      "epoch": 4.214521452145214,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.028893580282704653,
      "loss": 0.2353,
      "num_input_tokens_seen": 8083072,
      "step": 38310
    },
    {
      "epoch": 4.215071507150715,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.028893037413327223,
      "loss": 0.2289,
      "num_input_tokens_seen": 8084160,
      "step": 38315
    },
    {
      "epoch": 4.215621562156215,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02889249441590434,
      "loss": 0.2294,
      "num_input_tokens_seen": 8085152,
      "step": 38320
    },
    {
      "epoch": 4.2161716171617165,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.028891951290440993,
      "loss": 0.2289,
      "num_input_tokens_seen": 8086176,
      "step": 38325
    },
    {
      "epoch": 4.216721672167217,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.028891408036942203,
      "loss": 0.2273,
      "num_input_tokens_seen": 8087264,
      "step": 38330
    },
    {
      "epoch": 4.217271727172717,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.02889086465541297,
      "loss": 0.2239,
      "num_input_tokens_seen": 8088352,
      "step": 38335
    },
    {
      "epoch": 4.217821782178218,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.0288903211458583,
      "loss": 0.2305,
      "num_input_tokens_seen": 8089408,
      "step": 38340
    },
    {
      "epoch": 4.218371837183718,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.028889777508283213,
      "loss": 0.234,
      "num_input_tokens_seen": 8090496,
      "step": 38345
    },
    {
      "epoch": 4.218921892189219,
      "grad_norm": 0.007537841796875,
      "learning_rate": 0.028889233742692703,
      "loss": 0.2318,
      "num_input_tokens_seen": 8091552,
      "step": 38350
    },
    {
      "epoch": 4.2194719471947195,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.028888689849091792,
      "loss": 0.2318,
      "num_input_tokens_seen": 8092576,
      "step": 38355
    },
    {
      "epoch": 4.22002200220022,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.028888145827485494,
      "loss": 0.2265,
      "num_input_tokens_seen": 8093600,
      "step": 38360
    },
    {
      "epoch": 4.220572057205721,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.028887601677878818,
      "loss": 0.2355,
      "num_input_tokens_seen": 8094656,
      "step": 38365
    },
    {
      "epoch": 4.221122112211221,
      "grad_norm": 0.0084228515625,
      "learning_rate": 0.02888705740027678,
      "loss": 0.2407,
      "num_input_tokens_seen": 8095712,
      "step": 38370
    },
    {
      "epoch": 4.221672167216722,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.0288865129946844,
      "loss": 0.2294,
      "num_input_tokens_seen": 8096768,
      "step": 38375
    },
    {
      "epoch": 4.222222222222222,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.028885968461106696,
      "loss": 0.234,
      "num_input_tokens_seen": 8097856,
      "step": 38380
    },
    {
      "epoch": 4.2227722772277225,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028885423799548678,
      "loss": 0.2291,
      "num_input_tokens_seen": 8098912,
      "step": 38385
    },
    {
      "epoch": 4.223322332233224,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.028884879010015373,
      "loss": 0.2264,
      "num_input_tokens_seen": 8100000,
      "step": 38390
    },
    {
      "epoch": 4.223872387238724,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.028884334092511805,
      "loss": 0.2311,
      "num_input_tokens_seen": 8101088,
      "step": 38395
    },
    {
      "epoch": 4.224422442244224,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.028883789047042988,
      "loss": 0.2265,
      "num_input_tokens_seen": 8102112,
      "step": 38400
    },
    {
      "epoch": 4.224972497249725,
      "grad_norm": 0.001953125,
      "learning_rate": 0.02888324387361395,
      "loss": 0.2307,
      "num_input_tokens_seen": 8103232,
      "step": 38405
    },
    {
      "epoch": 4.225522552255225,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.028882698572229713,
      "loss": 0.2333,
      "num_input_tokens_seen": 8104288,
      "step": 38410
    },
    {
      "epoch": 4.226072607260726,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.028882153142895307,
      "loss": 0.2311,
      "num_input_tokens_seen": 8105376,
      "step": 38415
    },
    {
      "epoch": 4.226622662266227,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.02888160758561576,
      "loss": 0.2313,
      "num_input_tokens_seen": 8106432,
      "step": 38420
    },
    {
      "epoch": 4.227172717271727,
      "grad_norm": 0.01171875,
      "learning_rate": 0.02888106190039609,
      "loss": 0.2275,
      "num_input_tokens_seen": 8107424,
      "step": 38425
    },
    {
      "epoch": 4.227722772277228,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02888051608724134,
      "loss": 0.2266,
      "num_input_tokens_seen": 8108480,
      "step": 38430
    },
    {
      "epoch": 4.228272827282728,
      "grad_norm": 0.00762939453125,
      "learning_rate": 0.028879970146156533,
      "loss": 0.2313,
      "num_input_tokens_seen": 8109536,
      "step": 38435
    },
    {
      "epoch": 4.228822882288229,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0288794240771467,
      "loss": 0.2289,
      "num_input_tokens_seen": 8110560,
      "step": 38440
    },
    {
      "epoch": 4.229372937293729,
      "grad_norm": 0.007568359375,
      "learning_rate": 0.028878877880216872,
      "loss": 0.2319,
      "num_input_tokens_seen": 8111584,
      "step": 38445
    },
    {
      "epoch": 4.22992299229923,
      "grad_norm": 0.00726318359375,
      "learning_rate": 0.02887833155537209,
      "loss": 0.2367,
      "num_input_tokens_seen": 8112640,
      "step": 38450
    },
    {
      "epoch": 4.230473047304731,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.028877785102617383,
      "loss": 0.2359,
      "num_input_tokens_seen": 8113696,
      "step": 38455
    },
    {
      "epoch": 4.231023102310231,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.028877238521957796,
      "loss": 0.2353,
      "num_input_tokens_seen": 8114784,
      "step": 38460
    },
    {
      "epoch": 4.231573157315731,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02887669181339835,
      "loss": 0.2335,
      "num_input_tokens_seen": 8115840,
      "step": 38465
    },
    {
      "epoch": 4.232123212321232,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.0288761449769441,
      "loss": 0.2319,
      "num_input_tokens_seen": 8116864,
      "step": 38470
    },
    {
      "epoch": 4.232673267326732,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.028875598012600083,
      "loss": 0.233,
      "num_input_tokens_seen": 8117856,
      "step": 38475
    },
    {
      "epoch": 4.2332233223322335,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.028875050920371334,
      "loss": 0.2293,
      "num_input_tokens_seen": 8118880,
      "step": 38480
    },
    {
      "epoch": 4.233773377337734,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.0288745037002629,
      "loss": 0.234,
      "num_input_tokens_seen": 8119904,
      "step": 38485
    },
    {
      "epoch": 4.234323432343234,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.028873956352279823,
      "loss": 0.2303,
      "num_input_tokens_seen": 8121024,
      "step": 38490
    },
    {
      "epoch": 4.234873487348735,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028873408876427146,
      "loss": 0.2329,
      "num_input_tokens_seen": 8122048,
      "step": 38495
    },
    {
      "epoch": 4.235423542354235,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.028872861272709917,
      "loss": 0.2308,
      "num_input_tokens_seen": 8123040,
      "step": 38500
    },
    {
      "epoch": 4.235973597359736,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.028872313541133184,
      "loss": 0.2298,
      "num_input_tokens_seen": 8124096,
      "step": 38505
    },
    {
      "epoch": 4.2365236523652365,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.02887176568170199,
      "loss": 0.2313,
      "num_input_tokens_seen": 8125120,
      "step": 38510
    },
    {
      "epoch": 4.237073707370737,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.028871217694421393,
      "loss": 0.2319,
      "num_input_tokens_seen": 8126240,
      "step": 38515
    },
    {
      "epoch": 4.237623762376238,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.028870669579296435,
      "loss": 0.2303,
      "num_input_tokens_seen": 8127328,
      "step": 38520
    },
    {
      "epoch": 4.238173817381738,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.028870121336332173,
      "loss": 0.2288,
      "num_input_tokens_seen": 8128416,
      "step": 38525
    },
    {
      "epoch": 4.238723872387239,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.028869572965533656,
      "loss": 0.2319,
      "num_input_tokens_seen": 8129408,
      "step": 38530
    },
    {
      "epoch": 4.239273927392739,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.028869024466905945,
      "loss": 0.2298,
      "num_input_tokens_seen": 8130432,
      "step": 38535
    },
    {
      "epoch": 4.2398239823982395,
      "grad_norm": 0.0007476806640625,
      "learning_rate": 0.028868475840454082,
      "loss": 0.2324,
      "num_input_tokens_seen": 8131456,
      "step": 38540
    },
    {
      "epoch": 4.240374037403741,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.02886792708618314,
      "loss": 0.234,
      "num_input_tokens_seen": 8132544,
      "step": 38545
    },
    {
      "epoch": 4.240924092409241,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.028867378204098168,
      "loss": 0.2319,
      "num_input_tokens_seen": 8133568,
      "step": 38550
    },
    {
      "epoch": 4.241474147414741,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.028866829194204222,
      "loss": 0.2314,
      "num_input_tokens_seen": 8134656,
      "step": 38555
    },
    {
      "epoch": 4.242024202420242,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.028866280056506367,
      "loss": 0.2319,
      "num_input_tokens_seen": 8135776,
      "step": 38560
    },
    {
      "epoch": 4.242574257425742,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02886573079100966,
      "loss": 0.2335,
      "num_input_tokens_seen": 8136800,
      "step": 38565
    },
    {
      "epoch": 4.243124312431243,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02886518139771917,
      "loss": 0.2298,
      "num_input_tokens_seen": 8137888,
      "step": 38570
    },
    {
      "epoch": 4.243674367436744,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.02886463187663995,
      "loss": 0.2345,
      "num_input_tokens_seen": 8138976,
      "step": 38575
    },
    {
      "epoch": 4.244224422442244,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028864082227777078,
      "loss": 0.233,
      "num_input_tokens_seen": 8140032,
      "step": 38580
    },
    {
      "epoch": 4.244774477447745,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.028863532451135605,
      "loss": 0.2309,
      "num_input_tokens_seen": 8141056,
      "step": 38585
    },
    {
      "epoch": 4.245324532453245,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.028862982546720612,
      "loss": 0.2298,
      "num_input_tokens_seen": 8142112,
      "step": 38590
    },
    {
      "epoch": 4.245874587458746,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02886243251453716,
      "loss": 0.2303,
      "num_input_tokens_seen": 8143136,
      "step": 38595
    },
    {
      "epoch": 4.2464246424642464,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.028861882354590317,
      "loss": 0.2324,
      "num_input_tokens_seen": 8144160,
      "step": 38600
    },
    {
      "epoch": 4.246974697469747,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02886133206688516,
      "loss": 0.2288,
      "num_input_tokens_seen": 8145248,
      "step": 38605
    },
    {
      "epoch": 4.247524752475248,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.02886078165142675,
      "loss": 0.2298,
      "num_input_tokens_seen": 8146304,
      "step": 38610
    },
    {
      "epoch": 4.248074807480748,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.028860231108220172,
      "loss": 0.233,
      "num_input_tokens_seen": 8147392,
      "step": 38615
    },
    {
      "epoch": 4.248624862486249,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.028859680437270493,
      "loss": 0.2309,
      "num_input_tokens_seen": 8148448,
      "step": 38620
    },
    {
      "epoch": 4.249174917491749,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02885912963858279,
      "loss": 0.2324,
      "num_input_tokens_seen": 8149568,
      "step": 38625
    },
    {
      "epoch": 4.2497249724972495,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.028858578712162145,
      "loss": 0.2319,
      "num_input_tokens_seen": 8150624,
      "step": 38630
    },
    {
      "epoch": 4.2502750275027505,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.02885802765801362,
      "loss": 0.2288,
      "num_input_tokens_seen": 8151712,
      "step": 38635
    },
    {
      "epoch": 4.250825082508251,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.028857476476142306,
      "loss": 0.2314,
      "num_input_tokens_seen": 8152832,
      "step": 38640
    },
    {
      "epoch": 4.251375137513751,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028856925166553287,
      "loss": 0.2309,
      "num_input_tokens_seen": 8153888,
      "step": 38645
    },
    {
      "epoch": 4.251925192519252,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.028856373729251632,
      "loss": 0.2314,
      "num_input_tokens_seen": 8154912,
      "step": 38650
    },
    {
      "epoch": 4.252475247524752,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.028855822164242432,
      "loss": 0.232,
      "num_input_tokens_seen": 8156032,
      "step": 38655
    },
    {
      "epoch": 4.253025302530253,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.028855270471530768,
      "loss": 0.233,
      "num_input_tokens_seen": 8157120,
      "step": 38660
    },
    {
      "epoch": 4.2535753575357536,
      "grad_norm": 0.0028076171875,
      "learning_rate": 0.028854718651121724,
      "loss": 0.2335,
      "num_input_tokens_seen": 8158208,
      "step": 38665
    },
    {
      "epoch": 4.254125412541254,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.028854166703020385,
      "loss": 0.2314,
      "num_input_tokens_seen": 8159328,
      "step": 38670
    },
    {
      "epoch": 4.254675467546755,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.028853614627231843,
      "loss": 0.2372,
      "num_input_tokens_seen": 8160352,
      "step": 38675
    },
    {
      "epoch": 4.255225522552255,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.028853062423761178,
      "loss": 0.2335,
      "num_input_tokens_seen": 8161440,
      "step": 38680
    },
    {
      "epoch": 4.255775577557756,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.028852510092613484,
      "loss": 0.2314,
      "num_input_tokens_seen": 8162496,
      "step": 38685
    },
    {
      "epoch": 4.256325632563256,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.028851957633793855,
      "loss": 0.2308,
      "num_input_tokens_seen": 8163552,
      "step": 38690
    },
    {
      "epoch": 4.256875687568757,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028851405047307377,
      "loss": 0.2324,
      "num_input_tokens_seen": 8164608,
      "step": 38695
    },
    {
      "epoch": 4.257425742574258,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028850852333159142,
      "loss": 0.2314,
      "num_input_tokens_seen": 8165664,
      "step": 38700
    },
    {
      "epoch": 4.257975797579758,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028850299491354255,
      "loss": 0.2303,
      "num_input_tokens_seen": 8166720,
      "step": 38705
    },
    {
      "epoch": 4.258525852585258,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0288497465218978,
      "loss": 0.2303,
      "num_input_tokens_seen": 8167712,
      "step": 38710
    },
    {
      "epoch": 4.259075907590759,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.028849193424794875,
      "loss": 0.2288,
      "num_input_tokens_seen": 8168768,
      "step": 38715
    },
    {
      "epoch": 4.259625962596259,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.028848640200050586,
      "loss": 0.2319,
      "num_input_tokens_seen": 8169824,
      "step": 38720
    },
    {
      "epoch": 4.2601760176017605,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.02884808684767002,
      "loss": 0.2309,
      "num_input_tokens_seen": 8170848,
      "step": 38725
    },
    {
      "epoch": 4.260726072607261,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028847533367658284,
      "loss": 0.2303,
      "num_input_tokens_seen": 8171968,
      "step": 38730
    },
    {
      "epoch": 4.261276127612761,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.028846979760020472,
      "loss": 0.2319,
      "num_input_tokens_seen": 8173024,
      "step": 38735
    },
    {
      "epoch": 4.261826182618262,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.0288464260247617,
      "loss": 0.2313,
      "num_input_tokens_seen": 8174048,
      "step": 38740
    },
    {
      "epoch": 4.262376237623762,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.02884587216188706,
      "loss": 0.2313,
      "num_input_tokens_seen": 8175072,
      "step": 38745
    },
    {
      "epoch": 4.262926292629263,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.028845318171401665,
      "loss": 0.2313,
      "num_input_tokens_seen": 8176096,
      "step": 38750
    },
    {
      "epoch": 4.2634763476347635,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.02884476405331061,
      "loss": 0.2288,
      "num_input_tokens_seen": 8177056,
      "step": 38755
    },
    {
      "epoch": 4.264026402640264,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.028844209807619012,
      "loss": 0.2308,
      "num_input_tokens_seen": 8178112,
      "step": 38760
    },
    {
      "epoch": 4.264576457645765,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02884365543433197,
      "loss": 0.2298,
      "num_input_tokens_seen": 8179168,
      "step": 38765
    },
    {
      "epoch": 4.265126512651265,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.028843100933454602,
      "loss": 0.2329,
      "num_input_tokens_seen": 8180192,
      "step": 38770
    },
    {
      "epoch": 4.265676567656766,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.028842546304992018,
      "loss": 0.2313,
      "num_input_tokens_seen": 8181280,
      "step": 38775
    },
    {
      "epoch": 4.266226622662266,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.02884199154894932,
      "loss": 0.2303,
      "num_input_tokens_seen": 8182336,
      "step": 38780
    },
    {
      "epoch": 4.2667766776677665,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028841436665331635,
      "loss": 0.2324,
      "num_input_tokens_seen": 8183360,
      "step": 38785
    },
    {
      "epoch": 4.267326732673268,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028840881654144064,
      "loss": 0.2319,
      "num_input_tokens_seen": 8184384,
      "step": 38790
    },
    {
      "epoch": 4.267876787678768,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.028840326515391736,
      "loss": 0.2335,
      "num_input_tokens_seen": 8185472,
      "step": 38795
    },
    {
      "epoch": 4.268426842684269,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.028839771249079754,
      "loss": 0.233,
      "num_input_tokens_seen": 8186528,
      "step": 38800
    },
    {
      "epoch": 4.268976897689769,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.028839215855213245,
      "loss": 0.2319,
      "num_input_tokens_seen": 8187616,
      "step": 38805
    },
    {
      "epoch": 4.269526952695269,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.028838660333797318,
      "loss": 0.2325,
      "num_input_tokens_seen": 8188736,
      "step": 38810
    },
    {
      "epoch": 4.27007700770077,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.028838104684837107,
      "loss": 0.2324,
      "num_input_tokens_seen": 8189792,
      "step": 38815
    },
    {
      "epoch": 4.270627062706271,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.028837548908337715,
      "loss": 0.2299,
      "num_input_tokens_seen": 8190848,
      "step": 38820
    },
    {
      "epoch": 4.271177117711771,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.028836993004304284,
      "loss": 0.233,
      "num_input_tokens_seen": 8191840,
      "step": 38825
    },
    {
      "epoch": 4.271727172717272,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.028836436972741925,
      "loss": 0.233,
      "num_input_tokens_seen": 8192864,
      "step": 38830
    },
    {
      "epoch": 4.272277227722772,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.028835880813655766,
      "loss": 0.2325,
      "num_input_tokens_seen": 8193920,
      "step": 38835
    },
    {
      "epoch": 4.272827282728273,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028835324527050933,
      "loss": 0.2314,
      "num_input_tokens_seen": 8194944,
      "step": 38840
    },
    {
      "epoch": 4.273377337733773,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.028834768112932552,
      "loss": 0.2293,
      "num_input_tokens_seen": 8196000,
      "step": 38845
    },
    {
      "epoch": 4.273927392739274,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.028834211571305748,
      "loss": 0.2304,
      "num_input_tokens_seen": 8197120,
      "step": 38850
    },
    {
      "epoch": 4.274477447744775,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02883365490217566,
      "loss": 0.2299,
      "num_input_tokens_seen": 8198144,
      "step": 38855
    },
    {
      "epoch": 4.275027502750275,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.028833098105547413,
      "loss": 0.2336,
      "num_input_tokens_seen": 8199232,
      "step": 38860
    },
    {
      "epoch": 4.275577557755776,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.028832541181426134,
      "loss": 0.2315,
      "num_input_tokens_seen": 8200352,
      "step": 38865
    },
    {
      "epoch": 4.276127612761276,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.028831984129816968,
      "loss": 0.2325,
      "num_input_tokens_seen": 8201376,
      "step": 38870
    },
    {
      "epoch": 4.276677667766776,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.028831426950725034,
      "loss": 0.2309,
      "num_input_tokens_seen": 8202368,
      "step": 38875
    },
    {
      "epoch": 4.2772277227722775,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.028830869644155474,
      "loss": 0.234,
      "num_input_tokens_seen": 8203424,
      "step": 38880
    },
    {
      "epoch": 4.277777777777778,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02883031221011343,
      "loss": 0.232,
      "num_input_tokens_seen": 8204480,
      "step": 38885
    },
    {
      "epoch": 4.278327832783278,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.028829754648604034,
      "loss": 0.2278,
      "num_input_tokens_seen": 8205472,
      "step": 38890
    },
    {
      "epoch": 4.278877887788779,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028829196959632424,
      "loss": 0.2367,
      "num_input_tokens_seen": 8206496,
      "step": 38895
    },
    {
      "epoch": 4.279427942794279,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.028828639143203744,
      "loss": 0.2278,
      "num_input_tokens_seen": 8207616,
      "step": 38900
    },
    {
      "epoch": 4.27997799779978,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.02882808119932313,
      "loss": 0.234,
      "num_input_tokens_seen": 8208640,
      "step": 38905
    },
    {
      "epoch": 4.2805280528052805,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02882752312799573,
      "loss": 0.2299,
      "num_input_tokens_seen": 8209664,
      "step": 38910
    },
    {
      "epoch": 4.281078107810781,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02882696492922668,
      "loss": 0.2309,
      "num_input_tokens_seen": 8210688,
      "step": 38915
    },
    {
      "epoch": 4.281628162816282,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.028826406603021135,
      "loss": 0.2325,
      "num_input_tokens_seen": 8211744,
      "step": 38920
    },
    {
      "epoch": 4.282178217821782,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02882584814938423,
      "loss": 0.233,
      "num_input_tokens_seen": 8212800,
      "step": 38925
    },
    {
      "epoch": 4.282728272827283,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.028825289568321116,
      "loss": 0.2293,
      "num_input_tokens_seen": 8213888,
      "step": 38930
    },
    {
      "epoch": 4.283278327832783,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.028824730859836944,
      "loss": 0.2346,
      "num_input_tokens_seen": 8214912,
      "step": 38935
    },
    {
      "epoch": 4.2838283828382835,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.028824172023936865,
      "loss": 0.234,
      "num_input_tokens_seen": 8215936,
      "step": 38940
    },
    {
      "epoch": 4.284378437843785,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02882361306062602,
      "loss": 0.2319,
      "num_input_tokens_seen": 8217024,
      "step": 38945
    },
    {
      "epoch": 4.284928492849285,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02882305396990957,
      "loss": 0.233,
      "num_input_tokens_seen": 8218144,
      "step": 38950
    },
    {
      "epoch": 4.285478547854786,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.028822494751792663,
      "loss": 0.233,
      "num_input_tokens_seen": 8219168,
      "step": 38955
    },
    {
      "epoch": 4.286028602860286,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.028821935406280457,
      "loss": 0.2298,
      "num_input_tokens_seen": 8220192,
      "step": 38960
    },
    {
      "epoch": 4.286578657865786,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.028821375933378104,
      "loss": 0.2309,
      "num_input_tokens_seen": 8221312,
      "step": 38965
    },
    {
      "epoch": 4.287128712871287,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.02882081633309076,
      "loss": 0.2283,
      "num_input_tokens_seen": 8222400,
      "step": 38970
    },
    {
      "epoch": 4.287678767876788,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02882025660542358,
      "loss": 0.2299,
      "num_input_tokens_seen": 8223552,
      "step": 38975
    },
    {
      "epoch": 4.288228822882289,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02881969675038173,
      "loss": 0.233,
      "num_input_tokens_seen": 8224640,
      "step": 38980
    },
    {
      "epoch": 4.288778877887789,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028819136767970367,
      "loss": 0.2299,
      "num_input_tokens_seen": 8225728,
      "step": 38985
    },
    {
      "epoch": 4.289328932893289,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02881857665819465,
      "loss": 0.2299,
      "num_input_tokens_seen": 8226752,
      "step": 38990
    },
    {
      "epoch": 4.28987898789879,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.028818016421059748,
      "loss": 0.233,
      "num_input_tokens_seen": 8227840,
      "step": 38995
    },
    {
      "epoch": 4.29042904290429,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.028817456056570812,
      "loss": 0.2315,
      "num_input_tokens_seen": 8228928,
      "step": 39000
    },
    {
      "epoch": 4.290979097909791,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02881689556473302,
      "loss": 0.2299,
      "num_input_tokens_seen": 8229984,
      "step": 39005
    },
    {
      "epoch": 4.291529152915292,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.02881633494555153,
      "loss": 0.2325,
      "num_input_tokens_seen": 8231072,
      "step": 39010
    },
    {
      "epoch": 4.292079207920792,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.028815774199031505,
      "loss": 0.2325,
      "num_input_tokens_seen": 8232096,
      "step": 39015
    },
    {
      "epoch": 4.292629262926293,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028815213325178127,
      "loss": 0.2351,
      "num_input_tokens_seen": 8233088,
      "step": 39020
    },
    {
      "epoch": 4.293179317931793,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.028814652323996548,
      "loss": 0.233,
      "num_input_tokens_seen": 8234208,
      "step": 39025
    },
    {
      "epoch": 4.293729372937293,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.028814091195491956,
      "loss": 0.2299,
      "num_input_tokens_seen": 8235264,
      "step": 39030
    },
    {
      "epoch": 4.2942794279427945,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02881352993966951,
      "loss": 0.232,
      "num_input_tokens_seen": 8236352,
      "step": 39035
    },
    {
      "epoch": 4.294829482948295,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028812968556534386,
      "loss": 0.2314,
      "num_input_tokens_seen": 8237408,
      "step": 39040
    },
    {
      "epoch": 4.295379537953796,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02881240704609176,
      "loss": 0.2325,
      "num_input_tokens_seen": 8238496,
      "step": 39045
    },
    {
      "epoch": 4.295929592959296,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.028811845408346803,
      "loss": 0.2304,
      "num_input_tokens_seen": 8239616,
      "step": 39050
    },
    {
      "epoch": 4.296479647964796,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.0288112836433047,
      "loss": 0.2351,
      "num_input_tokens_seen": 8240736,
      "step": 39055
    },
    {
      "epoch": 4.297029702970297,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.02881072175097062,
      "loss": 0.2283,
      "num_input_tokens_seen": 8241824,
      "step": 39060
    },
    {
      "epoch": 4.2975797579757975,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.028810159731349743,
      "loss": 0.2325,
      "num_input_tokens_seen": 8242848,
      "step": 39065
    },
    {
      "epoch": 4.298129812981298,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.028809597584447257,
      "loss": 0.2325,
      "num_input_tokens_seen": 8243872,
      "step": 39070
    },
    {
      "epoch": 4.298679867986799,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.028809035310268332,
      "loss": 0.2325,
      "num_input_tokens_seen": 8244896,
      "step": 39075
    },
    {
      "epoch": 4.299229922992299,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.028808472908818154,
      "loss": 0.2309,
      "num_input_tokens_seen": 8245952,
      "step": 39080
    },
    {
      "epoch": 4.2997799779978,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.028807910380101907,
      "loss": 0.2304,
      "num_input_tokens_seen": 8246976,
      "step": 39085
    },
    {
      "epoch": 4.3003300330033,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.028807347724124778,
      "loss": 0.2304,
      "num_input_tokens_seen": 8247936,
      "step": 39090
    },
    {
      "epoch": 4.3008800880088005,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02880678494089195,
      "loss": 0.2293,
      "num_input_tokens_seen": 8248992,
      "step": 39095
    },
    {
      "epoch": 4.301430143014302,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02880622203040861,
      "loss": 0.2304,
      "num_input_tokens_seen": 8250016,
      "step": 39100
    },
    {
      "epoch": 4.301980198019802,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.028805658992679943,
      "loss": 0.232,
      "num_input_tokens_seen": 8251136,
      "step": 39105
    },
    {
      "epoch": 4.302530253025303,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028805095827711145,
      "loss": 0.2319,
      "num_input_tokens_seen": 8252192,
      "step": 39110
    },
    {
      "epoch": 4.303080308030803,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.028804532535507403,
      "loss": 0.2278,
      "num_input_tokens_seen": 8253216,
      "step": 39115
    },
    {
      "epoch": 4.303630363036303,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.028803969116073906,
      "loss": 0.232,
      "num_input_tokens_seen": 8254240,
      "step": 39120
    },
    {
      "epoch": 4.304180418041804,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.028803405569415855,
      "loss": 0.234,
      "num_input_tokens_seen": 8255328,
      "step": 39125
    },
    {
      "epoch": 4.304730473047305,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.028802841895538434,
      "loss": 0.2319,
      "num_input_tokens_seen": 8256352,
      "step": 39130
    },
    {
      "epoch": 4.305280528052805,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.028802278094446847,
      "loss": 0.2346,
      "num_input_tokens_seen": 8257472,
      "step": 39135
    },
    {
      "epoch": 4.305830583058306,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.028801714166146283,
      "loss": 0.2309,
      "num_input_tokens_seen": 8258496,
      "step": 39140
    },
    {
      "epoch": 4.306380638063806,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.02880115011064194,
      "loss": 0.2294,
      "num_input_tokens_seen": 8259520,
      "step": 39145
    },
    {
      "epoch": 4.306930693069307,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.028800585927939017,
      "loss": 0.2304,
      "num_input_tokens_seen": 8260576,
      "step": 39150
    },
    {
      "epoch": 4.307480748074807,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.028800021618042723,
      "loss": 0.2314,
      "num_input_tokens_seen": 8261600,
      "step": 39155
    },
    {
      "epoch": 4.3080308030803085,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.02879945718095825,
      "loss": 0.2283,
      "num_input_tokens_seen": 8262624,
      "step": 39160
    },
    {
      "epoch": 4.308580858085809,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0287988926166908,
      "loss": 0.234,
      "num_input_tokens_seen": 8263648,
      "step": 39165
    },
    {
      "epoch": 4.309130913091309,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.028798327925245577,
      "loss": 0.2304,
      "num_input_tokens_seen": 8264768,
      "step": 39170
    },
    {
      "epoch": 4.30968096809681,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.028797763106627787,
      "loss": 0.2294,
      "num_input_tokens_seen": 8265888,
      "step": 39175
    },
    {
      "epoch": 4.31023102310231,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.028797198160842637,
      "loss": 0.234,
      "num_input_tokens_seen": 8266976,
      "step": 39180
    },
    {
      "epoch": 4.31078107810781,
      "grad_norm": 0.0006256103515625,
      "learning_rate": 0.02879663308789533,
      "loss": 0.2315,
      "num_input_tokens_seen": 8268064,
      "step": 39185
    },
    {
      "epoch": 4.3113311331133115,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.02879606788779108,
      "loss": 0.2335,
      "num_input_tokens_seen": 8269152,
      "step": 39190
    },
    {
      "epoch": 4.311881188118812,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.028795502560535084,
      "loss": 0.2335,
      "num_input_tokens_seen": 8270240,
      "step": 39195
    },
    {
      "epoch": 4.312431243124313,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.028794937106132568,
      "loss": 0.2325,
      "num_input_tokens_seen": 8271360,
      "step": 39200
    },
    {
      "epoch": 4.312981298129813,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.028794371524588733,
      "loss": 0.2299,
      "num_input_tokens_seen": 8272480,
      "step": 39205
    },
    {
      "epoch": 4.313531353135313,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.028793805815908798,
      "loss": 0.2319,
      "num_input_tokens_seen": 8273536,
      "step": 39210
    },
    {
      "epoch": 4.314081408140814,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.028793239980097973,
      "loss": 0.2304,
      "num_input_tokens_seen": 8274592,
      "step": 39215
    },
    {
      "epoch": 4.3146314631463145,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02879267401716147,
      "loss": 0.2309,
      "num_input_tokens_seen": 8275680,
      "step": 39220
    },
    {
      "epoch": 4.315181518151816,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.028792107927104512,
      "loss": 0.233,
      "num_input_tokens_seen": 8276736,
      "step": 39225
    },
    {
      "epoch": 4.315731573157316,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.02879154170993231,
      "loss": 0.2319,
      "num_input_tokens_seen": 8277760,
      "step": 39230
    },
    {
      "epoch": 4.316281628162816,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.02879097536565009,
      "loss": 0.233,
      "num_input_tokens_seen": 8278848,
      "step": 39235
    },
    {
      "epoch": 4.316831683168317,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.028790408894263066,
      "loss": 0.2319,
      "num_input_tokens_seen": 8279904,
      "step": 39240
    },
    {
      "epoch": 4.317381738173817,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02878984229577646,
      "loss": 0.2314,
      "num_input_tokens_seen": 8280928,
      "step": 39245
    },
    {
      "epoch": 4.3179317931793175,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.028789275570195495,
      "loss": 0.2298,
      "num_input_tokens_seen": 8282016,
      "step": 39250
    },
    {
      "epoch": 4.318481848184819,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028788708717525396,
      "loss": 0.2298,
      "num_input_tokens_seen": 8283104,
      "step": 39255
    },
    {
      "epoch": 4.319031903190319,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.028788141737771382,
      "loss": 0.2303,
      "num_input_tokens_seen": 8284128,
      "step": 39260
    },
    {
      "epoch": 4.31958195819582,
      "grad_norm": 0.000797271728515625,
      "learning_rate": 0.028787574630938687,
      "loss": 0.2308,
      "num_input_tokens_seen": 8285184,
      "step": 39265
    },
    {
      "epoch": 4.32013201320132,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.028787007397032525,
      "loss": 0.2314,
      "num_input_tokens_seen": 8286272,
      "step": 39270
    },
    {
      "epoch": 4.32068206820682,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.028786440036058136,
      "loss": 0.2298,
      "num_input_tokens_seen": 8287360,
      "step": 39275
    },
    {
      "epoch": 4.321232123212321,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.028785872548020744,
      "loss": 0.2319,
      "num_input_tokens_seen": 8288416,
      "step": 39280
    },
    {
      "epoch": 4.321782178217822,
      "grad_norm": 0.00045013427734375,
      "learning_rate": 0.028785304932925575,
      "loss": 0.2319,
      "num_input_tokens_seen": 8289472,
      "step": 39285
    },
    {
      "epoch": 4.322332233223323,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02878473719077787,
      "loss": 0.2351,
      "num_input_tokens_seen": 8290528,
      "step": 39290
    },
    {
      "epoch": 4.322882288228823,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.028784169321582856,
      "loss": 0.2319,
      "num_input_tokens_seen": 8291584,
      "step": 39295
    },
    {
      "epoch": 4.323432343234323,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02878360132534577,
      "loss": 0.2314,
      "num_input_tokens_seen": 8292672,
      "step": 39300
    },
    {
      "epoch": 4.323982398239824,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.028783033202071845,
      "loss": 0.2324,
      "num_input_tokens_seen": 8293728,
      "step": 39305
    },
    {
      "epoch": 4.324532453245324,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028782464951766316,
      "loss": 0.2324,
      "num_input_tokens_seen": 8294784,
      "step": 39310
    },
    {
      "epoch": 4.325082508250825,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.028781896574434414,
      "loss": 0.2314,
      "num_input_tokens_seen": 8295840,
      "step": 39315
    },
    {
      "epoch": 4.325632563256326,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.028781328070081394,
      "loss": 0.2335,
      "num_input_tokens_seen": 8296960,
      "step": 39320
    },
    {
      "epoch": 4.326182618261826,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.02878075943871248,
      "loss": 0.2298,
      "num_input_tokens_seen": 8298016,
      "step": 39325
    },
    {
      "epoch": 4.326732673267327,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.028780190680332923,
      "loss": 0.2335,
      "num_input_tokens_seen": 8299040,
      "step": 39330
    },
    {
      "epoch": 4.327282728272827,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.028779621794947956,
      "loss": 0.2319,
      "num_input_tokens_seen": 8300128,
      "step": 39335
    },
    {
      "epoch": 4.327832783278327,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02877905278256283,
      "loss": 0.2273,
      "num_input_tokens_seen": 8301184,
      "step": 39340
    },
    {
      "epoch": 4.3283828382838285,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02877848364318279,
      "loss": 0.233,
      "num_input_tokens_seen": 8302272,
      "step": 39345
    },
    {
      "epoch": 4.328932893289329,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.028777914376813072,
      "loss": 0.234,
      "num_input_tokens_seen": 8303360,
      "step": 39350
    },
    {
      "epoch": 4.32948294829483,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02877734498345893,
      "loss": 0.2329,
      "num_input_tokens_seen": 8304416,
      "step": 39355
    },
    {
      "epoch": 4.33003300330033,
      "grad_norm": 0.000640869140625,
      "learning_rate": 0.028776775463125612,
      "loss": 0.2325,
      "num_input_tokens_seen": 8305408,
      "step": 39360
    },
    {
      "epoch": 4.33058305830583,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.028776205815818363,
      "loss": 0.2309,
      "num_input_tokens_seen": 8306464,
      "step": 39365
    },
    {
      "epoch": 4.331133113311331,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.028775636041542436,
      "loss": 0.2314,
      "num_input_tokens_seen": 8307552,
      "step": 39370
    },
    {
      "epoch": 4.3316831683168315,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.02877506614030308,
      "loss": 0.2314,
      "num_input_tokens_seen": 8308640,
      "step": 39375
    },
    {
      "epoch": 4.332233223322333,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.028774496112105554,
      "loss": 0.2309,
      "num_input_tokens_seen": 8309728,
      "step": 39380
    },
    {
      "epoch": 4.332783278327833,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.028773925956955104,
      "loss": 0.2298,
      "num_input_tokens_seen": 8310784,
      "step": 39385
    },
    {
      "epoch": 4.333333333333333,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.028773355674856987,
      "loss": 0.2298,
      "num_input_tokens_seen": 8311872,
      "step": 39390
    },
    {
      "epoch": 4.333883388338834,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.028772785265816465,
      "loss": 0.2319,
      "num_input_tokens_seen": 8312896,
      "step": 39395
    },
    {
      "epoch": 4.334433443344334,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.028772214729838784,
      "loss": 0.2314,
      "num_input_tokens_seen": 8313920,
      "step": 39400
    },
    {
      "epoch": 4.334983498349835,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.02877164406692921,
      "loss": 0.2324,
      "num_input_tokens_seen": 8314976,
      "step": 39405
    },
    {
      "epoch": 4.335533553355336,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.028771073277093006,
      "loss": 0.2329,
      "num_input_tokens_seen": 8316032,
      "step": 39410
    },
    {
      "epoch": 4.336083608360836,
      "grad_norm": 0.000774383544921875,
      "learning_rate": 0.02877050236033542,
      "loss": 0.2329,
      "num_input_tokens_seen": 8317056,
      "step": 39415
    },
    {
      "epoch": 4.336633663366337,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.028769931316661723,
      "loss": 0.2314,
      "num_input_tokens_seen": 8318112,
      "step": 39420
    },
    {
      "epoch": 4.337183718371837,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.028769360146077182,
      "loss": 0.2335,
      "num_input_tokens_seen": 8319168,
      "step": 39425
    },
    {
      "epoch": 4.337733773377337,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.028768788848587055,
      "loss": 0.2324,
      "num_input_tokens_seen": 8320256,
      "step": 39430
    },
    {
      "epoch": 4.338283828382838,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.0287682174241966,
      "loss": 0.2324,
      "num_input_tokens_seen": 8321248,
      "step": 39435
    },
    {
      "epoch": 4.338833883388339,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.0287676458729111,
      "loss": 0.2319,
      "num_input_tokens_seen": 8322272,
      "step": 39440
    },
    {
      "epoch": 4.33938393839384,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02876707419473581,
      "loss": 0.2324,
      "num_input_tokens_seen": 8323328,
      "step": 39445
    },
    {
      "epoch": 4.33993399339934,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.028766502389676003,
      "loss": 0.2324,
      "num_input_tokens_seen": 8324416,
      "step": 39450
    },
    {
      "epoch": 4.34048404840484,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02876593045773695,
      "loss": 0.2324,
      "num_input_tokens_seen": 8325472,
      "step": 39455
    },
    {
      "epoch": 4.341034103410341,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.02876535839892392,
      "loss": 0.2309,
      "num_input_tokens_seen": 8326592,
      "step": 39460
    },
    {
      "epoch": 4.341584158415841,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.028764786213242192,
      "loss": 0.2319,
      "num_input_tokens_seen": 8327584,
      "step": 39465
    },
    {
      "epoch": 4.3421342134213425,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02876421390069703,
      "loss": 0.2319,
      "num_input_tokens_seen": 8328576,
      "step": 39470
    },
    {
      "epoch": 4.342684268426843,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.028763641461293718,
      "loss": 0.2288,
      "num_input_tokens_seen": 8329568,
      "step": 39475
    },
    {
      "epoch": 4.343234323432343,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.028763068895037523,
      "loss": 0.2325,
      "num_input_tokens_seen": 8330624,
      "step": 39480
    },
    {
      "epoch": 4.343784378437844,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.028762496201933725,
      "loss": 0.2304,
      "num_input_tokens_seen": 8331680,
      "step": 39485
    },
    {
      "epoch": 4.344334433443344,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.028761923381987606,
      "loss": 0.234,
      "num_input_tokens_seen": 8332768,
      "step": 39490
    },
    {
      "epoch": 4.3448844884488445,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.028761350435204443,
      "loss": 0.2309,
      "num_input_tokens_seen": 8333856,
      "step": 39495
    },
    {
      "epoch": 4.3454345434543455,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.028760777361589513,
      "loss": 0.2325,
      "num_input_tokens_seen": 8335008,
      "step": 39500
    },
    {
      "epoch": 4.345984598459846,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.028760204161148103,
      "loss": 0.2314,
      "num_input_tokens_seen": 8336096,
      "step": 39505
    },
    {
      "epoch": 4.346534653465347,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.028759630833885497,
      "loss": 0.2309,
      "num_input_tokens_seen": 8337088,
      "step": 39510
    },
    {
      "epoch": 4.347084708470847,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.028759057379806975,
      "loss": 0.2329,
      "num_input_tokens_seen": 8338176,
      "step": 39515
    },
    {
      "epoch": 4.347634763476347,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02875848379891782,
      "loss": 0.2324,
      "num_input_tokens_seen": 8339264,
      "step": 39520
    },
    {
      "epoch": 4.348184818481848,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.028757910091223325,
      "loss": 0.2319,
      "num_input_tokens_seen": 8340352,
      "step": 39525
    },
    {
      "epoch": 4.3487348734873486,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.02875733625672877,
      "loss": 0.2319,
      "num_input_tokens_seen": 8341376,
      "step": 39530
    },
    {
      "epoch": 4.34928492849285,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.028756762295439456,
      "loss": 0.2309,
      "num_input_tokens_seen": 8342432,
      "step": 39535
    },
    {
      "epoch": 4.34983498349835,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.02875618820736066,
      "loss": 0.2319,
      "num_input_tokens_seen": 8343488,
      "step": 39540
    },
    {
      "epoch": 4.35038503850385,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.028755613992497674,
      "loss": 0.2319,
      "num_input_tokens_seen": 8344576,
      "step": 39545
    },
    {
      "epoch": 4.350935093509351,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.028755039650855803,
      "loss": 0.2319,
      "num_input_tokens_seen": 8345632,
      "step": 39550
    },
    {
      "epoch": 4.351485148514851,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.028754465182440323,
      "loss": 0.2298,
      "num_input_tokens_seen": 8346784,
      "step": 39555
    },
    {
      "epoch": 4.3520352035203524,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028753890587256543,
      "loss": 0.2335,
      "num_input_tokens_seen": 8347808,
      "step": 39560
    },
    {
      "epoch": 4.352585258525853,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.028753315865309756,
      "loss": 0.2329,
      "num_input_tokens_seen": 8348832,
      "step": 39565
    },
    {
      "epoch": 4.353135313531353,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.028752741016605254,
      "loss": 0.2298,
      "num_input_tokens_seen": 8349856,
      "step": 39570
    },
    {
      "epoch": 4.353685368536854,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.028752166041148334,
      "loss": 0.2303,
      "num_input_tokens_seen": 8350816,
      "step": 39575
    },
    {
      "epoch": 4.354235423542354,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.0287515909389443,
      "loss": 0.2309,
      "num_input_tokens_seen": 8351840,
      "step": 39580
    },
    {
      "epoch": 4.354785478547855,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.028751015709998454,
      "loss": 0.2309,
      "num_input_tokens_seen": 8352864,
      "step": 39585
    },
    {
      "epoch": 4.3553355335533555,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02875044035431609,
      "loss": 0.2324,
      "num_input_tokens_seen": 8353888,
      "step": 39590
    },
    {
      "epoch": 4.355885588558856,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.02874986487190252,
      "loss": 0.2303,
      "num_input_tokens_seen": 8354944,
      "step": 39595
    },
    {
      "epoch": 4.356435643564357,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.028749289262763038,
      "loss": 0.2324,
      "num_input_tokens_seen": 8356032,
      "step": 39600
    },
    {
      "epoch": 4.356985698569857,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.02874871352690296,
      "loss": 0.2303,
      "num_input_tokens_seen": 8357088,
      "step": 39605
    },
    {
      "epoch": 4.357535753575357,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.028748137664327583,
      "loss": 0.2319,
      "num_input_tokens_seen": 8358176,
      "step": 39610
    },
    {
      "epoch": 4.358085808580858,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.028747561675042218,
      "loss": 0.2298,
      "num_input_tokens_seen": 8359264,
      "step": 39615
    },
    {
      "epoch": 4.3586358635863585,
      "grad_norm": 0.00067138671875,
      "learning_rate": 0.028746985559052172,
      "loss": 0.2324,
      "num_input_tokens_seen": 8360288,
      "step": 39620
    },
    {
      "epoch": 4.3591859185918596,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02874640931636276,
      "loss": 0.2319,
      "num_input_tokens_seen": 8361344,
      "step": 39625
    },
    {
      "epoch": 4.35973597359736,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.028745832946979286,
      "loss": 0.2309,
      "num_input_tokens_seen": 8362432,
      "step": 39630
    },
    {
      "epoch": 4.36028602860286,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.028745256450907072,
      "loss": 0.2314,
      "num_input_tokens_seen": 8363488,
      "step": 39635
    },
    {
      "epoch": 4.360836083608361,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02874467982815142,
      "loss": 0.2314,
      "num_input_tokens_seen": 8364576,
      "step": 39640
    },
    {
      "epoch": 4.361386138613861,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02874410307871765,
      "loss": 0.2335,
      "num_input_tokens_seen": 8365632,
      "step": 39645
    },
    {
      "epoch": 4.361936193619362,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02874352620261108,
      "loss": 0.233,
      "num_input_tokens_seen": 8366688,
      "step": 39650
    },
    {
      "epoch": 4.362486248624863,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.028742949199837022,
      "loss": 0.2319,
      "num_input_tokens_seen": 8367712,
      "step": 39655
    },
    {
      "epoch": 4.363036303630363,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.028742372070400795,
      "loss": 0.2325,
      "num_input_tokens_seen": 8368800,
      "step": 39660
    },
    {
      "epoch": 4.363586358635864,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.028741794814307716,
      "loss": 0.2293,
      "num_input_tokens_seen": 8369856,
      "step": 39665
    },
    {
      "epoch": 4.364136413641364,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02874121743156311,
      "loss": 0.2335,
      "num_input_tokens_seen": 8370880,
      "step": 39670
    },
    {
      "epoch": 4.364686468646864,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.0287406399221723,
      "loss": 0.2293,
      "num_input_tokens_seen": 8372000,
      "step": 39675
    },
    {
      "epoch": 4.365236523652365,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.028740062286140605,
      "loss": 0.2303,
      "num_input_tokens_seen": 8373056,
      "step": 39680
    },
    {
      "epoch": 4.365786578657866,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028739484523473347,
      "loss": 0.2329,
      "num_input_tokens_seen": 8374144,
      "step": 39685
    },
    {
      "epoch": 4.366336633663367,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028738906634175852,
      "loss": 0.2303,
      "num_input_tokens_seen": 8375264,
      "step": 39690
    },
    {
      "epoch": 4.366886688668867,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.028738328618253445,
      "loss": 0.2303,
      "num_input_tokens_seen": 8376256,
      "step": 39695
    },
    {
      "epoch": 4.367436743674367,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02873775047571146,
      "loss": 0.2293,
      "num_input_tokens_seen": 8377376,
      "step": 39700
    },
    {
      "epoch": 4.367986798679868,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.02873717220655522,
      "loss": 0.2308,
      "num_input_tokens_seen": 8378432,
      "step": 39705
    },
    {
      "epoch": 4.368536853685368,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.028736593810790054,
      "loss": 0.2314,
      "num_input_tokens_seen": 8379520,
      "step": 39710
    },
    {
      "epoch": 4.3690869086908695,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.028736015288421297,
      "loss": 0.2303,
      "num_input_tokens_seen": 8380512,
      "step": 39715
    },
    {
      "epoch": 4.36963696369637,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028735436639454277,
      "loss": 0.2308,
      "num_input_tokens_seen": 8381568,
      "step": 39720
    },
    {
      "epoch": 4.37018701870187,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.028734857863894327,
      "loss": 0.2303,
      "num_input_tokens_seen": 8382592,
      "step": 39725
    },
    {
      "epoch": 4.370737073707371,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.02873427896174678,
      "loss": 0.2314,
      "num_input_tokens_seen": 8383584,
      "step": 39730
    },
    {
      "epoch": 4.371287128712871,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.02873369993301698,
      "loss": 0.2308,
      "num_input_tokens_seen": 8384576,
      "step": 39735
    },
    {
      "epoch": 4.371837183718371,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.028733120777710258,
      "loss": 0.2319,
      "num_input_tokens_seen": 8385632,
      "step": 39740
    },
    {
      "epoch": 4.3723872387238725,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.028732541495831945,
      "loss": 0.2351,
      "num_input_tokens_seen": 8386720,
      "step": 39745
    },
    {
      "epoch": 4.372937293729373,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.028731962087387396,
      "loss": 0.2298,
      "num_input_tokens_seen": 8387808,
      "step": 39750
    },
    {
      "epoch": 4.373487348734874,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02873138255238193,
      "loss": 0.2313,
      "num_input_tokens_seen": 8388832,
      "step": 39755
    },
    {
      "epoch": 4.374037403740374,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.02873080289082091,
      "loss": 0.2329,
      "num_input_tokens_seen": 8389824,
      "step": 39760
    },
    {
      "epoch": 4.374587458745874,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.028730223102709666,
      "loss": 0.2308,
      "num_input_tokens_seen": 8390848,
      "step": 39765
    },
    {
      "epoch": 4.375137513751375,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.028729643188053544,
      "loss": 0.2309,
      "num_input_tokens_seen": 8391872,
      "step": 39770
    },
    {
      "epoch": 4.3756875687568755,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.028729063146857886,
      "loss": 0.2335,
      "num_input_tokens_seen": 8392864,
      "step": 39775
    },
    {
      "epoch": 4.376237623762377,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.028728482979128042,
      "loss": 0.2329,
      "num_input_tokens_seen": 8393952,
      "step": 39780
    },
    {
      "epoch": 4.376787678767877,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.02872790268486936,
      "loss": 0.2335,
      "num_input_tokens_seen": 8395104,
      "step": 39785
    },
    {
      "epoch": 4.377337733773377,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.028727322264087182,
      "loss": 0.2324,
      "num_input_tokens_seen": 8396192,
      "step": 39790
    },
    {
      "epoch": 4.377887788778878,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.028726741716786866,
      "loss": 0.2308,
      "num_input_tokens_seen": 8397280,
      "step": 39795
    },
    {
      "epoch": 4.378437843784378,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.028726161042973757,
      "loss": 0.2324,
      "num_input_tokens_seen": 8398368,
      "step": 39800
    },
    {
      "epoch": 4.378987898789879,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.028725580242653204,
      "loss": 0.2324,
      "num_input_tokens_seen": 8399392,
      "step": 39805
    },
    {
      "epoch": 4.37953795379538,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02872499931583057,
      "loss": 0.2319,
      "num_input_tokens_seen": 8400512,
      "step": 39810
    },
    {
      "epoch": 4.38008800880088,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.028724418262511202,
      "loss": 0.2293,
      "num_input_tokens_seen": 8401600,
      "step": 39815
    },
    {
      "epoch": 4.380638063806381,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.028723837082700456,
      "loss": 0.2329,
      "num_input_tokens_seen": 8402656,
      "step": 39820
    },
    {
      "epoch": 4.381188118811881,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.02872325577640369,
      "loss": 0.2324,
      "num_input_tokens_seen": 8403712,
      "step": 39825
    },
    {
      "epoch": 4.381738173817382,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02872267434362626,
      "loss": 0.2303,
      "num_input_tokens_seen": 8404736,
      "step": 39830
    },
    {
      "epoch": 4.382288228822882,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.028722092784373522,
      "loss": 0.2319,
      "num_input_tokens_seen": 8405888,
      "step": 39835
    },
    {
      "epoch": 4.382838283828383,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.028721511098650843,
      "loss": 0.2308,
      "num_input_tokens_seen": 8406976,
      "step": 39840
    },
    {
      "epoch": 4.383388338833884,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02872092928646358,
      "loss": 0.2313,
      "num_input_tokens_seen": 8408032,
      "step": 39845
    },
    {
      "epoch": 4.383938393839384,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02872034734781709,
      "loss": 0.2283,
      "num_input_tokens_seen": 8409056,
      "step": 39850
    },
    {
      "epoch": 4.384488448844884,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02871976528271675,
      "loss": 0.2329,
      "num_input_tokens_seen": 8410176,
      "step": 39855
    },
    {
      "epoch": 4.385038503850385,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.028719183091167912,
      "loss": 0.2314,
      "num_input_tokens_seen": 8411232,
      "step": 39860
    },
    {
      "epoch": 4.385588558855885,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.028718600773175947,
      "loss": 0.2314,
      "num_input_tokens_seen": 8412256,
      "step": 39865
    },
    {
      "epoch": 4.3861386138613865,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028718018328746226,
      "loss": 0.2314,
      "num_input_tokens_seen": 8413376,
      "step": 39870
    },
    {
      "epoch": 4.386688668866887,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.02871743575788411,
      "loss": 0.2314,
      "num_input_tokens_seen": 8414432,
      "step": 39875
    },
    {
      "epoch": 4.387238723872387,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02871685306059497,
      "loss": 0.2303,
      "num_input_tokens_seen": 8415456,
      "step": 39880
    },
    {
      "epoch": 4.387788778877888,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.028716270236884178,
      "loss": 0.2293,
      "num_input_tokens_seen": 8416512,
      "step": 39885
    },
    {
      "epoch": 4.388338833883388,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.028715687286757104,
      "loss": 0.2324,
      "num_input_tokens_seen": 8417664,
      "step": 39890
    },
    {
      "epoch": 4.388888888888889,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02871510421021912,
      "loss": 0.2298,
      "num_input_tokens_seen": 8418720,
      "step": 39895
    },
    {
      "epoch": 4.3894389438943895,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.028714521007275607,
      "loss": 0.2334,
      "num_input_tokens_seen": 8419744,
      "step": 39900
    },
    {
      "epoch": 4.38998899889989,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.02871393767793193,
      "loss": 0.2313,
      "num_input_tokens_seen": 8420832,
      "step": 39905
    },
    {
      "epoch": 4.390539053905391,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028713354222193472,
      "loss": 0.2304,
      "num_input_tokens_seen": 8421888,
      "step": 39910
    },
    {
      "epoch": 4.391089108910891,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02871277064006561,
      "loss": 0.2303,
      "num_input_tokens_seen": 8422976,
      "step": 39915
    },
    {
      "epoch": 4.391639163916391,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.028712186931553717,
      "loss": 0.2304,
      "num_input_tokens_seen": 8424032,
      "step": 39920
    },
    {
      "epoch": 4.392189218921892,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.028711603096663176,
      "loss": 0.2304,
      "num_input_tokens_seen": 8425056,
      "step": 39925
    },
    {
      "epoch": 4.3927392739273925,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028711019135399377,
      "loss": 0.2304,
      "num_input_tokens_seen": 8426176,
      "step": 39930
    },
    {
      "epoch": 4.393289328932894,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.028710435047767685,
      "loss": 0.2288,
      "num_input_tokens_seen": 8427296,
      "step": 39935
    },
    {
      "epoch": 4.393839383938394,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.028709850833773497,
      "loss": 0.233,
      "num_input_tokens_seen": 8428352,
      "step": 39940
    },
    {
      "epoch": 4.394389438943894,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.02870926649342219,
      "loss": 0.2351,
      "num_input_tokens_seen": 8429440,
      "step": 39945
    },
    {
      "epoch": 4.394939493949395,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.028708682026719154,
      "loss": 0.2283,
      "num_input_tokens_seen": 8430496,
      "step": 39950
    },
    {
      "epoch": 4.395489548954895,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.028708097433669772,
      "loss": 0.2319,
      "num_input_tokens_seen": 8431520,
      "step": 39955
    },
    {
      "epoch": 4.396039603960396,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02870751271427944,
      "loss": 0.2319,
      "num_input_tokens_seen": 8432608,
      "step": 39960
    },
    {
      "epoch": 4.396589658965897,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.02870692786855353,
      "loss": 0.2298,
      "num_input_tokens_seen": 8433696,
      "step": 39965
    },
    {
      "epoch": 4.397139713971397,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.028706342896497452,
      "loss": 0.2309,
      "num_input_tokens_seen": 8434752,
      "step": 39970
    },
    {
      "epoch": 4.397689768976898,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028705757798116585,
      "loss": 0.2304,
      "num_input_tokens_seen": 8435872,
      "step": 39975
    },
    {
      "epoch": 4.398239823982398,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.028705172573416326,
      "loss": 0.2325,
      "num_input_tokens_seen": 8436928,
      "step": 39980
    },
    {
      "epoch": 4.398789878987899,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.028704587222402066,
      "loss": 0.2303,
      "num_input_tokens_seen": 8438048,
      "step": 39985
    },
    {
      "epoch": 4.399339933993399,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.028704001745079203,
      "loss": 0.2298,
      "num_input_tokens_seen": 8439104,
      "step": 39990
    },
    {
      "epoch": 4.3998899889989,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.028703416141453133,
      "loss": 0.2309,
      "num_input_tokens_seen": 8440192,
      "step": 39995
    },
    {
      "epoch": 4.400440044004401,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.02870283041152925,
      "loss": 0.2298,
      "num_input_tokens_seen": 8441280,
      "step": 40000
    },
    {
      "epoch": 4.400990099009901,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.028702244555312954,
      "loss": 0.2319,
      "num_input_tokens_seen": 8442304,
      "step": 40005
    },
    {
      "epoch": 4.401540154015402,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.02870165857280964,
      "loss": 0.2335,
      "num_input_tokens_seen": 8443328,
      "step": 40010
    },
    {
      "epoch": 4.402090209020902,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.028701072464024723,
      "loss": 0.2278,
      "num_input_tokens_seen": 8444320,
      "step": 40015
    },
    {
      "epoch": 4.402640264026402,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.028700486228963592,
      "loss": 0.2288,
      "num_input_tokens_seen": 8445344,
      "step": 40020
    },
    {
      "epoch": 4.4031903190319035,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02869989986763165,
      "loss": 0.2325,
      "num_input_tokens_seen": 8446368,
      "step": 40025
    },
    {
      "epoch": 4.403740374037404,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.02869931338003431,
      "loss": 0.2299,
      "num_input_tokens_seen": 8447456,
      "step": 40030
    },
    {
      "epoch": 4.404290429042904,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.028698726766176973,
      "loss": 0.2304,
      "num_input_tokens_seen": 8448480,
      "step": 40035
    },
    {
      "epoch": 4.404840484048405,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02869814002606504,
      "loss": 0.2294,
      "num_input_tokens_seen": 8449568,
      "step": 40040
    },
    {
      "epoch": 4.405390539053905,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.028697553159703924,
      "loss": 0.2341,
      "num_input_tokens_seen": 8450528,
      "step": 40045
    },
    {
      "epoch": 4.405940594059406,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.028696966167099033,
      "loss": 0.2273,
      "num_input_tokens_seen": 8451552,
      "step": 40050
    },
    {
      "epoch": 4.4064906490649065,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02869637904825578,
      "loss": 0.2315,
      "num_input_tokens_seen": 8452608,
      "step": 40055
    },
    {
      "epoch": 4.407040704070407,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.028695791803179573,
      "loss": 0.2304,
      "num_input_tokens_seen": 8453632,
      "step": 40060
    },
    {
      "epoch": 4.407590759075908,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.028695204431875825,
      "loss": 0.232,
      "num_input_tokens_seen": 8454688,
      "step": 40065
    },
    {
      "epoch": 4.408140814081408,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02869461693434995,
      "loss": 0.2357,
      "num_input_tokens_seen": 8455648,
      "step": 40070
    },
    {
      "epoch": 4.408690869086909,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.02869402931060736,
      "loss": 0.2357,
      "num_input_tokens_seen": 8456672,
      "step": 40075
    },
    {
      "epoch": 4.409240924092409,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02869344156065347,
      "loss": 0.2356,
      "num_input_tokens_seen": 8457728,
      "step": 40080
    },
    {
      "epoch": 4.4097909790979095,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.028692853684493706,
      "loss": 0.2325,
      "num_input_tokens_seen": 8458784,
      "step": 40085
    },
    {
      "epoch": 4.410341034103411,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.028692265682133476,
      "loss": 0.2319,
      "num_input_tokens_seen": 8459840,
      "step": 40090
    },
    {
      "epoch": 4.410891089108911,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.028691677553578206,
      "loss": 0.2309,
      "num_input_tokens_seen": 8460896,
      "step": 40095
    },
    {
      "epoch": 4.411441144114411,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.028691089298833312,
      "loss": 0.2319,
      "num_input_tokens_seen": 8461888,
      "step": 40100
    },
    {
      "epoch": 4.411991199119912,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.02869050091790422,
      "loss": 0.2309,
      "num_input_tokens_seen": 8462944,
      "step": 40105
    },
    {
      "epoch": 4.412541254125412,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.028689912410796346,
      "loss": 0.2314,
      "num_input_tokens_seen": 8464032,
      "step": 40110
    },
    {
      "epoch": 4.413091309130913,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.02868932377751512,
      "loss": 0.2299,
      "num_input_tokens_seen": 8465088,
      "step": 40115
    },
    {
      "epoch": 4.413641364136414,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02868873501806597,
      "loss": 0.2319,
      "num_input_tokens_seen": 8466144,
      "step": 40120
    },
    {
      "epoch": 4.414191419141914,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.028688146132454313,
      "loss": 0.2309,
      "num_input_tokens_seen": 8467200,
      "step": 40125
    },
    {
      "epoch": 4.414741474147415,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028687557120685582,
      "loss": 0.2314,
      "num_input_tokens_seen": 8468256,
      "step": 40130
    },
    {
      "epoch": 4.415291529152915,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.028686967982765207,
      "loss": 0.2324,
      "num_input_tokens_seen": 8469248,
      "step": 40135
    },
    {
      "epoch": 4.415841584158416,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.028686378718698613,
      "loss": 0.2298,
      "num_input_tokens_seen": 8470304,
      "step": 40140
    },
    {
      "epoch": 4.416391639163916,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.028685789328491234,
      "loss": 0.2288,
      "num_input_tokens_seen": 8471424,
      "step": 40145
    },
    {
      "epoch": 4.416941694169417,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.028685199812148503,
      "loss": 0.2298,
      "num_input_tokens_seen": 8472512,
      "step": 40150
    },
    {
      "epoch": 4.417491749174918,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028684610169675853,
      "loss": 0.2304,
      "num_input_tokens_seen": 8473600,
      "step": 40155
    },
    {
      "epoch": 4.418041804180418,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028684020401078716,
      "loss": 0.2325,
      "num_input_tokens_seen": 8474656,
      "step": 40160
    },
    {
      "epoch": 4.418591859185918,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.02868343050636253,
      "loss": 0.2326,
      "num_input_tokens_seen": 8475776,
      "step": 40165
    },
    {
      "epoch": 4.419141914191419,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02868284048553273,
      "loss": 0.231,
      "num_input_tokens_seen": 8476832,
      "step": 40170
    },
    {
      "epoch": 4.419691969196919,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02868225033859476,
      "loss": 0.2305,
      "num_input_tokens_seen": 8477920,
      "step": 40175
    },
    {
      "epoch": 4.4202420242024205,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028681660065554046,
      "loss": 0.2295,
      "num_input_tokens_seen": 8479072,
      "step": 40180
    },
    {
      "epoch": 4.420792079207921,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.02868106966641604,
      "loss": 0.232,
      "num_input_tokens_seen": 8480160,
      "step": 40185
    },
    {
      "epoch": 4.421342134213421,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02868047914118618,
      "loss": 0.2309,
      "num_input_tokens_seen": 8481152,
      "step": 40190
    },
    {
      "epoch": 4.421892189218922,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02867988848986991,
      "loss": 0.231,
      "num_input_tokens_seen": 8482112,
      "step": 40195
    },
    {
      "epoch": 4.422442244224422,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.028679297712472667,
      "loss": 0.2325,
      "num_input_tokens_seen": 8483200,
      "step": 40200
    },
    {
      "epoch": 4.422992299229923,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.028678706808999904,
      "loss": 0.2325,
      "num_input_tokens_seen": 8484288,
      "step": 40205
    },
    {
      "epoch": 4.4235423542354235,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.028678115779457062,
      "loss": 0.2346,
      "num_input_tokens_seen": 8485344,
      "step": 40210
    },
    {
      "epoch": 4.424092409240924,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.028677524623849593,
      "loss": 0.2304,
      "num_input_tokens_seen": 8486368,
      "step": 40215
    },
    {
      "epoch": 4.424642464246425,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.028676933342182943,
      "loss": 0.2294,
      "num_input_tokens_seen": 8487424,
      "step": 40220
    },
    {
      "epoch": 4.425192519251925,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02867634193446256,
      "loss": 0.2314,
      "num_input_tokens_seen": 8488448,
      "step": 40225
    },
    {
      "epoch": 4.425742574257426,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.028675750400693895,
      "loss": 0.234,
      "num_input_tokens_seen": 8489440,
      "step": 40230
    },
    {
      "epoch": 4.426292629262926,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.028675158740882403,
      "loss": 0.2308,
      "num_input_tokens_seen": 8490496,
      "step": 40235
    },
    {
      "epoch": 4.4268426842684265,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.028674566955033533,
      "loss": 0.2324,
      "num_input_tokens_seen": 8491552,
      "step": 40240
    },
    {
      "epoch": 4.427392739273928,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.028673975043152743,
      "loss": 0.2314,
      "num_input_tokens_seen": 8492640,
      "step": 40245
    },
    {
      "epoch": 4.427942794279428,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.028673383005245485,
      "loss": 0.2314,
      "num_input_tokens_seen": 8493664,
      "step": 40250
    },
    {
      "epoch": 4.428492849284929,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.028672790841317214,
      "loss": 0.2303,
      "num_input_tokens_seen": 8494688,
      "step": 40255
    },
    {
      "epoch": 4.429042904290429,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.028672198551373394,
      "loss": 0.2314,
      "num_input_tokens_seen": 8495680,
      "step": 40260
    },
    {
      "epoch": 4.429592959295929,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.028671606135419483,
      "loss": 0.2314,
      "num_input_tokens_seen": 8496736,
      "step": 40265
    },
    {
      "epoch": 4.43014301430143,
      "grad_norm": 0.009765625,
      "learning_rate": 0.028671013593460935,
      "loss": 0.2319,
      "num_input_tokens_seen": 8497824,
      "step": 40270
    },
    {
      "epoch": 4.430693069306931,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.028670420925503214,
      "loss": 0.2319,
      "num_input_tokens_seen": 8498880,
      "step": 40275
    },
    {
      "epoch": 4.431243124312431,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.028669828131551785,
      "loss": 0.2324,
      "num_input_tokens_seen": 8499936,
      "step": 40280
    },
    {
      "epoch": 4.431793179317932,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02866923521161211,
      "loss": 0.2303,
      "num_input_tokens_seen": 8501024,
      "step": 40285
    },
    {
      "epoch": 4.432343234323432,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02866864216568965,
      "loss": 0.2313,
      "num_input_tokens_seen": 8502080,
      "step": 40290
    },
    {
      "epoch": 4.432893289328933,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02866804899378988,
      "loss": 0.2319,
      "num_input_tokens_seen": 8503168,
      "step": 40295
    },
    {
      "epoch": 4.433443344334433,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02866745569591825,
      "loss": 0.2303,
      "num_input_tokens_seen": 8504192,
      "step": 40300
    },
    {
      "epoch": 4.433993399339934,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02866686227208025,
      "loss": 0.2293,
      "num_input_tokens_seen": 8505312,
      "step": 40305
    },
    {
      "epoch": 4.434543454345435,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.028666268722281337,
      "loss": 0.2319,
      "num_input_tokens_seen": 8506304,
      "step": 40310
    },
    {
      "epoch": 4.435093509350935,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02866567504652698,
      "loss": 0.2324,
      "num_input_tokens_seen": 8507360,
      "step": 40315
    },
    {
      "epoch": 4.435643564356436,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.028665081244822657,
      "loss": 0.2318,
      "num_input_tokens_seen": 8508448,
      "step": 40320
    },
    {
      "epoch": 4.436193619361936,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.028664487317173835,
      "loss": 0.2314,
      "num_input_tokens_seen": 8509504,
      "step": 40325
    },
    {
      "epoch": 4.436743674367436,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.028663893263585986,
      "loss": 0.2324,
      "num_input_tokens_seen": 8510528,
      "step": 40330
    },
    {
      "epoch": 4.4372937293729375,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.028663299084064595,
      "loss": 0.2314,
      "num_input_tokens_seen": 8511552,
      "step": 40335
    },
    {
      "epoch": 4.437843784378438,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.028662704778615128,
      "loss": 0.2319,
      "num_input_tokens_seen": 8512640,
      "step": 40340
    },
    {
      "epoch": 4.438393839383938,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02866211034724307,
      "loss": 0.2309,
      "num_input_tokens_seen": 8513696,
      "step": 40345
    },
    {
      "epoch": 4.438943894389439,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.028661515789953896,
      "loss": 0.2329,
      "num_input_tokens_seen": 8514816,
      "step": 40350
    },
    {
      "epoch": 4.439493949394939,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.028660921106753088,
      "loss": 0.2303,
      "num_input_tokens_seen": 8515840,
      "step": 40355
    },
    {
      "epoch": 4.44004400440044,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.028660326297646123,
      "loss": 0.2308,
      "num_input_tokens_seen": 8516864,
      "step": 40360
    },
    {
      "epoch": 4.4405940594059405,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.028659731362638486,
      "loss": 0.2319,
      "num_input_tokens_seen": 8517920,
      "step": 40365
    },
    {
      "epoch": 4.441144114411441,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.028659136301735666,
      "loss": 0.2324,
      "num_input_tokens_seen": 8519008,
      "step": 40370
    },
    {
      "epoch": 4.441694169416942,
      "grad_norm": 0.00055694580078125,
      "learning_rate": 0.028658541114943134,
      "loss": 0.2335,
      "num_input_tokens_seen": 8520032,
      "step": 40375
    },
    {
      "epoch": 4.442244224422442,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02865794580226638,
      "loss": 0.2288,
      "num_input_tokens_seen": 8521056,
      "step": 40380
    },
    {
      "epoch": 4.442794279427943,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.028657350363710896,
      "loss": 0.2318,
      "num_input_tokens_seen": 8522080,
      "step": 40385
    },
    {
      "epoch": 4.443344334433443,
      "grad_norm": 0.000762939453125,
      "learning_rate": 0.02865675479928217,
      "loss": 0.2314,
      "num_input_tokens_seen": 8523200,
      "step": 40390
    },
    {
      "epoch": 4.4438943894389435,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.028656159108985686,
      "loss": 0.2314,
      "num_input_tokens_seen": 8524224,
      "step": 40395
    },
    {
      "epoch": 4.444444444444445,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.028655563292826938,
      "loss": 0.2304,
      "num_input_tokens_seen": 8525312,
      "step": 40400
    },
    {
      "epoch": 4.444994499449945,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.028654967350811415,
      "loss": 0.2324,
      "num_input_tokens_seen": 8526368,
      "step": 40405
    },
    {
      "epoch": 4.445544554455446,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02865437128294461,
      "loss": 0.2303,
      "num_input_tokens_seen": 8527488,
      "step": 40410
    },
    {
      "epoch": 4.446094609460946,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.028653775089232017,
      "loss": 0.2319,
      "num_input_tokens_seen": 8528512,
      "step": 40415
    },
    {
      "epoch": 4.446644664466446,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.028653178769679127,
      "loss": 0.2303,
      "num_input_tokens_seen": 8529536,
      "step": 40420
    },
    {
      "epoch": 4.447194719471947,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.028652582324291444,
      "loss": 0.2303,
      "num_input_tokens_seen": 8530592,
      "step": 40425
    },
    {
      "epoch": 4.447744774477448,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.028651985753074462,
      "loss": 0.2334,
      "num_input_tokens_seen": 8531648,
      "step": 40430
    },
    {
      "epoch": 4.448294829482949,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.028651389056033674,
      "loss": 0.2314,
      "num_input_tokens_seen": 8532672,
      "step": 40435
    },
    {
      "epoch": 4.448844884488449,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.028650792233174585,
      "loss": 0.2329,
      "num_input_tokens_seen": 8533728,
      "step": 40440
    },
    {
      "epoch": 4.449394939493949,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.028650195284502696,
      "loss": 0.2303,
      "num_input_tokens_seen": 8534752,
      "step": 40445
    },
    {
      "epoch": 4.44994499449945,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.028649598210023504,
      "loss": 0.2309,
      "num_input_tokens_seen": 8535776,
      "step": 40450
    },
    {
      "epoch": 4.4504950495049505,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.028649001009742516,
      "loss": 0.2303,
      "num_input_tokens_seen": 8536832,
      "step": 40455
    },
    {
      "epoch": 4.451045104510451,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02864840368366524,
      "loss": 0.234,
      "num_input_tokens_seen": 8537888,
      "step": 40460
    },
    {
      "epoch": 4.451595159515952,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.028647806231797173,
      "loss": 0.2313,
      "num_input_tokens_seen": 8538976,
      "step": 40465
    },
    {
      "epoch": 4.452145214521452,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02864720865414382,
      "loss": 0.2314,
      "num_input_tokens_seen": 8540128,
      "step": 40470
    },
    {
      "epoch": 4.452695269526953,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.028646610950710697,
      "loss": 0.2319,
      "num_input_tokens_seen": 8541216,
      "step": 40475
    },
    {
      "epoch": 4.453245324532453,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.028646013121503312,
      "loss": 0.2314,
      "num_input_tokens_seen": 8542304,
      "step": 40480
    },
    {
      "epoch": 4.4537953795379535,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.02864541516652717,
      "loss": 0.2314,
      "num_input_tokens_seen": 8543360,
      "step": 40485
    },
    {
      "epoch": 4.4543454345434546,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.028644817085787786,
      "loss": 0.2319,
      "num_input_tokens_seen": 8544480,
      "step": 40490
    },
    {
      "epoch": 4.454895489548955,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.028644218879290667,
      "loss": 0.2303,
      "num_input_tokens_seen": 8545536,
      "step": 40495
    },
    {
      "epoch": 4.455445544554456,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02864362054704133,
      "loss": 0.2309,
      "num_input_tokens_seen": 8546688,
      "step": 40500
    },
    {
      "epoch": 4.455995599559956,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02864302208904529,
      "loss": 0.2313,
      "num_input_tokens_seen": 8547744,
      "step": 40505
    },
    {
      "epoch": 4.456545654565456,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.028642423505308063,
      "loss": 0.2303,
      "num_input_tokens_seen": 8548736,
      "step": 40510
    },
    {
      "epoch": 4.457095709570957,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.028641824795835165,
      "loss": 0.2303,
      "num_input_tokens_seen": 8549792,
      "step": 40515
    },
    {
      "epoch": 4.457645764576458,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.028641225960632115,
      "loss": 0.2329,
      "num_input_tokens_seen": 8550784,
      "step": 40520
    },
    {
      "epoch": 4.458195819581958,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.028640626999704426,
      "loss": 0.2313,
      "num_input_tokens_seen": 8551872,
      "step": 40525
    },
    {
      "epoch": 4.458745874587459,
      "grad_norm": 0.009765625,
      "learning_rate": 0.028640027913057628,
      "loss": 0.2319,
      "num_input_tokens_seen": 8552864,
      "step": 40530
    },
    {
      "epoch": 4.459295929592959,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.028639428700697238,
      "loss": 0.2356,
      "num_input_tokens_seen": 8553984,
      "step": 40535
    },
    {
      "epoch": 4.45984598459846,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028638829362628772,
      "loss": 0.2314,
      "num_input_tokens_seen": 8555040,
      "step": 40540
    },
    {
      "epoch": 4.46039603960396,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.028638229898857767,
      "loss": 0.234,
      "num_input_tokens_seen": 8556096,
      "step": 40545
    },
    {
      "epoch": 4.460946094609461,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.028637630309389737,
      "loss": 0.2324,
      "num_input_tokens_seen": 8557152,
      "step": 40550
    },
    {
      "epoch": 4.461496149614962,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.028637030594230214,
      "loss": 0.2329,
      "num_input_tokens_seen": 8558240,
      "step": 40555
    },
    {
      "epoch": 4.462046204620462,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.028636430753384724,
      "loss": 0.2329,
      "num_input_tokens_seen": 8559296,
      "step": 40560
    },
    {
      "epoch": 4.462596259625963,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028635830786858798,
      "loss": 0.2335,
      "num_input_tokens_seen": 8560384,
      "step": 40565
    },
    {
      "epoch": 4.463146314631463,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02863523069465796,
      "loss": 0.2309,
      "num_input_tokens_seen": 8561472,
      "step": 40570
    },
    {
      "epoch": 4.463696369636963,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.028634630476787742,
      "loss": 0.2319,
      "num_input_tokens_seen": 8562528,
      "step": 40575
    },
    {
      "epoch": 4.4642464246424645,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02863403013325368,
      "loss": 0.2314,
      "num_input_tokens_seen": 8563552,
      "step": 40580
    },
    {
      "epoch": 4.464796479647965,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.028633429664061308,
      "loss": 0.2324,
      "num_input_tokens_seen": 8564608,
      "step": 40585
    },
    {
      "epoch": 4.465346534653466,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02863282906921615,
      "loss": 0.2308,
      "num_input_tokens_seen": 8565664,
      "step": 40590
    },
    {
      "epoch": 4.465896589658966,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02863222834872375,
      "loss": 0.2324,
      "num_input_tokens_seen": 8566688,
      "step": 40595
    },
    {
      "epoch": 4.466446644664466,
      "grad_norm": 0.009765625,
      "learning_rate": 0.028631627502589643,
      "loss": 0.2288,
      "num_input_tokens_seen": 8567744,
      "step": 40600
    },
    {
      "epoch": 4.466996699669967,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.028631026530819365,
      "loss": 0.2293,
      "num_input_tokens_seen": 8568704,
      "step": 40605
    },
    {
      "epoch": 4.4675467546754675,
      "grad_norm": 0.000843048095703125,
      "learning_rate": 0.02863042543341846,
      "loss": 0.234,
      "num_input_tokens_seen": 8569696,
      "step": 40610
    },
    {
      "epoch": 4.468096809680969,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.02862982421039246,
      "loss": 0.2358,
      "num_input_tokens_seen": 8570784,
      "step": 40615
    },
    {
      "epoch": 4.468646864686469,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.028629222861746913,
      "loss": 0.233,
      "num_input_tokens_seen": 8571808,
      "step": 40620
    },
    {
      "epoch": 4.469196919691969,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02862862138748736,
      "loss": 0.2267,
      "num_input_tokens_seen": 8572864,
      "step": 40625
    },
    {
      "epoch": 4.46974697469747,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02862801978761934,
      "loss": 0.2337,
      "num_input_tokens_seen": 8573952,
      "step": 40630
    },
    {
      "epoch": 4.47029702970297,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.028627418062148403,
      "loss": 0.2284,
      "num_input_tokens_seen": 8574976,
      "step": 40635
    },
    {
      "epoch": 4.4708470847084705,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.028626816211080094,
      "loss": 0.2389,
      "num_input_tokens_seen": 8576000,
      "step": 40640
    },
    {
      "epoch": 4.471397139713972,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.028626214234419956,
      "loss": 0.2393,
      "num_input_tokens_seen": 8577056,
      "step": 40645
    },
    {
      "epoch": 4.471947194719472,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02862561213217354,
      "loss": 0.2319,
      "num_input_tokens_seen": 8578112,
      "step": 40650
    },
    {
      "epoch": 4.472497249724973,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.0286250099043464,
      "loss": 0.233,
      "num_input_tokens_seen": 8579136,
      "step": 40655
    },
    {
      "epoch": 4.473047304730473,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.028624407550944076,
      "loss": 0.2309,
      "num_input_tokens_seen": 8580224,
      "step": 40660
    },
    {
      "epoch": 4.473597359735973,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.02862380507197213,
      "loss": 0.2299,
      "num_input_tokens_seen": 8581280,
      "step": 40665
    },
    {
      "epoch": 4.474147414741474,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.028623202467436108,
      "loss": 0.2319,
      "num_input_tokens_seen": 8582432,
      "step": 40670
    },
    {
      "epoch": 4.474697469746975,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02862259973734156,
      "loss": 0.2304,
      "num_input_tokens_seen": 8583488,
      "step": 40675
    },
    {
      "epoch": 4.475247524752476,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.02862199688169406,
      "loss": 0.2309,
      "num_input_tokens_seen": 8584608,
      "step": 40680
    },
    {
      "epoch": 4.475797579757976,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.028621393900499136,
      "loss": 0.2319,
      "num_input_tokens_seen": 8585728,
      "step": 40685
    },
    {
      "epoch": 4.476347634763476,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.028620790793762372,
      "loss": 0.2304,
      "num_input_tokens_seen": 8586720,
      "step": 40690
    },
    {
      "epoch": 4.476897689768977,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.02862018756148931,
      "loss": 0.2314,
      "num_input_tokens_seen": 8587808,
      "step": 40695
    },
    {
      "epoch": 4.477447744774477,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.028619584203685516,
      "loss": 0.2308,
      "num_input_tokens_seen": 8588832,
      "step": 40700
    },
    {
      "epoch": 4.477997799779978,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02861898072035655,
      "loss": 0.2324,
      "num_input_tokens_seen": 8589856,
      "step": 40705
    },
    {
      "epoch": 4.478547854785479,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.028618377111507973,
      "loss": 0.2329,
      "num_input_tokens_seen": 8591008,
      "step": 40710
    },
    {
      "epoch": 4.479097909790979,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02861777337714535,
      "loss": 0.234,
      "num_input_tokens_seen": 8592032,
      "step": 40715
    },
    {
      "epoch": 4.47964796479648,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.028617169517274242,
      "loss": 0.2303,
      "num_input_tokens_seen": 8593056,
      "step": 40720
    },
    {
      "epoch": 4.48019801980198,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.02861656553190022,
      "loss": 0.2304,
      "num_input_tokens_seen": 8594176,
      "step": 40725
    },
    {
      "epoch": 4.48074807480748,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.028615961421028843,
      "loss": 0.2329,
      "num_input_tokens_seen": 8595168,
      "step": 40730
    },
    {
      "epoch": 4.4812981298129815,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.02861535718466569,
      "loss": 0.2298,
      "num_input_tokens_seen": 8596256,
      "step": 40735
    },
    {
      "epoch": 4.481848184818482,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02861475282281632,
      "loss": 0.2309,
      "num_input_tokens_seen": 8597280,
      "step": 40740
    },
    {
      "epoch": 4.482398239823983,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.028614148335486297,
      "loss": 0.2309,
      "num_input_tokens_seen": 8598336,
      "step": 40745
    },
    {
      "epoch": 4.482948294829483,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.028613543722681215,
      "loss": 0.2345,
      "num_input_tokens_seen": 8599392,
      "step": 40750
    },
    {
      "epoch": 4.483498349834983,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.028612938984406627,
      "loss": 0.2356,
      "num_input_tokens_seen": 8600416,
      "step": 40755
    },
    {
      "epoch": 4.484048404840484,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.02861233412066811,
      "loss": 0.2298,
      "num_input_tokens_seen": 8601504,
      "step": 40760
    },
    {
      "epoch": 4.4845984598459845,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.028611729131471246,
      "loss": 0.2319,
      "num_input_tokens_seen": 8602592,
      "step": 40765
    },
    {
      "epoch": 4.485148514851485,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028611124016821605,
      "loss": 0.2325,
      "num_input_tokens_seen": 8603616,
      "step": 40770
    },
    {
      "epoch": 4.485698569856986,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.028610518776724762,
      "loss": 0.2319,
      "num_input_tokens_seen": 8604736,
      "step": 40775
    },
    {
      "epoch": 4.486248624862486,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.028609913411186304,
      "loss": 0.2288,
      "num_input_tokens_seen": 8605824,
      "step": 40780
    },
    {
      "epoch": 4.486798679867987,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.0286093079202118,
      "loss": 0.2314,
      "num_input_tokens_seen": 8606976,
      "step": 40785
    },
    {
      "epoch": 4.487348734873487,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.028608702303806837,
      "loss": 0.233,
      "num_input_tokens_seen": 8608000,
      "step": 40790
    },
    {
      "epoch": 4.4878987898789875,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.028608096561976996,
      "loss": 0.2324,
      "num_input_tokens_seen": 8609024,
      "step": 40795
    },
    {
      "epoch": 4.488448844884489,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.028607490694727858,
      "loss": 0.2325,
      "num_input_tokens_seen": 8610080,
      "step": 40800
    },
    {
      "epoch": 4.488998899889989,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.028606884702065006,
      "loss": 0.2335,
      "num_input_tokens_seen": 8611072,
      "step": 40805
    },
    {
      "epoch": 4.48954895489549,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02860627858399403,
      "loss": 0.233,
      "num_input_tokens_seen": 8612128,
      "step": 40810
    },
    {
      "epoch": 4.49009900990099,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.02860567234052051,
      "loss": 0.2293,
      "num_input_tokens_seen": 8613184,
      "step": 40815
    },
    {
      "epoch": 4.49064906490649,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.028605065971650037,
      "loss": 0.2293,
      "num_input_tokens_seen": 8614208,
      "step": 40820
    },
    {
      "epoch": 4.491199119911991,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.028604459477388204,
      "loss": 0.2299,
      "num_input_tokens_seen": 8615232,
      "step": 40825
    },
    {
      "epoch": 4.491749174917492,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.028603852857740593,
      "loss": 0.2314,
      "num_input_tokens_seen": 8616288,
      "step": 40830
    },
    {
      "epoch": 4.492299229922993,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.028603246112712796,
      "loss": 0.233,
      "num_input_tokens_seen": 8617312,
      "step": 40835
    },
    {
      "epoch": 4.492849284928493,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.028602639242310407,
      "loss": 0.2299,
      "num_input_tokens_seen": 8618400,
      "step": 40840
    },
    {
      "epoch": 4.493399339933993,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.028602032246539025,
      "loss": 0.2304,
      "num_input_tokens_seen": 8619456,
      "step": 40845
    },
    {
      "epoch": 4.493949394939494,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.028601425125404232,
      "loss": 0.2362,
      "num_input_tokens_seen": 8620544,
      "step": 40850
    },
    {
      "epoch": 4.494499449944994,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028600817878911635,
      "loss": 0.2304,
      "num_input_tokens_seen": 8621632,
      "step": 40855
    },
    {
      "epoch": 4.4950495049504955,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028600210507066823,
      "loss": 0.2309,
      "num_input_tokens_seen": 8622784,
      "step": 40860
    },
    {
      "epoch": 4.495599559955996,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0285996030098754,
      "loss": 0.2299,
      "num_input_tokens_seen": 8623808,
      "step": 40865
    },
    {
      "epoch": 4.496149614961496,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02859899538734296,
      "loss": 0.2309,
      "num_input_tokens_seen": 8624800,
      "step": 40870
    },
    {
      "epoch": 4.496699669966997,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.0285983876394751,
      "loss": 0.231,
      "num_input_tokens_seen": 8625920,
      "step": 40875
    },
    {
      "epoch": 4.497249724972497,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02859777976627743,
      "loss": 0.232,
      "num_input_tokens_seen": 8626976,
      "step": 40880
    },
    {
      "epoch": 4.497799779977997,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.02859717176775555,
      "loss": 0.2294,
      "num_input_tokens_seen": 8628000,
      "step": 40885
    },
    {
      "epoch": 4.4983498349834985,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.028596563643915062,
      "loss": 0.2315,
      "num_input_tokens_seen": 8629088,
      "step": 40890
    },
    {
      "epoch": 4.498899889988999,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02859595539476157,
      "loss": 0.2335,
      "num_input_tokens_seen": 8630080,
      "step": 40895
    },
    {
      "epoch": 4.4994499449945,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02859534702030068,
      "loss": 0.234,
      "num_input_tokens_seen": 8631136,
      "step": 40900
    },
    {
      "epoch": 4.5,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.028594738520538003,
      "loss": 0.2268,
      "num_input_tokens_seen": 8632224,
      "step": 40905
    },
    {
      "epoch": 4.5005500550055,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.028594129895479145,
      "loss": 0.2299,
      "num_input_tokens_seen": 8633344,
      "step": 40910
    },
    {
      "epoch": 4.501100110011001,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.028593521145129714,
      "loss": 0.2319,
      "num_input_tokens_seen": 8634368,
      "step": 40915
    },
    {
      "epoch": 4.5016501650165015,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02859291226949532,
      "loss": 0.2277,
      "num_input_tokens_seen": 8635392,
      "step": 40920
    },
    {
      "epoch": 4.502200220022003,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.028592303268581577,
      "loss": 0.232,
      "num_input_tokens_seen": 8636448,
      "step": 40925
    },
    {
      "epoch": 4.502750275027503,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.028591694142394097,
      "loss": 0.2314,
      "num_input_tokens_seen": 8637536,
      "step": 40930
    },
    {
      "epoch": 4.503300330033003,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.028591084890938492,
      "loss": 0.2283,
      "num_input_tokens_seen": 8638624,
      "step": 40935
    },
    {
      "epoch": 4.503850385038504,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.028590475514220382,
      "loss": 0.2335,
      "num_input_tokens_seen": 8639648,
      "step": 40940
    },
    {
      "epoch": 4.504400440044004,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028589866012245384,
      "loss": 0.233,
      "num_input_tokens_seen": 8640672,
      "step": 40945
    },
    {
      "epoch": 4.5049504950495045,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.028589256385019107,
      "loss": 0.2314,
      "num_input_tokens_seen": 8641792,
      "step": 40950
    },
    {
      "epoch": 4.505500550055006,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.028588646632547177,
      "loss": 0.233,
      "num_input_tokens_seen": 8642880,
      "step": 40955
    },
    {
      "epoch": 4.506050605060506,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.028588036754835205,
      "loss": 0.2351,
      "num_input_tokens_seen": 8644000,
      "step": 40960
    },
    {
      "epoch": 4.506600660066007,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.028587426751888825,
      "loss": 0.2314,
      "num_input_tokens_seen": 8645024,
      "step": 40965
    },
    {
      "epoch": 4.507150715071507,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.02858681662371365,
      "loss": 0.2309,
      "num_input_tokens_seen": 8646112,
      "step": 40970
    },
    {
      "epoch": 4.507700770077007,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02858620637031531,
      "loss": 0.2303,
      "num_input_tokens_seen": 8647200,
      "step": 40975
    },
    {
      "epoch": 4.508250825082508,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.028585595991699422,
      "loss": 0.2309,
      "num_input_tokens_seen": 8648256,
      "step": 40980
    },
    {
      "epoch": 4.508800880088009,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02858498548787161,
      "loss": 0.2319,
      "num_input_tokens_seen": 8649280,
      "step": 40985
    },
    {
      "epoch": 4.50935093509351,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02858437485883751,
      "loss": 0.2303,
      "num_input_tokens_seen": 8650304,
      "step": 40990
    },
    {
      "epoch": 4.50990099009901,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.028583764104602744,
      "loss": 0.234,
      "num_input_tokens_seen": 8651360,
      "step": 40995
    },
    {
      "epoch": 4.51045104510451,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.028583153225172946,
      "loss": 0.2314,
      "num_input_tokens_seen": 8652416,
      "step": 41000
    },
    {
      "epoch": 4.511001100110011,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.028582542220553737,
      "loss": 0.2304,
      "num_input_tokens_seen": 8653440,
      "step": 41005
    },
    {
      "epoch": 4.511551155115511,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.028581931090750762,
      "loss": 0.2298,
      "num_input_tokens_seen": 8654464,
      "step": 41010
    },
    {
      "epoch": 4.512101210121012,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.02858131983576964,
      "loss": 0.233,
      "num_input_tokens_seen": 8655552,
      "step": 41015
    },
    {
      "epoch": 4.512651265126513,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02858070845561601,
      "loss": 0.2324,
      "num_input_tokens_seen": 8656608,
      "step": 41020
    },
    {
      "epoch": 4.513201320132013,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.028580096950295508,
      "loss": 0.2298,
      "num_input_tokens_seen": 8657600,
      "step": 41025
    },
    {
      "epoch": 4.513751375137514,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.028579485319813764,
      "loss": 0.2308,
      "num_input_tokens_seen": 8658624,
      "step": 41030
    },
    {
      "epoch": 4.514301430143014,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.02857887356417643,
      "loss": 0.2329,
      "num_input_tokens_seen": 8659648,
      "step": 41035
    },
    {
      "epoch": 4.514851485148515,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.028578261683389127,
      "loss": 0.2324,
      "num_input_tokens_seen": 8660608,
      "step": 41040
    },
    {
      "epoch": 4.5154015401540155,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.0285776496774575,
      "loss": 0.2324,
      "num_input_tokens_seen": 8661632,
      "step": 41045
    },
    {
      "epoch": 4.515951595159516,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.028577037546387195,
      "loss": 0.2303,
      "num_input_tokens_seen": 8662720,
      "step": 41050
    },
    {
      "epoch": 4.516501650165017,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.028576425290183848,
      "loss": 0.2313,
      "num_input_tokens_seen": 8663712,
      "step": 41055
    },
    {
      "epoch": 4.517051705170517,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.028575812908853103,
      "loss": 0.2329,
      "num_input_tokens_seen": 8664768,
      "step": 41060
    },
    {
      "epoch": 4.517601760176017,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.028575200402400608,
      "loss": 0.2308,
      "num_input_tokens_seen": 8665760,
      "step": 41065
    },
    {
      "epoch": 4.518151815181518,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.028574587770832,
      "loss": 0.2314,
      "num_input_tokens_seen": 8666816,
      "step": 41070
    },
    {
      "epoch": 4.5187018701870185,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.028573975014152934,
      "loss": 0.2303,
      "num_input_tokens_seen": 8667904,
      "step": 41075
    },
    {
      "epoch": 4.51925192519252,
      "grad_norm": 0.00074005126953125,
      "learning_rate": 0.028573362132369055,
      "loss": 0.2319,
      "num_input_tokens_seen": 8668960,
      "step": 41080
    },
    {
      "epoch": 4.51980198019802,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.028572749125486005,
      "loss": 0.2304,
      "num_input_tokens_seen": 8670016,
      "step": 41085
    },
    {
      "epoch": 4.52035203520352,
      "grad_norm": 0.000720977783203125,
      "learning_rate": 0.028572135993509443,
      "loss": 0.2314,
      "num_input_tokens_seen": 8671072,
      "step": 41090
    },
    {
      "epoch": 4.520902090209021,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.028571522736445018,
      "loss": 0.2308,
      "num_input_tokens_seen": 8672096,
      "step": 41095
    },
    {
      "epoch": 4.521452145214521,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.028570909354298377,
      "loss": 0.2308,
      "num_input_tokens_seen": 8673152,
      "step": 41100
    },
    {
      "epoch": 4.522002200220022,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.028570295847075176,
      "loss": 0.2319,
      "num_input_tokens_seen": 8674240,
      "step": 41105
    },
    {
      "epoch": 4.522552255225523,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.02856968221478107,
      "loss": 0.2319,
      "num_input_tokens_seen": 8675264,
      "step": 41110
    },
    {
      "epoch": 4.523102310231023,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.028569068457421718,
      "loss": 0.2293,
      "num_input_tokens_seen": 8676288,
      "step": 41115
    },
    {
      "epoch": 4.523652365236524,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.02856845457500277,
      "loss": 0.2329,
      "num_input_tokens_seen": 8677408,
      "step": 41120
    },
    {
      "epoch": 4.524202420242024,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.028567840567529888,
      "loss": 0.2319,
      "num_input_tokens_seen": 8678496,
      "step": 41125
    },
    {
      "epoch": 4.524752475247524,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.028567226435008728,
      "loss": 0.2293,
      "num_input_tokens_seen": 8679552,
      "step": 41130
    },
    {
      "epoch": 4.525302530253025,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.028566612177444953,
      "loss": 0.2303,
      "num_input_tokens_seen": 8680608,
      "step": 41135
    },
    {
      "epoch": 4.525852585258526,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.028565997794844224,
      "loss": 0.2303,
      "num_input_tokens_seen": 8681696,
      "step": 41140
    },
    {
      "epoch": 4.526402640264027,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.028565383287212205,
      "loss": 0.2329,
      "num_input_tokens_seen": 8682720,
      "step": 41145
    },
    {
      "epoch": 4.526952695269527,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.028564768654554557,
      "loss": 0.2293,
      "num_input_tokens_seen": 8683808,
      "step": 41150
    },
    {
      "epoch": 4.527502750275027,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.02856415389687694,
      "loss": 0.2345,
      "num_input_tokens_seen": 8684864,
      "step": 41155
    },
    {
      "epoch": 4.528052805280528,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.02856353901418503,
      "loss": 0.2293,
      "num_input_tokens_seen": 8685888,
      "step": 41160
    },
    {
      "epoch": 4.528602860286028,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02856292400648449,
      "loss": 0.2293,
      "num_input_tokens_seen": 8686944,
      "step": 41165
    },
    {
      "epoch": 4.5291529152915295,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.028562308873780983,
      "loss": 0.2308,
      "num_input_tokens_seen": 8687936,
      "step": 41170
    },
    {
      "epoch": 4.52970297029703,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02856169361608019,
      "loss": 0.2321,
      "num_input_tokens_seen": 8689024,
      "step": 41175
    },
    {
      "epoch": 4.53025302530253,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.028561078233387764,
      "loss": 0.2319,
      "num_input_tokens_seen": 8690048,
      "step": 41180
    },
    {
      "epoch": 4.530803080308031,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.028560462725709397,
      "loss": 0.2298,
      "num_input_tokens_seen": 8691104,
      "step": 41185
    },
    {
      "epoch": 4.531353135313531,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.028559847093050747,
      "loss": 0.2314,
      "num_input_tokens_seen": 8692224,
      "step": 41190
    },
    {
      "epoch": 4.531903190319031,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.028559231335417497,
      "loss": 0.2308,
      "num_input_tokens_seen": 8693184,
      "step": 41195
    },
    {
      "epoch": 4.5324532453245325,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.028558615452815315,
      "loss": 0.2329,
      "num_input_tokens_seen": 8694240,
      "step": 41200
    },
    {
      "epoch": 4.533003300330033,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.02855799944524988,
      "loss": 0.2298,
      "num_input_tokens_seen": 8695360,
      "step": 41205
    },
    {
      "epoch": 4.533553355335534,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.02855738331272687,
      "loss": 0.2308,
      "num_input_tokens_seen": 8696352,
      "step": 41210
    },
    {
      "epoch": 4.534103410341034,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.028556767055251966,
      "loss": 0.2324,
      "num_input_tokens_seen": 8697376,
      "step": 41215
    },
    {
      "epoch": 4.534653465346535,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.028556150672830842,
      "loss": 0.2314,
      "num_input_tokens_seen": 8698464,
      "step": 41220
    },
    {
      "epoch": 4.535203520352035,
      "grad_norm": 0.0006866455078125,
      "learning_rate": 0.028555534165469184,
      "loss": 0.2324,
      "num_input_tokens_seen": 8699520,
      "step": 41225
    },
    {
      "epoch": 4.5357535753575355,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.028554917533172672,
      "loss": 0.2308,
      "num_input_tokens_seen": 8700512,
      "step": 41230
    },
    {
      "epoch": 4.536303630363037,
      "grad_norm": 0.0150146484375,
      "learning_rate": 0.028554300775946992,
      "loss": 0.2334,
      "num_input_tokens_seen": 8701536,
      "step": 41235
    },
    {
      "epoch": 4.536853685368537,
      "grad_norm": 0.007598876953125,
      "learning_rate": 0.02855368389379782,
      "loss": 0.2303,
      "num_input_tokens_seen": 8702688,
      "step": 41240
    },
    {
      "epoch": 4.537403740374037,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.028553066886730854,
      "loss": 0.2319,
      "num_input_tokens_seen": 8703712,
      "step": 41245
    },
    {
      "epoch": 4.537953795379538,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.028552449754751765,
      "loss": 0.2324,
      "num_input_tokens_seen": 8704704,
      "step": 41250
    },
    {
      "epoch": 4.538503850385038,
      "grad_norm": 0.0074462890625,
      "learning_rate": 0.028551832497866254,
      "loss": 0.2303,
      "num_input_tokens_seen": 8705760,
      "step": 41255
    },
    {
      "epoch": 4.539053905390539,
      "grad_norm": 0.002716064453125,
      "learning_rate": 0.02855121511608001,
      "loss": 0.2319,
      "num_input_tokens_seen": 8706784,
      "step": 41260
    },
    {
      "epoch": 4.53960396039604,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.028550597609398716,
      "loss": 0.2329,
      "num_input_tokens_seen": 8707840,
      "step": 41265
    },
    {
      "epoch": 4.54015401540154,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.028549979977828065,
      "loss": 0.2313,
      "num_input_tokens_seen": 8708928,
      "step": 41270
    },
    {
      "epoch": 4.540704070407041,
      "grad_norm": 0.00372314453125,
      "learning_rate": 0.02854936222137375,
      "loss": 0.2303,
      "num_input_tokens_seen": 8709952,
      "step": 41275
    },
    {
      "epoch": 4.541254125412541,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.028548744340041465,
      "loss": 0.2324,
      "num_input_tokens_seen": 8711008,
      "step": 41280
    },
    {
      "epoch": 4.541804180418042,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.028548126333836906,
      "loss": 0.3231,
      "num_input_tokens_seen": 8712064,
      "step": 41285
    },
    {
      "epoch": 4.542354235423542,
      "grad_norm": 0.00347900390625,
      "learning_rate": 0.02854750820276577,
      "loss": 0.2319,
      "num_input_tokens_seen": 8713152,
      "step": 41290
    },
    {
      "epoch": 4.542904290429043,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.028546889946833746,
      "loss": 0.2326,
      "num_input_tokens_seen": 8714272,
      "step": 41295
    },
    {
      "epoch": 4.543454345434544,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02854627156604654,
      "loss": 0.2313,
      "num_input_tokens_seen": 8715296,
      "step": 41300
    },
    {
      "epoch": 4.544004400440044,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.028545653060409848,
      "loss": 0.2296,
      "num_input_tokens_seen": 8716320,
      "step": 41305
    },
    {
      "epoch": 4.544554455445544,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.028545034429929377,
      "loss": 0.2275,
      "num_input_tokens_seen": 8717344,
      "step": 41310
    },
    {
      "epoch": 4.545104510451045,
      "grad_norm": 0.0076904296875,
      "learning_rate": 0.02854441567461082,
      "loss": 0.2298,
      "num_input_tokens_seen": 8718368,
      "step": 41315
    },
    {
      "epoch": 4.5456545654565454,
      "grad_norm": 0.0166015625,
      "learning_rate": 0.028543796794459883,
      "loss": 0.2362,
      "num_input_tokens_seen": 8719456,
      "step": 41320
    },
    {
      "epoch": 4.5462046204620465,
      "grad_norm": 0.0164794921875,
      "learning_rate": 0.02854317778948227,
      "loss": 0.2344,
      "num_input_tokens_seen": 8720480,
      "step": 41325
    },
    {
      "epoch": 4.546754675467547,
      "grad_norm": 0.01544189453125,
      "learning_rate": 0.028542558659683682,
      "loss": 0.2299,
      "num_input_tokens_seen": 8721600,
      "step": 41330
    },
    {
      "epoch": 4.547304730473047,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.028541939405069834,
      "loss": 0.2278,
      "num_input_tokens_seen": 8722656,
      "step": 41335
    },
    {
      "epoch": 4.547854785478548,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.028541320025646424,
      "loss": 0.2299,
      "num_input_tokens_seen": 8723744,
      "step": 41340
    },
    {
      "epoch": 4.548404840484048,
      "grad_norm": 0.0078125,
      "learning_rate": 0.028540700521419174,
      "loss": 0.2341,
      "num_input_tokens_seen": 8724800,
      "step": 41345
    },
    {
      "epoch": 4.548954895489549,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.028540080892393776,
      "loss": 0.23,
      "num_input_tokens_seen": 8725856,
      "step": 41350
    },
    {
      "epoch": 4.5495049504950495,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.02853946113857595,
      "loss": 0.2264,
      "num_input_tokens_seen": 8726848,
      "step": 41355
    },
    {
      "epoch": 4.55005500550055,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.02853884125997141,
      "loss": 0.229,
      "num_input_tokens_seen": 8727872,
      "step": 41360
    },
    {
      "epoch": 4.550605060506051,
      "grad_norm": 0.01190185546875,
      "learning_rate": 0.028538221256585866,
      "loss": 0.2366,
      "num_input_tokens_seen": 8728992,
      "step": 41365
    },
    {
      "epoch": 4.551155115511551,
      "grad_norm": 0.003814697265625,
      "learning_rate": 0.028537601128425037,
      "loss": 0.2297,
      "num_input_tokens_seen": 8730016,
      "step": 41370
    },
    {
      "epoch": 4.551705170517051,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.028536980875494628,
      "loss": 0.2303,
      "num_input_tokens_seen": 8731072,
      "step": 41375
    },
    {
      "epoch": 4.552255225522552,
      "grad_norm": 0.015625,
      "learning_rate": 0.028536360497800364,
      "loss": 0.2339,
      "num_input_tokens_seen": 8732128,
      "step": 41380
    },
    {
      "epoch": 4.552805280528053,
      "grad_norm": 0.0040283203125,
      "learning_rate": 0.028535739995347965,
      "loss": 0.2322,
      "num_input_tokens_seen": 8733216,
      "step": 41385
    },
    {
      "epoch": 4.553355335533554,
      "grad_norm": 0.0150146484375,
      "learning_rate": 0.028535119368143142,
      "loss": 0.2337,
      "num_input_tokens_seen": 8734272,
      "step": 41390
    },
    {
      "epoch": 4.553905390539054,
      "grad_norm": 0.002838134765625,
      "learning_rate": 0.028534498616191617,
      "loss": 0.2309,
      "num_input_tokens_seen": 8735360,
      "step": 41395
    },
    {
      "epoch": 4.554455445544555,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.028533877739499117,
      "loss": 0.2314,
      "num_input_tokens_seen": 8736480,
      "step": 41400
    },
    {
      "epoch": 4.555005500550055,
      "grad_norm": 0.0130615234375,
      "learning_rate": 0.028533256738071357,
      "loss": 0.2287,
      "num_input_tokens_seen": 8737536,
      "step": 41405
    },
    {
      "epoch": 4.555555555555555,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.02853263561191406,
      "loss": 0.234,
      "num_input_tokens_seen": 8738560,
      "step": 41410
    },
    {
      "epoch": 4.5561056105610565,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.028532014361032962,
      "loss": 0.2303,
      "num_input_tokens_seen": 8739584,
      "step": 41415
    },
    {
      "epoch": 4.556655665566557,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.028531392985433775,
      "loss": 0.2292,
      "num_input_tokens_seen": 8740672,
      "step": 41420
    },
    {
      "epoch": 4.557205720572057,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.028530771485122234,
      "loss": 0.2334,
      "num_input_tokens_seen": 8741696,
      "step": 41425
    },
    {
      "epoch": 4.557755775577558,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.028530149860104066,
      "loss": 0.2318,
      "num_input_tokens_seen": 8742752,
      "step": 41430
    },
    {
      "epoch": 4.558305830583058,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.028529528110384998,
      "loss": 0.2297,
      "num_input_tokens_seen": 8743872,
      "step": 41435
    },
    {
      "epoch": 4.558855885588558,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.02852890623597076,
      "loss": 0.2313,
      "num_input_tokens_seen": 8744960,
      "step": 41440
    },
    {
      "epoch": 4.5594059405940595,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.028528284236867086,
      "loss": 0.2329,
      "num_input_tokens_seen": 8746048,
      "step": 41445
    },
    {
      "epoch": 4.55995599559956,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.02852766211307971,
      "loss": 0.2308,
      "num_input_tokens_seen": 8747136,
      "step": 41450
    },
    {
      "epoch": 4.560506050605061,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.028527039864614363,
      "loss": 0.2303,
      "num_input_tokens_seen": 8748256,
      "step": 41455
    },
    {
      "epoch": 4.561056105610561,
      "grad_norm": 0.007568359375,
      "learning_rate": 0.028526417491476778,
      "loss": 0.2293,
      "num_input_tokens_seen": 8749312,
      "step": 41460
    },
    {
      "epoch": 4.561606160616062,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.028525794993672693,
      "loss": 0.234,
      "num_input_tokens_seen": 8750400,
      "step": 41465
    },
    {
      "epoch": 4.562156215621562,
      "grad_norm": 0.0140380859375,
      "learning_rate": 0.028525172371207847,
      "loss": 0.2321,
      "num_input_tokens_seen": 8751392,
      "step": 41470
    },
    {
      "epoch": 4.5627062706270625,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.02852454962408798,
      "loss": 0.2319,
      "num_input_tokens_seen": 8752480,
      "step": 41475
    },
    {
      "epoch": 4.563256325632564,
      "grad_norm": 0.008056640625,
      "learning_rate": 0.028523926752318827,
      "loss": 0.2293,
      "num_input_tokens_seen": 8753504,
      "step": 41480
    },
    {
      "epoch": 4.563806380638064,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.02852330375590613,
      "loss": 0.234,
      "num_input_tokens_seen": 8754528,
      "step": 41485
    },
    {
      "epoch": 4.564356435643564,
      "grad_norm": 0.00860595703125,
      "learning_rate": 0.02852268063485563,
      "loss": 0.2298,
      "num_input_tokens_seen": 8755552,
      "step": 41490
    },
    {
      "epoch": 4.564906490649065,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.028522057389173076,
      "loss": 0.2314,
      "num_input_tokens_seen": 8756608,
      "step": 41495
    },
    {
      "epoch": 4.565456545654565,
      "grad_norm": 0.00750732421875,
      "learning_rate": 0.028521434018864204,
      "loss": 0.2323,
      "num_input_tokens_seen": 8757664,
      "step": 41500
    },
    {
      "epoch": 4.566006600660066,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.028520810523934764,
      "loss": 0.2303,
      "num_input_tokens_seen": 8758688,
      "step": 41505
    },
    {
      "epoch": 4.566556655665567,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.0285201869043905,
      "loss": 0.2313,
      "num_input_tokens_seen": 8759712,
      "step": 41510
    },
    {
      "epoch": 4.567106710671067,
      "grad_norm": 0.007415771484375,
      "learning_rate": 0.02851956316023716,
      "loss": 0.2308,
      "num_input_tokens_seen": 8760768,
      "step": 41515
    },
    {
      "epoch": 4.567656765676568,
      "grad_norm": 0.003509521484375,
      "learning_rate": 0.028518939291480496,
      "loss": 0.2287,
      "num_input_tokens_seen": 8761792,
      "step": 41520
    },
    {
      "epoch": 4.568206820682068,
      "grad_norm": 0.01385498046875,
      "learning_rate": 0.028518315298126255,
      "loss": 0.2283,
      "num_input_tokens_seen": 8762848,
      "step": 41525
    },
    {
      "epoch": 4.568756875687569,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.02851769118018019,
      "loss": 0.228,
      "num_input_tokens_seen": 8763872,
      "step": 41530
    },
    {
      "epoch": 4.569306930693069,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.02851706693764805,
      "loss": 0.2336,
      "num_input_tokens_seen": 8764928,
      "step": 41535
    },
    {
      "epoch": 4.56985698569857,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.028516442570535587,
      "loss": 0.2275,
      "num_input_tokens_seen": 8766016,
      "step": 41540
    },
    {
      "epoch": 4.570407040704071,
      "grad_norm": 0.017578125,
      "learning_rate": 0.028515818078848565,
      "loss": 0.2248,
      "num_input_tokens_seen": 8767104,
      "step": 41545
    },
    {
      "epoch": 4.570957095709571,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.028515193462592733,
      "loss": 0.235,
      "num_input_tokens_seen": 8768192,
      "step": 41550
    },
    {
      "epoch": 4.571507150715071,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.028514568721773843,
      "loss": 0.2369,
      "num_input_tokens_seen": 8769184,
      "step": 41555
    },
    {
      "epoch": 4.572057205720572,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.02851394385639766,
      "loss": 0.2335,
      "num_input_tokens_seen": 8770176,
      "step": 41560
    },
    {
      "epoch": 4.572607260726072,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.028513318866469944,
      "loss": 0.2286,
      "num_input_tokens_seen": 8771232,
      "step": 41565
    },
    {
      "epoch": 4.5731573157315735,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.02851269375199645,
      "loss": 0.228,
      "num_input_tokens_seen": 8772256,
      "step": 41570
    },
    {
      "epoch": 4.573707370737074,
      "grad_norm": 0.00775146484375,
      "learning_rate": 0.028512068512982942,
      "loss": 0.2318,
      "num_input_tokens_seen": 8773344,
      "step": 41575
    },
    {
      "epoch": 4.574257425742574,
      "grad_norm": 0.00823974609375,
      "learning_rate": 0.02851144314943518,
      "loss": 0.2342,
      "num_input_tokens_seen": 8774432,
      "step": 41580
    },
    {
      "epoch": 4.574807480748075,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.028510817661358934,
      "loss": 0.2347,
      "num_input_tokens_seen": 8775456,
      "step": 41585
    },
    {
      "epoch": 4.575357535753575,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.02851019204875996,
      "loss": 0.2294,
      "num_input_tokens_seen": 8776480,
      "step": 41590
    },
    {
      "epoch": 4.575907590759076,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.028509566311644034,
      "loss": 0.2336,
      "num_input_tokens_seen": 8777536,
      "step": 41595
    },
    {
      "epoch": 4.5764576457645765,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.028508940450016913,
      "loss": 0.232,
      "num_input_tokens_seen": 8778560,
      "step": 41600
    },
    {
      "epoch": 4.577007700770077,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.028508314463884375,
      "loss": 0.2351,
      "num_input_tokens_seen": 8779552,
      "step": 41605
    },
    {
      "epoch": 4.577557755775578,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02850768835325218,
      "loss": 0.2283,
      "num_input_tokens_seen": 8780672,
      "step": 41610
    },
    {
      "epoch": 4.578107810781078,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.028507062118126106,
      "loss": 0.231,
      "num_input_tokens_seen": 8781696,
      "step": 41615
    },
    {
      "epoch": 4.578657865786578,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.028506435758511917,
      "loss": 0.2294,
      "num_input_tokens_seen": 8782688,
      "step": 41620
    },
    {
      "epoch": 4.579207920792079,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.028505809274415397,
      "loss": 0.2357,
      "num_input_tokens_seen": 8783744,
      "step": 41625
    },
    {
      "epoch": 4.5797579757975795,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.02850518266584231,
      "loss": 0.232,
      "num_input_tokens_seen": 8784832,
      "step": 41630
    },
    {
      "epoch": 4.580308030803081,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.028504555932798433,
      "loss": 0.2294,
      "num_input_tokens_seen": 8785856,
      "step": 41635
    },
    {
      "epoch": 4.580858085808581,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.02850392907528955,
      "loss": 0.231,
      "num_input_tokens_seen": 8786944,
      "step": 41640
    },
    {
      "epoch": 4.581408140814082,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.028503302093321428,
      "loss": 0.2357,
      "num_input_tokens_seen": 8788000,
      "step": 41645
    },
    {
      "epoch": 4.581958195819582,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.028502674986899852,
      "loss": 0.233,
      "num_input_tokens_seen": 8789088,
      "step": 41650
    },
    {
      "epoch": 4.582508250825082,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.028502047756030602,
      "loss": 0.2314,
      "num_input_tokens_seen": 8790144,
      "step": 41655
    },
    {
      "epoch": 4.583058305830583,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.02850142040071945,
      "loss": 0.2319,
      "num_input_tokens_seen": 8791232,
      "step": 41660
    },
    {
      "epoch": 4.583608360836084,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.02850079292097219,
      "loss": 0.2319,
      "num_input_tokens_seen": 8792288,
      "step": 41665
    },
    {
      "epoch": 4.584158415841584,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.0285001653167946,
      "loss": 0.2319,
      "num_input_tokens_seen": 8793280,
      "step": 41670
    },
    {
      "epoch": 4.584708470847085,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.02849953758819246,
      "loss": 0.2324,
      "num_input_tokens_seen": 8794304,
      "step": 41675
    },
    {
      "epoch": 4.585258525852585,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.028498909735171565,
      "loss": 0.2309,
      "num_input_tokens_seen": 8795296,
      "step": 41680
    },
    {
      "epoch": 4.585808580858086,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.028498281757737692,
      "loss": 0.2309,
      "num_input_tokens_seen": 8796320,
      "step": 41685
    },
    {
      "epoch": 4.586358635863586,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.028497653655896637,
      "loss": 0.2324,
      "num_input_tokens_seen": 8797344,
      "step": 41690
    },
    {
      "epoch": 4.586908690869087,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.028497025429654183,
      "loss": 0.2329,
      "num_input_tokens_seen": 8798400,
      "step": 41695
    },
    {
      "epoch": 4.587458745874588,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.028496397079016123,
      "loss": 0.2319,
      "num_input_tokens_seen": 8799424,
      "step": 41700
    },
    {
      "epoch": 4.588008800880088,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.028495768603988246,
      "loss": 0.2293,
      "num_input_tokens_seen": 8800480,
      "step": 41705
    },
    {
      "epoch": 4.588558855885589,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.02849514000457635,
      "loss": 0.2319,
      "num_input_tokens_seen": 8801504,
      "step": 41710
    },
    {
      "epoch": 4.589108910891089,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.028494511280786217,
      "loss": 0.2314,
      "num_input_tokens_seen": 8802496,
      "step": 41715
    },
    {
      "epoch": 4.589658965896589,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02849388243262365,
      "loss": 0.2308,
      "num_input_tokens_seen": 8803488,
      "step": 41720
    },
    {
      "epoch": 4.5902090209020905,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.028493253460094452,
      "loss": 0.2309,
      "num_input_tokens_seen": 8804576,
      "step": 41725
    },
    {
      "epoch": 4.590759075907591,
      "grad_norm": 0.001953125,
      "learning_rate": 0.0284926243632044,
      "loss": 0.2308,
      "num_input_tokens_seen": 8805568,
      "step": 41730
    },
    {
      "epoch": 4.591309130913091,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.02849199514195931,
      "loss": 0.2308,
      "num_input_tokens_seen": 8806688,
      "step": 41735
    },
    {
      "epoch": 4.591859185918592,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02849136579636498,
      "loss": 0.2314,
      "num_input_tokens_seen": 8807776,
      "step": 41740
    },
    {
      "epoch": 4.592409240924092,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.028490736326427197,
      "loss": 0.2308,
      "num_input_tokens_seen": 8808864,
      "step": 41745
    },
    {
      "epoch": 4.592959295929593,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.028490106732151767,
      "loss": 0.2308,
      "num_input_tokens_seen": 8809920,
      "step": 41750
    },
    {
      "epoch": 4.5935093509350935,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.028489477013544505,
      "loss": 0.2303,
      "num_input_tokens_seen": 8810944,
      "step": 41755
    },
    {
      "epoch": 4.594059405940594,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.028488847170611207,
      "loss": 0.2309,
      "num_input_tokens_seen": 8812032,
      "step": 41760
    },
    {
      "epoch": 4.594609460946095,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.02848821720335767,
      "loss": 0.2335,
      "num_input_tokens_seen": 8813088,
      "step": 41765
    },
    {
      "epoch": 4.595159515951595,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.028487587111789715,
      "loss": 0.2309,
      "num_input_tokens_seen": 8814144,
      "step": 41770
    },
    {
      "epoch": 4.595709570957096,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.028486956895913136,
      "loss": 0.2314,
      "num_input_tokens_seen": 8815168,
      "step": 41775
    },
    {
      "epoch": 4.596259625962596,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.02848632655573375,
      "loss": 0.2325,
      "num_input_tokens_seen": 8816224,
      "step": 41780
    },
    {
      "epoch": 4.5968096809680965,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02848569609125736,
      "loss": 0.2293,
      "num_input_tokens_seen": 8817280,
      "step": 41785
    },
    {
      "epoch": 4.597359735973598,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.028485065502489786,
      "loss": 0.2294,
      "num_input_tokens_seen": 8818336,
      "step": 41790
    },
    {
      "epoch": 4.597909790979098,
      "grad_norm": 0.0029144287109375,
      "learning_rate": 0.028484434789436828,
      "loss": 0.2298,
      "num_input_tokens_seen": 8819360,
      "step": 41795
    },
    {
      "epoch": 4.598459845984598,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.028483803952104307,
      "loss": 0.2314,
      "num_input_tokens_seen": 8820384,
      "step": 41800
    },
    {
      "epoch": 4.599009900990099,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.02848317299049804,
      "loss": 0.2351,
      "num_input_tokens_seen": 8821440,
      "step": 41805
    },
    {
      "epoch": 4.599559955995599,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.02848254190462383,
      "loss": 0.2319,
      "num_input_tokens_seen": 8822528,
      "step": 41810
    },
    {
      "epoch": 4.6001100110011,
      "grad_norm": 0.00299072265625,
      "learning_rate": 0.028481910694487505,
      "loss": 0.2335,
      "num_input_tokens_seen": 8823584,
      "step": 41815
    },
    {
      "epoch": 4.600660066006601,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.028481279360094876,
      "loss": 0.2309,
      "num_input_tokens_seen": 8824640,
      "step": 41820
    },
    {
      "epoch": 4.601210121012102,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.028480647901451765,
      "loss": 0.2314,
      "num_input_tokens_seen": 8825696,
      "step": 41825
    },
    {
      "epoch": 4.601760176017602,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.02848001631856399,
      "loss": 0.2309,
      "num_input_tokens_seen": 8826784,
      "step": 41830
    },
    {
      "epoch": 4.602310231023102,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02847938461143738,
      "loss": 0.2304,
      "num_input_tokens_seen": 8827872,
      "step": 41835
    },
    {
      "epoch": 4.602860286028603,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.028478752780077738,
      "loss": 0.2356,
      "num_input_tokens_seen": 8828960,
      "step": 41840
    },
    {
      "epoch": 4.603410341034103,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.028478120824490905,
      "loss": 0.2319,
      "num_input_tokens_seen": 8829952,
      "step": 41845
    },
    {
      "epoch": 4.603960396039604,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.028477488744682707,
      "loss": 0.2314,
      "num_input_tokens_seen": 8831008,
      "step": 41850
    },
    {
      "epoch": 4.604510451045105,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.028476856540658953,
      "loss": 0.2309,
      "num_input_tokens_seen": 8832128,
      "step": 41855
    },
    {
      "epoch": 4.605060506050605,
      "grad_norm": 0.007171630859375,
      "learning_rate": 0.028476224212425478,
      "loss": 0.2314,
      "num_input_tokens_seen": 8833184,
      "step": 41860
    },
    {
      "epoch": 4.605610561056105,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.02847559175998812,
      "loss": 0.233,
      "num_input_tokens_seen": 8834272,
      "step": 41865
    },
    {
      "epoch": 4.606160616061606,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.028474959183352692,
      "loss": 0.2304,
      "num_input_tokens_seen": 8835264,
      "step": 41870
    },
    {
      "epoch": 4.606710671067106,
      "grad_norm": 0.0031585693359375,
      "learning_rate": 0.028474326482525035,
      "loss": 0.2303,
      "num_input_tokens_seen": 8836288,
      "step": 41875
    },
    {
      "epoch": 4.6072607260726075,
      "grad_norm": 0.001953125,
      "learning_rate": 0.02847369365751097,
      "loss": 0.2303,
      "num_input_tokens_seen": 8837312,
      "step": 41880
    },
    {
      "epoch": 4.607810781078108,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02847306070831634,
      "loss": 0.2324,
      "num_input_tokens_seen": 8838400,
      "step": 41885
    },
    {
      "epoch": 4.608360836083609,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.02847242763494698,
      "loss": 0.2293,
      "num_input_tokens_seen": 8839456,
      "step": 41890
    },
    {
      "epoch": 4.608910891089109,
      "grad_norm": 0.0128173828125,
      "learning_rate": 0.02847179443740871,
      "loss": 0.2305,
      "num_input_tokens_seen": 8840512,
      "step": 41895
    },
    {
      "epoch": 4.609460946094609,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02847116111570738,
      "loss": 0.2295,
      "num_input_tokens_seen": 8841504,
      "step": 41900
    },
    {
      "epoch": 4.61001100110011,
      "grad_norm": 0.00726318359375,
      "learning_rate": 0.028470527669848825,
      "loss": 0.2417,
      "num_input_tokens_seen": 8842528,
      "step": 41905
    },
    {
      "epoch": 4.6105610561056105,
      "grad_norm": 0.00762939453125,
      "learning_rate": 0.028469894099838872,
      "loss": 0.2379,
      "num_input_tokens_seen": 8843552,
      "step": 41910
    },
    {
      "epoch": 4.611111111111111,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.028469260405683372,
      "loss": 0.2314,
      "num_input_tokens_seen": 8844640,
      "step": 41915
    },
    {
      "epoch": 4.611661166116612,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.028468626587388163,
      "loss": 0.2319,
      "num_input_tokens_seen": 8845728,
      "step": 41920
    },
    {
      "epoch": 4.612211221122112,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.028467992644959084,
      "loss": 0.2319,
      "num_input_tokens_seen": 8846752,
      "step": 41925
    },
    {
      "epoch": 4.612761276127613,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.028467358578401984,
      "loss": 0.2324,
      "num_input_tokens_seen": 8847808,
      "step": 41930
    },
    {
      "epoch": 4.613311331133113,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.0284667243877227,
      "loss": 0.2308,
      "num_input_tokens_seen": 8848896,
      "step": 41935
    },
    {
      "epoch": 4.6138613861386135,
      "grad_norm": 0.003204345703125,
      "learning_rate": 0.028466090072927077,
      "loss": 0.233,
      "num_input_tokens_seen": 8849952,
      "step": 41940
    },
    {
      "epoch": 4.614411441144115,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.028465455634020965,
      "loss": 0.2345,
      "num_input_tokens_seen": 8851072,
      "step": 41945
    },
    {
      "epoch": 4.614961496149615,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.02846482107101021,
      "loss": 0.2319,
      "num_input_tokens_seen": 8852160,
      "step": 41950
    },
    {
      "epoch": 4.615511551155116,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.02846418638390066,
      "loss": 0.2298,
      "num_input_tokens_seen": 8853152,
      "step": 41955
    },
    {
      "epoch": 4.616061606160616,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.028463551572698163,
      "loss": 0.2313,
      "num_input_tokens_seen": 8854208,
      "step": 41960
    },
    {
      "epoch": 4.616611661166116,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.02846291663740857,
      "loss": 0.2313,
      "num_input_tokens_seen": 8855264,
      "step": 41965
    },
    {
      "epoch": 4.617161716171617,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.02846228157803774,
      "loss": 0.2313,
      "num_input_tokens_seen": 8856320,
      "step": 41970
    },
    {
      "epoch": 4.617711771177118,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.02846164639459152,
      "loss": 0.2329,
      "num_input_tokens_seen": 8857408,
      "step": 41975
    },
    {
      "epoch": 4.618261826182618,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02846101108707576,
      "loss": 0.2318,
      "num_input_tokens_seen": 8858368,
      "step": 41980
    },
    {
      "epoch": 4.618811881188119,
      "grad_norm": 0.002838134765625,
      "learning_rate": 0.028460375655496323,
      "loss": 0.2318,
      "num_input_tokens_seen": 8859456,
      "step": 41985
    },
    {
      "epoch": 4.619361936193619,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.028459740099859066,
      "loss": 0.2313,
      "num_input_tokens_seen": 8860576,
      "step": 41990
    },
    {
      "epoch": 4.61991199119912,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.02845910442016984,
      "loss": 0.2329,
      "num_input_tokens_seen": 8861600,
      "step": 41995
    },
    {
      "epoch": 4.62046204620462,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.028458468616434505,
      "loss": 0.2324,
      "num_input_tokens_seen": 8862624,
      "step": 42000
    },
    {
      "epoch": 4.621012101210121,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.028457832688658927,
      "loss": 0.2308,
      "num_input_tokens_seen": 8863744,
      "step": 42005
    },
    {
      "epoch": 4.621562156215622,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.02845719663684896,
      "loss": 0.234,
      "num_input_tokens_seen": 8864800,
      "step": 42010
    },
    {
      "epoch": 4.622112211221122,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.028456560461010468,
      "loss": 0.2329,
      "num_input_tokens_seen": 8865888,
      "step": 42015
    },
    {
      "epoch": 4.622662266226623,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.028455924161149317,
      "loss": 0.2324,
      "num_input_tokens_seen": 8866880,
      "step": 42020
    },
    {
      "epoch": 4.623212321232123,
      "grad_norm": 0.0027923583984375,
      "learning_rate": 0.028455287737271368,
      "loss": 0.2308,
      "num_input_tokens_seen": 8867872,
      "step": 42025
    },
    {
      "epoch": 4.623762376237623,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.028454651189382493,
      "loss": 0.2324,
      "num_input_tokens_seen": 8868928,
      "step": 42030
    },
    {
      "epoch": 4.6243124312431245,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.028454014517488548,
      "loss": 0.2303,
      "num_input_tokens_seen": 8869920,
      "step": 42035
    },
    {
      "epoch": 4.624862486248625,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02845337772159541,
      "loss": 0.2309,
      "num_input_tokens_seen": 8870976,
      "step": 42040
    },
    {
      "epoch": 4.625412541254125,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.028452740801708946,
      "loss": 0.2305,
      "num_input_tokens_seen": 8872000,
      "step": 42045
    },
    {
      "epoch": 4.625962596259626,
      "grad_norm": 0.002838134765625,
      "learning_rate": 0.028452103757835023,
      "loss": 0.228,
      "num_input_tokens_seen": 8873056,
      "step": 42050
    },
    {
      "epoch": 4.626512651265126,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.028451466589979515,
      "loss": 0.2297,
      "num_input_tokens_seen": 8874112,
      "step": 42055
    },
    {
      "epoch": 4.627062706270627,
      "grad_norm": 0.0069580078125,
      "learning_rate": 0.028450829298148295,
      "loss": 0.2336,
      "num_input_tokens_seen": 8875136,
      "step": 42060
    },
    {
      "epoch": 4.6276127612761275,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.028450191882347233,
      "loss": 0.2407,
      "num_input_tokens_seen": 8876224,
      "step": 42065
    },
    {
      "epoch": 4.628162816281629,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.028449554342582214,
      "loss": 0.2308,
      "num_input_tokens_seen": 8877312,
      "step": 42070
    },
    {
      "epoch": 4.628712871287129,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.0284489166788591,
      "loss": 0.2369,
      "num_input_tokens_seen": 8878336,
      "step": 42075
    },
    {
      "epoch": 4.629262926292629,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.028448278891183777,
      "loss": 0.2309,
      "num_input_tokens_seen": 8879360,
      "step": 42080
    },
    {
      "epoch": 4.62981298129813,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.02844764097956212,
      "loss": 0.2315,
      "num_input_tokens_seen": 8880384,
      "step": 42085
    },
    {
      "epoch": 4.63036303630363,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.028447002944000006,
      "loss": 0.2315,
      "num_input_tokens_seen": 8881408,
      "step": 42090
    },
    {
      "epoch": 4.6309130913091305,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02844636478450332,
      "loss": 0.232,
      "num_input_tokens_seen": 8882464,
      "step": 42095
    },
    {
      "epoch": 4.631463146314632,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02844572650107794,
      "loss": 0.233,
      "num_input_tokens_seen": 8883488,
      "step": 42100
    },
    {
      "epoch": 4.632013201320132,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.028445088093729753,
      "loss": 0.2304,
      "num_input_tokens_seen": 8884576,
      "step": 42105
    },
    {
      "epoch": 4.632563256325633,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.028444449562464644,
      "loss": 0.2309,
      "num_input_tokens_seen": 8885600,
      "step": 42110
    },
    {
      "epoch": 4.633113311331133,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02844381090728849,
      "loss": 0.2329,
      "num_input_tokens_seen": 8886688,
      "step": 42115
    },
    {
      "epoch": 4.633663366336633,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.028443172128207184,
      "loss": 0.2308,
      "num_input_tokens_seen": 8887712,
      "step": 42120
    },
    {
      "epoch": 4.634213421342134,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.02844253322522661,
      "loss": 0.2292,
      "num_input_tokens_seen": 8888736,
      "step": 42125
    },
    {
      "epoch": 4.634763476347635,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.02844189419835266,
      "loss": 0.2303,
      "num_input_tokens_seen": 8889760,
      "step": 42130
    },
    {
      "epoch": 4.635313531353136,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.028441255047591218,
      "loss": 0.2329,
      "num_input_tokens_seen": 8890752,
      "step": 42135
    },
    {
      "epoch": 4.635863586358636,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.028440615772948184,
      "loss": 0.2313,
      "num_input_tokens_seen": 8891840,
      "step": 42140
    },
    {
      "epoch": 4.636413641364136,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.028439976374429438,
      "loss": 0.2318,
      "num_input_tokens_seen": 8892896,
      "step": 42145
    },
    {
      "epoch": 4.636963696369637,
      "grad_norm": 0.005859375,
      "learning_rate": 0.028439336852040878,
      "loss": 0.2308,
      "num_input_tokens_seen": 8893888,
      "step": 42150
    },
    {
      "epoch": 4.637513751375137,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.0284386972057884,
      "loss": 0.2298,
      "num_input_tokens_seen": 8894944,
      "step": 42155
    },
    {
      "epoch": 4.638063806380638,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.028438057435677903,
      "loss": 0.2313,
      "num_input_tokens_seen": 8896064,
      "step": 42160
    },
    {
      "epoch": 4.638613861386139,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.028437417541715273,
      "loss": 0.2303,
      "num_input_tokens_seen": 8897152,
      "step": 42165
    },
    {
      "epoch": 4.639163916391639,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.028436777523906418,
      "loss": 0.2304,
      "num_input_tokens_seen": 8898240,
      "step": 42170
    },
    {
      "epoch": 4.63971397139714,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02843613738225723,
      "loss": 0.231,
      "num_input_tokens_seen": 8899360,
      "step": 42175
    },
    {
      "epoch": 4.64026402640264,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02843549711677361,
      "loss": 0.2338,
      "num_input_tokens_seen": 8900352,
      "step": 42180
    },
    {
      "epoch": 4.6408140814081404,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.028434856727461457,
      "loss": 0.2297,
      "num_input_tokens_seen": 8901440,
      "step": 42185
    },
    {
      "epoch": 4.6413641364136415,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.02843421621432668,
      "loss": 0.2332,
      "num_input_tokens_seen": 8902528,
      "step": 42190
    },
    {
      "epoch": 4.641914191419142,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.028433575577375178,
      "loss": 0.2302,
      "num_input_tokens_seen": 8903616,
      "step": 42195
    },
    {
      "epoch": 4.642464246424643,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.02843293481661286,
      "loss": 0.2274,
      "num_input_tokens_seen": 8904640,
      "step": 42200
    },
    {
      "epoch": 4.643014301430143,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.028432293932045626,
      "loss": 0.2391,
      "num_input_tokens_seen": 8905760,
      "step": 42205
    },
    {
      "epoch": 4.643564356435643,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02843165292367938,
      "loss": 0.2306,
      "num_input_tokens_seen": 8906816,
      "step": 42210
    },
    {
      "epoch": 4.644114411441144,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.02843101179152004,
      "loss": 0.2315,
      "num_input_tokens_seen": 8907776,
      "step": 42215
    },
    {
      "epoch": 4.6446644664466445,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.028430370535573502,
      "loss": 0.2253,
      "num_input_tokens_seen": 8908864,
      "step": 42220
    },
    {
      "epoch": 4.645214521452145,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.028429729155845688,
      "loss": 0.2239,
      "num_input_tokens_seen": 8909888,
      "step": 42225
    },
    {
      "epoch": 4.645764576457646,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.0284290876523425,
      "loss": 0.235,
      "num_input_tokens_seen": 8910880,
      "step": 42230
    },
    {
      "epoch": 4.646314631463146,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.028428446025069862,
      "loss": 0.2256,
      "num_input_tokens_seen": 8911968,
      "step": 42235
    },
    {
      "epoch": 4.646864686468647,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.02842780427403368,
      "loss": 0.2246,
      "num_input_tokens_seen": 8913056,
      "step": 42240
    },
    {
      "epoch": 4.647414741474147,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.028427162399239866,
      "loss": 0.2257,
      "num_input_tokens_seen": 8914080,
      "step": 42245
    },
    {
      "epoch": 4.647964796479648,
      "grad_norm": 0.01275634765625,
      "learning_rate": 0.02842652040069434,
      "loss": 0.2406,
      "num_input_tokens_seen": 8915136,
      "step": 42250
    },
    {
      "epoch": 4.648514851485149,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.028425878278403018,
      "loss": 0.2369,
      "num_input_tokens_seen": 8916224,
      "step": 42255
    },
    {
      "epoch": 4.649064906490649,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.028425236032371817,
      "loss": 0.2315,
      "num_input_tokens_seen": 8917280,
      "step": 42260
    },
    {
      "epoch": 4.64961496149615,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02842459366260666,
      "loss": 0.2389,
      "num_input_tokens_seen": 8918336,
      "step": 42265
    },
    {
      "epoch": 4.65016501650165,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.028423951169113462,
      "loss": 0.2308,
      "num_input_tokens_seen": 8919392,
      "step": 42270
    },
    {
      "epoch": 4.65071507150715,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.02842330855189815,
      "loss": 0.2339,
      "num_input_tokens_seen": 8920416,
      "step": 42275
    },
    {
      "epoch": 4.6512651265126514,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.028422665810966644,
      "loss": 0.235,
      "num_input_tokens_seen": 8921504,
      "step": 42280
    },
    {
      "epoch": 4.651815181518152,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.028422022946324867,
      "loss": 0.2359,
      "num_input_tokens_seen": 8922592,
      "step": 42285
    },
    {
      "epoch": 4.652365236523653,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.028421379957978745,
      "loss": 0.2301,
      "num_input_tokens_seen": 8923584,
      "step": 42290
    },
    {
      "epoch": 4.652915291529153,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.028420736845934203,
      "loss": 0.23,
      "num_input_tokens_seen": 8924736,
      "step": 42295
    },
    {
      "epoch": 4.653465346534653,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.028420093610197174,
      "loss": 0.2335,
      "num_input_tokens_seen": 8925760,
      "step": 42300
    },
    {
      "epoch": 4.654015401540154,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.028419450250773578,
      "loss": 0.2299,
      "num_input_tokens_seen": 8926720,
      "step": 42305
    },
    {
      "epoch": 4.6545654565456545,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.028418806767669346,
      "loss": 0.2309,
      "num_input_tokens_seen": 8927776,
      "step": 42310
    },
    {
      "epoch": 4.6551155115511555,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.02841816316089042,
      "loss": 0.2304,
      "num_input_tokens_seen": 8928864,
      "step": 42315
    },
    {
      "epoch": 4.655665566556656,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.02841751943044271,
      "loss": 0.2273,
      "num_input_tokens_seen": 8929984,
      "step": 42320
    },
    {
      "epoch": 4.656215621562156,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.02841687557633217,
      "loss": 0.2351,
      "num_input_tokens_seen": 8931008,
      "step": 42325
    },
    {
      "epoch": 4.656765676567657,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.028416231598564726,
      "loss": 0.2294,
      "num_input_tokens_seen": 8932064,
      "step": 42330
    },
    {
      "epoch": 4.657315731573157,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.028415587497146313,
      "loss": 0.2346,
      "num_input_tokens_seen": 8933056,
      "step": 42335
    },
    {
      "epoch": 4.6578657865786575,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.028414943272082865,
      "loss": 0.2257,
      "num_input_tokens_seen": 8934112,
      "step": 42340
    },
    {
      "epoch": 4.658415841584159,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02841429892338032,
      "loss": 0.2304,
      "num_input_tokens_seen": 8935168,
      "step": 42345
    },
    {
      "epoch": 4.658965896589659,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.028413654451044625,
      "loss": 0.233,
      "num_input_tokens_seen": 8936192,
      "step": 42350
    },
    {
      "epoch": 4.65951595159516,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.02841300985508171,
      "loss": 0.231,
      "num_input_tokens_seen": 8937248,
      "step": 42355
    },
    {
      "epoch": 4.66006600660066,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.028412365135497518,
      "loss": 0.2283,
      "num_input_tokens_seen": 8938368,
      "step": 42360
    },
    {
      "epoch": 4.66061606160616,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028411720292297993,
      "loss": 0.2268,
      "num_input_tokens_seen": 8939392,
      "step": 42365
    },
    {
      "epoch": 4.661166116611661,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02841107532548908,
      "loss": 0.2321,
      "num_input_tokens_seen": 8940480,
      "step": 42370
    },
    {
      "epoch": 4.661716171617162,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.02841043023507672,
      "loss": 0.2286,
      "num_input_tokens_seen": 8941504,
      "step": 42375
    },
    {
      "epoch": 4.662266226622663,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02840978502106686,
      "loss": 0.2301,
      "num_input_tokens_seen": 8942528,
      "step": 42380
    },
    {
      "epoch": 4.662816281628163,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.02840913968346544,
      "loss": 0.2276,
      "num_input_tokens_seen": 8943616,
      "step": 42385
    },
    {
      "epoch": 4.663366336633663,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.028408494222278424,
      "loss": 0.2277,
      "num_input_tokens_seen": 8944672,
      "step": 42390
    },
    {
      "epoch": 4.663916391639164,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.028407848637511744,
      "loss": 0.2362,
      "num_input_tokens_seen": 8945664,
      "step": 42395
    },
    {
      "epoch": 4.664466446644664,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.028407202929171357,
      "loss": 0.2388,
      "num_input_tokens_seen": 8946752,
      "step": 42400
    },
    {
      "epoch": 4.665016501650165,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.028406557097263214,
      "loss": 0.2376,
      "num_input_tokens_seen": 8947872,
      "step": 42405
    },
    {
      "epoch": 4.665566556655666,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.02840591114179327,
      "loss": 0.2281,
      "num_input_tokens_seen": 8948896,
      "step": 42410
    },
    {
      "epoch": 4.666116611661166,
      "grad_norm": 0.01202392578125,
      "learning_rate": 0.028405265062767478,
      "loss": 0.2344,
      "num_input_tokens_seen": 8949888,
      "step": 42415
    },
    {
      "epoch": 4.666666666666667,
      "grad_norm": 0.000812530517578125,
      "learning_rate": 0.028404618860191785,
      "loss": 0.2332,
      "num_input_tokens_seen": 8950944,
      "step": 42420
    },
    {
      "epoch": 4.667216721672167,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.028403972534072158,
      "loss": 0.2304,
      "num_input_tokens_seen": 8952032,
      "step": 42425
    },
    {
      "epoch": 4.667766776677668,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.028403326084414537,
      "loss": 0.2315,
      "num_input_tokens_seen": 8953056,
      "step": 42430
    },
    {
      "epoch": 4.6683168316831685,
      "grad_norm": 0.0027313232421875,
      "learning_rate": 0.028402679511224903,
      "loss": 0.233,
      "num_input_tokens_seen": 8954144,
      "step": 42435
    },
    {
      "epoch": 4.668866886688669,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.028402032814509197,
      "loss": 0.2335,
      "num_input_tokens_seen": 8955168,
      "step": 42440
    },
    {
      "epoch": 4.66941694169417,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.028401385994273387,
      "loss": 0.234,
      "num_input_tokens_seen": 8956192,
      "step": 42445
    },
    {
      "epoch": 4.66996699669967,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.02840073905052343,
      "loss": 0.2309,
      "num_input_tokens_seen": 8957216,
      "step": 42450
    },
    {
      "epoch": 4.67051705170517,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.028400091983265294,
      "loss": 0.2329,
      "num_input_tokens_seen": 8958304,
      "step": 42455
    },
    {
      "epoch": 4.671067106710671,
      "grad_norm": 0.00732421875,
      "learning_rate": 0.028399444792504943,
      "loss": 0.2313,
      "num_input_tokens_seen": 8959328,
      "step": 42460
    },
    {
      "epoch": 4.6716171617161715,
      "grad_norm": 0.00299072265625,
      "learning_rate": 0.028398797478248333,
      "loss": 0.2308,
      "num_input_tokens_seen": 8960384,
      "step": 42465
    },
    {
      "epoch": 4.672167216721672,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.028398150040501443,
      "loss": 0.2308,
      "num_input_tokens_seen": 8961472,
      "step": 42470
    },
    {
      "epoch": 4.672717271727173,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.02839750247927023,
      "loss": 0.2313,
      "num_input_tokens_seen": 8962528,
      "step": 42475
    },
    {
      "epoch": 4.673267326732673,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.028396854794560663,
      "loss": 0.2308,
      "num_input_tokens_seen": 8963616,
      "step": 42480
    },
    {
      "epoch": 4.673817381738174,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.028396206986378715,
      "loss": 0.2329,
      "num_input_tokens_seen": 8964640,
      "step": 42485
    },
    {
      "epoch": 4.674367436743674,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.02839555905473036,
      "loss": 0.2318,
      "num_input_tokens_seen": 8965760,
      "step": 42490
    },
    {
      "epoch": 4.674917491749175,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.028394910999621555,
      "loss": 0.2303,
      "num_input_tokens_seen": 8966784,
      "step": 42495
    },
    {
      "epoch": 4.675467546754676,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.028394262821058294,
      "loss": 0.2313,
      "num_input_tokens_seen": 8967872,
      "step": 42500
    },
    {
      "epoch": 4.676017601760176,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.028393614519046532,
      "loss": 0.2329,
      "num_input_tokens_seen": 8968960,
      "step": 42505
    },
    {
      "epoch": 4.676567656765677,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.028392966093592254,
      "loss": 0.2313,
      "num_input_tokens_seen": 8969952,
      "step": 42510
    },
    {
      "epoch": 4.677117711771177,
      "grad_norm": 0.01300048828125,
      "learning_rate": 0.028392317544701437,
      "loss": 0.2303,
      "num_input_tokens_seen": 8971008,
      "step": 42515
    },
    {
      "epoch": 4.677667766776677,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.028391668872380052,
      "loss": 0.2313,
      "num_input_tokens_seen": 8972096,
      "step": 42520
    },
    {
      "epoch": 4.678217821782178,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.02839102007663408,
      "loss": 0.2318,
      "num_input_tokens_seen": 8973120,
      "step": 42525
    },
    {
      "epoch": 4.678767876787679,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.028390371157469504,
      "loss": 0.2318,
      "num_input_tokens_seen": 8974240,
      "step": 42530
    },
    {
      "epoch": 4.67931793179318,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.028389722114892303,
      "loss": 0.2339,
      "num_input_tokens_seen": 8975296,
      "step": 42535
    },
    {
      "epoch": 4.67986798679868,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.028389072948908457,
      "loss": 0.2313,
      "num_input_tokens_seen": 8976256,
      "step": 42540
    },
    {
      "epoch": 4.68041804180418,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.028388423659523952,
      "loss": 0.2323,
      "num_input_tokens_seen": 8977312,
      "step": 42545
    },
    {
      "epoch": 4.680968096809681,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.028387774246744768,
      "loss": 0.2297,
      "num_input_tokens_seen": 8978336,
      "step": 42550
    },
    {
      "epoch": 4.681518151815181,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.028387124710576896,
      "loss": 0.2303,
      "num_input_tokens_seen": 8979360,
      "step": 42555
    },
    {
      "epoch": 4.6820682068206825,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.028386475051026313,
      "loss": 0.2293,
      "num_input_tokens_seen": 8980416,
      "step": 42560
    },
    {
      "epoch": 4.682618261826183,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02838582526809902,
      "loss": 0.2314,
      "num_input_tokens_seen": 8981440,
      "step": 42565
    },
    {
      "epoch": 4.683168316831683,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.028385175361800994,
      "loss": 0.2273,
      "num_input_tokens_seen": 8982528,
      "step": 42570
    },
    {
      "epoch": 4.683718371837184,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.02838452533213823,
      "loss": 0.2312,
      "num_input_tokens_seen": 8983584,
      "step": 42575
    },
    {
      "epoch": 4.684268426842684,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02838387517911672,
      "loss": 0.226,
      "num_input_tokens_seen": 8984704,
      "step": 42580
    },
    {
      "epoch": 4.684818481848184,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.028383224902742452,
      "loss": 0.2329,
      "num_input_tokens_seen": 8985728,
      "step": 42585
    },
    {
      "epoch": 4.6853685368536855,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.028382574503021427,
      "loss": 0.2296,
      "num_input_tokens_seen": 8986784,
      "step": 42590
    },
    {
      "epoch": 4.685918591859186,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.028381923979959634,
      "loss": 0.236,
      "num_input_tokens_seen": 8987840,
      "step": 42595
    },
    {
      "epoch": 4.686468646864687,
      "grad_norm": 0.0026397705078125,
      "learning_rate": 0.028381273333563068,
      "loss": 0.2313,
      "num_input_tokens_seen": 8988896,
      "step": 42600
    },
    {
      "epoch": 4.687018701870187,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.028380622563837724,
      "loss": 0.2334,
      "num_input_tokens_seen": 8989984,
      "step": 42605
    },
    {
      "epoch": 4.687568756875687,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.028379971670789605,
      "loss": 0.2366,
      "num_input_tokens_seen": 8991008,
      "step": 42610
    },
    {
      "epoch": 4.688118811881188,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.028379320654424705,
      "loss": 0.2355,
      "num_input_tokens_seen": 8992000,
      "step": 42615
    },
    {
      "epoch": 4.6886688668866885,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.028378669514749027,
      "loss": 0.2275,
      "num_input_tokens_seen": 8993056,
      "step": 42620
    },
    {
      "epoch": 4.68921892189219,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.028378018251768573,
      "loss": 0.2321,
      "num_input_tokens_seen": 8994112,
      "step": 42625
    },
    {
      "epoch": 4.68976897689769,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.028377366865489344,
      "loss": 0.2336,
      "num_input_tokens_seen": 8995168,
      "step": 42630
    },
    {
      "epoch": 4.69031903190319,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.028376715355917344,
      "loss": 0.2373,
      "num_input_tokens_seen": 8996192,
      "step": 42635
    },
    {
      "epoch": 4.690869086908691,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.028376063723058575,
      "loss": 0.2325,
      "num_input_tokens_seen": 8997248,
      "step": 42640
    },
    {
      "epoch": 4.691419141914191,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.028375411966919048,
      "loss": 0.2319,
      "num_input_tokens_seen": 8998304,
      "step": 42645
    },
    {
      "epoch": 4.6919691969196915,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.028374760087504766,
      "loss": 0.2309,
      "num_input_tokens_seen": 8999328,
      "step": 42650
    },
    {
      "epoch": 4.692519251925193,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.02837410808482174,
      "loss": 0.2324,
      "num_input_tokens_seen": 9000352,
      "step": 42655
    },
    {
      "epoch": 4.693069306930693,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.028373455958875974,
      "loss": 0.2314,
      "num_input_tokens_seen": 9001440,
      "step": 42660
    },
    {
      "epoch": 4.693619361936194,
      "grad_norm": 0.0006103515625,
      "learning_rate": 0.028372803709673485,
      "loss": 0.2303,
      "num_input_tokens_seen": 9002464,
      "step": 42665
    },
    {
      "epoch": 4.694169416941694,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.028372151337220278,
      "loss": 0.2308,
      "num_input_tokens_seen": 9003552,
      "step": 42670
    },
    {
      "epoch": 4.694719471947195,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.028371498841522368,
      "loss": 0.2319,
      "num_input_tokens_seen": 9004608,
      "step": 42675
    },
    {
      "epoch": 4.695269526952695,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.028370846222585774,
      "loss": 0.2314,
      "num_input_tokens_seen": 9005728,
      "step": 42680
    },
    {
      "epoch": 4.695819581958196,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0283701934804165,
      "loss": 0.2303,
      "num_input_tokens_seen": 9006848,
      "step": 42685
    },
    {
      "epoch": 4.696369636963697,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028369540615020575,
      "loss": 0.2303,
      "num_input_tokens_seen": 9007872,
      "step": 42690
    },
    {
      "epoch": 4.696919691969197,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.028368887626404007,
      "loss": 0.2314,
      "num_input_tokens_seen": 9008864,
      "step": 42695
    },
    {
      "epoch": 4.697469746974697,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.028368234514572814,
      "loss": 0.2298,
      "num_input_tokens_seen": 9009920,
      "step": 42700
    },
    {
      "epoch": 4.698019801980198,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.02836758127953302,
      "loss": 0.2308,
      "num_input_tokens_seen": 9010944,
      "step": 42705
    },
    {
      "epoch": 4.698569856985698,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.028366927921290645,
      "loss": 0.2303,
      "num_input_tokens_seen": 9012032,
      "step": 42710
    },
    {
      "epoch": 4.6991199119911995,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.02836627443985171,
      "loss": 0.234,
      "num_input_tokens_seen": 9013184,
      "step": 42715
    },
    {
      "epoch": 4.6996699669967,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.028365620835222236,
      "loss": 0.2309,
      "num_input_tokens_seen": 9014240,
      "step": 42720
    },
    {
      "epoch": 4.7002200220022,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.02836496710740825,
      "loss": 0.2319,
      "num_input_tokens_seen": 9015328,
      "step": 42725
    },
    {
      "epoch": 4.700770077007701,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.028364313256415773,
      "loss": 0.2319,
      "num_input_tokens_seen": 9016352,
      "step": 42730
    },
    {
      "epoch": 4.701320132013201,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.028363659282250833,
      "loss": 0.2345,
      "num_input_tokens_seen": 9017408,
      "step": 42735
    },
    {
      "epoch": 4.701870187018702,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.02836300518491946,
      "loss": 0.2314,
      "num_input_tokens_seen": 9018432,
      "step": 42740
    },
    {
      "epoch": 4.7024202420242025,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.028362350964427678,
      "loss": 0.2313,
      "num_input_tokens_seen": 9019456,
      "step": 42745
    },
    {
      "epoch": 4.702970297029703,
      "grad_norm": 0.00072479248046875,
      "learning_rate": 0.02836169662078152,
      "loss": 0.2324,
      "num_input_tokens_seen": 9020480,
      "step": 42750
    },
    {
      "epoch": 4.703520352035204,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.028361042153987014,
      "loss": 0.2319,
      "num_input_tokens_seen": 9021504,
      "step": 42755
    },
    {
      "epoch": 4.704070407040704,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0283603875640502,
      "loss": 0.233,
      "num_input_tokens_seen": 9022496,
      "step": 42760
    },
    {
      "epoch": 4.704620462046204,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.028359732850977096,
      "loss": 0.2314,
      "num_input_tokens_seen": 9023520,
      "step": 42765
    },
    {
      "epoch": 4.705170517051705,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02835907801477375,
      "loss": 0.2313,
      "num_input_tokens_seen": 9024512,
      "step": 42770
    },
    {
      "epoch": 4.7057205720572055,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.028358423055446193,
      "loss": 0.2329,
      "num_input_tokens_seen": 9025568,
      "step": 42775
    },
    {
      "epoch": 4.706270627062707,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02835776797300046,
      "loss": 0.2335,
      "num_input_tokens_seen": 9026656,
      "step": 42780
    },
    {
      "epoch": 4.706820682068207,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.028357112767442585,
      "loss": 0.2314,
      "num_input_tokens_seen": 9027744,
      "step": 42785
    },
    {
      "epoch": 4.707370737073707,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.028356457438778616,
      "loss": 0.2308,
      "num_input_tokens_seen": 9028768,
      "step": 42790
    },
    {
      "epoch": 4.707920792079208,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.028355801987014588,
      "loss": 0.2308,
      "num_input_tokens_seen": 9029792,
      "step": 42795
    },
    {
      "epoch": 4.708470847084708,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02835514641215654,
      "loss": 0.2319,
      "num_input_tokens_seen": 9030848,
      "step": 42800
    },
    {
      "epoch": 4.709020902090209,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.028354490714210517,
      "loss": 0.2314,
      "num_input_tokens_seen": 9031904,
      "step": 42805
    },
    {
      "epoch": 4.70957095709571,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.028353834893182564,
      "loss": 0.2325,
      "num_input_tokens_seen": 9032992,
      "step": 42810
    },
    {
      "epoch": 4.71012101210121,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.028353178949078715,
      "loss": 0.233,
      "num_input_tokens_seen": 9034048,
      "step": 42815
    },
    {
      "epoch": 4.710671067106711,
      "grad_norm": 0.0007476806640625,
      "learning_rate": 0.028352522881905026,
      "loss": 0.2289,
      "num_input_tokens_seen": 9035072,
      "step": 42820
    },
    {
      "epoch": 4.711221122112211,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.028351866691667543,
      "loss": 0.2299,
      "num_input_tokens_seen": 9036128,
      "step": 42825
    },
    {
      "epoch": 4.711771177117711,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02835121037837231,
      "loss": 0.2335,
      "num_input_tokens_seen": 9037152,
      "step": 42830
    },
    {
      "epoch": 4.712321232123212,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.02835055394202538,
      "loss": 0.2304,
      "num_input_tokens_seen": 9038208,
      "step": 42835
    },
    {
      "epoch": 4.712871287128713,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.028349897382632796,
      "loss": 0.2346,
      "num_input_tokens_seen": 9039264,
      "step": 42840
    },
    {
      "epoch": 4.713421342134214,
      "grad_norm": 0.005859375,
      "learning_rate": 0.028349240700200617,
      "loss": 0.2335,
      "num_input_tokens_seen": 9040288,
      "step": 42845
    },
    {
      "epoch": 4.713971397139714,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02834858389473489,
      "loss": 0.233,
      "num_input_tokens_seen": 9041344,
      "step": 42850
    },
    {
      "epoch": 4.714521452145215,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028347926966241672,
      "loss": 0.2324,
      "num_input_tokens_seen": 9042368,
      "step": 42855
    },
    {
      "epoch": 4.715071507150715,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.028347269914727018,
      "loss": 0.2298,
      "num_input_tokens_seen": 9043424,
      "step": 42860
    },
    {
      "epoch": 4.715621562156215,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02834661274019698,
      "loss": 0.2314,
      "num_input_tokens_seen": 9044448,
      "step": 42865
    },
    {
      "epoch": 4.7161716171617165,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028345955442657614,
      "loss": 0.2329,
      "num_input_tokens_seen": 9045504,
      "step": 42870
    },
    {
      "epoch": 4.716721672167217,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.02834529802211499,
      "loss": 0.2319,
      "num_input_tokens_seen": 9046560,
      "step": 42875
    },
    {
      "epoch": 4.717271727172717,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.028344640478575142,
      "loss": 0.2314,
      "num_input_tokens_seen": 9047648,
      "step": 42880
    },
    {
      "epoch": 4.717821782178218,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.028343982812044158,
      "loss": 0.2314,
      "num_input_tokens_seen": 9048672,
      "step": 42885
    },
    {
      "epoch": 4.718371837183718,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.028343325022528086,
      "loss": 0.2298,
      "num_input_tokens_seen": 9049728,
      "step": 42890
    },
    {
      "epoch": 4.718921892189218,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.028342667110032988,
      "loss": 0.2324,
      "num_input_tokens_seen": 9050784,
      "step": 42895
    },
    {
      "epoch": 4.7194719471947195,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.028342009074564932,
      "loss": 0.2313,
      "num_input_tokens_seen": 9051872,
      "step": 42900
    },
    {
      "epoch": 4.72002200220022,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.028341350916129975,
      "loss": 0.2314,
      "num_input_tokens_seen": 9052864,
      "step": 42905
    },
    {
      "epoch": 4.720572057205721,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.028340692634734196,
      "loss": 0.2314,
      "num_input_tokens_seen": 9053856,
      "step": 42910
    },
    {
      "epoch": 4.721122112211221,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02834003423038365,
      "loss": 0.2303,
      "num_input_tokens_seen": 9054912,
      "step": 42915
    },
    {
      "epoch": 4.721672167216722,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.02833937570308441,
      "loss": 0.2329,
      "num_input_tokens_seen": 9056000,
      "step": 42920
    },
    {
      "epoch": 4.722222222222222,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.028338717052842546,
      "loss": 0.2356,
      "num_input_tokens_seen": 9057088,
      "step": 42925
    },
    {
      "epoch": 4.7227722772277225,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.028338058279664124,
      "loss": 0.2329,
      "num_input_tokens_seen": 9058144,
      "step": 42930
    },
    {
      "epoch": 4.723322332233224,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.02833739938355522,
      "loss": 0.2314,
      "num_input_tokens_seen": 9059264,
      "step": 42935
    },
    {
      "epoch": 4.723872387238724,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.028336740364521912,
      "loss": 0.2309,
      "num_input_tokens_seen": 9060256,
      "step": 42940
    },
    {
      "epoch": 4.724422442244224,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.028336081222570263,
      "loss": 0.233,
      "num_input_tokens_seen": 9061344,
      "step": 42945
    },
    {
      "epoch": 4.724972497249725,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02833542195770635,
      "loss": 0.2324,
      "num_input_tokens_seen": 9062336,
      "step": 42950
    },
    {
      "epoch": 4.725522552255225,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028334762569936257,
      "loss": 0.2303,
      "num_input_tokens_seen": 9063328,
      "step": 42955
    },
    {
      "epoch": 4.726072607260726,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.028334103059266053,
      "loss": 0.2324,
      "num_input_tokens_seen": 9064384,
      "step": 42960
    },
    {
      "epoch": 4.726622662266227,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.02833344342570182,
      "loss": 0.2314,
      "num_input_tokens_seen": 9065408,
      "step": 42965
    },
    {
      "epoch": 4.727172717271727,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.028332783669249635,
      "loss": 0.2314,
      "num_input_tokens_seen": 9066432,
      "step": 42970
    },
    {
      "epoch": 4.727722772277228,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.028332123789915585,
      "loss": 0.2319,
      "num_input_tokens_seen": 9067520,
      "step": 42975
    },
    {
      "epoch": 4.728272827282728,
      "grad_norm": 0.000720977783203125,
      "learning_rate": 0.028331463787705742,
      "loss": 0.2308,
      "num_input_tokens_seen": 9068608,
      "step": 42980
    },
    {
      "epoch": 4.728822882288229,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.028330803662626202,
      "loss": 0.2309,
      "num_input_tokens_seen": 9069760,
      "step": 42985
    },
    {
      "epoch": 4.729372937293729,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.028330143414683037,
      "loss": 0.2314,
      "num_input_tokens_seen": 9070880,
      "step": 42990
    },
    {
      "epoch": 4.72992299229923,
      "grad_norm": 0.000629425048828125,
      "learning_rate": 0.028329483043882338,
      "loss": 0.2314,
      "num_input_tokens_seen": 9071968,
      "step": 42995
    },
    {
      "epoch": 4.730473047304731,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.028328822550230184,
      "loss": 0.2314,
      "num_input_tokens_seen": 9072960,
      "step": 43000
    },
    {
      "epoch": 4.731023102310231,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.028328161933732672,
      "loss": 0.2319,
      "num_input_tokens_seen": 9074048,
      "step": 43005
    },
    {
      "epoch": 4.731573157315731,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02832750119439589,
      "loss": 0.2319,
      "num_input_tokens_seen": 9075072,
      "step": 43010
    },
    {
      "epoch": 4.732123212321232,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02832684033222592,
      "loss": 0.2308,
      "num_input_tokens_seen": 9076096,
      "step": 43015
    },
    {
      "epoch": 4.732673267326732,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.02832617934722886,
      "loss": 0.2314,
      "num_input_tokens_seen": 9077184,
      "step": 43020
    },
    {
      "epoch": 4.7332233223322335,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0283255182394108,
      "loss": 0.2298,
      "num_input_tokens_seen": 9078240,
      "step": 43025
    },
    {
      "epoch": 4.733773377337734,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.028324857008777835,
      "loss": 0.2324,
      "num_input_tokens_seen": 9079200,
      "step": 43030
    },
    {
      "epoch": 4.734323432343234,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02832419565533605,
      "loss": 0.2293,
      "num_input_tokens_seen": 9080288,
      "step": 43035
    },
    {
      "epoch": 4.734873487348735,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.02832353417909155,
      "loss": 0.2351,
      "num_input_tokens_seen": 9081408,
      "step": 43040
    },
    {
      "epoch": 4.735423542354235,
      "grad_norm": 0.005859375,
      "learning_rate": 0.028322872580050434,
      "loss": 0.2304,
      "num_input_tokens_seen": 9082496,
      "step": 43045
    },
    {
      "epoch": 4.735973597359736,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.028322210858218785,
      "loss": 0.2303,
      "num_input_tokens_seen": 9083584,
      "step": 43050
    },
    {
      "epoch": 4.7365236523652365,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028321549013602718,
      "loss": 0.2314,
      "num_input_tokens_seen": 9084576,
      "step": 43055
    },
    {
      "epoch": 4.737073707370737,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.028320887046208325,
      "loss": 0.2304,
      "num_input_tokens_seen": 9085632,
      "step": 43060
    },
    {
      "epoch": 4.737623762376238,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028320224956041706,
      "loss": 0.2309,
      "num_input_tokens_seen": 9086656,
      "step": 43065
    },
    {
      "epoch": 4.738173817381738,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.028319562743108966,
      "loss": 0.2309,
      "num_input_tokens_seen": 9087776,
      "step": 43070
    },
    {
      "epoch": 4.738723872387238,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02831890040741621,
      "loss": 0.2325,
      "num_input_tokens_seen": 9088832,
      "step": 43075
    },
    {
      "epoch": 4.739273927392739,
      "grad_norm": 0.002838134765625,
      "learning_rate": 0.028318237948969534,
      "loss": 0.2314,
      "num_input_tokens_seen": 9089952,
      "step": 43080
    },
    {
      "epoch": 4.7398239823982395,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.028317575367775055,
      "loss": 0.2361,
      "num_input_tokens_seen": 9090976,
      "step": 43085
    },
    {
      "epoch": 4.740374037403741,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.028316912663838876,
      "loss": 0.2309,
      "num_input_tokens_seen": 9092032,
      "step": 43090
    },
    {
      "epoch": 4.740924092409241,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.028316249837167098,
      "loss": 0.2319,
      "num_input_tokens_seen": 9093024,
      "step": 43095
    },
    {
      "epoch": 4.741474147414742,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028315586887765837,
      "loss": 0.2309,
      "num_input_tokens_seen": 9094144,
      "step": 43100
    },
    {
      "epoch": 4.742024202420242,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0283149238156412,
      "loss": 0.233,
      "num_input_tokens_seen": 9095200,
      "step": 43105
    },
    {
      "epoch": 4.742574257425742,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.028314260620799298,
      "loss": 0.2308,
      "num_input_tokens_seen": 9096288,
      "step": 43110
    },
    {
      "epoch": 4.743124312431243,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028313597303246247,
      "loss": 0.2319,
      "num_input_tokens_seen": 9097312,
      "step": 43115
    },
    {
      "epoch": 4.743674367436744,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02831293386298816,
      "loss": 0.2309,
      "num_input_tokens_seen": 9098368,
      "step": 43120
    },
    {
      "epoch": 4.744224422442244,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02831227030003115,
      "loss": 0.233,
      "num_input_tokens_seen": 9099328,
      "step": 43125
    },
    {
      "epoch": 4.744774477447745,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.028311606614381327,
      "loss": 0.233,
      "num_input_tokens_seen": 9100320,
      "step": 43130
    },
    {
      "epoch": 4.745324532453245,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.028310942806044816,
      "loss": 0.2288,
      "num_input_tokens_seen": 9101312,
      "step": 43135
    },
    {
      "epoch": 4.745874587458746,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.028310278875027736,
      "loss": 0.2319,
      "num_input_tokens_seen": 9102368,
      "step": 43140
    },
    {
      "epoch": 4.7464246424642464,
      "grad_norm": 0.00079345703125,
      "learning_rate": 0.028309614821336197,
      "loss": 0.2324,
      "num_input_tokens_seen": 9103424,
      "step": 43145
    },
    {
      "epoch": 4.746974697469747,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02830895064497633,
      "loss": 0.2324,
      "num_input_tokens_seen": 9104512,
      "step": 43150
    },
    {
      "epoch": 4.747524752475248,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.028308286345954246,
      "loss": 0.2308,
      "num_input_tokens_seen": 9105600,
      "step": 43155
    },
    {
      "epoch": 4.748074807480748,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.028307621924276077,
      "loss": 0.2303,
      "num_input_tokens_seen": 9106688,
      "step": 43160
    },
    {
      "epoch": 4.748624862486249,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02830695737994794,
      "loss": 0.2324,
      "num_input_tokens_seen": 9107808,
      "step": 43165
    },
    {
      "epoch": 4.749174917491749,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.028306292712975963,
      "loss": 0.2324,
      "num_input_tokens_seen": 9108864,
      "step": 43170
    },
    {
      "epoch": 4.7497249724972495,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02830562792336627,
      "loss": 0.2319,
      "num_input_tokens_seen": 9109920,
      "step": 43175
    },
    {
      "epoch": 4.7502750275027505,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02830496301112499,
      "loss": 0.2319,
      "num_input_tokens_seen": 9110944,
      "step": 43180
    },
    {
      "epoch": 4.750825082508251,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.028304297976258247,
      "loss": 0.2324,
      "num_input_tokens_seen": 9112000,
      "step": 43185
    },
    {
      "epoch": 4.751375137513751,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.028303632818772177,
      "loss": 0.232,
      "num_input_tokens_seen": 9112992,
      "step": 43190
    },
    {
      "epoch": 4.751925192519252,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.028302967538672907,
      "loss": 0.233,
      "num_input_tokens_seen": 9114048,
      "step": 43195
    },
    {
      "epoch": 4.752475247524752,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.02830230213596657,
      "loss": 0.2335,
      "num_input_tokens_seen": 9115232,
      "step": 43200
    },
    {
      "epoch": 4.753025302530253,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028301636610659296,
      "loss": 0.2324,
      "num_input_tokens_seen": 9116288,
      "step": 43205
    },
    {
      "epoch": 4.7535753575357536,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.028300970962757216,
      "loss": 0.2268,
      "num_input_tokens_seen": 9117344,
      "step": 43210
    },
    {
      "epoch": 4.754125412541254,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.02830030519226647,
      "loss": 0.2304,
      "num_input_tokens_seen": 9118432,
      "step": 43215
    },
    {
      "epoch": 4.754675467546755,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.0282996392991932,
      "loss": 0.2294,
      "num_input_tokens_seen": 9119552,
      "step": 43220
    },
    {
      "epoch": 4.755225522552255,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.028298973283543534,
      "loss": 0.2372,
      "num_input_tokens_seen": 9120576,
      "step": 43225
    },
    {
      "epoch": 4.755775577557756,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.02829830714532361,
      "loss": 0.2319,
      "num_input_tokens_seen": 9121664,
      "step": 43230
    },
    {
      "epoch": 4.756325632563256,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.028297640884539572,
      "loss": 0.2308,
      "num_input_tokens_seen": 9122688,
      "step": 43235
    },
    {
      "epoch": 4.756875687568757,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.028296974501197558,
      "loss": 0.2313,
      "num_input_tokens_seen": 9123744,
      "step": 43240
    },
    {
      "epoch": 4.757425742574258,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028296307995303707,
      "loss": 0.2313,
      "num_input_tokens_seen": 9124768,
      "step": 43245
    },
    {
      "epoch": 4.757975797579758,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.028295641366864167,
      "loss": 0.2319,
      "num_input_tokens_seen": 9125888,
      "step": 43250
    },
    {
      "epoch": 4.758525852585258,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.028294974615885083,
      "loss": 0.2319,
      "num_input_tokens_seen": 9126880,
      "step": 43255
    },
    {
      "epoch": 4.759075907590759,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.0282943077423726,
      "loss": 0.2308,
      "num_input_tokens_seen": 9127904,
      "step": 43260
    },
    {
      "epoch": 4.759625962596259,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.02829364074633286,
      "loss": 0.2308,
      "num_input_tokens_seen": 9129024,
      "step": 43265
    },
    {
      "epoch": 4.7601760176017605,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.028292973627772007,
      "loss": 0.2303,
      "num_input_tokens_seen": 9130080,
      "step": 43270
    },
    {
      "epoch": 4.760726072607261,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0282923063866962,
      "loss": 0.2324,
      "num_input_tokens_seen": 9131104,
      "step": 43275
    },
    {
      "epoch": 4.761276127612762,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02829163902311158,
      "loss": 0.2303,
      "num_input_tokens_seen": 9132160,
      "step": 43280
    },
    {
      "epoch": 4.761826182618262,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.0282909715370243,
      "loss": 0.2319,
      "num_input_tokens_seen": 9133216,
      "step": 43285
    },
    {
      "epoch": 4.762376237623762,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.028290303928440515,
      "loss": 0.2324,
      "num_input_tokens_seen": 9134272,
      "step": 43290
    },
    {
      "epoch": 4.762926292629263,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02828963619736638,
      "loss": 0.2361,
      "num_input_tokens_seen": 9135296,
      "step": 43295
    },
    {
      "epoch": 4.7634763476347635,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02828896834380804,
      "loss": 0.2324,
      "num_input_tokens_seen": 9136320,
      "step": 43300
    },
    {
      "epoch": 4.764026402640264,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.028288300367771656,
      "loss": 0.2314,
      "num_input_tokens_seen": 9137408,
      "step": 43305
    },
    {
      "epoch": 4.764576457645765,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.028287632269263386,
      "loss": 0.2319,
      "num_input_tokens_seen": 9138432,
      "step": 43310
    },
    {
      "epoch": 4.765126512651265,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.028286964048289384,
      "loss": 0.2289,
      "num_input_tokens_seen": 9139552,
      "step": 43315
    },
    {
      "epoch": 4.765676567656766,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02828629570485581,
      "loss": 0.233,
      "num_input_tokens_seen": 9140576,
      "step": 43320
    },
    {
      "epoch": 4.766226622662266,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.028285627238968825,
      "loss": 0.2321,
      "num_input_tokens_seen": 9141632,
      "step": 43325
    },
    {
      "epoch": 4.7667766776677665,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.02828495865063459,
      "loss": 0.23,
      "num_input_tokens_seen": 9142688,
      "step": 43330
    },
    {
      "epoch": 4.767326732673268,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.028284289939859264,
      "loss": 0.2311,
      "num_input_tokens_seen": 9143776,
      "step": 43335
    },
    {
      "epoch": 4.767876787678768,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.02828362110664901,
      "loss": 0.231,
      "num_input_tokens_seen": 9144864,
      "step": 43340
    },
    {
      "epoch": 4.768426842684269,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.028282952151009996,
      "loss": 0.2336,
      "num_input_tokens_seen": 9145920,
      "step": 43345
    },
    {
      "epoch": 4.768976897689769,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02828228307294839,
      "loss": 0.2305,
      "num_input_tokens_seen": 9146912,
      "step": 43350
    },
    {
      "epoch": 4.769526952695269,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.028281613872470347,
      "loss": 0.231,
      "num_input_tokens_seen": 9147968,
      "step": 43355
    },
    {
      "epoch": 4.77007700770077,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02828094454958205,
      "loss": 0.2326,
      "num_input_tokens_seen": 9149024,
      "step": 43360
    },
    {
      "epoch": 4.770627062706271,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02828027510428965,
      "loss": 0.2315,
      "num_input_tokens_seen": 9150048,
      "step": 43365
    },
    {
      "epoch": 4.771177117711771,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.028279605536599334,
      "loss": 0.231,
      "num_input_tokens_seen": 9151104,
      "step": 43370
    },
    {
      "epoch": 4.771727172717272,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.028278935846517264,
      "loss": 0.2304,
      "num_input_tokens_seen": 9152160,
      "step": 43375
    },
    {
      "epoch": 4.772277227722772,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.028278266034049614,
      "loss": 0.2347,
      "num_input_tokens_seen": 9153152,
      "step": 43380
    },
    {
      "epoch": 4.772827282728273,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02827759609920256,
      "loss": 0.2331,
      "num_input_tokens_seen": 9154240,
      "step": 43385
    },
    {
      "epoch": 4.773377337733773,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.028276926041982273,
      "loss": 0.2325,
      "num_input_tokens_seen": 9155296,
      "step": 43390
    },
    {
      "epoch": 4.773927392739274,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028276255862394926,
      "loss": 0.2325,
      "num_input_tokens_seen": 9156384,
      "step": 43395
    },
    {
      "epoch": 4.774477447744775,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.028275585560446702,
      "loss": 0.2283,
      "num_input_tokens_seen": 9157472,
      "step": 43400
    },
    {
      "epoch": 4.775027502750275,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.028274915136143778,
      "loss": 0.233,
      "num_input_tokens_seen": 9158560,
      "step": 43405
    },
    {
      "epoch": 4.775577557755776,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.02827424458949233,
      "loss": 0.2294,
      "num_input_tokens_seen": 9159680,
      "step": 43410
    },
    {
      "epoch": 4.776127612761276,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.028273573920498535,
      "loss": 0.231,
      "num_input_tokens_seen": 9160736,
      "step": 43415
    },
    {
      "epoch": 4.776677667766776,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.028272903129168584,
      "loss": 0.2331,
      "num_input_tokens_seen": 9161792,
      "step": 43420
    },
    {
      "epoch": 4.7772277227722775,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02827223221550865,
      "loss": 0.23,
      "num_input_tokens_seen": 9162816,
      "step": 43425
    },
    {
      "epoch": 4.777777777777778,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.028271561179524923,
      "loss": 0.2316,
      "num_input_tokens_seen": 9163936,
      "step": 43430
    },
    {
      "epoch": 4.778327832783278,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.028270890021223585,
      "loss": 0.2263,
      "num_input_tokens_seen": 9165056,
      "step": 43435
    },
    {
      "epoch": 4.778877887788779,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.028270218740610822,
      "loss": 0.2316,
      "num_input_tokens_seen": 9166144,
      "step": 43440
    },
    {
      "epoch": 4.779427942794279,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.028269547337692814,
      "loss": 0.2285,
      "num_input_tokens_seen": 9167200,
      "step": 43445
    },
    {
      "epoch": 4.77997799779978,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.028268875812475763,
      "loss": 0.2321,
      "num_input_tokens_seen": 9168256,
      "step": 43450
    },
    {
      "epoch": 4.7805280528052805,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02826820416496585,
      "loss": 0.2277,
      "num_input_tokens_seen": 9169312,
      "step": 43455
    },
    {
      "epoch": 4.781078107810782,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.02826753239516926,
      "loss": 0.2266,
      "num_input_tokens_seen": 9170464,
      "step": 43460
    },
    {
      "epoch": 4.781628162816282,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.028266860503092198,
      "loss": 0.2304,
      "num_input_tokens_seen": 9171488,
      "step": 43465
    },
    {
      "epoch": 4.782178217821782,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.028266188488740844,
      "loss": 0.2278,
      "num_input_tokens_seen": 9172512,
      "step": 43470
    },
    {
      "epoch": 4.782728272827283,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.028265516352121395,
      "loss": 0.2294,
      "num_input_tokens_seen": 9173504,
      "step": 43475
    },
    {
      "epoch": 4.783278327832783,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.028264844093240052,
      "loss": 0.226,
      "num_input_tokens_seen": 9174624,
      "step": 43480
    },
    {
      "epoch": 4.7838283828382835,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.028264171712103003,
      "loss": 0.2421,
      "num_input_tokens_seen": 9175584,
      "step": 43485
    },
    {
      "epoch": 4.784378437843785,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.028263499208716448,
      "loss": 0.2397,
      "num_input_tokens_seen": 9176672,
      "step": 43490
    },
    {
      "epoch": 4.784928492849285,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.028262826583086587,
      "loss": 0.2299,
      "num_input_tokens_seen": 9177760,
      "step": 43495
    },
    {
      "epoch": 4.785478547854785,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.028262153835219617,
      "loss": 0.2324,
      "num_input_tokens_seen": 9178816,
      "step": 43500
    },
    {
      "epoch": 4.786028602860286,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02826148096512174,
      "loss": 0.2292,
      "num_input_tokens_seen": 9179840,
      "step": 43505
    },
    {
      "epoch": 4.786578657865786,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.028260807972799153,
      "loss": 0.2282,
      "num_input_tokens_seen": 9180896,
      "step": 43510
    },
    {
      "epoch": 4.787128712871287,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.028260134858258065,
      "loss": 0.2292,
      "num_input_tokens_seen": 9181984,
      "step": 43515
    },
    {
      "epoch": 4.787678767876788,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.028259461621504677,
      "loss": 0.2272,
      "num_input_tokens_seen": 9183008,
      "step": 43520
    },
    {
      "epoch": 4.788228822882289,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02825878826254519,
      "loss": 0.2292,
      "num_input_tokens_seen": 9184000,
      "step": 43525
    },
    {
      "epoch": 4.788778877887789,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02825811478138582,
      "loss": 0.2345,
      "num_input_tokens_seen": 9185024,
      "step": 43530
    },
    {
      "epoch": 4.789328932893289,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.028257441178032763,
      "loss": 0.2313,
      "num_input_tokens_seen": 9186112,
      "step": 43535
    },
    {
      "epoch": 4.78987898789879,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.028256767452492233,
      "loss": 0.2362,
      "num_input_tokens_seen": 9187136,
      "step": 43540
    },
    {
      "epoch": 4.79042904290429,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.028256093604770438,
      "loss": 0.2325,
      "num_input_tokens_seen": 9188128,
      "step": 43545
    },
    {
      "epoch": 4.790979097909791,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.028255419634873595,
      "loss": 0.2339,
      "num_input_tokens_seen": 9189216,
      "step": 43550
    },
    {
      "epoch": 4.791529152915292,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0282547455428079,
      "loss": 0.2286,
      "num_input_tokens_seen": 9190176,
      "step": 43555
    },
    {
      "epoch": 4.792079207920792,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.028254071328579582,
      "loss": 0.226,
      "num_input_tokens_seen": 9191232,
      "step": 43560
    },
    {
      "epoch": 4.792629262926293,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02825339699219485,
      "loss": 0.2354,
      "num_input_tokens_seen": 9192288,
      "step": 43565
    },
    {
      "epoch": 4.793179317931793,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.028252722533659916,
      "loss": 0.2234,
      "num_input_tokens_seen": 9193344,
      "step": 43570
    },
    {
      "epoch": 4.793729372937293,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.028252047952980995,
      "loss": 0.2324,
      "num_input_tokens_seen": 9194400,
      "step": 43575
    },
    {
      "epoch": 4.7942794279427945,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.02825137325016431,
      "loss": 0.2381,
      "num_input_tokens_seen": 9195456,
      "step": 43580
    },
    {
      "epoch": 4.794829482948295,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.028250698425216072,
      "loss": 0.2406,
      "num_input_tokens_seen": 9196480,
      "step": 43585
    },
    {
      "epoch": 4.795379537953796,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.028250023478142507,
      "loss": 0.2315,
      "num_input_tokens_seen": 9197536,
      "step": 43590
    },
    {
      "epoch": 4.795929592959296,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.028249348408949836,
      "loss": 0.2304,
      "num_input_tokens_seen": 9198624,
      "step": 43595
    },
    {
      "epoch": 4.796479647964796,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.028248673217644276,
      "loss": 0.2324,
      "num_input_tokens_seen": 9199648,
      "step": 43600
    },
    {
      "epoch": 4.797029702970297,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.028247997904232054,
      "loss": 0.2298,
      "num_input_tokens_seen": 9200768,
      "step": 43605
    },
    {
      "epoch": 4.7975797579757975,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.028247322468719388,
      "loss": 0.2313,
      "num_input_tokens_seen": 9201824,
      "step": 43610
    },
    {
      "epoch": 4.798129812981298,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.028246646911112512,
      "loss": 0.2324,
      "num_input_tokens_seen": 9202880,
      "step": 43615
    },
    {
      "epoch": 4.798679867986799,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.02824597123141765,
      "loss": 0.2329,
      "num_input_tokens_seen": 9203936,
      "step": 43620
    },
    {
      "epoch": 4.799229922992299,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.028245295429641024,
      "loss": 0.2308,
      "num_input_tokens_seen": 9205088,
      "step": 43625
    },
    {
      "epoch": 4.7997799779978,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02824461950578886,
      "loss": 0.2308,
      "num_input_tokens_seen": 9206144,
      "step": 43630
    },
    {
      "epoch": 4.8003300330033,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0282439434598674,
      "loss": 0.2313,
      "num_input_tokens_seen": 9207232,
      "step": 43635
    },
    {
      "epoch": 4.8008800880088005,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028243267291882868,
      "loss": 0.2313,
      "num_input_tokens_seen": 9208320,
      "step": 43640
    },
    {
      "epoch": 4.801430143014302,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.028242591001841497,
      "loss": 0.2318,
      "num_input_tokens_seen": 9209440,
      "step": 43645
    },
    {
      "epoch": 4.801980198019802,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.028241914589749513,
      "loss": 0.234,
      "num_input_tokens_seen": 9210464,
      "step": 43650
    },
    {
      "epoch": 4.802530253025303,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028241238055613162,
      "loss": 0.2303,
      "num_input_tokens_seen": 9211584,
      "step": 43655
    },
    {
      "epoch": 4.803080308030803,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.028240561399438677,
      "loss": 0.2329,
      "num_input_tokens_seen": 9212704,
      "step": 43660
    },
    {
      "epoch": 4.803630363036303,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.028239884621232284,
      "loss": 0.2329,
      "num_input_tokens_seen": 9213760,
      "step": 43665
    },
    {
      "epoch": 4.804180418041804,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.028239207721000227,
      "loss": 0.2324,
      "num_input_tokens_seen": 9214880,
      "step": 43670
    },
    {
      "epoch": 4.804730473047305,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.02823853069874875,
      "loss": 0.2324,
      "num_input_tokens_seen": 9215872,
      "step": 43675
    },
    {
      "epoch": 4.805280528052805,
      "grad_norm": 0.000614166259765625,
      "learning_rate": 0.028237853554484087,
      "loss": 0.2318,
      "num_input_tokens_seen": 9216928,
      "step": 43680
    },
    {
      "epoch": 4.805830583058306,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.028237176288212477,
      "loss": 0.2318,
      "num_input_tokens_seen": 9218016,
      "step": 43685
    },
    {
      "epoch": 4.806380638063806,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.028236498899940166,
      "loss": 0.2303,
      "num_input_tokens_seen": 9219040,
      "step": 43690
    },
    {
      "epoch": 4.806930693069307,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.0282358213896734,
      "loss": 0.2303,
      "num_input_tokens_seen": 9220128,
      "step": 43695
    },
    {
      "epoch": 4.807480748074807,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.028235143757418417,
      "loss": 0.233,
      "num_input_tokens_seen": 9221120,
      "step": 43700
    },
    {
      "epoch": 4.8080308030803085,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.028234466003181462,
      "loss": 0.234,
      "num_input_tokens_seen": 9222112,
      "step": 43705
    },
    {
      "epoch": 4.808580858085809,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.02823378812696879,
      "loss": 0.2325,
      "num_input_tokens_seen": 9223168,
      "step": 43710
    },
    {
      "epoch": 4.809130913091309,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.028233110128786636,
      "loss": 0.2309,
      "num_input_tokens_seen": 9224256,
      "step": 43715
    },
    {
      "epoch": 4.80968096809681,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02823243200864126,
      "loss": 0.2314,
      "num_input_tokens_seen": 9225312,
      "step": 43720
    },
    {
      "epoch": 4.81023102310231,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.028231753766538907,
      "loss": 0.2319,
      "num_input_tokens_seen": 9226368,
      "step": 43725
    },
    {
      "epoch": 4.81078107810781,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.02823107540248583,
      "loss": 0.2325,
      "num_input_tokens_seen": 9227424,
      "step": 43730
    },
    {
      "epoch": 4.8113311331133115,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.02823039691648828,
      "loss": 0.233,
      "num_input_tokens_seen": 9228416,
      "step": 43735
    },
    {
      "epoch": 4.811881188118812,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.028229718308552507,
      "loss": 0.2319,
      "num_input_tokens_seen": 9229472,
      "step": 43740
    },
    {
      "epoch": 4.812431243124313,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.02822903957868477,
      "loss": 0.233,
      "num_input_tokens_seen": 9230496,
      "step": 43745
    },
    {
      "epoch": 4.812981298129813,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.028228360726891326,
      "loss": 0.2314,
      "num_input_tokens_seen": 9231520,
      "step": 43750
    },
    {
      "epoch": 4.813531353135313,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.028227681753178427,
      "loss": 0.2293,
      "num_input_tokens_seen": 9232576,
      "step": 43755
    },
    {
      "epoch": 4.814081408140814,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.028227002657552333,
      "loss": 0.2308,
      "num_input_tokens_seen": 9233632,
      "step": 43760
    },
    {
      "epoch": 4.8146314631463145,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.028226323440019303,
      "loss": 0.2308,
      "num_input_tokens_seen": 9234688,
      "step": 43765
    },
    {
      "epoch": 4.815181518151816,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.028225644100585593,
      "loss": 0.2329,
      "num_input_tokens_seen": 9235712,
      "step": 43770
    },
    {
      "epoch": 4.815731573157316,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.028224964639257468,
      "loss": 0.2319,
      "num_input_tokens_seen": 9236832,
      "step": 43775
    },
    {
      "epoch": 4.816281628162816,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.028224285056041194,
      "loss": 0.2298,
      "num_input_tokens_seen": 9237920,
      "step": 43780
    },
    {
      "epoch": 4.816831683168317,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.028223605350943023,
      "loss": 0.2299,
      "num_input_tokens_seen": 9239008,
      "step": 43785
    },
    {
      "epoch": 4.817381738173817,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.028222925523969232,
      "loss": 0.2347,
      "num_input_tokens_seen": 9240032,
      "step": 43790
    },
    {
      "epoch": 4.8179317931793175,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.028222245575126082,
      "loss": 0.229,
      "num_input_tokens_seen": 9241120,
      "step": 43795
    },
    {
      "epoch": 4.818481848184819,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.02822156550441984,
      "loss": 0.2368,
      "num_input_tokens_seen": 9242176,
      "step": 43800
    },
    {
      "epoch": 4.819031903190319,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.02822088531185677,
      "loss": 0.23,
      "num_input_tokens_seen": 9243264,
      "step": 43805
    },
    {
      "epoch": 4.81958195819582,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.02822020499744314,
      "loss": 0.229,
      "num_input_tokens_seen": 9244352,
      "step": 43810
    },
    {
      "epoch": 4.82013201320132,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.028219524561185233,
      "loss": 0.2346,
      "num_input_tokens_seen": 9245408,
      "step": 43815
    },
    {
      "epoch": 4.82068206820682,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.02821884400308931,
      "loss": 0.2314,
      "num_input_tokens_seen": 9246400,
      "step": 43820
    },
    {
      "epoch": 4.821232123212321,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.028218163323161644,
      "loss": 0.2325,
      "num_input_tokens_seen": 9247424,
      "step": 43825
    },
    {
      "epoch": 4.821782178217822,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.028217482521408505,
      "loss": 0.2298,
      "num_input_tokens_seen": 9248448,
      "step": 43830
    },
    {
      "epoch": 4.822332233223323,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.028216801597836176,
      "loss": 0.2314,
      "num_input_tokens_seen": 9249504,
      "step": 43835
    },
    {
      "epoch": 4.822882288228823,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.028216120552450926,
      "loss": 0.2293,
      "num_input_tokens_seen": 9250560,
      "step": 43840
    },
    {
      "epoch": 4.823432343234323,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.02821543938525904,
      "loss": 0.2345,
      "num_input_tokens_seen": 9251648,
      "step": 43845
    },
    {
      "epoch": 4.823982398239824,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.028214758096266786,
      "loss": 0.2314,
      "num_input_tokens_seen": 9252704,
      "step": 43850
    },
    {
      "epoch": 4.824532453245324,
      "grad_norm": 0.000728607177734375,
      "learning_rate": 0.028214076685480448,
      "loss": 0.2324,
      "num_input_tokens_seen": 9253728,
      "step": 43855
    },
    {
      "epoch": 4.825082508250825,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.028213395152906304,
      "loss": 0.2319,
      "num_input_tokens_seen": 9254752,
      "step": 43860
    },
    {
      "epoch": 4.825632563256326,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.028212713498550646,
      "loss": 0.2314,
      "num_input_tokens_seen": 9255840,
      "step": 43865
    },
    {
      "epoch": 4.826182618261826,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.028212031722419737,
      "loss": 0.2303,
      "num_input_tokens_seen": 9256960,
      "step": 43870
    },
    {
      "epoch": 4.826732673267327,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.028211349824519876,
      "loss": 0.2324,
      "num_input_tokens_seen": 9257920,
      "step": 43875
    },
    {
      "epoch": 4.827282728272827,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.028210667804857342,
      "loss": 0.2309,
      "num_input_tokens_seen": 9259040,
      "step": 43880
    },
    {
      "epoch": 4.827832783278328,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.028209985663438426,
      "loss": 0.2324,
      "num_input_tokens_seen": 9260096,
      "step": 43885
    },
    {
      "epoch": 4.8283828382838285,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.028209303400269406,
      "loss": 0.2329,
      "num_input_tokens_seen": 9261152,
      "step": 43890
    },
    {
      "epoch": 4.828932893289329,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02820862101535658,
      "loss": 0.2313,
      "num_input_tokens_seen": 9262176,
      "step": 43895
    },
    {
      "epoch": 4.82948294829483,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.028207938508706225,
      "loss": 0.2303,
      "num_input_tokens_seen": 9263296,
      "step": 43900
    },
    {
      "epoch": 4.83003300330033,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.028207255880324647,
      "loss": 0.2329,
      "num_input_tokens_seen": 9264320,
      "step": 43905
    },
    {
      "epoch": 4.83058305830583,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.02820657313021812,
      "loss": 0.2298,
      "num_input_tokens_seen": 9265344,
      "step": 43910
    },
    {
      "epoch": 4.831133113311331,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.02820589025839295,
      "loss": 0.2303,
      "num_input_tokens_seen": 9266400,
      "step": 43915
    },
    {
      "epoch": 4.8316831683168315,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.02820520726485543,
      "loss": 0.2324,
      "num_input_tokens_seen": 9267520,
      "step": 43920
    },
    {
      "epoch": 4.832233223322332,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.028204524149611845,
      "loss": 0.2324,
      "num_input_tokens_seen": 9268544,
      "step": 43925
    },
    {
      "epoch": 4.832783278327833,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.0282038409126685,
      "loss": 0.2303,
      "num_input_tokens_seen": 9269568,
      "step": 43930
    },
    {
      "epoch": 4.833333333333333,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.02820315755403169,
      "loss": 0.2287,
      "num_input_tokens_seen": 9270560,
      "step": 43935
    },
    {
      "epoch": 4.833883388338834,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.028202474073707712,
      "loss": 0.233,
      "num_input_tokens_seen": 9271616,
      "step": 43940
    },
    {
      "epoch": 4.834433443344334,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.028201790471702865,
      "loss": 0.2314,
      "num_input_tokens_seen": 9272640,
      "step": 43945
    },
    {
      "epoch": 4.834983498349835,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02820110674802345,
      "loss": 0.2303,
      "num_input_tokens_seen": 9273728,
      "step": 43950
    },
    {
      "epoch": 4.835533553355336,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02820042290267577,
      "loss": 0.2293,
      "num_input_tokens_seen": 9274848,
      "step": 43955
    },
    {
      "epoch": 4.836083608360836,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.02819973893566612,
      "loss": 0.2319,
      "num_input_tokens_seen": 9275904,
      "step": 43960
    },
    {
      "epoch": 4.836633663366337,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.028199054847000818,
      "loss": 0.2309,
      "num_input_tokens_seen": 9276960,
      "step": 43965
    },
    {
      "epoch": 4.837183718371837,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.028198370636686156,
      "loss": 0.2324,
      "num_input_tokens_seen": 9278016,
      "step": 43970
    },
    {
      "epoch": 4.837733773377337,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.028197686304728447,
      "loss": 0.2293,
      "num_input_tokens_seen": 9279136,
      "step": 43975
    },
    {
      "epoch": 4.838283828382838,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.028197001851133996,
      "loss": 0.2325,
      "num_input_tokens_seen": 9280224,
      "step": 43980
    },
    {
      "epoch": 4.838833883388339,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.02819631727590911,
      "loss": 0.2346,
      "num_input_tokens_seen": 9281216,
      "step": 43985
    },
    {
      "epoch": 4.83938393839384,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.028195632579060104,
      "loss": 0.2285,
      "num_input_tokens_seen": 9282240,
      "step": 43990
    },
    {
      "epoch": 4.83993399339934,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02819494776059328,
      "loss": 0.232,
      "num_input_tokens_seen": 9283232,
      "step": 43995
    },
    {
      "epoch": 4.84048404840484,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.028194262820514956,
      "loss": 0.2341,
      "num_input_tokens_seen": 9284288,
      "step": 44000
    },
    {
      "epoch": 4.841034103410341,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.028193577758831443,
      "loss": 0.2336,
      "num_input_tokens_seen": 9285312,
      "step": 44005
    },
    {
      "epoch": 4.841584158415841,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.028192892575549057,
      "loss": 0.2314,
      "num_input_tokens_seen": 9286272,
      "step": 44010
    },
    {
      "epoch": 4.8421342134213425,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.028192207270674105,
      "loss": 0.232,
      "num_input_tokens_seen": 9287360,
      "step": 44015
    },
    {
      "epoch": 4.842684268426843,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.028191521844212912,
      "loss": 0.2324,
      "num_input_tokens_seen": 9288352,
      "step": 44020
    },
    {
      "epoch": 4.843234323432343,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.028190836296171792,
      "loss": 0.2294,
      "num_input_tokens_seen": 9289408,
      "step": 44025
    },
    {
      "epoch": 4.843784378437844,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.028190150626557067,
      "loss": 0.233,
      "num_input_tokens_seen": 9290464,
      "step": 44030
    },
    {
      "epoch": 4.844334433443344,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.028189464835375048,
      "loss": 0.2309,
      "num_input_tokens_seen": 9291520,
      "step": 44035
    },
    {
      "epoch": 4.8448844884488445,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.02818877892263206,
      "loss": 0.2319,
      "num_input_tokens_seen": 9292576,
      "step": 44040
    },
    {
      "epoch": 4.8454345434543455,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.028188092888334428,
      "loss": 0.2309,
      "num_input_tokens_seen": 9293632,
      "step": 44045
    },
    {
      "epoch": 4.845984598459846,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.028187406732488467,
      "loss": 0.2309,
      "num_input_tokens_seen": 9294656,
      "step": 44050
    },
    {
      "epoch": 4.846534653465347,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.028186720455100515,
      "loss": 0.2314,
      "num_input_tokens_seen": 9295680,
      "step": 44055
    },
    {
      "epoch": 4.847084708470847,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.028186034056176885,
      "loss": 0.2336,
      "num_input_tokens_seen": 9296672,
      "step": 44060
    },
    {
      "epoch": 4.847634763476347,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.028185347535723902,
      "loss": 0.2298,
      "num_input_tokens_seen": 9297728,
      "step": 44065
    },
    {
      "epoch": 4.848184818481848,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0281846608937479,
      "loss": 0.234,
      "num_input_tokens_seen": 9298784,
      "step": 44070
    },
    {
      "epoch": 4.8487348734873486,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.028183974130255206,
      "loss": 0.2325,
      "num_input_tokens_seen": 9299872,
      "step": 44075
    },
    {
      "epoch": 4.84928492849285,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02818328724525215,
      "loss": 0.2299,
      "num_input_tokens_seen": 9300960,
      "step": 44080
    },
    {
      "epoch": 4.84983498349835,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02818260023874506,
      "loss": 0.234,
      "num_input_tokens_seen": 9302048,
      "step": 44085
    },
    {
      "epoch": 4.85038503850385,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.028181913110740267,
      "loss": 0.2293,
      "num_input_tokens_seen": 9303072,
      "step": 44090
    },
    {
      "epoch": 4.850935093509351,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.028181225861244108,
      "loss": 0.2288,
      "num_input_tokens_seen": 9304096,
      "step": 44095
    },
    {
      "epoch": 4.851485148514851,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.028180538490262915,
      "loss": 0.2325,
      "num_input_tokens_seen": 9305088,
      "step": 44100
    },
    {
      "epoch": 4.852035203520352,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.028179850997803027,
      "loss": 0.2299,
      "num_input_tokens_seen": 9306176,
      "step": 44105
    },
    {
      "epoch": 4.852585258525853,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.028179163383870772,
      "loss": 0.2325,
      "num_input_tokens_seen": 9307200,
      "step": 44110
    },
    {
      "epoch": 4.853135313531353,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.028178475648472496,
      "loss": 0.2324,
      "num_input_tokens_seen": 9308256,
      "step": 44115
    },
    {
      "epoch": 4.853685368536854,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02817778779161453,
      "loss": 0.2309,
      "num_input_tokens_seen": 9309344,
      "step": 44120
    },
    {
      "epoch": 4.854235423542354,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.028177099813303217,
      "loss": 0.2314,
      "num_input_tokens_seen": 9310400,
      "step": 44125
    },
    {
      "epoch": 4.854785478547855,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.028176411713544895,
      "loss": 0.2314,
      "num_input_tokens_seen": 9311456,
      "step": 44130
    },
    {
      "epoch": 4.8553355335533555,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.028175723492345917,
      "loss": 0.2309,
      "num_input_tokens_seen": 9312512,
      "step": 44135
    },
    {
      "epoch": 4.855885588558856,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.028175035149712615,
      "loss": 0.2314,
      "num_input_tokens_seen": 9313568,
      "step": 44140
    },
    {
      "epoch": 4.856435643564357,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.028174346685651334,
      "loss": 0.2278,
      "num_input_tokens_seen": 9314624,
      "step": 44145
    },
    {
      "epoch": 4.856985698569857,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02817365810016842,
      "loss": 0.2305,
      "num_input_tokens_seen": 9315680,
      "step": 44150
    },
    {
      "epoch": 4.857535753575357,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.028172969393270225,
      "loss": 0.2331,
      "num_input_tokens_seen": 9316736,
      "step": 44155
    },
    {
      "epoch": 4.858085808580858,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.028172280564963088,
      "loss": 0.2316,
      "num_input_tokens_seen": 9317760,
      "step": 44160
    },
    {
      "epoch": 4.8586358635863585,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.02817159161525337,
      "loss": 0.2336,
      "num_input_tokens_seen": 9318816,
      "step": 44165
    },
    {
      "epoch": 4.8591859185918596,
      "grad_norm": 0.011962890625,
      "learning_rate": 0.028170902544147402,
      "loss": 0.2315,
      "num_input_tokens_seen": 9319840,
      "step": 44170
    },
    {
      "epoch": 4.85973597359736,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02817021335165155,
      "loss": 0.2284,
      "num_input_tokens_seen": 9320800,
      "step": 44175
    },
    {
      "epoch": 4.86028602860286,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028169524037772162,
      "loss": 0.2283,
      "num_input_tokens_seen": 9321888,
      "step": 44180
    },
    {
      "epoch": 4.860836083608361,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.02816883460251559,
      "loss": 0.2331,
      "num_input_tokens_seen": 9322912,
      "step": 44185
    },
    {
      "epoch": 4.861386138613861,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.028168145045888188,
      "loss": 0.232,
      "num_input_tokens_seen": 9323968,
      "step": 44190
    },
    {
      "epoch": 4.861936193619362,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.02816745536789631,
      "loss": 0.2314,
      "num_input_tokens_seen": 9325024,
      "step": 44195
    },
    {
      "epoch": 4.862486248624863,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.028166765568546314,
      "loss": 0.2341,
      "num_input_tokens_seen": 9326080,
      "step": 44200
    },
    {
      "epoch": 4.863036303630363,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02816607564784456,
      "loss": 0.2304,
      "num_input_tokens_seen": 9327104,
      "step": 44205
    },
    {
      "epoch": 4.863586358635864,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.028165385605797403,
      "loss": 0.2293,
      "num_input_tokens_seen": 9328224,
      "step": 44210
    },
    {
      "epoch": 4.864136413641364,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.028164695442411205,
      "loss": 0.2335,
      "num_input_tokens_seen": 9329280,
      "step": 44215
    },
    {
      "epoch": 4.864686468646864,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.028164005157692327,
      "loss": 0.2325,
      "num_input_tokens_seen": 9330368,
      "step": 44220
    },
    {
      "epoch": 4.865236523652365,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.028163314751647132,
      "loss": 0.2298,
      "num_input_tokens_seen": 9331456,
      "step": 44225
    },
    {
      "epoch": 4.865786578657866,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02816262422428198,
      "loss": 0.234,
      "num_input_tokens_seen": 9332512,
      "step": 44230
    },
    {
      "epoch": 4.866336633663367,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.028161933575603232,
      "loss": 0.235,
      "num_input_tokens_seen": 9333536,
      "step": 44235
    },
    {
      "epoch": 4.866886688668867,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02816124280561726,
      "loss": 0.2309,
      "num_input_tokens_seen": 9334560,
      "step": 44240
    },
    {
      "epoch": 4.867436743674367,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.028160551914330433,
      "loss": 0.2324,
      "num_input_tokens_seen": 9335648,
      "step": 44245
    },
    {
      "epoch": 4.867986798679868,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.028159860901749108,
      "loss": 0.2324,
      "num_input_tokens_seen": 9336672,
      "step": 44250
    },
    {
      "epoch": 4.868536853685368,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.028159169767879664,
      "loss": 0.2303,
      "num_input_tokens_seen": 9337696,
      "step": 44255
    },
    {
      "epoch": 4.8690869086908695,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.028158478512728465,
      "loss": 0.2309,
      "num_input_tokens_seen": 9338752,
      "step": 44260
    },
    {
      "epoch": 4.86963696369637,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02815778713630188,
      "loss": 0.2309,
      "num_input_tokens_seen": 9339808,
      "step": 44265
    },
    {
      "epoch": 4.87018701870187,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.02815709563860629,
      "loss": 0.2278,
      "num_input_tokens_seen": 9340832,
      "step": 44270
    },
    {
      "epoch": 4.870737073707371,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.028156404019648064,
      "loss": 0.2293,
      "num_input_tokens_seen": 9341888,
      "step": 44275
    },
    {
      "epoch": 4.871287128712871,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02815571227943357,
      "loss": 0.2324,
      "num_input_tokens_seen": 9342944,
      "step": 44280
    },
    {
      "epoch": 4.871837183718371,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.02815502041796919,
      "loss": 0.232,
      "num_input_tokens_seen": 9344032,
      "step": 44285
    },
    {
      "epoch": 4.8723872387238725,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.028154328435261302,
      "loss": 0.2304,
      "num_input_tokens_seen": 9345088,
      "step": 44290
    },
    {
      "epoch": 4.872937293729373,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.028153636331316283,
      "loss": 0.2319,
      "num_input_tokens_seen": 9346080,
      "step": 44295
    },
    {
      "epoch": 4.873487348734874,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.028152944106140502,
      "loss": 0.2314,
      "num_input_tokens_seen": 9347200,
      "step": 44300
    },
    {
      "epoch": 4.874037403740374,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.02815225175974035,
      "loss": 0.2298,
      "num_input_tokens_seen": 9348320,
      "step": 44305
    },
    {
      "epoch": 4.874587458745875,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028151559292122205,
      "loss": 0.2314,
      "num_input_tokens_seen": 9349280,
      "step": 44310
    },
    {
      "epoch": 4.875137513751375,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.028150866703292452,
      "loss": 0.2298,
      "num_input_tokens_seen": 9350368,
      "step": 44315
    },
    {
      "epoch": 4.8756875687568755,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.028150173993257463,
      "loss": 0.2319,
      "num_input_tokens_seen": 9351424,
      "step": 44320
    },
    {
      "epoch": 4.876237623762377,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.028149481162023638,
      "loss": 0.2314,
      "num_input_tokens_seen": 9352512,
      "step": 44325
    },
    {
      "epoch": 4.876787678767877,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.02814878820959735,
      "loss": 0.2293,
      "num_input_tokens_seen": 9353536,
      "step": 44330
    },
    {
      "epoch": 4.877337733773377,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.028148095135984992,
      "loss": 0.2313,
      "num_input_tokens_seen": 9354624,
      "step": 44335
    },
    {
      "epoch": 4.877887788778878,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.028147401941192952,
      "loss": 0.2324,
      "num_input_tokens_seen": 9355776,
      "step": 44340
    },
    {
      "epoch": 4.878437843784378,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.028146708625227616,
      "loss": 0.2288,
      "num_input_tokens_seen": 9356832,
      "step": 44345
    },
    {
      "epoch": 4.878987898789879,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.028146015188095376,
      "loss": 0.233,
      "num_input_tokens_seen": 9357920,
      "step": 44350
    },
    {
      "epoch": 4.87953795379538,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.02814532162980262,
      "loss": 0.2299,
      "num_input_tokens_seen": 9358912,
      "step": 44355
    },
    {
      "epoch": 4.88008800880088,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.028144627950355744,
      "loss": 0.2303,
      "num_input_tokens_seen": 9360000,
      "step": 44360
    },
    {
      "epoch": 4.880638063806381,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.028143934149761136,
      "loss": 0.2319,
      "num_input_tokens_seen": 9361056,
      "step": 44365
    },
    {
      "epoch": 4.881188118811881,
      "grad_norm": 0.01336669921875,
      "learning_rate": 0.0281432402280252,
      "loss": 0.2315,
      "num_input_tokens_seen": 9362112,
      "step": 44370
    },
    {
      "epoch": 4.881738173817382,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.02814254618515432,
      "loss": 0.2304,
      "num_input_tokens_seen": 9363232,
      "step": 44375
    },
    {
      "epoch": 4.882288228822882,
      "grad_norm": 0.01336669921875,
      "learning_rate": 0.0281418520211549,
      "loss": 0.2346,
      "num_input_tokens_seen": 9364224,
      "step": 44380
    },
    {
      "epoch": 4.882838283828383,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.02814115773603334,
      "loss": 0.2361,
      "num_input_tokens_seen": 9365312,
      "step": 44385
    },
    {
      "epoch": 4.883388338833884,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02814046332979603,
      "loss": 0.2324,
      "num_input_tokens_seen": 9366400,
      "step": 44390
    },
    {
      "epoch": 4.883938393839384,
      "grad_norm": 0.001953125,
      "learning_rate": 0.028139768802449373,
      "loss": 0.2277,
      "num_input_tokens_seen": 9367456,
      "step": 44395
    },
    {
      "epoch": 4.884488448844884,
      "grad_norm": 0.00048828125,
      "learning_rate": 0.028139074153999778,
      "loss": 0.2319,
      "num_input_tokens_seen": 9368512,
      "step": 44400
    },
    {
      "epoch": 4.885038503850385,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.028138379384453637,
      "loss": 0.2319,
      "num_input_tokens_seen": 9369600,
      "step": 44405
    },
    {
      "epoch": 4.885588558855885,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.02813768449381736,
      "loss": 0.2288,
      "num_input_tokens_seen": 9370720,
      "step": 44410
    },
    {
      "epoch": 4.8861386138613865,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02813698948209735,
      "loss": 0.2314,
      "num_input_tokens_seen": 9371776,
      "step": 44415
    },
    {
      "epoch": 4.886688668866887,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.028136294349300008,
      "loss": 0.2309,
      "num_input_tokens_seen": 9372800,
      "step": 44420
    },
    {
      "epoch": 4.887238723872387,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.028135599095431747,
      "loss": 0.233,
      "num_input_tokens_seen": 9373856,
      "step": 44425
    },
    {
      "epoch": 4.887788778877888,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.028134903720498972,
      "loss": 0.2319,
      "num_input_tokens_seen": 9374848,
      "step": 44430
    },
    {
      "epoch": 4.888338833883388,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.028134208224508093,
      "loss": 0.2278,
      "num_input_tokens_seen": 9375904,
      "step": 44435
    },
    {
      "epoch": 4.888888888888889,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.028133512607465515,
      "loss": 0.2299,
      "num_input_tokens_seen": 9376992,
      "step": 44440
    },
    {
      "epoch": 4.8894389438943895,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.028132816869377658,
      "loss": 0.2346,
      "num_input_tokens_seen": 9378016,
      "step": 44445
    },
    {
      "epoch": 4.88998899889989,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.028132121010250928,
      "loss": 0.2347,
      "num_input_tokens_seen": 9379072,
      "step": 44450
    },
    {
      "epoch": 4.890539053905391,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02813142503009174,
      "loss": 0.2294,
      "num_input_tokens_seen": 9380128,
      "step": 44455
    },
    {
      "epoch": 4.891089108910891,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.028130728928906508,
      "loss": 0.2325,
      "num_input_tokens_seen": 9381088,
      "step": 44460
    },
    {
      "epoch": 4.891639163916391,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028130032706701648,
      "loss": 0.2346,
      "num_input_tokens_seen": 9382144,
      "step": 44465
    },
    {
      "epoch": 4.892189218921892,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.028129336363483578,
      "loss": 0.2314,
      "num_input_tokens_seen": 9383200,
      "step": 44470
    },
    {
      "epoch": 4.8927392739273925,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.028128639899258713,
      "loss": 0.2351,
      "num_input_tokens_seen": 9384320,
      "step": 44475
    },
    {
      "epoch": 4.893289328932894,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.02812794331403348,
      "loss": 0.2314,
      "num_input_tokens_seen": 9385344,
      "step": 44480
    },
    {
      "epoch": 4.893839383938394,
      "grad_norm": 0.0004558563232421875,
      "learning_rate": 0.028127246607814283,
      "loss": 0.2313,
      "num_input_tokens_seen": 9386400,
      "step": 44485
    },
    {
      "epoch": 4.894389438943895,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02812654978060756,
      "loss": 0.2314,
      "num_input_tokens_seen": 9387424,
      "step": 44490
    },
    {
      "epoch": 4.894939493949395,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.028125852832419723,
      "loss": 0.2319,
      "num_input_tokens_seen": 9388480,
      "step": 44495
    },
    {
      "epoch": 4.895489548954895,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.028125155763257203,
      "loss": 0.2303,
      "num_input_tokens_seen": 9389504,
      "step": 44500
    },
    {
      "epoch": 4.896039603960396,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028124458573126416,
      "loss": 0.2314,
      "num_input_tokens_seen": 9390560,
      "step": 44505
    },
    {
      "epoch": 4.896589658965897,
      "grad_norm": 0.005859375,
      "learning_rate": 0.028123761262033798,
      "loss": 0.233,
      "num_input_tokens_seen": 9391680,
      "step": 44510
    },
    {
      "epoch": 4.897139713971397,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.028123063829985764,
      "loss": 0.2351,
      "num_input_tokens_seen": 9392768,
      "step": 44515
    },
    {
      "epoch": 4.897689768976898,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.028122366276988747,
      "loss": 0.2293,
      "num_input_tokens_seen": 9393888,
      "step": 44520
    },
    {
      "epoch": 4.898239823982398,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.028121668603049182,
      "loss": 0.2293,
      "num_input_tokens_seen": 9395040,
      "step": 44525
    },
    {
      "epoch": 4.898789878987898,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.02812097080817349,
      "loss": 0.2319,
      "num_input_tokens_seen": 9396064,
      "step": 44530
    },
    {
      "epoch": 4.899339933993399,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.028120272892368108,
      "loss": 0.2319,
      "num_input_tokens_seen": 9397120,
      "step": 44535
    },
    {
      "epoch": 4.8998899889989,
      "grad_norm": 0.005859375,
      "learning_rate": 0.028119574855639463,
      "loss": 0.2304,
      "num_input_tokens_seen": 9398144,
      "step": 44540
    },
    {
      "epoch": 4.900440044004401,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.028118876697993993,
      "loss": 0.2288,
      "num_input_tokens_seen": 9399200,
      "step": 44545
    },
    {
      "epoch": 4.900990099009901,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.028118178419438138,
      "loss": 0.2319,
      "num_input_tokens_seen": 9400256,
      "step": 44550
    },
    {
      "epoch": 4.901540154015402,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028117480019978318,
      "loss": 0.2304,
      "num_input_tokens_seen": 9401312,
      "step": 44555
    },
    {
      "epoch": 4.902090209020902,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.028116781499620987,
      "loss": 0.2335,
      "num_input_tokens_seen": 9402368,
      "step": 44560
    },
    {
      "epoch": 4.902640264026402,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02811608285837257,
      "loss": 0.2299,
      "num_input_tokens_seen": 9403360,
      "step": 44565
    },
    {
      "epoch": 4.9031903190319035,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.028115384096239513,
      "loss": 0.2324,
      "num_input_tokens_seen": 9404448,
      "step": 44570
    },
    {
      "epoch": 4.903740374037404,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02811468521322825,
      "loss": 0.2325,
      "num_input_tokens_seen": 9405536,
      "step": 44575
    },
    {
      "epoch": 4.904290429042904,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.028113986209345236,
      "loss": 0.2324,
      "num_input_tokens_seen": 9406656,
      "step": 44580
    },
    {
      "epoch": 4.904840484048405,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.028113287084596896,
      "loss": 0.2303,
      "num_input_tokens_seen": 9407680,
      "step": 44585
    },
    {
      "epoch": 4.905390539053905,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.02811258783898968,
      "loss": 0.2309,
      "num_input_tokens_seen": 9408736,
      "step": 44590
    },
    {
      "epoch": 4.905940594059406,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.02811188847253004,
      "loss": 0.2324,
      "num_input_tokens_seen": 9409792,
      "step": 44595
    },
    {
      "epoch": 4.9064906490649065,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02811118898522441,
      "loss": 0.234,
      "num_input_tokens_seen": 9410848,
      "step": 44600
    },
    {
      "epoch": 4.907040704070407,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02811048937707925,
      "loss": 0.2303,
      "num_input_tokens_seen": 9411936,
      "step": 44605
    },
    {
      "epoch": 4.907590759075908,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.028109789648100996,
      "loss": 0.2309,
      "num_input_tokens_seen": 9412992,
      "step": 44610
    },
    {
      "epoch": 4.908140814081408,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.0281090897982961,
      "loss": 0.2346,
      "num_input_tokens_seen": 9414112,
      "step": 44615
    },
    {
      "epoch": 4.908690869086909,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.028108389827671013,
      "loss": 0.2304,
      "num_input_tokens_seen": 9415168,
      "step": 44620
    },
    {
      "epoch": 4.909240924092409,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02810768973623219,
      "loss": 0.2329,
      "num_input_tokens_seen": 9416256,
      "step": 44625
    },
    {
      "epoch": 4.9097909790979095,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02810698952398608,
      "loss": 0.2298,
      "num_input_tokens_seen": 9417280,
      "step": 44630
    },
    {
      "epoch": 4.910341034103411,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.028106289190939132,
      "loss": 0.2309,
      "num_input_tokens_seen": 9418336,
      "step": 44635
    },
    {
      "epoch": 4.910891089108911,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.02810558873709781,
      "loss": 0.2319,
      "num_input_tokens_seen": 9419392,
      "step": 44640
    },
    {
      "epoch": 4.911441144114411,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.028104888162468564,
      "loss": 0.2308,
      "num_input_tokens_seen": 9420384,
      "step": 44645
    },
    {
      "epoch": 4.911991199119912,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028104187467057857,
      "loss": 0.2308,
      "num_input_tokens_seen": 9421440,
      "step": 44650
    },
    {
      "epoch": 4.912541254125412,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.028103486650872134,
      "loss": 0.2309,
      "num_input_tokens_seen": 9422464,
      "step": 44655
    },
    {
      "epoch": 4.913091309130913,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02810278571391787,
      "loss": 0.2324,
      "num_input_tokens_seen": 9423488,
      "step": 44660
    },
    {
      "epoch": 4.913641364136414,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02810208465620151,
      "loss": 0.2325,
      "num_input_tokens_seen": 9424576,
      "step": 44665
    },
    {
      "epoch": 4.914191419141914,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.028101383477729527,
      "loss": 0.2309,
      "num_input_tokens_seen": 9425664,
      "step": 44670
    },
    {
      "epoch": 4.914741474147415,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.02810068217850838,
      "loss": 0.2298,
      "num_input_tokens_seen": 9426688,
      "step": 44675
    },
    {
      "epoch": 4.915291529152915,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.028099980758544533,
      "loss": 0.2319,
      "num_input_tokens_seen": 9427712,
      "step": 44680
    },
    {
      "epoch": 4.915841584158416,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.028099279217844445,
      "loss": 0.2298,
      "num_input_tokens_seen": 9428768,
      "step": 44685
    },
    {
      "epoch": 4.916391639163916,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.028098577556414584,
      "loss": 0.2319,
      "num_input_tokens_seen": 9429824,
      "step": 44690
    },
    {
      "epoch": 4.916941694169417,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.028097875774261427,
      "loss": 0.2314,
      "num_input_tokens_seen": 9430880,
      "step": 44695
    },
    {
      "epoch": 4.917491749174918,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.028097173871391427,
      "loss": 0.233,
      "num_input_tokens_seen": 9432000,
      "step": 44700
    },
    {
      "epoch": 4.918041804180418,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028096471847811066,
      "loss": 0.2319,
      "num_input_tokens_seen": 9432992,
      "step": 44705
    },
    {
      "epoch": 4.918591859185918,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028095769703526802,
      "loss": 0.2314,
      "num_input_tokens_seen": 9434016,
      "step": 44710
    },
    {
      "epoch": 4.919141914191419,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.028095067438545118,
      "loss": 0.2313,
      "num_input_tokens_seen": 9435040,
      "step": 44715
    },
    {
      "epoch": 4.919691969196919,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.028094365052872477,
      "loss": 0.2314,
      "num_input_tokens_seen": 9436064,
      "step": 44720
    },
    {
      "epoch": 4.9202420242024205,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.028093662546515356,
      "loss": 0.2324,
      "num_input_tokens_seen": 9437088,
      "step": 44725
    },
    {
      "epoch": 4.920792079207921,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.028092959919480232,
      "loss": 0.2288,
      "num_input_tokens_seen": 9438144,
      "step": 44730
    },
    {
      "epoch": 4.921342134213422,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02809225717177358,
      "loss": 0.2314,
      "num_input_tokens_seen": 9439200,
      "step": 44735
    },
    {
      "epoch": 4.921892189218922,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.028091554303401878,
      "loss": 0.2303,
      "num_input_tokens_seen": 9440192,
      "step": 44740
    },
    {
      "epoch": 4.922442244224422,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.0280908513143716,
      "loss": 0.2324,
      "num_input_tokens_seen": 9441280,
      "step": 44745
    },
    {
      "epoch": 4.922992299229923,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028090148204689223,
      "loss": 0.234,
      "num_input_tokens_seen": 9442368,
      "step": 44750
    },
    {
      "epoch": 4.9235423542354235,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.028089444974361236,
      "loss": 0.2298,
      "num_input_tokens_seen": 9443456,
      "step": 44755
    },
    {
      "epoch": 4.924092409240924,
      "grad_norm": 0.0007171630859375,
      "learning_rate": 0.02808874162339411,
      "loss": 0.2303,
      "num_input_tokens_seen": 9444480,
      "step": 44760
    },
    {
      "epoch": 4.924642464246425,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02808803815179434,
      "loss": 0.2303,
      "num_input_tokens_seen": 9445568,
      "step": 44765
    },
    {
      "epoch": 4.925192519251925,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.0280873345595684,
      "loss": 0.2325,
      "num_input_tokens_seen": 9446688,
      "step": 44770
    },
    {
      "epoch": 4.925742574257426,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.028086630846722777,
      "loss": 0.2325,
      "num_input_tokens_seen": 9447840,
      "step": 44775
    },
    {
      "epoch": 4.926292629262926,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02808592701326396,
      "loss": 0.2305,
      "num_input_tokens_seen": 9448928,
      "step": 44780
    },
    {
      "epoch": 4.9268426842684265,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02808522305919843,
      "loss": 0.2305,
      "num_input_tokens_seen": 9450016,
      "step": 44785
    },
    {
      "epoch": 4.927392739273928,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02808451898453268,
      "loss": 0.2326,
      "num_input_tokens_seen": 9451040,
      "step": 44790
    },
    {
      "epoch": 4.927942794279428,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02808381478927319,
      "loss": 0.2268,
      "num_input_tokens_seen": 9452128,
      "step": 44795
    },
    {
      "epoch": 4.928492849284929,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.028083110473426463,
      "loss": 0.2336,
      "num_input_tokens_seen": 9453216,
      "step": 44800
    },
    {
      "epoch": 4.929042904290429,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.028082406036998988,
      "loss": 0.2295,
      "num_input_tokens_seen": 9454272,
      "step": 44805
    },
    {
      "epoch": 4.929592959295929,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.028081701479997247,
      "loss": 0.2274,
      "num_input_tokens_seen": 9455296,
      "step": 44810
    },
    {
      "epoch": 4.93014301430143,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.028080996802427743,
      "loss": 0.2316,
      "num_input_tokens_seen": 9456288,
      "step": 44815
    },
    {
      "epoch": 4.930693069306931,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.028080292004296972,
      "loss": 0.2338,
      "num_input_tokens_seen": 9457408,
      "step": 44820
    },
    {
      "epoch": 4.931243124312431,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.028079587085611424,
      "loss": 0.2327,
      "num_input_tokens_seen": 9458432,
      "step": 44825
    },
    {
      "epoch": 4.931793179317932,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.028078882046377596,
      "loss": 0.2317,
      "num_input_tokens_seen": 9459488,
      "step": 44830
    },
    {
      "epoch": 4.932343234323432,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.02807817688660199,
      "loss": 0.2326,
      "num_input_tokens_seen": 9460480,
      "step": 44835
    },
    {
      "epoch": 4.932893289328933,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.028077471606291104,
      "loss": 0.2341,
      "num_input_tokens_seen": 9461504,
      "step": 44840
    },
    {
      "epoch": 4.933443344334433,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.028076766205451433,
      "loss": 0.2345,
      "num_input_tokens_seen": 9462560,
      "step": 44845
    },
    {
      "epoch": 4.933993399339934,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.028076060684089486,
      "loss": 0.234,
      "num_input_tokens_seen": 9463584,
      "step": 44850
    },
    {
      "epoch": 4.934543454345435,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.02807535504221176,
      "loss": 0.2324,
      "num_input_tokens_seen": 9464608,
      "step": 44855
    },
    {
      "epoch": 4.935093509350935,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.028074649279824768,
      "loss": 0.2313,
      "num_input_tokens_seen": 9465664,
      "step": 44860
    },
    {
      "epoch": 4.935643564356436,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.028073943396935003,
      "loss": 0.2324,
      "num_input_tokens_seen": 9466720,
      "step": 44865
    },
    {
      "epoch": 4.936193619361936,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.02807323739354897,
      "loss": 0.2324,
      "num_input_tokens_seen": 9467776,
      "step": 44870
    },
    {
      "epoch": 4.936743674367436,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.028072531269673184,
      "loss": 0.2324,
      "num_input_tokens_seen": 9468864,
      "step": 44875
    },
    {
      "epoch": 4.9372937293729375,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.028071825025314155,
      "loss": 0.2319,
      "num_input_tokens_seen": 9469984,
      "step": 44880
    },
    {
      "epoch": 4.937843784378438,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028071118660478383,
      "loss": 0.2309,
      "num_input_tokens_seen": 9471072,
      "step": 44885
    },
    {
      "epoch": 4.938393839383938,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.028070412175172383,
      "loss": 0.2319,
      "num_input_tokens_seen": 9472128,
      "step": 44890
    },
    {
      "epoch": 4.938943894389439,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.028069705569402664,
      "loss": 0.2288,
      "num_input_tokens_seen": 9473216,
      "step": 44895
    },
    {
      "epoch": 4.939493949394939,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.028068998843175742,
      "loss": 0.2293,
      "num_input_tokens_seen": 9474240,
      "step": 44900
    },
    {
      "epoch": 4.94004400440044,
      "grad_norm": 0.0006103515625,
      "learning_rate": 0.028068291996498133,
      "loss": 0.233,
      "num_input_tokens_seen": 9475360,
      "step": 44905
    },
    {
      "epoch": 4.9405940594059405,
      "grad_norm": 0.000476837158203125,
      "learning_rate": 0.02806758502937634,
      "loss": 0.2314,
      "num_input_tokens_seen": 9476416,
      "step": 44910
    },
    {
      "epoch": 4.941144114411442,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.028066877941816892,
      "loss": 0.2309,
      "num_input_tokens_seen": 9477504,
      "step": 44915
    },
    {
      "epoch": 4.941694169416942,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.028066170733826296,
      "loss": 0.2309,
      "num_input_tokens_seen": 9478592,
      "step": 44920
    },
    {
      "epoch": 4.942244224422442,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.028065463405411075,
      "loss": 0.2335,
      "num_input_tokens_seen": 9479680,
      "step": 44925
    },
    {
      "epoch": 4.942794279427943,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.028064755956577748,
      "loss": 0.234,
      "num_input_tokens_seen": 9480704,
      "step": 44930
    },
    {
      "epoch": 4.943344334433443,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.028064048387332837,
      "loss": 0.2288,
      "num_input_tokens_seen": 9481760,
      "step": 44935
    },
    {
      "epoch": 4.9438943894389435,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.02806334069768286,
      "loss": 0.2304,
      "num_input_tokens_seen": 9482720,
      "step": 44940
    },
    {
      "epoch": 4.944444444444445,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.028062632887634335,
      "loss": 0.232,
      "num_input_tokens_seen": 9483776,
      "step": 44945
    },
    {
      "epoch": 4.944994499449945,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.0280619249571938,
      "loss": 0.2325,
      "num_input_tokens_seen": 9484832,
      "step": 44950
    },
    {
      "epoch": 4.945544554455445,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02806121690636776,
      "loss": 0.2319,
      "num_input_tokens_seen": 9485856,
      "step": 44955
    },
    {
      "epoch": 4.946094609460946,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.028060508735162755,
      "loss": 0.234,
      "num_input_tokens_seen": 9486912,
      "step": 44960
    },
    {
      "epoch": 4.946644664466446,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.028059800443585307,
      "loss": 0.2293,
      "num_input_tokens_seen": 9487968,
      "step": 44965
    },
    {
      "epoch": 4.947194719471947,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.028059092031641947,
      "loss": 0.2324,
      "num_input_tokens_seen": 9489088,
      "step": 44970
    },
    {
      "epoch": 4.947744774477448,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.028058383499339205,
      "loss": 0.2298,
      "num_input_tokens_seen": 9490144,
      "step": 44975
    },
    {
      "epoch": 4.948294829482949,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.028057674846683603,
      "loss": 0.2308,
      "num_input_tokens_seen": 9491200,
      "step": 44980
    },
    {
      "epoch": 4.948844884488449,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028056966073681678,
      "loss": 0.2308,
      "num_input_tokens_seen": 9492224,
      "step": 44985
    },
    {
      "epoch": 4.949394939493949,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.028056257180339963,
      "loss": 0.2308,
      "num_input_tokens_seen": 9493248,
      "step": 44990
    },
    {
      "epoch": 4.94994499449945,
      "grad_norm": 0.005859375,
      "learning_rate": 0.028055548166664994,
      "loss": 0.2293,
      "num_input_tokens_seen": 9494304,
      "step": 44995
    },
    {
      "epoch": 4.9504950495049505,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.0280548390326633,
      "loss": 0.2303,
      "num_input_tokens_seen": 9495328,
      "step": 45000
    },
    {
      "epoch": 4.951045104510451,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02805412977834142,
      "loss": 0.2314,
      "num_input_tokens_seen": 9496320,
      "step": 45005
    },
    {
      "epoch": 4.951595159515952,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.028053420403705887,
      "loss": 0.2319,
      "num_input_tokens_seen": 9497408,
      "step": 45010
    },
    {
      "epoch": 4.952145214521452,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.028052710908763243,
      "loss": 0.2303,
      "num_input_tokens_seen": 9498496,
      "step": 45015
    },
    {
      "epoch": 4.952695269526953,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.028052001293520024,
      "loss": 0.2329,
      "num_input_tokens_seen": 9499520,
      "step": 45020
    },
    {
      "epoch": 4.953245324532453,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.028051291557982777,
      "loss": 0.234,
      "num_input_tokens_seen": 9500608,
      "step": 45025
    },
    {
      "epoch": 4.9537953795379535,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.028050581702158035,
      "loss": 0.2298,
      "num_input_tokens_seen": 9501664,
      "step": 45030
    },
    {
      "epoch": 4.9543454345434546,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028049871726052344,
      "loss": 0.2324,
      "num_input_tokens_seen": 9502752,
      "step": 45035
    },
    {
      "epoch": 4.954895489548955,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.028049161629672248,
      "loss": 0.2324,
      "num_input_tokens_seen": 9503776,
      "step": 45040
    },
    {
      "epoch": 4.955445544554456,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.028048451413024287,
      "loss": 0.2319,
      "num_input_tokens_seen": 9504832,
      "step": 45045
    },
    {
      "epoch": 4.955995599559956,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.028047741076115015,
      "loss": 0.2324,
      "num_input_tokens_seen": 9505856,
      "step": 45050
    },
    {
      "epoch": 4.956545654565456,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.028047030618950974,
      "loss": 0.2324,
      "num_input_tokens_seen": 9506976,
      "step": 45055
    },
    {
      "epoch": 4.957095709570957,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.028046320041538713,
      "loss": 0.2329,
      "num_input_tokens_seen": 9507968,
      "step": 45060
    },
    {
      "epoch": 4.957645764576458,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.028045609343884777,
      "loss": 0.2308,
      "num_input_tokens_seen": 9509120,
      "step": 45065
    },
    {
      "epoch": 4.958195819581958,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02804489852599572,
      "loss": 0.2319,
      "num_input_tokens_seen": 9510208,
      "step": 45070
    },
    {
      "epoch": 4.958745874587459,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.028044187587878095,
      "loss": 0.2308,
      "num_input_tokens_seen": 9511296,
      "step": 45075
    },
    {
      "epoch": 4.959295929592959,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.02804347652953845,
      "loss": 0.2324,
      "num_input_tokens_seen": 9512448,
      "step": 45080
    },
    {
      "epoch": 4.95984598459846,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028042765350983344,
      "loss": 0.2308,
      "num_input_tokens_seen": 9513472,
      "step": 45085
    },
    {
      "epoch": 4.96039603960396,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02804205405221933,
      "loss": 0.2308,
      "num_input_tokens_seen": 9514464,
      "step": 45090
    },
    {
      "epoch": 4.960946094609461,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.028041342633252957,
      "loss": 0.2334,
      "num_input_tokens_seen": 9515616,
      "step": 45095
    },
    {
      "epoch": 4.961496149614962,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.028040631094090787,
      "loss": 0.2329,
      "num_input_tokens_seen": 9516608,
      "step": 45100
    },
    {
      "epoch": 4.962046204620462,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.02803991943473938,
      "loss": 0.2308,
      "num_input_tokens_seen": 9517632,
      "step": 45105
    },
    {
      "epoch": 4.962596259625963,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.028039207655205293,
      "loss": 0.2324,
      "num_input_tokens_seen": 9518656,
      "step": 45110
    },
    {
      "epoch": 4.963146314631463,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.02803849575549508,
      "loss": 0.2298,
      "num_input_tokens_seen": 9519648,
      "step": 45115
    },
    {
      "epoch": 4.963696369636963,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.028037783735615315,
      "loss": 0.2329,
      "num_input_tokens_seen": 9520672,
      "step": 45120
    },
    {
      "epoch": 4.9642464246424645,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.028037071595572553,
      "loss": 0.2325,
      "num_input_tokens_seen": 9521760,
      "step": 45125
    },
    {
      "epoch": 4.964796479647965,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02803635933537335,
      "loss": 0.2309,
      "num_input_tokens_seen": 9522816,
      "step": 45130
    },
    {
      "epoch": 4.965346534653465,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02803564695502429,
      "loss": 0.2299,
      "num_input_tokens_seen": 9523936,
      "step": 45135
    },
    {
      "epoch": 4.965896589658966,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.02803493445453192,
      "loss": 0.2326,
      "num_input_tokens_seen": 9524928,
      "step": 45140
    },
    {
      "epoch": 4.966446644664466,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028034221833902816,
      "loss": 0.231,
      "num_input_tokens_seen": 9525888,
      "step": 45145
    },
    {
      "epoch": 4.966996699669967,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.028033509093143543,
      "loss": 0.2279,
      "num_input_tokens_seen": 9526944,
      "step": 45150
    },
    {
      "epoch": 4.9675467546754675,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.028032796232260675,
      "loss": 0.2321,
      "num_input_tokens_seen": 9528032,
      "step": 45155
    },
    {
      "epoch": 4.968096809680969,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02803208325126077,
      "loss": 0.2295,
      "num_input_tokens_seen": 9529056,
      "step": 45160
    },
    {
      "epoch": 4.968646864686469,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02803137015015041,
      "loss": 0.2321,
      "num_input_tokens_seen": 9530080,
      "step": 45165
    },
    {
      "epoch": 4.969196919691969,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.028030656928936165,
      "loss": 0.23,
      "num_input_tokens_seen": 9531072,
      "step": 45170
    },
    {
      "epoch": 4.96974697469747,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.02802994358762461,
      "loss": 0.2311,
      "num_input_tokens_seen": 9532096,
      "step": 45175
    },
    {
      "epoch": 4.97029702970297,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.028029230126222316,
      "loss": 0.2348,
      "num_input_tokens_seen": 9533120,
      "step": 45180
    },
    {
      "epoch": 4.9708470847084705,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02802851654473586,
      "loss": 0.2353,
      "num_input_tokens_seen": 9534144,
      "step": 45185
    },
    {
      "epoch": 4.971397139713972,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.028027802843171818,
      "loss": 0.2352,
      "num_input_tokens_seen": 9535264,
      "step": 45190
    },
    {
      "epoch": 4.971947194719472,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.028027089021536767,
      "loss": 0.2305,
      "num_input_tokens_seen": 9536288,
      "step": 45195
    },
    {
      "epoch": 4.972497249724973,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.028026375079837292,
      "loss": 0.2294,
      "num_input_tokens_seen": 9537376,
      "step": 45200
    },
    {
      "epoch": 4.973047304730473,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.02802566101807996,
      "loss": 0.232,
      "num_input_tokens_seen": 9538432,
      "step": 45205
    },
    {
      "epoch": 4.973597359735973,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.02802494683627137,
      "loss": 0.2341,
      "num_input_tokens_seen": 9539488,
      "step": 45210
    },
    {
      "epoch": 4.974147414741474,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.028024232534418093,
      "loss": 0.233,
      "num_input_tokens_seen": 9540512,
      "step": 45215
    },
    {
      "epoch": 4.974697469746975,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02802351811252671,
      "loss": 0.232,
      "num_input_tokens_seen": 9541600,
      "step": 45220
    },
    {
      "epoch": 4.975247524752476,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.028022803570603817,
      "loss": 0.232,
      "num_input_tokens_seen": 9542624,
      "step": 45225
    },
    {
      "epoch": 4.975797579757976,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028022088908655988,
      "loss": 0.2341,
      "num_input_tokens_seen": 9543584,
      "step": 45230
    },
    {
      "epoch": 4.976347634763476,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.028021374126689813,
      "loss": 0.2288,
      "num_input_tokens_seen": 9544608,
      "step": 45235
    },
    {
      "epoch": 4.976897689768977,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02802065922471188,
      "loss": 0.2283,
      "num_input_tokens_seen": 9545728,
      "step": 45240
    },
    {
      "epoch": 4.977447744774477,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02801994420272878,
      "loss": 0.2314,
      "num_input_tokens_seen": 9546816,
      "step": 45245
    },
    {
      "epoch": 4.977997799779978,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02801922906074711,
      "loss": 0.2289,
      "num_input_tokens_seen": 9547872,
      "step": 45250
    },
    {
      "epoch": 4.978547854785479,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02801851379877344,
      "loss": 0.2294,
      "num_input_tokens_seen": 9548864,
      "step": 45255
    },
    {
      "epoch": 4.979097909790979,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.02801779841681438,
      "loss": 0.2346,
      "num_input_tokens_seen": 9549952,
      "step": 45260
    },
    {
      "epoch": 4.97964796479648,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.028017082914876524,
      "loss": 0.2325,
      "num_input_tokens_seen": 9551040,
      "step": 45265
    },
    {
      "epoch": 4.98019801980198,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.028016367292966454,
      "loss": 0.2294,
      "num_input_tokens_seen": 9552064,
      "step": 45270
    },
    {
      "epoch": 4.98074807480748,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.028015651551090776,
      "loss": 0.233,
      "num_input_tokens_seen": 9553120,
      "step": 45275
    },
    {
      "epoch": 4.9812981298129815,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.02801493568925608,
      "loss": 0.2351,
      "num_input_tokens_seen": 9554240,
      "step": 45280
    },
    {
      "epoch": 4.981848184818482,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02801421970746897,
      "loss": 0.2319,
      "num_input_tokens_seen": 9555360,
      "step": 45285
    },
    {
      "epoch": 4.982398239823983,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.028013503605736043,
      "loss": 0.2319,
      "num_input_tokens_seen": 9556384,
      "step": 45290
    },
    {
      "epoch": 4.982948294829483,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.028012787384063893,
      "loss": 0.2329,
      "num_input_tokens_seen": 9557472,
      "step": 45295
    },
    {
      "epoch": 4.983498349834983,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.02801207104245913,
      "loss": 0.2329,
      "num_input_tokens_seen": 9558528,
      "step": 45300
    },
    {
      "epoch": 4.984048404840484,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.02801135458092835,
      "loss": 0.2324,
      "num_input_tokens_seen": 9559616,
      "step": 45305
    },
    {
      "epoch": 4.9845984598459845,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02801063799947816,
      "loss": 0.2324,
      "num_input_tokens_seen": 9560672,
      "step": 45310
    },
    {
      "epoch": 4.985148514851485,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.02800992129811516,
      "loss": 0.2324,
      "num_input_tokens_seen": 9561760,
      "step": 45315
    },
    {
      "epoch": 4.985698569856986,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.02800920447684596,
      "loss": 0.2318,
      "num_input_tokens_seen": 9562784,
      "step": 45320
    },
    {
      "epoch": 4.986248624862486,
      "grad_norm": 0.000522613525390625,
      "learning_rate": 0.028008487535677168,
      "loss": 0.2303,
      "num_input_tokens_seen": 9563808,
      "step": 45325
    },
    {
      "epoch": 4.986798679867987,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.028007770474615383,
      "loss": 0.2304,
      "num_input_tokens_seen": 9564832,
      "step": 45330
    },
    {
      "epoch": 4.987348734873487,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.028007053293667223,
      "loss": 0.2298,
      "num_input_tokens_seen": 9565888,
      "step": 45335
    },
    {
      "epoch": 4.987898789878988,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02800633599283929,
      "loss": 0.2299,
      "num_input_tokens_seen": 9566880,
      "step": 45340
    },
    {
      "epoch": 4.988448844884489,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.028005618572138204,
      "loss": 0.2341,
      "num_input_tokens_seen": 9567936,
      "step": 45345
    },
    {
      "epoch": 4.988998899889989,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.028004901031570568,
      "loss": 0.232,
      "num_input_tokens_seen": 9569024,
      "step": 45350
    },
    {
      "epoch": 4.98954895489549,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.028004183371143004,
      "loss": 0.233,
      "num_input_tokens_seen": 9570112,
      "step": 45355
    },
    {
      "epoch": 4.99009900990099,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.028003465590862118,
      "loss": 0.2324,
      "num_input_tokens_seen": 9571136,
      "step": 45360
    },
    {
      "epoch": 4.99064906490649,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.028002747690734534,
      "loss": 0.2303,
      "num_input_tokens_seen": 9572224,
      "step": 45365
    },
    {
      "epoch": 4.991199119911991,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.028002029670766856,
      "loss": 0.2303,
      "num_input_tokens_seen": 9573312,
      "step": 45370
    },
    {
      "epoch": 4.991749174917492,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.028001311530965713,
      "loss": 0.2339,
      "num_input_tokens_seen": 9574336,
      "step": 45375
    },
    {
      "epoch": 4.992299229922993,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.028000593271337724,
      "loss": 0.2308,
      "num_input_tokens_seen": 9575360,
      "step": 45380
    },
    {
      "epoch": 4.992849284928493,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0279998748918895,
      "loss": 0.2324,
      "num_input_tokens_seen": 9576416,
      "step": 45385
    },
    {
      "epoch": 4.993399339933993,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.027999156392627667,
      "loss": 0.2308,
      "num_input_tokens_seen": 9577408,
      "step": 45390
    },
    {
      "epoch": 4.993949394939494,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.027998437773558853,
      "loss": 0.2324,
      "num_input_tokens_seen": 9578432,
      "step": 45395
    },
    {
      "epoch": 4.994499449944994,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.027997719034689672,
      "loss": 0.2329,
      "num_input_tokens_seen": 9579520,
      "step": 45400
    },
    {
      "epoch": 4.9950495049504955,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.027997000176026753,
      "loss": 0.2313,
      "num_input_tokens_seen": 9580576,
      "step": 45405
    },
    {
      "epoch": 4.995599559955996,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.027996281197576717,
      "loss": 0.2335,
      "num_input_tokens_seen": 9581632,
      "step": 45410
    },
    {
      "epoch": 4.996149614961496,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.027995562099346196,
      "loss": 0.2319,
      "num_input_tokens_seen": 9582688,
      "step": 45415
    },
    {
      "epoch": 4.996699669966997,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.027994842881341814,
      "loss": 0.2298,
      "num_input_tokens_seen": 9583744,
      "step": 45420
    },
    {
      "epoch": 4.997249724972497,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027994123543570202,
      "loss": 0.2313,
      "num_input_tokens_seen": 9584800,
      "step": 45425
    },
    {
      "epoch": 4.997799779977997,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.027993404086037985,
      "loss": 0.2319,
      "num_input_tokens_seen": 9585920,
      "step": 45430
    },
    {
      "epoch": 4.9983498349834985,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.0279926845087518,
      "loss": 0.2324,
      "num_input_tokens_seen": 9587008,
      "step": 45435
    },
    {
      "epoch": 4.998899889988999,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027991964811718274,
      "loss": 0.2278,
      "num_input_tokens_seen": 9587968,
      "step": 45440
    },
    {
      "epoch": 4.9994499449945,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.027991244994944046,
      "loss": 0.2341,
      "num_input_tokens_seen": 9589056,
      "step": 45445
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.027990525058435745,
      "loss": 0.2319,
      "num_input_tokens_seen": 9590080,
      "step": 45450
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.2314261496067047,
      "eval_runtime": 60.5636,
      "eval_samples_per_second": 66.707,
      "eval_steps_per_second": 16.677,
      "num_input_tokens_seen": 9590080,
      "step": 45450
    },
    {
      "epoch": 5.0005500550055,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027989805002200007,
      "loss": 0.2324,
      "num_input_tokens_seen": 9591168,
      "step": 45455
    },
    {
      "epoch": 5.001100110011001,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02798908482624347,
      "loss": 0.2293,
      "num_input_tokens_seen": 9592192,
      "step": 45460
    },
    {
      "epoch": 5.0016501650165015,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02798836453057277,
      "loss": 0.2346,
      "num_input_tokens_seen": 9593248,
      "step": 45465
    },
    {
      "epoch": 5.002200220022003,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.027987644115194547,
      "loss": 0.233,
      "num_input_tokens_seen": 9594304,
      "step": 45470
    },
    {
      "epoch": 5.002750275027503,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.027986923580115442,
      "loss": 0.232,
      "num_input_tokens_seen": 9595328,
      "step": 45475
    },
    {
      "epoch": 5.003300330033003,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.027986202925342093,
      "loss": 0.232,
      "num_input_tokens_seen": 9596320,
      "step": 45480
    },
    {
      "epoch": 5.003850385038504,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.027985482150881143,
      "loss": 0.2299,
      "num_input_tokens_seen": 9597408,
      "step": 45485
    },
    {
      "epoch": 5.004400440044004,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.027984761256739238,
      "loss": 0.2304,
      "num_input_tokens_seen": 9598464,
      "step": 45490
    },
    {
      "epoch": 5.0049504950495045,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.027984040242923017,
      "loss": 0.2324,
      "num_input_tokens_seen": 9599520,
      "step": 45495
    },
    {
      "epoch": 5.005500550055006,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.027983319109439127,
      "loss": 0.233,
      "num_input_tokens_seen": 9600576,
      "step": 45500
    },
    {
      "epoch": 5.006050605060506,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.027982597856294214,
      "loss": 0.2298,
      "num_input_tokens_seen": 9601664,
      "step": 45505
    },
    {
      "epoch": 5.006600660066007,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027981876483494924,
      "loss": 0.2309,
      "num_input_tokens_seen": 9602784,
      "step": 45510
    },
    {
      "epoch": 5.007150715071507,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.02798115499104791,
      "loss": 0.235,
      "num_input_tokens_seen": 9603808,
      "step": 45515
    },
    {
      "epoch": 5.007700770077007,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027980433378959824,
      "loss": 0.2319,
      "num_input_tokens_seen": 9604864,
      "step": 45520
    },
    {
      "epoch": 5.008250825082508,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.027979711647237306,
      "loss": 0.2303,
      "num_input_tokens_seen": 9605984,
      "step": 45525
    },
    {
      "epoch": 5.008800880088009,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02797898979588702,
      "loss": 0.2319,
      "num_input_tokens_seen": 9607072,
      "step": 45530
    },
    {
      "epoch": 5.00935093509351,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.027978267824915606,
      "loss": 0.2313,
      "num_input_tokens_seen": 9608096,
      "step": 45535
    },
    {
      "epoch": 5.00990099009901,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.02797754573432973,
      "loss": 0.2308,
      "num_input_tokens_seen": 9609216,
      "step": 45540
    },
    {
      "epoch": 5.01045104510451,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02797682352413604,
      "loss": 0.2319,
      "num_input_tokens_seen": 9610272,
      "step": 45545
    },
    {
      "epoch": 5.011001100110011,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027976101194341196,
      "loss": 0.2318,
      "num_input_tokens_seen": 9611328,
      "step": 45550
    },
    {
      "epoch": 5.011551155115511,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.02797537874495185,
      "loss": 0.2308,
      "num_input_tokens_seen": 9612384,
      "step": 45555
    },
    {
      "epoch": 5.0121012101210125,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02797465617597467,
      "loss": 0.2308,
      "num_input_tokens_seen": 9613504,
      "step": 45560
    },
    {
      "epoch": 5.012651265126513,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.027973933487416305,
      "loss": 0.2293,
      "num_input_tokens_seen": 9614560,
      "step": 45565
    },
    {
      "epoch": 5.013201320132013,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027973210679283423,
      "loss": 0.234,
      "num_input_tokens_seen": 9615680,
      "step": 45570
    },
    {
      "epoch": 5.013751375137514,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.027972487751582682,
      "loss": 0.2341,
      "num_input_tokens_seen": 9616704,
      "step": 45575
    },
    {
      "epoch": 5.014301430143014,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.027971764704320747,
      "loss": 0.2315,
      "num_input_tokens_seen": 9617760,
      "step": 45580
    },
    {
      "epoch": 5.014851485148514,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.027971041537504283,
      "loss": 0.232,
      "num_input_tokens_seen": 9618784,
      "step": 45585
    },
    {
      "epoch": 5.0154015401540155,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.02797031825113995,
      "loss": 0.2325,
      "num_input_tokens_seen": 9619872,
      "step": 45590
    },
    {
      "epoch": 5.015951595159516,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.027969594845234418,
      "loss": 0.2324,
      "num_input_tokens_seen": 9620896,
      "step": 45595
    },
    {
      "epoch": 5.016501650165017,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.027968871319794354,
      "loss": 0.2309,
      "num_input_tokens_seen": 9621920,
      "step": 45600
    },
    {
      "epoch": 5.017051705170517,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.027968147674826426,
      "loss": 0.2293,
      "num_input_tokens_seen": 9622976,
      "step": 45605
    },
    {
      "epoch": 5.017601760176017,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.027967423910337307,
      "loss": 0.2331,
      "num_input_tokens_seen": 9623968,
      "step": 45610
    },
    {
      "epoch": 5.018151815181518,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.027966700026333658,
      "loss": 0.2289,
      "num_input_tokens_seen": 9625056,
      "step": 45615
    },
    {
      "epoch": 5.0187018701870185,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.027965976022822164,
      "loss": 0.2299,
      "num_input_tokens_seen": 9626112,
      "step": 45620
    },
    {
      "epoch": 5.01925192519252,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.027965251899809487,
      "loss": 0.232,
      "num_input_tokens_seen": 9627168,
      "step": 45625
    },
    {
      "epoch": 5.01980198019802,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.027964527657302302,
      "loss": 0.2315,
      "num_input_tokens_seen": 9628256,
      "step": 45630
    },
    {
      "epoch": 5.02035203520352,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.027963803295307288,
      "loss": 0.2315,
      "num_input_tokens_seen": 9629280,
      "step": 45635
    },
    {
      "epoch": 5.020902090209021,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.027963078813831126,
      "loss": 0.232,
      "num_input_tokens_seen": 9630304,
      "step": 45640
    },
    {
      "epoch": 5.021452145214521,
      "grad_norm": 0.00055694580078125,
      "learning_rate": 0.02796235421288048,
      "loss": 0.2309,
      "num_input_tokens_seen": 9631328,
      "step": 45645
    },
    {
      "epoch": 5.022002200220022,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.027961629492462038,
      "loss": 0.232,
      "num_input_tokens_seen": 9632352,
      "step": 45650
    },
    {
      "epoch": 5.022552255225523,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.027960904652582476,
      "loss": 0.233,
      "num_input_tokens_seen": 9633408,
      "step": 45655
    },
    {
      "epoch": 5.023102310231023,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02796017969324847,
      "loss": 0.2314,
      "num_input_tokens_seen": 9634528,
      "step": 45660
    },
    {
      "epoch": 5.023652365236524,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.027959454614466715,
      "loss": 0.2341,
      "num_input_tokens_seen": 9635616,
      "step": 45665
    },
    {
      "epoch": 5.024202420242024,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.02795872941624388,
      "loss": 0.2324,
      "num_input_tokens_seen": 9636672,
      "step": 45670
    },
    {
      "epoch": 5.024752475247524,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.027958004098586654,
      "loss": 0.2288,
      "num_input_tokens_seen": 9637664,
      "step": 45675
    },
    {
      "epoch": 5.025302530253025,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.027957278661501724,
      "loss": 0.2314,
      "num_input_tokens_seen": 9638752,
      "step": 45680
    },
    {
      "epoch": 5.025852585258526,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027956553104995775,
      "loss": 0.2298,
      "num_input_tokens_seen": 9639872,
      "step": 45685
    },
    {
      "epoch": 5.026402640264027,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.027955827429075493,
      "loss": 0.2309,
      "num_input_tokens_seen": 9640960,
      "step": 45690
    },
    {
      "epoch": 5.026952695269527,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.027955101633747564,
      "loss": 0.2314,
      "num_input_tokens_seen": 9642016,
      "step": 45695
    },
    {
      "epoch": 5.027502750275027,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.02795437571901868,
      "loss": 0.2308,
      "num_input_tokens_seen": 9643072,
      "step": 45700
    },
    {
      "epoch": 5.028052805280528,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.027953649684895533,
      "loss": 0.2313,
      "num_input_tokens_seen": 9644096,
      "step": 45705
    },
    {
      "epoch": 5.028602860286028,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027952923531384812,
      "loss": 0.2298,
      "num_input_tokens_seen": 9645184,
      "step": 45710
    },
    {
      "epoch": 5.0291529152915295,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.02795219725849321,
      "loss": 0.2334,
      "num_input_tokens_seen": 9646208,
      "step": 45715
    },
    {
      "epoch": 5.02970297029703,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.027951470866227422,
      "loss": 0.2313,
      "num_input_tokens_seen": 9647296,
      "step": 45720
    },
    {
      "epoch": 5.03025302530253,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.027950744354594142,
      "loss": 0.2324,
      "num_input_tokens_seen": 9648480,
      "step": 45725
    },
    {
      "epoch": 5.030803080308031,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.027950017723600066,
      "loss": 0.2308,
      "num_input_tokens_seen": 9649568,
      "step": 45730
    },
    {
      "epoch": 5.031353135313531,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.027949290973251887,
      "loss": 0.2319,
      "num_input_tokens_seen": 9650656,
      "step": 45735
    },
    {
      "epoch": 5.031903190319032,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02794856410355631,
      "loss": 0.2288,
      "num_input_tokens_seen": 9651712,
      "step": 45740
    },
    {
      "epoch": 5.0324532453245325,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02794783711452003,
      "loss": 0.2304,
      "num_input_tokens_seen": 9652768,
      "step": 45745
    },
    {
      "epoch": 5.033003300330033,
      "grad_norm": 0.000705718994140625,
      "learning_rate": 0.027947110006149748,
      "loss": 0.2331,
      "num_input_tokens_seen": 9653824,
      "step": 45750
    },
    {
      "epoch": 5.033553355335534,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.027946382778452166,
      "loss": 0.2352,
      "num_input_tokens_seen": 9654880,
      "step": 45755
    },
    {
      "epoch": 5.034103410341034,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.027945655431433986,
      "loss": 0.23,
      "num_input_tokens_seen": 9656000,
      "step": 45760
    },
    {
      "epoch": 5.034653465346534,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.027944927965101913,
      "loss": 0.231,
      "num_input_tokens_seen": 9657024,
      "step": 45765
    },
    {
      "epoch": 5.035203520352035,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.027944200379462648,
      "loss": 0.2274,
      "num_input_tokens_seen": 9658176,
      "step": 45770
    },
    {
      "epoch": 5.0357535753575355,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0279434726745229,
      "loss": 0.2357,
      "num_input_tokens_seen": 9659168,
      "step": 45775
    },
    {
      "epoch": 5.036303630363037,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.027942744850289376,
      "loss": 0.2295,
      "num_input_tokens_seen": 9660288,
      "step": 45780
    },
    {
      "epoch": 5.036853685368537,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.027942016906768785,
      "loss": 0.2326,
      "num_input_tokens_seen": 9661344,
      "step": 45785
    },
    {
      "epoch": 5.037403740374037,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.02794128884396783,
      "loss": 0.2295,
      "num_input_tokens_seen": 9662400,
      "step": 45790
    },
    {
      "epoch": 5.037953795379538,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.02794056066189323,
      "loss": 0.2363,
      "num_input_tokens_seen": 9663488,
      "step": 45795
    },
    {
      "epoch": 5.038503850385038,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02793983236055169,
      "loss": 0.2326,
      "num_input_tokens_seen": 9664576,
      "step": 45800
    },
    {
      "epoch": 5.039053905390539,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.027939103939949923,
      "loss": 0.2309,
      "num_input_tokens_seen": 9665632,
      "step": 45805
    },
    {
      "epoch": 5.03960396039604,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02793837540009465,
      "loss": 0.233,
      "num_input_tokens_seen": 9666656,
      "step": 45810
    },
    {
      "epoch": 5.04015401540154,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.027937646740992573,
      "loss": 0.2319,
      "num_input_tokens_seen": 9667680,
      "step": 45815
    },
    {
      "epoch": 5.040704070407041,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027936917962650413,
      "loss": 0.2304,
      "num_input_tokens_seen": 9668672,
      "step": 45820
    },
    {
      "epoch": 5.041254125412541,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02793618906507489,
      "loss": 0.233,
      "num_input_tokens_seen": 9669728,
      "step": 45825
    },
    {
      "epoch": 5.041804180418042,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02793546004827272,
      "loss": 0.2309,
      "num_input_tokens_seen": 9670816,
      "step": 45830
    },
    {
      "epoch": 5.042354235423542,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.027934730912250624,
      "loss": 0.2293,
      "num_input_tokens_seen": 9671936,
      "step": 45835
    },
    {
      "epoch": 5.042904290429043,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027934001657015317,
      "loss": 0.2294,
      "num_input_tokens_seen": 9672960,
      "step": 45840
    },
    {
      "epoch": 5.043454345434544,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.027933272282573526,
      "loss": 0.2299,
      "num_input_tokens_seen": 9674016,
      "step": 45845
    },
    {
      "epoch": 5.044004400440044,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.02793254278893197,
      "loss": 0.231,
      "num_input_tokens_seen": 9675008,
      "step": 45850
    },
    {
      "epoch": 5.044554455445544,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.027931813176097366,
      "loss": 0.23,
      "num_input_tokens_seen": 9676032,
      "step": 45855
    },
    {
      "epoch": 5.045104510451045,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02793108344407645,
      "loss": 0.232,
      "num_input_tokens_seen": 9677120,
      "step": 45860
    },
    {
      "epoch": 5.0456545654565454,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.027930353592875948,
      "loss": 0.2295,
      "num_input_tokens_seen": 9678144,
      "step": 45865
    },
    {
      "epoch": 5.0462046204620465,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.02792962362250257,
      "loss": 0.2305,
      "num_input_tokens_seen": 9679232,
      "step": 45870
    },
    {
      "epoch": 5.046754675467547,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.027928893532963066,
      "loss": 0.2331,
      "num_input_tokens_seen": 9680288,
      "step": 45875
    },
    {
      "epoch": 5.047304730473047,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.027928163324264146,
      "loss": 0.2331,
      "num_input_tokens_seen": 9681408,
      "step": 45880
    },
    {
      "epoch": 5.047854785478548,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.027927432996412555,
      "loss": 0.2309,
      "num_input_tokens_seen": 9682528,
      "step": 45885
    },
    {
      "epoch": 5.048404840484048,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.027926702549415015,
      "loss": 0.233,
      "num_input_tokens_seen": 9683584,
      "step": 45890
    },
    {
      "epoch": 5.048954895489549,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.027925971983278258,
      "loss": 0.2325,
      "num_input_tokens_seen": 9684608,
      "step": 45895
    },
    {
      "epoch": 5.0495049504950495,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02792524129800902,
      "loss": 0.232,
      "num_input_tokens_seen": 9685696,
      "step": 45900
    },
    {
      "epoch": 5.05005500550055,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02792451049361404,
      "loss": 0.2299,
      "num_input_tokens_seen": 9686784,
      "step": 45905
    },
    {
      "epoch": 5.050605060506051,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02792377957010005,
      "loss": 0.2257,
      "num_input_tokens_seen": 9687840,
      "step": 45910
    },
    {
      "epoch": 5.051155115511551,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.02792304852747378,
      "loss": 0.2305,
      "num_input_tokens_seen": 9688864,
      "step": 45915
    },
    {
      "epoch": 5.051705170517051,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.027922317365741967,
      "loss": 0.2259,
      "num_input_tokens_seen": 9689920,
      "step": 45920
    },
    {
      "epoch": 5.052255225522552,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.027921586084911364,
      "loss": 0.2261,
      "num_input_tokens_seen": 9690976,
      "step": 45925
    },
    {
      "epoch": 5.052805280528053,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.0279208546849887,
      "loss": 0.2336,
      "num_input_tokens_seen": 9691968,
      "step": 45930
    },
    {
      "epoch": 5.053355335533554,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.02792012316598072,
      "loss": 0.2384,
      "num_input_tokens_seen": 9693024,
      "step": 45935
    },
    {
      "epoch": 5.053905390539054,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02791939152789416,
      "loss": 0.2315,
      "num_input_tokens_seen": 9694144,
      "step": 45940
    },
    {
      "epoch": 5.054455445544554,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.027918659770735772,
      "loss": 0.2315,
      "num_input_tokens_seen": 9695232,
      "step": 45945
    },
    {
      "epoch": 5.055005500550055,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.027917927894512296,
      "loss": 0.2393,
      "num_input_tokens_seen": 9696288,
      "step": 45950
    },
    {
      "epoch": 5.055555555555555,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.027917195899230477,
      "loss": 0.2345,
      "num_input_tokens_seen": 9697344,
      "step": 45955
    },
    {
      "epoch": 5.0561056105610565,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.02791646378489706,
      "loss": 0.2324,
      "num_input_tokens_seen": 9698432,
      "step": 45960
    },
    {
      "epoch": 5.056655665566557,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.027915731551518795,
      "loss": 0.2262,
      "num_input_tokens_seen": 9699488,
      "step": 45965
    },
    {
      "epoch": 5.057205720572057,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.02791499919910243,
      "loss": 0.2272,
      "num_input_tokens_seen": 9700544,
      "step": 45970
    },
    {
      "epoch": 5.057755775577558,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02791426672765471,
      "loss": 0.2318,
      "num_input_tokens_seen": 9701696,
      "step": 45975
    },
    {
      "epoch": 5.058305830583058,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.0279135341371824,
      "loss": 0.2376,
      "num_input_tokens_seen": 9702720,
      "step": 45980
    },
    {
      "epoch": 5.058855885588559,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.027912801427692237,
      "loss": 0.2345,
      "num_input_tokens_seen": 9703680,
      "step": 45985
    },
    {
      "epoch": 5.0594059405940595,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.027912068599190976,
      "loss": 0.2353,
      "num_input_tokens_seen": 9704768,
      "step": 45990
    },
    {
      "epoch": 5.05995599559956,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.027911335651685378,
      "loss": 0.2285,
      "num_input_tokens_seen": 9705792,
      "step": 45995
    },
    {
      "epoch": 5.060506050605061,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.027910602585182194,
      "loss": 0.2336,
      "num_input_tokens_seen": 9706880,
      "step": 46000
    },
    {
      "epoch": 5.061056105610561,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.027909869399688186,
      "loss": 0.2311,
      "num_input_tokens_seen": 9708000,
      "step": 46005
    },
    {
      "epoch": 5.061606160616061,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0279091360952101,
      "loss": 0.2321,
      "num_input_tokens_seen": 9709024,
      "step": 46010
    },
    {
      "epoch": 5.062156215621562,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.027908402671754703,
      "loss": 0.2352,
      "num_input_tokens_seen": 9709984,
      "step": 46015
    },
    {
      "epoch": 5.0627062706270625,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.027907669129328756,
      "loss": 0.2341,
      "num_input_tokens_seen": 9710944,
      "step": 46020
    },
    {
      "epoch": 5.063256325632564,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027906935467939008,
      "loss": 0.2309,
      "num_input_tokens_seen": 9712000,
      "step": 46025
    },
    {
      "epoch": 5.063806380638064,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02790620168759224,
      "loss": 0.2283,
      "num_input_tokens_seen": 9713024,
      "step": 46030
    },
    {
      "epoch": 5.064356435643564,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027905467788295196,
      "loss": 0.2304,
      "num_input_tokens_seen": 9714048,
      "step": 46035
    },
    {
      "epoch": 5.064906490649065,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.02790473377005465,
      "loss": 0.2325,
      "num_input_tokens_seen": 9715168,
      "step": 46040
    },
    {
      "epoch": 5.065456545654565,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.027903999632877364,
      "loss": 0.2294,
      "num_input_tokens_seen": 9716256,
      "step": 46045
    },
    {
      "epoch": 5.066006600660066,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.027903265376770108,
      "loss": 0.2315,
      "num_input_tokens_seen": 9717344,
      "step": 46050
    },
    {
      "epoch": 5.066556655665567,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.027902531001739644,
      "loss": 0.232,
      "num_input_tokens_seen": 9718432,
      "step": 46055
    },
    {
      "epoch": 5.067106710671067,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.027901796507792745,
      "loss": 0.233,
      "num_input_tokens_seen": 9719488,
      "step": 46060
    },
    {
      "epoch": 5.067656765676568,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.02790106189493618,
      "loss": 0.232,
      "num_input_tokens_seen": 9720512,
      "step": 46065
    },
    {
      "epoch": 5.068206820682068,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.027900327163176718,
      "loss": 0.2293,
      "num_input_tokens_seen": 9721632,
      "step": 46070
    },
    {
      "epoch": 5.068756875687569,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.027899592312521128,
      "loss": 0.2304,
      "num_input_tokens_seen": 9722656,
      "step": 46075
    },
    {
      "epoch": 5.069306930693069,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.027898857342976185,
      "loss": 0.2314,
      "num_input_tokens_seen": 9723744,
      "step": 46080
    },
    {
      "epoch": 5.06985698569857,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.027898122254548664,
      "loss": 0.2325,
      "num_input_tokens_seen": 9724832,
      "step": 46085
    },
    {
      "epoch": 5.070407040704071,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.02789738704724534,
      "loss": 0.2314,
      "num_input_tokens_seen": 9725984,
      "step": 46090
    },
    {
      "epoch": 5.070957095709571,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02789665172107299,
      "loss": 0.2293,
      "num_input_tokens_seen": 9727104,
      "step": 46095
    },
    {
      "epoch": 5.071507150715071,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.027895916276038386,
      "loss": 0.2299,
      "num_input_tokens_seen": 9728096,
      "step": 46100
    },
    {
      "epoch": 5.072057205720572,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02789518071214831,
      "loss": 0.2314,
      "num_input_tokens_seen": 9729216,
      "step": 46105
    },
    {
      "epoch": 5.072607260726072,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.027894445029409545,
      "loss": 0.2335,
      "num_input_tokens_seen": 9730304,
      "step": 46110
    },
    {
      "epoch": 5.0731573157315735,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.027893709227828863,
      "loss": 0.2304,
      "num_input_tokens_seen": 9731328,
      "step": 46115
    },
    {
      "epoch": 5.073707370737074,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027892973307413055,
      "loss": 0.232,
      "num_input_tokens_seen": 9732320,
      "step": 46120
    },
    {
      "epoch": 5.074257425742574,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027892237268168894,
      "loss": 0.2325,
      "num_input_tokens_seen": 9733312,
      "step": 46125
    },
    {
      "epoch": 5.074807480748075,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.027891501110103172,
      "loss": 0.2288,
      "num_input_tokens_seen": 9734368,
      "step": 46130
    },
    {
      "epoch": 5.075357535753575,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.027890764833222666,
      "loss": 0.2346,
      "num_input_tokens_seen": 9735456,
      "step": 46135
    },
    {
      "epoch": 5.075907590759076,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.02789002843753417,
      "loss": 0.2314,
      "num_input_tokens_seen": 9736576,
      "step": 46140
    },
    {
      "epoch": 5.0764576457645765,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.027889291923044467,
      "loss": 0.2315,
      "num_input_tokens_seen": 9737632,
      "step": 46145
    },
    {
      "epoch": 5.077007700770077,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.027888555289760347,
      "loss": 0.2314,
      "num_input_tokens_seen": 9738720,
      "step": 46150
    },
    {
      "epoch": 5.077557755775578,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.027887818537688586,
      "loss": 0.2299,
      "num_input_tokens_seen": 9739744,
      "step": 46155
    },
    {
      "epoch": 5.078107810781078,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.027887081666836,
      "loss": 0.2314,
      "num_input_tokens_seen": 9740832,
      "step": 46160
    },
    {
      "epoch": 5.078657865786579,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02788634467720936,
      "loss": 0.2319,
      "num_input_tokens_seen": 9741888,
      "step": 46165
    },
    {
      "epoch": 5.079207920792079,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.027885607568815463,
      "loss": 0.2314,
      "num_input_tokens_seen": 9742976,
      "step": 46170
    },
    {
      "epoch": 5.0797579757975795,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027884870341661108,
      "loss": 0.2298,
      "num_input_tokens_seen": 9744032,
      "step": 46175
    },
    {
      "epoch": 5.080308030803081,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.027884132995753083,
      "loss": 0.232,
      "num_input_tokens_seen": 9745056,
      "step": 46180
    },
    {
      "epoch": 5.080858085808581,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027883395531098188,
      "loss": 0.232,
      "num_input_tokens_seen": 9746208,
      "step": 46185
    },
    {
      "epoch": 5.081408140814081,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.027882657947703216,
      "loss": 0.2345,
      "num_input_tokens_seen": 9747200,
      "step": 46190
    },
    {
      "epoch": 5.081958195819582,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.027881920245574968,
      "loss": 0.2303,
      "num_input_tokens_seen": 9748192,
      "step": 46195
    },
    {
      "epoch": 5.082508250825082,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.027881182424720242,
      "loss": 0.234,
      "num_input_tokens_seen": 9749248,
      "step": 46200
    },
    {
      "epoch": 5.083058305830583,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.02788044448514584,
      "loss": 0.2324,
      "num_input_tokens_seen": 9750304,
      "step": 46205
    },
    {
      "epoch": 5.083608360836084,
      "grad_norm": 0.0006866455078125,
      "learning_rate": 0.027879706426858563,
      "loss": 0.2319,
      "num_input_tokens_seen": 9751360,
      "step": 46210
    },
    {
      "epoch": 5.084158415841584,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.027878968249865212,
      "loss": 0.2319,
      "num_input_tokens_seen": 9752384,
      "step": 46215
    },
    {
      "epoch": 5.084708470847085,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02787822995417259,
      "loss": 0.2314,
      "num_input_tokens_seen": 9753472,
      "step": 46220
    },
    {
      "epoch": 5.085258525852585,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.027877491539787504,
      "loss": 0.2319,
      "num_input_tokens_seen": 9754464,
      "step": 46225
    },
    {
      "epoch": 5.085808580858086,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.027876753006716753,
      "loss": 0.2313,
      "num_input_tokens_seen": 9755520,
      "step": 46230
    },
    {
      "epoch": 5.086358635863586,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.02787601435496715,
      "loss": 0.2298,
      "num_input_tokens_seen": 9756640,
      "step": 46235
    },
    {
      "epoch": 5.086908690869087,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.027875275584545504,
      "loss": 0.233,
      "num_input_tokens_seen": 9757664,
      "step": 46240
    },
    {
      "epoch": 5.087458745874588,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.027874536695458623,
      "loss": 0.2335,
      "num_input_tokens_seen": 9758656,
      "step": 46245
    },
    {
      "epoch": 5.088008800880088,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02787379768771331,
      "loss": 0.2304,
      "num_input_tokens_seen": 9759680,
      "step": 46250
    },
    {
      "epoch": 5.088558855885589,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.027873058561316378,
      "loss": 0.2289,
      "num_input_tokens_seen": 9760736,
      "step": 46255
    },
    {
      "epoch": 5.089108910891089,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02787231931627465,
      "loss": 0.2273,
      "num_input_tokens_seen": 9761792,
      "step": 46260
    },
    {
      "epoch": 5.089658965896589,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.027871579952594927,
      "loss": 0.2332,
      "num_input_tokens_seen": 9762848,
      "step": 46265
    },
    {
      "epoch": 5.0902090209020905,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02787084047028403,
      "loss": 0.2286,
      "num_input_tokens_seen": 9763904,
      "step": 46270
    },
    {
      "epoch": 5.090759075907591,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.027870100869348775,
      "loss": 0.2394,
      "num_input_tokens_seen": 9764896,
      "step": 46275
    },
    {
      "epoch": 5.091309130913091,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.02786936114979597,
      "loss": 0.2326,
      "num_input_tokens_seen": 9765952,
      "step": 46280
    },
    {
      "epoch": 5.091859185918592,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.027868621311632442,
      "loss": 0.2263,
      "num_input_tokens_seen": 9766976,
      "step": 46285
    },
    {
      "epoch": 5.092409240924092,
      "grad_norm": 0.0026397705078125,
      "learning_rate": 0.027867881354865003,
      "loss": 0.2284,
      "num_input_tokens_seen": 9768128,
      "step": 46290
    },
    {
      "epoch": 5.092959295929593,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02786714127950048,
      "loss": 0.2326,
      "num_input_tokens_seen": 9769184,
      "step": 46295
    },
    {
      "epoch": 5.0935093509350935,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02786640108554569,
      "loss": 0.2337,
      "num_input_tokens_seen": 9770240,
      "step": 46300
    },
    {
      "epoch": 5.094059405940594,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.02786566077300745,
      "loss": 0.2351,
      "num_input_tokens_seen": 9771232,
      "step": 46305
    },
    {
      "epoch": 5.094609460946095,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0278649203418926,
      "loss": 0.2304,
      "num_input_tokens_seen": 9772288,
      "step": 46310
    },
    {
      "epoch": 5.095159515951595,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.02786417979220794,
      "loss": 0.2336,
      "num_input_tokens_seen": 9773312,
      "step": 46315
    },
    {
      "epoch": 5.095709570957096,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.02786343912396031,
      "loss": 0.2309,
      "num_input_tokens_seen": 9774368,
      "step": 46320
    },
    {
      "epoch": 5.096259625962596,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.02786269833715654,
      "loss": 0.233,
      "num_input_tokens_seen": 9775424,
      "step": 46325
    },
    {
      "epoch": 5.0968096809680965,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.027861957431803447,
      "loss": 0.2314,
      "num_input_tokens_seen": 9776416,
      "step": 46330
    },
    {
      "epoch": 5.097359735973598,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.027861216407907865,
      "loss": 0.2335,
      "num_input_tokens_seen": 9777440,
      "step": 46335
    },
    {
      "epoch": 5.097909790979098,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.027860475265476622,
      "loss": 0.2298,
      "num_input_tokens_seen": 9778432,
      "step": 46340
    },
    {
      "epoch": 5.098459845984599,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02785973400451655,
      "loss": 0.2293,
      "num_input_tokens_seen": 9779488,
      "step": 46345
    },
    {
      "epoch": 5.099009900990099,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.027858992625034485,
      "loss": 0.2335,
      "num_input_tokens_seen": 9780512,
      "step": 46350
    },
    {
      "epoch": 5.099559955995599,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.027858251127037246,
      "loss": 0.2288,
      "num_input_tokens_seen": 9781568,
      "step": 46355
    },
    {
      "epoch": 5.1001100110011,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027857509510531685,
      "loss": 0.2309,
      "num_input_tokens_seen": 9782592,
      "step": 46360
    },
    {
      "epoch": 5.100660066006601,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.02785676777552462,
      "loss": 0.2304,
      "num_input_tokens_seen": 9783680,
      "step": 46365
    },
    {
      "epoch": 5.101210121012101,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.027856025922022902,
      "loss": 0.2345,
      "num_input_tokens_seen": 9784768,
      "step": 46370
    },
    {
      "epoch": 5.101760176017602,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.027855283950033363,
      "loss": 0.2309,
      "num_input_tokens_seen": 9785824,
      "step": 46375
    },
    {
      "epoch": 5.102310231023102,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.027854541859562838,
      "loss": 0.233,
      "num_input_tokens_seen": 9786880,
      "step": 46380
    },
    {
      "epoch": 5.102860286028603,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.027853799650618172,
      "loss": 0.2313,
      "num_input_tokens_seen": 9787968,
      "step": 46385
    },
    {
      "epoch": 5.103410341034103,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.0278530573232062,
      "loss": 0.2314,
      "num_input_tokens_seen": 9789024,
      "step": 46390
    },
    {
      "epoch": 5.103960396039604,
      "grad_norm": 0.007598876953125,
      "learning_rate": 0.027852314877333764,
      "loss": 0.231,
      "num_input_tokens_seen": 9790080,
      "step": 46395
    },
    {
      "epoch": 5.104510451045105,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.027851572313007708,
      "loss": 0.2294,
      "num_input_tokens_seen": 9791168,
      "step": 46400
    },
    {
      "epoch": 5.105060506050605,
      "grad_norm": 0.0079345703125,
      "learning_rate": 0.02785082963023488,
      "loss": 0.2326,
      "num_input_tokens_seen": 9792192,
      "step": 46405
    },
    {
      "epoch": 5.105610561056106,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.02785008682902212,
      "loss": 0.2285,
      "num_input_tokens_seen": 9793312,
      "step": 46410
    },
    {
      "epoch": 5.106160616061606,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.027849343909376278,
      "loss": 0.2322,
      "num_input_tokens_seen": 9794368,
      "step": 46415
    },
    {
      "epoch": 5.106710671067106,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.027848600871304197,
      "loss": 0.2367,
      "num_input_tokens_seen": 9795456,
      "step": 46420
    },
    {
      "epoch": 5.1072607260726075,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.027847857714812723,
      "loss": 0.2314,
      "num_input_tokens_seen": 9796576,
      "step": 46425
    },
    {
      "epoch": 5.107810781078108,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02784711443990872,
      "loss": 0.2304,
      "num_input_tokens_seen": 9797632,
      "step": 46430
    },
    {
      "epoch": 5.108360836083609,
      "grad_norm": 0.000732421875,
      "learning_rate": 0.02784637104659902,
      "loss": 0.2319,
      "num_input_tokens_seen": 9798688,
      "step": 46435
    },
    {
      "epoch": 5.108910891089109,
      "grad_norm": 0.000492095947265625,
      "learning_rate": 0.027845627534890484,
      "loss": 0.2303,
      "num_input_tokens_seen": 9799680,
      "step": 46440
    },
    {
      "epoch": 5.109460946094609,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.027844883904789966,
      "loss": 0.2324,
      "num_input_tokens_seen": 9800800,
      "step": 46445
    },
    {
      "epoch": 5.11001100110011,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.027844140156304313,
      "loss": 0.2308,
      "num_input_tokens_seen": 9801856,
      "step": 46450
    },
    {
      "epoch": 5.1105610561056105,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.027843396289440384,
      "loss": 0.2288,
      "num_input_tokens_seen": 9802912,
      "step": 46455
    },
    {
      "epoch": 5.111111111111111,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027842652304205035,
      "loss": 0.2324,
      "num_input_tokens_seen": 9804000,
      "step": 46460
    },
    {
      "epoch": 5.111661166116612,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.027841908200605118,
      "loss": 0.2314,
      "num_input_tokens_seen": 9805024,
      "step": 46465
    },
    {
      "epoch": 5.112211221122112,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.027841163978647502,
      "loss": 0.2303,
      "num_input_tokens_seen": 9806080,
      "step": 46470
    },
    {
      "epoch": 5.112761276127613,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.027840419638339036,
      "loss": 0.2293,
      "num_input_tokens_seen": 9807136,
      "step": 46475
    },
    {
      "epoch": 5.113311331133113,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027839675179686583,
      "loss": 0.2304,
      "num_input_tokens_seen": 9808160,
      "step": 46480
    },
    {
      "epoch": 5.1138613861386135,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.027838930602697005,
      "loss": 0.2325,
      "num_input_tokens_seen": 9809248,
      "step": 46485
    },
    {
      "epoch": 5.114411441144115,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.027838185907377168,
      "loss": 0.233,
      "num_input_tokens_seen": 9810240,
      "step": 46490
    },
    {
      "epoch": 5.114961496149615,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027837441093733928,
      "loss": 0.2283,
      "num_input_tokens_seen": 9811296,
      "step": 46495
    },
    {
      "epoch": 5.115511551155116,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.027836696161774154,
      "loss": 0.2303,
      "num_input_tokens_seen": 9812320,
      "step": 46500
    },
    {
      "epoch": 5.116061606160616,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027835951111504713,
      "loss": 0.2314,
      "num_input_tokens_seen": 9813376,
      "step": 46505
    },
    {
      "epoch": 5.116611661166116,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02783520594293247,
      "loss": 0.2335,
      "num_input_tokens_seen": 9814400,
      "step": 46510
    },
    {
      "epoch": 5.117161716171617,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.02783446065606429,
      "loss": 0.2335,
      "num_input_tokens_seen": 9815488,
      "step": 46515
    },
    {
      "epoch": 5.117711771177118,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.027833715250907047,
      "loss": 0.2293,
      "num_input_tokens_seen": 9816480,
      "step": 46520
    },
    {
      "epoch": 5.118261826182618,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.027832969727467606,
      "loss": 0.2324,
      "num_input_tokens_seen": 9817568,
      "step": 46525
    },
    {
      "epoch": 5.118811881188119,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027832224085752845,
      "loss": 0.2319,
      "num_input_tokens_seen": 9818656,
      "step": 46530
    },
    {
      "epoch": 5.119361936193619,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02783147832576963,
      "loss": 0.2314,
      "num_input_tokens_seen": 9819744,
      "step": 46535
    },
    {
      "epoch": 5.11991199119912,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.027830732447524837,
      "loss": 0.2309,
      "num_input_tokens_seen": 9820800,
      "step": 46540
    },
    {
      "epoch": 5.12046204620462,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.027829986451025336,
      "loss": 0.2329,
      "num_input_tokens_seen": 9821856,
      "step": 46545
    },
    {
      "epoch": 5.121012101210121,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 0.02782924033627801,
      "loss": 0.2308,
      "num_input_tokens_seen": 9822880,
      "step": 46550
    },
    {
      "epoch": 5.121562156215622,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.027828494103289733,
      "loss": 0.2308,
      "num_input_tokens_seen": 9823968,
      "step": 46555
    },
    {
      "epoch": 5.122112211221122,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.027827747752067377,
      "loss": 0.2314,
      "num_input_tokens_seen": 9825056,
      "step": 46560
    },
    {
      "epoch": 5.122662266226623,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02782700128261783,
      "loss": 0.2308,
      "num_input_tokens_seen": 9826144,
      "step": 46565
    },
    {
      "epoch": 5.123212321232123,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027826254694947965,
      "loss": 0.2303,
      "num_input_tokens_seen": 9827200,
      "step": 46570
    },
    {
      "epoch": 5.123762376237623,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.027825507989064666,
      "loss": 0.2308,
      "num_input_tokens_seen": 9828224,
      "step": 46575
    },
    {
      "epoch": 5.1243124312431245,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.027824761164974815,
      "loss": 0.2303,
      "num_input_tokens_seen": 9829248,
      "step": 46580
    },
    {
      "epoch": 5.124862486248625,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02782401422268529,
      "loss": 0.2314,
      "num_input_tokens_seen": 9830304,
      "step": 46585
    },
    {
      "epoch": 5.125412541254126,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.027823267162202982,
      "loss": 0.2329,
      "num_input_tokens_seen": 9831392,
      "step": 46590
    },
    {
      "epoch": 5.125962596259626,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.027822519983534775,
      "loss": 0.2314,
      "num_input_tokens_seen": 9832384,
      "step": 46595
    },
    {
      "epoch": 5.126512651265126,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.027821772686687552,
      "loss": 0.2308,
      "num_input_tokens_seen": 9833440,
      "step": 46600
    },
    {
      "epoch": 5.127062706270627,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.027821025271668205,
      "loss": 0.2314,
      "num_input_tokens_seen": 9834528,
      "step": 46605
    },
    {
      "epoch": 5.1276127612761275,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.027820277738483622,
      "loss": 0.2283,
      "num_input_tokens_seen": 9835616,
      "step": 46610
    },
    {
      "epoch": 5.128162816281628,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.027819530087140686,
      "loss": 0.2315,
      "num_input_tokens_seen": 9836672,
      "step": 46615
    },
    {
      "epoch": 5.128712871287129,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.0278187823176463,
      "loss": 0.2357,
      "num_input_tokens_seen": 9837728,
      "step": 46620
    },
    {
      "epoch": 5.129262926292629,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.027818034430007337,
      "loss": 0.233,
      "num_input_tokens_seen": 9838688,
      "step": 46625
    },
    {
      "epoch": 5.12981298129813,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.027817286424230706,
      "loss": 0.2304,
      "num_input_tokens_seen": 9839712,
      "step": 46630
    },
    {
      "epoch": 5.13036303630363,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.0278165383003233,
      "loss": 0.2299,
      "num_input_tokens_seen": 9840768,
      "step": 46635
    },
    {
      "epoch": 5.1309130913091305,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.027815790058292007,
      "loss": 0.2299,
      "num_input_tokens_seen": 9841824,
      "step": 46640
    },
    {
      "epoch": 5.131463146314632,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02781504169814373,
      "loss": 0.2283,
      "num_input_tokens_seen": 9842944,
      "step": 46645
    },
    {
      "epoch": 5.132013201320132,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02781429321988536,
      "loss": 0.2336,
      "num_input_tokens_seen": 9844000,
      "step": 46650
    },
    {
      "epoch": 5.132563256325633,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.0278135446235238,
      "loss": 0.2273,
      "num_input_tokens_seen": 9845024,
      "step": 46655
    },
    {
      "epoch": 5.133113311331133,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.02781279590906595,
      "loss": 0.2346,
      "num_input_tokens_seen": 9846080,
      "step": 46660
    },
    {
      "epoch": 5.133663366336633,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0278120470765187,
      "loss": 0.2357,
      "num_input_tokens_seen": 9847136,
      "step": 46665
    },
    {
      "epoch": 5.134213421342134,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027811298125888964,
      "loss": 0.2309,
      "num_input_tokens_seen": 9848256,
      "step": 46670
    },
    {
      "epoch": 5.134763476347635,
      "grad_norm": 0.000797271728515625,
      "learning_rate": 0.027810549057183644,
      "loss": 0.2319,
      "num_input_tokens_seen": 9849344,
      "step": 46675
    },
    {
      "epoch": 5.135313531353136,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.027809799870409633,
      "loss": 0.233,
      "num_input_tokens_seen": 9850464,
      "step": 46680
    },
    {
      "epoch": 5.135863586358636,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.027809050565573853,
      "loss": 0.2278,
      "num_input_tokens_seen": 9851520,
      "step": 46685
    },
    {
      "epoch": 5.136413641364136,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.027808301142683196,
      "loss": 0.234,
      "num_input_tokens_seen": 9852512,
      "step": 46690
    },
    {
      "epoch": 5.136963696369637,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02780755160174457,
      "loss": 0.2304,
      "num_input_tokens_seen": 9853504,
      "step": 46695
    },
    {
      "epoch": 5.137513751375137,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027806801942764893,
      "loss": 0.2309,
      "num_input_tokens_seen": 9854592,
      "step": 46700
    },
    {
      "epoch": 5.138063806380638,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02780605216575106,
      "loss": 0.2314,
      "num_input_tokens_seen": 9855648,
      "step": 46705
    },
    {
      "epoch": 5.138613861386139,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.027805302270709997,
      "loss": 0.233,
      "num_input_tokens_seen": 9856704,
      "step": 46710
    },
    {
      "epoch": 5.139163916391639,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.027804552257648604,
      "loss": 0.2298,
      "num_input_tokens_seen": 9857728,
      "step": 46715
    },
    {
      "epoch": 5.13971397139714,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.027803802126573796,
      "loss": 0.2335,
      "num_input_tokens_seen": 9858752,
      "step": 46720
    },
    {
      "epoch": 5.14026402640264,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02780305187749249,
      "loss": 0.2298,
      "num_input_tokens_seen": 9859776,
      "step": 46725
    },
    {
      "epoch": 5.1408140814081404,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.027802301510411596,
      "loss": 0.2314,
      "num_input_tokens_seen": 9860768,
      "step": 46730
    },
    {
      "epoch": 5.1413641364136415,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.027801551025338034,
      "loss": 0.233,
      "num_input_tokens_seen": 9861824,
      "step": 46735
    },
    {
      "epoch": 5.141914191419142,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.02780080042227872,
      "loss": 0.2303,
      "num_input_tokens_seen": 9862912,
      "step": 46740
    },
    {
      "epoch": 5.142464246424643,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.027800049701240566,
      "loss": 0.2319,
      "num_input_tokens_seen": 9864000,
      "step": 46745
    },
    {
      "epoch": 5.143014301430143,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0277992988622305,
      "loss": 0.234,
      "num_input_tokens_seen": 9864992,
      "step": 46750
    },
    {
      "epoch": 5.143564356435643,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.027798547905255436,
      "loss": 0.234,
      "num_input_tokens_seen": 9866016,
      "step": 46755
    },
    {
      "epoch": 5.144114411441144,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.0277977968303223,
      "loss": 0.232,
      "num_input_tokens_seen": 9867104,
      "step": 46760
    },
    {
      "epoch": 5.1446644664466445,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02779704563743801,
      "loss": 0.2294,
      "num_input_tokens_seen": 9868160,
      "step": 46765
    },
    {
      "epoch": 5.145214521452146,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.027796294326609492,
      "loss": 0.2314,
      "num_input_tokens_seen": 9869184,
      "step": 46770
    },
    {
      "epoch": 5.145764576457646,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02779554289784367,
      "loss": 0.2304,
      "num_input_tokens_seen": 9870240,
      "step": 46775
    },
    {
      "epoch": 5.146314631463146,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.027794791351147468,
      "loss": 0.2335,
      "num_input_tokens_seen": 9871328,
      "step": 46780
    },
    {
      "epoch": 5.146864686468647,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02779403968652781,
      "loss": 0.232,
      "num_input_tokens_seen": 9872384,
      "step": 46785
    },
    {
      "epoch": 5.147414741474147,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.027793287903991628,
      "loss": 0.2293,
      "num_input_tokens_seen": 9873376,
      "step": 46790
    },
    {
      "epoch": 5.1479647964796476,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.027792536003545858,
      "loss": 0.2335,
      "num_input_tokens_seen": 9874400,
      "step": 46795
    },
    {
      "epoch": 5.148514851485149,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.027791783985197414,
      "loss": 0.2304,
      "num_input_tokens_seen": 9875360,
      "step": 46800
    },
    {
      "epoch": 5.149064906490649,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.02779103184895324,
      "loss": 0.233,
      "num_input_tokens_seen": 9876480,
      "step": 46805
    },
    {
      "epoch": 5.14961496149615,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.02779027959482026,
      "loss": 0.2309,
      "num_input_tokens_seen": 9877504,
      "step": 46810
    },
    {
      "epoch": 5.15016501650165,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.02778952722280541,
      "loss": 0.2319,
      "num_input_tokens_seen": 9878528,
      "step": 46815
    },
    {
      "epoch": 5.15071507150715,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.02778877473291563,
      "loss": 0.2309,
      "num_input_tokens_seen": 9879552,
      "step": 46820
    },
    {
      "epoch": 5.1512651265126514,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.02778802212515784,
      "loss": 0.2346,
      "num_input_tokens_seen": 9880640,
      "step": 46825
    },
    {
      "epoch": 5.151815181518152,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.027787269399538995,
      "loss": 0.2325,
      "num_input_tokens_seen": 9881664,
      "step": 46830
    },
    {
      "epoch": 5.152365236523653,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.027786516556066024,
      "loss": 0.2298,
      "num_input_tokens_seen": 9882688,
      "step": 46835
    },
    {
      "epoch": 5.152915291529153,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.02778576359474586,
      "loss": 0.2314,
      "num_input_tokens_seen": 9883744,
      "step": 46840
    },
    {
      "epoch": 5.153465346534653,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02778501051558545,
      "loss": 0.2293,
      "num_input_tokens_seen": 9884864,
      "step": 46845
    },
    {
      "epoch": 5.154015401540154,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.027784257318591736,
      "loss": 0.233,
      "num_input_tokens_seen": 9885856,
      "step": 46850
    },
    {
      "epoch": 5.1545654565456545,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.027783504003771654,
      "loss": 0.2303,
      "num_input_tokens_seen": 9886944,
      "step": 46855
    },
    {
      "epoch": 5.1551155115511555,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.02778275057113215,
      "loss": 0.2325,
      "num_input_tokens_seen": 9888000,
      "step": 46860
    },
    {
      "epoch": 5.155665566556656,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02778199702068017,
      "loss": 0.2314,
      "num_input_tokens_seen": 9889120,
      "step": 46865
    },
    {
      "epoch": 5.156215621562156,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.027781243352422658,
      "loss": 0.2308,
      "num_input_tokens_seen": 9890208,
      "step": 46870
    },
    {
      "epoch": 5.156765676567657,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02778048956636655,
      "loss": 0.2319,
      "num_input_tokens_seen": 9891264,
      "step": 46875
    },
    {
      "epoch": 5.157315731573157,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02777973566251881,
      "loss": 0.2293,
      "num_input_tokens_seen": 9892256,
      "step": 46880
    },
    {
      "epoch": 5.1578657865786575,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.027778981640886376,
      "loss": 0.2308,
      "num_input_tokens_seen": 9893312,
      "step": 46885
    },
    {
      "epoch": 5.158415841584159,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0277782275014762,
      "loss": 0.2324,
      "num_input_tokens_seen": 9894368,
      "step": 46890
    },
    {
      "epoch": 5.158965896589659,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027777473244295237,
      "loss": 0.2308,
      "num_input_tokens_seen": 9895328,
      "step": 46895
    },
    {
      "epoch": 5.15951595159516,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.02777671886935043,
      "loss": 0.2329,
      "num_input_tokens_seen": 9896352,
      "step": 46900
    },
    {
      "epoch": 5.16006600660066,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.02777596437664874,
      "loss": 0.2324,
      "num_input_tokens_seen": 9897408,
      "step": 46905
    },
    {
      "epoch": 5.16061606160616,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.02777520976619711,
      "loss": 0.2314,
      "num_input_tokens_seen": 9898464,
      "step": 46910
    },
    {
      "epoch": 5.161166116611661,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.02777445503800251,
      "loss": 0.2314,
      "num_input_tokens_seen": 9899488,
      "step": 46915
    },
    {
      "epoch": 5.161716171617162,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02777370019207188,
      "loss": 0.2324,
      "num_input_tokens_seen": 9900576,
      "step": 46920
    },
    {
      "epoch": 5.162266226622663,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02777294522841219,
      "loss": 0.2319,
      "num_input_tokens_seen": 9901600,
      "step": 46925
    },
    {
      "epoch": 5.162816281628163,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.027772190147030393,
      "loss": 0.2313,
      "num_input_tokens_seen": 9902688,
      "step": 46930
    },
    {
      "epoch": 5.163366336633663,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.027771434947933445,
      "loss": 0.2298,
      "num_input_tokens_seen": 9903808,
      "step": 46935
    },
    {
      "epoch": 5.163916391639164,
      "grad_norm": 0.000675201416015625,
      "learning_rate": 0.02777067963112831,
      "loss": 0.2308,
      "num_input_tokens_seen": 9904864,
      "step": 46940
    },
    {
      "epoch": 5.164466446644664,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.02776992419662195,
      "loss": 0.2314,
      "num_input_tokens_seen": 9905888,
      "step": 46945
    },
    {
      "epoch": 5.165016501650165,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027769168644421326,
      "loss": 0.2303,
      "num_input_tokens_seen": 9906912,
      "step": 46950
    },
    {
      "epoch": 5.165566556655666,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.0277684129745334,
      "loss": 0.2324,
      "num_input_tokens_seen": 9908000,
      "step": 46955
    },
    {
      "epoch": 5.166116611661166,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02776765718696514,
      "loss": 0.2324,
      "num_input_tokens_seen": 9909088,
      "step": 46960
    },
    {
      "epoch": 5.166666666666667,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.027766901281723513,
      "loss": 0.2288,
      "num_input_tokens_seen": 9910144,
      "step": 46965
    },
    {
      "epoch": 5.167216721672167,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.027766145258815482,
      "loss": 0.2329,
      "num_input_tokens_seen": 9911264,
      "step": 46970
    },
    {
      "epoch": 5.167766776677667,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.027765389118248014,
      "loss": 0.2314,
      "num_input_tokens_seen": 9912416,
      "step": 46975
    },
    {
      "epoch": 5.1683168316831685,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.027764632860028077,
      "loss": 0.2309,
      "num_input_tokens_seen": 9913472,
      "step": 46980
    },
    {
      "epoch": 5.168866886688669,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.027763876484162647,
      "loss": 0.2298,
      "num_input_tokens_seen": 9914496,
      "step": 46985
    },
    {
      "epoch": 5.16941694169417,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027763119990658693,
      "loss": 0.2325,
      "num_input_tokens_seen": 9915584,
      "step": 46990
    },
    {
      "epoch": 5.16996699669967,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02776236337952319,
      "loss": 0.233,
      "num_input_tokens_seen": 9916608,
      "step": 46995
    },
    {
      "epoch": 5.17051705170517,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.027761606650763104,
      "loss": 0.2314,
      "num_input_tokens_seen": 9917632,
      "step": 47000
    },
    {
      "epoch": 5.171067106710671,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.027760849804385406,
      "loss": 0.2309,
      "num_input_tokens_seen": 9918688,
      "step": 47005
    },
    {
      "epoch": 5.1716171617161715,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02776009284039709,
      "loss": 0.2309,
      "num_input_tokens_seen": 9919776,
      "step": 47010
    },
    {
      "epoch": 5.172167216721673,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.027759335758805116,
      "loss": 0.2278,
      "num_input_tokens_seen": 9920864,
      "step": 47015
    },
    {
      "epoch": 5.172717271727173,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.027758578559616467,
      "loss": 0.2288,
      "num_input_tokens_seen": 9921888,
      "step": 47020
    },
    {
      "epoch": 5.173267326732673,
      "grad_norm": 0.0030517578125,
      "learning_rate": 0.02775782124283812,
      "loss": 0.2298,
      "num_input_tokens_seen": 9922944,
      "step": 47025
    },
    {
      "epoch": 5.173817381738174,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.02775706380847706,
      "loss": 0.2305,
      "num_input_tokens_seen": 9924000,
      "step": 47030
    },
    {
      "epoch": 5.174367436743674,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.02775630625654026,
      "loss": 0.2331,
      "num_input_tokens_seen": 9925120,
      "step": 47035
    },
    {
      "epoch": 5.174917491749175,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02775554858703471,
      "loss": 0.2294,
      "num_input_tokens_seen": 9926176,
      "step": 47040
    },
    {
      "epoch": 5.175467546754676,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02775479079996739,
      "loss": 0.2269,
      "num_input_tokens_seen": 9927136,
      "step": 47045
    },
    {
      "epoch": 5.176017601760176,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.027754032895345282,
      "loss": 0.2301,
      "num_input_tokens_seen": 9928224,
      "step": 47050
    },
    {
      "epoch": 5.176567656765677,
      "grad_norm": 0.0076904296875,
      "learning_rate": 0.02775327487317537,
      "loss": 0.2348,
      "num_input_tokens_seen": 9929312,
      "step": 47055
    },
    {
      "epoch": 5.177117711771177,
      "grad_norm": 0.013427734375,
      "learning_rate": 0.027752516733464648,
      "loss": 0.2334,
      "num_input_tokens_seen": 9930304,
      "step": 47060
    },
    {
      "epoch": 5.177667766776677,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.027751758476220096,
      "loss": 0.2342,
      "num_input_tokens_seen": 9931392,
      "step": 47065
    },
    {
      "epoch": 5.178217821782178,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0277510001014487,
      "loss": 0.2373,
      "num_input_tokens_seen": 9932448,
      "step": 47070
    },
    {
      "epoch": 5.178767876787679,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.027750241609157466,
      "loss": 0.2336,
      "num_input_tokens_seen": 9933536,
      "step": 47075
    },
    {
      "epoch": 5.17931793179318,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.027749482999353366,
      "loss": 0.2335,
      "num_input_tokens_seen": 9934528,
      "step": 47080
    },
    {
      "epoch": 5.17986798679868,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.0277487242720434,
      "loss": 0.2319,
      "num_input_tokens_seen": 9935616,
      "step": 47085
    },
    {
      "epoch": 5.18041804180418,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02774796542723456,
      "loss": 0.2325,
      "num_input_tokens_seen": 9936704,
      "step": 47090
    },
    {
      "epoch": 5.180968096809681,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.02774720646493384,
      "loss": 0.2309,
      "num_input_tokens_seen": 9937728,
      "step": 47095
    },
    {
      "epoch": 5.181518151815181,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.027746447385148237,
      "loss": 0.2319,
      "num_input_tokens_seen": 9938848,
      "step": 47100
    },
    {
      "epoch": 5.1820682068206825,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02774568818788474,
      "loss": 0.2314,
      "num_input_tokens_seen": 9939872,
      "step": 47105
    },
    {
      "epoch": 5.182618261826183,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.02774492887315035,
      "loss": 0.2303,
      "num_input_tokens_seen": 9940960,
      "step": 47110
    },
    {
      "epoch": 5.183168316831683,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.02774416944095207,
      "loss": 0.2314,
      "num_input_tokens_seen": 9941984,
      "step": 47115
    },
    {
      "epoch": 5.183718371837184,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.027743409891296894,
      "loss": 0.232,
      "num_input_tokens_seen": 9943008,
      "step": 47120
    },
    {
      "epoch": 5.184268426842684,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02774265022419183,
      "loss": 0.2294,
      "num_input_tokens_seen": 9944064,
      "step": 47125
    },
    {
      "epoch": 5.184818481848184,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.027741890439643868,
      "loss": 0.2341,
      "num_input_tokens_seen": 9945120,
      "step": 47130
    },
    {
      "epoch": 5.1853685368536855,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.027741130537660013,
      "loss": 0.2362,
      "num_input_tokens_seen": 9946176,
      "step": 47135
    },
    {
      "epoch": 5.185918591859186,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.027740370518247275,
      "loss": 0.2288,
      "num_input_tokens_seen": 9947168,
      "step": 47140
    },
    {
      "epoch": 5.186468646864687,
      "grad_norm": 0.0004520416259765625,
      "learning_rate": 0.027739610381412656,
      "loss": 0.2309,
      "num_input_tokens_seen": 9948192,
      "step": 47145
    },
    {
      "epoch": 5.187018701870187,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02773885012716316,
      "loss": 0.233,
      "num_input_tokens_seen": 9949280,
      "step": 47150
    },
    {
      "epoch": 5.187568756875687,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027738089755505797,
      "loss": 0.2283,
      "num_input_tokens_seen": 9950336,
      "step": 47155
    },
    {
      "epoch": 5.188118811881188,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02773732926644757,
      "loss": 0.2325,
      "num_input_tokens_seen": 9951360,
      "step": 47160
    },
    {
      "epoch": 5.1886688668866885,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.027736568659995492,
      "loss": 0.2314,
      "num_input_tokens_seen": 9952416,
      "step": 47165
    },
    {
      "epoch": 5.18921892189219,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.027735807936156574,
      "loss": 0.2278,
      "num_input_tokens_seen": 9953472,
      "step": 47170
    },
    {
      "epoch": 5.18976897689769,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.027735047094937824,
      "loss": 0.2325,
      "num_input_tokens_seen": 9954528,
      "step": 47175
    },
    {
      "epoch": 5.19031903190319,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.027734286136346254,
      "loss": 0.231,
      "num_input_tokens_seen": 9955584,
      "step": 47180
    },
    {
      "epoch": 5.190869086908691,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.027733525060388883,
      "loss": 0.232,
      "num_input_tokens_seen": 9956640,
      "step": 47185
    },
    {
      "epoch": 5.191419141914191,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.027732763867072716,
      "loss": 0.2316,
      "num_input_tokens_seen": 9957664,
      "step": 47190
    },
    {
      "epoch": 5.191969196919692,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.02773200255640478,
      "loss": 0.2326,
      "num_input_tokens_seen": 9958784,
      "step": 47195
    },
    {
      "epoch": 5.192519251925193,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.027731241128392082,
      "loss": 0.23,
      "num_input_tokens_seen": 9959840,
      "step": 47200
    },
    {
      "epoch": 5.193069306930693,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02773047958304165,
      "loss": 0.2301,
      "num_input_tokens_seen": 9960800,
      "step": 47205
    },
    {
      "epoch": 5.193619361936194,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.02772971792036049,
      "loss": 0.2322,
      "num_input_tokens_seen": 9961888,
      "step": 47210
    },
    {
      "epoch": 5.194169416941694,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.02772895614035563,
      "loss": 0.2337,
      "num_input_tokens_seen": 9962944,
      "step": 47215
    },
    {
      "epoch": 5.194719471947194,
      "grad_norm": 0.00726318359375,
      "learning_rate": 0.027728194243034087,
      "loss": 0.2255,
      "num_input_tokens_seen": 9964000,
      "step": 47220
    },
    {
      "epoch": 5.195269526952695,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.02772743222840289,
      "loss": 0.2276,
      "num_input_tokens_seen": 9965120,
      "step": 47225
    },
    {
      "epoch": 5.195819581958196,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.027726670096469053,
      "loss": 0.2349,
      "num_input_tokens_seen": 9966176,
      "step": 47230
    },
    {
      "epoch": 5.196369636963697,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.027725907847239606,
      "loss": 0.2229,
      "num_input_tokens_seen": 9967264,
      "step": 47235
    },
    {
      "epoch": 5.196919691969197,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.027725145480721573,
      "loss": 0.2315,
      "num_input_tokens_seen": 9968384,
      "step": 47240
    },
    {
      "epoch": 5.197469746974697,
      "grad_norm": 0.0084228515625,
      "learning_rate": 0.02772438299692198,
      "loss": 0.234,
      "num_input_tokens_seen": 9969408,
      "step": 47245
    },
    {
      "epoch": 5.198019801980198,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.02772362039584786,
      "loss": 0.2305,
      "num_input_tokens_seen": 9970432,
      "step": 47250
    },
    {
      "epoch": 5.198569856985698,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.027722857677506225,
      "loss": 0.2288,
      "num_input_tokens_seen": 9971488,
      "step": 47255
    },
    {
      "epoch": 5.1991199119911995,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.027722094841904125,
      "loss": 0.2414,
      "num_input_tokens_seen": 9972576,
      "step": 47260
    },
    {
      "epoch": 5.1996699669967,
      "grad_norm": 0.0078125,
      "learning_rate": 0.02772133188904858,
      "loss": 0.234,
      "num_input_tokens_seen": 9973632,
      "step": 47265
    },
    {
      "epoch": 5.2002200220022,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.027720568818946623,
      "loss": 0.2313,
      "num_input_tokens_seen": 9974752,
      "step": 47270
    },
    {
      "epoch": 5.200770077007701,
      "grad_norm": 0.012451171875,
      "learning_rate": 0.027719805631605286,
      "loss": 0.2291,
      "num_input_tokens_seen": 9975840,
      "step": 47275
    },
    {
      "epoch": 5.201320132013201,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.027719042327031602,
      "loss": 0.2265,
      "num_input_tokens_seen": 9976896,
      "step": 47280
    },
    {
      "epoch": 5.201870187018702,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.027718278905232614,
      "loss": 0.2286,
      "num_input_tokens_seen": 9977952,
      "step": 47285
    },
    {
      "epoch": 5.2024202420242025,
      "grad_norm": 0.01422119140625,
      "learning_rate": 0.027717515366215348,
      "loss": 0.2345,
      "num_input_tokens_seen": 9978944,
      "step": 47290
    },
    {
      "epoch": 5.202970297029703,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.02771675170998684,
      "loss": 0.2344,
      "num_input_tokens_seen": 9980000,
      "step": 47295
    },
    {
      "epoch": 5.203520352035204,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02771598793655414,
      "loss": 0.2291,
      "num_input_tokens_seen": 9981056,
      "step": 47300
    },
    {
      "epoch": 5.204070407040704,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.027715224045924285,
      "loss": 0.2317,
      "num_input_tokens_seen": 9982080,
      "step": 47305
    },
    {
      "epoch": 5.204620462046204,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0277144600381043,
      "loss": 0.2348,
      "num_input_tokens_seen": 9983104,
      "step": 47310
    },
    {
      "epoch": 5.205170517051705,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.02771369591310125,
      "loss": 0.2321,
      "num_input_tokens_seen": 9984128,
      "step": 47315
    },
    {
      "epoch": 5.2057205720572055,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.027712931670922158,
      "loss": 0.2247,
      "num_input_tokens_seen": 9985152,
      "step": 47320
    },
    {
      "epoch": 5.206270627062707,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.027712167311574075,
      "loss": 0.2337,
      "num_input_tokens_seen": 9986144,
      "step": 47325
    },
    {
      "epoch": 5.206820682068207,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027711402835064052,
      "loss": 0.2316,
      "num_input_tokens_seen": 9987200,
      "step": 47330
    },
    {
      "epoch": 5.207370737073707,
      "grad_norm": 0.01190185546875,
      "learning_rate": 0.027710638241399122,
      "loss": 0.2368,
      "num_input_tokens_seen": 9988224,
      "step": 47335
    },
    {
      "epoch": 5.207920792079208,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.027709873530586338,
      "loss": 0.2325,
      "num_input_tokens_seen": 9989344,
      "step": 47340
    },
    {
      "epoch": 5.208470847084708,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027709108702632754,
      "loss": 0.2299,
      "num_input_tokens_seen": 9990368,
      "step": 47345
    },
    {
      "epoch": 5.209020902090209,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.02770834375754541,
      "loss": 0.2299,
      "num_input_tokens_seen": 9991392,
      "step": 47350
    },
    {
      "epoch": 5.20957095709571,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.02770757869533136,
      "loss": 0.2314,
      "num_input_tokens_seen": 9992416,
      "step": 47355
    },
    {
      "epoch": 5.21012101210121,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.027706813515997655,
      "loss": 0.232,
      "num_input_tokens_seen": 9993504,
      "step": 47360
    },
    {
      "epoch": 5.210671067106711,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.02770604821955135,
      "loss": 0.2299,
      "num_input_tokens_seen": 9994624,
      "step": 47365
    },
    {
      "epoch": 5.211221122112211,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02770528280599949,
      "loss": 0.233,
      "num_input_tokens_seen": 9995648,
      "step": 47370
    },
    {
      "epoch": 5.211771177117711,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02770451727534914,
      "loss": 0.2314,
      "num_input_tokens_seen": 9996736,
      "step": 47375
    },
    {
      "epoch": 5.212321232123212,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.027703751627607347,
      "loss": 0.2315,
      "num_input_tokens_seen": 9997824,
      "step": 47380
    },
    {
      "epoch": 5.212871287128713,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.02770298586278117,
      "loss": 0.2299,
      "num_input_tokens_seen": 9998880,
      "step": 47385
    },
    {
      "epoch": 5.213421342134214,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.027702219980877674,
      "loss": 0.2335,
      "num_input_tokens_seen": 9999936,
      "step": 47390
    },
    {
      "epoch": 5.213971397139714,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.027701453981903906,
      "loss": 0.234,
      "num_input_tokens_seen": 10000960,
      "step": 47395
    },
    {
      "epoch": 5.214521452145214,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.027700687865866932,
      "loss": 0.2346,
      "num_input_tokens_seen": 10002016,
      "step": 47400
    },
    {
      "epoch": 5.215071507150715,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.027699921632773815,
      "loss": 0.2293,
      "num_input_tokens_seen": 10003040,
      "step": 47405
    },
    {
      "epoch": 5.215621562156215,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.027699155282631612,
      "loss": 0.2319,
      "num_input_tokens_seen": 10004096,
      "step": 47410
    },
    {
      "epoch": 5.2161716171617165,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02769838881544739,
      "loss": 0.2324,
      "num_input_tokens_seen": 10005216,
      "step": 47415
    },
    {
      "epoch": 5.216721672167217,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.02769762223122821,
      "loss": 0.2314,
      "num_input_tokens_seen": 10006272,
      "step": 47420
    },
    {
      "epoch": 5.217271727172717,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.027696855529981138,
      "loss": 0.2309,
      "num_input_tokens_seen": 10007328,
      "step": 47425
    },
    {
      "epoch": 5.217821782178218,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02769608871171324,
      "loss": 0.2308,
      "num_input_tokens_seen": 10008384,
      "step": 47430
    },
    {
      "epoch": 5.218371837183718,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02769532177643159,
      "loss": 0.2303,
      "num_input_tokens_seen": 10009472,
      "step": 47435
    },
    {
      "epoch": 5.218921892189219,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.027694554724143245,
      "loss": 0.2329,
      "num_input_tokens_seen": 10010528,
      "step": 47440
    },
    {
      "epoch": 5.2194719471947195,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.027693787554855288,
      "loss": 0.2319,
      "num_input_tokens_seen": 10011616,
      "step": 47445
    },
    {
      "epoch": 5.22002200220022,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.02769302026857478,
      "loss": 0.2324,
      "num_input_tokens_seen": 10012640,
      "step": 47450
    },
    {
      "epoch": 5.220572057205721,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02769225286530879,
      "loss": 0.2319,
      "num_input_tokens_seen": 10013728,
      "step": 47455
    },
    {
      "epoch": 5.221122112211221,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0276914853450644,
      "loss": 0.2303,
      "num_input_tokens_seen": 10014784,
      "step": 47460
    },
    {
      "epoch": 5.221672167216722,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.027690717707848676,
      "loss": 0.2298,
      "num_input_tokens_seen": 10015904,
      "step": 47465
    },
    {
      "epoch": 5.222222222222222,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.027689949953668698,
      "loss": 0.2303,
      "num_input_tokens_seen": 10016992,
      "step": 47470
    },
    {
      "epoch": 5.2227722772277225,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.02768918208253154,
      "loss": 0.2319,
      "num_input_tokens_seen": 10018176,
      "step": 47475
    },
    {
      "epoch": 5.223322332233224,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.02768841409444428,
      "loss": 0.2314,
      "num_input_tokens_seen": 10019296,
      "step": 47480
    },
    {
      "epoch": 5.223872387238724,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.027687645989414,
      "loss": 0.2304,
      "num_input_tokens_seen": 10020352,
      "step": 47485
    },
    {
      "epoch": 5.224422442244224,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.027686877767447767,
      "loss": 0.2314,
      "num_input_tokens_seen": 10021376,
      "step": 47490
    },
    {
      "epoch": 5.224972497249725,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.027686109428552676,
      "loss": 0.233,
      "num_input_tokens_seen": 10022464,
      "step": 47495
    },
    {
      "epoch": 5.225522552255225,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0276853409727358,
      "loss": 0.2315,
      "num_input_tokens_seen": 10023488,
      "step": 47500
    },
    {
      "epoch": 5.226072607260726,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027684572400004222,
      "loss": 0.2268,
      "num_input_tokens_seen": 10024544,
      "step": 47505
    },
    {
      "epoch": 5.226622662266227,
      "grad_norm": 0.005859375,
      "learning_rate": 0.027683803710365027,
      "loss": 0.2326,
      "num_input_tokens_seen": 10025568,
      "step": 47510
    },
    {
      "epoch": 5.227172717271727,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0276830349038253,
      "loss": 0.2254,
      "num_input_tokens_seen": 10026592,
      "step": 47515
    },
    {
      "epoch": 5.227722772277228,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027682265980392126,
      "loss": 0.2265,
      "num_input_tokens_seen": 10027648,
      "step": 47520
    },
    {
      "epoch": 5.228272827282728,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.027681496940072593,
      "loss": 0.2343,
      "num_input_tokens_seen": 10028704,
      "step": 47525
    },
    {
      "epoch": 5.228822882288229,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.02768072778287379,
      "loss": 0.2298,
      "num_input_tokens_seen": 10029792,
      "step": 47530
    },
    {
      "epoch": 5.229372937293729,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0276799585088028,
      "loss": 0.2338,
      "num_input_tokens_seen": 10030880,
      "step": 47535
    },
    {
      "epoch": 5.22992299229923,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.02767918911786672,
      "loss": 0.2308,
      "num_input_tokens_seen": 10031968,
      "step": 47540
    },
    {
      "epoch": 5.230473047304731,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.027678419610072634,
      "loss": 0.2364,
      "num_input_tokens_seen": 10033024,
      "step": 47545
    },
    {
      "epoch": 5.231023102310231,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.02767764998542764,
      "loss": 0.2317,
      "num_input_tokens_seen": 10034080,
      "step": 47550
    },
    {
      "epoch": 5.231573157315731,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.02767688024393883,
      "loss": 0.2306,
      "num_input_tokens_seen": 10035168,
      "step": 47555
    },
    {
      "epoch": 5.232123212321232,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.0276761103856133,
      "loss": 0.2395,
      "num_input_tokens_seen": 10036288,
      "step": 47560
    },
    {
      "epoch": 5.232673267326732,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.027675340410458136,
      "loss": 0.2331,
      "num_input_tokens_seen": 10037408,
      "step": 47565
    },
    {
      "epoch": 5.2332233223322335,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027674570318480442,
      "loss": 0.2341,
      "num_input_tokens_seen": 10038528,
      "step": 47570
    },
    {
      "epoch": 5.233773377337734,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027673800109687324,
      "loss": 0.233,
      "num_input_tokens_seen": 10039584,
      "step": 47575
    },
    {
      "epoch": 5.234323432343234,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027673029784085867,
      "loss": 0.2313,
      "num_input_tokens_seen": 10040640,
      "step": 47580
    },
    {
      "epoch": 5.234873487348735,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.027672259341683176,
      "loss": 0.2308,
      "num_input_tokens_seen": 10041696,
      "step": 47585
    },
    {
      "epoch": 5.235423542354235,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.02767148878248635,
      "loss": 0.2308,
      "num_input_tokens_seen": 10042720,
      "step": 47590
    },
    {
      "epoch": 5.235973597359736,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.027670718106502495,
      "loss": 0.2329,
      "num_input_tokens_seen": 10043808,
      "step": 47595
    },
    {
      "epoch": 5.2365236523652365,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02766994731373871,
      "loss": 0.2324,
      "num_input_tokens_seen": 10044896,
      "step": 47600
    },
    {
      "epoch": 5.237073707370737,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0276691764042021,
      "loss": 0.2319,
      "num_input_tokens_seen": 10045920,
      "step": 47605
    },
    {
      "epoch": 5.237623762376238,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.027668405377899767,
      "loss": 0.2335,
      "num_input_tokens_seen": 10046880,
      "step": 47610
    },
    {
      "epoch": 5.238173817381738,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.027667634234838825,
      "loss": 0.2309,
      "num_input_tokens_seen": 10048000,
      "step": 47615
    },
    {
      "epoch": 5.238723872387239,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.027666862975026376,
      "loss": 0.2298,
      "num_input_tokens_seen": 10049056,
      "step": 47620
    },
    {
      "epoch": 5.239273927392739,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.027666091598469524,
      "loss": 0.2309,
      "num_input_tokens_seen": 10050176,
      "step": 47625
    },
    {
      "epoch": 5.2398239823982395,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02766532010517539,
      "loss": 0.2299,
      "num_input_tokens_seen": 10051232,
      "step": 47630
    },
    {
      "epoch": 5.240374037403741,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.027664548495151077,
      "loss": 0.2314,
      "num_input_tokens_seen": 10052288,
      "step": 47635
    },
    {
      "epoch": 5.240924092409241,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.0276637767684037,
      "loss": 0.2294,
      "num_input_tokens_seen": 10053312,
      "step": 47640
    },
    {
      "epoch": 5.241474147414741,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.027663004924940365,
      "loss": 0.2304,
      "num_input_tokens_seen": 10054368,
      "step": 47645
    },
    {
      "epoch": 5.242024202420242,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.02766223296476819,
      "loss": 0.2351,
      "num_input_tokens_seen": 10055424,
      "step": 47650
    },
    {
      "epoch": 5.242574257425742,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.027661460887894288,
      "loss": 0.2309,
      "num_input_tokens_seen": 10056512,
      "step": 47655
    },
    {
      "epoch": 5.243124312431243,
      "grad_norm": 0.01287841796875,
      "learning_rate": 0.027660688694325777,
      "loss": 0.2309,
      "num_input_tokens_seen": 10057568,
      "step": 47660
    },
    {
      "epoch": 5.243674367436744,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.027659916384069776,
      "loss": 0.2299,
      "num_input_tokens_seen": 10058592,
      "step": 47665
    },
    {
      "epoch": 5.244224422442244,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.027659143957133404,
      "loss": 0.2335,
      "num_input_tokens_seen": 10059648,
      "step": 47670
    },
    {
      "epoch": 5.244774477447745,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.027658371413523775,
      "loss": 0.2293,
      "num_input_tokens_seen": 10060736,
      "step": 47675
    },
    {
      "epoch": 5.245324532453245,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.027657598753248008,
      "loss": 0.2293,
      "num_input_tokens_seen": 10061728,
      "step": 47680
    },
    {
      "epoch": 5.245874587458746,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.027656825976313224,
      "loss": 0.2325,
      "num_input_tokens_seen": 10062816,
      "step": 47685
    },
    {
      "epoch": 5.2464246424642464,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.02765605308272655,
      "loss": 0.2288,
      "num_input_tokens_seen": 10063840,
      "step": 47690
    },
    {
      "epoch": 5.246974697469747,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.027655280072495115,
      "loss": 0.2294,
      "num_input_tokens_seen": 10064864,
      "step": 47695
    },
    {
      "epoch": 5.247524752475248,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.027654506945626028,
      "loss": 0.2299,
      "num_input_tokens_seen": 10065952,
      "step": 47700
    },
    {
      "epoch": 5.248074807480748,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.027653733702126426,
      "loss": 0.2288,
      "num_input_tokens_seen": 10067072,
      "step": 47705
    },
    {
      "epoch": 5.248624862486249,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.02765296034200343,
      "loss": 0.2314,
      "num_input_tokens_seen": 10068128,
      "step": 47710
    },
    {
      "epoch": 5.249174917491749,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.027652186865264175,
      "loss": 0.2309,
      "num_input_tokens_seen": 10069184,
      "step": 47715
    },
    {
      "epoch": 5.2497249724972495,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.027651413271915778,
      "loss": 0.2335,
      "num_input_tokens_seen": 10070240,
      "step": 47720
    },
    {
      "epoch": 5.2502750275027505,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.027650639561965378,
      "loss": 0.232,
      "num_input_tokens_seen": 10071264,
      "step": 47725
    },
    {
      "epoch": 5.250825082508251,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.027649865735420102,
      "loss": 0.232,
      "num_input_tokens_seen": 10072288,
      "step": 47730
    },
    {
      "epoch": 5.251375137513751,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02764909179228709,
      "loss": 0.2288,
      "num_input_tokens_seen": 10073312,
      "step": 47735
    },
    {
      "epoch": 5.251925192519252,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.027648317732573462,
      "loss": 0.2335,
      "num_input_tokens_seen": 10074432,
      "step": 47740
    },
    {
      "epoch": 5.252475247524752,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.027647543556286364,
      "loss": 0.2314,
      "num_input_tokens_seen": 10075488,
      "step": 47745
    },
    {
      "epoch": 5.253025302530253,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027646769263432923,
      "loss": 0.2309,
      "num_input_tokens_seen": 10076608,
      "step": 47750
    },
    {
      "epoch": 5.2535753575357536,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.027645994854020275,
      "loss": 0.232,
      "num_input_tokens_seen": 10077696,
      "step": 47755
    },
    {
      "epoch": 5.254125412541254,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.027645220328055567,
      "loss": 0.2314,
      "num_input_tokens_seen": 10078720,
      "step": 47760
    },
    {
      "epoch": 5.254675467546755,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.02764444568554593,
      "loss": 0.2294,
      "num_input_tokens_seen": 10079744,
      "step": 47765
    },
    {
      "epoch": 5.255225522552255,
      "grad_norm": 0.012939453125,
      "learning_rate": 0.0276436709264985,
      "loss": 0.2341,
      "num_input_tokens_seen": 10080768,
      "step": 47770
    },
    {
      "epoch": 5.255775577557756,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.027642896050920424,
      "loss": 0.2273,
      "num_input_tokens_seen": 10081856,
      "step": 47775
    },
    {
      "epoch": 5.256325632563256,
      "grad_norm": 0.007171630859375,
      "learning_rate": 0.02764212105881884,
      "loss": 0.2357,
      "num_input_tokens_seen": 10082912,
      "step": 47780
    },
    {
      "epoch": 5.256875687568757,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.0276413459502009,
      "loss": 0.233,
      "num_input_tokens_seen": 10084000,
      "step": 47785
    },
    {
      "epoch": 5.257425742574258,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.027640570725073733,
      "loss": 0.232,
      "num_input_tokens_seen": 10085056,
      "step": 47790
    },
    {
      "epoch": 5.257975797579758,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.027639795383444493,
      "loss": 0.233,
      "num_input_tokens_seen": 10086080,
      "step": 47795
    },
    {
      "epoch": 5.258525852585258,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.027639019925320325,
      "loss": 0.2324,
      "num_input_tokens_seen": 10087136,
      "step": 47800
    },
    {
      "epoch": 5.259075907590759,
      "grad_norm": 0.005859375,
      "learning_rate": 0.027638244350708375,
      "loss": 0.2304,
      "num_input_tokens_seen": 10088128,
      "step": 47805
    },
    {
      "epoch": 5.259625962596259,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027637468659615787,
      "loss": 0.2309,
      "num_input_tokens_seen": 10089184,
      "step": 47810
    },
    {
      "epoch": 5.2601760176017605,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02763669285204972,
      "loss": 0.2319,
      "num_input_tokens_seen": 10090272,
      "step": 47815
    },
    {
      "epoch": 5.260726072607261,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.02763591692801732,
      "loss": 0.2309,
      "num_input_tokens_seen": 10091264,
      "step": 47820
    },
    {
      "epoch": 5.261276127612761,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.027635140887525732,
      "loss": 0.2309,
      "num_input_tokens_seen": 10092288,
      "step": 47825
    },
    {
      "epoch": 5.261826182618262,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.027634364730582117,
      "loss": 0.2335,
      "num_input_tokens_seen": 10093312,
      "step": 47830
    },
    {
      "epoch": 5.262376237623762,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 0.02763358845719362,
      "loss": 0.2345,
      "num_input_tokens_seen": 10094336,
      "step": 47835
    },
    {
      "epoch": 5.262926292629263,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.027632812067367403,
      "loss": 0.2304,
      "num_input_tokens_seen": 10095392,
      "step": 47840
    },
    {
      "epoch": 5.2634763476347635,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.027632035561110622,
      "loss": 0.2304,
      "num_input_tokens_seen": 10096416,
      "step": 47845
    },
    {
      "epoch": 5.264026402640264,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027631258938430426,
      "loss": 0.2304,
      "num_input_tokens_seen": 10097440,
      "step": 47850
    },
    {
      "epoch": 5.264576457645765,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027630482199333978,
      "loss": 0.232,
      "num_input_tokens_seen": 10098560,
      "step": 47855
    },
    {
      "epoch": 5.265126512651265,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.02762970534382844,
      "loss": 0.2351,
      "num_input_tokens_seen": 10099584,
      "step": 47860
    },
    {
      "epoch": 5.265676567656766,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.027628928371920967,
      "loss": 0.2309,
      "num_input_tokens_seen": 10100640,
      "step": 47865
    },
    {
      "epoch": 5.266226622662266,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.02762815128361872,
      "loss": 0.2299,
      "num_input_tokens_seen": 10101664,
      "step": 47870
    },
    {
      "epoch": 5.2667766776677665,
      "grad_norm": 0.0006256103515625,
      "learning_rate": 0.02762737407892886,
      "loss": 0.2351,
      "num_input_tokens_seen": 10102688,
      "step": 47875
    },
    {
      "epoch": 5.267326732673268,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.027626596757858557,
      "loss": 0.2283,
      "num_input_tokens_seen": 10103680,
      "step": 47880
    },
    {
      "epoch": 5.267876787678768,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.027625819320414968,
      "loss": 0.2308,
      "num_input_tokens_seen": 10104768,
      "step": 47885
    },
    {
      "epoch": 5.268426842684269,
      "grad_norm": 0.000762939453125,
      "learning_rate": 0.027625041766605258,
      "loss": 0.2308,
      "num_input_tokens_seen": 10105760,
      "step": 47890
    },
    {
      "epoch": 5.268976897689769,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0276242640964366,
      "loss": 0.2345,
      "num_input_tokens_seen": 10106816,
      "step": 47895
    },
    {
      "epoch": 5.269526952695269,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.027623486309916158,
      "loss": 0.2293,
      "num_input_tokens_seen": 10107808,
      "step": 47900
    },
    {
      "epoch": 5.27007700770077,
      "grad_norm": 0.00060272216796875,
      "learning_rate": 0.027622708407051097,
      "loss": 0.2319,
      "num_input_tokens_seen": 10108832,
      "step": 47905
    },
    {
      "epoch": 5.270627062706271,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.027621930387848593,
      "loss": 0.2324,
      "num_input_tokens_seen": 10109920,
      "step": 47910
    },
    {
      "epoch": 5.271177117711771,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02762115225231581,
      "loss": 0.2308,
      "num_input_tokens_seen": 10110976,
      "step": 47915
    },
    {
      "epoch": 5.271727172717272,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.027620374000459923,
      "loss": 0.2329,
      "num_input_tokens_seen": 10112032,
      "step": 47920
    },
    {
      "epoch": 5.272277227722772,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027619595632288108,
      "loss": 0.2318,
      "num_input_tokens_seen": 10113024,
      "step": 47925
    },
    {
      "epoch": 5.272827282728273,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.027618817147807534,
      "loss": 0.2313,
      "num_input_tokens_seen": 10114080,
      "step": 47930
    },
    {
      "epoch": 5.273377337733773,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.027618038547025374,
      "loss": 0.2303,
      "num_input_tokens_seen": 10115136,
      "step": 47935
    },
    {
      "epoch": 5.273927392739274,
      "grad_norm": 0.00072479248046875,
      "learning_rate": 0.027617259829948813,
      "loss": 0.2298,
      "num_input_tokens_seen": 10116160,
      "step": 47940
    },
    {
      "epoch": 5.274477447744775,
      "grad_norm": 0.000713348388671875,
      "learning_rate": 0.027616480996585018,
      "loss": 0.234,
      "num_input_tokens_seen": 10117248,
      "step": 47945
    },
    {
      "epoch": 5.275027502750275,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.027615702046941172,
      "loss": 0.2282,
      "num_input_tokens_seen": 10118304,
      "step": 47950
    },
    {
      "epoch": 5.275577557755776,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.02761492298102446,
      "loss": 0.2324,
      "num_input_tokens_seen": 10119424,
      "step": 47955
    },
    {
      "epoch": 5.276127612761276,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.02761414379884205,
      "loss": 0.2324,
      "num_input_tokens_seen": 10120512,
      "step": 47960
    },
    {
      "epoch": 5.276677667766776,
      "grad_norm": 0.000652313232421875,
      "learning_rate": 0.02761336450040113,
      "loss": 0.2308,
      "num_input_tokens_seen": 10121536,
      "step": 47965
    },
    {
      "epoch": 5.2772277227722775,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.027612585085708883,
      "loss": 0.2334,
      "num_input_tokens_seen": 10122656,
      "step": 47970
    },
    {
      "epoch": 5.277777777777778,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.027611805554772488,
      "loss": 0.2329,
      "num_input_tokens_seen": 10123680,
      "step": 47975
    },
    {
      "epoch": 5.278327832783278,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02761102590759914,
      "loss": 0.2329,
      "num_input_tokens_seen": 10124736,
      "step": 47980
    },
    {
      "epoch": 5.278877887788779,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02761024614419601,
      "loss": 0.2324,
      "num_input_tokens_seen": 10125824,
      "step": 47985
    },
    {
      "epoch": 5.279427942794279,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.027609466264570295,
      "loss": 0.2303,
      "num_input_tokens_seen": 10126880,
      "step": 47990
    },
    {
      "epoch": 5.27997799779978,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.027608686268729185,
      "loss": 0.2324,
      "num_input_tokens_seen": 10127904,
      "step": 47995
    },
    {
      "epoch": 5.2805280528052805,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.027607906156679858,
      "loss": 0.233,
      "num_input_tokens_seen": 10129024,
      "step": 48000
    },
    {
      "epoch": 5.281078107810781,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027607125928429514,
      "loss": 0.2329,
      "num_input_tokens_seen": 10130112,
      "step": 48005
    },
    {
      "epoch": 5.281628162816282,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.027606345583985337,
      "loss": 0.2324,
      "num_input_tokens_seen": 10131200,
      "step": 48010
    },
    {
      "epoch": 5.282178217821782,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.027605565123354523,
      "loss": 0.2313,
      "num_input_tokens_seen": 10132288,
      "step": 48015
    },
    {
      "epoch": 5.282728272827283,
      "grad_norm": 0.00061798095703125,
      "learning_rate": 0.02760478454654426,
      "loss": 0.2313,
      "num_input_tokens_seen": 10133280,
      "step": 48020
    },
    {
      "epoch": 5.283278327832783,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.027604003853561754,
      "loss": 0.2293,
      "num_input_tokens_seen": 10134336,
      "step": 48025
    },
    {
      "epoch": 5.2838283828382835,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02760322304441418,
      "loss": 0.2313,
      "num_input_tokens_seen": 10135360,
      "step": 48030
    },
    {
      "epoch": 5.284378437843785,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.02760244211910876,
      "loss": 0.2308,
      "num_input_tokens_seen": 10136480,
      "step": 48035
    },
    {
      "epoch": 5.284928492849285,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.027601661077652672,
      "loss": 0.2319,
      "num_input_tokens_seen": 10137568,
      "step": 48040
    },
    {
      "epoch": 5.285478547854786,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.027600879920053126,
      "loss": 0.2319,
      "num_input_tokens_seen": 10138592,
      "step": 48045
    },
    {
      "epoch": 5.286028602860286,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027600098646317305,
      "loss": 0.2298,
      "num_input_tokens_seen": 10139680,
      "step": 48050
    },
    {
      "epoch": 5.286578657865786,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.027599317256452434,
      "loss": 0.2319,
      "num_input_tokens_seen": 10140736,
      "step": 48055
    },
    {
      "epoch": 5.287128712871287,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02759853575046569,
      "loss": 0.2314,
      "num_input_tokens_seen": 10141760,
      "step": 48060
    },
    {
      "epoch": 5.287678767876788,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027597754128364294,
      "loss": 0.2319,
      "num_input_tokens_seen": 10142912,
      "step": 48065
    },
    {
      "epoch": 5.288228822882289,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027596972390155444,
      "loss": 0.234,
      "num_input_tokens_seen": 10144000,
      "step": 48070
    },
    {
      "epoch": 5.288778877887789,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027596190535846343,
      "loss": 0.2298,
      "num_input_tokens_seen": 10145088,
      "step": 48075
    },
    {
      "epoch": 5.289328932893289,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.027595408565444193,
      "loss": 0.233,
      "num_input_tokens_seen": 10146176,
      "step": 48080
    },
    {
      "epoch": 5.28987898789879,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027594626478956212,
      "loss": 0.2313,
      "num_input_tokens_seen": 10147264,
      "step": 48085
    },
    {
      "epoch": 5.29042904290429,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0275938442763896,
      "loss": 0.2313,
      "num_input_tokens_seen": 10148288,
      "step": 48090
    },
    {
      "epoch": 5.290979097909791,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02759306195775157,
      "loss": 0.2313,
      "num_input_tokens_seen": 10149376,
      "step": 48095
    },
    {
      "epoch": 5.291529152915292,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02759227952304933,
      "loss": 0.2319,
      "num_input_tokens_seen": 10150368,
      "step": 48100
    },
    {
      "epoch": 5.292079207920792,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.027591496972290092,
      "loss": 0.2319,
      "num_input_tokens_seen": 10151424,
      "step": 48105
    },
    {
      "epoch": 5.292629262926293,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.02759071430548107,
      "loss": 0.2303,
      "num_input_tokens_seen": 10152384,
      "step": 48110
    },
    {
      "epoch": 5.293179317931793,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.02758993152262947,
      "loss": 0.2324,
      "num_input_tokens_seen": 10153408,
      "step": 48115
    },
    {
      "epoch": 5.293729372937293,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.027589148623742517,
      "loss": 0.2319,
      "num_input_tokens_seen": 10154432,
      "step": 48120
    },
    {
      "epoch": 5.2942794279427945,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027588365608827423,
      "loss": 0.2303,
      "num_input_tokens_seen": 10155456,
      "step": 48125
    },
    {
      "epoch": 5.294829482948295,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0275875824778914,
      "loss": 0.2314,
      "num_input_tokens_seen": 10156512,
      "step": 48130
    },
    {
      "epoch": 5.295379537953796,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.027586799230941674,
      "loss": 0.2304,
      "num_input_tokens_seen": 10157664,
      "step": 48135
    },
    {
      "epoch": 5.295929592959296,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.027586015867985456,
      "loss": 0.2308,
      "num_input_tokens_seen": 10158688,
      "step": 48140
    },
    {
      "epoch": 5.296479647964796,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.02758523238902997,
      "loss": 0.2303,
      "num_input_tokens_seen": 10159712,
      "step": 48145
    },
    {
      "epoch": 5.297029702970297,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.027584448794082436,
      "loss": 0.2314,
      "num_input_tokens_seen": 10160736,
      "step": 48150
    },
    {
      "epoch": 5.2975797579757975,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02758366508315007,
      "loss": 0.2308,
      "num_input_tokens_seen": 10161888,
      "step": 48155
    },
    {
      "epoch": 5.298129812981298,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02758288125624011,
      "loss": 0.2313,
      "num_input_tokens_seen": 10162976,
      "step": 48160
    },
    {
      "epoch": 5.298679867986799,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02758209731335977,
      "loss": 0.2314,
      "num_input_tokens_seen": 10164032,
      "step": 48165
    },
    {
      "epoch": 5.299229922992299,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.027581313254516272,
      "loss": 0.2309,
      "num_input_tokens_seen": 10165056,
      "step": 48170
    },
    {
      "epoch": 5.2997799779978,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.02758052907971685,
      "loss": 0.2319,
      "num_input_tokens_seen": 10166144,
      "step": 48175
    },
    {
      "epoch": 5.3003300330033,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.02757974478896873,
      "loss": 0.2345,
      "num_input_tokens_seen": 10167232,
      "step": 48180
    },
    {
      "epoch": 5.3008800880088005,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.027578960382279135,
      "loss": 0.2313,
      "num_input_tokens_seen": 10168288,
      "step": 48185
    },
    {
      "epoch": 5.301430143014302,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.0275781758596553,
      "loss": 0.2324,
      "num_input_tokens_seen": 10169344,
      "step": 48190
    },
    {
      "epoch": 5.301980198019802,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02757739122110445,
      "loss": 0.2308,
      "num_input_tokens_seen": 10170400,
      "step": 48195
    },
    {
      "epoch": 5.302530253025303,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02757660646663383,
      "loss": 0.2304,
      "num_input_tokens_seen": 10171424,
      "step": 48200
    },
    {
      "epoch": 5.303080308030803,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.027575821596250653,
      "loss": 0.2314,
      "num_input_tokens_seen": 10172448,
      "step": 48205
    },
    {
      "epoch": 5.303630363036303,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02757503660996217,
      "loss": 0.232,
      "num_input_tokens_seen": 10173504,
      "step": 48210
    },
    {
      "epoch": 5.304180418041804,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.027574251507775604,
      "loss": 0.2293,
      "num_input_tokens_seen": 10174592,
      "step": 48215
    },
    {
      "epoch": 5.304730473047305,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.027573466289698194,
      "loss": 0.231,
      "num_input_tokens_seen": 10175648,
      "step": 48220
    },
    {
      "epoch": 5.305280528052805,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.027572680955737186,
      "loss": 0.2346,
      "num_input_tokens_seen": 10176736,
      "step": 48225
    },
    {
      "epoch": 5.305830583058306,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.0275718955058998,
      "loss": 0.2294,
      "num_input_tokens_seen": 10177728,
      "step": 48230
    },
    {
      "epoch": 5.306380638063806,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.027571109940193295,
      "loss": 0.232,
      "num_input_tokens_seen": 10178784,
      "step": 48235
    },
    {
      "epoch": 5.306930693069307,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.027570324258624896,
      "loss": 0.2325,
      "num_input_tokens_seen": 10179840,
      "step": 48240
    },
    {
      "epoch": 5.307480748074807,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.027569538461201852,
      "loss": 0.2315,
      "num_input_tokens_seen": 10180928,
      "step": 48245
    },
    {
      "epoch": 5.3080308030803085,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.027568752547931404,
      "loss": 0.2319,
      "num_input_tokens_seen": 10182016,
      "step": 48250
    },
    {
      "epoch": 5.308580858085809,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.027567966518820798,
      "loss": 0.2283,
      "num_input_tokens_seen": 10183072,
      "step": 48255
    },
    {
      "epoch": 5.309130913091309,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027567180373877273,
      "loss": 0.2293,
      "num_input_tokens_seen": 10184096,
      "step": 48260
    },
    {
      "epoch": 5.30968096809681,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.027566394113108076,
      "loss": 0.233,
      "num_input_tokens_seen": 10185152,
      "step": 48265
    },
    {
      "epoch": 5.31023102310231,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.02756560773652045,
      "loss": 0.2335,
      "num_input_tokens_seen": 10186208,
      "step": 48270
    },
    {
      "epoch": 5.31078107810781,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027564821244121653,
      "loss": 0.2324,
      "num_input_tokens_seen": 10187200,
      "step": 48275
    },
    {
      "epoch": 5.3113311331133115,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.027564034635918926,
      "loss": 0.2298,
      "num_input_tokens_seen": 10188256,
      "step": 48280
    },
    {
      "epoch": 5.311881188118812,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.02756324791191952,
      "loss": 0.2277,
      "num_input_tokens_seen": 10189312,
      "step": 48285
    },
    {
      "epoch": 5.312431243124313,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02756246107213069,
      "loss": 0.2325,
      "num_input_tokens_seen": 10190400,
      "step": 48290
    },
    {
      "epoch": 5.312981298129813,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.02756167411655968,
      "loss": 0.2309,
      "num_input_tokens_seen": 10191424,
      "step": 48295
    },
    {
      "epoch": 5.313531353135313,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02756088704521375,
      "loss": 0.2304,
      "num_input_tokens_seen": 10192512,
      "step": 48300
    },
    {
      "epoch": 5.314081408140814,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.02756009985810015,
      "loss": 0.2314,
      "num_input_tokens_seen": 10193600,
      "step": 48305
    },
    {
      "epoch": 5.3146314631463145,
      "grad_norm": 0.0006561279296875,
      "learning_rate": 0.027559312555226135,
      "loss": 0.2299,
      "num_input_tokens_seen": 10194624,
      "step": 48310
    },
    {
      "epoch": 5.315181518151816,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02755852513659897,
      "loss": 0.2283,
      "num_input_tokens_seen": 10195616,
      "step": 48315
    },
    {
      "epoch": 5.315731573157316,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.027557737602225896,
      "loss": 0.2325,
      "num_input_tokens_seen": 10196672,
      "step": 48320
    },
    {
      "epoch": 5.316281628162816,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.027556949952114187,
      "loss": 0.232,
      "num_input_tokens_seen": 10197792,
      "step": 48325
    },
    {
      "epoch": 5.316831683168317,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.027556162186271087,
      "loss": 0.231,
      "num_input_tokens_seen": 10198784,
      "step": 48330
    },
    {
      "epoch": 5.317381738173817,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.027555374304703874,
      "loss": 0.2331,
      "num_input_tokens_seen": 10199776,
      "step": 48335
    },
    {
      "epoch": 5.3179317931793175,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027554586307419797,
      "loss": 0.231,
      "num_input_tokens_seen": 10200864,
      "step": 48340
    },
    {
      "epoch": 5.318481848184819,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.027553798194426122,
      "loss": 0.2351,
      "num_input_tokens_seen": 10201824,
      "step": 48345
    },
    {
      "epoch": 5.319031903190319,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.027553009965730114,
      "loss": 0.2341,
      "num_input_tokens_seen": 10202912,
      "step": 48350
    },
    {
      "epoch": 5.31958195819582,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027552221621339038,
      "loss": 0.2299,
      "num_input_tokens_seen": 10203968,
      "step": 48355
    },
    {
      "epoch": 5.32013201320132,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.027551433161260155,
      "loss": 0.2289,
      "num_input_tokens_seen": 10204992,
      "step": 48360
    },
    {
      "epoch": 5.32068206820682,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.027550644585500737,
      "loss": 0.232,
      "num_input_tokens_seen": 10206016,
      "step": 48365
    },
    {
      "epoch": 5.321232123212321,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.027549855894068047,
      "loss": 0.2325,
      "num_input_tokens_seen": 10207104,
      "step": 48370
    },
    {
      "epoch": 5.321782178217822,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.027549067086969362,
      "loss": 0.2288,
      "num_input_tokens_seen": 10208192,
      "step": 48375
    },
    {
      "epoch": 5.322332233223323,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.02754827816421195,
      "loss": 0.2325,
      "num_input_tokens_seen": 10209216,
      "step": 48380
    },
    {
      "epoch": 5.322882288228823,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02754748912580307,
      "loss": 0.2345,
      "num_input_tokens_seen": 10210240,
      "step": 48385
    },
    {
      "epoch": 5.323432343234323,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.02754669997175001,
      "loss": 0.2335,
      "num_input_tokens_seen": 10211232,
      "step": 48390
    },
    {
      "epoch": 5.323982398239824,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02754591070206004,
      "loss": 0.2293,
      "num_input_tokens_seen": 10212288,
      "step": 48395
    },
    {
      "epoch": 5.324532453245324,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.027545121316740424,
      "loss": 0.2314,
      "num_input_tokens_seen": 10213408,
      "step": 48400
    },
    {
      "epoch": 5.325082508250825,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.027544331815798448,
      "loss": 0.2319,
      "num_input_tokens_seen": 10214464,
      "step": 48405
    },
    {
      "epoch": 5.325632563256326,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02754354219924138,
      "loss": 0.234,
      "num_input_tokens_seen": 10215488,
      "step": 48410
    },
    {
      "epoch": 5.326182618261826,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.02754275246707651,
      "loss": 0.2319,
      "num_input_tokens_seen": 10216544,
      "step": 48415
    },
    {
      "epoch": 5.326732673267327,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.027541962619311106,
      "loss": 0.2319,
      "num_input_tokens_seen": 10217632,
      "step": 48420
    },
    {
      "epoch": 5.327282728272827,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02754117265595245,
      "loss": 0.2309,
      "num_input_tokens_seen": 10218688,
      "step": 48425
    },
    {
      "epoch": 5.327832783278327,
      "grad_norm": 0.000637054443359375,
      "learning_rate": 0.027540382577007822,
      "loss": 0.2293,
      "num_input_tokens_seen": 10219712,
      "step": 48430
    },
    {
      "epoch": 5.3283828382838285,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.027539592382484507,
      "loss": 0.2313,
      "num_input_tokens_seen": 10220736,
      "step": 48435
    },
    {
      "epoch": 5.328932893289329,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02753880207238979,
      "loss": 0.2314,
      "num_input_tokens_seen": 10221792,
      "step": 48440
    },
    {
      "epoch": 5.32948294829483,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02753801164673095,
      "loss": 0.2308,
      "num_input_tokens_seen": 10222816,
      "step": 48445
    },
    {
      "epoch": 5.33003300330033,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027537221105515267,
      "loss": 0.2293,
      "num_input_tokens_seen": 10223904,
      "step": 48450
    },
    {
      "epoch": 5.33058305830583,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02753643044875004,
      "loss": 0.2309,
      "num_input_tokens_seen": 10224992,
      "step": 48455
    },
    {
      "epoch": 5.331133113311331,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.027535639676442545,
      "loss": 0.2314,
      "num_input_tokens_seen": 10226080,
      "step": 48460
    },
    {
      "epoch": 5.3316831683168315,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.027534848788600075,
      "loss": 0.2272,
      "num_input_tokens_seen": 10227072,
      "step": 48465
    },
    {
      "epoch": 5.332233223322333,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02753405778522992,
      "loss": 0.234,
      "num_input_tokens_seen": 10228128,
      "step": 48470
    },
    {
      "epoch": 5.332783278327833,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027533266666339366,
      "loss": 0.2267,
      "num_input_tokens_seen": 10229248,
      "step": 48475
    },
    {
      "epoch": 5.333333333333333,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027532475431935707,
      "loss": 0.2324,
      "num_input_tokens_seen": 10230240,
      "step": 48480
    },
    {
      "epoch": 5.333883388338834,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.027531684082026237,
      "loss": 0.2314,
      "num_input_tokens_seen": 10231328,
      "step": 48485
    },
    {
      "epoch": 5.334433443344334,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02753089261661825,
      "loss": 0.2314,
      "num_input_tokens_seen": 10232384,
      "step": 48490
    },
    {
      "epoch": 5.334983498349835,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.027530101035719037,
      "loss": 0.2304,
      "num_input_tokens_seen": 10233376,
      "step": 48495
    },
    {
      "epoch": 5.335533553355336,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.027529309339335895,
      "loss": 0.2351,
      "num_input_tokens_seen": 10234400,
      "step": 48500
    },
    {
      "epoch": 5.336083608360836,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.027528517527476117,
      "loss": 0.2283,
      "num_input_tokens_seen": 10235424,
      "step": 48505
    },
    {
      "epoch": 5.336633663366337,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02752772560014701,
      "loss": 0.234,
      "num_input_tokens_seen": 10236448,
      "step": 48510
    },
    {
      "epoch": 5.337183718371837,
      "grad_norm": 0.00080108642578125,
      "learning_rate": 0.02752693355735587,
      "loss": 0.2346,
      "num_input_tokens_seen": 10237472,
      "step": 48515
    },
    {
      "epoch": 5.337733773377337,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.027526141399109992,
      "loss": 0.2283,
      "num_input_tokens_seen": 10238496,
      "step": 48520
    },
    {
      "epoch": 5.338283828382838,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.027525349125416675,
      "loss": 0.2309,
      "num_input_tokens_seen": 10239552,
      "step": 48525
    },
    {
      "epoch": 5.338833883388339,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02752455673628323,
      "loss": 0.232,
      "num_input_tokens_seen": 10240576,
      "step": 48530
    },
    {
      "epoch": 5.33938393839384,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.027523764231716954,
      "loss": 0.2325,
      "num_input_tokens_seen": 10241632,
      "step": 48535
    },
    {
      "epoch": 5.33993399339934,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02752297161172515,
      "loss": 0.2314,
      "num_input_tokens_seen": 10242592,
      "step": 48540
    },
    {
      "epoch": 5.34048404840484,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.027522178876315133,
      "loss": 0.2304,
      "num_input_tokens_seen": 10243584,
      "step": 48545
    },
    {
      "epoch": 5.341034103410341,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.027521386025494195,
      "loss": 0.2304,
      "num_input_tokens_seen": 10244608,
      "step": 48550
    },
    {
      "epoch": 5.341584158415841,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02752059305926965,
      "loss": 0.2325,
      "num_input_tokens_seen": 10245696,
      "step": 48555
    },
    {
      "epoch": 5.3421342134213425,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02751979997764881,
      "loss": 0.2319,
      "num_input_tokens_seen": 10246688,
      "step": 48560
    },
    {
      "epoch": 5.342684268426843,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.027519006780638983,
      "loss": 0.2319,
      "num_input_tokens_seen": 10247744,
      "step": 48565
    },
    {
      "epoch": 5.343234323432343,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.027518213468247475,
      "loss": 0.2335,
      "num_input_tokens_seen": 10248800,
      "step": 48570
    },
    {
      "epoch": 5.343784378437844,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.027517420040481598,
      "loss": 0.2329,
      "num_input_tokens_seen": 10249952,
      "step": 48575
    },
    {
      "epoch": 5.344334433443344,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02751662649734867,
      "loss": 0.2329,
      "num_input_tokens_seen": 10251040,
      "step": 48580
    },
    {
      "epoch": 5.3448844884488445,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.027515832838856003,
      "loss": 0.2314,
      "num_input_tokens_seen": 10252128,
      "step": 48585
    },
    {
      "epoch": 5.3454345434543455,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027515039065010908,
      "loss": 0.2308,
      "num_input_tokens_seen": 10253184,
      "step": 48590
    },
    {
      "epoch": 5.345984598459846,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.027514245175820705,
      "loss": 0.2298,
      "num_input_tokens_seen": 10254240,
      "step": 48595
    },
    {
      "epoch": 5.346534653465347,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027513451171292707,
      "loss": 0.2293,
      "num_input_tokens_seen": 10255360,
      "step": 48600
    },
    {
      "epoch": 5.347084708470847,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.027512657051434235,
      "loss": 0.2319,
      "num_input_tokens_seen": 10256448,
      "step": 48605
    },
    {
      "epoch": 5.347634763476347,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027511862816252605,
      "loss": 0.2299,
      "num_input_tokens_seen": 10257504,
      "step": 48610
    },
    {
      "epoch": 5.348184818481848,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.027511068465755146,
      "loss": 0.2314,
      "num_input_tokens_seen": 10258528,
      "step": 48615
    },
    {
      "epoch": 5.3487348734873486,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.027510273999949163,
      "loss": 0.2324,
      "num_input_tokens_seen": 10259552,
      "step": 48620
    },
    {
      "epoch": 5.34928492849285,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.027509479418841992,
      "loss": 0.2288,
      "num_input_tokens_seen": 10260544,
      "step": 48625
    },
    {
      "epoch": 5.34983498349835,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.027508684722440953,
      "loss": 0.234,
      "num_input_tokens_seen": 10261568,
      "step": 48630
    },
    {
      "epoch": 5.35038503850385,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.02750788991075337,
      "loss": 0.2309,
      "num_input_tokens_seen": 10262560,
      "step": 48635
    },
    {
      "epoch": 5.350935093509351,
      "grad_norm": 0.00054931640625,
      "learning_rate": 0.027507094983786562,
      "loss": 0.2303,
      "num_input_tokens_seen": 10263584,
      "step": 48640
    },
    {
      "epoch": 5.351485148514851,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.027506299941547865,
      "loss": 0.2335,
      "num_input_tokens_seen": 10264704,
      "step": 48645
    },
    {
      "epoch": 5.3520352035203524,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027505504784044604,
      "loss": 0.2314,
      "num_input_tokens_seen": 10265728,
      "step": 48650
    },
    {
      "epoch": 5.352585258525853,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.027504709511284104,
      "loss": 0.2324,
      "num_input_tokens_seen": 10266816,
      "step": 48655
    },
    {
      "epoch": 5.353135313531353,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.027503914123273698,
      "loss": 0.2298,
      "num_input_tokens_seen": 10267808,
      "step": 48660
    },
    {
      "epoch": 5.353685368536854,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.027503118620020712,
      "loss": 0.2314,
      "num_input_tokens_seen": 10268832,
      "step": 48665
    },
    {
      "epoch": 5.354235423542354,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.027502323001532488,
      "loss": 0.2319,
      "num_input_tokens_seen": 10269856,
      "step": 48670
    },
    {
      "epoch": 5.354785478547855,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02750152726781635,
      "loss": 0.233,
      "num_input_tokens_seen": 10270976,
      "step": 48675
    },
    {
      "epoch": 5.3553355335533555,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02750073141887963,
      "loss": 0.2309,
      "num_input_tokens_seen": 10272000,
      "step": 48680
    },
    {
      "epoch": 5.355885588558856,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02749993545472967,
      "loss": 0.2314,
      "num_input_tokens_seen": 10272992,
      "step": 48685
    },
    {
      "epoch": 5.356435643564357,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.0274991393753738,
      "loss": 0.2319,
      "num_input_tokens_seen": 10273984,
      "step": 48690
    },
    {
      "epoch": 5.356985698569857,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027498343180819364,
      "loss": 0.233,
      "num_input_tokens_seen": 10275040,
      "step": 48695
    },
    {
      "epoch": 5.357535753575357,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027497546871073694,
      "loss": 0.2283,
      "num_input_tokens_seen": 10276096,
      "step": 48700
    },
    {
      "epoch": 5.358085808580858,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.02749675044614413,
      "loss": 0.2345,
      "num_input_tokens_seen": 10277216,
      "step": 48705
    },
    {
      "epoch": 5.3586358635863585,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.027495953906038016,
      "loss": 0.2299,
      "num_input_tokens_seen": 10278304,
      "step": 48710
    },
    {
      "epoch": 5.3591859185918596,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.02749515725076269,
      "loss": 0.2304,
      "num_input_tokens_seen": 10279392,
      "step": 48715
    },
    {
      "epoch": 5.35973597359736,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.027494360480325497,
      "loss": 0.2361,
      "num_input_tokens_seen": 10280512,
      "step": 48720
    },
    {
      "epoch": 5.36028602860286,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.027493563594733778,
      "loss": 0.2309,
      "num_input_tokens_seen": 10281536,
      "step": 48725
    },
    {
      "epoch": 5.360836083608361,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027492766593994876,
      "loss": 0.2308,
      "num_input_tokens_seen": 10282656,
      "step": 48730
    },
    {
      "epoch": 5.361386138613861,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027491969478116142,
      "loss": 0.2324,
      "num_input_tokens_seen": 10283744,
      "step": 48735
    },
    {
      "epoch": 5.361936193619362,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.027491172247104918,
      "loss": 0.2345,
      "num_input_tokens_seen": 10284832,
      "step": 48740
    },
    {
      "epoch": 5.362486248624863,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02749037490096855,
      "loss": 0.2345,
      "num_input_tokens_seen": 10285920,
      "step": 48745
    },
    {
      "epoch": 5.363036303630363,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.027489577439714398,
      "loss": 0.2303,
      "num_input_tokens_seen": 10286944,
      "step": 48750
    },
    {
      "epoch": 5.363586358635864,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.027488779863349796,
      "loss": 0.2319,
      "num_input_tokens_seen": 10287968,
      "step": 48755
    },
    {
      "epoch": 5.364136413641364,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.027487982171882107,
      "loss": 0.2319,
      "num_input_tokens_seen": 10288992,
      "step": 48760
    },
    {
      "epoch": 5.364686468646864,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02748718436531868,
      "loss": 0.2324,
      "num_input_tokens_seen": 10290048,
      "step": 48765
    },
    {
      "epoch": 5.365236523652365,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.02748638644366686,
      "loss": 0.2319,
      "num_input_tokens_seen": 10291136,
      "step": 48770
    },
    {
      "epoch": 5.365786578657866,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.027485588406934013,
      "loss": 0.2324,
      "num_input_tokens_seen": 10292192,
      "step": 48775
    },
    {
      "epoch": 5.366336633663367,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02748479025512749,
      "loss": 0.2298,
      "num_input_tokens_seen": 10293248,
      "step": 48780
    },
    {
      "epoch": 5.366886688668867,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027483991988254644,
      "loss": 0.2314,
      "num_input_tokens_seen": 10294272,
      "step": 48785
    },
    {
      "epoch": 5.367436743674367,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.02748319360632283,
      "loss": 0.2329,
      "num_input_tokens_seen": 10295296,
      "step": 48790
    },
    {
      "epoch": 5.367986798679868,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027482395109339416,
      "loss": 0.2298,
      "num_input_tokens_seen": 10296416,
      "step": 48795
    },
    {
      "epoch": 5.368536853685368,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.027481596497311755,
      "loss": 0.2319,
      "num_input_tokens_seen": 10297504,
      "step": 48800
    },
    {
      "epoch": 5.3690869086908695,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02748079777024721,
      "loss": 0.2319,
      "num_input_tokens_seen": 10298560,
      "step": 48805
    },
    {
      "epoch": 5.36963696369637,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02747999892815314,
      "loss": 0.2319,
      "num_input_tokens_seen": 10299616,
      "step": 48810
    },
    {
      "epoch": 5.37018701870187,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.027479199971036906,
      "loss": 0.2314,
      "num_input_tokens_seen": 10300672,
      "step": 48815
    },
    {
      "epoch": 5.370737073707371,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.027478400898905877,
      "loss": 0.2319,
      "num_input_tokens_seen": 10301824,
      "step": 48820
    },
    {
      "epoch": 5.371287128712871,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02747760171176741,
      "loss": 0.2319,
      "num_input_tokens_seen": 10302848,
      "step": 48825
    },
    {
      "epoch": 5.371837183718371,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.027476802409628878,
      "loss": 0.2325,
      "num_input_tokens_seen": 10303840,
      "step": 48830
    },
    {
      "epoch": 5.3723872387238725,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.027476002992497646,
      "loss": 0.2299,
      "num_input_tokens_seen": 10304832,
      "step": 48835
    },
    {
      "epoch": 5.372937293729373,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.02747520346038108,
      "loss": 0.2314,
      "num_input_tokens_seen": 10305824,
      "step": 48840
    },
    {
      "epoch": 5.373487348734874,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02747440381328655,
      "loss": 0.2304,
      "num_input_tokens_seen": 10306816,
      "step": 48845
    },
    {
      "epoch": 5.374037403740374,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027473604051221425,
      "loss": 0.233,
      "num_input_tokens_seen": 10307872,
      "step": 48850
    },
    {
      "epoch": 5.374587458745874,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02747280417419308,
      "loss": 0.2309,
      "num_input_tokens_seen": 10308832,
      "step": 48855
    },
    {
      "epoch": 5.375137513751375,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.02747200418220888,
      "loss": 0.2335,
      "num_input_tokens_seen": 10309888,
      "step": 48860
    },
    {
      "epoch": 5.3756875687568755,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.0274712040752762,
      "loss": 0.2314,
      "num_input_tokens_seen": 10311008,
      "step": 48865
    },
    {
      "epoch": 5.376237623762377,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.027470403853402426,
      "loss": 0.2319,
      "num_input_tokens_seen": 10312096,
      "step": 48870
    },
    {
      "epoch": 5.376787678767877,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.02746960351659492,
      "loss": 0.2309,
      "num_input_tokens_seen": 10313120,
      "step": 48875
    },
    {
      "epoch": 5.377337733773377,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.027468803064861054,
      "loss": 0.2324,
      "num_input_tokens_seen": 10314144,
      "step": 48880
    },
    {
      "epoch": 5.377887788778878,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02746800249820822,
      "loss": 0.2283,
      "num_input_tokens_seen": 10315200,
      "step": 48885
    },
    {
      "epoch": 5.378437843784378,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.027467201816643787,
      "loss": 0.2325,
      "num_input_tokens_seen": 10316192,
      "step": 48890
    },
    {
      "epoch": 5.378987898789879,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.027466401020175137,
      "loss": 0.2324,
      "num_input_tokens_seen": 10317216,
      "step": 48895
    },
    {
      "epoch": 5.37953795379538,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.027465600108809655,
      "loss": 0.2314,
      "num_input_tokens_seen": 10318272,
      "step": 48900
    },
    {
      "epoch": 5.38008800880088,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027464799082554718,
      "loss": 0.2319,
      "num_input_tokens_seen": 10319328,
      "step": 48905
    },
    {
      "epoch": 5.380638063806381,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027463997941417708,
      "loss": 0.2309,
      "num_input_tokens_seen": 10320384,
      "step": 48910
    },
    {
      "epoch": 5.381188118811881,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02746319668540601,
      "loss": 0.2329,
      "num_input_tokens_seen": 10321472,
      "step": 48915
    },
    {
      "epoch": 5.381738173817382,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.027462395314527006,
      "loss": 0.2309,
      "num_input_tokens_seen": 10322528,
      "step": 48920
    },
    {
      "epoch": 5.382288228822882,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.027461593828788086,
      "loss": 0.2309,
      "num_input_tokens_seen": 10323584,
      "step": 48925
    },
    {
      "epoch": 5.382838283828383,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.027460792228196633,
      "loss": 0.2314,
      "num_input_tokens_seen": 10324640,
      "step": 48930
    },
    {
      "epoch": 5.383388338833884,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02745999051276004,
      "loss": 0.2309,
      "num_input_tokens_seen": 10325728,
      "step": 48935
    },
    {
      "epoch": 5.383938393839384,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.027459188682485695,
      "loss": 0.2324,
      "num_input_tokens_seen": 10326816,
      "step": 48940
    },
    {
      "epoch": 5.384488448844884,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02745838673738098,
      "loss": 0.2303,
      "num_input_tokens_seen": 10327840,
      "step": 48945
    },
    {
      "epoch": 5.385038503850385,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.0274575846774533,
      "loss": 0.2293,
      "num_input_tokens_seen": 10328960,
      "step": 48950
    },
    {
      "epoch": 5.385588558855885,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027456782502710036,
      "loss": 0.2324,
      "num_input_tokens_seen": 10330016,
      "step": 48955
    },
    {
      "epoch": 5.3861386138613865,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.027455980213158582,
      "loss": 0.2324,
      "num_input_tokens_seen": 10331040,
      "step": 48960
    },
    {
      "epoch": 5.386688668866887,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027455177808806342,
      "loss": 0.233,
      "num_input_tokens_seen": 10332032,
      "step": 48965
    },
    {
      "epoch": 5.387238723872387,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.0274543752896607,
      "loss": 0.2309,
      "num_input_tokens_seen": 10333120,
      "step": 48970
    },
    {
      "epoch": 5.387788778877888,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02745357265572906,
      "loss": 0.2314,
      "num_input_tokens_seen": 10334144,
      "step": 48975
    },
    {
      "epoch": 5.388338833883388,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.027452769907018816,
      "loss": 0.233,
      "num_input_tokens_seen": 10335168,
      "step": 48980
    },
    {
      "epoch": 5.388888888888889,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.027451967043537365,
      "loss": 0.2325,
      "num_input_tokens_seen": 10336256,
      "step": 48985
    },
    {
      "epoch": 5.3894389438943895,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.027451164065292112,
      "loss": 0.2319,
      "num_input_tokens_seen": 10337376,
      "step": 48990
    },
    {
      "epoch": 5.38998899889989,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.027450360972290453,
      "loss": 0.2329,
      "num_input_tokens_seen": 10338464,
      "step": 48995
    },
    {
      "epoch": 5.390539053905391,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02744955776453979,
      "loss": 0.2314,
      "num_input_tokens_seen": 10339552,
      "step": 49000
    },
    {
      "epoch": 5.391089108910891,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.02744875444204753,
      "loss": 0.2313,
      "num_input_tokens_seen": 10340608,
      "step": 49005
    },
    {
      "epoch": 5.391639163916391,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027447951004821065,
      "loss": 0.2308,
      "num_input_tokens_seen": 10341664,
      "step": 49010
    },
    {
      "epoch": 5.392189218921892,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.027447147452867818,
      "loss": 0.2329,
      "num_input_tokens_seen": 10342752,
      "step": 49015
    },
    {
      "epoch": 5.3927392739273925,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027446343786195187,
      "loss": 0.2329,
      "num_input_tokens_seen": 10343776,
      "step": 49020
    },
    {
      "epoch": 5.393289328932894,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02744554000481057,
      "loss": 0.2314,
      "num_input_tokens_seen": 10344896,
      "step": 49025
    },
    {
      "epoch": 5.393839383938394,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.027444736108721383,
      "loss": 0.2298,
      "num_input_tokens_seen": 10345920,
      "step": 49030
    },
    {
      "epoch": 5.394389438943894,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.027443932097935043,
      "loss": 0.2319,
      "num_input_tokens_seen": 10346912,
      "step": 49035
    },
    {
      "epoch": 5.394939493949395,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.027443127972458944,
      "loss": 0.2303,
      "num_input_tokens_seen": 10347904,
      "step": 49040
    },
    {
      "epoch": 5.395489548954895,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.02744232373230051,
      "loss": 0.2319,
      "num_input_tokens_seen": 10348992,
      "step": 49045
    },
    {
      "epoch": 5.396039603960396,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02744151937746715,
      "loss": 0.2319,
      "num_input_tokens_seen": 10350016,
      "step": 49050
    },
    {
      "epoch": 5.396589658965897,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02744071490796627,
      "loss": 0.2309,
      "num_input_tokens_seen": 10351072,
      "step": 49055
    },
    {
      "epoch": 5.397139713971397,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027439910323805293,
      "loss": 0.2324,
      "num_input_tokens_seen": 10352096,
      "step": 49060
    },
    {
      "epoch": 5.397689768976898,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027439105624991633,
      "loss": 0.2308,
      "num_input_tokens_seen": 10353152,
      "step": 49065
    },
    {
      "epoch": 5.398239823982398,
      "grad_norm": 0.00080108642578125,
      "learning_rate": 0.027438300811532704,
      "loss": 0.2303,
      "num_input_tokens_seen": 10354272,
      "step": 49070
    },
    {
      "epoch": 5.398789878987899,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.027437495883435924,
      "loss": 0.2319,
      "num_input_tokens_seen": 10355296,
      "step": 49075
    },
    {
      "epoch": 5.399339933993399,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027436690840708718,
      "loss": 0.2319,
      "num_input_tokens_seen": 10356288,
      "step": 49080
    },
    {
      "epoch": 5.3998899889989,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.027435885683358495,
      "loss": 0.2283,
      "num_input_tokens_seen": 10357344,
      "step": 49085
    },
    {
      "epoch": 5.400440044004401,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.02743508041139268,
      "loss": 0.2314,
      "num_input_tokens_seen": 10358336,
      "step": 49090
    },
    {
      "epoch": 5.400990099009901,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.0274342750248187,
      "loss": 0.2308,
      "num_input_tokens_seen": 10359328,
      "step": 49095
    },
    {
      "epoch": 5.401540154015402,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027433469523643974,
      "loss": 0.2298,
      "num_input_tokens_seen": 10360352,
      "step": 49100
    },
    {
      "epoch": 5.402090209020902,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.02743266390787592,
      "loss": 0.2308,
      "num_input_tokens_seen": 10361344,
      "step": 49105
    },
    {
      "epoch": 5.402640264026402,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.027431858177521973,
      "loss": 0.2319,
      "num_input_tokens_seen": 10362368,
      "step": 49110
    },
    {
      "epoch": 5.4031903190319035,
      "grad_norm": 0.0006561279296875,
      "learning_rate": 0.02743105233258955,
      "loss": 0.2324,
      "num_input_tokens_seen": 10363424,
      "step": 49115
    },
    {
      "epoch": 5.403740374037404,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.027430246373086088,
      "loss": 0.2298,
      "num_input_tokens_seen": 10364480,
      "step": 49120
    },
    {
      "epoch": 5.404290429042904,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.027429440299019007,
      "loss": 0.2308,
      "num_input_tokens_seen": 10365472,
      "step": 49125
    },
    {
      "epoch": 5.404840484048405,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02742863411039574,
      "loss": 0.2314,
      "num_input_tokens_seen": 10366528,
      "step": 49130
    },
    {
      "epoch": 5.405390539053905,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.027427827807223713,
      "loss": 0.2324,
      "num_input_tokens_seen": 10367552,
      "step": 49135
    },
    {
      "epoch": 5.405940594059406,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.027427021389510364,
      "loss": 0.2298,
      "num_input_tokens_seen": 10368608,
      "step": 49140
    },
    {
      "epoch": 5.4064906490649065,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.027426214857263122,
      "loss": 0.2319,
      "num_input_tokens_seen": 10369664,
      "step": 49145
    },
    {
      "epoch": 5.407040704070407,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02742540821048942,
      "loss": 0.2319,
      "num_input_tokens_seen": 10370720,
      "step": 49150
    },
    {
      "epoch": 5.407590759075908,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.027424601449196694,
      "loss": 0.233,
      "num_input_tokens_seen": 10371680,
      "step": 49155
    },
    {
      "epoch": 5.408140814081408,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02742379457339238,
      "loss": 0.2309,
      "num_input_tokens_seen": 10372704,
      "step": 49160
    },
    {
      "epoch": 5.408690869086909,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.027422987583083907,
      "loss": 0.2314,
      "num_input_tokens_seen": 10373760,
      "step": 49165
    },
    {
      "epoch": 5.409240924092409,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02742218047827872,
      "loss": 0.2309,
      "num_input_tokens_seen": 10374752,
      "step": 49170
    },
    {
      "epoch": 5.4097909790979095,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.027421373258984257,
      "loss": 0.234,
      "num_input_tokens_seen": 10375808,
      "step": 49175
    },
    {
      "epoch": 5.410341034103411,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.02742056592520796,
      "loss": 0.2325,
      "num_input_tokens_seen": 10376800,
      "step": 49180
    },
    {
      "epoch": 5.410891089108911,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.027419758476957264,
      "loss": 0.2318,
      "num_input_tokens_seen": 10377824,
      "step": 49185
    },
    {
      "epoch": 5.411441144114411,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.027418950914239616,
      "loss": 0.2314,
      "num_input_tokens_seen": 10379008,
      "step": 49190
    },
    {
      "epoch": 5.411991199119912,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027418143237062455,
      "loss": 0.2309,
      "num_input_tokens_seen": 10380032,
      "step": 49195
    },
    {
      "epoch": 5.412541254125412,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.027417335445433223,
      "loss": 0.2308,
      "num_input_tokens_seen": 10381056,
      "step": 49200
    },
    {
      "epoch": 5.413091309130913,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.027416527539359373,
      "loss": 0.2309,
      "num_input_tokens_seen": 10382176,
      "step": 49205
    },
    {
      "epoch": 5.413641364136414,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027415719518848348,
      "loss": 0.2314,
      "num_input_tokens_seen": 10383232,
      "step": 49210
    },
    {
      "epoch": 5.414191419141914,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.027414911383907586,
      "loss": 0.2314,
      "num_input_tokens_seen": 10384288,
      "step": 49215
    },
    {
      "epoch": 5.414741474147415,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.02741410313454455,
      "loss": 0.233,
      "num_input_tokens_seen": 10385376,
      "step": 49220
    },
    {
      "epoch": 5.415291529152915,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.027413294770766684,
      "loss": 0.2314,
      "num_input_tokens_seen": 10386464,
      "step": 49225
    },
    {
      "epoch": 5.415841584158416,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.027412486292581435,
      "loss": 0.2314,
      "num_input_tokens_seen": 10387584,
      "step": 49230
    },
    {
      "epoch": 5.416391639163916,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.02741167769999625,
      "loss": 0.2356,
      "num_input_tokens_seen": 10388608,
      "step": 49235
    },
    {
      "epoch": 5.416941694169417,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.027410868993018596,
      "loss": 0.2313,
      "num_input_tokens_seen": 10389568,
      "step": 49240
    },
    {
      "epoch": 5.417491749174918,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.027410060171655908,
      "loss": 0.2314,
      "num_input_tokens_seen": 10390592,
      "step": 49245
    },
    {
      "epoch": 5.418041804180418,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.027409251235915658,
      "loss": 0.2313,
      "num_input_tokens_seen": 10391648,
      "step": 49250
    },
    {
      "epoch": 5.418591859185918,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.027408442185805293,
      "loss": 0.2303,
      "num_input_tokens_seen": 10392768,
      "step": 49255
    },
    {
      "epoch": 5.419141914191419,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.02740763302133227,
      "loss": 0.2319,
      "num_input_tokens_seen": 10393824,
      "step": 49260
    },
    {
      "epoch": 5.419691969196919,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.027406823742504045,
      "loss": 0.2303,
      "num_input_tokens_seen": 10394880,
      "step": 49265
    },
    {
      "epoch": 5.4202420242024205,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02740601434932808,
      "loss": 0.2329,
      "num_input_tokens_seen": 10396000,
      "step": 49270
    },
    {
      "epoch": 5.420792079207921,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.027405204841811834,
      "loss": 0.2298,
      "num_input_tokens_seen": 10397088,
      "step": 49275
    },
    {
      "epoch": 5.421342134213421,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02740439521996277,
      "loss": 0.233,
      "num_input_tokens_seen": 10398144,
      "step": 49280
    },
    {
      "epoch": 5.421892189218922,
      "grad_norm": 0.000667572021484375,
      "learning_rate": 0.027403585483788344,
      "loss": 0.2324,
      "num_input_tokens_seen": 10399136,
      "step": 49285
    },
    {
      "epoch": 5.422442244224422,
      "grad_norm": 0.000713348388671875,
      "learning_rate": 0.027402775633296024,
      "loss": 0.2309,
      "num_input_tokens_seen": 10400128,
      "step": 49290
    },
    {
      "epoch": 5.422992299229923,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02740196566849327,
      "loss": 0.2319,
      "num_input_tokens_seen": 10401184,
      "step": 49295
    },
    {
      "epoch": 5.4235423542354235,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.027401155589387553,
      "loss": 0.2324,
      "num_input_tokens_seen": 10402272,
      "step": 49300
    },
    {
      "epoch": 5.424092409240924,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.027400345395986334,
      "loss": 0.2325,
      "num_input_tokens_seen": 10403360,
      "step": 49305
    },
    {
      "epoch": 5.424642464246425,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.02739953508829708,
      "loss": 0.2293,
      "num_input_tokens_seen": 10404352,
      "step": 49310
    },
    {
      "epoch": 5.425192519251925,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.027398724666327263,
      "loss": 0.2304,
      "num_input_tokens_seen": 10405376,
      "step": 49315
    },
    {
      "epoch": 5.425742574257426,
      "grad_norm": 0.009765625,
      "learning_rate": 0.027397914130084352,
      "loss": 0.2299,
      "num_input_tokens_seen": 10406432,
      "step": 49320
    },
    {
      "epoch": 5.426292629262926,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027397103479575816,
      "loss": 0.2319,
      "num_input_tokens_seen": 10407520,
      "step": 49325
    },
    {
      "epoch": 5.4268426842684265,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027396292714809117,
      "loss": 0.2325,
      "num_input_tokens_seen": 10408512,
      "step": 49330
    },
    {
      "epoch": 5.427392739273928,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027395481835791744,
      "loss": 0.2319,
      "num_input_tokens_seen": 10409536,
      "step": 49335
    },
    {
      "epoch": 5.427942794279428,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.027394670842531166,
      "loss": 0.2329,
      "num_input_tokens_seen": 10410528,
      "step": 49340
    },
    {
      "epoch": 5.428492849284929,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02739385973503485,
      "loss": 0.2319,
      "num_input_tokens_seen": 10411552,
      "step": 49345
    },
    {
      "epoch": 5.429042904290429,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.027393048513310276,
      "loss": 0.2303,
      "num_input_tokens_seen": 10412640,
      "step": 49350
    },
    {
      "epoch": 5.429592959295929,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.027392237177364918,
      "loss": 0.2313,
      "num_input_tokens_seen": 10413760,
      "step": 49355
    },
    {
      "epoch": 5.43014301430143,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.027391425727206258,
      "loss": 0.2334,
      "num_input_tokens_seen": 10414816,
      "step": 49360
    },
    {
      "epoch": 5.430693069306931,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.027390614162841775,
      "loss": 0.2313,
      "num_input_tokens_seen": 10415840,
      "step": 49365
    },
    {
      "epoch": 5.431243124312431,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027389802484278948,
      "loss": 0.2334,
      "num_input_tokens_seen": 10416896,
      "step": 49370
    },
    {
      "epoch": 5.431793179317932,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.027388990691525252,
      "loss": 0.2308,
      "num_input_tokens_seen": 10417984,
      "step": 49375
    },
    {
      "epoch": 5.432343234323432,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.027388178784588175,
      "loss": 0.2308,
      "num_input_tokens_seen": 10419008,
      "step": 49380
    },
    {
      "epoch": 5.432893289328933,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.027387366763475203,
      "loss": 0.2329,
      "num_input_tokens_seen": 10420032,
      "step": 49385
    },
    {
      "epoch": 5.433443344334433,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.027386554628193813,
      "loss": 0.2319,
      "num_input_tokens_seen": 10421056,
      "step": 49390
    },
    {
      "epoch": 5.433993399339934,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02738574237875149,
      "loss": 0.2314,
      "num_input_tokens_seen": 10422112,
      "step": 49395
    },
    {
      "epoch": 5.434543454345435,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.02738493001515573,
      "loss": 0.2313,
      "num_input_tokens_seen": 10423136,
      "step": 49400
    },
    {
      "epoch": 5.435093509350935,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.027384117537414,
      "loss": 0.2319,
      "num_input_tokens_seen": 10424128,
      "step": 49405
    },
    {
      "epoch": 5.435643564356436,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.02738330494553381,
      "loss": 0.2308,
      "num_input_tokens_seen": 10425248,
      "step": 49410
    },
    {
      "epoch": 5.436193619361936,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02738249223952264,
      "loss": 0.2303,
      "num_input_tokens_seen": 10426240,
      "step": 49415
    },
    {
      "epoch": 5.436743674367436,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.02738167941938798,
      "loss": 0.2324,
      "num_input_tokens_seen": 10427296,
      "step": 49420
    },
    {
      "epoch": 5.4372937293729375,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.027380866485137323,
      "loss": 0.2324,
      "num_input_tokens_seen": 10428352,
      "step": 49425
    },
    {
      "epoch": 5.437843784378438,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.027380053436778157,
      "loss": 0.2324,
      "num_input_tokens_seen": 10429376,
      "step": 49430
    },
    {
      "epoch": 5.438393839383938,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.027379240274317984,
      "loss": 0.2319,
      "num_input_tokens_seen": 10430400,
      "step": 49435
    },
    {
      "epoch": 5.438943894389439,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02737842699776429,
      "loss": 0.2303,
      "num_input_tokens_seen": 10431456,
      "step": 49440
    },
    {
      "epoch": 5.439493949394939,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.027377613607124575,
      "loss": 0.2303,
      "num_input_tokens_seen": 10432512,
      "step": 49445
    },
    {
      "epoch": 5.44004400440044,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.027376800102406337,
      "loss": 0.2324,
      "num_input_tokens_seen": 10433568,
      "step": 49450
    },
    {
      "epoch": 5.4405940594059405,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027375986483617068,
      "loss": 0.2303,
      "num_input_tokens_seen": 10434688,
      "step": 49455
    },
    {
      "epoch": 5.441144114411441,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.027375172750764273,
      "loss": 0.2309,
      "num_input_tokens_seen": 10435776,
      "step": 49460
    },
    {
      "epoch": 5.441694169416942,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.027374358903855447,
      "loss": 0.2319,
      "num_input_tokens_seen": 10436832,
      "step": 49465
    },
    {
      "epoch": 5.442244224422442,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.027373544942898096,
      "loss": 0.2324,
      "num_input_tokens_seen": 10437952,
      "step": 49470
    },
    {
      "epoch": 5.442794279427943,
      "grad_norm": 0.00077056884765625,
      "learning_rate": 0.027372730867899712,
      "loss": 0.2303,
      "num_input_tokens_seen": 10439040,
      "step": 49475
    },
    {
      "epoch": 5.443344334433443,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02737191667886781,
      "loss": 0.2324,
      "num_input_tokens_seen": 10440064,
      "step": 49480
    },
    {
      "epoch": 5.4438943894389435,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.027371102375809885,
      "loss": 0.2319,
      "num_input_tokens_seen": 10441152,
      "step": 49485
    },
    {
      "epoch": 5.444444444444445,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.027370287958733447,
      "loss": 0.2324,
      "num_input_tokens_seen": 10442208,
      "step": 49490
    },
    {
      "epoch": 5.444994499449945,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.027369473427646,
      "loss": 0.2308,
      "num_input_tokens_seen": 10443232,
      "step": 49495
    },
    {
      "epoch": 5.445544554455446,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.027368658782555056,
      "loss": 0.2314,
      "num_input_tokens_seen": 10444320,
      "step": 49500
    },
    {
      "epoch": 5.446094609460946,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02736784402346812,
      "loss": 0.2303,
      "num_input_tokens_seen": 10445376,
      "step": 49505
    },
    {
      "epoch": 5.446644664466446,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.027367029150392696,
      "loss": 0.2314,
      "num_input_tokens_seen": 10446400,
      "step": 49510
    },
    {
      "epoch": 5.447194719471947,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027366214163336296,
      "loss": 0.2324,
      "num_input_tokens_seen": 10447456,
      "step": 49515
    },
    {
      "epoch": 5.447744774477448,
      "grad_norm": 0.00054931640625,
      "learning_rate": 0.027365399062306436,
      "loss": 0.2298,
      "num_input_tokens_seen": 10448480,
      "step": 49520
    },
    {
      "epoch": 5.448294829482949,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02736458384731063,
      "loss": 0.2309,
      "num_input_tokens_seen": 10449536,
      "step": 49525
    },
    {
      "epoch": 5.448844884488449,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027363768518356386,
      "loss": 0.2293,
      "num_input_tokens_seen": 10450592,
      "step": 49530
    },
    {
      "epoch": 5.449394939493949,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.027362953075451217,
      "loss": 0.234,
      "num_input_tokens_seen": 10451680,
      "step": 49535
    },
    {
      "epoch": 5.44994499449945,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02736213751860265,
      "loss": 0.2298,
      "num_input_tokens_seen": 10452704,
      "step": 49540
    },
    {
      "epoch": 5.4504950495049505,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.027361321847818188,
      "loss": 0.2319,
      "num_input_tokens_seen": 10453696,
      "step": 49545
    },
    {
      "epoch": 5.451045104510451,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.02736050606310535,
      "loss": 0.2319,
      "num_input_tokens_seen": 10454752,
      "step": 49550
    },
    {
      "epoch": 5.451595159515952,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.027359690164471664,
      "loss": 0.2308,
      "num_input_tokens_seen": 10455840,
      "step": 49555
    },
    {
      "epoch": 5.452145214521452,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02735887415192464,
      "loss": 0.2324,
      "num_input_tokens_seen": 10456928,
      "step": 49560
    },
    {
      "epoch": 5.452695269526953,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02735805802547181,
      "loss": 0.2319,
      "num_input_tokens_seen": 10457920,
      "step": 49565
    },
    {
      "epoch": 5.453245324532453,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.02735724178512068,
      "loss": 0.2308,
      "num_input_tokens_seen": 10458944,
      "step": 49570
    },
    {
      "epoch": 5.4537953795379535,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02735642543087879,
      "loss": 0.2319,
      "num_input_tokens_seen": 10460000,
      "step": 49575
    },
    {
      "epoch": 5.4543454345434546,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.027355608962753655,
      "loss": 0.2309,
      "num_input_tokens_seen": 10461088,
      "step": 49580
    },
    {
      "epoch": 5.454895489548955,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0273547923807528,
      "loss": 0.232,
      "num_input_tokens_seen": 10462176,
      "step": 49585
    },
    {
      "epoch": 5.455445544554456,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.02735397568488375,
      "loss": 0.2283,
      "num_input_tokens_seen": 10463232,
      "step": 49590
    },
    {
      "epoch": 5.455995599559956,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02735315887515404,
      "loss": 0.2305,
      "num_input_tokens_seen": 10464256,
      "step": 49595
    },
    {
      "epoch": 5.456545654565456,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.027352341951571185,
      "loss": 0.2294,
      "num_input_tokens_seen": 10465312,
      "step": 49600
    },
    {
      "epoch": 5.457095709570957,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027351524914142722,
      "loss": 0.2299,
      "num_input_tokens_seen": 10466304,
      "step": 49605
    },
    {
      "epoch": 5.457645764576458,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.027350707762876188,
      "loss": 0.2331,
      "num_input_tokens_seen": 10467360,
      "step": 49610
    },
    {
      "epoch": 5.458195819581958,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0273498904977791,
      "loss": 0.2299,
      "num_input_tokens_seen": 10468352,
      "step": 49615
    },
    {
      "epoch": 5.458745874587459,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.027349073118858997,
      "loss": 0.2289,
      "num_input_tokens_seen": 10469472,
      "step": 49620
    },
    {
      "epoch": 5.459295929592959,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.02734825562612342,
      "loss": 0.232,
      "num_input_tokens_seen": 10470624,
      "step": 49625
    },
    {
      "epoch": 5.45984598459846,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.02734743801957989,
      "loss": 0.2326,
      "num_input_tokens_seen": 10471648,
      "step": 49630
    },
    {
      "epoch": 5.46039603960396,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02734662029923595,
      "loss": 0.2248,
      "num_input_tokens_seen": 10472704,
      "step": 49635
    },
    {
      "epoch": 5.460946094609461,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.027345802465099135,
      "loss": 0.2296,
      "num_input_tokens_seen": 10473792,
      "step": 49640
    },
    {
      "epoch": 5.461496149614962,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.027344984517176985,
      "loss": 0.229,
      "num_input_tokens_seen": 10474848,
      "step": 49645
    },
    {
      "epoch": 5.462046204620462,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.027344166455477034,
      "loss": 0.2324,
      "num_input_tokens_seen": 10475936,
      "step": 49650
    },
    {
      "epoch": 5.462596259625963,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.027343348280006827,
      "loss": 0.2412,
      "num_input_tokens_seen": 10476960,
      "step": 49655
    },
    {
      "epoch": 5.463146314631463,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.0273425299907739,
      "loss": 0.2327,
      "num_input_tokens_seen": 10477984,
      "step": 49660
    },
    {
      "epoch": 5.463696369636963,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.027341711587785793,
      "loss": 0.2316,
      "num_input_tokens_seen": 10479040,
      "step": 49665
    },
    {
      "epoch": 5.4642464246424645,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.027340893071050056,
      "loss": 0.2321,
      "num_input_tokens_seen": 10480064,
      "step": 49670
    },
    {
      "epoch": 5.464796479647965,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.02734007444057423,
      "loss": 0.2269,
      "num_input_tokens_seen": 10481088,
      "step": 49675
    },
    {
      "epoch": 5.465346534653466,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.027339255696365857,
      "loss": 0.2347,
      "num_input_tokens_seen": 10482208,
      "step": 49680
    },
    {
      "epoch": 5.465896589658966,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027338436838432487,
      "loss": 0.2331,
      "num_input_tokens_seen": 10483232,
      "step": 49685
    },
    {
      "epoch": 5.466446644664466,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.027337617866781665,
      "loss": 0.2331,
      "num_input_tokens_seen": 10484256,
      "step": 49690
    },
    {
      "epoch": 5.466996699669967,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.027336798781420937,
      "loss": 0.2305,
      "num_input_tokens_seen": 10485312,
      "step": 49695
    },
    {
      "epoch": 5.4675467546754675,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.027335979582357856,
      "loss": 0.2357,
      "num_input_tokens_seen": 10486400,
      "step": 49700
    },
    {
      "epoch": 5.468096809680969,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02733516026959997,
      "loss": 0.233,
      "num_input_tokens_seen": 10487424,
      "step": 49705
    },
    {
      "epoch": 5.468646864686469,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.027334340843154833,
      "loss": 0.2309,
      "num_input_tokens_seen": 10488480,
      "step": 49710
    },
    {
      "epoch": 5.469196919691969,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.027333521303029993,
      "loss": 0.2304,
      "num_input_tokens_seen": 10489536,
      "step": 49715
    },
    {
      "epoch": 5.46974697469747,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02733270164923301,
      "loss": 0.2293,
      "num_input_tokens_seen": 10490688,
      "step": 49720
    },
    {
      "epoch": 5.47029702970297,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.027331881881771426,
      "loss": 0.2325,
      "num_input_tokens_seen": 10491712,
      "step": 49725
    },
    {
      "epoch": 5.4708470847084705,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.027331062000652804,
      "loss": 0.2303,
      "num_input_tokens_seen": 10492832,
      "step": 49730
    },
    {
      "epoch": 5.471397139713972,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.027330242005884704,
      "loss": 0.2304,
      "num_input_tokens_seen": 10493888,
      "step": 49735
    },
    {
      "epoch": 5.471947194719472,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.027329421897474684,
      "loss": 0.2335,
      "num_input_tokens_seen": 10494912,
      "step": 49740
    },
    {
      "epoch": 5.472497249724973,
      "grad_norm": 0.000797271728515625,
      "learning_rate": 0.027328601675430293,
      "loss": 0.2288,
      "num_input_tokens_seen": 10495968,
      "step": 49745
    },
    {
      "epoch": 5.473047304730473,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.027327781339759098,
      "loss": 0.2315,
      "num_input_tokens_seen": 10497056,
      "step": 49750
    },
    {
      "epoch": 5.473597359735973,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02732696089046866,
      "loss": 0.2289,
      "num_input_tokens_seen": 10498112,
      "step": 49755
    },
    {
      "epoch": 5.474147414741474,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.027326140327566535,
      "loss": 0.2326,
      "num_input_tokens_seen": 10499136,
      "step": 49760
    },
    {
      "epoch": 5.474697469746975,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.027325319651060293,
      "loss": 0.23,
      "num_input_tokens_seen": 10500128,
      "step": 49765
    },
    {
      "epoch": 5.475247524752476,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.027324498860957492,
      "loss": 0.2305,
      "num_input_tokens_seen": 10501152,
      "step": 49770
    },
    {
      "epoch": 5.475797579757976,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027323677957265695,
      "loss": 0.2346,
      "num_input_tokens_seen": 10502240,
      "step": 49775
    },
    {
      "epoch": 5.476347634763476,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.027322856939992476,
      "loss": 0.2325,
      "num_input_tokens_seen": 10503264,
      "step": 49780
    },
    {
      "epoch": 5.476897689768977,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.027322035809145398,
      "loss": 0.232,
      "num_input_tokens_seen": 10504384,
      "step": 49785
    },
    {
      "epoch": 5.477447744774477,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02732121456473203,
      "loss": 0.2315,
      "num_input_tokens_seen": 10505440,
      "step": 49790
    },
    {
      "epoch": 5.477997799779978,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.027320393206759934,
      "loss": 0.233,
      "num_input_tokens_seen": 10506496,
      "step": 49795
    },
    {
      "epoch": 5.478547854785479,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02731957173523669,
      "loss": 0.2304,
      "num_input_tokens_seen": 10507616,
      "step": 49800
    },
    {
      "epoch": 5.479097909790979,
      "grad_norm": 0.009765625,
      "learning_rate": 0.027318750150169862,
      "loss": 0.2263,
      "num_input_tokens_seen": 10508704,
      "step": 49805
    },
    {
      "epoch": 5.47964796479648,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.02731792845156703,
      "loss": 0.232,
      "num_input_tokens_seen": 10509728,
      "step": 49810
    },
    {
      "epoch": 5.48019801980198,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.027317106639435755,
      "loss": 0.231,
      "num_input_tokens_seen": 10510784,
      "step": 49815
    },
    {
      "epoch": 5.48074807480748,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027316284713783626,
      "loss": 0.2353,
      "num_input_tokens_seen": 10511808,
      "step": 49820
    },
    {
      "epoch": 5.4812981298129815,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027315462674618207,
      "loss": 0.233,
      "num_input_tokens_seen": 10512864,
      "step": 49825
    },
    {
      "epoch": 5.481848184818482,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.027314640521947077,
      "loss": 0.2326,
      "num_input_tokens_seen": 10513920,
      "step": 49830
    },
    {
      "epoch": 5.482398239823983,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.027313818255777816,
      "loss": 0.2325,
      "num_input_tokens_seen": 10515040,
      "step": 49835
    },
    {
      "epoch": 5.482948294829483,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.027312995876118,
      "loss": 0.234,
      "num_input_tokens_seen": 10516128,
      "step": 49840
    },
    {
      "epoch": 5.483498349834983,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02731217338297521,
      "loss": 0.233,
      "num_input_tokens_seen": 10517152,
      "step": 49845
    },
    {
      "epoch": 5.484048404840484,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027311350776357025,
      "loss": 0.2298,
      "num_input_tokens_seen": 10518144,
      "step": 49850
    },
    {
      "epoch": 5.4845984598459845,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.02731052805627103,
      "loss": 0.2314,
      "num_input_tokens_seen": 10519168,
      "step": 49855
    },
    {
      "epoch": 5.485148514851485,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0273097052227248,
      "loss": 0.2319,
      "num_input_tokens_seen": 10520256,
      "step": 49860
    },
    {
      "epoch": 5.485698569856986,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02730888227572593,
      "loss": 0.2303,
      "num_input_tokens_seen": 10521344,
      "step": 49865
    },
    {
      "epoch": 5.486248624862486,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.027308059215281993,
      "loss": 0.2324,
      "num_input_tokens_seen": 10522368,
      "step": 49870
    },
    {
      "epoch": 5.486798679867987,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.027307236041400583,
      "loss": 0.2329,
      "num_input_tokens_seen": 10523392,
      "step": 49875
    },
    {
      "epoch": 5.487348734873487,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.027306412754089286,
      "loss": 0.2283,
      "num_input_tokens_seen": 10524512,
      "step": 49880
    },
    {
      "epoch": 5.4878987898789875,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.027305589353355687,
      "loss": 0.2288,
      "num_input_tokens_seen": 10525568,
      "step": 49885
    },
    {
      "epoch": 5.488448844884489,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.027304765839207372,
      "loss": 0.234,
      "num_input_tokens_seen": 10526656,
      "step": 49890
    },
    {
      "epoch": 5.488998899889989,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.027303942211651937,
      "loss": 0.2288,
      "num_input_tokens_seen": 10527680,
      "step": 49895
    },
    {
      "epoch": 5.48954895489549,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.02730311847069697,
      "loss": 0.2299,
      "num_input_tokens_seen": 10528768,
      "step": 49900
    },
    {
      "epoch": 5.49009900990099,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.027302294616350066,
      "loss": 0.2325,
      "num_input_tokens_seen": 10529824,
      "step": 49905
    },
    {
      "epoch": 5.49064906490649,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027301470648618813,
      "loss": 0.2331,
      "num_input_tokens_seen": 10530880,
      "step": 49910
    },
    {
      "epoch": 5.491199119911991,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027300646567510807,
      "loss": 0.2335,
      "num_input_tokens_seen": 10531904,
      "step": 49915
    },
    {
      "epoch": 5.491749174917492,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.027299822373033647,
      "loss": 0.2315,
      "num_input_tokens_seen": 10532960,
      "step": 49920
    },
    {
      "epoch": 5.492299229922993,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027298998065194927,
      "loss": 0.2314,
      "num_input_tokens_seen": 10534016,
      "step": 49925
    },
    {
      "epoch": 5.492849284928493,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02729817364400224,
      "loss": 0.2325,
      "num_input_tokens_seen": 10535104,
      "step": 49930
    },
    {
      "epoch": 5.493399339933993,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.027297349109463187,
      "loss": 0.2336,
      "num_input_tokens_seen": 10536096,
      "step": 49935
    },
    {
      "epoch": 5.493949394939494,
      "grad_norm": 0.000812530517578125,
      "learning_rate": 0.02729652446158537,
      "loss": 0.2298,
      "num_input_tokens_seen": 10537120,
      "step": 49940
    },
    {
      "epoch": 5.494499449944994,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02729569970037639,
      "loss": 0.2299,
      "num_input_tokens_seen": 10538176,
      "step": 49945
    },
    {
      "epoch": 5.4950495049504955,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.027294874825843845,
      "loss": 0.2346,
      "num_input_tokens_seen": 10539296,
      "step": 49950
    },
    {
      "epoch": 5.495599559955996,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.027294049837995334,
      "loss": 0.2314,
      "num_input_tokens_seen": 10540352,
      "step": 49955
    },
    {
      "epoch": 5.496149614961496,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02729322473683847,
      "loss": 0.2293,
      "num_input_tokens_seen": 10541440,
      "step": 49960
    },
    {
      "epoch": 5.496699669966997,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02729239952238085,
      "loss": 0.2335,
      "num_input_tokens_seen": 10542496,
      "step": 49965
    },
    {
      "epoch": 5.497249724972497,
      "grad_norm": 0.000888824462890625,
      "learning_rate": 0.027291574194630082,
      "loss": 0.2315,
      "num_input_tokens_seen": 10543552,
      "step": 49970
    },
    {
      "epoch": 5.497799779977997,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.027290748753593774,
      "loss": 0.2346,
      "num_input_tokens_seen": 10544640,
      "step": 49975
    },
    {
      "epoch": 5.4983498349834985,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.027289923199279534,
      "loss": 0.2283,
      "num_input_tokens_seen": 10545632,
      "step": 49980
    },
    {
      "epoch": 5.498899889988999,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027289097531694964,
      "loss": 0.2283,
      "num_input_tokens_seen": 10546752,
      "step": 49985
    },
    {
      "epoch": 5.4994499449945,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.027288271750847682,
      "loss": 0.2336,
      "num_input_tokens_seen": 10547744,
      "step": 49990
    },
    {
      "epoch": 5.5,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.027287445856745297,
      "loss": 0.2274,
      "num_input_tokens_seen": 10548768,
      "step": 49995
    },
    {
      "epoch": 5.5005500550055,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.027286619849395417,
      "loss": 0.2337,
      "num_input_tokens_seen": 10549856,
      "step": 50000
    },
    {
      "epoch": 5.501100110011001,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02728579372880566,
      "loss": 0.2326,
      "num_input_tokens_seen": 10551040,
      "step": 50005
    },
    {
      "epoch": 5.5016501650165015,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.027284967494983634,
      "loss": 0.232,
      "num_input_tokens_seen": 10552160,
      "step": 50010
    },
    {
      "epoch": 5.502200220022003,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.027284141147936963,
      "loss": 0.232,
      "num_input_tokens_seen": 10553216,
      "step": 50015
    },
    {
      "epoch": 5.502750275027503,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02728331468767325,
      "loss": 0.2284,
      "num_input_tokens_seen": 10554240,
      "step": 50020
    },
    {
      "epoch": 5.503300330033003,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.027282488114200126,
      "loss": 0.2289,
      "num_input_tokens_seen": 10555360,
      "step": 50025
    },
    {
      "epoch": 5.503850385038504,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.0272816614275252,
      "loss": 0.23,
      "num_input_tokens_seen": 10556448,
      "step": 50030
    },
    {
      "epoch": 5.504400440044004,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027280834627656094,
      "loss": 0.2378,
      "num_input_tokens_seen": 10557472,
      "step": 50035
    },
    {
      "epoch": 5.5049504950495045,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02728000771460043,
      "loss": 0.2233,
      "num_input_tokens_seen": 10558560,
      "step": 50040
    },
    {
      "epoch": 5.505500550055006,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.027279180688365826,
      "loss": 0.2264,
      "num_input_tokens_seen": 10559584,
      "step": 50045
    },
    {
      "epoch": 5.506050605060506,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.027278353548959907,
      "loss": 0.2322,
      "num_input_tokens_seen": 10560608,
      "step": 50050
    },
    {
      "epoch": 5.506600660066007,
      "grad_norm": 0.000690460205078125,
      "learning_rate": 0.027277526296390294,
      "loss": 0.2328,
      "num_input_tokens_seen": 10561664,
      "step": 50055
    },
    {
      "epoch": 5.507150715071507,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.02727669893066461,
      "loss": 0.2307,
      "num_input_tokens_seen": 10562688,
      "step": 50060
    },
    {
      "epoch": 5.507700770077007,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.027275871451790487,
      "loss": 0.2317,
      "num_input_tokens_seen": 10563776,
      "step": 50065
    },
    {
      "epoch": 5.508250825082508,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.027275043859775545,
      "loss": 0.2344,
      "num_input_tokens_seen": 10564864,
      "step": 50070
    },
    {
      "epoch": 5.508800880088009,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.027274216154627415,
      "loss": 0.2343,
      "num_input_tokens_seen": 10565920,
      "step": 50075
    },
    {
      "epoch": 5.50935093509351,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.027273388336353722,
      "loss": 0.2322,
      "num_input_tokens_seen": 10567008,
      "step": 50080
    },
    {
      "epoch": 5.50990099009901,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.027272560404962103,
      "loss": 0.2301,
      "num_input_tokens_seen": 10568064,
      "step": 50085
    },
    {
      "epoch": 5.51045104510451,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02727173236046018,
      "loss": 0.2321,
      "num_input_tokens_seen": 10569152,
      "step": 50090
    },
    {
      "epoch": 5.511001100110011,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02727090420285559,
      "loss": 0.2379,
      "num_input_tokens_seen": 10570208,
      "step": 50095
    },
    {
      "epoch": 5.511551155115511,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02727007593215596,
      "loss": 0.2295,
      "num_input_tokens_seen": 10571296,
      "step": 50100
    },
    {
      "epoch": 5.512101210121012,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02726924754836893,
      "loss": 0.2362,
      "num_input_tokens_seen": 10572320,
      "step": 50105
    },
    {
      "epoch": 5.512651265126513,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.027268419051502138,
      "loss": 0.232,
      "num_input_tokens_seen": 10573376,
      "step": 50110
    },
    {
      "epoch": 5.513201320132013,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.027267590441563207,
      "loss": 0.233,
      "num_input_tokens_seen": 10574432,
      "step": 50115
    },
    {
      "epoch": 5.513751375137514,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02726676171855979,
      "loss": 0.2314,
      "num_input_tokens_seen": 10575520,
      "step": 50120
    },
    {
      "epoch": 5.514301430143014,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02726593288249951,
      "loss": 0.2309,
      "num_input_tokens_seen": 10576544,
      "step": 50125
    },
    {
      "epoch": 5.514851485148515,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.02726510393339001,
      "loss": 0.2335,
      "num_input_tokens_seen": 10577536,
      "step": 50130
    },
    {
      "epoch": 5.5154015401540155,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.027264274871238937,
      "loss": 0.2304,
      "num_input_tokens_seen": 10578560,
      "step": 50135
    },
    {
      "epoch": 5.515951595159516,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.027263445696053927,
      "loss": 0.2319,
      "num_input_tokens_seen": 10579584,
      "step": 50140
    },
    {
      "epoch": 5.516501650165017,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.027262616407842624,
      "loss": 0.2288,
      "num_input_tokens_seen": 10580608,
      "step": 50145
    },
    {
      "epoch": 5.517051705170517,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.027261787006612668,
      "loss": 0.2293,
      "num_input_tokens_seen": 10581632,
      "step": 50150
    },
    {
      "epoch": 5.517601760176017,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.027260957492371705,
      "loss": 0.2309,
      "num_input_tokens_seen": 10582688,
      "step": 50155
    },
    {
      "epoch": 5.518151815181518,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.027260127865127378,
      "loss": 0.2278,
      "num_input_tokens_seen": 10583744,
      "step": 50160
    },
    {
      "epoch": 5.5187018701870185,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.027259298124887338,
      "loss": 0.2336,
      "num_input_tokens_seen": 10584800,
      "step": 50165
    },
    {
      "epoch": 5.51925192519252,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.02725846827165923,
      "loss": 0.232,
      "num_input_tokens_seen": 10585888,
      "step": 50170
    },
    {
      "epoch": 5.51980198019802,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027257638305450697,
      "loss": 0.2326,
      "num_input_tokens_seen": 10587008,
      "step": 50175
    },
    {
      "epoch": 5.52035203520352,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0272568082262694,
      "loss": 0.2336,
      "num_input_tokens_seen": 10588064,
      "step": 50180
    },
    {
      "epoch": 5.520902090209021,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.027255978034122982,
      "loss": 0.2336,
      "num_input_tokens_seen": 10589184,
      "step": 50185
    },
    {
      "epoch": 5.521452145214521,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.027255147729019092,
      "loss": 0.2263,
      "num_input_tokens_seen": 10590240,
      "step": 50190
    },
    {
      "epoch": 5.522002200220022,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.027254317310965392,
      "loss": 0.2274,
      "num_input_tokens_seen": 10591296,
      "step": 50195
    },
    {
      "epoch": 5.522552255225523,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.02725348677996952,
      "loss": 0.2315,
      "num_input_tokens_seen": 10592416,
      "step": 50200
    },
    {
      "epoch": 5.523102310231023,
      "grad_norm": 0.00457763671875,
      "learning_rate": 0.02725265613603915,
      "loss": 0.2284,
      "num_input_tokens_seen": 10593440,
      "step": 50205
    },
    {
      "epoch": 5.523652365236524,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.027251825379181922,
      "loss": 0.2331,
      "num_input_tokens_seen": 10594528,
      "step": 50210
    },
    {
      "epoch": 5.524202420242024,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.027250994509405504,
      "loss": 0.23,
      "num_input_tokens_seen": 10595584,
      "step": 50215
    },
    {
      "epoch": 5.524752475247524,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027250163526717542,
      "loss": 0.2295,
      "num_input_tokens_seen": 10596608,
      "step": 50220
    },
    {
      "epoch": 5.525302530253025,
      "grad_norm": 0.00080108642578125,
      "learning_rate": 0.027249332431125702,
      "loss": 0.2311,
      "num_input_tokens_seen": 10597632,
      "step": 50225
    },
    {
      "epoch": 5.525852585258526,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.027248501222637642,
      "loss": 0.2306,
      "num_input_tokens_seen": 10598656,
      "step": 50230
    },
    {
      "epoch": 5.526402640264027,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02724766990126103,
      "loss": 0.2285,
      "num_input_tokens_seen": 10599712,
      "step": 50235
    },
    {
      "epoch": 5.526952695269527,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.027246838467003518,
      "loss": 0.2306,
      "num_input_tokens_seen": 10600736,
      "step": 50240
    },
    {
      "epoch": 5.527502750275027,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02724600691987277,
      "loss": 0.2331,
      "num_input_tokens_seen": 10601824,
      "step": 50245
    },
    {
      "epoch": 5.528052805280528,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027245175259876454,
      "loss": 0.2333,
      "num_input_tokens_seen": 10602880,
      "step": 50250
    },
    {
      "epoch": 5.528602860286028,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.027244343487022236,
      "loss": 0.2342,
      "num_input_tokens_seen": 10603872,
      "step": 50255
    },
    {
      "epoch": 5.5291529152915295,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.02724351160131777,
      "loss": 0.2306,
      "num_input_tokens_seen": 10604896,
      "step": 50260
    },
    {
      "epoch": 5.52970297029703,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.027242679602770745,
      "loss": 0.2326,
      "num_input_tokens_seen": 10605952,
      "step": 50265
    },
    {
      "epoch": 5.53025302530253,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02724184749138881,
      "loss": 0.2331,
      "num_input_tokens_seen": 10606976,
      "step": 50270
    },
    {
      "epoch": 5.530803080308031,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.027241015267179643,
      "loss": 0.2341,
      "num_input_tokens_seen": 10608064,
      "step": 50275
    },
    {
      "epoch": 5.531353135313531,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.02724018293015091,
      "loss": 0.2335,
      "num_input_tokens_seen": 10609152,
      "step": 50280
    },
    {
      "epoch": 5.531903190319031,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.027239350480310284,
      "loss": 0.2335,
      "num_input_tokens_seen": 10610240,
      "step": 50285
    },
    {
      "epoch": 5.5324532453245325,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02723851791766544,
      "loss": 0.2308,
      "num_input_tokens_seen": 10611360,
      "step": 50290
    },
    {
      "epoch": 5.533003300330033,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.02723768524222405,
      "loss": 0.2309,
      "num_input_tokens_seen": 10612384,
      "step": 50295
    },
    {
      "epoch": 5.533553355335534,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.027236852453993782,
      "loss": 0.2314,
      "num_input_tokens_seen": 10613440,
      "step": 50300
    },
    {
      "epoch": 5.534103410341034,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.027236019552982323,
      "loss": 0.2308,
      "num_input_tokens_seen": 10614528,
      "step": 50305
    },
    {
      "epoch": 5.534653465346535,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.027235186539197338,
      "loss": 0.2314,
      "num_input_tokens_seen": 10615616,
      "step": 50310
    },
    {
      "epoch": 5.535203520352035,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.027234353412646517,
      "loss": 0.2335,
      "num_input_tokens_seen": 10616704,
      "step": 50315
    },
    {
      "epoch": 5.5357535753575355,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.027233520173337526,
      "loss": 0.2303,
      "num_input_tokens_seen": 10617856,
      "step": 50320
    },
    {
      "epoch": 5.536303630363037,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.027232686821278048,
      "loss": 0.2303,
      "num_input_tokens_seen": 10618912,
      "step": 50325
    },
    {
      "epoch": 5.536853685368537,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027231853356475766,
      "loss": 0.2324,
      "num_input_tokens_seen": 10619936,
      "step": 50330
    },
    {
      "epoch": 5.537403740374037,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.027231019778938364,
      "loss": 0.2314,
      "num_input_tokens_seen": 10620960,
      "step": 50335
    },
    {
      "epoch": 5.537953795379538,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02723018608867352,
      "loss": 0.2298,
      "num_input_tokens_seen": 10622048,
      "step": 50340
    },
    {
      "epoch": 5.538503850385038,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.027229352285688922,
      "loss": 0.2314,
      "num_input_tokens_seen": 10623072,
      "step": 50345
    },
    {
      "epoch": 5.539053905390539,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02722851836999225,
      "loss": 0.2325,
      "num_input_tokens_seen": 10624064,
      "step": 50350
    },
    {
      "epoch": 5.53960396039604,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02722768434159119,
      "loss": 0.2298,
      "num_input_tokens_seen": 10625120,
      "step": 50355
    },
    {
      "epoch": 5.54015401540154,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02722685020049343,
      "loss": 0.2319,
      "num_input_tokens_seen": 10626144,
      "step": 50360
    },
    {
      "epoch": 5.540704070407041,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.027226015946706667,
      "loss": 0.2314,
      "num_input_tokens_seen": 10627168,
      "step": 50365
    },
    {
      "epoch": 5.541254125412541,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.027225181580238577,
      "loss": 0.2314,
      "num_input_tokens_seen": 10628192,
      "step": 50370
    },
    {
      "epoch": 5.541804180418042,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.02722434710109685,
      "loss": 0.2319,
      "num_input_tokens_seen": 10629152,
      "step": 50375
    },
    {
      "epoch": 5.542354235423542,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027223512509289186,
      "loss": 0.2319,
      "num_input_tokens_seen": 10630304,
      "step": 50380
    },
    {
      "epoch": 5.542904290429043,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.027222677804823275,
      "loss": 0.2304,
      "num_input_tokens_seen": 10631360,
      "step": 50385
    },
    {
      "epoch": 5.543454345434544,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.027221842987706805,
      "loss": 0.2309,
      "num_input_tokens_seen": 10632352,
      "step": 50390
    },
    {
      "epoch": 5.544004400440044,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.027221008057947468,
      "loss": 0.2299,
      "num_input_tokens_seen": 10633440,
      "step": 50395
    },
    {
      "epoch": 5.544554455445544,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.02722017301555297,
      "loss": 0.2286,
      "num_input_tokens_seen": 10634560,
      "step": 50400
    },
    {
      "epoch": 5.545104510451045,
      "grad_norm": 0.0152587890625,
      "learning_rate": 0.027219337860530997,
      "loss": 0.2406,
      "num_input_tokens_seen": 10635616,
      "step": 50405
    },
    {
      "epoch": 5.5456545654565454,
      "grad_norm": 0.01422119140625,
      "learning_rate": 0.02721850259288926,
      "loss": 0.229,
      "num_input_tokens_seen": 10636672,
      "step": 50410
    },
    {
      "epoch": 5.5462046204620465,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.027217667212635438,
      "loss": 0.2327,
      "num_input_tokens_seen": 10637792,
      "step": 50415
    },
    {
      "epoch": 5.546754675467547,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.02721683171977724,
      "loss": 0.2322,
      "num_input_tokens_seen": 10638816,
      "step": 50420
    },
    {
      "epoch": 5.547304730473047,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.02721599611432237,
      "loss": 0.2289,
      "num_input_tokens_seen": 10639904,
      "step": 50425
    },
    {
      "epoch": 5.547854785478548,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.027215160396278526,
      "loss": 0.2327,
      "num_input_tokens_seen": 10640992,
      "step": 50430
    },
    {
      "epoch": 5.548404840484048,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.027214324565653408,
      "loss": 0.2315,
      "num_input_tokens_seen": 10642016,
      "step": 50435
    },
    {
      "epoch": 5.548954895489549,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02721348862245472,
      "loss": 0.2335,
      "num_input_tokens_seen": 10643072,
      "step": 50440
    },
    {
      "epoch": 5.5495049504950495,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02721265256669017,
      "loss": 0.2345,
      "num_input_tokens_seen": 10644096,
      "step": 50445
    },
    {
      "epoch": 5.55005500550055,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027211816398367464,
      "loss": 0.233,
      "num_input_tokens_seen": 10645152,
      "step": 50450
    },
    {
      "epoch": 5.550605060506051,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.027210980117494298,
      "loss": 0.2303,
      "num_input_tokens_seen": 10646240,
      "step": 50455
    },
    {
      "epoch": 5.551155115511551,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027210143724078394,
      "loss": 0.2329,
      "num_input_tokens_seen": 10647264,
      "step": 50460
    },
    {
      "epoch": 5.551705170517051,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.027209307218127452,
      "loss": 0.2319,
      "num_input_tokens_seen": 10648384,
      "step": 50465
    },
    {
      "epoch": 5.552255225522552,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.027208470599649186,
      "loss": 0.2298,
      "num_input_tokens_seen": 10649408,
      "step": 50470
    },
    {
      "epoch": 5.552805280528053,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.027207633868651306,
      "loss": 0.2324,
      "num_input_tokens_seen": 10650496,
      "step": 50475
    },
    {
      "epoch": 5.553355335533554,
      "grad_norm": 0.000732421875,
      "learning_rate": 0.02720679702514152,
      "loss": 0.2303,
      "num_input_tokens_seen": 10651552,
      "step": 50480
    },
    {
      "epoch": 5.553905390539054,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027205960069127542,
      "loss": 0.2314,
      "num_input_tokens_seen": 10652512,
      "step": 50485
    },
    {
      "epoch": 5.554455445544555,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.027205123000617087,
      "loss": 0.2314,
      "num_input_tokens_seen": 10653600,
      "step": 50490
    },
    {
      "epoch": 5.555005500550055,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.027204285819617875,
      "loss": 0.2313,
      "num_input_tokens_seen": 10654592,
      "step": 50495
    },
    {
      "epoch": 5.555555555555555,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.02720344852613761,
      "loss": 0.2319,
      "num_input_tokens_seen": 10655648,
      "step": 50500
    },
    {
      "epoch": 5.5561056105610565,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02720261112018402,
      "loss": 0.2283,
      "num_input_tokens_seen": 10656832,
      "step": 50505
    },
    {
      "epoch": 5.556655665566557,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.027201773601764817,
      "loss": 0.2298,
      "num_input_tokens_seen": 10657856,
      "step": 50510
    },
    {
      "epoch": 5.557205720572057,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.027200935970887723,
      "loss": 0.2298,
      "num_input_tokens_seen": 10658880,
      "step": 50515
    },
    {
      "epoch": 5.557755775577558,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.027200098227560456,
      "loss": 0.2325,
      "num_input_tokens_seen": 10659936,
      "step": 50520
    },
    {
      "epoch": 5.558305830583058,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.027199260371790736,
      "loss": 0.2295,
      "num_input_tokens_seen": 10660960,
      "step": 50525
    },
    {
      "epoch": 5.558855885588558,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.02719842240358629,
      "loss": 0.2332,
      "num_input_tokens_seen": 10662016,
      "step": 50530
    },
    {
      "epoch": 5.5594059405940595,
      "grad_norm": 0.0181884765625,
      "learning_rate": 0.027197584322954837,
      "loss": 0.2332,
      "num_input_tokens_seen": 10663136,
      "step": 50535
    },
    {
      "epoch": 5.55995599559956,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.027196746129904102,
      "loss": 0.2274,
      "num_input_tokens_seen": 10664224,
      "step": 50540
    },
    {
      "epoch": 5.560506050605061,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.027195907824441812,
      "loss": 0.2315,
      "num_input_tokens_seen": 10665312,
      "step": 50545
    },
    {
      "epoch": 5.561056105610561,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.02719506940657569,
      "loss": 0.23,
      "num_input_tokens_seen": 10666368,
      "step": 50550
    },
    {
      "epoch": 5.561606160616062,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.027194230876313466,
      "loss": 0.2321,
      "num_input_tokens_seen": 10667456,
      "step": 50555
    },
    {
      "epoch": 5.562156215621562,
      "grad_norm": 0.000667572021484375,
      "learning_rate": 0.027193392233662864,
      "loss": 0.2336,
      "num_input_tokens_seen": 10668544,
      "step": 50560
    },
    {
      "epoch": 5.5627062706270625,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.02719255347863162,
      "loss": 0.2367,
      "num_input_tokens_seen": 10669568,
      "step": 50565
    },
    {
      "epoch": 5.563256325632564,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02719171461122746,
      "loss": 0.2315,
      "num_input_tokens_seen": 10670592,
      "step": 50570
    },
    {
      "epoch": 5.563806380638064,
      "grad_norm": 0.00066375732421875,
      "learning_rate": 0.02719087563145812,
      "loss": 0.233,
      "num_input_tokens_seen": 10671648,
      "step": 50575
    },
    {
      "epoch": 5.564356435643564,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.027190036539331323,
      "loss": 0.2294,
      "num_input_tokens_seen": 10672672,
      "step": 50580
    },
    {
      "epoch": 5.564906490649065,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.027189197334854812,
      "loss": 0.232,
      "num_input_tokens_seen": 10673728,
      "step": 50585
    },
    {
      "epoch": 5.565456545654565,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.027188358018036318,
      "loss": 0.2299,
      "num_input_tokens_seen": 10674784,
      "step": 50590
    },
    {
      "epoch": 5.566006600660066,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027187518588883578,
      "loss": 0.2325,
      "num_input_tokens_seen": 10675776,
      "step": 50595
    },
    {
      "epoch": 5.566556655665567,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02718667904740433,
      "loss": 0.2278,
      "num_input_tokens_seen": 10676896,
      "step": 50600
    },
    {
      "epoch": 5.567106710671067,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0271858393936063,
      "loss": 0.2289,
      "num_input_tokens_seen": 10677952,
      "step": 50605
    },
    {
      "epoch": 5.567656765676568,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.02718499962749724,
      "loss": 0.232,
      "num_input_tokens_seen": 10678976,
      "step": 50610
    },
    {
      "epoch": 5.568206820682068,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02718415974908489,
      "loss": 0.2315,
      "num_input_tokens_seen": 10680128,
      "step": 50615
    },
    {
      "epoch": 5.568756875687569,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.02718331975837698,
      "loss": 0.2335,
      "num_input_tokens_seen": 10681088,
      "step": 50620
    },
    {
      "epoch": 5.569306930693069,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.02718247965538126,
      "loss": 0.2315,
      "num_input_tokens_seen": 10682144,
      "step": 50625
    },
    {
      "epoch": 5.56985698569857,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02718163944010547,
      "loss": 0.2325,
      "num_input_tokens_seen": 10683136,
      "step": 50630
    },
    {
      "epoch": 5.570407040704071,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.027180799112557354,
      "loss": 0.2341,
      "num_input_tokens_seen": 10684192,
      "step": 50635
    },
    {
      "epoch": 5.570957095709571,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.02717995867274466,
      "loss": 0.2336,
      "num_input_tokens_seen": 10685280,
      "step": 50640
    },
    {
      "epoch": 5.571507150715071,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.02717911812067513,
      "loss": 0.2299,
      "num_input_tokens_seen": 10686368,
      "step": 50645
    },
    {
      "epoch": 5.572057205720572,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.027178277456356512,
      "loss": 0.2283,
      "num_input_tokens_seen": 10687360,
      "step": 50650
    },
    {
      "epoch": 5.572607260726072,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.027177436679796555,
      "loss": 0.2315,
      "num_input_tokens_seen": 10688384,
      "step": 50655
    },
    {
      "epoch": 5.5731573157315735,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02717659579100301,
      "loss": 0.2309,
      "num_input_tokens_seen": 10689408,
      "step": 50660
    },
    {
      "epoch": 5.573707370737074,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.02717575478998362,
      "loss": 0.2304,
      "num_input_tokens_seen": 10690400,
      "step": 50665
    },
    {
      "epoch": 5.574257425742574,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.027174913676746144,
      "loss": 0.2293,
      "num_input_tokens_seen": 10691456,
      "step": 50670
    },
    {
      "epoch": 5.574807480748075,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02717407245129833,
      "loss": 0.2304,
      "num_input_tokens_seen": 10692480,
      "step": 50675
    },
    {
      "epoch": 5.575357535753575,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02717323111364793,
      "loss": 0.233,
      "num_input_tokens_seen": 10693568,
      "step": 50680
    },
    {
      "epoch": 5.575907590759076,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.027172389663802702,
      "loss": 0.232,
      "num_input_tokens_seen": 10694560,
      "step": 50685
    },
    {
      "epoch": 5.5764576457645765,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.027171548101770398,
      "loss": 0.2283,
      "num_input_tokens_seen": 10695616,
      "step": 50690
    },
    {
      "epoch": 5.577007700770077,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.027170706427558772,
      "loss": 0.2294,
      "num_input_tokens_seen": 10696672,
      "step": 50695
    },
    {
      "epoch": 5.577557755775578,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027169864641175594,
      "loss": 0.2325,
      "num_input_tokens_seen": 10697728,
      "step": 50700
    },
    {
      "epoch": 5.578107810781078,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02716902274262861,
      "loss": 0.2341,
      "num_input_tokens_seen": 10698816,
      "step": 50705
    },
    {
      "epoch": 5.578657865786578,
      "grad_norm": 0.0008392333984375,
      "learning_rate": 0.027168180731925576,
      "loss": 0.232,
      "num_input_tokens_seen": 10699904,
      "step": 50710
    },
    {
      "epoch": 5.579207920792079,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.027167338609074266,
      "loss": 0.2341,
      "num_input_tokens_seen": 10700960,
      "step": 50715
    },
    {
      "epoch": 5.5797579757975795,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.027166496374082433,
      "loss": 0.2356,
      "num_input_tokens_seen": 10702016,
      "step": 50720
    },
    {
      "epoch": 5.580308030803081,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02716565402695784,
      "loss": 0.2345,
      "num_input_tokens_seen": 10703040,
      "step": 50725
    },
    {
      "epoch": 5.580858085808581,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.027164811567708256,
      "loss": 0.2319,
      "num_input_tokens_seen": 10704128,
      "step": 50730
    },
    {
      "epoch": 5.581408140814082,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.027163968996341433,
      "loss": 0.2304,
      "num_input_tokens_seen": 10705248,
      "step": 50735
    },
    {
      "epoch": 5.581958195819582,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02716312631286515,
      "loss": 0.2351,
      "num_input_tokens_seen": 10706304,
      "step": 50740
    },
    {
      "epoch": 5.582508250825082,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.027162283517287167,
      "loss": 0.2335,
      "num_input_tokens_seen": 10707328,
      "step": 50745
    },
    {
      "epoch": 5.583058305830583,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.027161440609615253,
      "loss": 0.2293,
      "num_input_tokens_seen": 10708384,
      "step": 50750
    },
    {
      "epoch": 5.583608360836084,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.02716059758985718,
      "loss": 0.2309,
      "num_input_tokens_seen": 10709536,
      "step": 50755
    },
    {
      "epoch": 5.584158415841584,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.027159754458020706,
      "loss": 0.2314,
      "num_input_tokens_seen": 10710592,
      "step": 50760
    },
    {
      "epoch": 5.584708470847085,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.027158911214113617,
      "loss": 0.2314,
      "num_input_tokens_seen": 10711616,
      "step": 50765
    },
    {
      "epoch": 5.585258525852585,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02715806785814367,
      "loss": 0.2314,
      "num_input_tokens_seen": 10712640,
      "step": 50770
    },
    {
      "epoch": 5.585808580858086,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.027157224390118653,
      "loss": 0.2308,
      "num_input_tokens_seen": 10713696,
      "step": 50775
    },
    {
      "epoch": 5.586358635863586,
      "grad_norm": 0.0007476806640625,
      "learning_rate": 0.02715638081004633,
      "loss": 0.2319,
      "num_input_tokens_seen": 10714816,
      "step": 50780
    },
    {
      "epoch": 5.586908690869087,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02715553711793448,
      "loss": 0.2303,
      "num_input_tokens_seen": 10715840,
      "step": 50785
    },
    {
      "epoch": 5.587458745874588,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.027154693313790876,
      "loss": 0.2319,
      "num_input_tokens_seen": 10716992,
      "step": 50790
    },
    {
      "epoch": 5.588008800880088,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.027153849397623296,
      "loss": 0.2298,
      "num_input_tokens_seen": 10718048,
      "step": 50795
    },
    {
      "epoch": 5.588558855885589,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.027153005369439518,
      "loss": 0.2309,
      "num_input_tokens_seen": 10719104,
      "step": 50800
    },
    {
      "epoch": 5.589108910891089,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02715216122924732,
      "loss": 0.2325,
      "num_input_tokens_seen": 10720192,
      "step": 50805
    },
    {
      "epoch": 5.589658965896589,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.027151316977054484,
      "loss": 0.2314,
      "num_input_tokens_seen": 10721312,
      "step": 50810
    },
    {
      "epoch": 5.5902090209020905,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02715047261286879,
      "loss": 0.2319,
      "num_input_tokens_seen": 10722400,
      "step": 50815
    },
    {
      "epoch": 5.590759075907591,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.027149628136698023,
      "loss": 0.2319,
      "num_input_tokens_seen": 10723488,
      "step": 50820
    },
    {
      "epoch": 5.591309130913091,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.027148783548549958,
      "loss": 0.2298,
      "num_input_tokens_seen": 10724480,
      "step": 50825
    },
    {
      "epoch": 5.591859185918592,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02714793884843239,
      "loss": 0.2298,
      "num_input_tokens_seen": 10725504,
      "step": 50830
    },
    {
      "epoch": 5.592409240924092,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027147094036353094,
      "loss": 0.2303,
      "num_input_tokens_seen": 10726560,
      "step": 50835
    },
    {
      "epoch": 5.592959295929593,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.027146249112319866,
      "loss": 0.2303,
      "num_input_tokens_seen": 10727552,
      "step": 50840
    },
    {
      "epoch": 5.5935093509350935,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02714540407634048,
      "loss": 0.2335,
      "num_input_tokens_seen": 10728576,
      "step": 50845
    },
    {
      "epoch": 5.594059405940594,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02714455892842274,
      "loss": 0.2319,
      "num_input_tokens_seen": 10729632,
      "step": 50850
    },
    {
      "epoch": 5.594609460946095,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02714371366857443,
      "loss": 0.2314,
      "num_input_tokens_seen": 10730720,
      "step": 50855
    },
    {
      "epoch": 5.595159515951595,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.027142868296803332,
      "loss": 0.2278,
      "num_input_tokens_seen": 10731776,
      "step": 50860
    },
    {
      "epoch": 5.595709570957096,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.027142022813117245,
      "loss": 0.2335,
      "num_input_tokens_seen": 10732832,
      "step": 50865
    },
    {
      "epoch": 5.596259625962596,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.027141177217523963,
      "loss": 0.2346,
      "num_input_tokens_seen": 10733888,
      "step": 50870
    },
    {
      "epoch": 5.5968096809680965,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.027140331510031274,
      "loss": 0.2288,
      "num_input_tokens_seen": 10734976,
      "step": 50875
    },
    {
      "epoch": 5.597359735973598,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.027139485690646975,
      "loss": 0.2293,
      "num_input_tokens_seen": 10735968,
      "step": 50880
    },
    {
      "epoch": 5.597909790979098,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02713863975937886,
      "loss": 0.2325,
      "num_input_tokens_seen": 10737056,
      "step": 50885
    },
    {
      "epoch": 5.598459845984598,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02713779371623473,
      "loss": 0.2325,
      "num_input_tokens_seen": 10738080,
      "step": 50890
    },
    {
      "epoch": 5.599009900990099,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.027136947561222383,
      "loss": 0.2324,
      "num_input_tokens_seen": 10739168,
      "step": 50895
    },
    {
      "epoch": 5.599559955995599,
      "grad_norm": 0.009765625,
      "learning_rate": 0.027136101294349606,
      "loss": 0.2314,
      "num_input_tokens_seen": 10740288,
      "step": 50900
    },
    {
      "epoch": 5.6001100110011,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02713525491562421,
      "loss": 0.2324,
      "num_input_tokens_seen": 10741440,
      "step": 50905
    },
    {
      "epoch": 5.600660066006601,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02713440842505399,
      "loss": 0.2298,
      "num_input_tokens_seen": 10742464,
      "step": 50910
    },
    {
      "epoch": 5.601210121012102,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.027133561822646757,
      "loss": 0.2308,
      "num_input_tokens_seen": 10743488,
      "step": 50915
    },
    {
      "epoch": 5.601760176017602,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.027132715108410303,
      "loss": 0.2319,
      "num_input_tokens_seen": 10744512,
      "step": 50920
    },
    {
      "epoch": 5.602310231023102,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027131868282352437,
      "loss": 0.2303,
      "num_input_tokens_seen": 10745536,
      "step": 50925
    },
    {
      "epoch": 5.602860286028603,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.027131021344480963,
      "loss": 0.2308,
      "num_input_tokens_seen": 10746592,
      "step": 50930
    },
    {
      "epoch": 5.603410341034103,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02713017429480368,
      "loss": 0.2314,
      "num_input_tokens_seen": 10747712,
      "step": 50935
    },
    {
      "epoch": 5.603960396039604,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02712932713332841,
      "loss": 0.2293,
      "num_input_tokens_seen": 10748736,
      "step": 50940
    },
    {
      "epoch": 5.604510451045105,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02712847986006295,
      "loss": 0.2283,
      "num_input_tokens_seen": 10749824,
      "step": 50945
    },
    {
      "epoch": 5.605060506050605,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.027127632475015102,
      "loss": 0.2294,
      "num_input_tokens_seen": 10750944,
      "step": 50950
    },
    {
      "epoch": 5.605610561056105,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.027126784978192698,
      "loss": 0.2258,
      "num_input_tokens_seen": 10752032,
      "step": 50955
    },
    {
      "epoch": 5.606160616061606,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.027125937369603528,
      "loss": 0.2315,
      "num_input_tokens_seen": 10753120,
      "step": 50960
    },
    {
      "epoch": 5.606710671067106,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.027125089649255408,
      "loss": 0.227,
      "num_input_tokens_seen": 10754208,
      "step": 50965
    },
    {
      "epoch": 5.6072607260726075,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.027124241817156163,
      "loss": 0.225,
      "num_input_tokens_seen": 10755296,
      "step": 50970
    },
    {
      "epoch": 5.607810781078108,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.027123393873313598,
      "loss": 0.2272,
      "num_input_tokens_seen": 10756384,
      "step": 50975
    },
    {
      "epoch": 5.608360836083609,
      "grad_norm": 0.01611328125,
      "learning_rate": 0.027122545817735527,
      "loss": 0.235,
      "num_input_tokens_seen": 10757408,
      "step": 50980
    },
    {
      "epoch": 5.608910891089109,
      "grad_norm": 0.01263427734375,
      "learning_rate": 0.02712169765042977,
      "loss": 0.237,
      "num_input_tokens_seen": 10758464,
      "step": 50985
    },
    {
      "epoch": 5.609460946094609,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.02712084937140414,
      "loss": 0.2311,
      "num_input_tokens_seen": 10759584,
      "step": 50990
    },
    {
      "epoch": 5.61001100110011,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02712000098066646,
      "loss": 0.2293,
      "num_input_tokens_seen": 10760704,
      "step": 50995
    },
    {
      "epoch": 5.6105610561056105,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.027119152478224544,
      "loss": 0.235,
      "num_input_tokens_seen": 10761760,
      "step": 51000
    },
    {
      "epoch": 5.611111111111111,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.027118303864086212,
      "loss": 0.2309,
      "num_input_tokens_seen": 10762816,
      "step": 51005
    },
    {
      "epoch": 5.611661166116612,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02711745513825929,
      "loss": 0.2204,
      "num_input_tokens_seen": 10763840,
      "step": 51010
    },
    {
      "epoch": 5.612211221122112,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.027116606300751602,
      "loss": 0.2326,
      "num_input_tokens_seen": 10764896,
      "step": 51015
    },
    {
      "epoch": 5.612761276127613,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.027115757351570968,
      "loss": 0.2316,
      "num_input_tokens_seen": 10765888,
      "step": 51020
    },
    {
      "epoch": 5.613311331133113,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.027114908290725208,
      "loss": 0.2296,
      "num_input_tokens_seen": 10766976,
      "step": 51025
    },
    {
      "epoch": 5.6138613861386135,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.027114059118222154,
      "loss": 0.2336,
      "num_input_tokens_seen": 10768000,
      "step": 51030
    },
    {
      "epoch": 5.614411441144115,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.027113209834069636,
      "loss": 0.2252,
      "num_input_tokens_seen": 10769024,
      "step": 51035
    },
    {
      "epoch": 5.614961496149615,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.027112360438275467,
      "loss": 0.2291,
      "num_input_tokens_seen": 10770112,
      "step": 51040
    },
    {
      "epoch": 5.615511551155116,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.027111510930847488,
      "loss": 0.2373,
      "num_input_tokens_seen": 10771136,
      "step": 51045
    },
    {
      "epoch": 5.616061606160616,
      "grad_norm": 0.01171875,
      "learning_rate": 0.027110661311793523,
      "loss": 0.2304,
      "num_input_tokens_seen": 10772192,
      "step": 51050
    },
    {
      "epoch": 5.616611661166116,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.027109811581121405,
      "loss": 0.236,
      "num_input_tokens_seen": 10773216,
      "step": 51055
    },
    {
      "epoch": 5.617161716171617,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02710896173883896,
      "loss": 0.2263,
      "num_input_tokens_seen": 10774336,
      "step": 51060
    },
    {
      "epoch": 5.617711771177118,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.02710811178495403,
      "loss": 0.2315,
      "num_input_tokens_seen": 10775424,
      "step": 51065
    },
    {
      "epoch": 5.618261826182618,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.027107261719474442,
      "loss": 0.221,
      "num_input_tokens_seen": 10776448,
      "step": 51070
    },
    {
      "epoch": 5.618811881188119,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02710641154240803,
      "loss": 0.2387,
      "num_input_tokens_seen": 10777504,
      "step": 51075
    },
    {
      "epoch": 5.619361936193619,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.027105561253762636,
      "loss": 0.2373,
      "num_input_tokens_seen": 10778560,
      "step": 51080
    },
    {
      "epoch": 5.61991199119912,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.027104710853546092,
      "loss": 0.2371,
      "num_input_tokens_seen": 10779616,
      "step": 51085
    },
    {
      "epoch": 5.62046204620462,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.027103860341766234,
      "loss": 0.2392,
      "num_input_tokens_seen": 10780672,
      "step": 51090
    },
    {
      "epoch": 5.621012101210121,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.027103009718430906,
      "loss": 0.2337,
      "num_input_tokens_seen": 10781728,
      "step": 51095
    },
    {
      "epoch": 5.621562156215622,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.027102158983547946,
      "loss": 0.231,
      "num_input_tokens_seen": 10782784,
      "step": 51100
    },
    {
      "epoch": 5.622112211221122,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02710130813712519,
      "loss": 0.231,
      "num_input_tokens_seen": 10783872,
      "step": 51105
    },
    {
      "epoch": 5.622662266226623,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.027100457179170485,
      "loss": 0.2284,
      "num_input_tokens_seen": 10784960,
      "step": 51110
    },
    {
      "epoch": 5.623212321232123,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.027099606109691674,
      "loss": 0.2331,
      "num_input_tokens_seen": 10786016,
      "step": 51115
    },
    {
      "epoch": 5.623762376237623,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0270987549286966,
      "loss": 0.2315,
      "num_input_tokens_seen": 10787136,
      "step": 51120
    },
    {
      "epoch": 5.6243124312431245,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.027097903636193105,
      "loss": 0.231,
      "num_input_tokens_seen": 10788192,
      "step": 51125
    },
    {
      "epoch": 5.624862486248625,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.02709705223218904,
      "loss": 0.233,
      "num_input_tokens_seen": 10789280,
      "step": 51130
    },
    {
      "epoch": 5.625412541254125,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.027096200716692245,
      "loss": 0.232,
      "num_input_tokens_seen": 10790304,
      "step": 51135
    },
    {
      "epoch": 5.625962596259626,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.027095349089710577,
      "loss": 0.2362,
      "num_input_tokens_seen": 10791360,
      "step": 51140
    },
    {
      "epoch": 5.626512651265126,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027094497351251873,
      "loss": 0.2299,
      "num_input_tokens_seen": 10792480,
      "step": 51145
    },
    {
      "epoch": 5.627062706270627,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.027093645501324,
      "loss": 0.2304,
      "num_input_tokens_seen": 10793568,
      "step": 51150
    },
    {
      "epoch": 5.6276127612761275,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.027092793539934795,
      "loss": 0.231,
      "num_input_tokens_seen": 10794656,
      "step": 51155
    },
    {
      "epoch": 5.628162816281629,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.027091941467092116,
      "loss": 0.2335,
      "num_input_tokens_seen": 10795744,
      "step": 51160
    },
    {
      "epoch": 5.628712871287129,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.027091089282803814,
      "loss": 0.233,
      "num_input_tokens_seen": 10796800,
      "step": 51165
    },
    {
      "epoch": 5.629262926292629,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.027090236987077743,
      "loss": 0.2329,
      "num_input_tokens_seen": 10797824,
      "step": 51170
    },
    {
      "epoch": 5.62981298129813,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.027089384579921758,
      "loss": 0.2293,
      "num_input_tokens_seen": 10798944,
      "step": 51175
    },
    {
      "epoch": 5.63036303630363,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02708853206134372,
      "loss": 0.2325,
      "num_input_tokens_seen": 10799968,
      "step": 51180
    },
    {
      "epoch": 5.6309130913091305,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.02708767943135148,
      "loss": 0.2325,
      "num_input_tokens_seen": 10801088,
      "step": 51185
    },
    {
      "epoch": 5.631463146314632,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.027086826689952898,
      "loss": 0.2309,
      "num_input_tokens_seen": 10802208,
      "step": 51190
    },
    {
      "epoch": 5.632013201320132,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.027085973837155837,
      "loss": 0.2319,
      "num_input_tokens_seen": 10803232,
      "step": 51195
    },
    {
      "epoch": 5.632563256325633,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.027085120872968153,
      "loss": 0.2325,
      "num_input_tokens_seen": 10804288,
      "step": 51200
    },
    {
      "epoch": 5.633113311331133,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.027084267797397713,
      "loss": 0.2304,
      "num_input_tokens_seen": 10805344,
      "step": 51205
    },
    {
      "epoch": 5.633663366336633,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.027083414610452373,
      "loss": 0.2304,
      "num_input_tokens_seen": 10806400,
      "step": 51210
    },
    {
      "epoch": 5.634213421342134,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.027082561312139992,
      "loss": 0.2299,
      "num_input_tokens_seen": 10807520,
      "step": 51215
    },
    {
      "epoch": 5.634763476347635,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.027081707902468444,
      "loss": 0.2309,
      "num_input_tokens_seen": 10808608,
      "step": 51220
    },
    {
      "epoch": 5.635313531353136,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0270808543814456,
      "loss": 0.2293,
      "num_input_tokens_seen": 10809664,
      "step": 51225
    },
    {
      "epoch": 5.635863586358636,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.027080000749079312,
      "loss": 0.232,
      "num_input_tokens_seen": 10810688,
      "step": 51230
    },
    {
      "epoch": 5.636413641364136,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.027079147005377453,
      "loss": 0.2319,
      "num_input_tokens_seen": 10811840,
      "step": 51235
    },
    {
      "epoch": 5.636963696369637,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02707829315034789,
      "loss": 0.2329,
      "num_input_tokens_seen": 10812896,
      "step": 51240
    },
    {
      "epoch": 5.637513751375137,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0270774391839985,
      "loss": 0.2299,
      "num_input_tokens_seen": 10814016,
      "step": 51245
    },
    {
      "epoch": 5.638063806380638,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.027076585106337144,
      "loss": 0.2299,
      "num_input_tokens_seen": 10815072,
      "step": 51250
    },
    {
      "epoch": 5.638613861386139,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.027075730917371702,
      "loss": 0.2314,
      "num_input_tokens_seen": 10816128,
      "step": 51255
    },
    {
      "epoch": 5.639163916391639,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.027074876617110045,
      "loss": 0.2319,
      "num_input_tokens_seen": 10817280,
      "step": 51260
    },
    {
      "epoch": 5.63971397139714,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.027074022205560034,
      "loss": 0.2309,
      "num_input_tokens_seen": 10818304,
      "step": 51265
    },
    {
      "epoch": 5.64026402640264,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027073167682729563,
      "loss": 0.2319,
      "num_input_tokens_seen": 10819328,
      "step": 51270
    },
    {
      "epoch": 5.6408140814081404,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.027072313048626496,
      "loss": 0.2345,
      "num_input_tokens_seen": 10820384,
      "step": 51275
    },
    {
      "epoch": 5.6413641364136415,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.027071458303258713,
      "loss": 0.2262,
      "num_input_tokens_seen": 10821440,
      "step": 51280
    },
    {
      "epoch": 5.641914191419142,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02707060344663409,
      "loss": 0.2325,
      "num_input_tokens_seen": 10822464,
      "step": 51285
    },
    {
      "epoch": 5.642464246424643,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.027069748478760505,
      "loss": 0.234,
      "num_input_tokens_seen": 10823520,
      "step": 51290
    },
    {
      "epoch": 5.643014301430143,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.027068893399645844,
      "loss": 0.233,
      "num_input_tokens_seen": 10824576,
      "step": 51295
    },
    {
      "epoch": 5.643564356435643,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.02706803820929798,
      "loss": 0.233,
      "num_input_tokens_seen": 10825664,
      "step": 51300
    },
    {
      "epoch": 5.644114411441144,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.027067182907724804,
      "loss": 0.2298,
      "num_input_tokens_seen": 10826816,
      "step": 51305
    },
    {
      "epoch": 5.6446644664466445,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.027066327494934192,
      "loss": 0.2314,
      "num_input_tokens_seen": 10827808,
      "step": 51310
    },
    {
      "epoch": 5.645214521452145,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.027065471970934025,
      "loss": 0.2308,
      "num_input_tokens_seen": 10828800,
      "step": 51315
    },
    {
      "epoch": 5.645764576457646,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.027064616335732195,
      "loss": 0.2319,
      "num_input_tokens_seen": 10829920,
      "step": 51320
    },
    {
      "epoch": 5.646314631463146,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02706376058933659,
      "loss": 0.2309,
      "num_input_tokens_seen": 10830976,
      "step": 51325
    },
    {
      "epoch": 5.646864686468647,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.027062904731755082,
      "loss": 0.2308,
      "num_input_tokens_seen": 10832000,
      "step": 51330
    },
    {
      "epoch": 5.647414741474147,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.027062048762995578,
      "loss": 0.2303,
      "num_input_tokens_seen": 10833056,
      "step": 51335
    },
    {
      "epoch": 5.647964796479648,
      "grad_norm": 0.000843048095703125,
      "learning_rate": 0.02706119268306596,
      "loss": 0.2309,
      "num_input_tokens_seen": 10834112,
      "step": 51340
    },
    {
      "epoch": 5.648514851485149,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.027060336491974108,
      "loss": 0.2293,
      "num_input_tokens_seen": 10835136,
      "step": 51345
    },
    {
      "epoch": 5.649064906490649,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.02705948018972793,
      "loss": 0.2335,
      "num_input_tokens_seen": 10836224,
      "step": 51350
    },
    {
      "epoch": 5.64961496149615,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.027058623776335305,
      "loss": 0.2308,
      "num_input_tokens_seen": 10837344,
      "step": 51355
    },
    {
      "epoch": 5.65016501650165,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.027057767251804133,
      "loss": 0.2314,
      "num_input_tokens_seen": 10838368,
      "step": 51360
    },
    {
      "epoch": 5.65071507150715,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.027056910616142304,
      "loss": 0.2308,
      "num_input_tokens_seen": 10839456,
      "step": 51365
    },
    {
      "epoch": 5.6512651265126514,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.027056053869357718,
      "loss": 0.2308,
      "num_input_tokens_seen": 10840512,
      "step": 51370
    },
    {
      "epoch": 5.651815181518152,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.027055197011458264,
      "loss": 0.2313,
      "num_input_tokens_seen": 10841568,
      "step": 51375
    },
    {
      "epoch": 5.652365236523653,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.027054340042451847,
      "loss": 0.2303,
      "num_input_tokens_seen": 10842592,
      "step": 51380
    },
    {
      "epoch": 5.652915291529153,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.027053482962346366,
      "loss": 0.2293,
      "num_input_tokens_seen": 10843680,
      "step": 51385
    },
    {
      "epoch": 5.653465346534653,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02705262577114971,
      "loss": 0.233,
      "num_input_tokens_seen": 10844768,
      "step": 51390
    },
    {
      "epoch": 5.654015401540154,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02705176846886979,
      "loss": 0.233,
      "num_input_tokens_seen": 10845856,
      "step": 51395
    },
    {
      "epoch": 5.6545654565456545,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.0270509110555145,
      "loss": 0.2309,
      "num_input_tokens_seen": 10846912,
      "step": 51400
    },
    {
      "epoch": 5.6551155115511555,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.027050053531091744,
      "loss": 0.2324,
      "num_input_tokens_seen": 10847936,
      "step": 51405
    },
    {
      "epoch": 5.655665566556656,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.027049195895609432,
      "loss": 0.233,
      "num_input_tokens_seen": 10849024,
      "step": 51410
    },
    {
      "epoch": 5.656215621562156,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.02704833814907546,
      "loss": 0.2319,
      "num_input_tokens_seen": 10850112,
      "step": 51415
    },
    {
      "epoch": 5.656765676567657,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.027047480291497738,
      "loss": 0.2293,
      "num_input_tokens_seen": 10851168,
      "step": 51420
    },
    {
      "epoch": 5.657315731573157,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.02704662232288417,
      "loss": 0.233,
      "num_input_tokens_seen": 10852192,
      "step": 51425
    },
    {
      "epoch": 5.6578657865786575,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.027045764243242662,
      "loss": 0.233,
      "num_input_tokens_seen": 10853280,
      "step": 51430
    },
    {
      "epoch": 5.658415841584159,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.027044906052581127,
      "loss": 0.232,
      "num_input_tokens_seen": 10854336,
      "step": 51435
    },
    {
      "epoch": 5.658965896589659,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.027044047750907473,
      "loss": 0.2288,
      "num_input_tokens_seen": 10855328,
      "step": 51440
    },
    {
      "epoch": 5.65951595159516,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.027043189338229613,
      "loss": 0.2335,
      "num_input_tokens_seen": 10856352,
      "step": 51445
    },
    {
      "epoch": 5.66006600660066,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.027042330814555454,
      "loss": 0.232,
      "num_input_tokens_seen": 10857408,
      "step": 51450
    },
    {
      "epoch": 5.66061606160616,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02704147217989291,
      "loss": 0.2288,
      "num_input_tokens_seen": 10858528,
      "step": 51455
    },
    {
      "epoch": 5.661166116611661,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.027040613434249895,
      "loss": 0.2315,
      "num_input_tokens_seen": 10859680,
      "step": 51460
    },
    {
      "epoch": 5.661716171617162,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02703975457763432,
      "loss": 0.2361,
      "num_input_tokens_seen": 10860704,
      "step": 51465
    },
    {
      "epoch": 5.662266226622663,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02703889561005411,
      "loss": 0.2309,
      "num_input_tokens_seen": 10861792,
      "step": 51470
    },
    {
      "epoch": 5.662816281628163,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02703803653151717,
      "loss": 0.2278,
      "num_input_tokens_seen": 10862784,
      "step": 51475
    },
    {
      "epoch": 5.663366336633663,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.027037177342031433,
      "loss": 0.2278,
      "num_input_tokens_seen": 10863872,
      "step": 51480
    },
    {
      "epoch": 5.663916391639164,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.027036318041604796,
      "loss": 0.2304,
      "num_input_tokens_seen": 10864960,
      "step": 51485
    },
    {
      "epoch": 5.664466446644664,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.0270354586302452,
      "loss": 0.2299,
      "num_input_tokens_seen": 10866016,
      "step": 51490
    },
    {
      "epoch": 5.665016501650165,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.027034599107960553,
      "loss": 0.232,
      "num_input_tokens_seen": 10867040,
      "step": 51495
    },
    {
      "epoch": 5.665566556655666,
      "grad_norm": 0.01177978515625,
      "learning_rate": 0.027033739474758785,
      "loss": 0.2351,
      "num_input_tokens_seen": 10868128,
      "step": 51500
    },
    {
      "epoch": 5.666116611661166,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.027032879730647816,
      "loss": 0.2351,
      "num_input_tokens_seen": 10869152,
      "step": 51505
    },
    {
      "epoch": 5.666666666666667,
      "grad_norm": 0.0028076171875,
      "learning_rate": 0.027032019875635565,
      "loss": 0.2288,
      "num_input_tokens_seen": 10870240,
      "step": 51510
    },
    {
      "epoch": 5.667216721672167,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.027031159909729957,
      "loss": 0.2288,
      "num_input_tokens_seen": 10871296,
      "step": 51515
    },
    {
      "epoch": 5.667766776677668,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.027030299832938923,
      "loss": 0.2298,
      "num_input_tokens_seen": 10872320,
      "step": 51520
    },
    {
      "epoch": 5.6683168316831685,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027029439645270388,
      "loss": 0.2288,
      "num_input_tokens_seen": 10873408,
      "step": 51525
    },
    {
      "epoch": 5.668866886688669,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.027028579346732286,
      "loss": 0.2294,
      "num_input_tokens_seen": 10874432,
      "step": 51530
    },
    {
      "epoch": 5.66941694169417,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.027027718937332534,
      "loss": 0.2305,
      "num_input_tokens_seen": 10875488,
      "step": 51535
    },
    {
      "epoch": 5.66996699669967,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02702685841707907,
      "loss": 0.2317,
      "num_input_tokens_seen": 10876512,
      "step": 51540
    },
    {
      "epoch": 5.67051705170517,
      "grad_norm": 0.01171875,
      "learning_rate": 0.027025997785979822,
      "loss": 0.2285,
      "num_input_tokens_seen": 10877536,
      "step": 51545
    },
    {
      "epoch": 5.671067106710671,
      "grad_norm": 0.007476806640625,
      "learning_rate": 0.027025137044042725,
      "loss": 0.2323,
      "num_input_tokens_seen": 10878528,
      "step": 51550
    },
    {
      "epoch": 5.6716171617161715,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.027024276191275707,
      "loss": 0.2271,
      "num_input_tokens_seen": 10879584,
      "step": 51555
    },
    {
      "epoch": 5.672167216721672,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.02702341522768671,
      "loss": 0.2292,
      "num_input_tokens_seen": 10880576,
      "step": 51560
    },
    {
      "epoch": 5.672717271727173,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.02702255415328366,
      "loss": 0.225,
      "num_input_tokens_seen": 10881600,
      "step": 51565
    },
    {
      "epoch": 5.673267326732673,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.027021692968074496,
      "loss": 0.2283,
      "num_input_tokens_seen": 10882688,
      "step": 51570
    },
    {
      "epoch": 5.673817381738174,
      "grad_norm": 0.001953125,
      "learning_rate": 0.02702083167206716,
      "loss": 0.2279,
      "num_input_tokens_seen": 10883712,
      "step": 51575
    },
    {
      "epoch": 5.674367436743674,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.027019970265269585,
      "loss": 0.2358,
      "num_input_tokens_seen": 10884736,
      "step": 51580
    },
    {
      "epoch": 5.674917491749175,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.027019108747689716,
      "loss": 0.2369,
      "num_input_tokens_seen": 10885760,
      "step": 51585
    },
    {
      "epoch": 5.675467546754676,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.02701824711933548,
      "loss": 0.2351,
      "num_input_tokens_seen": 10886816,
      "step": 51590
    },
    {
      "epoch": 5.676017601760176,
      "grad_norm": 0.01348876953125,
      "learning_rate": 0.027017385380214833,
      "loss": 0.2355,
      "num_input_tokens_seen": 10887840,
      "step": 51595
    },
    {
      "epoch": 5.676567656765677,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.02701652353033571,
      "loss": 0.2323,
      "num_input_tokens_seen": 10888864,
      "step": 51600
    },
    {
      "epoch": 5.677117711771177,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02701566156970606,
      "loss": 0.2337,
      "num_input_tokens_seen": 10889888,
      "step": 51605
    },
    {
      "epoch": 5.677667766776677,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.027014799498333818,
      "loss": 0.2316,
      "num_input_tokens_seen": 10890944,
      "step": 51610
    },
    {
      "epoch": 5.678217821782178,
      "grad_norm": 0.0005950927734375,
      "learning_rate": 0.027013937316226935,
      "loss": 0.2269,
      "num_input_tokens_seen": 10892032,
      "step": 51615
    },
    {
      "epoch": 5.678767876787679,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.027013075023393358,
      "loss": 0.2305,
      "num_input_tokens_seen": 10893056,
      "step": 51620
    },
    {
      "epoch": 5.67931793179318,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.027012212619841037,
      "loss": 0.2284,
      "num_input_tokens_seen": 10894176,
      "step": 51625
    },
    {
      "epoch": 5.67986798679868,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02701135010557791,
      "loss": 0.229,
      "num_input_tokens_seen": 10895232,
      "step": 51630
    },
    {
      "epoch": 5.68041804180418,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.027010487480611932,
      "loss": 0.2279,
      "num_input_tokens_seen": 10896320,
      "step": 51635
    },
    {
      "epoch": 5.680968096809681,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.027009624744951062,
      "loss": 0.2275,
      "num_input_tokens_seen": 10897312,
      "step": 51640
    },
    {
      "epoch": 5.681518151815181,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.027008761898603236,
      "loss": 0.2265,
      "num_input_tokens_seen": 10898336,
      "step": 51645
    },
    {
      "epoch": 5.6820682068206825,
      "grad_norm": 0.0072021484375,
      "learning_rate": 0.027007898941576413,
      "loss": 0.232,
      "num_input_tokens_seen": 10899360,
      "step": 51650
    },
    {
      "epoch": 5.682618261826183,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.027007035873878555,
      "loss": 0.2223,
      "num_input_tokens_seen": 10900480,
      "step": 51655
    },
    {
      "epoch": 5.683168316831683,
      "grad_norm": 0.0140380859375,
      "learning_rate": 0.0270061726955176,
      "loss": 0.241,
      "num_input_tokens_seen": 10901536,
      "step": 51660
    },
    {
      "epoch": 5.683718371837184,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.027005309406501518,
      "loss": 0.2305,
      "num_input_tokens_seen": 10902624,
      "step": 51665
    },
    {
      "epoch": 5.684268426842684,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.027004446006838257,
      "loss": 0.2247,
      "num_input_tokens_seen": 10903648,
      "step": 51670
    },
    {
      "epoch": 5.684818481848184,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.027003582496535783,
      "loss": 0.2322,
      "num_input_tokens_seen": 10904672,
      "step": 51675
    },
    {
      "epoch": 5.6853685368536855,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.02700271887560204,
      "loss": 0.2222,
      "num_input_tokens_seen": 10905792,
      "step": 51680
    },
    {
      "epoch": 5.685918591859186,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.027001855144045007,
      "loss": 0.2279,
      "num_input_tokens_seen": 10906816,
      "step": 51685
    },
    {
      "epoch": 5.686468646864687,
      "grad_norm": 0.00811767578125,
      "learning_rate": 0.02700099130187263,
      "loss": 0.2328,
      "num_input_tokens_seen": 10907776,
      "step": 51690
    },
    {
      "epoch": 5.687018701870187,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.02700012734909287,
      "loss": 0.237,
      "num_input_tokens_seen": 10908768,
      "step": 51695
    },
    {
      "epoch": 5.687568756875687,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.026999263285713702,
      "loss": 0.2354,
      "num_input_tokens_seen": 10909760,
      "step": 51700
    },
    {
      "epoch": 5.688118811881188,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.02699839911174308,
      "loss": 0.2232,
      "num_input_tokens_seen": 10910752,
      "step": 51705
    },
    {
      "epoch": 5.6886688668866885,
      "grad_norm": 0.0033416748046875,
      "learning_rate": 0.026997534827188967,
      "loss": 0.2239,
      "num_input_tokens_seen": 10911776,
      "step": 51710
    },
    {
      "epoch": 5.68921892189219,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.026996670432059333,
      "loss": 0.2301,
      "num_input_tokens_seen": 10912768,
      "step": 51715
    },
    {
      "epoch": 5.68976897689769,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.026995805926362144,
      "loss": 0.2365,
      "num_input_tokens_seen": 10913824,
      "step": 51720
    },
    {
      "epoch": 5.69031903190319,
      "grad_norm": 0.007659912109375,
      "learning_rate": 0.026994941310105373,
      "loss": 0.2359,
      "num_input_tokens_seen": 10914880,
      "step": 51725
    },
    {
      "epoch": 5.690869086908691,
      "grad_norm": 0.007659912109375,
      "learning_rate": 0.02699407658329698,
      "loss": 0.2344,
      "num_input_tokens_seen": 10916000,
      "step": 51730
    },
    {
      "epoch": 5.691419141914191,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.026993211745944934,
      "loss": 0.2316,
      "num_input_tokens_seen": 10917056,
      "step": 51735
    },
    {
      "epoch": 5.6919691969196915,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.026992346798057218,
      "loss": 0.232,
      "num_input_tokens_seen": 10918112,
      "step": 51740
    },
    {
      "epoch": 5.692519251925193,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.026991481739641793,
      "loss": 0.2356,
      "num_input_tokens_seen": 10919136,
      "step": 51745
    },
    {
      "epoch": 5.693069306930693,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.026990616570706633,
      "loss": 0.2371,
      "num_input_tokens_seen": 10920192,
      "step": 51750
    },
    {
      "epoch": 5.693619361936194,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.026989751291259717,
      "loss": 0.2255,
      "num_input_tokens_seen": 10921312,
      "step": 51755
    },
    {
      "epoch": 5.694169416941694,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.026988885901309014,
      "loss": 0.2368,
      "num_input_tokens_seen": 10922400,
      "step": 51760
    },
    {
      "epoch": 5.694719471947195,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.026988020400862504,
      "loss": 0.2291,
      "num_input_tokens_seen": 10923520,
      "step": 51765
    },
    {
      "epoch": 5.695269526952695,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.026987154789928164,
      "loss": 0.2264,
      "num_input_tokens_seen": 10924608,
      "step": 51770
    },
    {
      "epoch": 5.695819581958196,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.02698628906851397,
      "loss": 0.2301,
      "num_input_tokens_seen": 10925664,
      "step": 51775
    },
    {
      "epoch": 5.696369636963697,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.026985423236627896,
      "loss": 0.2391,
      "num_input_tokens_seen": 10926720,
      "step": 51780
    },
    {
      "epoch": 5.696919691969197,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.02698455729427793,
      "loss": 0.2317,
      "num_input_tokens_seen": 10927808,
      "step": 51785
    },
    {
      "epoch": 5.697469746974697,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.026983691241472055,
      "loss": 0.2311,
      "num_input_tokens_seen": 10928800,
      "step": 51790
    },
    {
      "epoch": 5.698019801980198,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.026982825078218243,
      "loss": 0.2316,
      "num_input_tokens_seen": 10929856,
      "step": 51795
    },
    {
      "epoch": 5.698569856985698,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.026981958804524482,
      "loss": 0.2301,
      "num_input_tokens_seen": 10930912,
      "step": 51800
    },
    {
      "epoch": 5.6991199119911995,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.02698109242039876,
      "loss": 0.2388,
      "num_input_tokens_seen": 10931936,
      "step": 51805
    },
    {
      "epoch": 5.6996699669967,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.026980225925849054,
      "loss": 0.2319,
      "num_input_tokens_seen": 10932992,
      "step": 51810
    },
    {
      "epoch": 5.7002200220022,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.026979359320883356,
      "loss": 0.2308,
      "num_input_tokens_seen": 10934048,
      "step": 51815
    },
    {
      "epoch": 5.700770077007701,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.026978492605509653,
      "loss": 0.2345,
      "num_input_tokens_seen": 10935072,
      "step": 51820
    },
    {
      "epoch": 5.701320132013201,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.02697762577973593,
      "loss": 0.2303,
      "num_input_tokens_seen": 10936160,
      "step": 51825
    },
    {
      "epoch": 5.701870187018702,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.026976758843570172,
      "loss": 0.2308,
      "num_input_tokens_seen": 10937216,
      "step": 51830
    },
    {
      "epoch": 5.7024202420242025,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.026975891797020386,
      "loss": 0.2323,
      "num_input_tokens_seen": 10938304,
      "step": 51835
    },
    {
      "epoch": 5.702970297029703,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.026975024640094544,
      "loss": 0.2313,
      "num_input_tokens_seen": 10939296,
      "step": 51840
    },
    {
      "epoch": 5.703520352035204,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.02697415737280065,
      "loss": 0.2313,
      "num_input_tokens_seen": 10940384,
      "step": 51845
    },
    {
      "epoch": 5.704070407040704,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.026973289995146688,
      "loss": 0.2318,
      "num_input_tokens_seen": 10941472,
      "step": 51850
    },
    {
      "epoch": 5.704620462046204,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.026972422507140662,
      "loss": 0.2308,
      "num_input_tokens_seen": 10942528,
      "step": 51855
    },
    {
      "epoch": 5.705170517051705,
      "grad_norm": 0.0128173828125,
      "learning_rate": 0.026971554908790565,
      "loss": 0.2323,
      "num_input_tokens_seen": 10943520,
      "step": 51860
    },
    {
      "epoch": 5.7057205720572055,
      "grad_norm": 0.007537841796875,
      "learning_rate": 0.026970687200104388,
      "loss": 0.2318,
      "num_input_tokens_seen": 10944608,
      "step": 51865
    },
    {
      "epoch": 5.706270627062707,
      "grad_norm": 0.01470947265625,
      "learning_rate": 0.02696981938109013,
      "loss": 0.2302,
      "num_input_tokens_seen": 10945696,
      "step": 51870
    },
    {
      "epoch": 5.706820682068207,
      "grad_norm": 0.00872802734375,
      "learning_rate": 0.026968951451755787,
      "loss": 0.2313,
      "num_input_tokens_seen": 10946816,
      "step": 51875
    },
    {
      "epoch": 5.707370737073707,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.026968083412109368,
      "loss": 0.2292,
      "num_input_tokens_seen": 10947872,
      "step": 51880
    },
    {
      "epoch": 5.707920792079208,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.026967215262158863,
      "loss": 0.2319,
      "num_input_tokens_seen": 10948992,
      "step": 51885
    },
    {
      "epoch": 5.708470847084708,
      "grad_norm": 0.00799560546875,
      "learning_rate": 0.026966347001912284,
      "loss": 0.2303,
      "num_input_tokens_seen": 10950048,
      "step": 51890
    },
    {
      "epoch": 5.709020902090209,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.026965478631377618,
      "loss": 0.2329,
      "num_input_tokens_seen": 10951168,
      "step": 51895
    },
    {
      "epoch": 5.70957095709571,
      "grad_norm": 0.000701904296875,
      "learning_rate": 0.026964610150562885,
      "loss": 0.2329,
      "num_input_tokens_seen": 10952224,
      "step": 51900
    },
    {
      "epoch": 5.71012101210121,
      "grad_norm": 0.00885009765625,
      "learning_rate": 0.02696374155947608,
      "loss": 0.2324,
      "num_input_tokens_seen": 10953312,
      "step": 51905
    },
    {
      "epoch": 5.710671067106711,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.026962872858125205,
      "loss": 0.2308,
      "num_input_tokens_seen": 10954368,
      "step": 51910
    },
    {
      "epoch": 5.711221122112211,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.026962004046518273,
      "loss": 0.2339,
      "num_input_tokens_seen": 10955424,
      "step": 51915
    },
    {
      "epoch": 5.711771177117711,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.02696113512466329,
      "loss": 0.2319,
      "num_input_tokens_seen": 10956480,
      "step": 51920
    },
    {
      "epoch": 5.712321232123212,
      "grad_norm": 0.0186767578125,
      "learning_rate": 0.026960266092568262,
      "loss": 0.2303,
      "num_input_tokens_seen": 10957536,
      "step": 51925
    },
    {
      "epoch": 5.712871287128713,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.026959396950241205,
      "loss": 0.2309,
      "num_input_tokens_seen": 10958592,
      "step": 51930
    },
    {
      "epoch": 5.713421342134214,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.026958527697690124,
      "loss": 0.2329,
      "num_input_tokens_seen": 10959616,
      "step": 51935
    },
    {
      "epoch": 5.713971397139714,
      "grad_norm": 0.007781982421875,
      "learning_rate": 0.026957658334923026,
      "loss": 0.2314,
      "num_input_tokens_seen": 10960672,
      "step": 51940
    },
    {
      "epoch": 5.714521452145215,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.026956788861947932,
      "loss": 0.2287,
      "num_input_tokens_seen": 10961696,
      "step": 51945
    },
    {
      "epoch": 5.715071507150715,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.026955919278772857,
      "loss": 0.2277,
      "num_input_tokens_seen": 10962784,
      "step": 51950
    },
    {
      "epoch": 5.715621562156215,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02695504958540581,
      "loss": 0.2319,
      "num_input_tokens_seen": 10963872,
      "step": 51955
    },
    {
      "epoch": 5.7161716171617165,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.026954179781854805,
      "loss": 0.2319,
      "num_input_tokens_seen": 10964928,
      "step": 51960
    },
    {
      "epoch": 5.716721672167217,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.02695330986812786,
      "loss": 0.2345,
      "num_input_tokens_seen": 10965952,
      "step": 51965
    },
    {
      "epoch": 5.717271727172717,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.026952439844232994,
      "loss": 0.2308,
      "num_input_tokens_seen": 10966976,
      "step": 51970
    },
    {
      "epoch": 5.717821782178218,
      "grad_norm": 0.007720947265625,
      "learning_rate": 0.026951569710178227,
      "loss": 0.2329,
      "num_input_tokens_seen": 10968032,
      "step": 51975
    },
    {
      "epoch": 5.718371837183718,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.02695069946597158,
      "loss": 0.2302,
      "num_input_tokens_seen": 10969088,
      "step": 51980
    },
    {
      "epoch": 5.718921892189218,
      "grad_norm": 0.007598876953125,
      "learning_rate": 0.02694982911162106,
      "loss": 0.2303,
      "num_input_tokens_seen": 10970144,
      "step": 51985
    },
    {
      "epoch": 5.7194719471947195,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.02694895864713471,
      "loss": 0.2293,
      "num_input_tokens_seen": 10971296,
      "step": 51990
    },
    {
      "epoch": 5.72002200220022,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.02694808807252054,
      "loss": 0.2277,
      "num_input_tokens_seen": 10972416,
      "step": 51995
    },
    {
      "epoch": 5.720572057205721,
      "grad_norm": 0.01519775390625,
      "learning_rate": 0.02694721738778657,
      "loss": 0.2358,
      "num_input_tokens_seen": 10973472,
      "step": 52000
    },
    {
      "epoch": 5.721122112211221,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.026946346592940835,
      "loss": 0.2335,
      "num_input_tokens_seen": 10974496,
      "step": 52005
    },
    {
      "epoch": 5.721672167216722,
      "grad_norm": 0.0152587890625,
      "learning_rate": 0.02694547568799135,
      "loss": 0.2361,
      "num_input_tokens_seen": 10975616,
      "step": 52010
    },
    {
      "epoch": 5.722222222222222,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.026944604672946158,
      "loss": 0.2303,
      "num_input_tokens_seen": 10976672,
      "step": 52015
    },
    {
      "epoch": 5.7227722772277225,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.026943733547813268,
      "loss": 0.2308,
      "num_input_tokens_seen": 10977696,
      "step": 52020
    },
    {
      "epoch": 5.723322332233224,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.026942862312600718,
      "loss": 0.2324,
      "num_input_tokens_seen": 10978752,
      "step": 52025
    },
    {
      "epoch": 5.723872387238724,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02694199096731654,
      "loss": 0.2313,
      "num_input_tokens_seen": 10979808,
      "step": 52030
    },
    {
      "epoch": 5.724422442244224,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02694111951196876,
      "loss": 0.2303,
      "num_input_tokens_seen": 10980800,
      "step": 52035
    },
    {
      "epoch": 5.724972497249725,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02694024794656541,
      "loss": 0.2298,
      "num_input_tokens_seen": 10981824,
      "step": 52040
    },
    {
      "epoch": 5.725522552255225,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.026939376271114524,
      "loss": 0.2308,
      "num_input_tokens_seen": 10982880,
      "step": 52045
    },
    {
      "epoch": 5.726072607260726,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02693850448562414,
      "loss": 0.2313,
      "num_input_tokens_seen": 10983840,
      "step": 52050
    },
    {
      "epoch": 5.726622662266227,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026937632590102283,
      "loss": 0.2303,
      "num_input_tokens_seen": 10984864,
      "step": 52055
    },
    {
      "epoch": 5.727172717271727,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.026936760584557,
      "loss": 0.2303,
      "num_input_tokens_seen": 10985920,
      "step": 52060
    },
    {
      "epoch": 5.727722772277228,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02693588846899632,
      "loss": 0.2308,
      "num_input_tokens_seen": 10987008,
      "step": 52065
    },
    {
      "epoch": 5.728272827282728,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02693501624342828,
      "loss": 0.2288,
      "num_input_tokens_seen": 10988064,
      "step": 52070
    },
    {
      "epoch": 5.728822882288229,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.026934143907860926,
      "loss": 0.2325,
      "num_input_tokens_seen": 10989056,
      "step": 52075
    },
    {
      "epoch": 5.729372937293729,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.026933271462302295,
      "loss": 0.2289,
      "num_input_tokens_seen": 10990144,
      "step": 52080
    },
    {
      "epoch": 5.72992299229923,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.026932398906760427,
      "loss": 0.2243,
      "num_input_tokens_seen": 10991296,
      "step": 52085
    },
    {
      "epoch": 5.730473047304731,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.026931526241243363,
      "loss": 0.2296,
      "num_input_tokens_seen": 10992320,
      "step": 52090
    },
    {
      "epoch": 5.731023102310231,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.02693065346575915,
      "loss": 0.2319,
      "num_input_tokens_seen": 10993440,
      "step": 52095
    },
    {
      "epoch": 5.731573157315731,
      "grad_norm": 0.0032501220703125,
      "learning_rate": 0.026929780580315824,
      "loss": 0.2381,
      "num_input_tokens_seen": 10994464,
      "step": 52100
    },
    {
      "epoch": 5.732123212321232,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.026928907584921436,
      "loss": 0.2317,
      "num_input_tokens_seen": 10995488,
      "step": 52105
    },
    {
      "epoch": 5.732673267326732,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02692803447958403,
      "loss": 0.2343,
      "num_input_tokens_seen": 10996544,
      "step": 52110
    },
    {
      "epoch": 5.7332233223322335,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.026927161264311654,
      "loss": 0.2363,
      "num_input_tokens_seen": 10997600,
      "step": 52115
    },
    {
      "epoch": 5.733773377337734,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.026926287939112357,
      "loss": 0.2305,
      "num_input_tokens_seen": 10998688,
      "step": 52120
    },
    {
      "epoch": 5.734323432343234,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.026925414503994185,
      "loss": 0.2299,
      "num_input_tokens_seen": 10999712,
      "step": 52125
    },
    {
      "epoch": 5.734873487348735,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.026924540958965196,
      "loss": 0.233,
      "num_input_tokens_seen": 11000704,
      "step": 52130
    },
    {
      "epoch": 5.735423542354235,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02692366730403343,
      "loss": 0.233,
      "num_input_tokens_seen": 11001792,
      "step": 52135
    },
    {
      "epoch": 5.735973597359736,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.026922793539206943,
      "loss": 0.2309,
      "num_input_tokens_seen": 11002816,
      "step": 52140
    },
    {
      "epoch": 5.7365236523652365,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.026921919664493792,
      "loss": 0.232,
      "num_input_tokens_seen": 11003936,
      "step": 52145
    },
    {
      "epoch": 5.737073707370737,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026921045679902025,
      "loss": 0.233,
      "num_input_tokens_seen": 11004992,
      "step": 52150
    },
    {
      "epoch": 5.737623762376238,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.026920171585439707,
      "loss": 0.2314,
      "num_input_tokens_seen": 11006048,
      "step": 52155
    },
    {
      "epoch": 5.738173817381738,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.026919297381114878,
      "loss": 0.2324,
      "num_input_tokens_seen": 11007072,
      "step": 52160
    },
    {
      "epoch": 5.738723872387238,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02691842306693561,
      "loss": 0.2299,
      "num_input_tokens_seen": 11008096,
      "step": 52165
    },
    {
      "epoch": 5.739273927392739,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.026917548642909954,
      "loss": 0.2319,
      "num_input_tokens_seen": 11009120,
      "step": 52170
    },
    {
      "epoch": 5.7398239823982395,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.02691667410904597,
      "loss": 0.2314,
      "num_input_tokens_seen": 11010144,
      "step": 52175
    },
    {
      "epoch": 5.740374037403741,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.026915799465351727,
      "loss": 0.2298,
      "num_input_tokens_seen": 11011232,
      "step": 52180
    },
    {
      "epoch": 5.740924092409241,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.02691492471183527,
      "loss": 0.2319,
      "num_input_tokens_seen": 11012352,
      "step": 52185
    },
    {
      "epoch": 5.741474147414742,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.026914049848504668,
      "loss": 0.2314,
      "num_input_tokens_seen": 11013408,
      "step": 52190
    },
    {
      "epoch": 5.742024202420242,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.026913174875367988,
      "loss": 0.2324,
      "num_input_tokens_seen": 11014528,
      "step": 52195
    },
    {
      "epoch": 5.742574257425742,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.02691229979243329,
      "loss": 0.2324,
      "num_input_tokens_seen": 11015648,
      "step": 52200
    },
    {
      "epoch": 5.743124312431243,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.026911424599708647,
      "loss": 0.2308,
      "num_input_tokens_seen": 11016640,
      "step": 52205
    },
    {
      "epoch": 5.743674367436744,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.026910549297202113,
      "loss": 0.2308,
      "num_input_tokens_seen": 11017696,
      "step": 52210
    },
    {
      "epoch": 5.744224422442244,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.026909673884921763,
      "loss": 0.233,
      "num_input_tokens_seen": 11018720,
      "step": 52215
    },
    {
      "epoch": 5.744774477447745,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02690879836287566,
      "loss": 0.2324,
      "num_input_tokens_seen": 11019712,
      "step": 52220
    },
    {
      "epoch": 5.745324532453245,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.026907922731071884,
      "loss": 0.2308,
      "num_input_tokens_seen": 11020768,
      "step": 52225
    },
    {
      "epoch": 5.745874587458746,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.026907046989518495,
      "loss": 0.2314,
      "num_input_tokens_seen": 11021792,
      "step": 52230
    },
    {
      "epoch": 5.7464246424642464,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.026906171138223568,
      "loss": 0.2298,
      "num_input_tokens_seen": 11022880,
      "step": 52235
    },
    {
      "epoch": 5.746974697469747,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.026905295177195174,
      "loss": 0.2314,
      "num_input_tokens_seen": 11023968,
      "step": 52240
    },
    {
      "epoch": 5.747524752475248,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.026904419106441386,
      "loss": 0.232,
      "num_input_tokens_seen": 11025024,
      "step": 52245
    },
    {
      "epoch": 5.748074807480748,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.026903542925970282,
      "loss": 0.234,
      "num_input_tokens_seen": 11026112,
      "step": 52250
    },
    {
      "epoch": 5.748624862486249,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.026902666635789933,
      "loss": 0.2293,
      "num_input_tokens_seen": 11027168,
      "step": 52255
    },
    {
      "epoch": 5.749174917491749,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.026901790235908416,
      "loss": 0.2351,
      "num_input_tokens_seen": 11028224,
      "step": 52260
    },
    {
      "epoch": 5.7497249724972495,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.026900913726333815,
      "loss": 0.2319,
      "num_input_tokens_seen": 11029248,
      "step": 52265
    },
    {
      "epoch": 5.7502750275027505,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.026900037107074195,
      "loss": 0.2325,
      "num_input_tokens_seen": 11030272,
      "step": 52270
    },
    {
      "epoch": 5.750825082508251,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.026899160378137648,
      "loss": 0.2309,
      "num_input_tokens_seen": 11031360,
      "step": 52275
    },
    {
      "epoch": 5.751375137513751,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.026898283539532247,
      "loss": 0.2335,
      "num_input_tokens_seen": 11032352,
      "step": 52280
    },
    {
      "epoch": 5.751925192519252,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.02689740659126608,
      "loss": 0.2329,
      "num_input_tokens_seen": 11033440,
      "step": 52285
    },
    {
      "epoch": 5.752475247524752,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.026896529533347223,
      "loss": 0.2335,
      "num_input_tokens_seen": 11034496,
      "step": 52290
    },
    {
      "epoch": 5.753025302530253,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.02689565236578376,
      "loss": 0.2314,
      "num_input_tokens_seen": 11035488,
      "step": 52295
    },
    {
      "epoch": 5.7535753575357536,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.02689477508858378,
      "loss": 0.2324,
      "num_input_tokens_seen": 11036608,
      "step": 52300
    },
    {
      "epoch": 5.754125412541254,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026893897701755362,
      "loss": 0.2314,
      "num_input_tokens_seen": 11037664,
      "step": 52305
    },
    {
      "epoch": 5.754675467546755,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.026893020205306598,
      "loss": 0.2293,
      "num_input_tokens_seen": 11038688,
      "step": 52310
    },
    {
      "epoch": 5.755225522552255,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.026892142599245578,
      "loss": 0.2303,
      "num_input_tokens_seen": 11039712,
      "step": 52315
    },
    {
      "epoch": 5.755775577557756,
      "grad_norm": 0.001953125,
      "learning_rate": 0.026891264883580377,
      "loss": 0.2319,
      "num_input_tokens_seen": 11040800,
      "step": 52320
    },
    {
      "epoch": 5.756325632563256,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0268903870583191,
      "loss": 0.2314,
      "num_input_tokens_seen": 11041824,
      "step": 52325
    },
    {
      "epoch": 5.756875687568757,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.02688950912346983,
      "loss": 0.2324,
      "num_input_tokens_seen": 11042912,
      "step": 52330
    },
    {
      "epoch": 5.757425742574258,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.02688863107904066,
      "loss": 0.2319,
      "num_input_tokens_seen": 11043968,
      "step": 52335
    },
    {
      "epoch": 5.757975797579758,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.026887752925039685,
      "loss": 0.2319,
      "num_input_tokens_seen": 11045056,
      "step": 52340
    },
    {
      "epoch": 5.758525852585258,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.02688687466147499,
      "loss": 0.2308,
      "num_input_tokens_seen": 11046048,
      "step": 52345
    },
    {
      "epoch": 5.759075907590759,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02688599628835468,
      "loss": 0.2308,
      "num_input_tokens_seen": 11047072,
      "step": 52350
    },
    {
      "epoch": 5.759625962596259,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.026885117805686837,
      "loss": 0.2314,
      "num_input_tokens_seen": 11048128,
      "step": 52355
    },
    {
      "epoch": 5.7601760176017605,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.026884239213479575,
      "loss": 0.2309,
      "num_input_tokens_seen": 11049184,
      "step": 52360
    },
    {
      "epoch": 5.760726072607261,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.026883360511740978,
      "loss": 0.233,
      "num_input_tokens_seen": 11050304,
      "step": 52365
    },
    {
      "epoch": 5.761276127612762,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.026882481700479154,
      "loss": 0.2324,
      "num_input_tokens_seen": 11051392,
      "step": 52370
    },
    {
      "epoch": 5.761826182618262,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.026881602779702193,
      "loss": 0.2314,
      "num_input_tokens_seen": 11052384,
      "step": 52375
    },
    {
      "epoch": 5.762376237623762,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.026880723749418203,
      "loss": 0.2283,
      "num_input_tokens_seen": 11053440,
      "step": 52380
    },
    {
      "epoch": 5.762926292629263,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02687984460963528,
      "loss": 0.2324,
      "num_input_tokens_seen": 11054528,
      "step": 52385
    },
    {
      "epoch": 5.7634763476347635,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.026878965360361533,
      "loss": 0.2303,
      "num_input_tokens_seen": 11055520,
      "step": 52390
    },
    {
      "epoch": 5.764026402640264,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.026878086001605062,
      "loss": 0.233,
      "num_input_tokens_seen": 11056576,
      "step": 52395
    },
    {
      "epoch": 5.764576457645765,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.026877206533373974,
      "loss": 0.2325,
      "num_input_tokens_seen": 11057728,
      "step": 52400
    },
    {
      "epoch": 5.765126512651265,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02687632695567637,
      "loss": 0.2325,
      "num_input_tokens_seen": 11058848,
      "step": 52405
    },
    {
      "epoch": 5.765676567656766,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.02687544726852036,
      "loss": 0.2283,
      "num_input_tokens_seen": 11059936,
      "step": 52410
    },
    {
      "epoch": 5.766226622662266,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.026874567471914048,
      "loss": 0.2283,
      "num_input_tokens_seen": 11060928,
      "step": 52415
    },
    {
      "epoch": 5.7667766776677665,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.02687368756586555,
      "loss": 0.231,
      "num_input_tokens_seen": 11062016,
      "step": 52420
    },
    {
      "epoch": 5.767326732673268,
      "grad_norm": 0.01171875,
      "learning_rate": 0.026872807550382967,
      "loss": 0.2352,
      "num_input_tokens_seen": 11063072,
      "step": 52425
    },
    {
      "epoch": 5.767876787678768,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02687192742547442,
      "loss": 0.2304,
      "num_input_tokens_seen": 11064096,
      "step": 52430
    },
    {
      "epoch": 5.768426842684269,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02687104719114801,
      "loss": 0.2299,
      "num_input_tokens_seen": 11065216,
      "step": 52435
    },
    {
      "epoch": 5.768976897689769,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.026870166847411857,
      "loss": 0.2279,
      "num_input_tokens_seen": 11066240,
      "step": 52440
    },
    {
      "epoch": 5.769526952695269,
      "grad_norm": 0.01239013671875,
      "learning_rate": 0.026869286394274067,
      "loss": 0.231,
      "num_input_tokens_seen": 11067232,
      "step": 52445
    },
    {
      "epoch": 5.77007700770077,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.026868405831742763,
      "loss": 0.231,
      "num_input_tokens_seen": 11068256,
      "step": 52450
    },
    {
      "epoch": 5.770627062706271,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.02686752515982606,
      "loss": 0.2378,
      "num_input_tokens_seen": 11069344,
      "step": 52455
    },
    {
      "epoch": 5.771177117711771,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.026866644378532072,
      "loss": 0.2299,
      "num_input_tokens_seen": 11070400,
      "step": 52460
    },
    {
      "epoch": 5.771727172717272,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.026865763487868913,
      "loss": 0.2298,
      "num_input_tokens_seen": 11071456,
      "step": 52465
    },
    {
      "epoch": 5.772277227722772,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.026864882487844705,
      "loss": 0.2351,
      "num_input_tokens_seen": 11072480,
      "step": 52470
    },
    {
      "epoch": 5.772827282728273,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02686400137846757,
      "loss": 0.2316,
      "num_input_tokens_seen": 11073472,
      "step": 52475
    },
    {
      "epoch": 5.773377337733773,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02686312015974563,
      "loss": 0.233,
      "num_input_tokens_seen": 11074496,
      "step": 52480
    },
    {
      "epoch": 5.773927392739274,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.026862238831687,
      "loss": 0.2299,
      "num_input_tokens_seen": 11075616,
      "step": 52485
    },
    {
      "epoch": 5.774477447744775,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.02686135739429981,
      "loss": 0.2267,
      "num_input_tokens_seen": 11076768,
      "step": 52490
    },
    {
      "epoch": 5.775027502750275,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026860475847592182,
      "loss": 0.2321,
      "num_input_tokens_seen": 11077888,
      "step": 52495
    },
    {
      "epoch": 5.775577557755776,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026859594191572238,
      "loss": 0.2309,
      "num_input_tokens_seen": 11078912,
      "step": 52500
    },
    {
      "epoch": 5.776127612761276,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.0268587124262481,
      "loss": 0.2294,
      "num_input_tokens_seen": 11079968,
      "step": 52505
    },
    {
      "epoch": 5.776677667766776,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.026857830551627906,
      "loss": 0.2274,
      "num_input_tokens_seen": 11080992,
      "step": 52510
    },
    {
      "epoch": 5.7772277227722775,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026856948567719773,
      "loss": 0.2269,
      "num_input_tokens_seen": 11081984,
      "step": 52515
    },
    {
      "epoch": 5.777777777777778,
      "grad_norm": 0.004302978515625,
      "learning_rate": 0.02685606647453184,
      "loss": 0.2347,
      "num_input_tokens_seen": 11083040,
      "step": 52520
    },
    {
      "epoch": 5.778327832783278,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.026855184272072227,
      "loss": 0.2311,
      "num_input_tokens_seen": 11084128,
      "step": 52525
    },
    {
      "epoch": 5.778877887788779,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.02685430196034907,
      "loss": 0.2326,
      "num_input_tokens_seen": 11085184,
      "step": 52530
    },
    {
      "epoch": 5.779427942794279,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0268534195393705,
      "loss": 0.2326,
      "num_input_tokens_seen": 11086208,
      "step": 52535
    },
    {
      "epoch": 5.77997799779978,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.026852537009144653,
      "loss": 0.2341,
      "num_input_tokens_seen": 11087264,
      "step": 52540
    },
    {
      "epoch": 5.7805280528052805,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.026851654369679654,
      "loss": 0.2289,
      "num_input_tokens_seen": 11088320,
      "step": 52545
    },
    {
      "epoch": 5.781078107810782,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.02685077162098365,
      "loss": 0.2299,
      "num_input_tokens_seen": 11089312,
      "step": 52550
    },
    {
      "epoch": 5.781628162816282,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.026849888763064763,
      "loss": 0.2299,
      "num_input_tokens_seen": 11090432,
      "step": 52555
    },
    {
      "epoch": 5.782178217821782,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.026849005795931146,
      "loss": 0.2325,
      "num_input_tokens_seen": 11091456,
      "step": 52560
    },
    {
      "epoch": 5.782728272827283,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.026848122719590924,
      "loss": 0.2314,
      "num_input_tokens_seen": 11092480,
      "step": 52565
    },
    {
      "epoch": 5.783278327832783,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02684723953405224,
      "loss": 0.2314,
      "num_input_tokens_seen": 11093568,
      "step": 52570
    },
    {
      "epoch": 5.7838283828382835,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.026846356239323233,
      "loss": 0.2336,
      "num_input_tokens_seen": 11094624,
      "step": 52575
    },
    {
      "epoch": 5.784378437843785,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.026845472835412044,
      "loss": 0.2367,
      "num_input_tokens_seen": 11095680,
      "step": 52580
    },
    {
      "epoch": 5.784928492849285,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02684458932232682,
      "loss": 0.2325,
      "num_input_tokens_seen": 11096800,
      "step": 52585
    },
    {
      "epoch": 5.785478547854785,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.0268437057000757,
      "loss": 0.2298,
      "num_input_tokens_seen": 11097824,
      "step": 52590
    },
    {
      "epoch": 5.786028602860286,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.026842821968666823,
      "loss": 0.2293,
      "num_input_tokens_seen": 11098880,
      "step": 52595
    },
    {
      "epoch": 5.786578657865786,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026841938128108343,
      "loss": 0.2324,
      "num_input_tokens_seen": 11099936,
      "step": 52600
    },
    {
      "epoch": 5.787128712871287,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.0268410541784084,
      "loss": 0.2314,
      "num_input_tokens_seen": 11100992,
      "step": 52605
    },
    {
      "epoch": 5.787678767876788,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02684017011957514,
      "loss": 0.2298,
      "num_input_tokens_seen": 11102112,
      "step": 52610
    },
    {
      "epoch": 5.788228822882289,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02683928595161672,
      "loss": 0.2298,
      "num_input_tokens_seen": 11103104,
      "step": 52615
    },
    {
      "epoch": 5.788778877887789,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.026838401674541275,
      "loss": 0.2324,
      "num_input_tokens_seen": 11104128,
      "step": 52620
    },
    {
      "epoch": 5.789328932893289,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.026837517288356972,
      "loss": 0.2309,
      "num_input_tokens_seen": 11105152,
      "step": 52625
    },
    {
      "epoch": 5.78987898789879,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026836632793071944,
      "loss": 0.2309,
      "num_input_tokens_seen": 11106144,
      "step": 52630
    },
    {
      "epoch": 5.79042904290429,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.026835748188694355,
      "loss": 0.233,
      "num_input_tokens_seen": 11107264,
      "step": 52635
    },
    {
      "epoch": 5.790979097909791,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.02683486347523235,
      "loss": 0.2304,
      "num_input_tokens_seen": 11108288,
      "step": 52640
    },
    {
      "epoch": 5.791529152915292,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02683397865269409,
      "loss": 0.2309,
      "num_input_tokens_seen": 11109280,
      "step": 52645
    },
    {
      "epoch": 5.792079207920792,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02683309372108773,
      "loss": 0.2319,
      "num_input_tokens_seen": 11110304,
      "step": 52650
    },
    {
      "epoch": 5.792629262926293,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.026832208680421417,
      "loss": 0.2325,
      "num_input_tokens_seen": 11111392,
      "step": 52655
    },
    {
      "epoch": 5.793179317931793,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02683132353070332,
      "loss": 0.2319,
      "num_input_tokens_seen": 11112448,
      "step": 52660
    },
    {
      "epoch": 5.793729372937293,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.026830438271941585,
      "loss": 0.2304,
      "num_input_tokens_seen": 11113504,
      "step": 52665
    },
    {
      "epoch": 5.7942794279427945,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.026829552904144384,
      "loss": 0.2309,
      "num_input_tokens_seen": 11114560,
      "step": 52670
    },
    {
      "epoch": 5.794829482948295,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026828667427319867,
      "loss": 0.2335,
      "num_input_tokens_seen": 11115584,
      "step": 52675
    },
    {
      "epoch": 5.795379537953796,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.0268277818414762,
      "loss": 0.2304,
      "num_input_tokens_seen": 11116544,
      "step": 52680
    },
    {
      "epoch": 5.795929592959296,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02682689614662154,
      "loss": 0.2303,
      "num_input_tokens_seen": 11117568,
      "step": 52685
    },
    {
      "epoch": 5.796479647964796,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.026826010342764058,
      "loss": 0.234,
      "num_input_tokens_seen": 11118592,
      "step": 52690
    },
    {
      "epoch": 5.797029702970297,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02682512442991191,
      "loss": 0.2319,
      "num_input_tokens_seen": 11119584,
      "step": 52695
    },
    {
      "epoch": 5.7975797579757975,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.026824238408073266,
      "loss": 0.2309,
      "num_input_tokens_seen": 11120704,
      "step": 52700
    },
    {
      "epoch": 5.798129812981298,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.02682335227725629,
      "loss": 0.233,
      "num_input_tokens_seen": 11121728,
      "step": 52705
    },
    {
      "epoch": 5.798679867986799,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.02682246603746915,
      "loss": 0.2309,
      "num_input_tokens_seen": 11122784,
      "step": 52710
    },
    {
      "epoch": 5.799229922992299,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.026821579688720013,
      "loss": 0.2324,
      "num_input_tokens_seen": 11123872,
      "step": 52715
    },
    {
      "epoch": 5.7997799779978,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026820693231017045,
      "loss": 0.2319,
      "num_input_tokens_seen": 11124960,
      "step": 52720
    },
    {
      "epoch": 5.8003300330033,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.026819806664368426,
      "loss": 0.2309,
      "num_input_tokens_seen": 11126048,
      "step": 52725
    },
    {
      "epoch": 5.8008800880088005,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.026818919988782317,
      "loss": 0.2345,
      "num_input_tokens_seen": 11127104,
      "step": 52730
    },
    {
      "epoch": 5.801430143014302,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026818033204266893,
      "loss": 0.2319,
      "num_input_tokens_seen": 11128160,
      "step": 52735
    },
    {
      "epoch": 5.801980198019802,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.026817146310830327,
      "loss": 0.2293,
      "num_input_tokens_seen": 11129216,
      "step": 52740
    },
    {
      "epoch": 5.802530253025303,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0268162593084808,
      "loss": 0.2247,
      "num_input_tokens_seen": 11130336,
      "step": 52745
    },
    {
      "epoch": 5.803080308030803,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.026815372197226475,
      "loss": 0.2305,
      "num_input_tokens_seen": 11131424,
      "step": 52750
    },
    {
      "epoch": 5.803630363036303,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.026814484977075535,
      "loss": 0.2253,
      "num_input_tokens_seen": 11132448,
      "step": 52755
    },
    {
      "epoch": 5.804180418041804,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.026813597648036158,
      "loss": 0.2266,
      "num_input_tokens_seen": 11133504,
      "step": 52760
    },
    {
      "epoch": 5.804730473047305,
      "grad_norm": 0.007171630859375,
      "learning_rate": 0.026812710210116518,
      "loss": 0.2227,
      "num_input_tokens_seen": 11134592,
      "step": 52765
    },
    {
      "epoch": 5.805280528052805,
      "grad_norm": 0.0084228515625,
      "learning_rate": 0.026811822663324797,
      "loss": 0.2308,
      "num_input_tokens_seen": 11135648,
      "step": 52770
    },
    {
      "epoch": 5.805830583058306,
      "grad_norm": 0.0086669921875,
      "learning_rate": 0.026810935007669172,
      "loss": 0.2476,
      "num_input_tokens_seen": 11136736,
      "step": 52775
    },
    {
      "epoch": 5.806380638063806,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02681004724315783,
      "loss": 0.2368,
      "num_input_tokens_seen": 11137792,
      "step": 52780
    },
    {
      "epoch": 5.806930693069307,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.02680915936979895,
      "loss": 0.2346,
      "num_input_tokens_seen": 11138784,
      "step": 52785
    },
    {
      "epoch": 5.807480748074807,
      "grad_norm": 0.00421142578125,
      "learning_rate": 0.026808271387600714,
      "loss": 0.233,
      "num_input_tokens_seen": 11139808,
      "step": 52790
    },
    {
      "epoch": 5.8080308030803085,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.026807383296571306,
      "loss": 0.2328,
      "num_input_tokens_seen": 11140864,
      "step": 52795
    },
    {
      "epoch": 5.808580858085809,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.026806495096718914,
      "loss": 0.2297,
      "num_input_tokens_seen": 11141952,
      "step": 52800
    },
    {
      "epoch": 5.809130913091309,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.026805606788051717,
      "loss": 0.2322,
      "num_input_tokens_seen": 11143072,
      "step": 52805
    },
    {
      "epoch": 5.80968096809681,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02680471837057791,
      "loss": 0.2281,
      "num_input_tokens_seen": 11144128,
      "step": 52810
    },
    {
      "epoch": 5.81023102310231,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.02680382984430568,
      "loss": 0.2333,
      "num_input_tokens_seen": 11145184,
      "step": 52815
    },
    {
      "epoch": 5.81078107810781,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.026802941209243217,
      "loss": 0.2337,
      "num_input_tokens_seen": 11146272,
      "step": 52820
    },
    {
      "epoch": 5.8113311331133115,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.026802052465398702,
      "loss": 0.239,
      "num_input_tokens_seen": 11147360,
      "step": 52825
    },
    {
      "epoch": 5.811881188118812,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026801163612780333,
      "loss": 0.2264,
      "num_input_tokens_seen": 11148416,
      "step": 52830
    },
    {
      "epoch": 5.812431243124313,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.026800274651396304,
      "loss": 0.23,
      "num_input_tokens_seen": 11149536,
      "step": 52835
    },
    {
      "epoch": 5.812981298129813,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.026799385581254803,
      "loss": 0.2311,
      "num_input_tokens_seen": 11150592,
      "step": 52840
    },
    {
      "epoch": 5.813531353135313,
      "grad_norm": 0.0005340576171875,
      "learning_rate": 0.026798496402364032,
      "loss": 0.2299,
      "num_input_tokens_seen": 11151584,
      "step": 52845
    },
    {
      "epoch": 5.814081408140814,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02679760711473218,
      "loss": 0.2331,
      "num_input_tokens_seen": 11152576,
      "step": 52850
    },
    {
      "epoch": 5.8146314631463145,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02679671771836744,
      "loss": 0.2336,
      "num_input_tokens_seen": 11153664,
      "step": 52855
    },
    {
      "epoch": 5.815181518151816,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.026795828213278017,
      "loss": 0.2337,
      "num_input_tokens_seen": 11154688,
      "step": 52860
    },
    {
      "epoch": 5.815731573157316,
      "grad_norm": 0.000720977783203125,
      "learning_rate": 0.026794938599472105,
      "loss": 0.2315,
      "num_input_tokens_seen": 11155744,
      "step": 52865
    },
    {
      "epoch": 5.816281628162816,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.026794048876957903,
      "loss": 0.2305,
      "num_input_tokens_seen": 11156800,
      "step": 52870
    },
    {
      "epoch": 5.816831683168317,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.026793159045743614,
      "loss": 0.231,
      "num_input_tokens_seen": 11157792,
      "step": 52875
    },
    {
      "epoch": 5.817381738173817,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.026792269105837433,
      "loss": 0.232,
      "num_input_tokens_seen": 11158848,
      "step": 52880
    },
    {
      "epoch": 5.8179317931793175,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02679137905724757,
      "loss": 0.2351,
      "num_input_tokens_seen": 11159872,
      "step": 52885
    },
    {
      "epoch": 5.818481848184819,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.026790488899982225,
      "loss": 0.233,
      "num_input_tokens_seen": 11160928,
      "step": 52890
    },
    {
      "epoch": 5.819031903190319,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.0267895986340496,
      "loss": 0.2304,
      "num_input_tokens_seen": 11161984,
      "step": 52895
    },
    {
      "epoch": 5.81958195819582,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.026788708259457894,
      "loss": 0.2341,
      "num_input_tokens_seen": 11163104,
      "step": 52900
    },
    {
      "epoch": 5.82013201320132,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02678781777621533,
      "loss": 0.2314,
      "num_input_tokens_seen": 11164224,
      "step": 52905
    },
    {
      "epoch": 5.82068206820682,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.026786927184330105,
      "loss": 0.2325,
      "num_input_tokens_seen": 11165280,
      "step": 52910
    },
    {
      "epoch": 5.821232123212321,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.026786036483810425,
      "loss": 0.2314,
      "num_input_tokens_seen": 11166272,
      "step": 52915
    },
    {
      "epoch": 5.821782178217822,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.026785145674664505,
      "loss": 0.2309,
      "num_input_tokens_seen": 11167296,
      "step": 52920
    },
    {
      "epoch": 5.822332233223323,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02678425475690055,
      "loss": 0.2319,
      "num_input_tokens_seen": 11168384,
      "step": 52925
    },
    {
      "epoch": 5.822882288228823,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02678336373052678,
      "loss": 0.2325,
      "num_input_tokens_seen": 11169440,
      "step": 52930
    },
    {
      "epoch": 5.823432343234323,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02678247259555139,
      "loss": 0.2283,
      "num_input_tokens_seen": 11170464,
      "step": 52935
    },
    {
      "epoch": 5.823982398239824,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.026781581351982612,
      "loss": 0.2303,
      "num_input_tokens_seen": 11171456,
      "step": 52940
    },
    {
      "epoch": 5.824532453245324,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.026780689999828647,
      "loss": 0.2314,
      "num_input_tokens_seen": 11172608,
      "step": 52945
    },
    {
      "epoch": 5.825082508250825,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.02677979853909772,
      "loss": 0.2319,
      "num_input_tokens_seen": 11173696,
      "step": 52950
    },
    {
      "epoch": 5.825632563256326,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.02677890696979804,
      "loss": 0.2304,
      "num_input_tokens_seen": 11174816,
      "step": 52955
    },
    {
      "epoch": 5.826182618261826,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.026778015291937823,
      "loss": 0.2314,
      "num_input_tokens_seen": 11175904,
      "step": 52960
    },
    {
      "epoch": 5.826732673267327,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026777123505525295,
      "loss": 0.2304,
      "num_input_tokens_seen": 11176864,
      "step": 52965
    },
    {
      "epoch": 5.827282728272827,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.026776231610568673,
      "loss": 0.2319,
      "num_input_tokens_seen": 11177952,
      "step": 52970
    },
    {
      "epoch": 5.827832783278328,
      "grad_norm": 0.000690460205078125,
      "learning_rate": 0.02677533960707617,
      "loss": 0.2304,
      "num_input_tokens_seen": 11178976,
      "step": 52975
    },
    {
      "epoch": 5.8283828382838285,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026774447495056013,
      "loss": 0.2329,
      "num_input_tokens_seen": 11180032,
      "step": 52980
    },
    {
      "epoch": 5.828932893289329,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02677355527451642,
      "loss": 0.2319,
      "num_input_tokens_seen": 11181088,
      "step": 52985
    },
    {
      "epoch": 5.82948294829483,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.026772662945465625,
      "loss": 0.2314,
      "num_input_tokens_seen": 11182112,
      "step": 52990
    },
    {
      "epoch": 5.83003300330033,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.02677177050791184,
      "loss": 0.2314,
      "num_input_tokens_seen": 11183168,
      "step": 52995
    },
    {
      "epoch": 5.83058305830583,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0267708779618633,
      "loss": 0.2314,
      "num_input_tokens_seen": 11184192,
      "step": 53000
    },
    {
      "epoch": 5.831133113311331,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.02676998530732822,
      "loss": 0.2304,
      "num_input_tokens_seen": 11185216,
      "step": 53005
    },
    {
      "epoch": 5.8316831683168315,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02676909254431484,
      "loss": 0.2334,
      "num_input_tokens_seen": 11186272,
      "step": 53010
    },
    {
      "epoch": 5.832233223322332,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.026768199672831373,
      "loss": 0.2309,
      "num_input_tokens_seen": 11187392,
      "step": 53015
    },
    {
      "epoch": 5.832783278327833,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.026767306692886062,
      "loss": 0.2319,
      "num_input_tokens_seen": 11188448,
      "step": 53020
    },
    {
      "epoch": 5.833333333333333,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.026766413604487133,
      "loss": 0.234,
      "num_input_tokens_seen": 11189472,
      "step": 53025
    },
    {
      "epoch": 5.833883388338834,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.026765520407642813,
      "loss": 0.2309,
      "num_input_tokens_seen": 11190560,
      "step": 53030
    },
    {
      "epoch": 5.834433443344334,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.026764627102361334,
      "loss": 0.2319,
      "num_input_tokens_seen": 11191616,
      "step": 53035
    },
    {
      "epoch": 5.834983498349835,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.026763733688650936,
      "loss": 0.234,
      "num_input_tokens_seen": 11192640,
      "step": 53040
    },
    {
      "epoch": 5.835533553355336,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.026762840166519853,
      "loss": 0.2319,
      "num_input_tokens_seen": 11193696,
      "step": 53045
    },
    {
      "epoch": 5.836083608360836,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.026761946535976307,
      "loss": 0.2319,
      "num_input_tokens_seen": 11194720,
      "step": 53050
    },
    {
      "epoch": 5.836633663366337,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02676105279702855,
      "loss": 0.2309,
      "num_input_tokens_seen": 11195808,
      "step": 53055
    },
    {
      "epoch": 5.837183718371837,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.026760158949684816,
      "loss": 0.2303,
      "num_input_tokens_seen": 11196896,
      "step": 53060
    },
    {
      "epoch": 5.837733773377337,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.026759264993953334,
      "loss": 0.2309,
      "num_input_tokens_seen": 11197920,
      "step": 53065
    },
    {
      "epoch": 5.838283828382838,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026758370929842352,
      "loss": 0.2314,
      "num_input_tokens_seen": 11198976,
      "step": 53070
    },
    {
      "epoch": 5.838833883388339,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026757476757360107,
      "loss": 0.2303,
      "num_input_tokens_seen": 11199968,
      "step": 53075
    },
    {
      "epoch": 5.83938393839384,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02675658247651484,
      "loss": 0.2329,
      "num_input_tokens_seen": 11201056,
      "step": 53080
    },
    {
      "epoch": 5.83993399339934,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.026755688087314797,
      "loss": 0.233,
      "num_input_tokens_seen": 11202112,
      "step": 53085
    },
    {
      "epoch": 5.84048404840484,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02675479358976821,
      "loss": 0.2314,
      "num_input_tokens_seen": 11203168,
      "step": 53090
    },
    {
      "epoch": 5.841034103410341,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.026753898983883338,
      "loss": 0.2319,
      "num_input_tokens_seen": 11204224,
      "step": 53095
    },
    {
      "epoch": 5.841584158415841,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.026753004269668416,
      "loss": 0.2324,
      "num_input_tokens_seen": 11205344,
      "step": 53100
    },
    {
      "epoch": 5.8421342134213425,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.026752109447131696,
      "loss": 0.2324,
      "num_input_tokens_seen": 11206336,
      "step": 53105
    },
    {
      "epoch": 5.842684268426843,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.02675121451628142,
      "loss": 0.2324,
      "num_input_tokens_seen": 11207392,
      "step": 53110
    },
    {
      "epoch": 5.843234323432343,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.026750319477125837,
      "loss": 0.2325,
      "num_input_tokens_seen": 11208416,
      "step": 53115
    },
    {
      "epoch": 5.843784378437844,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.026749424329673203,
      "loss": 0.2309,
      "num_input_tokens_seen": 11209472,
      "step": 53120
    },
    {
      "epoch": 5.844334433443344,
      "grad_norm": 0.000614166259765625,
      "learning_rate": 0.026748529073931757,
      "loss": 0.2335,
      "num_input_tokens_seen": 11210496,
      "step": 53125
    },
    {
      "epoch": 5.8448844884488445,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02674763370990976,
      "loss": 0.2335,
      "num_input_tokens_seen": 11211552,
      "step": 53130
    },
    {
      "epoch": 5.8454345434543455,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.026746738237615458,
      "loss": 0.2319,
      "num_input_tokens_seen": 11212608,
      "step": 53135
    },
    {
      "epoch": 5.845984598459846,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.026745842657057105,
      "loss": 0.2303,
      "num_input_tokens_seen": 11213696,
      "step": 53140
    },
    {
      "epoch": 5.846534653465347,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02674494696824296,
      "loss": 0.2309,
      "num_input_tokens_seen": 11214688,
      "step": 53145
    },
    {
      "epoch": 5.847084708470847,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.026744051171181275,
      "loss": 0.2314,
      "num_input_tokens_seen": 11215776,
      "step": 53150
    },
    {
      "epoch": 5.847634763476347,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.026743155265880305,
      "loss": 0.2303,
      "num_input_tokens_seen": 11216832,
      "step": 53155
    },
    {
      "epoch": 5.848184818481848,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.026742259252348303,
      "loss": 0.2288,
      "num_input_tokens_seen": 11217920,
      "step": 53160
    },
    {
      "epoch": 5.8487348734873486,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.02674136313059354,
      "loss": 0.232,
      "num_input_tokens_seen": 11218912,
      "step": 53165
    },
    {
      "epoch": 5.84928492849285,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02674046690062426,
      "loss": 0.2304,
      "num_input_tokens_seen": 11219936,
      "step": 53170
    },
    {
      "epoch": 5.84983498349835,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.026739570562448734,
      "loss": 0.2289,
      "num_input_tokens_seen": 11220960,
      "step": 53175
    },
    {
      "epoch": 5.85038503850385,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02673867411607522,
      "loss": 0.2315,
      "num_input_tokens_seen": 11221984,
      "step": 53180
    },
    {
      "epoch": 5.850935093509351,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.026737777561511977,
      "loss": 0.2315,
      "num_input_tokens_seen": 11223072,
      "step": 53185
    },
    {
      "epoch": 5.851485148514851,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02673688089876727,
      "loss": 0.2341,
      "num_input_tokens_seen": 11224096,
      "step": 53190
    },
    {
      "epoch": 5.852035203520352,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.026735984127849364,
      "loss": 0.2319,
      "num_input_tokens_seen": 11225152,
      "step": 53195
    },
    {
      "epoch": 5.852585258525853,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026735087248766522,
      "loss": 0.2288,
      "num_input_tokens_seen": 11226304,
      "step": 53200
    },
    {
      "epoch": 5.853135313531353,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02673419026152702,
      "loss": 0.2299,
      "num_input_tokens_seen": 11227392,
      "step": 53205
    },
    {
      "epoch": 5.853685368536854,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02673329316613911,
      "loss": 0.233,
      "num_input_tokens_seen": 11228448,
      "step": 53210
    },
    {
      "epoch": 5.854235423542354,
      "grad_norm": 0.0006866455078125,
      "learning_rate": 0.02673239596261107,
      "loss": 0.2325,
      "num_input_tokens_seen": 11229536,
      "step": 53215
    },
    {
      "epoch": 5.854785478547855,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.026731498650951164,
      "loss": 0.233,
      "num_input_tokens_seen": 11230592,
      "step": 53220
    },
    {
      "epoch": 5.8553355335533555,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.026730601231167665,
      "loss": 0.2304,
      "num_input_tokens_seen": 11231680,
      "step": 53225
    },
    {
      "epoch": 5.855885588558856,
      "grad_norm": 0.004547119140625,
      "learning_rate": 0.026729703703268843,
      "loss": 0.2299,
      "num_input_tokens_seen": 11232736,
      "step": 53230
    },
    {
      "epoch": 5.856435643564357,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.02672880606726297,
      "loss": 0.2325,
      "num_input_tokens_seen": 11233824,
      "step": 53235
    },
    {
      "epoch": 5.856985698569857,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.02672790832315832,
      "loss": 0.2341,
      "num_input_tokens_seen": 11234912,
      "step": 53240
    },
    {
      "epoch": 5.857535753575357,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.02672701047096317,
      "loss": 0.2273,
      "num_input_tokens_seen": 11235936,
      "step": 53245
    },
    {
      "epoch": 5.858085808580858,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.02672611251068579,
      "loss": 0.2289,
      "num_input_tokens_seen": 11236992,
      "step": 53250
    },
    {
      "epoch": 5.8586358635863585,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.02672521444233445,
      "loss": 0.2315,
      "num_input_tokens_seen": 11238048,
      "step": 53255
    },
    {
      "epoch": 5.8591859185918596,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026724316265917442,
      "loss": 0.2325,
      "num_input_tokens_seen": 11239040,
      "step": 53260
    },
    {
      "epoch": 5.85973597359736,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.026723417981443037,
      "loss": 0.2304,
      "num_input_tokens_seen": 11240032,
      "step": 53265
    },
    {
      "epoch": 5.86028602860286,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.026722519588919514,
      "loss": 0.231,
      "num_input_tokens_seen": 11241088,
      "step": 53270
    },
    {
      "epoch": 5.860836083608361,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.026721621088355154,
      "loss": 0.2325,
      "num_input_tokens_seen": 11242144,
      "step": 53275
    },
    {
      "epoch": 5.861386138613861,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.026720722479758232,
      "loss": 0.2325,
      "num_input_tokens_seen": 11243200,
      "step": 53280
    },
    {
      "epoch": 5.861936193619362,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.026719823763137036,
      "loss": 0.2341,
      "num_input_tokens_seen": 11244320,
      "step": 53285
    },
    {
      "epoch": 5.862486248624863,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.026718924938499852,
      "loss": 0.2304,
      "num_input_tokens_seen": 11245376,
      "step": 53290
    },
    {
      "epoch": 5.863036303630363,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026718026005854958,
      "loss": 0.2304,
      "num_input_tokens_seen": 11246400,
      "step": 53295
    },
    {
      "epoch": 5.863586358635864,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.026717126965210638,
      "loss": 0.2268,
      "num_input_tokens_seen": 11247488,
      "step": 53300
    },
    {
      "epoch": 5.864136413641364,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.026716227816575185,
      "loss": 0.232,
      "num_input_tokens_seen": 11248576,
      "step": 53305
    },
    {
      "epoch": 5.864686468646864,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.026715328559956883,
      "loss": 0.231,
      "num_input_tokens_seen": 11249632,
      "step": 53310
    },
    {
      "epoch": 5.865236523652365,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.026714429195364018,
      "loss": 0.232,
      "num_input_tokens_seen": 11250656,
      "step": 53315
    },
    {
      "epoch": 5.865786578657866,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.02671352972280488,
      "loss": 0.232,
      "num_input_tokens_seen": 11251680,
      "step": 53320
    },
    {
      "epoch": 5.866336633663367,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.026712630142287763,
      "loss": 0.2289,
      "num_input_tokens_seen": 11252704,
      "step": 53325
    },
    {
      "epoch": 5.866886688668867,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.02671173045382095,
      "loss": 0.2315,
      "num_input_tokens_seen": 11253792,
      "step": 53330
    },
    {
      "epoch": 5.867436743674367,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.026710830657412737,
      "loss": 0.2341,
      "num_input_tokens_seen": 11254880,
      "step": 53335
    },
    {
      "epoch": 5.867986798679868,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026709930753071418,
      "loss": 0.233,
      "num_input_tokens_seen": 11255968,
      "step": 53340
    },
    {
      "epoch": 5.868536853685368,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.026709030740805283,
      "loss": 0.2304,
      "num_input_tokens_seen": 11256992,
      "step": 53345
    },
    {
      "epoch": 5.8690869086908695,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.026708130620622633,
      "loss": 0.2315,
      "num_input_tokens_seen": 11258016,
      "step": 53350
    },
    {
      "epoch": 5.86963696369637,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.026707230392531766,
      "loss": 0.2356,
      "num_input_tokens_seen": 11259072,
      "step": 53355
    },
    {
      "epoch": 5.87018701870187,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.026706330056540967,
      "loss": 0.2289,
      "num_input_tokens_seen": 11260096,
      "step": 53360
    },
    {
      "epoch": 5.870737073707371,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.026705429612658545,
      "loss": 0.2268,
      "num_input_tokens_seen": 11261056,
      "step": 53365
    },
    {
      "epoch": 5.871287128712871,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.026704529060892795,
      "loss": 0.231,
      "num_input_tokens_seen": 11262144,
      "step": 53370
    },
    {
      "epoch": 5.871837183718371,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.026703628401252018,
      "loss": 0.2279,
      "num_input_tokens_seen": 11263232,
      "step": 53375
    },
    {
      "epoch": 5.8723872387238725,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.02670272763374451,
      "loss": 0.232,
      "num_input_tokens_seen": 11264256,
      "step": 53380
    },
    {
      "epoch": 5.872937293729373,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.026701826758378577,
      "loss": 0.2325,
      "num_input_tokens_seen": 11265344,
      "step": 53385
    },
    {
      "epoch": 5.873487348734874,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.026700925775162525,
      "loss": 0.2315,
      "num_input_tokens_seen": 11266336,
      "step": 53390
    },
    {
      "epoch": 5.874037403740374,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.026700024684104654,
      "loss": 0.2315,
      "num_input_tokens_seen": 11267392,
      "step": 53395
    },
    {
      "epoch": 5.874587458745875,
      "grad_norm": 0.003173828125,
      "learning_rate": 0.026699123485213267,
      "loss": 0.2309,
      "num_input_tokens_seen": 11268448,
      "step": 53400
    },
    {
      "epoch": 5.875137513751375,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.026698222178496676,
      "loss": 0.2289,
      "num_input_tokens_seen": 11269536,
      "step": 53405
    },
    {
      "epoch": 5.8756875687568755,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.026697320763963184,
      "loss": 0.2294,
      "num_input_tokens_seen": 11270656,
      "step": 53410
    },
    {
      "epoch": 5.876237623762377,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.0266964192416211,
      "loss": 0.2314,
      "num_input_tokens_seen": 11271680,
      "step": 53415
    },
    {
      "epoch": 5.876787678767877,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.026695517611478727,
      "loss": 0.2335,
      "num_input_tokens_seen": 11272800,
      "step": 53420
    },
    {
      "epoch": 5.877337733773377,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.026694615873544388,
      "loss": 0.2331,
      "num_input_tokens_seen": 11273856,
      "step": 53425
    },
    {
      "epoch": 5.877887788778878,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02669371402782638,
      "loss": 0.234,
      "num_input_tokens_seen": 11274912,
      "step": 53430
    },
    {
      "epoch": 5.878437843784378,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.026692812074333026,
      "loss": 0.233,
      "num_input_tokens_seen": 11275968,
      "step": 53435
    },
    {
      "epoch": 5.878987898789879,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.02669191001307263,
      "loss": 0.2325,
      "num_input_tokens_seen": 11277024,
      "step": 53440
    },
    {
      "epoch": 5.87953795379538,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02669100784405351,
      "loss": 0.2319,
      "num_input_tokens_seen": 11278080,
      "step": 53445
    },
    {
      "epoch": 5.88008800880088,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.026690105567283986,
      "loss": 0.234,
      "num_input_tokens_seen": 11279072,
      "step": 53450
    },
    {
      "epoch": 5.880638063806381,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02668920318277236,
      "loss": 0.2314,
      "num_input_tokens_seen": 11280192,
      "step": 53455
    },
    {
      "epoch": 5.881188118811881,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.02668830069052696,
      "loss": 0.2309,
      "num_input_tokens_seen": 11281248,
      "step": 53460
    },
    {
      "epoch": 5.881738173817382,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.026687398090556105,
      "loss": 0.2319,
      "num_input_tokens_seen": 11282336,
      "step": 53465
    },
    {
      "epoch": 5.882288228822882,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.026686495382868106,
      "loss": 0.2314,
      "num_input_tokens_seen": 11283392,
      "step": 53470
    },
    {
      "epoch": 5.882838283828383,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02668559256747129,
      "loss": 0.2303,
      "num_input_tokens_seen": 11284416,
      "step": 53475
    },
    {
      "epoch": 5.883388338833884,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.026684689644373972,
      "loss": 0.2324,
      "num_input_tokens_seen": 11285536,
      "step": 53480
    },
    {
      "epoch": 5.883938393839384,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02668378661358448,
      "loss": 0.2335,
      "num_input_tokens_seen": 11286592,
      "step": 53485
    },
    {
      "epoch": 5.884488448844884,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.026682883475111125,
      "loss": 0.2324,
      "num_input_tokens_seen": 11287680,
      "step": 53490
    },
    {
      "epoch": 5.885038503850385,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.02668198022896225,
      "loss": 0.2309,
      "num_input_tokens_seen": 11288736,
      "step": 53495
    },
    {
      "epoch": 5.885588558855885,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.026681076875146162,
      "loss": 0.2319,
      "num_input_tokens_seen": 11289792,
      "step": 53500
    },
    {
      "epoch": 5.8861386138613865,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.026680173413671196,
      "loss": 0.2314,
      "num_input_tokens_seen": 11290848,
      "step": 53505
    },
    {
      "epoch": 5.886688668866887,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026679269844545675,
      "loss": 0.2314,
      "num_input_tokens_seen": 11291872,
      "step": 53510
    },
    {
      "epoch": 5.887238723872387,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.02667836616777793,
      "loss": 0.2309,
      "num_input_tokens_seen": 11292864,
      "step": 53515
    },
    {
      "epoch": 5.887788778877888,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.02667746238337628,
      "loss": 0.2293,
      "num_input_tokens_seen": 11293888,
      "step": 53520
    },
    {
      "epoch": 5.888338833883388,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.026676558491349072,
      "loss": 0.2309,
      "num_input_tokens_seen": 11294944,
      "step": 53525
    },
    {
      "epoch": 5.888888888888889,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.026675654491704625,
      "loss": 0.2319,
      "num_input_tokens_seen": 11296000,
      "step": 53530
    },
    {
      "epoch": 5.8894389438943895,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02667475038445127,
      "loss": 0.2314,
      "num_input_tokens_seen": 11297088,
      "step": 53535
    },
    {
      "epoch": 5.88998899889989,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.026673846169597348,
      "loss": 0.2319,
      "num_input_tokens_seen": 11298176,
      "step": 53540
    },
    {
      "epoch": 5.890539053905391,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.02667294184715118,
      "loss": 0.2309,
      "num_input_tokens_seen": 11299232,
      "step": 53545
    },
    {
      "epoch": 5.891089108910891,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026672037417121115,
      "loss": 0.2309,
      "num_input_tokens_seen": 11300352,
      "step": 53550
    },
    {
      "epoch": 5.891639163916391,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02667113287951548,
      "loss": 0.2314,
      "num_input_tokens_seen": 11301344,
      "step": 53555
    },
    {
      "epoch": 5.892189218921892,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.026670228234342612,
      "loss": 0.2325,
      "num_input_tokens_seen": 11302400,
      "step": 53560
    },
    {
      "epoch": 5.8927392739273925,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.026669323481610854,
      "loss": 0.2309,
      "num_input_tokens_seen": 11303392,
      "step": 53565
    },
    {
      "epoch": 5.893289328932894,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.026668418621328537,
      "loss": 0.2319,
      "num_input_tokens_seen": 11304416,
      "step": 53570
    },
    {
      "epoch": 5.893839383938394,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026667513653504008,
      "loss": 0.2319,
      "num_input_tokens_seen": 11305472,
      "step": 53575
    },
    {
      "epoch": 5.894389438943895,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.026666608578145602,
      "loss": 0.2309,
      "num_input_tokens_seen": 11306496,
      "step": 53580
    },
    {
      "epoch": 5.894939493949395,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.026665703395261663,
      "loss": 0.2324,
      "num_input_tokens_seen": 11307552,
      "step": 53585
    },
    {
      "epoch": 5.895489548954895,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.026664798104860533,
      "loss": 0.234,
      "num_input_tokens_seen": 11308640,
      "step": 53590
    },
    {
      "epoch": 5.896039603960396,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.026663892706950555,
      "loss": 0.2298,
      "num_input_tokens_seen": 11309664,
      "step": 53595
    },
    {
      "epoch": 5.896589658965897,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.02666298720154008,
      "loss": 0.2314,
      "num_input_tokens_seen": 11310688,
      "step": 53600
    },
    {
      "epoch": 5.897139713971397,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.026662081588637445,
      "loss": 0.233,
      "num_input_tokens_seen": 11311744,
      "step": 53605
    },
    {
      "epoch": 5.897689768976898,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.026661175868250998,
      "loss": 0.2335,
      "num_input_tokens_seen": 11312768,
      "step": 53610
    },
    {
      "epoch": 5.898239823982398,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02666027004038909,
      "loss": 0.233,
      "num_input_tokens_seen": 11313920,
      "step": 53615
    },
    {
      "epoch": 5.898789878987898,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02665936410506007,
      "loss": 0.2319,
      "num_input_tokens_seen": 11314944,
      "step": 53620
    },
    {
      "epoch": 5.899339933993399,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.026658458062272287,
      "loss": 0.2314,
      "num_input_tokens_seen": 11316064,
      "step": 53625
    },
    {
      "epoch": 5.8998899889989,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02665755191203409,
      "loss": 0.2324,
      "num_input_tokens_seen": 11317120,
      "step": 53630
    },
    {
      "epoch": 5.900440044004401,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.026656645654353824,
      "loss": 0.2309,
      "num_input_tokens_seen": 11318208,
      "step": 53635
    },
    {
      "epoch": 5.900990099009901,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.026655739289239855,
      "loss": 0.2303,
      "num_input_tokens_seen": 11319328,
      "step": 53640
    },
    {
      "epoch": 5.901540154015402,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02665483281670053,
      "loss": 0.2308,
      "num_input_tokens_seen": 11320384,
      "step": 53645
    },
    {
      "epoch": 5.902090209020902,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0266539262367442,
      "loss": 0.2304,
      "num_input_tokens_seen": 11321440,
      "step": 53650
    },
    {
      "epoch": 5.902640264026402,
      "grad_norm": 0.0026702880859375,
      "learning_rate": 0.026653019549379226,
      "loss": 0.2309,
      "num_input_tokens_seen": 11322464,
      "step": 53655
    },
    {
      "epoch": 5.9031903190319035,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02665211275461396,
      "loss": 0.2314,
      "num_input_tokens_seen": 11323488,
      "step": 53660
    },
    {
      "epoch": 5.903740374037404,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02665120585245677,
      "loss": 0.2335,
      "num_input_tokens_seen": 11324512,
      "step": 53665
    },
    {
      "epoch": 5.904290429042904,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.026650298842915993,
      "loss": 0.2299,
      "num_input_tokens_seen": 11325536,
      "step": 53670
    },
    {
      "epoch": 5.904840484048405,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.02664939172600001,
      "loss": 0.2324,
      "num_input_tokens_seen": 11326592,
      "step": 53675
    },
    {
      "epoch": 5.905390539053905,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.026648484501717175,
      "loss": 0.2304,
      "num_input_tokens_seen": 11327648,
      "step": 53680
    },
    {
      "epoch": 5.905940594059406,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.026647577170075848,
      "loss": 0.232,
      "num_input_tokens_seen": 11328672,
      "step": 53685
    },
    {
      "epoch": 5.9064906490649065,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.026646669731084385,
      "loss": 0.2268,
      "num_input_tokens_seen": 11329664,
      "step": 53690
    },
    {
      "epoch": 5.907040704070407,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.026645762184751162,
      "loss": 0.2289,
      "num_input_tokens_seen": 11330688,
      "step": 53695
    },
    {
      "epoch": 5.907590759075908,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.026644854531084537,
      "loss": 0.2305,
      "num_input_tokens_seen": 11331808,
      "step": 53700
    },
    {
      "epoch": 5.908140814081408,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.026643946770092873,
      "loss": 0.2326,
      "num_input_tokens_seen": 11332832,
      "step": 53705
    },
    {
      "epoch": 5.908690869086909,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02664303890178454,
      "loss": 0.2326,
      "num_input_tokens_seen": 11333856,
      "step": 53710
    },
    {
      "epoch": 5.909240924092409,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.026642130926167903,
      "loss": 0.2284,
      "num_input_tokens_seen": 11334912,
      "step": 53715
    },
    {
      "epoch": 5.9097909790979095,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.026641222843251332,
      "loss": 0.2326,
      "num_input_tokens_seen": 11336032,
      "step": 53720
    },
    {
      "epoch": 5.910341034103411,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.026640314653043198,
      "loss": 0.2321,
      "num_input_tokens_seen": 11337120,
      "step": 53725
    },
    {
      "epoch": 5.910891089108911,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.026639406355551863,
      "loss": 0.2299,
      "num_input_tokens_seen": 11338176,
      "step": 53730
    },
    {
      "epoch": 5.911441144114411,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.026638497950785715,
      "loss": 0.2305,
      "num_input_tokens_seen": 11339232,
      "step": 53735
    },
    {
      "epoch": 5.911991199119912,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02663758943875311,
      "loss": 0.2347,
      "num_input_tokens_seen": 11340256,
      "step": 53740
    },
    {
      "epoch": 5.912541254125412,
      "grad_norm": 0.00054168701171875,
      "learning_rate": 0.026636680819462428,
      "loss": 0.2362,
      "num_input_tokens_seen": 11341312,
      "step": 53745
    },
    {
      "epoch": 5.913091309130913,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.02663577209292204,
      "loss": 0.2325,
      "num_input_tokens_seen": 11342368,
      "step": 53750
    },
    {
      "epoch": 5.913641364136414,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.02663486325914033,
      "loss": 0.2324,
      "num_input_tokens_seen": 11343488,
      "step": 53755
    },
    {
      "epoch": 5.914191419141914,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.026633954318125663,
      "loss": 0.2319,
      "num_input_tokens_seen": 11344544,
      "step": 53760
    },
    {
      "epoch": 5.914741474147415,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.026633045269886423,
      "loss": 0.2303,
      "num_input_tokens_seen": 11345568,
      "step": 53765
    },
    {
      "epoch": 5.915291529152915,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02663213611443099,
      "loss": 0.2314,
      "num_input_tokens_seen": 11346592,
      "step": 53770
    },
    {
      "epoch": 5.915841584158416,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.026631226851767734,
      "loss": 0.2288,
      "num_input_tokens_seen": 11347648,
      "step": 53775
    },
    {
      "epoch": 5.916391639163916,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.026630317481905048,
      "loss": 0.2314,
      "num_input_tokens_seen": 11348736,
      "step": 53780
    },
    {
      "epoch": 5.916941694169417,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0266294080048513,
      "loss": 0.2303,
      "num_input_tokens_seen": 11349856,
      "step": 53785
    },
    {
      "epoch": 5.917491749174918,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.026628498420614884,
      "loss": 0.2319,
      "num_input_tokens_seen": 11350880,
      "step": 53790
    },
    {
      "epoch": 5.918041804180418,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.026627588729204177,
      "loss": 0.2314,
      "num_input_tokens_seen": 11352032,
      "step": 53795
    },
    {
      "epoch": 5.918591859185918,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.026626678930627562,
      "loss": 0.2303,
      "num_input_tokens_seen": 11353088,
      "step": 53800
    },
    {
      "epoch": 5.919141914191419,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.026625769024893428,
      "loss": 0.2324,
      "num_input_tokens_seen": 11354176,
      "step": 53805
    },
    {
      "epoch": 5.919691969196919,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.026624859012010157,
      "loss": 0.2345,
      "num_input_tokens_seen": 11355328,
      "step": 53810
    },
    {
      "epoch": 5.9202420242024205,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.026623948891986143,
      "loss": 0.2303,
      "num_input_tokens_seen": 11356352,
      "step": 53815
    },
    {
      "epoch": 5.920792079207921,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.026623038664829766,
      "loss": 0.2314,
      "num_input_tokens_seen": 11357440,
      "step": 53820
    },
    {
      "epoch": 5.921342134213422,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02662212833054942,
      "loss": 0.2308,
      "num_input_tokens_seen": 11358432,
      "step": 53825
    },
    {
      "epoch": 5.921892189218922,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.026621217889153493,
      "loss": 0.2319,
      "num_input_tokens_seen": 11359424,
      "step": 53830
    },
    {
      "epoch": 5.922442244224422,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.026620307340650382,
      "loss": 0.2283,
      "num_input_tokens_seen": 11360480,
      "step": 53835
    },
    {
      "epoch": 5.922992299229923,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02661939668504847,
      "loss": 0.2319,
      "num_input_tokens_seen": 11361536,
      "step": 53840
    },
    {
      "epoch": 5.9235423542354235,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.026618485922356154,
      "loss": 0.2267,
      "num_input_tokens_seen": 11362560,
      "step": 53845
    },
    {
      "epoch": 5.924092409240924,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.02661757505258183,
      "loss": 0.2315,
      "num_input_tokens_seen": 11363616,
      "step": 53850
    },
    {
      "epoch": 5.924642464246425,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.026616664075733893,
      "loss": 0.2274,
      "num_input_tokens_seen": 11364672,
      "step": 53855
    },
    {
      "epoch": 5.925192519251925,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.026615752991820733,
      "loss": 0.2347,
      "num_input_tokens_seen": 11365728,
      "step": 53860
    },
    {
      "epoch": 5.925742574257426,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02661484180085075,
      "loss": 0.2362,
      "num_input_tokens_seen": 11366784,
      "step": 53865
    },
    {
      "epoch": 5.926292629262926,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.026613930502832353,
      "loss": 0.2289,
      "num_input_tokens_seen": 11367840,
      "step": 53870
    },
    {
      "epoch": 5.9268426842684265,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.026613019097773923,
      "loss": 0.2336,
      "num_input_tokens_seen": 11368896,
      "step": 53875
    },
    {
      "epoch": 5.927392739273928,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026612107585683872,
      "loss": 0.2357,
      "num_input_tokens_seen": 11369952,
      "step": 53880
    },
    {
      "epoch": 5.927942794279428,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.026611195966570595,
      "loss": 0.232,
      "num_input_tokens_seen": 11371008,
      "step": 53885
    },
    {
      "epoch": 5.928492849284929,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.026610284240442497,
      "loss": 0.2335,
      "num_input_tokens_seen": 11372096,
      "step": 53890
    },
    {
      "epoch": 5.929042904290429,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.02660937240730798,
      "loss": 0.2314,
      "num_input_tokens_seen": 11373120,
      "step": 53895
    },
    {
      "epoch": 5.929592959295929,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.026608460467175448,
      "loss": 0.2319,
      "num_input_tokens_seen": 11374208,
      "step": 53900
    },
    {
      "epoch": 5.93014301430143,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02660754842005331,
      "loss": 0.2283,
      "num_input_tokens_seen": 11375200,
      "step": 53905
    },
    {
      "epoch": 5.930693069306931,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.026606636265949966,
      "loss": 0.233,
      "num_input_tokens_seen": 11376224,
      "step": 53910
    },
    {
      "epoch": 5.931243124312431,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.026605724004873824,
      "loss": 0.2309,
      "num_input_tokens_seen": 11377248,
      "step": 53915
    },
    {
      "epoch": 5.931793179317932,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.026604811636833297,
      "loss": 0.2289,
      "num_input_tokens_seen": 11378272,
      "step": 53920
    },
    {
      "epoch": 5.932343234323432,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02660389916183678,
      "loss": 0.2309,
      "num_input_tokens_seen": 11379296,
      "step": 53925
    },
    {
      "epoch": 5.932893289328933,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.026602986579892702,
      "loss": 0.2294,
      "num_input_tokens_seen": 11380352,
      "step": 53930
    },
    {
      "epoch": 5.933443344334433,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.026602073891009458,
      "loss": 0.2336,
      "num_input_tokens_seen": 11381376,
      "step": 53935
    },
    {
      "epoch": 5.933993399339934,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.02660116109519547,
      "loss": 0.232,
      "num_input_tokens_seen": 11382496,
      "step": 53940
    },
    {
      "epoch": 5.934543454345435,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.026600248192459146,
      "loss": 0.2346,
      "num_input_tokens_seen": 11383616,
      "step": 53945
    },
    {
      "epoch": 5.935093509350935,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0265993351828089,
      "loss": 0.232,
      "num_input_tokens_seen": 11384608,
      "step": 53950
    },
    {
      "epoch": 5.935643564356436,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.02659842206625315,
      "loss": 0.2315,
      "num_input_tokens_seen": 11385632,
      "step": 53955
    },
    {
      "epoch": 5.936193619361936,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.026597508842800308,
      "loss": 0.2315,
      "num_input_tokens_seen": 11386720,
      "step": 53960
    },
    {
      "epoch": 5.936743674367436,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.02659659551245879,
      "loss": 0.2268,
      "num_input_tokens_seen": 11387840,
      "step": 53965
    },
    {
      "epoch": 5.9372937293729375,
      "grad_norm": 0.009765625,
      "learning_rate": 0.026595682075237017,
      "loss": 0.2294,
      "num_input_tokens_seen": 11388864,
      "step": 53970
    },
    {
      "epoch": 5.937843784378438,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.026594768531143405,
      "loss": 0.2331,
      "num_input_tokens_seen": 11389856,
      "step": 53975
    },
    {
      "epoch": 5.938393839383938,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02659385488018638,
      "loss": 0.2326,
      "num_input_tokens_seen": 11390976,
      "step": 53980
    },
    {
      "epoch": 5.938943894389439,
      "grad_norm": 0.009765625,
      "learning_rate": 0.026592941122374352,
      "loss": 0.2284,
      "num_input_tokens_seen": 11392032,
      "step": 53985
    },
    {
      "epoch": 5.939493949394939,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.026592027257715752,
      "loss": 0.2346,
      "num_input_tokens_seen": 11393088,
      "step": 53990
    },
    {
      "epoch": 5.94004400440044,
      "grad_norm": 0.00074005126953125,
      "learning_rate": 0.026591113286219,
      "loss": 0.2274,
      "num_input_tokens_seen": 11394080,
      "step": 53995
    },
    {
      "epoch": 5.9405940594059405,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.026590199207892518,
      "loss": 0.2269,
      "num_input_tokens_seen": 11395168,
      "step": 54000
    },
    {
      "epoch": 5.941144114411442,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.026589285022744727,
      "loss": 0.238,
      "num_input_tokens_seen": 11396192,
      "step": 54005
    },
    {
      "epoch": 5.941694169416942,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.026588370730784062,
      "loss": 0.2266,
      "num_input_tokens_seen": 11397216,
      "step": 54010
    },
    {
      "epoch": 5.942244224422442,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.026587456332018942,
      "loss": 0.2346,
      "num_input_tokens_seen": 11398240,
      "step": 54015
    },
    {
      "epoch": 5.942794279427943,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.026586541826457803,
      "loss": 0.234,
      "num_input_tokens_seen": 11399296,
      "step": 54020
    },
    {
      "epoch": 5.943344334433443,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.026585627214109057,
      "loss": 0.2355,
      "num_input_tokens_seen": 11400352,
      "step": 54025
    },
    {
      "epoch": 5.9438943894389435,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.026584712494981153,
      "loss": 0.2291,
      "num_input_tokens_seen": 11401376,
      "step": 54030
    },
    {
      "epoch": 5.944444444444445,
      "grad_norm": 0.00067901611328125,
      "learning_rate": 0.026583797669082507,
      "loss": 0.2396,
      "num_input_tokens_seen": 11402400,
      "step": 54035
    },
    {
      "epoch": 5.944994499449945,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.02658288273642156,
      "loss": 0.2322,
      "num_input_tokens_seen": 11403424,
      "step": 54040
    },
    {
      "epoch": 5.945544554455445,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.026581967697006742,
      "loss": 0.2301,
      "num_input_tokens_seen": 11404512,
      "step": 54045
    },
    {
      "epoch": 5.946094609460946,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.026581052550846484,
      "loss": 0.2348,
      "num_input_tokens_seen": 11405568,
      "step": 54050
    },
    {
      "epoch": 5.946644664466446,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02658013729794922,
      "loss": 0.2301,
      "num_input_tokens_seen": 11406656,
      "step": 54055
    },
    {
      "epoch": 5.947194719471947,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02657922193832339,
      "loss": 0.2311,
      "num_input_tokens_seen": 11407712,
      "step": 54060
    },
    {
      "epoch": 5.947744774477448,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.026578306471977425,
      "loss": 0.228,
      "num_input_tokens_seen": 11408704,
      "step": 54065
    },
    {
      "epoch": 5.948294829482949,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.026577390898919768,
      "loss": 0.2311,
      "num_input_tokens_seen": 11409792,
      "step": 54070
    },
    {
      "epoch": 5.948844884488449,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.026576475219158853,
      "loss": 0.2369,
      "num_input_tokens_seen": 11410880,
      "step": 54075
    },
    {
      "epoch": 5.949394939493949,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.026575559432703122,
      "loss": 0.2307,
      "num_input_tokens_seen": 11411904,
      "step": 54080
    },
    {
      "epoch": 5.94994499449945,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.026574643539561012,
      "loss": 0.2291,
      "num_input_tokens_seen": 11412960,
      "step": 54085
    },
    {
      "epoch": 5.9504950495049505,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02657372753974097,
      "loss": 0.2333,
      "num_input_tokens_seen": 11413984,
      "step": 54090
    },
    {
      "epoch": 5.951045104510451,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02657281143325143,
      "loss": 0.2296,
      "num_input_tokens_seen": 11415072,
      "step": 54095
    },
    {
      "epoch": 5.951595159515952,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.026571895220100842,
      "loss": 0.2305,
      "num_input_tokens_seen": 11416096,
      "step": 54100
    },
    {
      "epoch": 5.952145214521452,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02657097890029765,
      "loss": 0.2404,
      "num_input_tokens_seen": 11417184,
      "step": 54105
    },
    {
      "epoch": 5.952695269526953,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.026570062473850298,
      "loss": 0.2315,
      "num_input_tokens_seen": 11418272,
      "step": 54110
    },
    {
      "epoch": 5.953245324532453,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.026569145940767227,
      "loss": 0.2325,
      "num_input_tokens_seen": 11419264,
      "step": 54115
    },
    {
      "epoch": 5.9537953795379535,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.026568229301056897,
      "loss": 0.2289,
      "num_input_tokens_seen": 11420288,
      "step": 54120
    },
    {
      "epoch": 5.9543454345434546,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.026567312554727742,
      "loss": 0.2336,
      "num_input_tokens_seen": 11421344,
      "step": 54125
    },
    {
      "epoch": 5.954895489548955,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02656639570178822,
      "loss": 0.2304,
      "num_input_tokens_seen": 11422336,
      "step": 54130
    },
    {
      "epoch": 5.955445544554456,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.026565478742246785,
      "loss": 0.2315,
      "num_input_tokens_seen": 11423456,
      "step": 54135
    },
    {
      "epoch": 5.955995599559956,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.026564561676111872,
      "loss": 0.2305,
      "num_input_tokens_seen": 11424512,
      "step": 54140
    },
    {
      "epoch": 5.956545654565456,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.026563644503391947,
      "loss": 0.2326,
      "num_input_tokens_seen": 11425632,
      "step": 54145
    },
    {
      "epoch": 5.957095709570957,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.026562727224095462,
      "loss": 0.2326,
      "num_input_tokens_seen": 11426752,
      "step": 54150
    },
    {
      "epoch": 5.957645764576458,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.026561809838230865,
      "loss": 0.232,
      "num_input_tokens_seen": 11427808,
      "step": 54155
    },
    {
      "epoch": 5.958195819581958,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.02656089234580662,
      "loss": 0.233,
      "num_input_tokens_seen": 11428832,
      "step": 54160
    },
    {
      "epoch": 5.958745874587459,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.02655997474683117,
      "loss": 0.2325,
      "num_input_tokens_seen": 11429888,
      "step": 54165
    },
    {
      "epoch": 5.959295929592959,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.026559057041312985,
      "loss": 0.2325,
      "num_input_tokens_seen": 11430912,
      "step": 54170
    },
    {
      "epoch": 5.95984598459846,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.026558139229260518,
      "loss": 0.2319,
      "num_input_tokens_seen": 11432032,
      "step": 54175
    },
    {
      "epoch": 5.96039603960396,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.026557221310682225,
      "loss": 0.233,
      "num_input_tokens_seen": 11433088,
      "step": 54180
    },
    {
      "epoch": 5.960946094609461,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02655630328558657,
      "loss": 0.2319,
      "num_input_tokens_seen": 11434240,
      "step": 54185
    },
    {
      "epoch": 5.961496149614962,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.026555385153982012,
      "loss": 0.2298,
      "num_input_tokens_seen": 11435296,
      "step": 54190
    },
    {
      "epoch": 5.962046204620462,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.026554466915877015,
      "loss": 0.2309,
      "num_input_tokens_seen": 11436320,
      "step": 54195
    },
    {
      "epoch": 5.962596259625963,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.02655354857128004,
      "loss": 0.2319,
      "num_input_tokens_seen": 11437408,
      "step": 54200
    },
    {
      "epoch": 5.963146314631463,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.026552630120199556,
      "loss": 0.2304,
      "num_input_tokens_seen": 11438432,
      "step": 54205
    },
    {
      "epoch": 5.963696369636963,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02655171156264402,
      "loss": 0.2298,
      "num_input_tokens_seen": 11439456,
      "step": 54210
    },
    {
      "epoch": 5.9642464246424645,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.0265507928986219,
      "loss": 0.2303,
      "num_input_tokens_seen": 11440544,
      "step": 54215
    },
    {
      "epoch": 5.964796479647965,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02654987412814166,
      "loss": 0.233,
      "num_input_tokens_seen": 11441568,
      "step": 54220
    },
    {
      "epoch": 5.965346534653465,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.026548955251211783,
      "loss": 0.2329,
      "num_input_tokens_seen": 11442688,
      "step": 54225
    },
    {
      "epoch": 5.965896589658966,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.026548036267840722,
      "loss": 0.2303,
      "num_input_tokens_seen": 11443904,
      "step": 54230
    },
    {
      "epoch": 5.966446644664466,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.026547117178036952,
      "loss": 0.2314,
      "num_input_tokens_seen": 11444992,
      "step": 54235
    },
    {
      "epoch": 5.966996699669967,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.026546197981808947,
      "loss": 0.2308,
      "num_input_tokens_seen": 11446176,
      "step": 54240
    },
    {
      "epoch": 5.9675467546754675,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.02654527867916517,
      "loss": 0.2309,
      "num_input_tokens_seen": 11447200,
      "step": 54245
    },
    {
      "epoch": 5.968096809680969,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.026544359270114105,
      "loss": 0.2324,
      "num_input_tokens_seen": 11448256,
      "step": 54250
    },
    {
      "epoch": 5.968646864686469,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.026543439754664215,
      "loss": 0.2298,
      "num_input_tokens_seen": 11449312,
      "step": 54255
    },
    {
      "epoch": 5.969196919691969,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.026542520132823984,
      "loss": 0.2314,
      "num_input_tokens_seen": 11450432,
      "step": 54260
    },
    {
      "epoch": 5.96974697469747,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.026541600404601882,
      "loss": 0.2314,
      "num_input_tokens_seen": 11451488,
      "step": 54265
    },
    {
      "epoch": 5.97029702970297,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.02654068057000639,
      "loss": 0.2314,
      "num_input_tokens_seen": 11452544,
      "step": 54270
    },
    {
      "epoch": 5.9708470847084705,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.026539760629045978,
      "loss": 0.2303,
      "num_input_tokens_seen": 11453536,
      "step": 54275
    },
    {
      "epoch": 5.971397139713972,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02653884058172913,
      "loss": 0.2298,
      "num_input_tokens_seen": 11454560,
      "step": 54280
    },
    {
      "epoch": 5.971947194719472,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.02653792042806433,
      "loss": 0.2314,
      "num_input_tokens_seen": 11455648,
      "step": 54285
    },
    {
      "epoch": 5.972497249724973,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.02653700016806005,
      "loss": 0.2309,
      "num_input_tokens_seen": 11456608,
      "step": 54290
    },
    {
      "epoch": 5.973047304730473,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.026536079801724777,
      "loss": 0.2319,
      "num_input_tokens_seen": 11457664,
      "step": 54295
    },
    {
      "epoch": 5.973597359735973,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02653515932906699,
      "loss": 0.2298,
      "num_input_tokens_seen": 11458784,
      "step": 54300
    },
    {
      "epoch": 5.974147414741474,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.026534238750095175,
      "loss": 0.2314,
      "num_input_tokens_seen": 11459776,
      "step": 54305
    },
    {
      "epoch": 5.974697469746975,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.026533318064817818,
      "loss": 0.2324,
      "num_input_tokens_seen": 11460768,
      "step": 54310
    },
    {
      "epoch": 5.975247524752476,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.026532397273243403,
      "loss": 0.2293,
      "num_input_tokens_seen": 11461792,
      "step": 54315
    },
    {
      "epoch": 5.975797579757976,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.026531476375380416,
      "loss": 0.2314,
      "num_input_tokens_seen": 11462784,
      "step": 54320
    },
    {
      "epoch": 5.976347634763476,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02653055537123734,
      "loss": 0.2319,
      "num_input_tokens_seen": 11463840,
      "step": 54325
    },
    {
      "epoch": 5.976897689768977,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.026529634260822672,
      "loss": 0.2314,
      "num_input_tokens_seen": 11464864,
      "step": 54330
    },
    {
      "epoch": 5.977447744774477,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.0265287130441449,
      "loss": 0.2335,
      "num_input_tokens_seen": 11465888,
      "step": 54335
    },
    {
      "epoch": 5.977997799779978,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.026527791721212505,
      "loss": 0.2304,
      "num_input_tokens_seen": 11466912,
      "step": 54340
    },
    {
      "epoch": 5.978547854785479,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.026526870292033992,
      "loss": 0.2329,
      "num_input_tokens_seen": 11468000,
      "step": 54345
    },
    {
      "epoch": 5.979097909790979,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.026525948756617845,
      "loss": 0.2324,
      "num_input_tokens_seen": 11469088,
      "step": 54350
    },
    {
      "epoch": 5.97964796479648,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.026525027114972555,
      "loss": 0.2319,
      "num_input_tokens_seen": 11470144,
      "step": 54355
    },
    {
      "epoch": 5.98019801980198,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.026524105367106622,
      "loss": 0.2309,
      "num_input_tokens_seen": 11471232,
      "step": 54360
    },
    {
      "epoch": 5.98074807480748,
      "grad_norm": 0.001953125,
      "learning_rate": 0.026523183513028542,
      "loss": 0.2324,
      "num_input_tokens_seen": 11472288,
      "step": 54365
    },
    {
      "epoch": 5.9812981298129815,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.02652226155274681,
      "loss": 0.2298,
      "num_input_tokens_seen": 11473344,
      "step": 54370
    },
    {
      "epoch": 5.981848184818482,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.026521339486269918,
      "loss": 0.233,
      "num_input_tokens_seen": 11474400,
      "step": 54375
    },
    {
      "epoch": 5.982398239823983,
      "grad_norm": 0.0008392333984375,
      "learning_rate": 0.026520417313606368,
      "loss": 0.233,
      "num_input_tokens_seen": 11475456,
      "step": 54380
    },
    {
      "epoch": 5.982948294829483,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.026519495034764664,
      "loss": 0.2309,
      "num_input_tokens_seen": 11476512,
      "step": 54385
    },
    {
      "epoch": 5.983498349834983,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0265185726497533,
      "loss": 0.232,
      "num_input_tokens_seen": 11477600,
      "step": 54390
    },
    {
      "epoch": 5.984048404840484,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.026517650158580777,
      "loss": 0.2309,
      "num_input_tokens_seen": 11478656,
      "step": 54395
    },
    {
      "epoch": 5.9845984598459845,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.0265167275612556,
      "loss": 0.2304,
      "num_input_tokens_seen": 11479680,
      "step": 54400
    },
    {
      "epoch": 5.985148514851485,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.026515804857786276,
      "loss": 0.233,
      "num_input_tokens_seen": 11480704,
      "step": 54405
    },
    {
      "epoch": 5.985698569856986,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.0265148820481813,
      "loss": 0.2289,
      "num_input_tokens_seen": 11481760,
      "step": 54410
    },
    {
      "epoch": 5.986248624862486,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02651395913244918,
      "loss": 0.2361,
      "num_input_tokens_seen": 11482848,
      "step": 54415
    },
    {
      "epoch": 5.986798679867987,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.026513036110598426,
      "loss": 0.2309,
      "num_input_tokens_seen": 11483904,
      "step": 54420
    },
    {
      "epoch": 5.987348734873487,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02651211298263754,
      "loss": 0.2329,
      "num_input_tokens_seen": 11484960,
      "step": 54425
    },
    {
      "epoch": 5.987898789878988,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02651118974857504,
      "loss": 0.2314,
      "num_input_tokens_seen": 11485984,
      "step": 54430
    },
    {
      "epoch": 5.988448844884489,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02651026640841942,
      "loss": 0.2335,
      "num_input_tokens_seen": 11487072,
      "step": 54435
    },
    {
      "epoch": 5.988998899889989,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0265093429621792,
      "loss": 0.2303,
      "num_input_tokens_seen": 11488064,
      "step": 54440
    },
    {
      "epoch": 5.98954895489549,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02650841940986289,
      "loss": 0.2304,
      "num_input_tokens_seen": 11489184,
      "step": 54445
    },
    {
      "epoch": 5.99009900990099,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.026507495751479003,
      "loss": 0.2294,
      "num_input_tokens_seen": 11490208,
      "step": 54450
    },
    {
      "epoch": 5.99064906490649,
      "grad_norm": 0.0006256103515625,
      "learning_rate": 0.026506571987036046,
      "loss": 0.2305,
      "num_input_tokens_seen": 11491232,
      "step": 54455
    },
    {
      "epoch": 5.991199119911991,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.026505648116542536,
      "loss": 0.2305,
      "num_input_tokens_seen": 11492320,
      "step": 54460
    },
    {
      "epoch": 5.991749174917492,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.02650472414000699,
      "loss": 0.2265,
      "num_input_tokens_seen": 11493376,
      "step": 54465
    },
    {
      "epoch": 5.992299229922993,
      "grad_norm": 0.00909423828125,
      "learning_rate": 0.026503800057437917,
      "loss": 0.2313,
      "num_input_tokens_seen": 11494432,
      "step": 54470
    },
    {
      "epoch": 5.992849284928493,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.026502875868843843,
      "loss": 0.2339,
      "num_input_tokens_seen": 11495456,
      "step": 54475
    },
    {
      "epoch": 5.993399339933993,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.026501951574233285,
      "loss": 0.237,
      "num_input_tokens_seen": 11496576,
      "step": 54480
    },
    {
      "epoch": 5.993949394939494,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.026501027173614758,
      "loss": 0.2276,
      "num_input_tokens_seen": 11497664,
      "step": 54485
    },
    {
      "epoch": 5.994499449944994,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.026500102666996775,
      "loss": 0.2281,
      "num_input_tokens_seen": 11498752,
      "step": 54490
    },
    {
      "epoch": 5.9950495049504955,
      "grad_norm": 0.0078125,
      "learning_rate": 0.026499178054387867,
      "loss": 0.2364,
      "num_input_tokens_seen": 11499776,
      "step": 54495
    },
    {
      "epoch": 5.995599559955996,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.02649825333579656,
      "loss": 0.2321,
      "num_input_tokens_seen": 11500800,
      "step": 54500
    },
    {
      "epoch": 5.996149614961496,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02649732851123136,
      "loss": 0.2274,
      "num_input_tokens_seen": 11501856,
      "step": 54505
    },
    {
      "epoch": 5.996699669966997,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.026496403580700805,
      "loss": 0.2291,
      "num_input_tokens_seen": 11502880,
      "step": 54510
    },
    {
      "epoch": 5.997249724972497,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.026495478544213413,
      "loss": 0.2322,
      "num_input_tokens_seen": 11503904,
      "step": 54515
    },
    {
      "epoch": 5.997799779977997,
      "grad_norm": 0.007415771484375,
      "learning_rate": 0.026494553401777712,
      "loss": 0.2358,
      "num_input_tokens_seen": 11504992,
      "step": 54520
    },
    {
      "epoch": 5.9983498349834985,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.026493628153402227,
      "loss": 0.2363,
      "num_input_tokens_seen": 11505984,
      "step": 54525
    },
    {
      "epoch": 5.998899889988999,
      "grad_norm": 0.005859375,
      "learning_rate": 0.026492702799095484,
      "loss": 0.2341,
      "num_input_tokens_seen": 11507072,
      "step": 54530
    },
    {
      "epoch": 5.9994499449945,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.026491777338866023,
      "loss": 0.2335,
      "num_input_tokens_seen": 11508096,
      "step": 54535
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.026490851772722354,
      "loss": 0.2293,
      "num_input_tokens_seen": 11509088,
      "step": 54540
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.23144161701202393,
      "eval_runtime": 60.5222,
      "eval_samples_per_second": 66.752,
      "eval_steps_per_second": 16.688,
      "num_input_tokens_seen": 11509088,
      "step": 54540
    },
    {
      "epoch": 6.0005500550055,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.026489926100673026,
      "loss": 0.2319,
      "num_input_tokens_seen": 11510144,
      "step": 54545
    },
    {
      "epoch": 6.001100110011001,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02648900032272656,
      "loss": 0.233,
      "num_input_tokens_seen": 11511168,
      "step": 54550
    },
    {
      "epoch": 6.0016501650165015,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02648807443889149,
      "loss": 0.2309,
      "num_input_tokens_seen": 11512224,
      "step": 54555
    },
    {
      "epoch": 6.002200220022003,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02648714844917635,
      "loss": 0.2309,
      "num_input_tokens_seen": 11513280,
      "step": 54560
    },
    {
      "epoch": 6.002750275027503,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.026486222353589675,
      "loss": 0.2288,
      "num_input_tokens_seen": 11514240,
      "step": 54565
    },
    {
      "epoch": 6.003300330033003,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.026485296152139997,
      "loss": 0.2325,
      "num_input_tokens_seen": 11515296,
      "step": 54570
    },
    {
      "epoch": 6.003850385038504,
      "grad_norm": 0.000762939453125,
      "learning_rate": 0.026484369844835865,
      "loss": 0.2298,
      "num_input_tokens_seen": 11516384,
      "step": 54575
    },
    {
      "epoch": 6.004400440044004,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.026483443431685797,
      "loss": 0.2309,
      "num_input_tokens_seen": 11517504,
      "step": 54580
    },
    {
      "epoch": 6.0049504950495045,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.026482516912698345,
      "loss": 0.2304,
      "num_input_tokens_seen": 11518464,
      "step": 54585
    },
    {
      "epoch": 6.005500550055006,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.026481590287882047,
      "loss": 0.2315,
      "num_input_tokens_seen": 11519552,
      "step": 54590
    },
    {
      "epoch": 6.006050605060506,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.02648066355724544,
      "loss": 0.2304,
      "num_input_tokens_seen": 11520544,
      "step": 54595
    },
    {
      "epoch": 6.006600660066007,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02647973672079706,
      "loss": 0.233,
      "num_input_tokens_seen": 11521664,
      "step": 54600
    },
    {
      "epoch": 6.007150715071507,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.026478809778545463,
      "loss": 0.232,
      "num_input_tokens_seen": 11522752,
      "step": 54605
    },
    {
      "epoch": 6.007700770077007,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02647788273049918,
      "loss": 0.233,
      "num_input_tokens_seen": 11523776,
      "step": 54610
    },
    {
      "epoch": 6.008250825082508,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.026476955576666763,
      "loss": 0.2288,
      "num_input_tokens_seen": 11524768,
      "step": 54615
    },
    {
      "epoch": 6.008800880088009,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.026476028317056748,
      "loss": 0.2325,
      "num_input_tokens_seen": 11525792,
      "step": 54620
    },
    {
      "epoch": 6.00935093509351,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.026475100951677693,
      "loss": 0.2304,
      "num_input_tokens_seen": 11526848,
      "step": 54625
    },
    {
      "epoch": 6.00990099009901,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026474173480538134,
      "loss": 0.233,
      "num_input_tokens_seen": 11527840,
      "step": 54630
    },
    {
      "epoch": 6.01045104510451,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.026473245903646622,
      "loss": 0.2314,
      "num_input_tokens_seen": 11528928,
      "step": 54635
    },
    {
      "epoch": 6.011001100110011,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.026472318221011717,
      "loss": 0.2293,
      "num_input_tokens_seen": 11529952,
      "step": 54640
    },
    {
      "epoch": 6.011551155115511,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02647139043264195,
      "loss": 0.233,
      "num_input_tokens_seen": 11530912,
      "step": 54645
    },
    {
      "epoch": 6.0121012101210125,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.026470462538545885,
      "loss": 0.2309,
      "num_input_tokens_seen": 11531936,
      "step": 54650
    },
    {
      "epoch": 6.012651265126513,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.026469534538732072,
      "loss": 0.2314,
      "num_input_tokens_seen": 11532992,
      "step": 54655
    },
    {
      "epoch": 6.013201320132013,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.026468606433209065,
      "loss": 0.2314,
      "num_input_tokens_seen": 11534080,
      "step": 54660
    },
    {
      "epoch": 6.013751375137514,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.026467678221985414,
      "loss": 0.2309,
      "num_input_tokens_seen": 11535136,
      "step": 54665
    },
    {
      "epoch": 6.014301430143014,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.026466749905069677,
      "loss": 0.234,
      "num_input_tokens_seen": 11536192,
      "step": 54670
    },
    {
      "epoch": 6.014851485148514,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026465821482470404,
      "loss": 0.232,
      "num_input_tokens_seen": 11537184,
      "step": 54675
    },
    {
      "epoch": 6.0154015401540155,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02646489295419616,
      "loss": 0.2278,
      "num_input_tokens_seen": 11538240,
      "step": 54680
    },
    {
      "epoch": 6.015951595159516,
      "grad_norm": 0.000774383544921875,
      "learning_rate": 0.026463964320255495,
      "loss": 0.2309,
      "num_input_tokens_seen": 11539296,
      "step": 54685
    },
    {
      "epoch": 6.016501650165017,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026463035580656977,
      "loss": 0.2288,
      "num_input_tokens_seen": 11540480,
      "step": 54690
    },
    {
      "epoch": 6.017051705170517,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.026462106735409156,
      "loss": 0.2299,
      "num_input_tokens_seen": 11541536,
      "step": 54695
    },
    {
      "epoch": 6.017601760176017,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.0264611777845206,
      "loss": 0.231,
      "num_input_tokens_seen": 11542528,
      "step": 54700
    },
    {
      "epoch": 6.018151815181518,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.026460248727999864,
      "loss": 0.2315,
      "num_input_tokens_seen": 11543584,
      "step": 54705
    },
    {
      "epoch": 6.0187018701870185,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02645931956585552,
      "loss": 0.2305,
      "num_input_tokens_seen": 11544640,
      "step": 54710
    },
    {
      "epoch": 6.01925192519252,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.026458390298096126,
      "loss": 0.2305,
      "num_input_tokens_seen": 11545696,
      "step": 54715
    },
    {
      "epoch": 6.01980198019802,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.026457460924730246,
      "loss": 0.2315,
      "num_input_tokens_seen": 11546720,
      "step": 54720
    },
    {
      "epoch": 6.02035203520352,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.026456531445766443,
      "loss": 0.2331,
      "num_input_tokens_seen": 11547776,
      "step": 54725
    },
    {
      "epoch": 6.020902090209021,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02645560186121329,
      "loss": 0.2304,
      "num_input_tokens_seen": 11548832,
      "step": 54730
    },
    {
      "epoch": 6.021452145214521,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.026454672171079354,
      "loss": 0.231,
      "num_input_tokens_seen": 11549856,
      "step": 54735
    },
    {
      "epoch": 6.022002200220022,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.026453742375373197,
      "loss": 0.2288,
      "num_input_tokens_seen": 11550912,
      "step": 54740
    },
    {
      "epoch": 6.022552255225523,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02645281247410339,
      "loss": 0.2289,
      "num_input_tokens_seen": 11551904,
      "step": 54745
    },
    {
      "epoch": 6.023102310231023,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.026451882467278514,
      "loss": 0.2279,
      "num_input_tokens_seen": 11552896,
      "step": 54750
    },
    {
      "epoch": 6.023652365236524,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.026450952354907124,
      "loss": 0.2289,
      "num_input_tokens_seen": 11553952,
      "step": 54755
    },
    {
      "epoch": 6.024202420242024,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.026450022136997813,
      "loss": 0.2332,
      "num_input_tokens_seen": 11555008,
      "step": 54760
    },
    {
      "epoch": 6.024752475247524,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.026449091813559134,
      "loss": 0.2316,
      "num_input_tokens_seen": 11556000,
      "step": 54765
    },
    {
      "epoch": 6.025302530253025,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.02644816138459967,
      "loss": 0.2315,
      "num_input_tokens_seen": 11557088,
      "step": 54770
    },
    {
      "epoch": 6.025852585258526,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02644723085012799,
      "loss": 0.2315,
      "num_input_tokens_seen": 11558176,
      "step": 54775
    },
    {
      "epoch": 6.026402640264027,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.026446300210152686,
      "loss": 0.2343,
      "num_input_tokens_seen": 11559264,
      "step": 54780
    },
    {
      "epoch": 6.026952695269527,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.02644536946468232,
      "loss": 0.231,
      "num_input_tokens_seen": 11560320,
      "step": 54785
    },
    {
      "epoch": 6.027502750275027,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.026444438613725477,
      "loss": 0.231,
      "num_input_tokens_seen": 11561376,
      "step": 54790
    },
    {
      "epoch": 6.028052805280528,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.026443507657290737,
      "loss": 0.2341,
      "num_input_tokens_seen": 11562464,
      "step": 54795
    },
    {
      "epoch": 6.028602860286028,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.026442576595386673,
      "loss": 0.232,
      "num_input_tokens_seen": 11563520,
      "step": 54800
    },
    {
      "epoch": 6.0291529152915295,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.026441645428021875,
      "loss": 0.233,
      "num_input_tokens_seen": 11564544,
      "step": 54805
    },
    {
      "epoch": 6.02970297029703,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.026440714155204922,
      "loss": 0.2309,
      "num_input_tokens_seen": 11565632,
      "step": 54810
    },
    {
      "epoch": 6.03025302530253,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.026439782776944396,
      "loss": 0.232,
      "num_input_tokens_seen": 11566688,
      "step": 54815
    },
    {
      "epoch": 6.030803080308031,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02643885129324888,
      "loss": 0.233,
      "num_input_tokens_seen": 11567840,
      "step": 54820
    },
    {
      "epoch": 6.031353135313531,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.026437919704126962,
      "loss": 0.2309,
      "num_input_tokens_seen": 11568832,
      "step": 54825
    },
    {
      "epoch": 6.031903190319032,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02643698800958722,
      "loss": 0.2304,
      "num_input_tokens_seen": 11569952,
      "step": 54830
    },
    {
      "epoch": 6.0324532453245325,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.026436056209638254,
      "loss": 0.2303,
      "num_input_tokens_seen": 11571008,
      "step": 54835
    },
    {
      "epoch": 6.033003300330033,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.026435124304288647,
      "loss": 0.2314,
      "num_input_tokens_seen": 11572096,
      "step": 54840
    },
    {
      "epoch": 6.033553355335534,
      "grad_norm": 0.00078582763671875,
      "learning_rate": 0.026434192293546984,
      "loss": 0.233,
      "num_input_tokens_seen": 11573120,
      "step": 54845
    },
    {
      "epoch": 6.034103410341034,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.02643326017742186,
      "loss": 0.2314,
      "num_input_tokens_seen": 11574208,
      "step": 54850
    },
    {
      "epoch": 6.034653465346534,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.026432327955921858,
      "loss": 0.2308,
      "num_input_tokens_seen": 11575232,
      "step": 54855
    },
    {
      "epoch": 6.035203520352035,
      "grad_norm": 0.00081634521484375,
      "learning_rate": 0.026431395629055575,
      "loss": 0.233,
      "num_input_tokens_seen": 11576288,
      "step": 54860
    },
    {
      "epoch": 6.0357535753575355,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.02643046319683161,
      "loss": 0.2288,
      "num_input_tokens_seen": 11577344,
      "step": 54865
    },
    {
      "epoch": 6.036303630363037,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.026429530659258543,
      "loss": 0.2298,
      "num_input_tokens_seen": 11578368,
      "step": 54870
    },
    {
      "epoch": 6.036853685368537,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.026428598016344982,
      "loss": 0.2319,
      "num_input_tokens_seen": 11579424,
      "step": 54875
    },
    {
      "epoch": 6.037403740374037,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026427665268099514,
      "loss": 0.232,
      "num_input_tokens_seen": 11580480,
      "step": 54880
    },
    {
      "epoch": 6.037953795379538,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.02642673241453074,
      "loss": 0.2303,
      "num_input_tokens_seen": 11581600,
      "step": 54885
    },
    {
      "epoch": 6.038503850385038,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026425799455647256,
      "loss": 0.2293,
      "num_input_tokens_seen": 11582624,
      "step": 54890
    },
    {
      "epoch": 6.039053905390539,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.026424866391457663,
      "loss": 0.2299,
      "num_input_tokens_seen": 11583648,
      "step": 54895
    },
    {
      "epoch": 6.03960396039604,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.026423933221970557,
      "loss": 0.233,
      "num_input_tokens_seen": 11584672,
      "step": 54900
    },
    {
      "epoch": 6.04015401540154,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02642299994719454,
      "loss": 0.2299,
      "num_input_tokens_seen": 11585696,
      "step": 54905
    },
    {
      "epoch": 6.040704070407041,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.026422066567138216,
      "loss": 0.2288,
      "num_input_tokens_seen": 11586720,
      "step": 54910
    },
    {
      "epoch": 6.041254125412541,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.026421133081810183,
      "loss": 0.2324,
      "num_input_tokens_seen": 11587872,
      "step": 54915
    },
    {
      "epoch": 6.041804180418042,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.026420199491219043,
      "loss": 0.233,
      "num_input_tokens_seen": 11588960,
      "step": 54920
    },
    {
      "epoch": 6.042354235423542,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.02641926579537341,
      "loss": 0.232,
      "num_input_tokens_seen": 11589984,
      "step": 54925
    },
    {
      "epoch": 6.042904290429043,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026418331994281882,
      "loss": 0.2304,
      "num_input_tokens_seen": 11590976,
      "step": 54930
    },
    {
      "epoch": 6.043454345434544,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02641739808795307,
      "loss": 0.2319,
      "num_input_tokens_seen": 11592064,
      "step": 54935
    },
    {
      "epoch": 6.044004400440044,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.026416464076395575,
      "loss": 0.234,
      "num_input_tokens_seen": 11593120,
      "step": 54940
    },
    {
      "epoch": 6.044554455445544,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.026415529959618007,
      "loss": 0.2278,
      "num_input_tokens_seen": 11594176,
      "step": 54945
    },
    {
      "epoch": 6.045104510451045,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02641459573762898,
      "loss": 0.2315,
      "num_input_tokens_seen": 11595200,
      "step": 54950
    },
    {
      "epoch": 6.0456545654565454,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.0264136614104371,
      "loss": 0.233,
      "num_input_tokens_seen": 11596288,
      "step": 54955
    },
    {
      "epoch": 6.0462046204620465,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.026412726978050978,
      "loss": 0.2299,
      "num_input_tokens_seen": 11597376,
      "step": 54960
    },
    {
      "epoch": 6.046754675467547,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02641179244047923,
      "loss": 0.231,
      "num_input_tokens_seen": 11598432,
      "step": 54965
    },
    {
      "epoch": 6.047304730473047,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.026410857797730468,
      "loss": 0.2314,
      "num_input_tokens_seen": 11599520,
      "step": 54970
    },
    {
      "epoch": 6.047854785478548,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.026409923049813305,
      "loss": 0.232,
      "num_input_tokens_seen": 11600480,
      "step": 54975
    },
    {
      "epoch": 6.048404840484048,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.026408988196736354,
      "loss": 0.2299,
      "num_input_tokens_seen": 11601472,
      "step": 54980
    },
    {
      "epoch": 6.048954895489549,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.026408053238508238,
      "loss": 0.2304,
      "num_input_tokens_seen": 11602528,
      "step": 54985
    },
    {
      "epoch": 6.0495049504950495,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.026407118175137564,
      "loss": 0.2304,
      "num_input_tokens_seen": 11603648,
      "step": 54990
    },
    {
      "epoch": 6.05005500550055,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02640618300663296,
      "loss": 0.2283,
      "num_input_tokens_seen": 11604704,
      "step": 54995
    },
    {
      "epoch": 6.050605060506051,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.026405247733003036,
      "loss": 0.2331,
      "num_input_tokens_seen": 11605792,
      "step": 55000
    },
    {
      "epoch": 6.051155115511551,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02640431235425642,
      "loss": 0.2325,
      "num_input_tokens_seen": 11606816,
      "step": 55005
    },
    {
      "epoch": 6.051705170517051,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.02640337687040173,
      "loss": 0.2315,
      "num_input_tokens_seen": 11607872,
      "step": 55010
    },
    {
      "epoch": 6.052255225522552,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.026402441281447583,
      "loss": 0.2346,
      "num_input_tokens_seen": 11608896,
      "step": 55015
    },
    {
      "epoch": 6.052805280528053,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.026401505587402608,
      "loss": 0.232,
      "num_input_tokens_seen": 11609920,
      "step": 55020
    },
    {
      "epoch": 6.053355335533554,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.026400569788275423,
      "loss": 0.232,
      "num_input_tokens_seen": 11610912,
      "step": 55025
    },
    {
      "epoch": 6.053905390539054,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.026399633884074666,
      "loss": 0.2273,
      "num_input_tokens_seen": 11611936,
      "step": 55030
    },
    {
      "epoch": 6.054455445544554,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.026398697874808946,
      "loss": 0.2319,
      "num_input_tokens_seen": 11612960,
      "step": 55035
    },
    {
      "epoch": 6.055005500550055,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.026397761760486898,
      "loss": 0.2288,
      "num_input_tokens_seen": 11613984,
      "step": 55040
    },
    {
      "epoch": 6.055555555555555,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.026396825541117157,
      "loss": 0.2319,
      "num_input_tokens_seen": 11615040,
      "step": 55045
    },
    {
      "epoch": 6.0561056105610565,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.026395889216708338,
      "loss": 0.2299,
      "num_input_tokens_seen": 11616064,
      "step": 55050
    },
    {
      "epoch": 6.056655665566557,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.026394952787269072,
      "loss": 0.2309,
      "num_input_tokens_seen": 11617120,
      "step": 55055
    },
    {
      "epoch": 6.057205720572057,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.026394016252808,
      "loss": 0.2294,
      "num_input_tokens_seen": 11618144,
      "step": 55060
    },
    {
      "epoch": 6.057755775577558,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02639307961333374,
      "loss": 0.2336,
      "num_input_tokens_seen": 11619232,
      "step": 55065
    },
    {
      "epoch": 6.058305830583058,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.026392142868854943,
      "loss": 0.231,
      "num_input_tokens_seen": 11620288,
      "step": 55070
    },
    {
      "epoch": 6.058855885588559,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.026391206019380227,
      "loss": 0.2346,
      "num_input_tokens_seen": 11621280,
      "step": 55075
    },
    {
      "epoch": 6.0594059405940595,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.026390269064918228,
      "loss": 0.2294,
      "num_input_tokens_seen": 11622304,
      "step": 55080
    },
    {
      "epoch": 6.05995599559956,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02638933200547759,
      "loss": 0.2294,
      "num_input_tokens_seen": 11623456,
      "step": 55085
    },
    {
      "epoch": 6.060506050605061,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.026388394841066937,
      "loss": 0.232,
      "num_input_tokens_seen": 11624544,
      "step": 55090
    },
    {
      "epoch": 6.061056105610561,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.026387457571694917,
      "loss": 0.2279,
      "num_input_tokens_seen": 11625664,
      "step": 55095
    },
    {
      "epoch": 6.061606160616061,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.02638652019737017,
      "loss": 0.2337,
      "num_input_tokens_seen": 11626688,
      "step": 55100
    },
    {
      "epoch": 6.062156215621562,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.026385582718101323,
      "loss": 0.2296,
      "num_input_tokens_seen": 11627744,
      "step": 55105
    },
    {
      "epoch": 6.0627062706270625,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.026384645133897027,
      "loss": 0.2285,
      "num_input_tokens_seen": 11628832,
      "step": 55110
    },
    {
      "epoch": 6.063256325632564,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.026383707444765918,
      "loss": 0.2332,
      "num_input_tokens_seen": 11629952,
      "step": 55115
    },
    {
      "epoch": 6.063806380638064,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.026382769650716644,
      "loss": 0.2286,
      "num_input_tokens_seen": 11631040,
      "step": 55120
    },
    {
      "epoch": 6.064356435643564,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.026381831751757834,
      "loss": 0.2336,
      "num_input_tokens_seen": 11632160,
      "step": 55125
    },
    {
      "epoch": 6.064906490649065,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.026380893747898156,
      "loss": 0.2265,
      "num_input_tokens_seen": 11633216,
      "step": 55130
    },
    {
      "epoch": 6.065456545654565,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.026379955639146233,
      "loss": 0.2281,
      "num_input_tokens_seen": 11634272,
      "step": 55135
    },
    {
      "epoch": 6.066006600660066,
      "grad_norm": 0.005859375,
      "learning_rate": 0.026379017425510718,
      "loss": 0.2322,
      "num_input_tokens_seen": 11635360,
      "step": 55140
    },
    {
      "epoch": 6.066556655665567,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.02637807910700026,
      "loss": 0.2293,
      "num_input_tokens_seen": 11636384,
      "step": 55145
    },
    {
      "epoch": 6.067106710671067,
      "grad_norm": 0.01171875,
      "learning_rate": 0.026377140683623512,
      "loss": 0.2342,
      "num_input_tokens_seen": 11637472,
      "step": 55150
    },
    {
      "epoch": 6.067656765676568,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026376202155389113,
      "loss": 0.2313,
      "num_input_tokens_seen": 11638496,
      "step": 55155
    },
    {
      "epoch": 6.068206820682068,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.026375263522305722,
      "loss": 0.2327,
      "num_input_tokens_seen": 11639616,
      "step": 55160
    },
    {
      "epoch": 6.068756875687569,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.02637432478438198,
      "loss": 0.2301,
      "num_input_tokens_seen": 11640672,
      "step": 55165
    },
    {
      "epoch": 6.069306930693069,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02637338594162655,
      "loss": 0.2331,
      "num_input_tokens_seen": 11641696,
      "step": 55170
    },
    {
      "epoch": 6.06985698569857,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.026372446994048074,
      "loss": 0.2258,
      "num_input_tokens_seen": 11642720,
      "step": 55175
    },
    {
      "epoch": 6.070407040704071,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.026371507941655215,
      "loss": 0.2315,
      "num_input_tokens_seen": 11643744,
      "step": 55180
    },
    {
      "epoch": 6.070957095709571,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.02637056878445662,
      "loss": 0.2316,
      "num_input_tokens_seen": 11644768,
      "step": 55185
    },
    {
      "epoch": 6.071507150715071,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.02636962952246095,
      "loss": 0.2357,
      "num_input_tokens_seen": 11645856,
      "step": 55190
    },
    {
      "epoch": 6.072057205720572,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02636869015567686,
      "loss": 0.2315,
      "num_input_tokens_seen": 11646912,
      "step": 55195
    },
    {
      "epoch": 6.072607260726072,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.026367750684113015,
      "loss": 0.2301,
      "num_input_tokens_seen": 11647968,
      "step": 55200
    },
    {
      "epoch": 6.0731573157315735,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.026366811107778058,
      "loss": 0.2341,
      "num_input_tokens_seen": 11649056,
      "step": 55205
    },
    {
      "epoch": 6.073707370737074,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.026365871426680662,
      "loss": 0.229,
      "num_input_tokens_seen": 11650080,
      "step": 55210
    },
    {
      "epoch": 6.074257425742574,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.02636493164082948,
      "loss": 0.2325,
      "num_input_tokens_seen": 11651200,
      "step": 55215
    },
    {
      "epoch": 6.074807480748075,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.026363991750233177,
      "loss": 0.2305,
      "num_input_tokens_seen": 11652192,
      "step": 55220
    },
    {
      "epoch": 6.075357535753575,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02636305175490042,
      "loss": 0.2347,
      "num_input_tokens_seen": 11653216,
      "step": 55225
    },
    {
      "epoch": 6.075907590759076,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.02636211165483986,
      "loss": 0.232,
      "num_input_tokens_seen": 11654272,
      "step": 55230
    },
    {
      "epoch": 6.0764576457645765,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.026361171450060176,
      "loss": 0.2305,
      "num_input_tokens_seen": 11655360,
      "step": 55235
    },
    {
      "epoch": 6.077007700770077,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.02636023114057002,
      "loss": 0.2311,
      "num_input_tokens_seen": 11656320,
      "step": 55240
    },
    {
      "epoch": 6.077557755775578,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.026359290726378066,
      "loss": 0.232,
      "num_input_tokens_seen": 11657344,
      "step": 55245
    },
    {
      "epoch": 6.078107810781078,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.026358350207492987,
      "loss": 0.2294,
      "num_input_tokens_seen": 11658400,
      "step": 55250
    },
    {
      "epoch": 6.078657865786579,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.026357409583923436,
      "loss": 0.231,
      "num_input_tokens_seen": 11659392,
      "step": 55255
    },
    {
      "epoch": 6.079207920792079,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.026356468855678092,
      "loss": 0.2304,
      "num_input_tokens_seen": 11660512,
      "step": 55260
    },
    {
      "epoch": 6.0797579757975795,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.026355528022765626,
      "loss": 0.2315,
      "num_input_tokens_seen": 11661568,
      "step": 55265
    },
    {
      "epoch": 6.080308030803081,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.026354587085194708,
      "loss": 0.231,
      "num_input_tokens_seen": 11662688,
      "step": 55270
    },
    {
      "epoch": 6.080858085808581,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.026353646042974004,
      "loss": 0.2357,
      "num_input_tokens_seen": 11663712,
      "step": 55275
    },
    {
      "epoch": 6.081408140814081,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.026352704896112203,
      "loss": 0.2341,
      "num_input_tokens_seen": 11664768,
      "step": 55280
    },
    {
      "epoch": 6.081958195819582,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02635176364461796,
      "loss": 0.2309,
      "num_input_tokens_seen": 11665856,
      "step": 55285
    },
    {
      "epoch": 6.082508250825082,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.02635082228849996,
      "loss": 0.2319,
      "num_input_tokens_seen": 11666880,
      "step": 55290
    },
    {
      "epoch": 6.083058305830583,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.02634988082776688,
      "loss": 0.2351,
      "num_input_tokens_seen": 11667872,
      "step": 55295
    },
    {
      "epoch": 6.083608360836084,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02634893926242739,
      "loss": 0.2314,
      "num_input_tokens_seen": 11668928,
      "step": 55300
    },
    {
      "epoch": 6.084158415841584,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.026347997592490176,
      "loss": 0.233,
      "num_input_tokens_seen": 11670016,
      "step": 55305
    },
    {
      "epoch": 6.084708470847085,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.026347055817963913,
      "loss": 0.2366,
      "num_input_tokens_seen": 11671040,
      "step": 55310
    },
    {
      "epoch": 6.085258525852585,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026346113938857278,
      "loss": 0.2319,
      "num_input_tokens_seen": 11672128,
      "step": 55315
    },
    {
      "epoch": 6.085808580858086,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02634517195517896,
      "loss": 0.2324,
      "num_input_tokens_seen": 11673152,
      "step": 55320
    },
    {
      "epoch": 6.086358635863586,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.026344229866937633,
      "loss": 0.2313,
      "num_input_tokens_seen": 11674176,
      "step": 55325
    },
    {
      "epoch": 6.086908690869087,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.026343287674141985,
      "loss": 0.2298,
      "num_input_tokens_seen": 11675264,
      "step": 55330
    },
    {
      "epoch": 6.087458745874588,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.026342345376800698,
      "loss": 0.2314,
      "num_input_tokens_seen": 11676320,
      "step": 55335
    },
    {
      "epoch": 6.088008800880088,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.02634140297492245,
      "loss": 0.2314,
      "num_input_tokens_seen": 11677312,
      "step": 55340
    },
    {
      "epoch": 6.088558855885589,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.026340460468515944,
      "loss": 0.2329,
      "num_input_tokens_seen": 11678368,
      "step": 55345
    },
    {
      "epoch": 6.089108910891089,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.026339517857589846,
      "loss": 0.2303,
      "num_input_tokens_seen": 11679392,
      "step": 55350
    },
    {
      "epoch": 6.089658965896589,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.02633857514215286,
      "loss": 0.2314,
      "num_input_tokens_seen": 11680448,
      "step": 55355
    },
    {
      "epoch": 6.0902090209020905,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.026337632322213665,
      "loss": 0.2324,
      "num_input_tokens_seen": 11681472,
      "step": 55360
    },
    {
      "epoch": 6.090759075907591,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.02633668939778095,
      "loss": 0.2313,
      "num_input_tokens_seen": 11682528,
      "step": 55365
    },
    {
      "epoch": 6.091309130913091,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02633574636886341,
      "loss": 0.2335,
      "num_input_tokens_seen": 11683584,
      "step": 55370
    },
    {
      "epoch": 6.091859185918592,
      "grad_norm": 0.000629425048828125,
      "learning_rate": 0.02633480323546974,
      "loss": 0.2314,
      "num_input_tokens_seen": 11684576,
      "step": 55375
    },
    {
      "epoch": 6.092409240924092,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.026333859997608622,
      "loss": 0.2314,
      "num_input_tokens_seen": 11685632,
      "step": 55380
    },
    {
      "epoch": 6.092959295929593,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.02633291665528876,
      "loss": 0.2319,
      "num_input_tokens_seen": 11686720,
      "step": 55385
    },
    {
      "epoch": 6.0935093509350935,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.026331973208518836,
      "loss": 0.2319,
      "num_input_tokens_seen": 11687808,
      "step": 55390
    },
    {
      "epoch": 6.094059405940594,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.02633102965730756,
      "loss": 0.2324,
      "num_input_tokens_seen": 11688832,
      "step": 55395
    },
    {
      "epoch": 6.094609460946095,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.026330086001663613,
      "loss": 0.2298,
      "num_input_tokens_seen": 11689888,
      "step": 55400
    },
    {
      "epoch": 6.095159515951595,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.026329142241595705,
      "loss": 0.2314,
      "num_input_tokens_seen": 11690976,
      "step": 55405
    },
    {
      "epoch": 6.095709570957096,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02632819837711253,
      "loss": 0.2309,
      "num_input_tokens_seen": 11691968,
      "step": 55410
    },
    {
      "epoch": 6.096259625962596,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.026327254408222787,
      "loss": 0.2329,
      "num_input_tokens_seen": 11693024,
      "step": 55415
    },
    {
      "epoch": 6.0968096809680965,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.026326310334935172,
      "loss": 0.2308,
      "num_input_tokens_seen": 11694144,
      "step": 55420
    },
    {
      "epoch": 6.097359735973598,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02632536615725839,
      "loss": 0.2329,
      "num_input_tokens_seen": 11695232,
      "step": 55425
    },
    {
      "epoch": 6.097909790979098,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.026324421875201145,
      "loss": 0.2329,
      "num_input_tokens_seen": 11696320,
      "step": 55430
    },
    {
      "epoch": 6.098459845984599,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.026323477488772135,
      "loss": 0.2308,
      "num_input_tokens_seen": 11697376,
      "step": 55435
    },
    {
      "epoch": 6.099009900990099,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02632253299798007,
      "loss": 0.2303,
      "num_input_tokens_seen": 11698432,
      "step": 55440
    },
    {
      "epoch": 6.099559955995599,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02632158840283365,
      "loss": 0.2308,
      "num_input_tokens_seen": 11699520,
      "step": 55445
    },
    {
      "epoch": 6.1001100110011,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02632064370334158,
      "loss": 0.2313,
      "num_input_tokens_seen": 11700544,
      "step": 55450
    },
    {
      "epoch": 6.100660066006601,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.026319698899512568,
      "loss": 0.2324,
      "num_input_tokens_seen": 11701600,
      "step": 55455
    },
    {
      "epoch": 6.101210121012101,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.026318753991355328,
      "loss": 0.2319,
      "num_input_tokens_seen": 11702688,
      "step": 55460
    },
    {
      "epoch": 6.101760176017602,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02631780897887856,
      "loss": 0.2315,
      "num_input_tokens_seen": 11703680,
      "step": 55465
    },
    {
      "epoch": 6.102310231023102,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.026316863862090978,
      "loss": 0.2319,
      "num_input_tokens_seen": 11704640,
      "step": 55470
    },
    {
      "epoch": 6.102860286028603,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.026315918641001294,
      "loss": 0.2351,
      "num_input_tokens_seen": 11705696,
      "step": 55475
    },
    {
      "epoch": 6.103410341034103,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.026314973315618214,
      "loss": 0.2319,
      "num_input_tokens_seen": 11706784,
      "step": 55480
    },
    {
      "epoch": 6.103960396039604,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.02631402788595046,
      "loss": 0.2308,
      "num_input_tokens_seen": 11707808,
      "step": 55485
    },
    {
      "epoch": 6.104510451045105,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.026313082352006738,
      "loss": 0.2334,
      "num_input_tokens_seen": 11708864,
      "step": 55490
    },
    {
      "epoch": 6.105060506050605,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.026312136713795762,
      "loss": 0.2313,
      "num_input_tokens_seen": 11709952,
      "step": 55495
    },
    {
      "epoch": 6.105610561056106,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02631119097132625,
      "loss": 0.2308,
      "num_input_tokens_seen": 11710976,
      "step": 55500
    },
    {
      "epoch": 6.106160616061606,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02631024512460692,
      "loss": 0.2324,
      "num_input_tokens_seen": 11712000,
      "step": 55505
    },
    {
      "epoch": 6.106710671067106,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.026309299173646487,
      "loss": 0.2308,
      "num_input_tokens_seen": 11713088,
      "step": 55510
    },
    {
      "epoch": 6.1072607260726075,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.026308353118453674,
      "loss": 0.2303,
      "num_input_tokens_seen": 11714144,
      "step": 55515
    },
    {
      "epoch": 6.107810781078108,
      "grad_norm": 0.009765625,
      "learning_rate": 0.026307406959037195,
      "loss": 0.2298,
      "num_input_tokens_seen": 11715232,
      "step": 55520
    },
    {
      "epoch": 6.108360836083609,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02630646069540577,
      "loss": 0.2314,
      "num_input_tokens_seen": 11716288,
      "step": 55525
    },
    {
      "epoch": 6.108910891089109,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02630551432756812,
      "loss": 0.2319,
      "num_input_tokens_seen": 11717408,
      "step": 55530
    },
    {
      "epoch": 6.109460946094609,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.026304567855532975,
      "loss": 0.2313,
      "num_input_tokens_seen": 11718400,
      "step": 55535
    },
    {
      "epoch": 6.11001100110011,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026303621279309047,
      "loss": 0.2324,
      "num_input_tokens_seen": 11719392,
      "step": 55540
    },
    {
      "epoch": 6.1105610561056105,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02630267459890507,
      "loss": 0.2319,
      "num_input_tokens_seen": 11720448,
      "step": 55545
    },
    {
      "epoch": 6.111111111111111,
      "grad_norm": 0.000446319580078125,
      "learning_rate": 0.026301727814329764,
      "loss": 0.2329,
      "num_input_tokens_seen": 11721504,
      "step": 55550
    },
    {
      "epoch": 6.111661166116612,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.026300780925591852,
      "loss": 0.2298,
      "num_input_tokens_seen": 11722592,
      "step": 55555
    },
    {
      "epoch": 6.112211221122112,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.026299833932700068,
      "loss": 0.2319,
      "num_input_tokens_seen": 11723616,
      "step": 55560
    },
    {
      "epoch": 6.112761276127613,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.02629888683566314,
      "loss": 0.2309,
      "num_input_tokens_seen": 11724768,
      "step": 55565
    },
    {
      "epoch": 6.113311331133113,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.02629793963448979,
      "loss": 0.2303,
      "num_input_tokens_seen": 11725888,
      "step": 55570
    },
    {
      "epoch": 6.1138613861386135,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02629699232918875,
      "loss": 0.2309,
      "num_input_tokens_seen": 11726976,
      "step": 55575
    },
    {
      "epoch": 6.114411441144115,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.026296044919768755,
      "loss": 0.2303,
      "num_input_tokens_seen": 11728032,
      "step": 55580
    },
    {
      "epoch": 6.114961496149615,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.026295097406238536,
      "loss": 0.2313,
      "num_input_tokens_seen": 11729152,
      "step": 55585
    },
    {
      "epoch": 6.115511551155116,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.026294149788606818,
      "loss": 0.2314,
      "num_input_tokens_seen": 11730240,
      "step": 55590
    },
    {
      "epoch": 6.116061606160616,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02629320206688235,
      "loss": 0.2324,
      "num_input_tokens_seen": 11731232,
      "step": 55595
    },
    {
      "epoch": 6.116611661166116,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02629225424107385,
      "loss": 0.2335,
      "num_input_tokens_seen": 11732256,
      "step": 55600
    },
    {
      "epoch": 6.117161716171617,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.026291306311190064,
      "loss": 0.234,
      "num_input_tokens_seen": 11733376,
      "step": 55605
    },
    {
      "epoch": 6.117711771177118,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.026290358277239724,
      "loss": 0.234,
      "num_input_tokens_seen": 11734400,
      "step": 55610
    },
    {
      "epoch": 6.118261826182618,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.026289410139231573,
      "loss": 0.2303,
      "num_input_tokens_seen": 11735456,
      "step": 55615
    },
    {
      "epoch": 6.118811881188119,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.026288461897174344,
      "loss": 0.2303,
      "num_input_tokens_seen": 11736512,
      "step": 55620
    },
    {
      "epoch": 6.119361936193619,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02628751355107678,
      "loss": 0.2319,
      "num_input_tokens_seen": 11737568,
      "step": 55625
    },
    {
      "epoch": 6.11991199119912,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.026286565100947613,
      "loss": 0.2314,
      "num_input_tokens_seen": 11738656,
      "step": 55630
    },
    {
      "epoch": 6.12046204620462,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.0262856165467956,
      "loss": 0.2309,
      "num_input_tokens_seen": 11739648,
      "step": 55635
    },
    {
      "epoch": 6.121012101210121,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02628466788862947,
      "loss": 0.2309,
      "num_input_tokens_seen": 11740704,
      "step": 55640
    },
    {
      "epoch": 6.121562156215622,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.02628371912645797,
      "loss": 0.2298,
      "num_input_tokens_seen": 11741728,
      "step": 55645
    },
    {
      "epoch": 6.122112211221122,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02628277026028985,
      "loss": 0.2298,
      "num_input_tokens_seen": 11742784,
      "step": 55650
    },
    {
      "epoch": 6.122662266226623,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.026281821290133848,
      "loss": 0.2309,
      "num_input_tokens_seen": 11743840,
      "step": 55655
    },
    {
      "epoch": 6.123212321232123,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.02628087221599871,
      "loss": 0.2314,
      "num_input_tokens_seen": 11744896,
      "step": 55660
    },
    {
      "epoch": 6.123762376237623,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02627992303789319,
      "loss": 0.2309,
      "num_input_tokens_seen": 11745984,
      "step": 55665
    },
    {
      "epoch": 6.1243124312431245,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.026278973755826027,
      "loss": 0.2319,
      "num_input_tokens_seen": 11747072,
      "step": 55670
    },
    {
      "epoch": 6.124862486248625,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.02627802436980598,
      "loss": 0.2283,
      "num_input_tokens_seen": 11748096,
      "step": 55675
    },
    {
      "epoch": 6.125412541254126,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.026277074879841794,
      "loss": 0.233,
      "num_input_tokens_seen": 11749184,
      "step": 55680
    },
    {
      "epoch": 6.125962596259626,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.026276125285942215,
      "loss": 0.2325,
      "num_input_tokens_seen": 11750272,
      "step": 55685
    },
    {
      "epoch": 6.126512651265126,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.026275175588116005,
      "loss": 0.2283,
      "num_input_tokens_seen": 11751296,
      "step": 55690
    },
    {
      "epoch": 6.127062706270627,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.026274225786371914,
      "loss": 0.2304,
      "num_input_tokens_seen": 11752384,
      "step": 55695
    },
    {
      "epoch": 6.1276127612761275,
      "grad_norm": 0.00457763671875,
      "learning_rate": 0.02627327588071869,
      "loss": 0.231,
      "num_input_tokens_seen": 11753376,
      "step": 55700
    },
    {
      "epoch": 6.128162816281628,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.026272325871165087,
      "loss": 0.2367,
      "num_input_tokens_seen": 11754400,
      "step": 55705
    },
    {
      "epoch": 6.128712871287129,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.026271375757719876,
      "loss": 0.2294,
      "num_input_tokens_seen": 11755488,
      "step": 55710
    },
    {
      "epoch": 6.129262926292629,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.026270425540391794,
      "loss": 0.232,
      "num_input_tokens_seen": 11756544,
      "step": 55715
    },
    {
      "epoch": 6.12981298129813,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.026269475219189615,
      "loss": 0.2336,
      "num_input_tokens_seen": 11757632,
      "step": 55720
    },
    {
      "epoch": 6.13036303630363,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.026268524794122085,
      "loss": 0.2299,
      "num_input_tokens_seen": 11758624,
      "step": 55725
    },
    {
      "epoch": 6.1309130913091305,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.026267574265197973,
      "loss": 0.2315,
      "num_input_tokens_seen": 11759680,
      "step": 55730
    },
    {
      "epoch": 6.131463146314632,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026266623632426035,
      "loss": 0.233,
      "num_input_tokens_seen": 11760704,
      "step": 55735
    },
    {
      "epoch": 6.132013201320132,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.026265672895815034,
      "loss": 0.2299,
      "num_input_tokens_seen": 11761760,
      "step": 55740
    },
    {
      "epoch": 6.132563256325633,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.02626472205537373,
      "loss": 0.232,
      "num_input_tokens_seen": 11762880,
      "step": 55745
    },
    {
      "epoch": 6.133113311331133,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.026263771111110883,
      "loss": 0.2304,
      "num_input_tokens_seen": 11763968,
      "step": 55750
    },
    {
      "epoch": 6.133663366336633,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02626282006303527,
      "loss": 0.232,
      "num_input_tokens_seen": 11765056,
      "step": 55755
    },
    {
      "epoch": 6.134213421342134,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.026261868911155645,
      "loss": 0.2299,
      "num_input_tokens_seen": 11766144,
      "step": 55760
    },
    {
      "epoch": 6.134763476347635,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.026260917655480782,
      "loss": 0.2336,
      "num_input_tokens_seen": 11767168,
      "step": 55765
    },
    {
      "epoch": 6.135313531353136,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026259966296019437,
      "loss": 0.2352,
      "num_input_tokens_seen": 11768224,
      "step": 55770
    },
    {
      "epoch": 6.135863586358636,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.026259014832780394,
      "loss": 0.2315,
      "num_input_tokens_seen": 11769216,
      "step": 55775
    },
    {
      "epoch": 6.136413641364136,
      "grad_norm": 0.00057220458984375,
      "learning_rate": 0.026258063265772413,
      "loss": 0.232,
      "num_input_tokens_seen": 11770272,
      "step": 55780
    },
    {
      "epoch": 6.136963696369637,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.02625711159500426,
      "loss": 0.232,
      "num_input_tokens_seen": 11771264,
      "step": 55785
    },
    {
      "epoch": 6.137513751375137,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.026256159820484716,
      "loss": 0.2294,
      "num_input_tokens_seen": 11772320,
      "step": 55790
    },
    {
      "epoch": 6.138063806380638,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.026255207942222546,
      "loss": 0.2304,
      "num_input_tokens_seen": 11773440,
      "step": 55795
    },
    {
      "epoch": 6.138613861386139,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.026254255960226527,
      "loss": 0.232,
      "num_input_tokens_seen": 11774528,
      "step": 55800
    },
    {
      "epoch": 6.139163916391639,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02625330387450543,
      "loss": 0.2315,
      "num_input_tokens_seen": 11775616,
      "step": 55805
    },
    {
      "epoch": 6.13971397139714,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.026252351685068027,
      "loss": 0.2314,
      "num_input_tokens_seen": 11776608,
      "step": 55810
    },
    {
      "epoch": 6.14026402640264,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.026251399391923103,
      "loss": 0.232,
      "num_input_tokens_seen": 11777664,
      "step": 55815
    },
    {
      "epoch": 6.1408140814081404,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.026250446995079425,
      "loss": 0.2278,
      "num_input_tokens_seen": 11778784,
      "step": 55820
    },
    {
      "epoch": 6.1413641364136415,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02624949449454578,
      "loss": 0.2309,
      "num_input_tokens_seen": 11779776,
      "step": 55825
    },
    {
      "epoch": 6.141914191419142,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.02624854189033094,
      "loss": 0.2304,
      "num_input_tokens_seen": 11780800,
      "step": 55830
    },
    {
      "epoch": 6.142464246424643,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.02624758918244369,
      "loss": 0.2309,
      "num_input_tokens_seen": 11781824,
      "step": 55835
    },
    {
      "epoch": 6.143014301430143,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.026246636370892806,
      "loss": 0.233,
      "num_input_tokens_seen": 11782880,
      "step": 55840
    },
    {
      "epoch": 6.143564356435643,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02624568345568707,
      "loss": 0.2283,
      "num_input_tokens_seen": 11784000,
      "step": 55845
    },
    {
      "epoch": 6.144114411441144,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.026244730436835267,
      "loss": 0.2304,
      "num_input_tokens_seen": 11785056,
      "step": 55850
    },
    {
      "epoch": 6.1446644664466445,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.02624377731434618,
      "loss": 0.2325,
      "num_input_tokens_seen": 11786080,
      "step": 55855
    },
    {
      "epoch": 6.145214521452146,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02624282408822859,
      "loss": 0.2325,
      "num_input_tokens_seen": 11787104,
      "step": 55860
    },
    {
      "epoch": 6.145764576457646,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.026241870758491284,
      "loss": 0.232,
      "num_input_tokens_seen": 11788096,
      "step": 55865
    },
    {
      "epoch": 6.146314631463146,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.026240917325143057,
      "loss": 0.2294,
      "num_input_tokens_seen": 11789120,
      "step": 55870
    },
    {
      "epoch": 6.146864686468647,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.026239963788192683,
      "loss": 0.2336,
      "num_input_tokens_seen": 11790240,
      "step": 55875
    },
    {
      "epoch": 6.147414741474147,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.02623901014764896,
      "loss": 0.2331,
      "num_input_tokens_seen": 11791328,
      "step": 55880
    },
    {
      "epoch": 6.1479647964796476,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.026238056403520667,
      "loss": 0.2293,
      "num_input_tokens_seen": 11792320,
      "step": 55885
    },
    {
      "epoch": 6.148514851485149,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026237102555816606,
      "loss": 0.2288,
      "num_input_tokens_seen": 11793376,
      "step": 55890
    },
    {
      "epoch": 6.149064906490649,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.026236148604545564,
      "loss": 0.231,
      "num_input_tokens_seen": 11794400,
      "step": 55895
    },
    {
      "epoch": 6.14961496149615,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.026235194549716328,
      "loss": 0.2299,
      "num_input_tokens_seen": 11795488,
      "step": 55900
    },
    {
      "epoch": 6.15016501650165,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.026234240391337697,
      "loss": 0.2263,
      "num_input_tokens_seen": 11796576,
      "step": 55905
    },
    {
      "epoch": 6.15071507150715,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.026233286129418464,
      "loss": 0.2311,
      "num_input_tokens_seen": 11797632,
      "step": 55910
    },
    {
      "epoch": 6.1512651265126514,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.026232331763967418,
      "loss": 0.2275,
      "num_input_tokens_seen": 11798688,
      "step": 55915
    },
    {
      "epoch": 6.151815181518152,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.026231377294993367,
      "loss": 0.2292,
      "num_input_tokens_seen": 11799808,
      "step": 55920
    },
    {
      "epoch": 6.152365236523653,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.026230422722505094,
      "loss": 0.2377,
      "num_input_tokens_seen": 11800896,
      "step": 55925
    },
    {
      "epoch": 6.152915291529153,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.026229468046511408,
      "loss": 0.2304,
      "num_input_tokens_seen": 11801920,
      "step": 55930
    },
    {
      "epoch": 6.153465346534653,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.026228513267021098,
      "loss": 0.236,
      "num_input_tokens_seen": 11802944,
      "step": 55935
    },
    {
      "epoch": 6.154015401540154,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.02622755838404298,
      "loss": 0.2283,
      "num_input_tokens_seen": 11804032,
      "step": 55940
    },
    {
      "epoch": 6.1545654565456545,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.026226603397585833,
      "loss": 0.235,
      "num_input_tokens_seen": 11805056,
      "step": 55945
    },
    {
      "epoch": 6.1551155115511555,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.026225648307658476,
      "loss": 0.2339,
      "num_input_tokens_seen": 11806176,
      "step": 55950
    },
    {
      "epoch": 6.155665566556656,
      "grad_norm": 0.00457763671875,
      "learning_rate": 0.026224693114269705,
      "loss": 0.2318,
      "num_input_tokens_seen": 11807232,
      "step": 55955
    },
    {
      "epoch": 6.156215621562156,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.026223737817428323,
      "loss": 0.2308,
      "num_input_tokens_seen": 11808288,
      "step": 55960
    },
    {
      "epoch": 6.156765676567657,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.02622278241714313,
      "loss": 0.2229,
      "num_input_tokens_seen": 11809312,
      "step": 55965
    },
    {
      "epoch": 6.157315731573157,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.026221826913422945,
      "loss": 0.2307,
      "num_input_tokens_seen": 11810400,
      "step": 55970
    },
    {
      "epoch": 6.1578657865786575,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.026220871306276556,
      "loss": 0.2361,
      "num_input_tokens_seen": 11811488,
      "step": 55975
    },
    {
      "epoch": 6.158415841584159,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.026219915595712788,
      "loss": 0.2303,
      "num_input_tokens_seen": 11812480,
      "step": 55980
    },
    {
      "epoch": 6.158965896589659,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02621895978174044,
      "loss": 0.235,
      "num_input_tokens_seen": 11813536,
      "step": 55985
    },
    {
      "epoch": 6.15951595159516,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.026218003864368324,
      "loss": 0.2307,
      "num_input_tokens_seen": 11814624,
      "step": 55990
    },
    {
      "epoch": 6.16006600660066,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02621704784360525,
      "loss": 0.2344,
      "num_input_tokens_seen": 11815680,
      "step": 55995
    },
    {
      "epoch": 6.16061606160616,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.026216091719460027,
      "loss": 0.2306,
      "num_input_tokens_seen": 11816800,
      "step": 56000
    },
    {
      "epoch": 6.161166116611661,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.02621513549194147,
      "loss": 0.2297,
      "num_input_tokens_seen": 11817856,
      "step": 56005
    },
    {
      "epoch": 6.161716171617162,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.026214179161058386,
      "loss": 0.2321,
      "num_input_tokens_seen": 11818976,
      "step": 56010
    },
    {
      "epoch": 6.162266226622663,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0262132227268196,
      "loss": 0.2343,
      "num_input_tokens_seen": 11820064,
      "step": 56015
    },
    {
      "epoch": 6.162816281628163,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.026212266189233918,
      "loss": 0.2275,
      "num_input_tokens_seen": 11821152,
      "step": 56020
    },
    {
      "epoch": 6.163366336633663,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.026211309548310153,
      "loss": 0.2317,
      "num_input_tokens_seen": 11822176,
      "step": 56025
    },
    {
      "epoch": 6.163916391639164,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.026210352804057136,
      "loss": 0.2275,
      "num_input_tokens_seen": 11823232,
      "step": 56030
    },
    {
      "epoch": 6.164466446644664,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.026209395956483673,
      "loss": 0.2301,
      "num_input_tokens_seen": 11824288,
      "step": 56035
    },
    {
      "epoch": 6.165016501650165,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.026208439005598588,
      "loss": 0.2287,
      "num_input_tokens_seen": 11825344,
      "step": 56040
    },
    {
      "epoch": 6.165566556655666,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.026207481951410698,
      "loss": 0.2353,
      "num_input_tokens_seen": 11826432,
      "step": 56045
    },
    {
      "epoch": 6.166116611661166,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.026206524793928824,
      "loss": 0.227,
      "num_input_tokens_seen": 11827520,
      "step": 56050
    },
    {
      "epoch": 6.166666666666667,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02620556753316179,
      "loss": 0.2264,
      "num_input_tokens_seen": 11828640,
      "step": 56055
    },
    {
      "epoch": 6.167216721672167,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.026204610169118417,
      "loss": 0.237,
      "num_input_tokens_seen": 11829728,
      "step": 56060
    },
    {
      "epoch": 6.167766776677667,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.026203652701807526,
      "loss": 0.2359,
      "num_input_tokens_seen": 11830848,
      "step": 56065
    },
    {
      "epoch": 6.1683168316831685,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.026202695131237946,
      "loss": 0.228,
      "num_input_tokens_seen": 11831904,
      "step": 56070
    },
    {
      "epoch": 6.168866886688669,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.026201737457418495,
      "loss": 0.2302,
      "num_input_tokens_seen": 11832928,
      "step": 56075
    },
    {
      "epoch": 6.16941694169417,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.026200779680358013,
      "loss": 0.2327,
      "num_input_tokens_seen": 11833952,
      "step": 56080
    },
    {
      "epoch": 6.16996699669967,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.026199821800065318,
      "loss": 0.2347,
      "num_input_tokens_seen": 11834976,
      "step": 56085
    },
    {
      "epoch": 6.17051705170517,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.026198863816549237,
      "loss": 0.23,
      "num_input_tokens_seen": 11836032,
      "step": 56090
    },
    {
      "epoch": 6.171067106710671,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.026197905729818605,
      "loss": 0.2327,
      "num_input_tokens_seen": 11837024,
      "step": 56095
    },
    {
      "epoch": 6.1716171617161715,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.026196947539882248,
      "loss": 0.231,
      "num_input_tokens_seen": 11838080,
      "step": 56100
    },
    {
      "epoch": 6.172167216721673,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.026195989246748996,
      "loss": 0.2284,
      "num_input_tokens_seen": 11839072,
      "step": 56105
    },
    {
      "epoch": 6.172717271727173,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026195030850427684,
      "loss": 0.2315,
      "num_input_tokens_seen": 11840128,
      "step": 56110
    },
    {
      "epoch": 6.173267326732673,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.026194072350927147,
      "loss": 0.2284,
      "num_input_tokens_seen": 11841216,
      "step": 56115
    },
    {
      "epoch": 6.173817381738174,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.026193113748256214,
      "loss": 0.2306,
      "num_input_tokens_seen": 11842240,
      "step": 56120
    },
    {
      "epoch": 6.174367436743674,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.026192155042423726,
      "loss": 0.2384,
      "num_input_tokens_seen": 11843296,
      "step": 56125
    },
    {
      "epoch": 6.174917491749175,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02619119623343851,
      "loss": 0.2341,
      "num_input_tokens_seen": 11844352,
      "step": 56130
    },
    {
      "epoch": 6.175467546754676,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.026190237321309415,
      "loss": 0.233,
      "num_input_tokens_seen": 11845472,
      "step": 56135
    },
    {
      "epoch": 6.176017601760176,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.02618927830604527,
      "loss": 0.2325,
      "num_input_tokens_seen": 11846560,
      "step": 56140
    },
    {
      "epoch": 6.176567656765677,
      "grad_norm": 0.005859375,
      "learning_rate": 0.026188319187654914,
      "loss": 0.2345,
      "num_input_tokens_seen": 11847552,
      "step": 56145
    },
    {
      "epoch": 6.177117711771177,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02618735996614719,
      "loss": 0.232,
      "num_input_tokens_seen": 11848608,
      "step": 56150
    },
    {
      "epoch": 6.177667766776677,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.026186400641530935,
      "loss": 0.233,
      "num_input_tokens_seen": 11849664,
      "step": 56155
    },
    {
      "epoch": 6.178217821782178,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.026185441213814997,
      "loss": 0.2319,
      "num_input_tokens_seen": 11850624,
      "step": 56160
    },
    {
      "epoch": 6.178767876787679,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.026184481683008208,
      "loss": 0.2319,
      "num_input_tokens_seen": 11851648,
      "step": 56165
    },
    {
      "epoch": 6.17931793179318,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.026183522049119425,
      "loss": 0.2303,
      "num_input_tokens_seen": 11852704,
      "step": 56170
    },
    {
      "epoch": 6.17986798679868,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.026182562312157483,
      "loss": 0.2314,
      "num_input_tokens_seen": 11853728,
      "step": 56175
    },
    {
      "epoch": 6.18041804180418,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.02618160247213123,
      "loss": 0.2298,
      "num_input_tokens_seen": 11854816,
      "step": 56180
    },
    {
      "epoch": 6.180968096809681,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.026180642529049513,
      "loss": 0.2308,
      "num_input_tokens_seen": 11855872,
      "step": 56185
    },
    {
      "epoch": 6.181518151815181,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.026179682482921178,
      "loss": 0.2314,
      "num_input_tokens_seen": 11856896,
      "step": 56190
    },
    {
      "epoch": 6.1820682068206825,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.026178722333755077,
      "loss": 0.2309,
      "num_input_tokens_seen": 11857984,
      "step": 56195
    },
    {
      "epoch": 6.182618261826183,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02617776208156005,
      "loss": 0.2314,
      "num_input_tokens_seen": 11859008,
      "step": 56200
    },
    {
      "epoch": 6.183168316831683,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.02617680172634496,
      "loss": 0.2314,
      "num_input_tokens_seen": 11860064,
      "step": 56205
    },
    {
      "epoch": 6.183718371837184,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02617584126811865,
      "loss": 0.2319,
      "num_input_tokens_seen": 11861184,
      "step": 56210
    },
    {
      "epoch": 6.184268426842684,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.026174880706889975,
      "loss": 0.2303,
      "num_input_tokens_seen": 11862240,
      "step": 56215
    },
    {
      "epoch": 6.184818481848184,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.026173920042667786,
      "loss": 0.2319,
      "num_input_tokens_seen": 11863328,
      "step": 56220
    },
    {
      "epoch": 6.1853685368536855,
      "grad_norm": 0.000579833984375,
      "learning_rate": 0.026172959275460934,
      "loss": 0.2309,
      "num_input_tokens_seen": 11864320,
      "step": 56225
    },
    {
      "epoch": 6.185918591859186,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.026171998405278278,
      "loss": 0.2278,
      "num_input_tokens_seen": 11865344,
      "step": 56230
    },
    {
      "epoch": 6.186468646864687,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.026171037432128678,
      "loss": 0.232,
      "num_input_tokens_seen": 11866464,
      "step": 56235
    },
    {
      "epoch": 6.187018701870187,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.026170076356020984,
      "loss": 0.232,
      "num_input_tokens_seen": 11867520,
      "step": 56240
    },
    {
      "epoch": 6.187568756875687,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.026169115176964058,
      "loss": 0.2304,
      "num_input_tokens_seen": 11868576,
      "step": 56245
    },
    {
      "epoch": 6.188118811881188,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026168153894966753,
      "loss": 0.2336,
      "num_input_tokens_seen": 11869696,
      "step": 56250
    },
    {
      "epoch": 6.1886688668866885,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.026167192510037934,
      "loss": 0.2326,
      "num_input_tokens_seen": 11870720,
      "step": 56255
    },
    {
      "epoch": 6.18921892189219,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.026166231022186463,
      "loss": 0.2352,
      "num_input_tokens_seen": 11871744,
      "step": 56260
    },
    {
      "epoch": 6.18976897689769,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.026165269431421195,
      "loss": 0.2315,
      "num_input_tokens_seen": 11872800,
      "step": 56265
    },
    {
      "epoch": 6.19031903190319,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.026164307737750997,
      "loss": 0.2294,
      "num_input_tokens_seen": 11873888,
      "step": 56270
    },
    {
      "epoch": 6.190869086908691,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.026163345941184735,
      "loss": 0.2309,
      "num_input_tokens_seen": 11874912,
      "step": 56275
    },
    {
      "epoch": 6.191419141914191,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.026162384041731265,
      "loss": 0.2294,
      "num_input_tokens_seen": 11875936,
      "step": 56280
    },
    {
      "epoch": 6.191969196919692,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.026161422039399463,
      "loss": 0.2304,
      "num_input_tokens_seen": 11876992,
      "step": 56285
    },
    {
      "epoch": 6.192519251925193,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026160459934198188,
      "loss": 0.2309,
      "num_input_tokens_seen": 11878016,
      "step": 56290
    },
    {
      "epoch": 6.193069306930693,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.026159497726136308,
      "loss": 0.2325,
      "num_input_tokens_seen": 11879008,
      "step": 56295
    },
    {
      "epoch": 6.193619361936194,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.02615853541522269,
      "loss": 0.2299,
      "num_input_tokens_seen": 11880064,
      "step": 56300
    },
    {
      "epoch": 6.194169416941694,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.026157573001466213,
      "loss": 0.2325,
      "num_input_tokens_seen": 11881152,
      "step": 56305
    },
    {
      "epoch": 6.194719471947194,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02615661048487573,
      "loss": 0.2299,
      "num_input_tokens_seen": 11882208,
      "step": 56310
    },
    {
      "epoch": 6.195269526952695,
      "grad_norm": 0.009765625,
      "learning_rate": 0.026155647865460128,
      "loss": 0.2283,
      "num_input_tokens_seen": 11883264,
      "step": 56315
    },
    {
      "epoch": 6.195819581958196,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.026154685143228266,
      "loss": 0.2304,
      "num_input_tokens_seen": 11884320,
      "step": 56320
    },
    {
      "epoch": 6.196369636963697,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.02615372231818903,
      "loss": 0.232,
      "num_input_tokens_seen": 11885376,
      "step": 56325
    },
    {
      "epoch": 6.196919691969197,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.026152759390351287,
      "loss": 0.232,
      "num_input_tokens_seen": 11886464,
      "step": 56330
    },
    {
      "epoch": 6.197469746974697,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026151796359723915,
      "loss": 0.2341,
      "num_input_tokens_seen": 11887520,
      "step": 56335
    },
    {
      "epoch": 6.198019801980198,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.02615083322631578,
      "loss": 0.232,
      "num_input_tokens_seen": 11888608,
      "step": 56340
    },
    {
      "epoch": 6.198569856985698,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.026149869990135768,
      "loss": 0.231,
      "num_input_tokens_seen": 11889728,
      "step": 56345
    },
    {
      "epoch": 6.1991199119911995,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.02614890665119276,
      "loss": 0.2315,
      "num_input_tokens_seen": 11890688,
      "step": 56350
    },
    {
      "epoch": 6.1996699669967,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.026147943209495626,
      "loss": 0.2279,
      "num_input_tokens_seen": 11891712,
      "step": 56355
    },
    {
      "epoch": 6.2002200220022,
      "grad_norm": 0.000713348388671875,
      "learning_rate": 0.026146979665053242,
      "loss": 0.2341,
      "num_input_tokens_seen": 11892704,
      "step": 56360
    },
    {
      "epoch": 6.200770077007701,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.026146016017874505,
      "loss": 0.233,
      "num_input_tokens_seen": 11893728,
      "step": 56365
    },
    {
      "epoch": 6.201320132013201,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.02614505226796828,
      "loss": 0.232,
      "num_input_tokens_seen": 11894752,
      "step": 56370
    },
    {
      "epoch": 6.201870187018702,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.026144088415343457,
      "loss": 0.2294,
      "num_input_tokens_seen": 11895808,
      "step": 56375
    },
    {
      "epoch": 6.2024202420242025,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.026143124460008924,
      "loss": 0.233,
      "num_input_tokens_seen": 11896864,
      "step": 56380
    },
    {
      "epoch": 6.202970297029703,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.026142160401973553,
      "loss": 0.2335,
      "num_input_tokens_seen": 11897920,
      "step": 56385
    },
    {
      "epoch": 6.203520352035204,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02614119624124624,
      "loss": 0.2346,
      "num_input_tokens_seen": 11899008,
      "step": 56390
    },
    {
      "epoch": 6.204070407040704,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.026140231977835864,
      "loss": 0.2309,
      "num_input_tokens_seen": 11900096,
      "step": 56395
    },
    {
      "epoch": 6.204620462046204,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.026139267611751324,
      "loss": 0.2294,
      "num_input_tokens_seen": 11901152,
      "step": 56400
    },
    {
      "epoch": 6.205170517051705,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02613830314300149,
      "loss": 0.2298,
      "num_input_tokens_seen": 11902240,
      "step": 56405
    },
    {
      "epoch": 6.2057205720572055,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.026137338571595264,
      "loss": 0.2298,
      "num_input_tokens_seen": 11903360,
      "step": 56410
    },
    {
      "epoch": 6.206270627062707,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.02613637389754153,
      "loss": 0.2325,
      "num_input_tokens_seen": 11904384,
      "step": 56415
    },
    {
      "epoch": 6.206820682068207,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.026135409120849187,
      "loss": 0.2304,
      "num_input_tokens_seen": 11905440,
      "step": 56420
    },
    {
      "epoch": 6.207370737073707,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02613444424152712,
      "loss": 0.2324,
      "num_input_tokens_seen": 11906592,
      "step": 56425
    },
    {
      "epoch": 6.207920792079208,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02613347925958422,
      "loss": 0.234,
      "num_input_tokens_seen": 11907712,
      "step": 56430
    },
    {
      "epoch": 6.208470847084708,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.02613251417502939,
      "loss": 0.2314,
      "num_input_tokens_seen": 11908800,
      "step": 56435
    },
    {
      "epoch": 6.209020902090209,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02613154898787151,
      "loss": 0.2309,
      "num_input_tokens_seen": 11909856,
      "step": 56440
    },
    {
      "epoch": 6.20957095709571,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02613058369811949,
      "loss": 0.2303,
      "num_input_tokens_seen": 11910976,
      "step": 56445
    },
    {
      "epoch": 6.21012101210121,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02612961830578222,
      "loss": 0.2319,
      "num_input_tokens_seen": 11912096,
      "step": 56450
    },
    {
      "epoch": 6.210671067106711,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0261286528108686,
      "loss": 0.2324,
      "num_input_tokens_seen": 11913152,
      "step": 56455
    },
    {
      "epoch": 6.211221122112211,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02612768721338753,
      "loss": 0.2303,
      "num_input_tokens_seen": 11914176,
      "step": 56460
    },
    {
      "epoch": 6.211771177117711,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0261267215133479,
      "loss": 0.2314,
      "num_input_tokens_seen": 11915232,
      "step": 56465
    },
    {
      "epoch": 6.212321232123212,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.026125755710758616,
      "loss": 0.2319,
      "num_input_tokens_seen": 11916256,
      "step": 56470
    },
    {
      "epoch": 6.212871287128713,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.026124789805628584,
      "loss": 0.2309,
      "num_input_tokens_seen": 11917280,
      "step": 56475
    },
    {
      "epoch": 6.213421342134214,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.0261238237979667,
      "loss": 0.2314,
      "num_input_tokens_seen": 11918304,
      "step": 56480
    },
    {
      "epoch": 6.213971397139714,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.026122857687781876,
      "loss": 0.2335,
      "num_input_tokens_seen": 11919328,
      "step": 56485
    },
    {
      "epoch": 6.214521452145214,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.026121891475083005,
      "loss": 0.2319,
      "num_input_tokens_seen": 11920416,
      "step": 56490
    },
    {
      "epoch": 6.215071507150715,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.026120925159879,
      "loss": 0.2293,
      "num_input_tokens_seen": 11921504,
      "step": 56495
    },
    {
      "epoch": 6.215621562156215,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.02611995874217876,
      "loss": 0.2329,
      "num_input_tokens_seen": 11922592,
      "step": 56500
    },
    {
      "epoch": 6.2161716171617165,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.0261189922219912,
      "loss": 0.2319,
      "num_input_tokens_seen": 11923744,
      "step": 56505
    },
    {
      "epoch": 6.216721672167217,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.02611802559932522,
      "loss": 0.2293,
      "num_input_tokens_seen": 11924704,
      "step": 56510
    },
    {
      "epoch": 6.217271727172717,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.026117058874189737,
      "loss": 0.2335,
      "num_input_tokens_seen": 11925696,
      "step": 56515
    },
    {
      "epoch": 6.217821782178218,
      "grad_norm": 0.00048828125,
      "learning_rate": 0.026116092046593653,
      "loss": 0.2298,
      "num_input_tokens_seen": 11926720,
      "step": 56520
    },
    {
      "epoch": 6.218371837183718,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026115125116545884,
      "loss": 0.2293,
      "num_input_tokens_seen": 11927744,
      "step": 56525
    },
    {
      "epoch": 6.218921892189219,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.026114158084055348,
      "loss": 0.2288,
      "num_input_tokens_seen": 11928800,
      "step": 56530
    },
    {
      "epoch": 6.2194719471947195,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.026113190949130942,
      "loss": 0.2314,
      "num_input_tokens_seen": 11929888,
      "step": 56535
    },
    {
      "epoch": 6.22002200220022,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.026112223711781585,
      "loss": 0.2308,
      "num_input_tokens_seen": 11930912,
      "step": 56540
    },
    {
      "epoch": 6.220572057205721,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.0261112563720162,
      "loss": 0.2298,
      "num_input_tokens_seen": 11932032,
      "step": 56545
    },
    {
      "epoch": 6.221122112211221,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02611028892984369,
      "loss": 0.2334,
      "num_input_tokens_seen": 11933184,
      "step": 56550
    },
    {
      "epoch": 6.221672167216722,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02610932138527299,
      "loss": 0.2314,
      "num_input_tokens_seen": 11934208,
      "step": 56555
    },
    {
      "epoch": 6.222222222222222,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.026108353738312995,
      "loss": 0.2314,
      "num_input_tokens_seen": 11935264,
      "step": 56560
    },
    {
      "epoch": 6.2227722772277225,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02610738598897264,
      "loss": 0.2293,
      "num_input_tokens_seen": 11936320,
      "step": 56565
    },
    {
      "epoch": 6.223322332233224,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.026106418137260834,
      "loss": 0.2319,
      "num_input_tokens_seen": 11937312,
      "step": 56570
    },
    {
      "epoch": 6.223872387238724,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.026105450183186503,
      "loss": 0.234,
      "num_input_tokens_seen": 11938368,
      "step": 56575
    },
    {
      "epoch": 6.224422442244224,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026104482126758564,
      "loss": 0.2304,
      "num_input_tokens_seen": 11939424,
      "step": 56580
    },
    {
      "epoch": 6.224972497249725,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.026103513967985947,
      "loss": 0.233,
      "num_input_tokens_seen": 11940544,
      "step": 56585
    },
    {
      "epoch": 6.225522552255225,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.026102545706877568,
      "loss": 0.2314,
      "num_input_tokens_seen": 11941536,
      "step": 56590
    },
    {
      "epoch": 6.226072607260726,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02610157734344235,
      "loss": 0.2309,
      "num_input_tokens_seen": 11942656,
      "step": 56595
    },
    {
      "epoch": 6.226622662266227,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.026100608877689224,
      "loss": 0.2303,
      "num_input_tokens_seen": 11943712,
      "step": 56600
    },
    {
      "epoch": 6.227172717271727,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02609964030962711,
      "loss": 0.2293,
      "num_input_tokens_seen": 11944704,
      "step": 56605
    },
    {
      "epoch": 6.227722772277228,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02609867163926494,
      "loss": 0.2345,
      "num_input_tokens_seen": 11945824,
      "step": 56610
    },
    {
      "epoch": 6.228272827282728,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.026097702866611634,
      "loss": 0.2314,
      "num_input_tokens_seen": 11946880,
      "step": 56615
    },
    {
      "epoch": 6.228822882288229,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02609673399167613,
      "loss": 0.2298,
      "num_input_tokens_seen": 11947936,
      "step": 56620
    },
    {
      "epoch": 6.229372937293729,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.026095765014467358,
      "loss": 0.2293,
      "num_input_tokens_seen": 11948992,
      "step": 56625
    },
    {
      "epoch": 6.22992299229923,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.026094795934994235,
      "loss": 0.2335,
      "num_input_tokens_seen": 11950048,
      "step": 56630
    },
    {
      "epoch": 6.230473047304731,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.026093826753265704,
      "loss": 0.2313,
      "num_input_tokens_seen": 11951072,
      "step": 56635
    },
    {
      "epoch": 6.231023102310231,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0260928574692907,
      "loss": 0.2308,
      "num_input_tokens_seen": 11952128,
      "step": 56640
    },
    {
      "epoch": 6.231573157315731,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.026091888083078146,
      "loss": 0.2313,
      "num_input_tokens_seen": 11953216,
      "step": 56645
    },
    {
      "epoch": 6.232123212321232,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.026090918594636987,
      "loss": 0.2329,
      "num_input_tokens_seen": 11954240,
      "step": 56650
    },
    {
      "epoch": 6.232673267326732,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02608994900397615,
      "loss": 0.2313,
      "num_input_tokens_seen": 11955296,
      "step": 56655
    },
    {
      "epoch": 6.2332233223322335,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.026088979311104574,
      "loss": 0.2324,
      "num_input_tokens_seen": 11956320,
      "step": 56660
    },
    {
      "epoch": 6.233773377337734,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.0260880095160312,
      "loss": 0.2309,
      "num_input_tokens_seen": 11957312,
      "step": 56665
    },
    {
      "epoch": 6.234323432343234,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.02608703961876496,
      "loss": 0.2309,
      "num_input_tokens_seen": 11958368,
      "step": 56670
    },
    {
      "epoch": 6.234873487348735,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.026086069619314797,
      "loss": 0.2335,
      "num_input_tokens_seen": 11959424,
      "step": 56675
    },
    {
      "epoch": 6.235423542354235,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.026085099517689645,
      "loss": 0.2298,
      "num_input_tokens_seen": 11960512,
      "step": 56680
    },
    {
      "epoch": 6.235973597359736,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.02608412931389846,
      "loss": 0.2324,
      "num_input_tokens_seen": 11961600,
      "step": 56685
    },
    {
      "epoch": 6.2365236523652365,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.026083159007950163,
      "loss": 0.2309,
      "num_input_tokens_seen": 11962624,
      "step": 56690
    },
    {
      "epoch": 6.237073707370737,
      "grad_norm": 0.000713348388671875,
      "learning_rate": 0.026082188599853718,
      "loss": 0.2298,
      "num_input_tokens_seen": 11963616,
      "step": 56695
    },
    {
      "epoch": 6.237623762376238,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.02608121808961805,
      "loss": 0.2309,
      "num_input_tokens_seen": 11964672,
      "step": 56700
    },
    {
      "epoch": 6.238173817381738,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.026080247477252117,
      "loss": 0.2308,
      "num_input_tokens_seen": 11965728,
      "step": 56705
    },
    {
      "epoch": 6.238723872387239,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.026079276762764857,
      "loss": 0.2314,
      "num_input_tokens_seen": 11966816,
      "step": 56710
    },
    {
      "epoch": 6.239273927392739,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02607830594616522,
      "loss": 0.2303,
      "num_input_tokens_seen": 11967872,
      "step": 56715
    },
    {
      "epoch": 6.2398239823982395,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.026077335027462158,
      "loss": 0.2308,
      "num_input_tokens_seen": 11968896,
      "step": 56720
    },
    {
      "epoch": 6.240374037403741,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.026076364006664608,
      "loss": 0.2324,
      "num_input_tokens_seen": 11969984,
      "step": 56725
    },
    {
      "epoch": 6.240924092409241,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.026075392883781526,
      "loss": 0.2314,
      "num_input_tokens_seen": 11971040,
      "step": 56730
    },
    {
      "epoch": 6.241474147414741,
      "grad_norm": 0.0004825592041015625,
      "learning_rate": 0.026074421658821862,
      "loss": 0.233,
      "num_input_tokens_seen": 11972192,
      "step": 56735
    },
    {
      "epoch": 6.242024202420242,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.026073450331794572,
      "loss": 0.2304,
      "num_input_tokens_seen": 11973216,
      "step": 56740
    },
    {
      "epoch": 6.242574257425742,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.026072478902708605,
      "loss": 0.2314,
      "num_input_tokens_seen": 11974304,
      "step": 56745
    },
    {
      "epoch": 6.243124312431243,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02607150737157291,
      "loss": 0.234,
      "num_input_tokens_seen": 11975296,
      "step": 56750
    },
    {
      "epoch": 6.243674367436744,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.026070535738396444,
      "loss": 0.2288,
      "num_input_tokens_seen": 11976352,
      "step": 56755
    },
    {
      "epoch": 6.244224422442244,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.02606956400318816,
      "loss": 0.2324,
      "num_input_tokens_seen": 11977408,
      "step": 56760
    },
    {
      "epoch": 6.244774477447745,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.026068592165957017,
      "loss": 0.2309,
      "num_input_tokens_seen": 11978496,
      "step": 56765
    },
    {
      "epoch": 6.245324532453245,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.026067620226711975,
      "loss": 0.233,
      "num_input_tokens_seen": 11979552,
      "step": 56770
    },
    {
      "epoch": 6.245874587458746,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.026066648185461987,
      "loss": 0.2298,
      "num_input_tokens_seen": 11980640,
      "step": 56775
    },
    {
      "epoch": 6.2464246424642464,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.026065676042216015,
      "loss": 0.2329,
      "num_input_tokens_seen": 11981760,
      "step": 56780
    },
    {
      "epoch": 6.246974697469747,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.026064703796983012,
      "loss": 0.234,
      "num_input_tokens_seen": 11982784,
      "step": 56785
    },
    {
      "epoch": 6.247524752475248,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.026063731449771945,
      "loss": 0.2324,
      "num_input_tokens_seen": 11983808,
      "step": 56790
    },
    {
      "epoch": 6.248074807480748,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.026062759000591774,
      "loss": 0.2304,
      "num_input_tokens_seen": 11984864,
      "step": 56795
    },
    {
      "epoch": 6.248624862486249,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.026061786449451462,
      "loss": 0.2309,
      "num_input_tokens_seen": 11985920,
      "step": 56800
    },
    {
      "epoch": 6.249174917491749,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026060813796359973,
      "loss": 0.2309,
      "num_input_tokens_seen": 11986976,
      "step": 56805
    },
    {
      "epoch": 6.2497249724972495,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02605984104132627,
      "loss": 0.2303,
      "num_input_tokens_seen": 11988000,
      "step": 56810
    },
    {
      "epoch": 6.2502750275027505,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02605886818435932,
      "loss": 0.2319,
      "num_input_tokens_seen": 11989088,
      "step": 56815
    },
    {
      "epoch": 6.250825082508251,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.026057895225468083,
      "loss": 0.2303,
      "num_input_tokens_seen": 11990144,
      "step": 56820
    },
    {
      "epoch": 6.251375137513751,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.026056922164661538,
      "loss": 0.2314,
      "num_input_tokens_seen": 11991200,
      "step": 56825
    },
    {
      "epoch": 6.251925192519252,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.026055949001948644,
      "loss": 0.2314,
      "num_input_tokens_seen": 11992288,
      "step": 56830
    },
    {
      "epoch": 6.252475247524752,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.026054975737338375,
      "loss": 0.2319,
      "num_input_tokens_seen": 11993312,
      "step": 56835
    },
    {
      "epoch": 6.253025302530253,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0260540023708397,
      "loss": 0.2324,
      "num_input_tokens_seen": 11994400,
      "step": 56840
    },
    {
      "epoch": 6.2535753575357536,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.026053028902461587,
      "loss": 0.2325,
      "num_input_tokens_seen": 11995456,
      "step": 56845
    },
    {
      "epoch": 6.254125412541254,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02605205533221301,
      "loss": 0.2324,
      "num_input_tokens_seen": 11996544,
      "step": 56850
    },
    {
      "epoch": 6.254675467546755,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.026051081660102948,
      "loss": 0.2319,
      "num_input_tokens_seen": 11997632,
      "step": 56855
    },
    {
      "epoch": 6.255225522552255,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.026050107886140362,
      "loss": 0.2309,
      "num_input_tokens_seen": 11998688,
      "step": 56860
    },
    {
      "epoch": 6.255775577557756,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.026049134010334242,
      "loss": 0.2329,
      "num_input_tokens_seen": 11999776,
      "step": 56865
    },
    {
      "epoch": 6.256325632563256,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.026048160032693547,
      "loss": 0.2314,
      "num_input_tokens_seen": 12000800,
      "step": 56870
    },
    {
      "epoch": 6.256875687568757,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02604718595322727,
      "loss": 0.2324,
      "num_input_tokens_seen": 12001920,
      "step": 56875
    },
    {
      "epoch": 6.257425742574258,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.026046211771944376,
      "loss": 0.2293,
      "num_input_tokens_seen": 12003008,
      "step": 56880
    },
    {
      "epoch": 6.257975797579758,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.026045237488853852,
      "loss": 0.2319,
      "num_input_tokens_seen": 12004064,
      "step": 56885
    },
    {
      "epoch": 6.258525852585258,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.02604426310396467,
      "loss": 0.2329,
      "num_input_tokens_seen": 12005152,
      "step": 56890
    },
    {
      "epoch": 6.259075907590759,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.026043288617285817,
      "loss": 0.2314,
      "num_input_tokens_seen": 12006176,
      "step": 56895
    },
    {
      "epoch": 6.259625962596259,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.02604231402882627,
      "loss": 0.2319,
      "num_input_tokens_seen": 12007328,
      "step": 56900
    },
    {
      "epoch": 6.2601760176017605,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.026041339338595013,
      "loss": 0.2298,
      "num_input_tokens_seen": 12008448,
      "step": 56905
    },
    {
      "epoch": 6.260726072607261,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.026040364546601035,
      "loss": 0.2324,
      "num_input_tokens_seen": 12009568,
      "step": 56910
    },
    {
      "epoch": 6.261276127612761,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.02603938965285331,
      "loss": 0.2293,
      "num_input_tokens_seen": 12010624,
      "step": 56915
    },
    {
      "epoch": 6.261826182618262,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02603841465736083,
      "loss": 0.2273,
      "num_input_tokens_seen": 12011680,
      "step": 56920
    },
    {
      "epoch": 6.262376237623762,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.026037439560132575,
      "loss": 0.2354,
      "num_input_tokens_seen": 12012800,
      "step": 56925
    },
    {
      "epoch": 6.262926292629263,
      "grad_norm": 0.001953125,
      "learning_rate": 0.026036464361177538,
      "loss": 0.225,
      "num_input_tokens_seen": 12013856,
      "step": 56930
    },
    {
      "epoch": 6.2634763476347635,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.026035489060504704,
      "loss": 0.2309,
      "num_input_tokens_seen": 12014880,
      "step": 56935
    },
    {
      "epoch": 6.264026402640264,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.026034513658123065,
      "loss": 0.2337,
      "num_input_tokens_seen": 12015936,
      "step": 56940
    },
    {
      "epoch": 6.264576457645765,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.026033538154041604,
      "loss": 0.2283,
      "num_input_tokens_seen": 12016928,
      "step": 56945
    },
    {
      "epoch": 6.265126512651265,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.026032562548269324,
      "loss": 0.2238,
      "num_input_tokens_seen": 12017952,
      "step": 56950
    },
    {
      "epoch": 6.265676567656766,
      "grad_norm": 0.00836181640625,
      "learning_rate": 0.0260315868408152,
      "loss": 0.2336,
      "num_input_tokens_seen": 12018976,
      "step": 56955
    },
    {
      "epoch": 6.266226622662266,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02603061103168824,
      "loss": 0.2242,
      "num_input_tokens_seen": 12020000,
      "step": 56960
    },
    {
      "epoch": 6.2667766776677665,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02602963512089743,
      "loss": 0.2312,
      "num_input_tokens_seen": 12021056,
      "step": 56965
    },
    {
      "epoch": 6.267326732673268,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.026028659108451764,
      "loss": 0.2392,
      "num_input_tokens_seen": 12022112,
      "step": 56970
    },
    {
      "epoch": 6.267876787678768,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.02602768299436024,
      "loss": 0.2256,
      "num_input_tokens_seen": 12023200,
      "step": 56975
    },
    {
      "epoch": 6.268426842684269,
      "grad_norm": 0.00714111328125,
      "learning_rate": 0.02602670677863185,
      "loss": 0.2262,
      "num_input_tokens_seen": 12024288,
      "step": 56980
    },
    {
      "epoch": 6.268976897689769,
      "grad_norm": 0.0027313232421875,
      "learning_rate": 0.0260257304612756,
      "loss": 0.2364,
      "num_input_tokens_seen": 12025312,
      "step": 56985
    },
    {
      "epoch": 6.269526952695269,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.02602475404230048,
      "loss": 0.2287,
      "num_input_tokens_seen": 12026336,
      "step": 56990
    },
    {
      "epoch": 6.27007700770077,
      "grad_norm": 0.00787353515625,
      "learning_rate": 0.026023777521715495,
      "loss": 0.2411,
      "num_input_tokens_seen": 12027392,
      "step": 56995
    },
    {
      "epoch": 6.270627062706271,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.02602280089952964,
      "loss": 0.2215,
      "num_input_tokens_seen": 12028416,
      "step": 57000
    },
    {
      "epoch": 6.271177117711771,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.026021824175751915,
      "loss": 0.2335,
      "num_input_tokens_seen": 12029408,
      "step": 57005
    },
    {
      "epoch": 6.271727172717272,
      "grad_norm": 0.007568359375,
      "learning_rate": 0.02602084735039133,
      "loss": 0.2377,
      "num_input_tokens_seen": 12030464,
      "step": 57010
    },
    {
      "epoch": 6.272277227722772,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.026019870423456885,
      "loss": 0.2262,
      "num_input_tokens_seen": 12031552,
      "step": 57015
    },
    {
      "epoch": 6.272827282728273,
      "grad_norm": 0.01336669921875,
      "learning_rate": 0.02601889339495758,
      "loss": 0.2318,
      "num_input_tokens_seen": 12032576,
      "step": 57020
    },
    {
      "epoch": 6.273377337733773,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.02601791626490242,
      "loss": 0.2291,
      "num_input_tokens_seen": 12033536,
      "step": 57025
    },
    {
      "epoch": 6.273927392739274,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.026016939033300417,
      "loss": 0.2271,
      "num_input_tokens_seen": 12034528,
      "step": 57030
    },
    {
      "epoch": 6.274477447744775,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.02601596170016057,
      "loss": 0.2327,
      "num_input_tokens_seen": 12035584,
      "step": 57035
    },
    {
      "epoch": 6.275027502750275,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.026014984265491887,
      "loss": 0.2354,
      "num_input_tokens_seen": 12036640,
      "step": 57040
    },
    {
      "epoch": 6.275577557755776,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.026014006729303388,
      "loss": 0.2344,
      "num_input_tokens_seen": 12037664,
      "step": 57045
    },
    {
      "epoch": 6.276127612761276,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.026013029091604065,
      "loss": 0.2307,
      "num_input_tokens_seen": 12038752,
      "step": 57050
    },
    {
      "epoch": 6.276677667766776,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.02601205135240294,
      "loss": 0.2359,
      "num_input_tokens_seen": 12039808,
      "step": 57055
    },
    {
      "epoch": 6.2772277227722775,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.026011073511709025,
      "loss": 0.2389,
      "num_input_tokens_seen": 12040800,
      "step": 57060
    },
    {
      "epoch": 6.277777777777778,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02601009556953133,
      "loss": 0.2249,
      "num_input_tokens_seen": 12041952,
      "step": 57065
    },
    {
      "epoch": 6.278327832783278,
      "grad_norm": 0.001953125,
      "learning_rate": 0.02600911752587886,
      "loss": 0.2379,
      "num_input_tokens_seen": 12043040,
      "step": 57070
    },
    {
      "epoch": 6.278877887788779,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.026008139380760645,
      "loss": 0.2289,
      "num_input_tokens_seen": 12044064,
      "step": 57075
    },
    {
      "epoch": 6.279427942794279,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.026007161134185688,
      "loss": 0.2305,
      "num_input_tokens_seen": 12045088,
      "step": 57080
    },
    {
      "epoch": 6.27997799779978,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.026006182786163005,
      "loss": 0.2336,
      "num_input_tokens_seen": 12046176,
      "step": 57085
    },
    {
      "epoch": 6.2805280528052805,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.026005204336701625,
      "loss": 0.232,
      "num_input_tokens_seen": 12047232,
      "step": 57090
    },
    {
      "epoch": 6.281078107810781,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.026004225785810547,
      "loss": 0.23,
      "num_input_tokens_seen": 12048320,
      "step": 57095
    },
    {
      "epoch": 6.281628162816282,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.02600324713349881,
      "loss": 0.2304,
      "num_input_tokens_seen": 12049408,
      "step": 57100
    },
    {
      "epoch": 6.282178217821782,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.026002268379775418,
      "loss": 0.2319,
      "num_input_tokens_seen": 12050464,
      "step": 57105
    },
    {
      "epoch": 6.282728272827283,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0260012895246494,
      "loss": 0.2283,
      "num_input_tokens_seen": 12051520,
      "step": 57110
    },
    {
      "epoch": 6.283278327832783,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.026000310568129775,
      "loss": 0.2325,
      "num_input_tokens_seen": 12052576,
      "step": 57115
    },
    {
      "epoch": 6.2838283828382835,
      "grad_norm": 0.005859375,
      "learning_rate": 0.025999331510225564,
      "loss": 0.233,
      "num_input_tokens_seen": 12053600,
      "step": 57120
    },
    {
      "epoch": 6.284378437843785,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.025998352350945792,
      "loss": 0.2283,
      "num_input_tokens_seen": 12054656,
      "step": 57125
    },
    {
      "epoch": 6.284928492849285,
      "grad_norm": 0.01171875,
      "learning_rate": 0.025997373090299487,
      "loss": 0.2309,
      "num_input_tokens_seen": 12055712,
      "step": 57130
    },
    {
      "epoch": 6.285478547854786,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.02599639372829567,
      "loss": 0.2288,
      "num_input_tokens_seen": 12056768,
      "step": 57135
    },
    {
      "epoch": 6.286028602860286,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02599541426494337,
      "loss": 0.2263,
      "num_input_tokens_seen": 12057856,
      "step": 57140
    },
    {
      "epoch": 6.286578657865786,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02599443470025161,
      "loss": 0.2274,
      "num_input_tokens_seen": 12058944,
      "step": 57145
    },
    {
      "epoch": 6.287128712871287,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.025993455034229423,
      "loss": 0.2284,
      "num_input_tokens_seen": 12059936,
      "step": 57150
    },
    {
      "epoch": 6.287678767876788,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.025992475266885837,
      "loss": 0.2276,
      "num_input_tokens_seen": 12060992,
      "step": 57155
    },
    {
      "epoch": 6.288228822882289,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.02599149539822988,
      "loss": 0.2334,
      "num_input_tokens_seen": 12062048,
      "step": 57160
    },
    {
      "epoch": 6.288778877887789,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.025990515428270584,
      "loss": 0.2251,
      "num_input_tokens_seen": 12063136,
      "step": 57165
    },
    {
      "epoch": 6.289328932893289,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.025989535357016983,
      "loss": 0.2328,
      "num_input_tokens_seen": 12064224,
      "step": 57170
    },
    {
      "epoch": 6.28987898789879,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.025988555184478106,
      "loss": 0.235,
      "num_input_tokens_seen": 12065312,
      "step": 57175
    },
    {
      "epoch": 6.29042904290429,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02598757491066299,
      "loss": 0.2391,
      "num_input_tokens_seen": 12066400,
      "step": 57180
    },
    {
      "epoch": 6.290979097909791,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.025986594535580665,
      "loss": 0.2209,
      "num_input_tokens_seen": 12067456,
      "step": 57185
    },
    {
      "epoch": 6.291529152915292,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.025985614059240172,
      "loss": 0.2261,
      "num_input_tokens_seen": 12068544,
      "step": 57190
    },
    {
      "epoch": 6.292079207920792,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.025984633481650546,
      "loss": 0.2255,
      "num_input_tokens_seen": 12069568,
      "step": 57195
    },
    {
      "epoch": 6.292629262926293,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.025983652802820823,
      "loss": 0.2292,
      "num_input_tokens_seen": 12070656,
      "step": 57200
    },
    {
      "epoch": 6.293179317931793,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.025982672022760046,
      "loss": 0.2323,
      "num_input_tokens_seen": 12071680,
      "step": 57205
    },
    {
      "epoch": 6.293729372937293,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.025981691141477244,
      "loss": 0.2273,
      "num_input_tokens_seen": 12072768,
      "step": 57210
    },
    {
      "epoch": 6.2942794279427945,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.025980710158981474,
      "loss": 0.2355,
      "num_input_tokens_seen": 12073792,
      "step": 57215
    },
    {
      "epoch": 6.294829482948295,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.025979729075281763,
      "loss": 0.2376,
      "num_input_tokens_seen": 12074848,
      "step": 57220
    },
    {
      "epoch": 6.295379537953796,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.025978747890387156,
      "loss": 0.2319,
      "num_input_tokens_seen": 12075872,
      "step": 57225
    },
    {
      "epoch": 6.295929592959296,
      "grad_norm": 0.001953125,
      "learning_rate": 0.025977766604306698,
      "loss": 0.2308,
      "num_input_tokens_seen": 12076928,
      "step": 57230
    },
    {
      "epoch": 6.296479647964796,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.02597678521704943,
      "loss": 0.2302,
      "num_input_tokens_seen": 12077920,
      "step": 57235
    },
    {
      "epoch": 6.297029702970297,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.025975803728624403,
      "loss": 0.2323,
      "num_input_tokens_seen": 12079008,
      "step": 57240
    },
    {
      "epoch": 6.2975797579757975,
      "grad_norm": 0.01226806640625,
      "learning_rate": 0.025974822139040663,
      "loss": 0.2335,
      "num_input_tokens_seen": 12080096,
      "step": 57245
    },
    {
      "epoch": 6.298129812981298,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02597384044830725,
      "loss": 0.2266,
      "num_input_tokens_seen": 12081152,
      "step": 57250
    },
    {
      "epoch": 6.298679867986799,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.025972858656433218,
      "loss": 0.2276,
      "num_input_tokens_seen": 12082176,
      "step": 57255
    },
    {
      "epoch": 6.299229922992299,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.02597187676342761,
      "loss": 0.2344,
      "num_input_tokens_seen": 12083168,
      "step": 57260
    },
    {
      "epoch": 6.2997799779978,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.025970894769299483,
      "loss": 0.2365,
      "num_input_tokens_seen": 12084160,
      "step": 57265
    },
    {
      "epoch": 6.3003300330033,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.02596991267405788,
      "loss": 0.2323,
      "num_input_tokens_seen": 12085248,
      "step": 57270
    },
    {
      "epoch": 6.3008800880088005,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.025968930477711854,
      "loss": 0.2407,
      "num_input_tokens_seen": 12086304,
      "step": 57275
    },
    {
      "epoch": 6.301430143014302,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.025967948180270463,
      "loss": 0.2359,
      "num_input_tokens_seen": 12087360,
      "step": 57280
    },
    {
      "epoch": 6.301980198019802,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.025966965781742752,
      "loss": 0.229,
      "num_input_tokens_seen": 12088384,
      "step": 57285
    },
    {
      "epoch": 6.302530253025303,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.025965983282137783,
      "loss": 0.2312,
      "num_input_tokens_seen": 12089504,
      "step": 57290
    },
    {
      "epoch": 6.303080308030803,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02596500068146461,
      "loss": 0.2332,
      "num_input_tokens_seen": 12090528,
      "step": 57295
    },
    {
      "epoch": 6.303630363036303,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.025964017979732284,
      "loss": 0.2331,
      "num_input_tokens_seen": 12091552,
      "step": 57300
    },
    {
      "epoch": 6.304180418041804,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.025963035176949863,
      "loss": 0.2295,
      "num_input_tokens_seen": 12092608,
      "step": 57305
    },
    {
      "epoch": 6.304730473047305,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.025962052273126413,
      "loss": 0.2268,
      "num_input_tokens_seen": 12093664,
      "step": 57310
    },
    {
      "epoch": 6.305280528052805,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.02596106926827098,
      "loss": 0.2312,
      "num_input_tokens_seen": 12094784,
      "step": 57315
    },
    {
      "epoch": 6.305830583058306,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.025960086162392638,
      "loss": 0.2306,
      "num_input_tokens_seen": 12095776,
      "step": 57320
    },
    {
      "epoch": 6.306380638063806,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.025959102955500444,
      "loss": 0.2379,
      "num_input_tokens_seen": 12096768,
      "step": 57325
    },
    {
      "epoch": 6.306930693069307,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02595811964760345,
      "loss": 0.2294,
      "num_input_tokens_seen": 12097792,
      "step": 57330
    },
    {
      "epoch": 6.307480748074807,
      "grad_norm": 0.005859375,
      "learning_rate": 0.025957136238710724,
      "loss": 0.2306,
      "num_input_tokens_seen": 12098880,
      "step": 57335
    },
    {
      "epoch": 6.3080308030803085,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.025956152728831335,
      "loss": 0.2321,
      "num_input_tokens_seen": 12099936,
      "step": 57340
    },
    {
      "epoch": 6.308580858085809,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.025955169117974342,
      "loss": 0.2342,
      "num_input_tokens_seen": 12101056,
      "step": 57345
    },
    {
      "epoch": 6.309130913091309,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.025954185406148817,
      "loss": 0.2367,
      "num_input_tokens_seen": 12102112,
      "step": 57350
    },
    {
      "epoch": 6.30968096809681,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.025953201593363815,
      "loss": 0.2335,
      "num_input_tokens_seen": 12103168,
      "step": 57355
    },
    {
      "epoch": 6.31023102310231,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.025952217679628416,
      "loss": 0.2309,
      "num_input_tokens_seen": 12104160,
      "step": 57360
    },
    {
      "epoch": 6.31078107810781,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.025951233664951674,
      "loss": 0.2319,
      "num_input_tokens_seen": 12105184,
      "step": 57365
    },
    {
      "epoch": 6.3113311331133115,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02595024954934267,
      "loss": 0.2314,
      "num_input_tokens_seen": 12106144,
      "step": 57370
    },
    {
      "epoch": 6.311881188118812,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.025949265332810473,
      "loss": 0.2298,
      "num_input_tokens_seen": 12107200,
      "step": 57375
    },
    {
      "epoch": 6.312431243124313,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.025948281015364148,
      "loss": 0.2288,
      "num_input_tokens_seen": 12108224,
      "step": 57380
    },
    {
      "epoch": 6.312981298129813,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02594729659701277,
      "loss": 0.2351,
      "num_input_tokens_seen": 12109248,
      "step": 57385
    },
    {
      "epoch": 6.313531353135313,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.025946312077765415,
      "loss": 0.2319,
      "num_input_tokens_seen": 12110240,
      "step": 57390
    },
    {
      "epoch": 6.314081408140814,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.025945327457631152,
      "loss": 0.2294,
      "num_input_tokens_seen": 12111200,
      "step": 57395
    },
    {
      "epoch": 6.3146314631463145,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.02594434273661906,
      "loss": 0.233,
      "num_input_tokens_seen": 12112256,
      "step": 57400
    },
    {
      "epoch": 6.315181518151816,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02594335791473821,
      "loss": 0.2351,
      "num_input_tokens_seen": 12113312,
      "step": 57405
    },
    {
      "epoch": 6.315731573157316,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.025942372991997684,
      "loss": 0.2283,
      "num_input_tokens_seen": 12114304,
      "step": 57410
    },
    {
      "epoch": 6.316281628162816,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.025941387968406553,
      "loss": 0.2325,
      "num_input_tokens_seen": 12115424,
      "step": 57415
    },
    {
      "epoch": 6.316831683168317,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.025940402843973906,
      "loss": 0.2288,
      "num_input_tokens_seen": 12116416,
      "step": 57420
    },
    {
      "epoch": 6.317381738173817,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02593941761870881,
      "loss": 0.2335,
      "num_input_tokens_seen": 12117472,
      "step": 57425
    },
    {
      "epoch": 6.3179317931793175,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.025938432292620352,
      "loss": 0.2289,
      "num_input_tokens_seen": 12118592,
      "step": 57430
    },
    {
      "epoch": 6.318481848184819,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02593744686571761,
      "loss": 0.2356,
      "num_input_tokens_seen": 12119648,
      "step": 57435
    },
    {
      "epoch": 6.319031903190319,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.025936461338009673,
      "loss": 0.2309,
      "num_input_tokens_seen": 12120672,
      "step": 57440
    },
    {
      "epoch": 6.31958195819582,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02593547570950562,
      "loss": 0.234,
      "num_input_tokens_seen": 12121664,
      "step": 57445
    },
    {
      "epoch": 6.32013201320132,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02593448998021453,
      "loss": 0.2293,
      "num_input_tokens_seen": 12122720,
      "step": 57450
    },
    {
      "epoch": 6.32068206820682,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.025933504150145496,
      "loss": 0.2314,
      "num_input_tokens_seen": 12123712,
      "step": 57455
    },
    {
      "epoch": 6.321232123212321,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0259325182193076,
      "loss": 0.2314,
      "num_input_tokens_seen": 12124768,
      "step": 57460
    },
    {
      "epoch": 6.321782178217822,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.025931532187709928,
      "loss": 0.232,
      "num_input_tokens_seen": 12125824,
      "step": 57465
    },
    {
      "epoch": 6.322332233223323,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.025930546055361575,
      "loss": 0.2315,
      "num_input_tokens_seen": 12126944,
      "step": 57470
    },
    {
      "epoch": 6.322882288228823,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02592955982227162,
      "loss": 0.2315,
      "num_input_tokens_seen": 12128000,
      "step": 57475
    },
    {
      "epoch": 6.323432343234323,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.025928573488449156,
      "loss": 0.2314,
      "num_input_tokens_seen": 12129088,
      "step": 57480
    },
    {
      "epoch": 6.323982398239824,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.025927587053903273,
      "loss": 0.2304,
      "num_input_tokens_seen": 12130144,
      "step": 57485
    },
    {
      "epoch": 6.324532453245324,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.025926600518643067,
      "loss": 0.2299,
      "num_input_tokens_seen": 12131232,
      "step": 57490
    },
    {
      "epoch": 6.325082508250825,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.025925613882677626,
      "loss": 0.2299,
      "num_input_tokens_seen": 12132256,
      "step": 57495
    },
    {
      "epoch": 6.325632563256326,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02592462714601604,
      "loss": 0.2299,
      "num_input_tokens_seen": 12133344,
      "step": 57500
    },
    {
      "epoch": 6.326182618261826,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.025923640308667416,
      "loss": 0.2309,
      "num_input_tokens_seen": 12134432,
      "step": 57505
    },
    {
      "epoch": 6.326732673267327,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02592265337064083,
      "loss": 0.2299,
      "num_input_tokens_seen": 12135456,
      "step": 57510
    },
    {
      "epoch": 6.327282728272827,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.0259216663319454,
      "loss": 0.2309,
      "num_input_tokens_seen": 12136608,
      "step": 57515
    },
    {
      "epoch": 6.327832783278327,
      "grad_norm": 0.00077056884765625,
      "learning_rate": 0.02592067919259021,
      "loss": 0.2289,
      "num_input_tokens_seen": 12137664,
      "step": 57520
    },
    {
      "epoch": 6.3283828382838285,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.025919691952584357,
      "loss": 0.2315,
      "num_input_tokens_seen": 12138688,
      "step": 57525
    },
    {
      "epoch": 6.328932893289329,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.025918704611936946,
      "loss": 0.2336,
      "num_input_tokens_seen": 12139712,
      "step": 57530
    },
    {
      "epoch": 6.32948294829483,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.025917717170657068,
      "loss": 0.2299,
      "num_input_tokens_seen": 12140800,
      "step": 57535
    },
    {
      "epoch": 6.33003300330033,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.025916729628753837,
      "loss": 0.2305,
      "num_input_tokens_seen": 12141856,
      "step": 57540
    },
    {
      "epoch": 6.33058305830583,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.02591574198623634,
      "loss": 0.232,
      "num_input_tokens_seen": 12142944,
      "step": 57545
    },
    {
      "epoch": 6.331133113311331,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.025914754243113694,
      "loss": 0.2294,
      "num_input_tokens_seen": 12144032,
      "step": 57550
    },
    {
      "epoch": 6.3316831683168315,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.025913766399394994,
      "loss": 0.232,
      "num_input_tokens_seen": 12145088,
      "step": 57555
    },
    {
      "epoch": 6.332233223322333,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.025912778455089342,
      "loss": 0.234,
      "num_input_tokens_seen": 12146112,
      "step": 57560
    },
    {
      "epoch": 6.332783278327833,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02591179041020585,
      "loss": 0.2299,
      "num_input_tokens_seen": 12147168,
      "step": 57565
    },
    {
      "epoch": 6.333333333333333,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.025910802264753622,
      "loss": 0.233,
      "num_input_tokens_seen": 12148160,
      "step": 57570
    },
    {
      "epoch": 6.333883388338834,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.025909814018741765,
      "loss": 0.2315,
      "num_input_tokens_seen": 12149184,
      "step": 57575
    },
    {
      "epoch": 6.334433443344334,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02590882567217939,
      "loss": 0.2314,
      "num_input_tokens_seen": 12150240,
      "step": 57580
    },
    {
      "epoch": 6.334983498349835,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.025907837225075597,
      "loss": 0.232,
      "num_input_tokens_seen": 12151328,
      "step": 57585
    },
    {
      "epoch": 6.335533553355336,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.025906848677439507,
      "loss": 0.2305,
      "num_input_tokens_seen": 12152416,
      "step": 57590
    },
    {
      "epoch": 6.336083608360836,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.025905860029280223,
      "loss": 0.2367,
      "num_input_tokens_seen": 12153472,
      "step": 57595
    },
    {
      "epoch": 6.336633663366337,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.025904871280606867,
      "loss": 0.2288,
      "num_input_tokens_seen": 12154560,
      "step": 57600
    },
    {
      "epoch": 6.337183718371837,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.02590388243142854,
      "loss": 0.2319,
      "num_input_tokens_seen": 12155584,
      "step": 57605
    },
    {
      "epoch": 6.337733773377337,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02590289348175436,
      "loss": 0.233,
      "num_input_tokens_seen": 12156672,
      "step": 57610
    },
    {
      "epoch": 6.338283828382838,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.02590190443159344,
      "loss": 0.232,
      "num_input_tokens_seen": 12157728,
      "step": 57615
    },
    {
      "epoch": 6.338833883388339,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.025900915280954905,
      "loss": 0.233,
      "num_input_tokens_seen": 12158752,
      "step": 57620
    },
    {
      "epoch": 6.33938393839384,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.02589992602984786,
      "loss": 0.2304,
      "num_input_tokens_seen": 12159808,
      "step": 57625
    },
    {
      "epoch": 6.33993399339934,
      "grad_norm": 0.0029144287109375,
      "learning_rate": 0.02589893667828143,
      "loss": 0.2284,
      "num_input_tokens_seen": 12160864,
      "step": 57630
    },
    {
      "epoch": 6.34048404840484,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02589794722626473,
      "loss": 0.2305,
      "num_input_tokens_seen": 12161920,
      "step": 57635
    },
    {
      "epoch": 6.341034103410341,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.025896957673806877,
      "loss": 0.233,
      "num_input_tokens_seen": 12162944,
      "step": 57640
    },
    {
      "epoch": 6.341584158415841,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.025895968020916996,
      "loss": 0.2315,
      "num_input_tokens_seen": 12163968,
      "step": 57645
    },
    {
      "epoch": 6.3421342134213425,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.02589497826760421,
      "loss": 0.2294,
      "num_input_tokens_seen": 12165024,
      "step": 57650
    },
    {
      "epoch": 6.342684268426843,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.02589398841387763,
      "loss": 0.2309,
      "num_input_tokens_seen": 12166080,
      "step": 57655
    },
    {
      "epoch": 6.343234323432343,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.025892998459746387,
      "loss": 0.2288,
      "num_input_tokens_seen": 12167168,
      "step": 57660
    },
    {
      "epoch": 6.343784378437844,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.02589200840521961,
      "loss": 0.2299,
      "num_input_tokens_seen": 12168256,
      "step": 57665
    },
    {
      "epoch": 6.344334433443344,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.025891018250306413,
      "loss": 0.2311,
      "num_input_tokens_seen": 12169248,
      "step": 57670
    },
    {
      "epoch": 6.3448844884488445,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.025890027995015927,
      "loss": 0.2306,
      "num_input_tokens_seen": 12170272,
      "step": 57675
    },
    {
      "epoch": 6.3454345434543455,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.025889037639357283,
      "loss": 0.2332,
      "num_input_tokens_seen": 12171296,
      "step": 57680
    },
    {
      "epoch": 6.345984598459846,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0258880471833396,
      "loss": 0.2301,
      "num_input_tokens_seen": 12172352,
      "step": 57685
    },
    {
      "epoch": 6.346534653465347,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.02588705662697201,
      "loss": 0.2332,
      "num_input_tokens_seen": 12173376,
      "step": 57690
    },
    {
      "epoch": 6.347084708470847,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.025886065970263644,
      "loss": 0.2305,
      "num_input_tokens_seen": 12174400,
      "step": 57695
    },
    {
      "epoch": 6.347634763476347,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.025885075213223634,
      "loss": 0.229,
      "num_input_tokens_seen": 12175584,
      "step": 57700
    },
    {
      "epoch": 6.348184818481848,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.02588408435586111,
      "loss": 0.233,
      "num_input_tokens_seen": 12176608,
      "step": 57705
    },
    {
      "epoch": 6.3487348734873486,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0258830933981852,
      "loss": 0.2326,
      "num_input_tokens_seen": 12177664,
      "step": 57710
    },
    {
      "epoch": 6.34928492849285,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02588210234020504,
      "loss": 0.231,
      "num_input_tokens_seen": 12178752,
      "step": 57715
    },
    {
      "epoch": 6.34983498349835,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.025881111181929765,
      "loss": 0.2284,
      "num_input_tokens_seen": 12179776,
      "step": 57720
    },
    {
      "epoch": 6.35038503850385,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.02588011992336851,
      "loss": 0.2326,
      "num_input_tokens_seen": 12180800,
      "step": 57725
    },
    {
      "epoch": 6.350935093509351,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.025879128564530413,
      "loss": 0.2336,
      "num_input_tokens_seen": 12181856,
      "step": 57730
    },
    {
      "epoch": 6.351485148514851,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.025878137105424602,
      "loss": 0.2274,
      "num_input_tokens_seen": 12182944,
      "step": 57735
    },
    {
      "epoch": 6.3520352035203524,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.025877145546060226,
      "loss": 0.2275,
      "num_input_tokens_seen": 12184000,
      "step": 57740
    },
    {
      "epoch": 6.352585258525853,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.02587615388644642,
      "loss": 0.2358,
      "num_input_tokens_seen": 12185088,
      "step": 57745
    },
    {
      "epoch": 6.353135313531353,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.025875162126592317,
      "loss": 0.2385,
      "num_input_tokens_seen": 12186144,
      "step": 57750
    },
    {
      "epoch": 6.353685368536854,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.02587417026650707,
      "loss": 0.2316,
      "num_input_tokens_seen": 12187264,
      "step": 57755
    },
    {
      "epoch": 6.354235423542354,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.025873178306199805,
      "loss": 0.2331,
      "num_input_tokens_seen": 12188320,
      "step": 57760
    },
    {
      "epoch": 6.354785478547855,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02587218624567968,
      "loss": 0.2294,
      "num_input_tokens_seen": 12189344,
      "step": 57765
    },
    {
      "epoch": 6.3553355335533555,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.025871194084955827,
      "loss": 0.232,
      "num_input_tokens_seen": 12190400,
      "step": 57770
    },
    {
      "epoch": 6.355885588558856,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.025870201824037396,
      "loss": 0.232,
      "num_input_tokens_seen": 12191488,
      "step": 57775
    },
    {
      "epoch": 6.356435643564357,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.025869209462933536,
      "loss": 0.2309,
      "num_input_tokens_seen": 12192544,
      "step": 57780
    },
    {
      "epoch": 6.356985698569857,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.02586821700165338,
      "loss": 0.2274,
      "num_input_tokens_seen": 12193568,
      "step": 57785
    },
    {
      "epoch": 6.357535753575357,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.025867224440206088,
      "loss": 0.2258,
      "num_input_tokens_seen": 12194592,
      "step": 57790
    },
    {
      "epoch": 6.358085808580858,
      "grad_norm": 0.005859375,
      "learning_rate": 0.025866231778600802,
      "loss": 0.228,
      "num_input_tokens_seen": 12195648,
      "step": 57795
    },
    {
      "epoch": 6.3586358635863585,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.02586523901684667,
      "loss": 0.2293,
      "num_input_tokens_seen": 12196704,
      "step": 57800
    },
    {
      "epoch": 6.3591859185918596,
      "grad_norm": 0.0155029296875,
      "learning_rate": 0.02586424615495285,
      "loss": 0.2394,
      "num_input_tokens_seen": 12197856,
      "step": 57805
    },
    {
      "epoch": 6.35973597359736,
      "grad_norm": 0.01312255859375,
      "learning_rate": 0.025863253192928478,
      "loss": 0.2233,
      "num_input_tokens_seen": 12198912,
      "step": 57810
    },
    {
      "epoch": 6.36028602860286,
      "grad_norm": 0.0159912109375,
      "learning_rate": 0.02586226013078272,
      "loss": 0.2352,
      "num_input_tokens_seen": 12199936,
      "step": 57815
    },
    {
      "epoch": 6.360836083608361,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.02586126696852472,
      "loss": 0.2318,
      "num_input_tokens_seen": 12200960,
      "step": 57820
    },
    {
      "epoch": 6.361386138613861,
      "grad_norm": 0.0137939453125,
      "learning_rate": 0.025860273706163634,
      "loss": 0.2228,
      "num_input_tokens_seen": 12201984,
      "step": 57825
    },
    {
      "epoch": 6.361936193619362,
      "grad_norm": 0.01446533203125,
      "learning_rate": 0.025859280343708616,
      "loss": 0.2261,
      "num_input_tokens_seen": 12203040,
      "step": 57830
    },
    {
      "epoch": 6.362486248624863,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.025858286881168823,
      "loss": 0.2418,
      "num_input_tokens_seen": 12204096,
      "step": 57835
    },
    {
      "epoch": 6.363036303630363,
      "grad_norm": 0.00738525390625,
      "learning_rate": 0.02585729331855341,
      "loss": 0.2235,
      "num_input_tokens_seen": 12205216,
      "step": 57840
    },
    {
      "epoch": 6.363586358635864,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.025856299655871533,
      "loss": 0.2304,
      "num_input_tokens_seen": 12206240,
      "step": 57845
    },
    {
      "epoch": 6.364136413641364,
      "grad_norm": 0.0157470703125,
      "learning_rate": 0.025855305893132355,
      "loss": 0.2443,
      "num_input_tokens_seen": 12207296,
      "step": 57850
    },
    {
      "epoch": 6.364686468646864,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.025854312030345026,
      "loss": 0.2325,
      "num_input_tokens_seen": 12208352,
      "step": 57855
    },
    {
      "epoch": 6.365236523652365,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.025853318067518713,
      "loss": 0.2303,
      "num_input_tokens_seen": 12209504,
      "step": 57860
    },
    {
      "epoch": 6.365786578657866,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.025852324004662575,
      "loss": 0.2358,
      "num_input_tokens_seen": 12210592,
      "step": 57865
    },
    {
      "epoch": 6.366336633663367,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.025851329841785776,
      "loss": 0.2348,
      "num_input_tokens_seen": 12211648,
      "step": 57870
    },
    {
      "epoch": 6.366886688668867,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.02585033557889748,
      "loss": 0.2361,
      "num_input_tokens_seen": 12212736,
      "step": 57875
    },
    {
      "epoch": 6.367436743674367,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.025849341216006842,
      "loss": 0.2329,
      "num_input_tokens_seen": 12213792,
      "step": 57880
    },
    {
      "epoch": 6.367986798679868,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.025848346753123034,
      "loss": 0.2318,
      "num_input_tokens_seen": 12214880,
      "step": 57885
    },
    {
      "epoch": 6.368536853685368,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.025847352190255222,
      "loss": 0.2334,
      "num_input_tokens_seen": 12215872,
      "step": 57890
    },
    {
      "epoch": 6.3690869086908695,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.025846357527412565,
      "loss": 0.2313,
      "num_input_tokens_seen": 12216896,
      "step": 57895
    },
    {
      "epoch": 6.36963696369637,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.02584536276460424,
      "loss": 0.2307,
      "num_input_tokens_seen": 12217984,
      "step": 57900
    },
    {
      "epoch": 6.37018701870187,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.025844367901839405,
      "loss": 0.2313,
      "num_input_tokens_seen": 12219040,
      "step": 57905
    },
    {
      "epoch": 6.370737073707371,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.02584337293912724,
      "loss": 0.2302,
      "num_input_tokens_seen": 12220096,
      "step": 57910
    },
    {
      "epoch": 6.371287128712871,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.025842377876476908,
      "loss": 0.2308,
      "num_input_tokens_seen": 12221120,
      "step": 57915
    },
    {
      "epoch": 6.371837183718371,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.02584138271389758,
      "loss": 0.2313,
      "num_input_tokens_seen": 12222176,
      "step": 57920
    },
    {
      "epoch": 6.3723872387238725,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.025840387451398432,
      "loss": 0.2328,
      "num_input_tokens_seen": 12223200,
      "step": 57925
    },
    {
      "epoch": 6.372937293729373,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.025839392088988635,
      "loss": 0.2302,
      "num_input_tokens_seen": 12224224,
      "step": 57930
    },
    {
      "epoch": 6.373487348734874,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.02583839662667736,
      "loss": 0.2313,
      "num_input_tokens_seen": 12225312,
      "step": 57935
    },
    {
      "epoch": 6.374037403740374,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.025837401064473783,
      "loss": 0.2308,
      "num_input_tokens_seen": 12226336,
      "step": 57940
    },
    {
      "epoch": 6.374587458745874,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.025836405402387085,
      "loss": 0.2318,
      "num_input_tokens_seen": 12227424,
      "step": 57945
    },
    {
      "epoch": 6.375137513751375,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.025835409640426438,
      "loss": 0.2334,
      "num_input_tokens_seen": 12228448,
      "step": 57950
    },
    {
      "epoch": 6.3756875687568755,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.025834413778601014,
      "loss": 0.2292,
      "num_input_tokens_seen": 12229472,
      "step": 57955
    },
    {
      "epoch": 6.376237623762377,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.02583341781692,
      "loss": 0.2309,
      "num_input_tokens_seen": 12230528,
      "step": 57960
    },
    {
      "epoch": 6.376787678767877,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.025832421755392576,
      "loss": 0.2278,
      "num_input_tokens_seen": 12231520,
      "step": 57965
    },
    {
      "epoch": 6.377337733773377,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.02583142559402792,
      "loss": 0.232,
      "num_input_tokens_seen": 12232576,
      "step": 57970
    },
    {
      "epoch": 6.377887788778878,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.025830429332835202,
      "loss": 0.2325,
      "num_input_tokens_seen": 12233568,
      "step": 57975
    },
    {
      "epoch": 6.378437843784378,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02582943297182362,
      "loss": 0.2294,
      "num_input_tokens_seen": 12234656,
      "step": 57980
    },
    {
      "epoch": 6.378987898789879,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02582843651100235,
      "loss": 0.2357,
      "num_input_tokens_seen": 12235712,
      "step": 57985
    },
    {
      "epoch": 6.37953795379538,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.025827439950380577,
      "loss": 0.2289,
      "num_input_tokens_seen": 12236800,
      "step": 57990
    },
    {
      "epoch": 6.38008800880088,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.025826443289967485,
      "loss": 0.2299,
      "num_input_tokens_seen": 12237920,
      "step": 57995
    },
    {
      "epoch": 6.380638063806381,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02582544652977226,
      "loss": 0.2331,
      "num_input_tokens_seen": 12239008,
      "step": 58000
    },
    {
      "epoch": 6.381188118811881,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.02582444966980409,
      "loss": 0.2336,
      "num_input_tokens_seen": 12240064,
      "step": 58005
    },
    {
      "epoch": 6.381738173817382,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.02582345271007216,
      "loss": 0.2314,
      "num_input_tokens_seen": 12241056,
      "step": 58010
    },
    {
      "epoch": 6.382288228822882,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.025822455650585663,
      "loss": 0.2315,
      "num_input_tokens_seen": 12242112,
      "step": 58015
    },
    {
      "epoch": 6.382838283828383,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.025821458491353783,
      "loss": 0.2288,
      "num_input_tokens_seen": 12243168,
      "step": 58020
    },
    {
      "epoch": 6.383388338833884,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.02582046123238571,
      "loss": 0.2288,
      "num_input_tokens_seen": 12244320,
      "step": 58025
    },
    {
      "epoch": 6.383938393839384,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02581946387369064,
      "loss": 0.2325,
      "num_input_tokens_seen": 12245344,
      "step": 58030
    },
    {
      "epoch": 6.384488448844884,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.025818466415277765,
      "loss": 0.2288,
      "num_input_tokens_seen": 12246432,
      "step": 58035
    },
    {
      "epoch": 6.385038503850385,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02581746885715628,
      "loss": 0.2362,
      "num_input_tokens_seen": 12247456,
      "step": 58040
    },
    {
      "epoch": 6.385588558855885,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.025816471199335368,
      "loss": 0.2299,
      "num_input_tokens_seen": 12248512,
      "step": 58045
    },
    {
      "epoch": 6.3861386138613865,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.025815473441824232,
      "loss": 0.232,
      "num_input_tokens_seen": 12249536,
      "step": 58050
    },
    {
      "epoch": 6.386688668866887,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.02581447558463207,
      "loss": 0.234,
      "num_input_tokens_seen": 12250560,
      "step": 58055
    },
    {
      "epoch": 6.387238723872387,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.025813477627768075,
      "loss": 0.2272,
      "num_input_tokens_seen": 12251616,
      "step": 58060
    },
    {
      "epoch": 6.387788778877888,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.025812479571241446,
      "loss": 0.233,
      "num_input_tokens_seen": 12252640,
      "step": 58065
    },
    {
      "epoch": 6.388338833883388,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.02581148141506138,
      "loss": 0.231,
      "num_input_tokens_seen": 12253696,
      "step": 58070
    },
    {
      "epoch": 6.388888888888889,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.025810483159237077,
      "loss": 0.2294,
      "num_input_tokens_seen": 12254784,
      "step": 58075
    },
    {
      "epoch": 6.3894389438943895,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.025809484803777736,
      "loss": 0.2309,
      "num_input_tokens_seen": 12255776,
      "step": 58080
    },
    {
      "epoch": 6.38998899889989,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.025808486348692566,
      "loss": 0.2294,
      "num_input_tokens_seen": 12256832,
      "step": 58085
    },
    {
      "epoch": 6.390539053905391,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.025807487793990763,
      "loss": 0.2341,
      "num_input_tokens_seen": 12257920,
      "step": 58090
    },
    {
      "epoch": 6.391089108910891,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.025806489139681525,
      "loss": 0.2243,
      "num_input_tokens_seen": 12258944,
      "step": 58095
    },
    {
      "epoch": 6.391639163916391,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.02580549038577407,
      "loss": 0.2336,
      "num_input_tokens_seen": 12260000,
      "step": 58100
    },
    {
      "epoch": 6.392189218921892,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02580449153227759,
      "loss": 0.2326,
      "num_input_tokens_seen": 12261088,
      "step": 58105
    },
    {
      "epoch": 6.3927392739273925,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.025803492579201295,
      "loss": 0.2305,
      "num_input_tokens_seen": 12262080,
      "step": 58110
    },
    {
      "epoch": 6.393289328932894,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0258024935265544,
      "loss": 0.2357,
      "num_input_tokens_seen": 12263168,
      "step": 58115
    },
    {
      "epoch": 6.393839383938394,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.0258014943743461,
      "loss": 0.23,
      "num_input_tokens_seen": 12264192,
      "step": 58120
    },
    {
      "epoch": 6.394389438943894,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.025800495122585608,
      "loss": 0.2278,
      "num_input_tokens_seen": 12265280,
      "step": 58125
    },
    {
      "epoch": 6.394939493949395,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02579949577128214,
      "loss": 0.2237,
      "num_input_tokens_seen": 12266368,
      "step": 58130
    },
    {
      "epoch": 6.395489548954895,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.025798496320444897,
      "loss": 0.2265,
      "num_input_tokens_seen": 12267424,
      "step": 58135
    },
    {
      "epoch": 6.396039603960396,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.025797496770083098,
      "loss": 0.2351,
      "num_input_tokens_seen": 12268416,
      "step": 58140
    },
    {
      "epoch": 6.396589658965897,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.025796497120205947,
      "loss": 0.2366,
      "num_input_tokens_seen": 12269472,
      "step": 58145
    },
    {
      "epoch": 6.397139713971397,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02579549737082267,
      "loss": 0.2326,
      "num_input_tokens_seen": 12270528,
      "step": 58150
    },
    {
      "epoch": 6.397689768976898,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.025794497521942475,
      "loss": 0.2336,
      "num_input_tokens_seen": 12271488,
      "step": 58155
    },
    {
      "epoch": 6.398239823982398,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.02579349757357457,
      "loss": 0.2299,
      "num_input_tokens_seen": 12272608,
      "step": 58160
    },
    {
      "epoch": 6.398789878987899,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.025792497525728177,
      "loss": 0.2429,
      "num_input_tokens_seen": 12273664,
      "step": 58165
    },
    {
      "epoch": 6.399339933993399,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02579149737841252,
      "loss": 0.2256,
      "num_input_tokens_seen": 12274720,
      "step": 58170
    },
    {
      "epoch": 6.3998899889989,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.025790497131636798,
      "loss": 0.2327,
      "num_input_tokens_seen": 12275776,
      "step": 58175
    },
    {
      "epoch": 6.400440044004401,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.025789496785410253,
      "loss": 0.2281,
      "num_input_tokens_seen": 12276800,
      "step": 58180
    },
    {
      "epoch": 6.400990099009901,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.02578849633974209,
      "loss": 0.2302,
      "num_input_tokens_seen": 12277856,
      "step": 58185
    },
    {
      "epoch": 6.401540154015402,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02578749579464153,
      "loss": 0.2338,
      "num_input_tokens_seen": 12278880,
      "step": 58190
    },
    {
      "epoch": 6.402090209020902,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.025786495150117803,
      "loss": 0.2297,
      "num_input_tokens_seen": 12279936,
      "step": 58195
    },
    {
      "epoch": 6.402640264026402,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02578549440618012,
      "loss": 0.239,
      "num_input_tokens_seen": 12280960,
      "step": 58200
    },
    {
      "epoch": 6.4031903190319035,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.025784493562837716,
      "loss": 0.2255,
      "num_input_tokens_seen": 12282016,
      "step": 58205
    },
    {
      "epoch": 6.403740374037404,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02578349262009981,
      "loss": 0.2405,
      "num_input_tokens_seen": 12283104,
      "step": 58210
    },
    {
      "epoch": 6.404290429042904,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.025782491577975623,
      "loss": 0.2284,
      "num_input_tokens_seen": 12284160,
      "step": 58215
    },
    {
      "epoch": 6.404840484048405,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02578149043647438,
      "loss": 0.2294,
      "num_input_tokens_seen": 12285312,
      "step": 58220
    },
    {
      "epoch": 6.405390539053905,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.02578048919560532,
      "loss": 0.2341,
      "num_input_tokens_seen": 12286336,
      "step": 58225
    },
    {
      "epoch": 6.405940594059406,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.025779487855377665,
      "loss": 0.231,
      "num_input_tokens_seen": 12287328,
      "step": 58230
    },
    {
      "epoch": 6.4064906490649065,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.025778486415800644,
      "loss": 0.2289,
      "num_input_tokens_seen": 12288352,
      "step": 58235
    },
    {
      "epoch": 6.407040704070407,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.025777484876883482,
      "loss": 0.2315,
      "num_input_tokens_seen": 12289408,
      "step": 58240
    },
    {
      "epoch": 6.407590759075908,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.025776483238635416,
      "loss": 0.231,
      "num_input_tokens_seen": 12290464,
      "step": 58245
    },
    {
      "epoch": 6.408140814081408,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.02577548150106567,
      "loss": 0.23,
      "num_input_tokens_seen": 12291520,
      "step": 58250
    },
    {
      "epoch": 6.408690869086909,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02577447966418349,
      "loss": 0.2326,
      "num_input_tokens_seen": 12292544,
      "step": 58255
    },
    {
      "epoch": 6.409240924092409,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.025773477727998092,
      "loss": 0.231,
      "num_input_tokens_seen": 12293600,
      "step": 58260
    },
    {
      "epoch": 6.4097909790979095,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.025772475692518726,
      "loss": 0.2294,
      "num_input_tokens_seen": 12294656,
      "step": 58265
    },
    {
      "epoch": 6.410341034103411,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.025771473557754616,
      "loss": 0.2305,
      "num_input_tokens_seen": 12295680,
      "step": 58270
    },
    {
      "epoch": 6.410891089108911,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02577047132371501,
      "loss": 0.231,
      "num_input_tokens_seen": 12296704,
      "step": 58275
    },
    {
      "epoch": 6.411441144114411,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02576946899040913,
      "loss": 0.2325,
      "num_input_tokens_seen": 12297792,
      "step": 58280
    },
    {
      "epoch": 6.411991199119912,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.02576846655784622,
      "loss": 0.2279,
      "num_input_tokens_seen": 12298848,
      "step": 58285
    },
    {
      "epoch": 6.412541254125412,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02576746402603553,
      "loss": 0.2325,
      "num_input_tokens_seen": 12299968,
      "step": 58290
    },
    {
      "epoch": 6.413091309130913,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02576646139498629,
      "loss": 0.2325,
      "num_input_tokens_seen": 12300992,
      "step": 58295
    },
    {
      "epoch": 6.413641364136414,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02576545866470774,
      "loss": 0.2305,
      "num_input_tokens_seen": 12302016,
      "step": 58300
    },
    {
      "epoch": 6.414191419141914,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02576445583520912,
      "loss": 0.2284,
      "num_input_tokens_seen": 12303104,
      "step": 58305
    },
    {
      "epoch": 6.414741474147415,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.025763452906499677,
      "loss": 0.2322,
      "num_input_tokens_seen": 12304224,
      "step": 58310
    },
    {
      "epoch": 6.415291529152915,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.025762449878588656,
      "loss": 0.2326,
      "num_input_tokens_seen": 12305312,
      "step": 58315
    },
    {
      "epoch": 6.415841584158416,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02576144675148529,
      "loss": 0.23,
      "num_input_tokens_seen": 12306368,
      "step": 58320
    },
    {
      "epoch": 6.416391639163916,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.025760443525198844,
      "loss": 0.2295,
      "num_input_tokens_seen": 12307456,
      "step": 58325
    },
    {
      "epoch": 6.416941694169417,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.025759440199738547,
      "loss": 0.2357,
      "num_input_tokens_seen": 12308448,
      "step": 58330
    },
    {
      "epoch": 6.417491749174918,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.025758436775113656,
      "loss": 0.2299,
      "num_input_tokens_seen": 12309504,
      "step": 58335
    },
    {
      "epoch": 6.418041804180418,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.025757433251333416,
      "loss": 0.2289,
      "num_input_tokens_seen": 12310560,
      "step": 58340
    },
    {
      "epoch": 6.418591859185918,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02575642962840707,
      "loss": 0.2295,
      "num_input_tokens_seen": 12311616,
      "step": 58345
    },
    {
      "epoch": 6.419141914191419,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02575542590634388,
      "loss": 0.23,
      "num_input_tokens_seen": 12312704,
      "step": 58350
    },
    {
      "epoch": 6.419691969196919,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02575442208515309,
      "loss": 0.2341,
      "num_input_tokens_seen": 12313760,
      "step": 58355
    },
    {
      "epoch": 6.4202420242024205,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.025753418164843948,
      "loss": 0.2294,
      "num_input_tokens_seen": 12314848,
      "step": 58360
    },
    {
      "epoch": 6.420792079207921,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.025752414145425714,
      "loss": 0.2295,
      "num_input_tokens_seen": 12315968,
      "step": 58365
    },
    {
      "epoch": 6.421342134213421,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.025751410026907638,
      "loss": 0.2295,
      "num_input_tokens_seen": 12317024,
      "step": 58370
    },
    {
      "epoch": 6.421892189218922,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.025750405809298972,
      "loss": 0.2336,
      "num_input_tokens_seen": 12318080,
      "step": 58375
    },
    {
      "epoch": 6.422442244224422,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.025749401492608976,
      "loss": 0.2316,
      "num_input_tokens_seen": 12319072,
      "step": 58380
    },
    {
      "epoch": 6.422992299229923,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.025748397076846903,
      "loss": 0.2321,
      "num_input_tokens_seen": 12320192,
      "step": 58385
    },
    {
      "epoch": 6.4235423542354235,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.025747392562022017,
      "loss": 0.2311,
      "num_input_tokens_seen": 12321248,
      "step": 58390
    },
    {
      "epoch": 6.424092409240924,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.025746387948143566,
      "loss": 0.2357,
      "num_input_tokens_seen": 12322336,
      "step": 58395
    },
    {
      "epoch": 6.424642464246425,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.025745383235220815,
      "loss": 0.2299,
      "num_input_tokens_seen": 12323392,
      "step": 58400
    },
    {
      "epoch": 6.425192519251925,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.025744378423263024,
      "loss": 0.2356,
      "num_input_tokens_seen": 12324512,
      "step": 58405
    },
    {
      "epoch": 6.425742574257426,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.025743373512279455,
      "loss": 0.2314,
      "num_input_tokens_seen": 12325536,
      "step": 58410
    },
    {
      "epoch": 6.426292629262926,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.02574236850227936,
      "loss": 0.2345,
      "num_input_tokens_seen": 12326592,
      "step": 58415
    },
    {
      "epoch": 6.4268426842684265,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.025741363393272015,
      "loss": 0.2288,
      "num_input_tokens_seen": 12327648,
      "step": 58420
    },
    {
      "epoch": 6.427392739273928,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.025740358185266675,
      "loss": 0.2298,
      "num_input_tokens_seen": 12328704,
      "step": 58425
    },
    {
      "epoch": 6.427942794279428,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02573935287827261,
      "loss": 0.234,
      "num_input_tokens_seen": 12329728,
      "step": 58430
    },
    {
      "epoch": 6.428492849284929,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.025738347472299083,
      "loss": 0.2299,
      "num_input_tokens_seen": 12330784,
      "step": 58435
    },
    {
      "epoch": 6.429042904290429,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.025737341967355356,
      "loss": 0.2293,
      "num_input_tokens_seen": 12331808,
      "step": 58440
    },
    {
      "epoch": 6.429592959295929,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.025736336363450706,
      "loss": 0.2294,
      "num_input_tokens_seen": 12332832,
      "step": 58445
    },
    {
      "epoch": 6.43014301430143,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.025735330660594392,
      "loss": 0.2325,
      "num_input_tokens_seen": 12333856,
      "step": 58450
    },
    {
      "epoch": 6.430693069306931,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.025734324858795687,
      "loss": 0.2346,
      "num_input_tokens_seen": 12334944,
      "step": 58455
    },
    {
      "epoch": 6.431243124312431,
      "grad_norm": 0.009765625,
      "learning_rate": 0.025733318958063855,
      "loss": 0.23,
      "num_input_tokens_seen": 12335936,
      "step": 58460
    },
    {
      "epoch": 6.431793179317932,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.025732312958408183,
      "loss": 0.232,
      "num_input_tokens_seen": 12337056,
      "step": 58465
    },
    {
      "epoch": 6.432343234323432,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02573130685983793,
      "loss": 0.2289,
      "num_input_tokens_seen": 12338112,
      "step": 58470
    },
    {
      "epoch": 6.432893289328933,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.025730300662362364,
      "loss": 0.2331,
      "num_input_tokens_seen": 12339168,
      "step": 58475
    },
    {
      "epoch": 6.433443344334433,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.025729294365990772,
      "loss": 0.2305,
      "num_input_tokens_seen": 12340224,
      "step": 58480
    },
    {
      "epoch": 6.433993399339934,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02572828797073242,
      "loss": 0.2279,
      "num_input_tokens_seen": 12341216,
      "step": 58485
    },
    {
      "epoch": 6.434543454345435,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.025727281476596586,
      "loss": 0.2279,
      "num_input_tokens_seen": 12342272,
      "step": 58490
    },
    {
      "epoch": 6.435093509350935,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.025726274883592543,
      "loss": 0.2316,
      "num_input_tokens_seen": 12343296,
      "step": 58495
    },
    {
      "epoch": 6.435643564356436,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02572526819172958,
      "loss": 0.2274,
      "num_input_tokens_seen": 12344352,
      "step": 58500
    },
    {
      "epoch": 6.436193619361936,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02572426140101696,
      "loss": 0.229,
      "num_input_tokens_seen": 12345344,
      "step": 58505
    },
    {
      "epoch": 6.436743674367436,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.02572325451146397,
      "loss": 0.2295,
      "num_input_tokens_seen": 12346432,
      "step": 58510
    },
    {
      "epoch": 6.4372937293729375,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02572224752307989,
      "loss": 0.228,
      "num_input_tokens_seen": 12347520,
      "step": 58515
    },
    {
      "epoch": 6.437843784378438,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.025721240435873996,
      "loss": 0.237,
      "num_input_tokens_seen": 12348544,
      "step": 58520
    },
    {
      "epoch": 6.438393839383938,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.025720233249855576,
      "loss": 0.2337,
      "num_input_tokens_seen": 12349632,
      "step": 58525
    },
    {
      "epoch": 6.438943894389439,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02571922596503391,
      "loss": 0.2338,
      "num_input_tokens_seen": 12350720,
      "step": 58530
    },
    {
      "epoch": 6.439493949394939,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.025718218581418283,
      "loss": 0.2311,
      "num_input_tokens_seen": 12351808,
      "step": 58535
    },
    {
      "epoch": 6.44004400440044,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.025717211099017978,
      "loss": 0.2249,
      "num_input_tokens_seen": 12352832,
      "step": 58540
    },
    {
      "epoch": 6.4405940594059405,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.025716203517842276,
      "loss": 0.2316,
      "num_input_tokens_seen": 12353856,
      "step": 58545
    },
    {
      "epoch": 6.441144114411441,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.025715195837900474,
      "loss": 0.2307,
      "num_input_tokens_seen": 12354976,
      "step": 58550
    },
    {
      "epoch": 6.441694169416942,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.025714188059201854,
      "loss": 0.2365,
      "num_input_tokens_seen": 12356000,
      "step": 58555
    },
    {
      "epoch": 6.442244224422442,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.0257131801817557,
      "loss": 0.2321,
      "num_input_tokens_seen": 12357088,
      "step": 58560
    },
    {
      "epoch": 6.442794279427943,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02571217220557131,
      "loss": 0.2364,
      "num_input_tokens_seen": 12358144,
      "step": 58565
    },
    {
      "epoch": 6.443344334433443,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.025711164130657962,
      "loss": 0.2384,
      "num_input_tokens_seen": 12359264,
      "step": 58570
    },
    {
      "epoch": 6.4438943894389435,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02571015595702496,
      "loss": 0.2294,
      "num_input_tokens_seen": 12360352,
      "step": 58575
    },
    {
      "epoch": 6.444444444444445,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.025709147684681587,
      "loss": 0.2294,
      "num_input_tokens_seen": 12361376,
      "step": 58580
    },
    {
      "epoch": 6.444994499449945,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02570813931363714,
      "loss": 0.2325,
      "num_input_tokens_seen": 12362400,
      "step": 58585
    },
    {
      "epoch": 6.445544554455446,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.025707130843900906,
      "loss": 0.2336,
      "num_input_tokens_seen": 12363488,
      "step": 58590
    },
    {
      "epoch": 6.446094609460946,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.02570612227548219,
      "loss": 0.2319,
      "num_input_tokens_seen": 12364608,
      "step": 58595
    },
    {
      "epoch": 6.446644664466446,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02570511360839028,
      "loss": 0.2314,
      "num_input_tokens_seen": 12365696,
      "step": 58600
    },
    {
      "epoch": 6.447194719471947,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02570410484263448,
      "loss": 0.2309,
      "num_input_tokens_seen": 12366720,
      "step": 58605
    },
    {
      "epoch": 6.447744774477448,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.025703095978224073,
      "loss": 0.2345,
      "num_input_tokens_seen": 12367808,
      "step": 58610
    },
    {
      "epoch": 6.448294829482949,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.025702087015168375,
      "loss": 0.2319,
      "num_input_tokens_seen": 12368800,
      "step": 58615
    },
    {
      "epoch": 6.448844884488449,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.025701077953476672,
      "loss": 0.2309,
      "num_input_tokens_seen": 12369792,
      "step": 58620
    },
    {
      "epoch": 6.449394939493949,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.02570006879315827,
      "loss": 0.2325,
      "num_input_tokens_seen": 12370880,
      "step": 58625
    },
    {
      "epoch": 6.44994499449945,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.025699059534222465,
      "loss": 0.2304,
      "num_input_tokens_seen": 12371936,
      "step": 58630
    },
    {
      "epoch": 6.4504950495049505,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.025698050176678564,
      "loss": 0.2314,
      "num_input_tokens_seen": 12372960,
      "step": 58635
    },
    {
      "epoch": 6.451045104510451,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02569704072053587,
      "loss": 0.2345,
      "num_input_tokens_seen": 12373984,
      "step": 58640
    },
    {
      "epoch": 6.451595159515952,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.025696031165803684,
      "loss": 0.2309,
      "num_input_tokens_seen": 12375040,
      "step": 58645
    },
    {
      "epoch": 6.452145214521452,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.025695021512491308,
      "loss": 0.2319,
      "num_input_tokens_seen": 12376128,
      "step": 58650
    },
    {
      "epoch": 6.452695269526953,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.025694011760608055,
      "loss": 0.2298,
      "num_input_tokens_seen": 12377216,
      "step": 58655
    },
    {
      "epoch": 6.453245324532453,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.025693001910163226,
      "loss": 0.2319,
      "num_input_tokens_seen": 12378336,
      "step": 58660
    },
    {
      "epoch": 6.4537953795379535,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.025691991961166127,
      "loss": 0.2293,
      "num_input_tokens_seen": 12379424,
      "step": 58665
    },
    {
      "epoch": 6.4543454345434546,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.02569098191362607,
      "loss": 0.2324,
      "num_input_tokens_seen": 12380512,
      "step": 58670
    },
    {
      "epoch": 6.454895489548955,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.02568997176755236,
      "loss": 0.2335,
      "num_input_tokens_seen": 12381600,
      "step": 58675
    },
    {
      "epoch": 6.455445544554456,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.025688961522954315,
      "loss": 0.2303,
      "num_input_tokens_seen": 12382592,
      "step": 58680
    },
    {
      "epoch": 6.455995599559956,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02568795117984124,
      "loss": 0.2298,
      "num_input_tokens_seen": 12383648,
      "step": 58685
    },
    {
      "epoch": 6.456545654565456,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.025686940738222443,
      "loss": 0.2319,
      "num_input_tokens_seen": 12384768,
      "step": 58690
    },
    {
      "epoch": 6.457095709570957,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.02568593019810724,
      "loss": 0.2309,
      "num_input_tokens_seen": 12385760,
      "step": 58695
    },
    {
      "epoch": 6.457645764576458,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.025684919559504953,
      "loss": 0.233,
      "num_input_tokens_seen": 12386784,
      "step": 58700
    },
    {
      "epoch": 6.458195819581958,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.025683908822424887,
      "loss": 0.2298,
      "num_input_tokens_seen": 12387872,
      "step": 58705
    },
    {
      "epoch": 6.458745874587459,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.025682897986876358,
      "loss": 0.2303,
      "num_input_tokens_seen": 12388896,
      "step": 58710
    },
    {
      "epoch": 6.459295929592959,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.025681887052868686,
      "loss": 0.2308,
      "num_input_tokens_seen": 12390080,
      "step": 58715
    },
    {
      "epoch": 6.45984598459846,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.025680876020411185,
      "loss": 0.234,
      "num_input_tokens_seen": 12391136,
      "step": 58720
    },
    {
      "epoch": 6.46039603960396,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.02567986488951318,
      "loss": 0.2319,
      "num_input_tokens_seen": 12392160,
      "step": 58725
    },
    {
      "epoch": 6.460946094609461,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.025678853660183982,
      "loss": 0.2298,
      "num_input_tokens_seen": 12393248,
      "step": 58730
    },
    {
      "epoch": 6.461496149614962,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.025677842332432914,
      "loss": 0.2303,
      "num_input_tokens_seen": 12394368,
      "step": 58735
    },
    {
      "epoch": 6.462046204620462,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.025676830906269297,
      "loss": 0.2324,
      "num_input_tokens_seen": 12395328,
      "step": 58740
    },
    {
      "epoch": 6.462596259625963,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02567581938170245,
      "loss": 0.2335,
      "num_input_tokens_seen": 12396352,
      "step": 58745
    },
    {
      "epoch": 6.463146314631463,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.025674807758741702,
      "loss": 0.2304,
      "num_input_tokens_seen": 12397344,
      "step": 58750
    },
    {
      "epoch": 6.463696369636963,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.025673796037396373,
      "loss": 0.2303,
      "num_input_tokens_seen": 12398432,
      "step": 58755
    },
    {
      "epoch": 6.4642464246424645,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.025672784217675788,
      "loss": 0.2314,
      "num_input_tokens_seen": 12399456,
      "step": 58760
    },
    {
      "epoch": 6.464796479647965,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02567177229958927,
      "loss": 0.2308,
      "num_input_tokens_seen": 12400480,
      "step": 58765
    },
    {
      "epoch": 6.465346534653466,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02567076028314615,
      "loss": 0.2345,
      "num_input_tokens_seen": 12401536,
      "step": 58770
    },
    {
      "epoch": 6.465896589658966,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.02566974816835575,
      "loss": 0.2303,
      "num_input_tokens_seen": 12402656,
      "step": 58775
    },
    {
      "epoch": 6.466446644664466,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0256687359552274,
      "loss": 0.2335,
      "num_input_tokens_seen": 12403712,
      "step": 58780
    },
    {
      "epoch": 6.466996699669967,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.025667723643770433,
      "loss": 0.2298,
      "num_input_tokens_seen": 12404832,
      "step": 58785
    },
    {
      "epoch": 6.4675467546754675,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.025666711233994177,
      "loss": 0.2319,
      "num_input_tokens_seen": 12405984,
      "step": 58790
    },
    {
      "epoch": 6.468096809680969,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.025665698725907956,
      "loss": 0.2298,
      "num_input_tokens_seen": 12406976,
      "step": 58795
    },
    {
      "epoch": 6.468646864686469,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.025664686119521114,
      "loss": 0.2335,
      "num_input_tokens_seen": 12408032,
      "step": 58800
    },
    {
      "epoch": 6.469196919691969,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.025663673414842977,
      "loss": 0.2308,
      "num_input_tokens_seen": 12409088,
      "step": 58805
    },
    {
      "epoch": 6.46974697469747,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.025662660611882878,
      "loss": 0.2298,
      "num_input_tokens_seen": 12410208,
      "step": 58810
    },
    {
      "epoch": 6.47029702970297,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.025661647710650153,
      "loss": 0.2309,
      "num_input_tokens_seen": 12411264,
      "step": 58815
    },
    {
      "epoch": 6.4708470847084705,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.025660634711154134,
      "loss": 0.2314,
      "num_input_tokens_seen": 12412352,
      "step": 58820
    },
    {
      "epoch": 6.471397139713972,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.02565962161340416,
      "loss": 0.2324,
      "num_input_tokens_seen": 12413440,
      "step": 58825
    },
    {
      "epoch": 6.471947194719472,
      "grad_norm": 0.000812530517578125,
      "learning_rate": 0.02565860841740957,
      "loss": 0.2309,
      "num_input_tokens_seen": 12414464,
      "step": 58830
    },
    {
      "epoch": 6.472497249724973,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.025657595123179704,
      "loss": 0.2335,
      "num_input_tokens_seen": 12415616,
      "step": 58835
    },
    {
      "epoch": 6.473047304730473,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.025656581730723896,
      "loss": 0.233,
      "num_input_tokens_seen": 12416608,
      "step": 58840
    },
    {
      "epoch": 6.473597359735973,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.025655568240051486,
      "loss": 0.2309,
      "num_input_tokens_seen": 12417664,
      "step": 58845
    },
    {
      "epoch": 6.474147414741474,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.025654554651171817,
      "loss": 0.2325,
      "num_input_tokens_seen": 12418784,
      "step": 58850
    },
    {
      "epoch": 6.474697469746975,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.02565354096409423,
      "loss": 0.233,
      "num_input_tokens_seen": 12419840,
      "step": 58855
    },
    {
      "epoch": 6.475247524752476,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.025652527178828072,
      "loss": 0.2314,
      "num_input_tokens_seen": 12420896,
      "step": 58860
    },
    {
      "epoch": 6.475797579757976,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02565151329538268,
      "loss": 0.234,
      "num_input_tokens_seen": 12421888,
      "step": 58865
    },
    {
      "epoch": 6.476347634763476,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0256504993137674,
      "loss": 0.2319,
      "num_input_tokens_seen": 12422976,
      "step": 58870
    },
    {
      "epoch": 6.476897689768977,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02564948523399158,
      "loss": 0.2324,
      "num_input_tokens_seen": 12424064,
      "step": 58875
    },
    {
      "epoch": 6.477447744774477,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.025648471056064562,
      "loss": 0.2324,
      "num_input_tokens_seen": 12425088,
      "step": 58880
    },
    {
      "epoch": 6.477997799779978,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0256474567799957,
      "loss": 0.2308,
      "num_input_tokens_seen": 12426208,
      "step": 58885
    },
    {
      "epoch": 6.478547854785479,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.025646442405794334,
      "loss": 0.2314,
      "num_input_tokens_seen": 12427264,
      "step": 58890
    },
    {
      "epoch": 6.479097909790979,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.025645427933469817,
      "loss": 0.2324,
      "num_input_tokens_seen": 12428320,
      "step": 58895
    },
    {
      "epoch": 6.47964796479648,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.025644413363031502,
      "loss": 0.2309,
      "num_input_tokens_seen": 12429408,
      "step": 58900
    },
    {
      "epoch": 6.48019801980198,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.025643398694488735,
      "loss": 0.2314,
      "num_input_tokens_seen": 12430432,
      "step": 58905
    },
    {
      "epoch": 6.48074807480748,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.02564238392785087,
      "loss": 0.2299,
      "num_input_tokens_seen": 12431488,
      "step": 58910
    },
    {
      "epoch": 6.4812981298129815,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.025641369063127258,
      "loss": 0.2325,
      "num_input_tokens_seen": 12432608,
      "step": 58915
    },
    {
      "epoch": 6.481848184818482,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.02564035410032726,
      "loss": 0.2288,
      "num_input_tokens_seen": 12433664,
      "step": 58920
    },
    {
      "epoch": 6.482398239823983,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.025639339039460213,
      "loss": 0.2331,
      "num_input_tokens_seen": 12434752,
      "step": 58925
    },
    {
      "epoch": 6.482948294829483,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.02563832388053549,
      "loss": 0.2299,
      "num_input_tokens_seen": 12435808,
      "step": 58930
    },
    {
      "epoch": 6.483498349834983,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02563730862356244,
      "loss": 0.2283,
      "num_input_tokens_seen": 12436864,
      "step": 58935
    },
    {
      "epoch": 6.484048404840484,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.025636293268550422,
      "loss": 0.2309,
      "num_input_tokens_seen": 12437920,
      "step": 58940
    },
    {
      "epoch": 6.4845984598459845,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.025635277815508787,
      "loss": 0.2314,
      "num_input_tokens_seen": 12439008,
      "step": 58945
    },
    {
      "epoch": 6.485148514851485,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.025634262264446907,
      "loss": 0.2309,
      "num_input_tokens_seen": 12440032,
      "step": 58950
    },
    {
      "epoch": 6.485698569856986,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02563324661537413,
      "loss": 0.2283,
      "num_input_tokens_seen": 12441088,
      "step": 58955
    },
    {
      "epoch": 6.486248624862486,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02563223086829982,
      "loss": 0.232,
      "num_input_tokens_seen": 12442112,
      "step": 58960
    },
    {
      "epoch": 6.486798679867987,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.025631215023233345,
      "loss": 0.2309,
      "num_input_tokens_seen": 12443168,
      "step": 58965
    },
    {
      "epoch": 6.487348734873487,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02563019908018406,
      "loss": 0.233,
      "num_input_tokens_seen": 12444288,
      "step": 58970
    },
    {
      "epoch": 6.4878987898789875,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02562918303916133,
      "loss": 0.2325,
      "num_input_tokens_seen": 12445344,
      "step": 58975
    },
    {
      "epoch": 6.488448844884489,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.025628166900174523,
      "loss": 0.2305,
      "num_input_tokens_seen": 12446368,
      "step": 58980
    },
    {
      "epoch": 6.488998899889989,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.025627150663232998,
      "loss": 0.2289,
      "num_input_tokens_seen": 12447424,
      "step": 58985
    },
    {
      "epoch": 6.48954895489549,
      "grad_norm": 0.0123291015625,
      "learning_rate": 0.025626134328346123,
      "loss": 0.2304,
      "num_input_tokens_seen": 12448416,
      "step": 58990
    },
    {
      "epoch": 6.49009900990099,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.025625117895523273,
      "loss": 0.2341,
      "num_input_tokens_seen": 12449504,
      "step": 58995
    },
    {
      "epoch": 6.49064906490649,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.025624101364773805,
      "loss": 0.2341,
      "num_input_tokens_seen": 12450560,
      "step": 59000
    },
    {
      "epoch": 6.491199119911991,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.025623084736107093,
      "loss": 0.2314,
      "num_input_tokens_seen": 12451616,
      "step": 59005
    },
    {
      "epoch": 6.491749174917492,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02562206800953251,
      "loss": 0.2351,
      "num_input_tokens_seen": 12452672,
      "step": 59010
    },
    {
      "epoch": 6.492299229922993,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.025621051185059418,
      "loss": 0.2319,
      "num_input_tokens_seen": 12453696,
      "step": 59015
    },
    {
      "epoch": 6.492849284928493,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.0256200342626972,
      "loss": 0.2319,
      "num_input_tokens_seen": 12454688,
      "step": 59020
    },
    {
      "epoch": 6.493399339933993,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.02561901724245522,
      "loss": 0.2319,
      "num_input_tokens_seen": 12455776,
      "step": 59025
    },
    {
      "epoch": 6.493949394939494,
      "grad_norm": 0.00060272216796875,
      "learning_rate": 0.02561800012434285,
      "loss": 0.2309,
      "num_input_tokens_seen": 12456864,
      "step": 59030
    },
    {
      "epoch": 6.494499449944994,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02561698290836947,
      "loss": 0.2325,
      "num_input_tokens_seen": 12457920,
      "step": 59035
    },
    {
      "epoch": 6.4950495049504955,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.025615965594544453,
      "loss": 0.2304,
      "num_input_tokens_seen": 12459040,
      "step": 59040
    },
    {
      "epoch": 6.495599559955996,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.025614948182877177,
      "loss": 0.2324,
      "num_input_tokens_seen": 12460128,
      "step": 59045
    },
    {
      "epoch": 6.496149614961496,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.025613930673377017,
      "loss": 0.2308,
      "num_input_tokens_seen": 12461216,
      "step": 59050
    },
    {
      "epoch": 6.496699669966997,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.025612913066053348,
      "loss": 0.2308,
      "num_input_tokens_seen": 12462336,
      "step": 59055
    },
    {
      "epoch": 6.497249724972497,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.025611895360915553,
      "loss": 0.2314,
      "num_input_tokens_seen": 12463392,
      "step": 59060
    },
    {
      "epoch": 6.497799779977997,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.025610877557973012,
      "loss": 0.2324,
      "num_input_tokens_seen": 12464384,
      "step": 59065
    },
    {
      "epoch": 6.4983498349834985,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.025609859657235102,
      "loss": 0.2335,
      "num_input_tokens_seen": 12465472,
      "step": 59070
    },
    {
      "epoch": 6.498899889988999,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.025608841658711207,
      "loss": 0.2314,
      "num_input_tokens_seen": 12466528,
      "step": 59075
    },
    {
      "epoch": 6.4994499449945,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02560782356241071,
      "loss": 0.2314,
      "num_input_tokens_seen": 12467552,
      "step": 59080
    },
    {
      "epoch": 6.5,
      "grad_norm": 0.005859375,
      "learning_rate": 0.025606805368343,
      "loss": 0.2304,
      "num_input_tokens_seen": 12468640,
      "step": 59085
    },
    {
      "epoch": 6.5005500550055,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.025605787076517443,
      "loss": 0.2303,
      "num_input_tokens_seen": 12469664,
      "step": 59090
    },
    {
      "epoch": 6.501100110011001,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02560476868694344,
      "loss": 0.2324,
      "num_input_tokens_seen": 12470720,
      "step": 59095
    },
    {
      "epoch": 6.5016501650165015,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.025603750199630373,
      "loss": 0.2303,
      "num_input_tokens_seen": 12471808,
      "step": 59100
    },
    {
      "epoch": 6.502200220022003,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02560273161458763,
      "loss": 0.2314,
      "num_input_tokens_seen": 12472832,
      "step": 59105
    },
    {
      "epoch": 6.502750275027503,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.025601712931824595,
      "loss": 0.2308,
      "num_input_tokens_seen": 12473920,
      "step": 59110
    },
    {
      "epoch": 6.503300330033003,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.025600694151350665,
      "loss": 0.2309,
      "num_input_tokens_seen": 12475008,
      "step": 59115
    },
    {
      "epoch": 6.503850385038504,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.025599675273175214,
      "loss": 0.2298,
      "num_input_tokens_seen": 12476064,
      "step": 59120
    },
    {
      "epoch": 6.504400440044004,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.02559865629730765,
      "loss": 0.2299,
      "num_input_tokens_seen": 12477152,
      "step": 59125
    },
    {
      "epoch": 6.5049504950495045,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.025597637223757352,
      "loss": 0.233,
      "num_input_tokens_seen": 12478240,
      "step": 59130
    },
    {
      "epoch": 6.505500550055006,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.025596618052533715,
      "loss": 0.2298,
      "num_input_tokens_seen": 12479296,
      "step": 59135
    },
    {
      "epoch": 6.506050605060506,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.025595598783646145,
      "loss": 0.2351,
      "num_input_tokens_seen": 12480416,
      "step": 59140
    },
    {
      "epoch": 6.506600660066007,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.025594579417104014,
      "loss": 0.2299,
      "num_input_tokens_seen": 12481472,
      "step": 59145
    },
    {
      "epoch": 6.507150715071507,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.025593559952916733,
      "loss": 0.2319,
      "num_input_tokens_seen": 12482496,
      "step": 59150
    },
    {
      "epoch": 6.507700770077007,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.025592540391093696,
      "loss": 0.2309,
      "num_input_tokens_seen": 12483552,
      "step": 59155
    },
    {
      "epoch": 6.508250825082508,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.025591520731644297,
      "loss": 0.2304,
      "num_input_tokens_seen": 12484704,
      "step": 59160
    },
    {
      "epoch": 6.508800880088009,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.025590500974577927,
      "loss": 0.2324,
      "num_input_tokens_seen": 12485728,
      "step": 59165
    },
    {
      "epoch": 6.50935093509351,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.025589481119903994,
      "loss": 0.2309,
      "num_input_tokens_seen": 12486720,
      "step": 59170
    },
    {
      "epoch": 6.50990099009901,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.025588461167631894,
      "loss": 0.2304,
      "num_input_tokens_seen": 12487808,
      "step": 59175
    },
    {
      "epoch": 6.51045104510451,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.025587441117771028,
      "loss": 0.2288,
      "num_input_tokens_seen": 12488800,
      "step": 59180
    },
    {
      "epoch": 6.511001100110011,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.025586420970330804,
      "loss": 0.2304,
      "num_input_tokens_seen": 12489856,
      "step": 59185
    },
    {
      "epoch": 6.511551155115511,
      "grad_norm": 0.01220703125,
      "learning_rate": 0.02558540072532061,
      "loss": 0.2351,
      "num_input_tokens_seen": 12490848,
      "step": 59190
    },
    {
      "epoch": 6.512101210121012,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.025584380382749863,
      "loss": 0.2335,
      "num_input_tokens_seen": 12491872,
      "step": 59195
    },
    {
      "epoch": 6.512651265126513,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.025583359942627955,
      "loss": 0.2314,
      "num_input_tokens_seen": 12492960,
      "step": 59200
    },
    {
      "epoch": 6.513201320132013,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0255823394049643,
      "loss": 0.2294,
      "num_input_tokens_seen": 12494048,
      "step": 59205
    },
    {
      "epoch": 6.513751375137514,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.0255813187697683,
      "loss": 0.2314,
      "num_input_tokens_seen": 12495136,
      "step": 59210
    },
    {
      "epoch": 6.514301430143014,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.025580298037049365,
      "loss": 0.2304,
      "num_input_tokens_seen": 12496224,
      "step": 59215
    },
    {
      "epoch": 6.514851485148515,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.0255792772068169,
      "loss": 0.2341,
      "num_input_tokens_seen": 12497248,
      "step": 59220
    },
    {
      "epoch": 6.5154015401540155,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02557825627908031,
      "loss": 0.2294,
      "num_input_tokens_seen": 12498336,
      "step": 59225
    },
    {
      "epoch": 6.515951595159516,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.025577235253849008,
      "loss": 0.2314,
      "num_input_tokens_seen": 12499328,
      "step": 59230
    },
    {
      "epoch": 6.516501650165017,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.025576214131132407,
      "loss": 0.2325,
      "num_input_tokens_seen": 12500352,
      "step": 59235
    },
    {
      "epoch": 6.517051705170517,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.025575192910939914,
      "loss": 0.2314,
      "num_input_tokens_seen": 12501376,
      "step": 59240
    },
    {
      "epoch": 6.517601760176017,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.025574171593280938,
      "loss": 0.2314,
      "num_input_tokens_seen": 12502432,
      "step": 59245
    },
    {
      "epoch": 6.518151815181518,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.025573150178164902,
      "loss": 0.2325,
      "num_input_tokens_seen": 12503520,
      "step": 59250
    },
    {
      "epoch": 6.5187018701870185,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.025572128665601213,
      "loss": 0.2325,
      "num_input_tokens_seen": 12504640,
      "step": 59255
    },
    {
      "epoch": 6.51925192519252,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.025571107055599285,
      "loss": 0.233,
      "num_input_tokens_seen": 12505664,
      "step": 59260
    },
    {
      "epoch": 6.51980198019802,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.025570085348168535,
      "loss": 0.2324,
      "num_input_tokens_seen": 12506816,
      "step": 59265
    },
    {
      "epoch": 6.52035203520352,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.025569063543318383,
      "loss": 0.2309,
      "num_input_tokens_seen": 12507904,
      "step": 59270
    },
    {
      "epoch": 6.520902090209021,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02556804164105824,
      "loss": 0.2303,
      "num_input_tokens_seen": 12508928,
      "step": 59275
    },
    {
      "epoch": 6.521452145214521,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.025567019641397536,
      "loss": 0.2319,
      "num_input_tokens_seen": 12509952,
      "step": 59280
    },
    {
      "epoch": 6.522002200220022,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.02556599754434568,
      "loss": 0.2314,
      "num_input_tokens_seen": 12511040,
      "step": 59285
    },
    {
      "epoch": 6.522552255225523,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02556497534991209,
      "loss": 0.2283,
      "num_input_tokens_seen": 12512096,
      "step": 59290
    },
    {
      "epoch": 6.523102310231023,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.025563953058106197,
      "loss": 0.2325,
      "num_input_tokens_seen": 12513152,
      "step": 59295
    },
    {
      "epoch": 6.523652365236524,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.025562930668937417,
      "loss": 0.2314,
      "num_input_tokens_seen": 12514144,
      "step": 59300
    },
    {
      "epoch": 6.524202420242024,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.025561908182415174,
      "loss": 0.2341,
      "num_input_tokens_seen": 12515200,
      "step": 59305
    },
    {
      "epoch": 6.524752475247524,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.025560885598548886,
      "loss": 0.233,
      "num_input_tokens_seen": 12516256,
      "step": 59310
    },
    {
      "epoch": 6.525302530253025,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.025559862917347985,
      "loss": 0.2335,
      "num_input_tokens_seen": 12517312,
      "step": 59315
    },
    {
      "epoch": 6.525852585258526,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0255588401388219,
      "loss": 0.2304,
      "num_input_tokens_seen": 12518368,
      "step": 59320
    },
    {
      "epoch": 6.526402640264027,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.025557817262980047,
      "loss": 0.2288,
      "num_input_tokens_seen": 12519424,
      "step": 59325
    },
    {
      "epoch": 6.526952695269527,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02555679428983186,
      "loss": 0.2314,
      "num_input_tokens_seen": 12520576,
      "step": 59330
    },
    {
      "epoch": 6.527502750275027,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.025555771219386764,
      "loss": 0.2288,
      "num_input_tokens_seen": 12521664,
      "step": 59335
    },
    {
      "epoch": 6.528052805280528,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02555474805165419,
      "loss": 0.2304,
      "num_input_tokens_seen": 12522688,
      "step": 59340
    },
    {
      "epoch": 6.528602860286028,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.025553724786643573,
      "loss": 0.2325,
      "num_input_tokens_seen": 12523744,
      "step": 59345
    },
    {
      "epoch": 6.5291529152915295,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.025552701424364332,
      "loss": 0.2304,
      "num_input_tokens_seen": 12524832,
      "step": 59350
    },
    {
      "epoch": 6.52970297029703,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.02555167796482591,
      "loss": 0.2288,
      "num_input_tokens_seen": 12525888,
      "step": 59355
    },
    {
      "epoch": 6.53025302530253,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.02555065440803773,
      "loss": 0.2325,
      "num_input_tokens_seen": 12527008,
      "step": 59360
    },
    {
      "epoch": 6.530803080308031,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02554963075400924,
      "loss": 0.2309,
      "num_input_tokens_seen": 12528064,
      "step": 59365
    },
    {
      "epoch": 6.531353135313531,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.025548607002749857,
      "loss": 0.2309,
      "num_input_tokens_seen": 12529120,
      "step": 59370
    },
    {
      "epoch": 6.531903190319031,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02554758315426903,
      "loss": 0.2361,
      "num_input_tokens_seen": 12530176,
      "step": 59375
    },
    {
      "epoch": 6.5324532453245325,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.025546559208576188,
      "loss": 0.2335,
      "num_input_tokens_seen": 12531264,
      "step": 59380
    },
    {
      "epoch": 6.533003300330033,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.025545535165680772,
      "loss": 0.2335,
      "num_input_tokens_seen": 12532320,
      "step": 59385
    },
    {
      "epoch": 6.533553355335534,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.025544511025592215,
      "loss": 0.2319,
      "num_input_tokens_seen": 12533344,
      "step": 59390
    },
    {
      "epoch": 6.534103410341034,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.025543486788319962,
      "loss": 0.2319,
      "num_input_tokens_seen": 12534400,
      "step": 59395
    },
    {
      "epoch": 6.534653465346535,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02554246245387345,
      "loss": 0.2293,
      "num_input_tokens_seen": 12535392,
      "step": 59400
    },
    {
      "epoch": 6.535203520352035,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.025541438022262126,
      "loss": 0.2325,
      "num_input_tokens_seen": 12536512,
      "step": 59405
    },
    {
      "epoch": 6.5357535753575355,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.02554041349349542,
      "loss": 0.2309,
      "num_input_tokens_seen": 12537504,
      "step": 59410
    },
    {
      "epoch": 6.536303630363037,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02553938886758278,
      "loss": 0.2304,
      "num_input_tokens_seen": 12538528,
      "step": 59415
    },
    {
      "epoch": 6.536853685368537,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.02553836414453365,
      "loss": 0.233,
      "num_input_tokens_seen": 12539584,
      "step": 59420
    },
    {
      "epoch": 6.537403740374037,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02553733932435748,
      "loss": 0.2336,
      "num_input_tokens_seen": 12540640,
      "step": 59425
    },
    {
      "epoch": 6.537953795379538,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.02553631440706371,
      "loss": 0.233,
      "num_input_tokens_seen": 12541728,
      "step": 59430
    },
    {
      "epoch": 6.538503850385038,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.025535289392661777,
      "loss": 0.2351,
      "num_input_tokens_seen": 12542752,
      "step": 59435
    },
    {
      "epoch": 6.539053905390539,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02553426428116114,
      "loss": 0.2308,
      "num_input_tokens_seen": 12543776,
      "step": 59440
    },
    {
      "epoch": 6.53960396039604,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02553323907257125,
      "loss": 0.2304,
      "num_input_tokens_seen": 12544800,
      "step": 59445
    },
    {
      "epoch": 6.54015401540154,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.025532213766901544,
      "loss": 0.2298,
      "num_input_tokens_seen": 12545856,
      "step": 59450
    },
    {
      "epoch": 6.540704070407041,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02553118836416148,
      "loss": 0.2319,
      "num_input_tokens_seen": 12546944,
      "step": 59455
    },
    {
      "epoch": 6.541254125412541,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.025530162864360505,
      "loss": 0.2319,
      "num_input_tokens_seen": 12548032,
      "step": 59460
    },
    {
      "epoch": 6.541804180418042,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.02552913726750807,
      "loss": 0.2318,
      "num_input_tokens_seen": 12549120,
      "step": 59465
    },
    {
      "epoch": 6.542354235423542,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.025528111573613635,
      "loss": 0.2324,
      "num_input_tokens_seen": 12550240,
      "step": 59470
    },
    {
      "epoch": 6.542904290429043,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.025527085782686646,
      "loss": 0.2308,
      "num_input_tokens_seen": 12551296,
      "step": 59475
    },
    {
      "epoch": 6.543454345434544,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.025526059894736556,
      "loss": 0.2325,
      "num_input_tokens_seen": 12552288,
      "step": 59480
    },
    {
      "epoch": 6.544004400440044,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.025525033909772822,
      "loss": 0.2309,
      "num_input_tokens_seen": 12553312,
      "step": 59485
    },
    {
      "epoch": 6.544554455445544,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.025524007827804902,
      "loss": 0.2319,
      "num_input_tokens_seen": 12554400,
      "step": 59490
    },
    {
      "epoch": 6.545104510451045,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.02552298164884225,
      "loss": 0.2304,
      "num_input_tokens_seen": 12555424,
      "step": 59495
    },
    {
      "epoch": 6.5456545654565454,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.025521955372894332,
      "loss": 0.2315,
      "num_input_tokens_seen": 12556448,
      "step": 59500
    },
    {
      "epoch": 6.5462046204620465,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.025520928999970598,
      "loss": 0.2288,
      "num_input_tokens_seen": 12557568,
      "step": 59505
    },
    {
      "epoch": 6.546754675467547,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.025519902530080504,
      "loss": 0.2325,
      "num_input_tokens_seen": 12558528,
      "step": 59510
    },
    {
      "epoch": 6.547304730473047,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.02551887596323352,
      "loss": 0.2325,
      "num_input_tokens_seen": 12559648,
      "step": 59515
    },
    {
      "epoch": 6.547854785478548,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.025517849299439108,
      "loss": 0.2314,
      "num_input_tokens_seen": 12560672,
      "step": 59520
    },
    {
      "epoch": 6.548404840484048,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.02551682253870672,
      "loss": 0.233,
      "num_input_tokens_seen": 12561696,
      "step": 59525
    },
    {
      "epoch": 6.548954895489549,
      "grad_norm": 0.009765625,
      "learning_rate": 0.025515795681045825,
      "loss": 0.2304,
      "num_input_tokens_seen": 12562816,
      "step": 59530
    },
    {
      "epoch": 6.5495049504950495,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02551476872646589,
      "loss": 0.2319,
      "num_input_tokens_seen": 12563840,
      "step": 59535
    },
    {
      "epoch": 6.55005500550055,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.025513741674976376,
      "loss": 0.2319,
      "num_input_tokens_seen": 12564960,
      "step": 59540
    },
    {
      "epoch": 6.550605060506051,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.025512714526586752,
      "loss": 0.2319,
      "num_input_tokens_seen": 12565984,
      "step": 59545
    },
    {
      "epoch": 6.551155115511551,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.02551168728130648,
      "loss": 0.2288,
      "num_input_tokens_seen": 12567072,
      "step": 59550
    },
    {
      "epoch": 6.551705170517051,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.02551065993914503,
      "loss": 0.2324,
      "num_input_tokens_seen": 12568128,
      "step": 59555
    },
    {
      "epoch": 6.552255225522552,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.025509632500111873,
      "loss": 0.2335,
      "num_input_tokens_seen": 12569280,
      "step": 59560
    },
    {
      "epoch": 6.552805280528053,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.025508604964216475,
      "loss": 0.2304,
      "num_input_tokens_seen": 12570336,
      "step": 59565
    },
    {
      "epoch": 6.553355335533554,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.025507577331468308,
      "loss": 0.233,
      "num_input_tokens_seen": 12571360,
      "step": 59570
    },
    {
      "epoch": 6.553905390539054,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.025506549601876843,
      "loss": 0.233,
      "num_input_tokens_seen": 12572416,
      "step": 59575
    },
    {
      "epoch": 6.554455445544555,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.025505521775451553,
      "loss": 0.234,
      "num_input_tokens_seen": 12573440,
      "step": 59580
    },
    {
      "epoch": 6.555005500550055,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.025504493852201906,
      "loss": 0.2309,
      "num_input_tokens_seen": 12574528,
      "step": 59585
    },
    {
      "epoch": 6.555555555555555,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.025503465832137386,
      "loss": 0.2304,
      "num_input_tokens_seen": 12575584,
      "step": 59590
    },
    {
      "epoch": 6.5561056105610565,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.025502437715267452,
      "loss": 0.234,
      "num_input_tokens_seen": 12576608,
      "step": 59595
    },
    {
      "epoch": 6.556655665566557,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.025501409501601596,
      "loss": 0.2309,
      "num_input_tokens_seen": 12577600,
      "step": 59600
    },
    {
      "epoch": 6.557205720572057,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.025500381191149286,
      "loss": 0.2288,
      "num_input_tokens_seen": 12578592,
      "step": 59605
    },
    {
      "epoch": 6.557755775577558,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.025499352783920005,
      "loss": 0.2335,
      "num_input_tokens_seen": 12579648,
      "step": 59610
    },
    {
      "epoch": 6.558305830583058,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.025498324279923223,
      "loss": 0.2309,
      "num_input_tokens_seen": 12580704,
      "step": 59615
    },
    {
      "epoch": 6.558855885588558,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.025497295679168426,
      "loss": 0.2315,
      "num_input_tokens_seen": 12581664,
      "step": 59620
    },
    {
      "epoch": 6.5594059405940595,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.02549626698166509,
      "loss": 0.2293,
      "num_input_tokens_seen": 12582720,
      "step": 59625
    },
    {
      "epoch": 6.55995599559956,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.0254952381874227,
      "loss": 0.2299,
      "num_input_tokens_seen": 12583776,
      "step": 59630
    },
    {
      "epoch": 6.560506050605061,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.025494209296450732,
      "loss": 0.2346,
      "num_input_tokens_seen": 12584768,
      "step": 59635
    },
    {
      "epoch": 6.561056105610561,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.025493180308758676,
      "loss": 0.2341,
      "num_input_tokens_seen": 12585824,
      "step": 59640
    },
    {
      "epoch": 6.561606160616062,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02549215122435601,
      "loss": 0.2356,
      "num_input_tokens_seen": 12586912,
      "step": 59645
    },
    {
      "epoch": 6.562156215621562,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.02549112204325222,
      "loss": 0.2325,
      "num_input_tokens_seen": 12587968,
      "step": 59650
    },
    {
      "epoch": 6.5627062706270625,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02549009276545679,
      "loss": 0.2283,
      "num_input_tokens_seen": 12589024,
      "step": 59655
    },
    {
      "epoch": 6.563256325632564,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.025489063390979217,
      "loss": 0.2298,
      "num_input_tokens_seen": 12590048,
      "step": 59660
    },
    {
      "epoch": 6.563806380638064,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.025488033919828974,
      "loss": 0.234,
      "num_input_tokens_seen": 12591072,
      "step": 59665
    },
    {
      "epoch": 6.564356435643564,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.025487004352015556,
      "loss": 0.2283,
      "num_input_tokens_seen": 12592096,
      "step": 59670
    },
    {
      "epoch": 6.564906490649065,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.025485974687548454,
      "loss": 0.23,
      "num_input_tokens_seen": 12593120,
      "step": 59675
    },
    {
      "epoch": 6.565456545654565,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02548494492643715,
      "loss": 0.2297,
      "num_input_tokens_seen": 12594112,
      "step": 59680
    },
    {
      "epoch": 6.566006600660066,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.025483915068691146,
      "loss": 0.229,
      "num_input_tokens_seen": 12595136,
      "step": 59685
    },
    {
      "epoch": 6.566556655665567,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.025482885114319922,
      "loss": 0.2339,
      "num_input_tokens_seen": 12596320,
      "step": 59690
    },
    {
      "epoch": 6.567106710671067,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.025481855063332975,
      "loss": 0.2313,
      "num_input_tokens_seen": 12597376,
      "step": 59695
    },
    {
      "epoch": 6.567656765676568,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.025480824915739806,
      "loss": 0.2333,
      "num_input_tokens_seen": 12598432,
      "step": 59700
    },
    {
      "epoch": 6.568206820682068,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0254797946715499,
      "loss": 0.2365,
      "num_input_tokens_seen": 12599456,
      "step": 59705
    },
    {
      "epoch": 6.568756875687569,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.025478764330772754,
      "loss": 0.2348,
      "num_input_tokens_seen": 12600512,
      "step": 59710
    },
    {
      "epoch": 6.569306930693069,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.025477733893417863,
      "loss": 0.2332,
      "num_input_tokens_seen": 12601600,
      "step": 59715
    },
    {
      "epoch": 6.56985698569857,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.025476703359494732,
      "loss": 0.2347,
      "num_input_tokens_seen": 12602592,
      "step": 59720
    },
    {
      "epoch": 6.570407040704071,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.025475672729012848,
      "loss": 0.2305,
      "num_input_tokens_seen": 12603616,
      "step": 59725
    },
    {
      "epoch": 6.570957095709571,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.025474642001981723,
      "loss": 0.233,
      "num_input_tokens_seen": 12604672,
      "step": 59730
    },
    {
      "epoch": 6.571507150715071,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.02547361117841084,
      "loss": 0.2324,
      "num_input_tokens_seen": 12605696,
      "step": 59735
    },
    {
      "epoch": 6.572057205720572,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.025472580258309712,
      "loss": 0.2314,
      "num_input_tokens_seen": 12606752,
      "step": 59740
    },
    {
      "epoch": 6.572607260726072,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.025471549241687836,
      "loss": 0.233,
      "num_input_tokens_seen": 12607808,
      "step": 59745
    },
    {
      "epoch": 6.5731573157315735,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.02547051812855472,
      "loss": 0.2319,
      "num_input_tokens_seen": 12608928,
      "step": 59750
    },
    {
      "epoch": 6.573707370737074,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.025469486918919856,
      "loss": 0.2308,
      "num_input_tokens_seen": 12610016,
      "step": 59755
    },
    {
      "epoch": 6.574257425742574,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.02546845561279276,
      "loss": 0.2319,
      "num_input_tokens_seen": 12611168,
      "step": 59760
    },
    {
      "epoch": 6.574807480748075,
      "grad_norm": 0.000659942626953125,
      "learning_rate": 0.02546742421018293,
      "loss": 0.2313,
      "num_input_tokens_seen": 12612160,
      "step": 59765
    },
    {
      "epoch": 6.575357535753575,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.02546639271109987,
      "loss": 0.2308,
      "num_input_tokens_seen": 12613216,
      "step": 59770
    },
    {
      "epoch": 6.575907590759076,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.025465361115553097,
      "loss": 0.2324,
      "num_input_tokens_seen": 12614272,
      "step": 59775
    },
    {
      "epoch": 6.5764576457645765,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.025464329423552107,
      "loss": 0.2313,
      "num_input_tokens_seen": 12615296,
      "step": 59780
    },
    {
      "epoch": 6.577007700770077,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02546329763510642,
      "loss": 0.2318,
      "num_input_tokens_seen": 12616288,
      "step": 59785
    },
    {
      "epoch": 6.577557755775578,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.025462265750225536,
      "loss": 0.2329,
      "num_input_tokens_seen": 12617344,
      "step": 59790
    },
    {
      "epoch": 6.578107810781078,
      "grad_norm": 0.000644683837890625,
      "learning_rate": 0.02546123376891897,
      "loss": 0.2303,
      "num_input_tokens_seen": 12618304,
      "step": 59795
    },
    {
      "epoch": 6.578657865786578,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02546020169119623,
      "loss": 0.2303,
      "num_input_tokens_seen": 12619360,
      "step": 59800
    },
    {
      "epoch": 6.579207920792079,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.02545916951706683,
      "loss": 0.2308,
      "num_input_tokens_seen": 12620416,
      "step": 59805
    },
    {
      "epoch": 6.5797579757975795,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.025458137246540287,
      "loss": 0.2324,
      "num_input_tokens_seen": 12621536,
      "step": 59810
    },
    {
      "epoch": 6.580308030803081,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.025457104879626106,
      "loss": 0.2314,
      "num_input_tokens_seen": 12622560,
      "step": 59815
    },
    {
      "epoch": 6.580858085808581,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.02545607241633381,
      "loss": 0.2303,
      "num_input_tokens_seen": 12623648,
      "step": 59820
    },
    {
      "epoch": 6.581408140814082,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.025455039856672915,
      "loss": 0.2309,
      "num_input_tokens_seen": 12624704,
      "step": 59825
    },
    {
      "epoch": 6.581958195819582,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02545400720065293,
      "loss": 0.234,
      "num_input_tokens_seen": 12625760,
      "step": 59830
    },
    {
      "epoch": 6.582508250825082,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.02545297444828338,
      "loss": 0.2298,
      "num_input_tokens_seen": 12626816,
      "step": 59835
    },
    {
      "epoch": 6.583058305830583,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02545194159957378,
      "loss": 0.2303,
      "num_input_tokens_seen": 12627904,
      "step": 59840
    },
    {
      "epoch": 6.583608360836084,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.025450908654533647,
      "loss": 0.2324,
      "num_input_tokens_seen": 12629024,
      "step": 59845
    },
    {
      "epoch": 6.584158415841584,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.025449875613172506,
      "loss": 0.2314,
      "num_input_tokens_seen": 12630016,
      "step": 59850
    },
    {
      "epoch": 6.584708470847085,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.025448842475499875,
      "loss": 0.2314,
      "num_input_tokens_seen": 12631104,
      "step": 59855
    },
    {
      "epoch": 6.585258525852585,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.025447809241525274,
      "loss": 0.2324,
      "num_input_tokens_seen": 12632224,
      "step": 59860
    },
    {
      "epoch": 6.585808580858086,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.02544677591125823,
      "loss": 0.2303,
      "num_input_tokens_seen": 12633280,
      "step": 59865
    },
    {
      "epoch": 6.586358635863586,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.025445742484708267,
      "loss": 0.2309,
      "num_input_tokens_seen": 12634336,
      "step": 59870
    },
    {
      "epoch": 6.586908690869087,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02544470896188491,
      "loss": 0.2324,
      "num_input_tokens_seen": 12635456,
      "step": 59875
    },
    {
      "epoch": 6.587458745874588,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.025443675342797678,
      "loss": 0.2319,
      "num_input_tokens_seen": 12636544,
      "step": 59880
    },
    {
      "epoch": 6.588008800880088,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.025442641627456103,
      "loss": 0.2309,
      "num_input_tokens_seen": 12637568,
      "step": 59885
    },
    {
      "epoch": 6.588558855885589,
      "grad_norm": 0.00060272216796875,
      "learning_rate": 0.02544160781586971,
      "loss": 0.2308,
      "num_input_tokens_seen": 12638688,
      "step": 59890
    },
    {
      "epoch": 6.589108910891089,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.025440573908048034,
      "loss": 0.2303,
      "num_input_tokens_seen": 12639712,
      "step": 59895
    },
    {
      "epoch": 6.589658965896589,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.025439539904000597,
      "loss": 0.2314,
      "num_input_tokens_seen": 12640736,
      "step": 59900
    },
    {
      "epoch": 6.5902090209020905,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.025438505803736924,
      "loss": 0.2303,
      "num_input_tokens_seen": 12641760,
      "step": 59905
    },
    {
      "epoch": 6.590759075907591,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.025437471607266553,
      "loss": 0.2298,
      "num_input_tokens_seen": 12642784,
      "step": 59910
    },
    {
      "epoch": 6.591309130913091,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.025436437314599016,
      "loss": 0.2313,
      "num_input_tokens_seen": 12643872,
      "step": 59915
    },
    {
      "epoch": 6.591859185918592,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.025435402925743845,
      "loss": 0.2308,
      "num_input_tokens_seen": 12644960,
      "step": 59920
    },
    {
      "epoch": 6.592409240924092,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.025434368440710575,
      "loss": 0.2313,
      "num_input_tokens_seen": 12646016,
      "step": 59925
    },
    {
      "epoch": 6.592959295929593,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.025433333859508734,
      "loss": 0.2303,
      "num_input_tokens_seen": 12647040,
      "step": 59930
    },
    {
      "epoch": 6.5935093509350935,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.025432299182147858,
      "loss": 0.2314,
      "num_input_tokens_seen": 12648096,
      "step": 59935
    },
    {
      "epoch": 6.594059405940594,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.025431264408637494,
      "loss": 0.2324,
      "num_input_tokens_seen": 12649152,
      "step": 59940
    },
    {
      "epoch": 6.594609460946095,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.025430229538987166,
      "loss": 0.2313,
      "num_input_tokens_seen": 12650176,
      "step": 59945
    },
    {
      "epoch": 6.595159515951595,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.025429194573206422,
      "loss": 0.2308,
      "num_input_tokens_seen": 12651264,
      "step": 59950
    },
    {
      "epoch": 6.595709570957096,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.025428159511304796,
      "loss": 0.2319,
      "num_input_tokens_seen": 12652320,
      "step": 59955
    },
    {
      "epoch": 6.596259625962596,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.025427124353291822,
      "loss": 0.2314,
      "num_input_tokens_seen": 12653376,
      "step": 59960
    },
    {
      "epoch": 6.5968096809680965,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.025426089099177047,
      "loss": 0.234,
      "num_input_tokens_seen": 12654432,
      "step": 59965
    },
    {
      "epoch": 6.597359735973598,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.025425053748970017,
      "loss": 0.2334,
      "num_input_tokens_seen": 12655520,
      "step": 59970
    },
    {
      "epoch": 6.597909790979098,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.02542401830268027,
      "loss": 0.2319,
      "num_input_tokens_seen": 12656608,
      "step": 59975
    },
    {
      "epoch": 6.598459845984598,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.025422982760317337,
      "loss": 0.2303,
      "num_input_tokens_seen": 12657600,
      "step": 59980
    },
    {
      "epoch": 6.599009900990099,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.025421947121890782,
      "loss": 0.2314,
      "num_input_tokens_seen": 12658720,
      "step": 59985
    },
    {
      "epoch": 6.599559955995599,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02542091138741014,
      "loss": 0.2308,
      "num_input_tokens_seen": 12659776,
      "step": 59990
    },
    {
      "epoch": 6.6001100110011,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.025419875556884956,
      "loss": 0.2319,
      "num_input_tokens_seen": 12660800,
      "step": 59995
    },
    {
      "epoch": 6.600660066006601,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.025418839630324784,
      "loss": 0.2298,
      "num_input_tokens_seen": 12661824,
      "step": 60000
    },
    {
      "epoch": 6.601210121012102,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.025417803607739164,
      "loss": 0.2329,
      "num_input_tokens_seen": 12662880,
      "step": 60005
    },
    {
      "epoch": 6.601760176017602,
      "grad_norm": 0.0006866455078125,
      "learning_rate": 0.025416767489137647,
      "loss": 0.2313,
      "num_input_tokens_seen": 12664000,
      "step": 60010
    },
    {
      "epoch": 6.602310231023102,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.02541573127452978,
      "loss": 0.2308,
      "num_input_tokens_seen": 12665056,
      "step": 60015
    },
    {
      "epoch": 6.602860286028603,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.02541469496392512,
      "loss": 0.2313,
      "num_input_tokens_seen": 12666112,
      "step": 60020
    },
    {
      "epoch": 6.603410341034103,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02541365855733321,
      "loss": 0.2314,
      "num_input_tokens_seen": 12667136,
      "step": 60025
    },
    {
      "epoch": 6.603960396039604,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02541262205476361,
      "loss": 0.2335,
      "num_input_tokens_seen": 12668224,
      "step": 60030
    },
    {
      "epoch": 6.604510451045105,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.025411585456225863,
      "loss": 0.2308,
      "num_input_tokens_seen": 12669312,
      "step": 60035
    },
    {
      "epoch": 6.605060506050605,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.025410548761729535,
      "loss": 0.2303,
      "num_input_tokens_seen": 12670336,
      "step": 60040
    },
    {
      "epoch": 6.605610561056105,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.025409511971284174,
      "loss": 0.2288,
      "num_input_tokens_seen": 12671328,
      "step": 60045
    },
    {
      "epoch": 6.606160616061606,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.025408475084899334,
      "loss": 0.2298,
      "num_input_tokens_seen": 12672384,
      "step": 60050
    },
    {
      "epoch": 6.606710671067106,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.025407438102584574,
      "loss": 0.2313,
      "num_input_tokens_seen": 12673472,
      "step": 60055
    },
    {
      "epoch": 6.6072607260726075,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.02540640102434945,
      "loss": 0.2313,
      "num_input_tokens_seen": 12674464,
      "step": 60060
    },
    {
      "epoch": 6.607810781078108,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.025405363850203526,
      "loss": 0.2319,
      "num_input_tokens_seen": 12675488,
      "step": 60065
    },
    {
      "epoch": 6.608360836083609,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.02540432658015635,
      "loss": 0.2303,
      "num_input_tokens_seen": 12676640,
      "step": 60070
    },
    {
      "epoch": 6.608910891089109,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02540328921421749,
      "loss": 0.2324,
      "num_input_tokens_seen": 12677696,
      "step": 60075
    },
    {
      "epoch": 6.609460946094609,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.02540225175239651,
      "loss": 0.2308,
      "num_input_tokens_seen": 12678720,
      "step": 60080
    },
    {
      "epoch": 6.61001100110011,
      "grad_norm": 0.0004825592041015625,
      "learning_rate": 0.02540121419470296,
      "loss": 0.2303,
      "num_input_tokens_seen": 12679744,
      "step": 60085
    },
    {
      "epoch": 6.6105610561056105,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.025400176541146416,
      "loss": 0.2303,
      "num_input_tokens_seen": 12680800,
      "step": 60090
    },
    {
      "epoch": 6.611111111111111,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.02539913879173643,
      "loss": 0.2304,
      "num_input_tokens_seen": 12681920,
      "step": 60095
    },
    {
      "epoch": 6.611661166116612,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.025398100946482573,
      "loss": 0.2298,
      "num_input_tokens_seen": 12683040,
      "step": 60100
    },
    {
      "epoch": 6.612211221122112,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02539706300539441,
      "loss": 0.2314,
      "num_input_tokens_seen": 12684128,
      "step": 60105
    },
    {
      "epoch": 6.612761276127613,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02539602496848151,
      "loss": 0.2298,
      "num_input_tokens_seen": 12685184,
      "step": 60110
    },
    {
      "epoch": 6.613311331133113,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.025394986835753433,
      "loss": 0.2325,
      "num_input_tokens_seen": 12686272,
      "step": 60115
    },
    {
      "epoch": 6.6138613861386135,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.02539394860721975,
      "loss": 0.2325,
      "num_input_tokens_seen": 12687360,
      "step": 60120
    },
    {
      "epoch": 6.614411441144115,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.02539291028289003,
      "loss": 0.2314,
      "num_input_tokens_seen": 12688384,
      "step": 60125
    },
    {
      "epoch": 6.614961496149615,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.025391871862773845,
      "loss": 0.2278,
      "num_input_tokens_seen": 12689376,
      "step": 60130
    },
    {
      "epoch": 6.615511551155116,
      "grad_norm": 0.00689697265625,
      "learning_rate": 0.02539083334688076,
      "loss": 0.2315,
      "num_input_tokens_seen": 12690336,
      "step": 60135
    },
    {
      "epoch": 6.616061606160616,
      "grad_norm": 0.0068359375,
      "learning_rate": 0.025389794735220354,
      "loss": 0.2304,
      "num_input_tokens_seen": 12691456,
      "step": 60140
    },
    {
      "epoch": 6.616611661166116,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.02538875602780219,
      "loss": 0.2335,
      "num_input_tokens_seen": 12692544,
      "step": 60145
    },
    {
      "epoch": 6.617161716171617,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.02538771722463585,
      "loss": 0.2278,
      "num_input_tokens_seen": 12693664,
      "step": 60150
    },
    {
      "epoch": 6.617711771177118,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.025386678325730907,
      "loss": 0.2346,
      "num_input_tokens_seen": 12694784,
      "step": 60155
    },
    {
      "epoch": 6.618261826182618,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.025385639331096933,
      "loss": 0.234,
      "num_input_tokens_seen": 12695808,
      "step": 60160
    },
    {
      "epoch": 6.618811881188119,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.025384600240743505,
      "loss": 0.2293,
      "num_input_tokens_seen": 12696800,
      "step": 60165
    },
    {
      "epoch": 6.619361936193619,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.025383561054680197,
      "loss": 0.2298,
      "num_input_tokens_seen": 12697792,
      "step": 60170
    },
    {
      "epoch": 6.61991199119912,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.025382521772916587,
      "loss": 0.2277,
      "num_input_tokens_seen": 12698816,
      "step": 60175
    },
    {
      "epoch": 6.62046204620462,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02538148239546226,
      "loss": 0.233,
      "num_input_tokens_seen": 12699872,
      "step": 60180
    },
    {
      "epoch": 6.621012101210121,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02538044292232679,
      "loss": 0.2293,
      "num_input_tokens_seen": 12700928,
      "step": 60185
    },
    {
      "epoch": 6.621562156215622,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02537940335351976,
      "loss": 0.233,
      "num_input_tokens_seen": 12701984,
      "step": 60190
    },
    {
      "epoch": 6.622112211221122,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.02537836368905075,
      "loss": 0.2288,
      "num_input_tokens_seen": 12703072,
      "step": 60195
    },
    {
      "epoch": 6.622662266226623,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.025377323928929343,
      "loss": 0.2242,
      "num_input_tokens_seen": 12704160,
      "step": 60200
    },
    {
      "epoch": 6.623212321232123,
      "grad_norm": 0.0030670166015625,
      "learning_rate": 0.025376284073165115,
      "loss": 0.2306,
      "num_input_tokens_seen": 12705248,
      "step": 60205
    },
    {
      "epoch": 6.623762376237623,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.025375244121767664,
      "loss": 0.2356,
      "num_input_tokens_seen": 12706368,
      "step": 60210
    },
    {
      "epoch": 6.6243124312431245,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.02537420407474656,
      "loss": 0.2295,
      "num_input_tokens_seen": 12707424,
      "step": 60215
    },
    {
      "epoch": 6.624862486248625,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.025373163932111396,
      "loss": 0.232,
      "num_input_tokens_seen": 12708480,
      "step": 60220
    },
    {
      "epoch": 6.625412541254125,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.02537212369387176,
      "loss": 0.2263,
      "num_input_tokens_seen": 12709472,
      "step": 60225
    },
    {
      "epoch": 6.625962596259626,
      "grad_norm": 0.01373291015625,
      "learning_rate": 0.025371083360037235,
      "loss": 0.2366,
      "num_input_tokens_seen": 12710528,
      "step": 60230
    },
    {
      "epoch": 6.626512651265126,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.025370042930617412,
      "loss": 0.2324,
      "num_input_tokens_seen": 12711584,
      "step": 60235
    },
    {
      "epoch": 6.627062706270627,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.025369002405621878,
      "loss": 0.2344,
      "num_input_tokens_seen": 12712608,
      "step": 60240
    },
    {
      "epoch": 6.6276127612761275,
      "grad_norm": 0.005859375,
      "learning_rate": 0.025367961785060227,
      "loss": 0.2328,
      "num_input_tokens_seen": 12713696,
      "step": 60245
    },
    {
      "epoch": 6.628162816281629,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.025366921068942047,
      "loss": 0.2306,
      "num_input_tokens_seen": 12714752,
      "step": 60250
    },
    {
      "epoch": 6.628712871287129,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.025365880257276926,
      "loss": 0.2379,
      "num_input_tokens_seen": 12715840,
      "step": 60255
    },
    {
      "epoch": 6.629262926292629,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.025364839350074463,
      "loss": 0.2383,
      "num_input_tokens_seen": 12716864,
      "step": 60260
    },
    {
      "epoch": 6.62981298129813,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.02536379834734425,
      "loss": 0.2304,
      "num_input_tokens_seen": 12717920,
      "step": 60265
    },
    {
      "epoch": 6.63036303630363,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02536275724909588,
      "loss": 0.2351,
      "num_input_tokens_seen": 12718912,
      "step": 60270
    },
    {
      "epoch": 6.6309130913091305,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02536171605533895,
      "loss": 0.2304,
      "num_input_tokens_seen": 12719936,
      "step": 60275
    },
    {
      "epoch": 6.631463146314632,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.025360674766083058,
      "loss": 0.2298,
      "num_input_tokens_seen": 12721024,
      "step": 60280
    },
    {
      "epoch": 6.632013201320132,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.025359633381337794,
      "loss": 0.2293,
      "num_input_tokens_seen": 12722112,
      "step": 60285
    },
    {
      "epoch": 6.632563256325633,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.02535859190111276,
      "loss": 0.2341,
      "num_input_tokens_seen": 12723104,
      "step": 60290
    },
    {
      "epoch": 6.633113311331133,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.025357550325417554,
      "loss": 0.233,
      "num_input_tokens_seen": 12724128,
      "step": 60295
    },
    {
      "epoch": 6.633663366336633,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.025356508654261777,
      "loss": 0.2325,
      "num_input_tokens_seen": 12725216,
      "step": 60300
    },
    {
      "epoch": 6.634213421342134,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.025355466887655036,
      "loss": 0.2289,
      "num_input_tokens_seen": 12726240,
      "step": 60305
    },
    {
      "epoch": 6.634763476347635,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02535442502560692,
      "loss": 0.2341,
      "num_input_tokens_seen": 12727296,
      "step": 60310
    },
    {
      "epoch": 6.635313531353136,
      "grad_norm": 0.0006256103515625,
      "learning_rate": 0.025353383068127035,
      "loss": 0.2303,
      "num_input_tokens_seen": 12728320,
      "step": 60315
    },
    {
      "epoch": 6.635863586358636,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.02535234101522499,
      "loss": 0.2288,
      "num_input_tokens_seen": 12729312,
      "step": 60320
    },
    {
      "epoch": 6.636413641364136,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.025351298866910385,
      "loss": 0.2319,
      "num_input_tokens_seen": 12730432,
      "step": 60325
    },
    {
      "epoch": 6.636963696369637,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.025350256623192825,
      "loss": 0.2319,
      "num_input_tokens_seen": 12731456,
      "step": 60330
    },
    {
      "epoch": 6.637513751375137,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.025349214284081916,
      "loss": 0.2298,
      "num_input_tokens_seen": 12732544,
      "step": 60335
    },
    {
      "epoch": 6.638063806380638,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.025348171849587264,
      "loss": 0.2304,
      "num_input_tokens_seen": 12733664,
      "step": 60340
    },
    {
      "epoch": 6.638613861386139,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.02534712931971848,
      "loss": 0.2314,
      "num_input_tokens_seen": 12734784,
      "step": 60345
    },
    {
      "epoch": 6.639163916391639,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.02534608669448517,
      "loss": 0.2314,
      "num_input_tokens_seen": 12735808,
      "step": 60350
    },
    {
      "epoch": 6.63971397139714,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02534504397389694,
      "loss": 0.2314,
      "num_input_tokens_seen": 12736896,
      "step": 60355
    },
    {
      "epoch": 6.64026402640264,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.025344001157963404,
      "loss": 0.2319,
      "num_input_tokens_seen": 12737888,
      "step": 60360
    },
    {
      "epoch": 6.6408140814081404,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.025342958246694174,
      "loss": 0.2319,
      "num_input_tokens_seen": 12738944,
      "step": 60365
    },
    {
      "epoch": 6.6413641364136415,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.02534191524009886,
      "loss": 0.2329,
      "num_input_tokens_seen": 12740000,
      "step": 60370
    },
    {
      "epoch": 6.641914191419142,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.02534087213818708,
      "loss": 0.2308,
      "num_input_tokens_seen": 12741120,
      "step": 60375
    },
    {
      "epoch": 6.642464246424643,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.02533982894096844,
      "loss": 0.2318,
      "num_input_tokens_seen": 12742176,
      "step": 60380
    },
    {
      "epoch": 6.643014301430143,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.02533878564845256,
      "loss": 0.2324,
      "num_input_tokens_seen": 12743200,
      "step": 60385
    },
    {
      "epoch": 6.643564356435643,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02533774226064905,
      "loss": 0.2319,
      "num_input_tokens_seen": 12744352,
      "step": 60390
    },
    {
      "epoch": 6.644114411441144,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.025336698777567536,
      "loss": 0.2319,
      "num_input_tokens_seen": 12745408,
      "step": 60395
    },
    {
      "epoch": 6.6446644664466445,
      "grad_norm": 0.00064849853515625,
      "learning_rate": 0.025335655199217627,
      "loss": 0.2324,
      "num_input_tokens_seen": 12746400,
      "step": 60400
    },
    {
      "epoch": 6.645214521452145,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.025334611525608945,
      "loss": 0.2308,
      "num_input_tokens_seen": 12747392,
      "step": 60405
    },
    {
      "epoch": 6.645764576457646,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.025333567756751104,
      "loss": 0.2334,
      "num_input_tokens_seen": 12748448,
      "step": 60410
    },
    {
      "epoch": 6.646314631463146,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.025332523892653724,
      "loss": 0.2314,
      "num_input_tokens_seen": 12749536,
      "step": 60415
    },
    {
      "epoch": 6.646864686468647,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02533147993332644,
      "loss": 0.2313,
      "num_input_tokens_seen": 12750592,
      "step": 60420
    },
    {
      "epoch": 6.647414741474147,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.025330435878778857,
      "loss": 0.2335,
      "num_input_tokens_seen": 12751648,
      "step": 60425
    },
    {
      "epoch": 6.647964796479648,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.0253293917290206,
      "loss": 0.2298,
      "num_input_tokens_seen": 12752736,
      "step": 60430
    },
    {
      "epoch": 6.648514851485149,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.0253283474840613,
      "loss": 0.2309,
      "num_input_tokens_seen": 12753760,
      "step": 60435
    },
    {
      "epoch": 6.649064906490649,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.025327303143910577,
      "loss": 0.2329,
      "num_input_tokens_seen": 12754848,
      "step": 60440
    },
    {
      "epoch": 6.64961496149615,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02532625870857806,
      "loss": 0.2314,
      "num_input_tokens_seen": 12755904,
      "step": 60445
    },
    {
      "epoch": 6.65016501650165,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.025325214178073364,
      "loss": 0.2335,
      "num_input_tokens_seen": 12756928,
      "step": 60450
    },
    {
      "epoch": 6.65071507150715,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.02532416955240613,
      "loss": 0.2319,
      "num_input_tokens_seen": 12757984,
      "step": 60455
    },
    {
      "epoch": 6.6512651265126514,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.025323124831585976,
      "loss": 0.2324,
      "num_input_tokens_seen": 12759040,
      "step": 60460
    },
    {
      "epoch": 6.651815181518152,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.025322080015622535,
      "loss": 0.2324,
      "num_input_tokens_seen": 12760160,
      "step": 60465
    },
    {
      "epoch": 6.652365236523653,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.025321035104525436,
      "loss": 0.2324,
      "num_input_tokens_seen": 12761248,
      "step": 60470
    },
    {
      "epoch": 6.652915291529153,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.02531999009830431,
      "loss": 0.2319,
      "num_input_tokens_seen": 12762400,
      "step": 60475
    },
    {
      "epoch": 6.653465346534653,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.02531894499696878,
      "loss": 0.2313,
      "num_input_tokens_seen": 12763456,
      "step": 60480
    },
    {
      "epoch": 6.654015401540154,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.025317899800528496,
      "loss": 0.2298,
      "num_input_tokens_seen": 12764512,
      "step": 60485
    },
    {
      "epoch": 6.6545654565456545,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.025316854508993073,
      "loss": 0.2314,
      "num_input_tokens_seen": 12765600,
      "step": 60490
    },
    {
      "epoch": 6.6551155115511555,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.025315809122372156,
      "loss": 0.2314,
      "num_input_tokens_seen": 12766688,
      "step": 60495
    },
    {
      "epoch": 6.655665566556656,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.025314763640675374,
      "loss": 0.2319,
      "num_input_tokens_seen": 12767744,
      "step": 60500
    },
    {
      "epoch": 6.656215621562156,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.025313718063912367,
      "loss": 0.2334,
      "num_input_tokens_seen": 12768832,
      "step": 60505
    },
    {
      "epoch": 6.656765676567657,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.025312672392092766,
      "loss": 0.2319,
      "num_input_tokens_seen": 12769888,
      "step": 60510
    },
    {
      "epoch": 6.657315731573157,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02531162662522622,
      "loss": 0.2319,
      "num_input_tokens_seen": 12770912,
      "step": 60515
    },
    {
      "epoch": 6.6578657865786575,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.025310580763322352,
      "loss": 0.2324,
      "num_input_tokens_seen": 12772032,
      "step": 60520
    },
    {
      "epoch": 6.658415841584159,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.02530953480639081,
      "loss": 0.2314,
      "num_input_tokens_seen": 12773088,
      "step": 60525
    },
    {
      "epoch": 6.658965896589659,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.025308488754441236,
      "loss": 0.2319,
      "num_input_tokens_seen": 12774112,
      "step": 60530
    },
    {
      "epoch": 6.65951595159516,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02530744260748326,
      "loss": 0.2298,
      "num_input_tokens_seen": 12775200,
      "step": 60535
    },
    {
      "epoch": 6.66006600660066,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.02530639636552654,
      "loss": 0.2329,
      "num_input_tokens_seen": 12776192,
      "step": 60540
    },
    {
      "epoch": 6.66061606160616,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.025305350028580708,
      "loss": 0.2314,
      "num_input_tokens_seen": 12777184,
      "step": 60545
    },
    {
      "epoch": 6.661166116611661,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.025304303596655406,
      "loss": 0.2319,
      "num_input_tokens_seen": 12778272,
      "step": 60550
    },
    {
      "epoch": 6.661716171617162,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.025303257069760287,
      "loss": 0.2324,
      "num_input_tokens_seen": 12779360,
      "step": 60555
    },
    {
      "epoch": 6.662266226622663,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.025302210447904986,
      "loss": 0.2308,
      "num_input_tokens_seen": 12780416,
      "step": 60560
    },
    {
      "epoch": 6.662816281628163,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.025301163731099162,
      "loss": 0.2303,
      "num_input_tokens_seen": 12781504,
      "step": 60565
    },
    {
      "epoch": 6.663366336633663,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.02530011691935245,
      "loss": 0.2324,
      "num_input_tokens_seen": 12782528,
      "step": 60570
    },
    {
      "epoch": 6.663916391639164,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.0252990700126745,
      "loss": 0.2329,
      "num_input_tokens_seen": 12783584,
      "step": 60575
    },
    {
      "epoch": 6.664466446644664,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.02529802301107497,
      "loss": 0.2308,
      "num_input_tokens_seen": 12784704,
      "step": 60580
    },
    {
      "epoch": 6.665016501650165,
      "grad_norm": 0.000499725341796875,
      "learning_rate": 0.0252969759145635,
      "loss": 0.2319,
      "num_input_tokens_seen": 12785696,
      "step": 60585
    },
    {
      "epoch": 6.665566556655666,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.025295928723149747,
      "loss": 0.2314,
      "num_input_tokens_seen": 12786720,
      "step": 60590
    },
    {
      "epoch": 6.666116611661166,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.02529488143684336,
      "loss": 0.2319,
      "num_input_tokens_seen": 12787840,
      "step": 60595
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.025293834055653986,
      "loss": 0.2308,
      "num_input_tokens_seen": 12788864,
      "step": 60600
    },
    {
      "epoch": 6.667216721672167,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.025292786579591286,
      "loss": 0.2314,
      "num_input_tokens_seen": 12789984,
      "step": 60605
    },
    {
      "epoch": 6.667766776677668,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.025291739008664906,
      "loss": 0.2304,
      "num_input_tokens_seen": 12791008,
      "step": 60610
    },
    {
      "epoch": 6.6683168316831685,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.025290691342884516,
      "loss": 0.2293,
      "num_input_tokens_seen": 12792064,
      "step": 60615
    },
    {
      "epoch": 6.668866886688669,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.025289643582259755,
      "loss": 0.2314,
      "num_input_tokens_seen": 12793120,
      "step": 60620
    },
    {
      "epoch": 6.66941694169417,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.025288595726800284,
      "loss": 0.2329,
      "num_input_tokens_seen": 12794240,
      "step": 60625
    },
    {
      "epoch": 6.66996699669967,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02528754777651577,
      "loss": 0.2314,
      "num_input_tokens_seen": 12795264,
      "step": 60630
    },
    {
      "epoch": 6.67051705170517,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02528649973141586,
      "loss": 0.2329,
      "num_input_tokens_seen": 12796384,
      "step": 60635
    },
    {
      "epoch": 6.671067106710671,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.02528545159151022,
      "loss": 0.2324,
      "num_input_tokens_seen": 12797344,
      "step": 60640
    },
    {
      "epoch": 6.6716171617161715,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.025284403356808507,
      "loss": 0.2325,
      "num_input_tokens_seen": 12798368,
      "step": 60645
    },
    {
      "epoch": 6.672167216721672,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.025283355027320382,
      "loss": 0.2309,
      "num_input_tokens_seen": 12799392,
      "step": 60650
    },
    {
      "epoch": 6.672717271727173,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.02528230660305551,
      "loss": 0.2298,
      "num_input_tokens_seen": 12800448,
      "step": 60655
    },
    {
      "epoch": 6.673267326732673,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.025281258084023553,
      "loss": 0.2298,
      "num_input_tokens_seen": 12801472,
      "step": 60660
    },
    {
      "epoch": 6.673817381738174,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.025280209470234172,
      "loss": 0.2314,
      "num_input_tokens_seen": 12802528,
      "step": 60665
    },
    {
      "epoch": 6.674367436743674,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.025279160761697034,
      "loss": 0.2319,
      "num_input_tokens_seen": 12803584,
      "step": 60670
    },
    {
      "epoch": 6.674917491749175,
      "grad_norm": 0.002777099609375,
      "learning_rate": 0.0252781119584218,
      "loss": 0.2304,
      "num_input_tokens_seen": 12804672,
      "step": 60675
    },
    {
      "epoch": 6.675467546754676,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02527706306041815,
      "loss": 0.2304,
      "num_input_tokens_seen": 12805632,
      "step": 60680
    },
    {
      "epoch": 6.676017601760176,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.02527601406769573,
      "loss": 0.2325,
      "num_input_tokens_seen": 12806656,
      "step": 60685
    },
    {
      "epoch": 6.676567656765677,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02527496498026422,
      "loss": 0.233,
      "num_input_tokens_seen": 12807712,
      "step": 60690
    },
    {
      "epoch": 6.677117711771177,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.025273915798133293,
      "loss": 0.2303,
      "num_input_tokens_seen": 12808736,
      "step": 60695
    },
    {
      "epoch": 6.677667766776677,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.025272866521312616,
      "loss": 0.2319,
      "num_input_tokens_seen": 12809760,
      "step": 60700
    },
    {
      "epoch": 6.678217821782178,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.025271817149811856,
      "loss": 0.2304,
      "num_input_tokens_seen": 12810816,
      "step": 60705
    },
    {
      "epoch": 6.678767876787679,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.025270767683640685,
      "loss": 0.233,
      "num_input_tokens_seen": 12811904,
      "step": 60710
    },
    {
      "epoch": 6.67931793179318,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.025269718122808777,
      "loss": 0.2314,
      "num_input_tokens_seen": 12812864,
      "step": 60715
    },
    {
      "epoch": 6.67986798679868,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.0252686684673258,
      "loss": 0.2298,
      "num_input_tokens_seen": 12813856,
      "step": 60720
    },
    {
      "epoch": 6.68041804180418,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.02526761871720144,
      "loss": 0.2303,
      "num_input_tokens_seen": 12814880,
      "step": 60725
    },
    {
      "epoch": 6.680968096809681,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.025266568872445366,
      "loss": 0.2325,
      "num_input_tokens_seen": 12815872,
      "step": 60730
    },
    {
      "epoch": 6.681518151815181,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.025265518933067246,
      "loss": 0.2319,
      "num_input_tokens_seen": 12816928,
      "step": 60735
    },
    {
      "epoch": 6.6820682068206825,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.025264468899076766,
      "loss": 0.2325,
      "num_input_tokens_seen": 12817952,
      "step": 60740
    },
    {
      "epoch": 6.682618261826183,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.025263418770483603,
      "loss": 0.2298,
      "num_input_tokens_seen": 12819040,
      "step": 60745
    },
    {
      "epoch": 6.683168316831683,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.025262368547297433,
      "loss": 0.2299,
      "num_input_tokens_seen": 12820000,
      "step": 60750
    },
    {
      "epoch": 6.683718371837184,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.02526131822952794,
      "loss": 0.2309,
      "num_input_tokens_seen": 12821056,
      "step": 60755
    },
    {
      "epoch": 6.684268426842684,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.0252602678171848,
      "loss": 0.2314,
      "num_input_tokens_seen": 12822144,
      "step": 60760
    },
    {
      "epoch": 6.684818481848184,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.025259217310277693,
      "loss": 0.2293,
      "num_input_tokens_seen": 12823200,
      "step": 60765
    },
    {
      "epoch": 6.6853685368536855,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.0252581667088163,
      "loss": 0.2309,
      "num_input_tokens_seen": 12824288,
      "step": 60770
    },
    {
      "epoch": 6.685918591859186,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.025257116012810315,
      "loss": 0.2314,
      "num_input_tokens_seen": 12825344,
      "step": 60775
    },
    {
      "epoch": 6.686468646864687,
      "grad_norm": 0.0007476806640625,
      "learning_rate": 0.02525606522226941,
      "loss": 0.2314,
      "num_input_tokens_seen": 12826368,
      "step": 60780
    },
    {
      "epoch": 6.687018701870187,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.025255014337203268,
      "loss": 0.2314,
      "num_input_tokens_seen": 12827456,
      "step": 60785
    },
    {
      "epoch": 6.687568756875687,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.025253963357621582,
      "loss": 0.2309,
      "num_input_tokens_seen": 12828544,
      "step": 60790
    },
    {
      "epoch": 6.688118811881188,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.02525291228353404,
      "loss": 0.2329,
      "num_input_tokens_seen": 12829536,
      "step": 60795
    },
    {
      "epoch": 6.6886688668866885,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.02525186111495032,
      "loss": 0.2335,
      "num_input_tokens_seen": 12830688,
      "step": 60800
    },
    {
      "epoch": 6.68921892189219,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.02525080985188012,
      "loss": 0.2319,
      "num_input_tokens_seen": 12831712,
      "step": 60805
    },
    {
      "epoch": 6.68976897689769,
      "grad_norm": 0.0030364990234375,
      "learning_rate": 0.025249758494333124,
      "loss": 0.2313,
      "num_input_tokens_seen": 12832768,
      "step": 60810
    },
    {
      "epoch": 6.69031903190319,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02524870704231902,
      "loss": 0.2313,
      "num_input_tokens_seen": 12833888,
      "step": 60815
    },
    {
      "epoch": 6.690869086908691,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.0252476554958475,
      "loss": 0.2313,
      "num_input_tokens_seen": 12835008,
      "step": 60820
    },
    {
      "epoch": 6.691419141914191,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.02524660385492826,
      "loss": 0.2324,
      "num_input_tokens_seen": 12836064,
      "step": 60825
    },
    {
      "epoch": 6.6919691969196915,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.025245552119570986,
      "loss": 0.2334,
      "num_input_tokens_seen": 12837088,
      "step": 60830
    },
    {
      "epoch": 6.692519251925193,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.025244500289785376,
      "loss": 0.2334,
      "num_input_tokens_seen": 12838144,
      "step": 60835
    },
    {
      "epoch": 6.693069306930693,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.025243448365581123,
      "loss": 0.2324,
      "num_input_tokens_seen": 12839200,
      "step": 60840
    },
    {
      "epoch": 6.693619361936194,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.025242396346967923,
      "loss": 0.2324,
      "num_input_tokens_seen": 12840320,
      "step": 60845
    },
    {
      "epoch": 6.694169416941694,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.02524134423395547,
      "loss": 0.2324,
      "num_input_tokens_seen": 12841408,
      "step": 60850
    },
    {
      "epoch": 6.694719471947195,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02524029202655346,
      "loss": 0.2314,
      "num_input_tokens_seen": 12842496,
      "step": 60855
    },
    {
      "epoch": 6.695269526952695,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02523923972477159,
      "loss": 0.2288,
      "num_input_tokens_seen": 12843648,
      "step": 60860
    },
    {
      "epoch": 6.695819581958196,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02523818732861956,
      "loss": 0.2314,
      "num_input_tokens_seen": 12844704,
      "step": 60865
    },
    {
      "epoch": 6.696369636963697,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.025237134838107078,
      "loss": 0.2308,
      "num_input_tokens_seen": 12845792,
      "step": 60870
    },
    {
      "epoch": 6.696919691969197,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.025236082253243832,
      "loss": 0.2319,
      "num_input_tokens_seen": 12846880,
      "step": 60875
    },
    {
      "epoch": 6.697469746974697,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.025235029574039527,
      "loss": 0.2314,
      "num_input_tokens_seen": 12847904,
      "step": 60880
    },
    {
      "epoch": 6.698019801980198,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.025233976800503866,
      "loss": 0.2303,
      "num_input_tokens_seen": 12848960,
      "step": 60885
    },
    {
      "epoch": 6.698569856985698,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.025232923932646548,
      "loss": 0.2308,
      "num_input_tokens_seen": 12849984,
      "step": 60890
    },
    {
      "epoch": 6.6991199119911995,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02523187097047729,
      "loss": 0.2314,
      "num_input_tokens_seen": 12850976,
      "step": 60895
    },
    {
      "epoch": 6.6996699669967,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.02523081791400578,
      "loss": 0.2329,
      "num_input_tokens_seen": 12852032,
      "step": 60900
    },
    {
      "epoch": 6.7002200220022,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.025229764763241732,
      "loss": 0.2308,
      "num_input_tokens_seen": 12853056,
      "step": 60905
    },
    {
      "epoch": 6.700770077007701,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02522871151819485,
      "loss": 0.2319,
      "num_input_tokens_seen": 12854144,
      "step": 60910
    },
    {
      "epoch": 6.701320132013201,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02522765817887484,
      "loss": 0.2314,
      "num_input_tokens_seen": 12855136,
      "step": 60915
    },
    {
      "epoch": 6.701870187018702,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.025226604745291415,
      "loss": 0.2308,
      "num_input_tokens_seen": 12856160,
      "step": 60920
    },
    {
      "epoch": 6.7024202420242025,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.025225551217454283,
      "loss": 0.2319,
      "num_input_tokens_seen": 12857248,
      "step": 60925
    },
    {
      "epoch": 6.702970297029703,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.025224497595373152,
      "loss": 0.2308,
      "num_input_tokens_seen": 12858272,
      "step": 60930
    },
    {
      "epoch": 6.703520352035204,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.025223443879057728,
      "loss": 0.2319,
      "num_input_tokens_seen": 12859392,
      "step": 60935
    },
    {
      "epoch": 6.704070407040704,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02522239006851773,
      "loss": 0.2308,
      "num_input_tokens_seen": 12860448,
      "step": 60940
    },
    {
      "epoch": 6.704620462046204,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.025221336163762872,
      "loss": 0.2314,
      "num_input_tokens_seen": 12861568,
      "step": 60945
    },
    {
      "epoch": 6.705170517051705,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.025220282164802856,
      "loss": 0.2308,
      "num_input_tokens_seen": 12862560,
      "step": 60950
    },
    {
      "epoch": 6.7057205720572055,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.025219228071647407,
      "loss": 0.2324,
      "num_input_tokens_seen": 12863712,
      "step": 60955
    },
    {
      "epoch": 6.706270627062707,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.02521817388430624,
      "loss": 0.2314,
      "num_input_tokens_seen": 12864768,
      "step": 60960
    },
    {
      "epoch": 6.706820682068207,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.02521711960278906,
      "loss": 0.2314,
      "num_input_tokens_seen": 12865792,
      "step": 60965
    },
    {
      "epoch": 6.707370737073707,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.0252160652271056,
      "loss": 0.2334,
      "num_input_tokens_seen": 12866784,
      "step": 60970
    },
    {
      "epoch": 6.707920792079208,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.025215010757265566,
      "loss": 0.2319,
      "num_input_tokens_seen": 12867840,
      "step": 60975
    },
    {
      "epoch": 6.708470847084708,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.025213956193278677,
      "loss": 0.2313,
      "num_input_tokens_seen": 12868864,
      "step": 60980
    },
    {
      "epoch": 6.709020902090209,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.025212901535154653,
      "loss": 0.2313,
      "num_input_tokens_seen": 12869888,
      "step": 60985
    },
    {
      "epoch": 6.70957095709571,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02521184678290322,
      "loss": 0.2324,
      "num_input_tokens_seen": 12870976,
      "step": 60990
    },
    {
      "epoch": 6.71012101210121,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.025210791936534093,
      "loss": 0.2319,
      "num_input_tokens_seen": 12872032,
      "step": 60995
    },
    {
      "epoch": 6.710671067106711,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.025209736996057,
      "loss": 0.2308,
      "num_input_tokens_seen": 12873088,
      "step": 61000
    },
    {
      "epoch": 6.711221122112211,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.025208681961481655,
      "loss": 0.2308,
      "num_input_tokens_seen": 12874144,
      "step": 61005
    },
    {
      "epoch": 6.711771177117711,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.025207626832817787,
      "loss": 0.2303,
      "num_input_tokens_seen": 12875200,
      "step": 61010
    },
    {
      "epoch": 6.712321232123212,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.025206571610075126,
      "loss": 0.2319,
      "num_input_tokens_seen": 12876192,
      "step": 61015
    },
    {
      "epoch": 6.712871287128713,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.02520551629326339,
      "loss": 0.2319,
      "num_input_tokens_seen": 12877216,
      "step": 61020
    },
    {
      "epoch": 6.713421342134214,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.025204460882392307,
      "loss": 0.2303,
      "num_input_tokens_seen": 12878272,
      "step": 61025
    },
    {
      "epoch": 6.713971397139714,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.025203405377471604,
      "loss": 0.2303,
      "num_input_tokens_seen": 12879328,
      "step": 61030
    },
    {
      "epoch": 6.714521452145215,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.025202349778511008,
      "loss": 0.2308,
      "num_input_tokens_seen": 12880416,
      "step": 61035
    },
    {
      "epoch": 6.715071507150715,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.025201294085520257,
      "loss": 0.2324,
      "num_input_tokens_seen": 12881504,
      "step": 61040
    },
    {
      "epoch": 6.715621562156215,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.025200238298509068,
      "loss": 0.2308,
      "num_input_tokens_seen": 12882560,
      "step": 61045
    },
    {
      "epoch": 6.7161716171617165,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.025199182417487177,
      "loss": 0.2308,
      "num_input_tokens_seen": 12883648,
      "step": 61050
    },
    {
      "epoch": 6.716721672167217,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.025198126442464313,
      "loss": 0.2329,
      "num_input_tokens_seen": 12884736,
      "step": 61055
    },
    {
      "epoch": 6.717271727172717,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.025197070373450217,
      "loss": 0.2308,
      "num_input_tokens_seen": 12885824,
      "step": 61060
    },
    {
      "epoch": 6.717821782178218,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.025196014210454615,
      "loss": 0.2308,
      "num_input_tokens_seen": 12886944,
      "step": 61065
    },
    {
      "epoch": 6.718371837183718,
      "grad_norm": 0.00061798095703125,
      "learning_rate": 0.02519495795348724,
      "loss": 0.2308,
      "num_input_tokens_seen": 12888000,
      "step": 61070
    },
    {
      "epoch": 6.718921892189218,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.025193901602557828,
      "loss": 0.2308,
      "num_input_tokens_seen": 12889088,
      "step": 61075
    },
    {
      "epoch": 6.7194719471947195,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.02519284515767612,
      "loss": 0.2313,
      "num_input_tokens_seen": 12890144,
      "step": 61080
    },
    {
      "epoch": 6.72002200220022,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.02519178861885185,
      "loss": 0.2303,
      "num_input_tokens_seen": 12891264,
      "step": 61085
    },
    {
      "epoch": 6.720572057205721,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.025190731986094752,
      "loss": 0.2309,
      "num_input_tokens_seen": 12892320,
      "step": 61090
    },
    {
      "epoch": 6.721122112211221,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02518967525941457,
      "loss": 0.2308,
      "num_input_tokens_seen": 12893440,
      "step": 61095
    },
    {
      "epoch": 6.721672167216722,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.025188618438821038,
      "loss": 0.2318,
      "num_input_tokens_seen": 12894560,
      "step": 61100
    },
    {
      "epoch": 6.722222222222222,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0251875615243239,
      "loss": 0.2313,
      "num_input_tokens_seen": 12895616,
      "step": 61105
    },
    {
      "epoch": 6.7227722772277225,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.025186504515932896,
      "loss": 0.2319,
      "num_input_tokens_seen": 12896736,
      "step": 61110
    },
    {
      "epoch": 6.723322332233224,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.025185447413657767,
      "loss": 0.2308,
      "num_input_tokens_seen": 12897728,
      "step": 61115
    },
    {
      "epoch": 6.723872387238724,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.02518439021750826,
      "loss": 0.2293,
      "num_input_tokens_seen": 12898816,
      "step": 61120
    },
    {
      "epoch": 6.724422442244224,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02518333292749411,
      "loss": 0.2319,
      "num_input_tokens_seen": 12899904,
      "step": 61125
    },
    {
      "epoch": 6.724972497249725,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02518227554362507,
      "loss": 0.2308,
      "num_input_tokens_seen": 12900960,
      "step": 61130
    },
    {
      "epoch": 6.725522552255225,
      "grad_norm": 0.005859375,
      "learning_rate": 0.025181218065910886,
      "loss": 0.2298,
      "num_input_tokens_seen": 12902016,
      "step": 61135
    },
    {
      "epoch": 6.726072607260726,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.02518016049436129,
      "loss": 0.2309,
      "num_input_tokens_seen": 12903104,
      "step": 61140
    },
    {
      "epoch": 6.726622662266227,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.02517910282898605,
      "loss": 0.2314,
      "num_input_tokens_seen": 12904192,
      "step": 61145
    },
    {
      "epoch": 6.727172717271727,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.025178045069794895,
      "loss": 0.2314,
      "num_input_tokens_seen": 12905216,
      "step": 61150
    },
    {
      "epoch": 6.727722772277228,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.025176987216797588,
      "loss": 0.2282,
      "num_input_tokens_seen": 12906336,
      "step": 61155
    },
    {
      "epoch": 6.728272827282728,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.025175929270003873,
      "loss": 0.2283,
      "num_input_tokens_seen": 12907392,
      "step": 61160
    },
    {
      "epoch": 6.728822882288229,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.025174871229423502,
      "loss": 0.2298,
      "num_input_tokens_seen": 12908544,
      "step": 61165
    },
    {
      "epoch": 6.729372937293729,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.025173813095066222,
      "loss": 0.2324,
      "num_input_tokens_seen": 12909600,
      "step": 61170
    },
    {
      "epoch": 6.72992299229923,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02517275486694179,
      "loss": 0.2319,
      "num_input_tokens_seen": 12910688,
      "step": 61175
    },
    {
      "epoch": 6.730473047304731,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02517169654505996,
      "loss": 0.2298,
      "num_input_tokens_seen": 12911680,
      "step": 61180
    },
    {
      "epoch": 6.731023102310231,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.025170638129430482,
      "loss": 0.2309,
      "num_input_tokens_seen": 12912704,
      "step": 61185
    },
    {
      "epoch": 6.731573157315731,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.02516957962006311,
      "loss": 0.2324,
      "num_input_tokens_seen": 12913792,
      "step": 61190
    },
    {
      "epoch": 6.732123212321232,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.025168521016967604,
      "loss": 0.232,
      "num_input_tokens_seen": 12914848,
      "step": 61195
    },
    {
      "epoch": 6.732673267326732,
      "grad_norm": 0.005859375,
      "learning_rate": 0.025167462320153718,
      "loss": 0.2367,
      "num_input_tokens_seen": 12915936,
      "step": 61200
    },
    {
      "epoch": 6.7332233223322335,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.025166403529631214,
      "loss": 0.2314,
      "num_input_tokens_seen": 12916928,
      "step": 61205
    },
    {
      "epoch": 6.733773377337734,
      "grad_norm": 0.00347900390625,
      "learning_rate": 0.025165344645409846,
      "loss": 0.2278,
      "num_input_tokens_seen": 12917920,
      "step": 61210
    },
    {
      "epoch": 6.734323432343234,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.025164285667499373,
      "loss": 0.2335,
      "num_input_tokens_seen": 12918944,
      "step": 61215
    },
    {
      "epoch": 6.734873487348735,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.025163226595909555,
      "loss": 0.2303,
      "num_input_tokens_seen": 12920000,
      "step": 61220
    },
    {
      "epoch": 6.735423542354235,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.025162167430650157,
      "loss": 0.2298,
      "num_input_tokens_seen": 12921088,
      "step": 61225
    },
    {
      "epoch": 6.735973597359736,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02516110817173094,
      "loss": 0.2335,
      "num_input_tokens_seen": 12922144,
      "step": 61230
    },
    {
      "epoch": 6.7365236523652365,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02516004881916166,
      "loss": 0.2293,
      "num_input_tokens_seen": 12923232,
      "step": 61235
    },
    {
      "epoch": 6.737073707370737,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.02515898937295209,
      "loss": 0.2314,
      "num_input_tokens_seen": 12924320,
      "step": 61240
    },
    {
      "epoch": 6.737623762376238,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02515792983311198,
      "loss": 0.2303,
      "num_input_tokens_seen": 12925344,
      "step": 61245
    },
    {
      "epoch": 6.738173817381738,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.025156870199651113,
      "loss": 0.2325,
      "num_input_tokens_seen": 12926368,
      "step": 61250
    },
    {
      "epoch": 6.738723872387238,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.025155810472579247,
      "loss": 0.2324,
      "num_input_tokens_seen": 12927488,
      "step": 61255
    },
    {
      "epoch": 6.739273927392739,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.02515475065190615,
      "loss": 0.2293,
      "num_input_tokens_seen": 12928544,
      "step": 61260
    },
    {
      "epoch": 6.7398239823982395,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.025153690737641583,
      "loss": 0.2324,
      "num_input_tokens_seen": 12929664,
      "step": 61265
    },
    {
      "epoch": 6.740374037403741,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02515263072979533,
      "loss": 0.2335,
      "num_input_tokens_seen": 12930720,
      "step": 61270
    },
    {
      "epoch": 6.740924092409241,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.025151570628377144,
      "loss": 0.2314,
      "num_input_tokens_seen": 12931744,
      "step": 61275
    },
    {
      "epoch": 6.741474147414742,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.025150510433396805,
      "loss": 0.2329,
      "num_input_tokens_seen": 12932800,
      "step": 61280
    },
    {
      "epoch": 6.742024202420242,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.025149450144864083,
      "loss": 0.2314,
      "num_input_tokens_seen": 12933856,
      "step": 61285
    },
    {
      "epoch": 6.742574257425742,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02514838976278875,
      "loss": 0.2308,
      "num_input_tokens_seen": 12934912,
      "step": 61290
    },
    {
      "epoch": 6.743124312431243,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.02514732928718057,
      "loss": 0.2308,
      "num_input_tokens_seen": 12936000,
      "step": 61295
    },
    {
      "epoch": 6.743674367436744,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.02514626871804933,
      "loss": 0.2314,
      "num_input_tokens_seen": 12937120,
      "step": 61300
    },
    {
      "epoch": 6.744224422442244,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.025145208055404804,
      "loss": 0.2324,
      "num_input_tokens_seen": 12938208,
      "step": 61305
    },
    {
      "epoch": 6.744774477447745,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.025144147299256764,
      "loss": 0.2309,
      "num_input_tokens_seen": 12939296,
      "step": 61310
    },
    {
      "epoch": 6.745324532453245,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.025143086449614983,
      "loss": 0.2303,
      "num_input_tokens_seen": 12940352,
      "step": 61315
    },
    {
      "epoch": 6.745874587458746,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.025142025506489238,
      "loss": 0.2309,
      "num_input_tokens_seen": 12941344,
      "step": 61320
    },
    {
      "epoch": 6.7464246424642464,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.025140964469889313,
      "loss": 0.2319,
      "num_input_tokens_seen": 12942368,
      "step": 61325
    },
    {
      "epoch": 6.746974697469747,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.025139903339824986,
      "loss": 0.2309,
      "num_input_tokens_seen": 12943456,
      "step": 61330
    },
    {
      "epoch": 6.747524752475248,
      "grad_norm": 0.000598907470703125,
      "learning_rate": 0.025138842116306036,
      "loss": 0.2308,
      "num_input_tokens_seen": 12944512,
      "step": 61335
    },
    {
      "epoch": 6.748074807480748,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02513778079934224,
      "loss": 0.2314,
      "num_input_tokens_seen": 12945568,
      "step": 61340
    },
    {
      "epoch": 6.748624862486249,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.025136719388943383,
      "loss": 0.2329,
      "num_input_tokens_seen": 12946592,
      "step": 61345
    },
    {
      "epoch": 6.749174917491749,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02513565788511925,
      "loss": 0.2324,
      "num_input_tokens_seen": 12947712,
      "step": 61350
    },
    {
      "epoch": 6.7497249724972495,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.02513459628787962,
      "loss": 0.2319,
      "num_input_tokens_seen": 12948736,
      "step": 61355
    },
    {
      "epoch": 6.7502750275027505,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.025133534597234282,
      "loss": 0.2314,
      "num_input_tokens_seen": 12949760,
      "step": 61360
    },
    {
      "epoch": 6.750825082508251,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02513247281319302,
      "loss": 0.2314,
      "num_input_tokens_seen": 12950880,
      "step": 61365
    },
    {
      "epoch": 6.751375137513751,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02513141093576561,
      "loss": 0.2313,
      "num_input_tokens_seen": 12951904,
      "step": 61370
    },
    {
      "epoch": 6.751925192519252,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.025130348964961857,
      "loss": 0.2303,
      "num_input_tokens_seen": 12952992,
      "step": 61375
    },
    {
      "epoch": 6.752475247524752,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02512928690079153,
      "loss": 0.2319,
      "num_input_tokens_seen": 12954016,
      "step": 61380
    },
    {
      "epoch": 6.753025302530253,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.025128224743264427,
      "loss": 0.2319,
      "num_input_tokens_seen": 12955136,
      "step": 61385
    },
    {
      "epoch": 6.7535753575357536,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.02512716249239034,
      "loss": 0.2303,
      "num_input_tokens_seen": 12956128,
      "step": 61390
    },
    {
      "epoch": 6.754125412541254,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.025126100148179054,
      "loss": 0.2324,
      "num_input_tokens_seen": 12957184,
      "step": 61395
    },
    {
      "epoch": 6.754675467546755,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.025125037710640365,
      "loss": 0.2308,
      "num_input_tokens_seen": 12958240,
      "step": 61400
    },
    {
      "epoch": 6.755225522552255,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.02512397517978406,
      "loss": 0.2324,
      "num_input_tokens_seen": 12959328,
      "step": 61405
    },
    {
      "epoch": 6.755775577557756,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.02512291255561993,
      "loss": 0.2308,
      "num_input_tokens_seen": 12960352,
      "step": 61410
    },
    {
      "epoch": 6.756325632563256,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.025121849838157777,
      "loss": 0.2314,
      "num_input_tokens_seen": 12961376,
      "step": 61415
    },
    {
      "epoch": 6.756875687568757,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.025120787027407387,
      "loss": 0.2308,
      "num_input_tokens_seen": 12962400,
      "step": 61420
    },
    {
      "epoch": 6.757425742574258,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.025119724123378564,
      "loss": 0.2308,
      "num_input_tokens_seen": 12963520,
      "step": 61425
    },
    {
      "epoch": 6.757975797579758,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.025118661126081097,
      "loss": 0.2314,
      "num_input_tokens_seen": 12964576,
      "step": 61430
    },
    {
      "epoch": 6.758525852585258,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.025117598035524786,
      "loss": 0.2313,
      "num_input_tokens_seen": 12965536,
      "step": 61435
    },
    {
      "epoch": 6.759075907590759,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.025116534851719427,
      "loss": 0.2298,
      "num_input_tokens_seen": 12966624,
      "step": 61440
    },
    {
      "epoch": 6.759625962596259,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.025115471574674825,
      "loss": 0.2319,
      "num_input_tokens_seen": 12967680,
      "step": 61445
    },
    {
      "epoch": 6.7601760176017605,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.025114408204400772,
      "loss": 0.2319,
      "num_input_tokens_seen": 12968736,
      "step": 61450
    },
    {
      "epoch": 6.760726072607261,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02511334474090707,
      "loss": 0.2329,
      "num_input_tokens_seen": 12969760,
      "step": 61455
    },
    {
      "epoch": 6.761276127612762,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.025112281184203528,
      "loss": 0.2319,
      "num_input_tokens_seen": 12970784,
      "step": 61460
    },
    {
      "epoch": 6.761826182618262,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.025111217534299943,
      "loss": 0.2319,
      "num_input_tokens_seen": 12971808,
      "step": 61465
    },
    {
      "epoch": 6.762376237623762,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.025110153791206114,
      "loss": 0.2288,
      "num_input_tokens_seen": 12972832,
      "step": 61470
    },
    {
      "epoch": 6.762926292629263,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.025109089954931847,
      "loss": 0.2303,
      "num_input_tokens_seen": 12973952,
      "step": 61475
    },
    {
      "epoch": 6.7634763476347635,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.025108026025486955,
      "loss": 0.2319,
      "num_input_tokens_seen": 12975008,
      "step": 61480
    },
    {
      "epoch": 6.764026402640264,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.025106962002881233,
      "loss": 0.2324,
      "num_input_tokens_seen": 12976032,
      "step": 61485
    },
    {
      "epoch": 6.764576457645765,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.025105897887124494,
      "loss": 0.2303,
      "num_input_tokens_seen": 12977088,
      "step": 61490
    },
    {
      "epoch": 6.765126512651265,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02510483367822654,
      "loss": 0.2313,
      "num_input_tokens_seen": 12978176,
      "step": 61495
    },
    {
      "epoch": 6.765676567656766,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.025103769376197186,
      "loss": 0.2303,
      "num_input_tokens_seen": 12979264,
      "step": 61500
    },
    {
      "epoch": 6.766226622662266,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.025102704981046237,
      "loss": 0.2299,
      "num_input_tokens_seen": 12980384,
      "step": 61505
    },
    {
      "epoch": 6.7667766776677665,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.025101640492783503,
      "loss": 0.2319,
      "num_input_tokens_seen": 12981376,
      "step": 61510
    },
    {
      "epoch": 6.767326732673268,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.0251005759114188,
      "loss": 0.2299,
      "num_input_tokens_seen": 12982464,
      "step": 61515
    },
    {
      "epoch": 6.767876787678768,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02509951123696193,
      "loss": 0.2335,
      "num_input_tokens_seen": 12983520,
      "step": 61520
    },
    {
      "epoch": 6.768426842684269,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.025098446469422712,
      "loss": 0.2293,
      "num_input_tokens_seen": 12984544,
      "step": 61525
    },
    {
      "epoch": 6.768976897689769,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.025097381608810965,
      "loss": 0.2304,
      "num_input_tokens_seen": 12985600,
      "step": 61530
    },
    {
      "epoch": 6.769526952695269,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.025096316655136487,
      "loss": 0.2288,
      "num_input_tokens_seen": 12986656,
      "step": 61535
    },
    {
      "epoch": 6.77007700770077,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.02509525160840911,
      "loss": 0.2345,
      "num_input_tokens_seen": 12987680,
      "step": 61540
    },
    {
      "epoch": 6.770627062706271,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.025094186468638643,
      "loss": 0.2319,
      "num_input_tokens_seen": 12988768,
      "step": 61545
    },
    {
      "epoch": 6.771177117711771,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.025093121235834896,
      "loss": 0.233,
      "num_input_tokens_seen": 12989824,
      "step": 61550
    },
    {
      "epoch": 6.771727172717272,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.025092055910007703,
      "loss": 0.2298,
      "num_input_tokens_seen": 12990912,
      "step": 61555
    },
    {
      "epoch": 6.772277227722772,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.025090990491166865,
      "loss": 0.2298,
      "num_input_tokens_seen": 12991968,
      "step": 61560
    },
    {
      "epoch": 6.772827282728273,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.025089924979322217,
      "loss": 0.2299,
      "num_input_tokens_seen": 12992992,
      "step": 61565
    },
    {
      "epoch": 6.773377337733773,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.025088859374483567,
      "loss": 0.2299,
      "num_input_tokens_seen": 12994048,
      "step": 61570
    },
    {
      "epoch": 6.773927392739274,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.025087793676660745,
      "loss": 0.2309,
      "num_input_tokens_seen": 12995168,
      "step": 61575
    },
    {
      "epoch": 6.774477447744775,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.02508672788586357,
      "loss": 0.231,
      "num_input_tokens_seen": 12996128,
      "step": 61580
    },
    {
      "epoch": 6.775027502750275,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.025085662002101862,
      "loss": 0.2284,
      "num_input_tokens_seen": 12997152,
      "step": 61585
    },
    {
      "epoch": 6.775577557755776,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.025084596025385444,
      "loss": 0.2269,
      "num_input_tokens_seen": 12998208,
      "step": 61590
    },
    {
      "epoch": 6.776127612761276,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.025083529955724146,
      "loss": 0.2321,
      "num_input_tokens_seen": 12999232,
      "step": 61595
    },
    {
      "epoch": 6.776677667766776,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.025082463793127792,
      "loss": 0.2312,
      "num_input_tokens_seen": 13000352,
      "step": 61600
    },
    {
      "epoch": 6.7772277227722775,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.025081397537606208,
      "loss": 0.225,
      "num_input_tokens_seen": 13001376,
      "step": 61605
    },
    {
      "epoch": 6.777777777777778,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.025080331189169216,
      "loss": 0.2311,
      "num_input_tokens_seen": 13002368,
      "step": 61610
    },
    {
      "epoch": 6.778327832783278,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02507926474782665,
      "loss": 0.2308,
      "num_input_tokens_seen": 13003392,
      "step": 61615
    },
    {
      "epoch": 6.778877887788779,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02507819821358834,
      "loss": 0.2266,
      "num_input_tokens_seen": 13004480,
      "step": 61620
    },
    {
      "epoch": 6.779427942794279,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.025077131586464112,
      "loss": 0.2395,
      "num_input_tokens_seen": 13005504,
      "step": 61625
    },
    {
      "epoch": 6.77997799779978,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.025076064866463795,
      "loss": 0.237,
      "num_input_tokens_seen": 13006624,
      "step": 61630
    },
    {
      "epoch": 6.7805280528052805,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02507499805359723,
      "loss": 0.2331,
      "num_input_tokens_seen": 13007616,
      "step": 61635
    },
    {
      "epoch": 6.781078107810782,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.025073931147874234,
      "loss": 0.2336,
      "num_input_tokens_seen": 13008672,
      "step": 61640
    },
    {
      "epoch": 6.781628162816282,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.025072864149304656,
      "loss": 0.232,
      "num_input_tokens_seen": 13009728,
      "step": 61645
    },
    {
      "epoch": 6.782178217821782,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.025071797057898314,
      "loss": 0.2273,
      "num_input_tokens_seen": 13010816,
      "step": 61650
    },
    {
      "epoch": 6.782728272827283,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02507072987366506,
      "loss": 0.2283,
      "num_input_tokens_seen": 13011872,
      "step": 61655
    },
    {
      "epoch": 6.783278327832783,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.02506966259661472,
      "loss": 0.2304,
      "num_input_tokens_seen": 13012992,
      "step": 61660
    },
    {
      "epoch": 6.7838283828382835,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.025068595226757126,
      "loss": 0.233,
      "num_input_tokens_seen": 13014016,
      "step": 61665
    },
    {
      "epoch": 6.784378437843785,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.025067527764102124,
      "loss": 0.2357,
      "num_input_tokens_seen": 13015040,
      "step": 61670
    },
    {
      "epoch": 6.784928492849285,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.02506646020865955,
      "loss": 0.2304,
      "num_input_tokens_seen": 13016128,
      "step": 61675
    },
    {
      "epoch": 6.785478547854785,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.025065392560439246,
      "loss": 0.2309,
      "num_input_tokens_seen": 13017120,
      "step": 61680
    },
    {
      "epoch": 6.786028602860286,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.02506432481945104,
      "loss": 0.233,
      "num_input_tokens_seen": 13018208,
      "step": 61685
    },
    {
      "epoch": 6.786578657865786,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.025063256985704788,
      "loss": 0.2304,
      "num_input_tokens_seen": 13019264,
      "step": 61690
    },
    {
      "epoch": 6.787128712871287,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.025062189059210323,
      "loss": 0.2335,
      "num_input_tokens_seen": 13020320,
      "step": 61695
    },
    {
      "epoch": 6.787678767876788,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.025061121039977492,
      "loss": 0.232,
      "num_input_tokens_seen": 13021440,
      "step": 61700
    },
    {
      "epoch": 6.788228822882289,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.025060052928016132,
      "loss": 0.2315,
      "num_input_tokens_seen": 13022496,
      "step": 61705
    },
    {
      "epoch": 6.788778877887789,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.025058984723336095,
      "loss": 0.234,
      "num_input_tokens_seen": 13023552,
      "step": 61710
    },
    {
      "epoch": 6.789328932893289,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.025057916425947218,
      "loss": 0.2304,
      "num_input_tokens_seen": 13024576,
      "step": 61715
    },
    {
      "epoch": 6.78987898789879,
      "grad_norm": 0.001953125,
      "learning_rate": 0.02505684803585936,
      "loss": 0.2273,
      "num_input_tokens_seen": 13025568,
      "step": 61720
    },
    {
      "epoch": 6.79042904290429,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.02505577955308235,
      "loss": 0.2314,
      "num_input_tokens_seen": 13026656,
      "step": 61725
    },
    {
      "epoch": 6.790979097909791,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.025054710977626054,
      "loss": 0.2273,
      "num_input_tokens_seen": 13027712,
      "step": 61730
    },
    {
      "epoch": 6.791529152915292,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.025053642309500305,
      "loss": 0.2304,
      "num_input_tokens_seen": 13028704,
      "step": 61735
    },
    {
      "epoch": 6.792079207920792,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.025052573548714964,
      "loss": 0.2299,
      "num_input_tokens_seen": 13029696,
      "step": 61740
    },
    {
      "epoch": 6.792629262926293,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.025051504695279874,
      "loss": 0.2294,
      "num_input_tokens_seen": 13030752,
      "step": 61745
    },
    {
      "epoch": 6.793179317931793,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.025050435749204885,
      "loss": 0.2284,
      "num_input_tokens_seen": 13031712,
      "step": 61750
    },
    {
      "epoch": 6.793729372937293,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.025049366710499857,
      "loss": 0.2337,
      "num_input_tokens_seen": 13032800,
      "step": 61755
    },
    {
      "epoch": 6.7942794279427945,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.025048297579174636,
      "loss": 0.2274,
      "num_input_tokens_seen": 13033792,
      "step": 61760
    },
    {
      "epoch": 6.794829482948295,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.025047228355239076,
      "loss": 0.23,
      "num_input_tokens_seen": 13034816,
      "step": 61765
    },
    {
      "epoch": 6.795379537953796,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.02504615903870304,
      "loss": 0.2337,
      "num_input_tokens_seen": 13035840,
      "step": 61770
    },
    {
      "epoch": 6.795929592959296,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.025045089629576373,
      "loss": 0.2326,
      "num_input_tokens_seen": 13036864,
      "step": 61775
    },
    {
      "epoch": 6.796479647964796,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.025044020127868934,
      "loss": 0.2295,
      "num_input_tokens_seen": 13037920,
      "step": 61780
    },
    {
      "epoch": 6.797029702970297,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.02504295053359058,
      "loss": 0.2353,
      "num_input_tokens_seen": 13038912,
      "step": 61785
    },
    {
      "epoch": 6.7975797579757975,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02504188084675117,
      "loss": 0.2311,
      "num_input_tokens_seen": 13040000,
      "step": 61790
    },
    {
      "epoch": 6.798129812981298,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.025040811067360565,
      "loss": 0.231,
      "num_input_tokens_seen": 13041120,
      "step": 61795
    },
    {
      "epoch": 6.798679867986799,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.025039741195428625,
      "loss": 0.231,
      "num_input_tokens_seen": 13042112,
      "step": 61800
    },
    {
      "epoch": 6.799229922992299,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.025038671230965204,
      "loss": 0.2322,
      "num_input_tokens_seen": 13043168,
      "step": 61805
    },
    {
      "epoch": 6.7997799779978,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.025037601173980167,
      "loss": 0.2316,
      "num_input_tokens_seen": 13044192,
      "step": 61810
    },
    {
      "epoch": 6.8003300330033,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.025036531024483378,
      "loss": 0.2347,
      "num_input_tokens_seen": 13045184,
      "step": 61815
    },
    {
      "epoch": 6.8008800880088005,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.025035460782484698,
      "loss": 0.2342,
      "num_input_tokens_seen": 13046208,
      "step": 61820
    },
    {
      "epoch": 6.801430143014302,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.025034390447993993,
      "loss": 0.2357,
      "num_input_tokens_seen": 13047264,
      "step": 61825
    },
    {
      "epoch": 6.801980198019802,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.025033320021021124,
      "loss": 0.2268,
      "num_input_tokens_seen": 13048320,
      "step": 61830
    },
    {
      "epoch": 6.802530253025303,
      "grad_norm": 0.00069427490234375,
      "learning_rate": 0.02503224950157596,
      "loss": 0.2352,
      "num_input_tokens_seen": 13049376,
      "step": 61835
    },
    {
      "epoch": 6.803080308030803,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.025031178889668364,
      "loss": 0.232,
      "num_input_tokens_seen": 13050432,
      "step": 61840
    },
    {
      "epoch": 6.803630363036303,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.025030108185308205,
      "loss": 0.2314,
      "num_input_tokens_seen": 13051520,
      "step": 61845
    },
    {
      "epoch": 6.804180418041804,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.02502903738850536,
      "loss": 0.2314,
      "num_input_tokens_seen": 13052608,
      "step": 61850
    },
    {
      "epoch": 6.804730473047305,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.025027966499269683,
      "loss": 0.2299,
      "num_input_tokens_seen": 13053696,
      "step": 61855
    },
    {
      "epoch": 6.805280528052805,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.025026895517611048,
      "loss": 0.2335,
      "num_input_tokens_seen": 13054784,
      "step": 61860
    },
    {
      "epoch": 6.805830583058306,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.025025824443539332,
      "loss": 0.2325,
      "num_input_tokens_seen": 13055776,
      "step": 61865
    },
    {
      "epoch": 6.806380638063806,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.025024753277064406,
      "loss": 0.2309,
      "num_input_tokens_seen": 13056896,
      "step": 61870
    },
    {
      "epoch": 6.806930693069307,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.025023682018196138,
      "loss": 0.2272,
      "num_input_tokens_seen": 13057856,
      "step": 61875
    },
    {
      "epoch": 6.807480748074807,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.0250226106669444,
      "loss": 0.2309,
      "num_input_tokens_seen": 13058848,
      "step": 61880
    },
    {
      "epoch": 6.8080308030803085,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02502153922331907,
      "loss": 0.2325,
      "num_input_tokens_seen": 13059872,
      "step": 61885
    },
    {
      "epoch": 6.808580858085809,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.025020467687330017,
      "loss": 0.2325,
      "num_input_tokens_seen": 13060928,
      "step": 61890
    },
    {
      "epoch": 6.809130913091309,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.02501939605898713,
      "loss": 0.2346,
      "num_input_tokens_seen": 13061952,
      "step": 61895
    },
    {
      "epoch": 6.80968096809681,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02501832433830027,
      "loss": 0.2298,
      "num_input_tokens_seen": 13063040,
      "step": 61900
    },
    {
      "epoch": 6.81023102310231,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.025017252525279327,
      "loss": 0.2325,
      "num_input_tokens_seen": 13064128,
      "step": 61905
    },
    {
      "epoch": 6.81078107810781,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.025016180619934176,
      "loss": 0.2304,
      "num_input_tokens_seen": 13065152,
      "step": 61910
    },
    {
      "epoch": 6.8113311331133115,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.02501510862227469,
      "loss": 0.2324,
      "num_input_tokens_seen": 13066176,
      "step": 61915
    },
    {
      "epoch": 6.811881188118812,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.025014036532310754,
      "loss": 0.2303,
      "num_input_tokens_seen": 13067264,
      "step": 61920
    },
    {
      "epoch": 6.812431243124313,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02501296435005225,
      "loss": 0.2324,
      "num_input_tokens_seen": 13068256,
      "step": 61925
    },
    {
      "epoch": 6.812981298129813,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.025011892075509058,
      "loss": 0.2324,
      "num_input_tokens_seen": 13069344,
      "step": 61930
    },
    {
      "epoch": 6.813531353135313,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02501081970869106,
      "loss": 0.234,
      "num_input_tokens_seen": 13070400,
      "step": 61935
    },
    {
      "epoch": 6.814081408140814,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02500974724960814,
      "loss": 0.2304,
      "num_input_tokens_seen": 13071456,
      "step": 61940
    },
    {
      "epoch": 6.8146314631463145,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.025008674698270186,
      "loss": 0.233,
      "num_input_tokens_seen": 13072512,
      "step": 61945
    },
    {
      "epoch": 6.815181518151816,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.025007602054687075,
      "loss": 0.2319,
      "num_input_tokens_seen": 13073536,
      "step": 61950
    },
    {
      "epoch": 6.815731573157316,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.0250065293188687,
      "loss": 0.2278,
      "num_input_tokens_seen": 13074624,
      "step": 61955
    },
    {
      "epoch": 6.816281628162816,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.025005456490824945,
      "loss": 0.2319,
      "num_input_tokens_seen": 13075680,
      "step": 61960
    },
    {
      "epoch": 6.816831683168317,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.025004383570565704,
      "loss": 0.2325,
      "num_input_tokens_seen": 13076704,
      "step": 61965
    },
    {
      "epoch": 6.817381738173817,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02500331055810085,
      "loss": 0.2304,
      "num_input_tokens_seen": 13077760,
      "step": 61970
    },
    {
      "epoch": 6.8179317931793175,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.025002237453440292,
      "loss": 0.2324,
      "num_input_tokens_seen": 13078848,
      "step": 61975
    },
    {
      "epoch": 6.818481848184819,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02500116425659391,
      "loss": 0.2304,
      "num_input_tokens_seen": 13079968,
      "step": 61980
    },
    {
      "epoch": 6.819031903190319,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02500009096757159,
      "loss": 0.2319,
      "num_input_tokens_seen": 13080960,
      "step": 61985
    },
    {
      "epoch": 6.81958195819582,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.024999017586383235,
      "loss": 0.2298,
      "num_input_tokens_seen": 13082016,
      "step": 61990
    },
    {
      "epoch": 6.82013201320132,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.024997944113038732,
      "loss": 0.2283,
      "num_input_tokens_seen": 13083136,
      "step": 61995
    },
    {
      "epoch": 6.82068206820682,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.024996870547547977,
      "loss": 0.2319,
      "num_input_tokens_seen": 13084224,
      "step": 62000
    },
    {
      "epoch": 6.821232123212321,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.024995796889920863,
      "loss": 0.234,
      "num_input_tokens_seen": 13085280,
      "step": 62005
    },
    {
      "epoch": 6.821782178217822,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02499472314016728,
      "loss": 0.233,
      "num_input_tokens_seen": 13086336,
      "step": 62010
    },
    {
      "epoch": 6.822332233223323,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.024993649298297137,
      "loss": 0.2309,
      "num_input_tokens_seen": 13087360,
      "step": 62015
    },
    {
      "epoch": 6.822882288228823,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02499257536432032,
      "loss": 0.233,
      "num_input_tokens_seen": 13088448,
      "step": 62020
    },
    {
      "epoch": 6.823432343234323,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02499150133824673,
      "loss": 0.2314,
      "num_input_tokens_seen": 13089440,
      "step": 62025
    },
    {
      "epoch": 6.823982398239824,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.024990427220086264,
      "loss": 0.2324,
      "num_input_tokens_seen": 13090464,
      "step": 62030
    },
    {
      "epoch": 6.824532453245324,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02498935300984883,
      "loss": 0.2309,
      "num_input_tokens_seen": 13091552,
      "step": 62035
    },
    {
      "epoch": 6.825082508250825,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02498827870754432,
      "loss": 0.2319,
      "num_input_tokens_seen": 13092608,
      "step": 62040
    },
    {
      "epoch": 6.825632563256326,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.024987204313182645,
      "loss": 0.2314,
      "num_input_tokens_seen": 13093696,
      "step": 62045
    },
    {
      "epoch": 6.826182618261826,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02498612982677369,
      "loss": 0.2335,
      "num_input_tokens_seen": 13094720,
      "step": 62050
    },
    {
      "epoch": 6.826732673267327,
      "grad_norm": 0.00070953369140625,
      "learning_rate": 0.02498505524832737,
      "loss": 0.2309,
      "num_input_tokens_seen": 13095776,
      "step": 62055
    },
    {
      "epoch": 6.827282728272827,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.024983980577853594,
      "loss": 0.2314,
      "num_input_tokens_seen": 13096896,
      "step": 62060
    },
    {
      "epoch": 6.827832783278328,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.024982905815362253,
      "loss": 0.2303,
      "num_input_tokens_seen": 13098048,
      "step": 62065
    },
    {
      "epoch": 6.8283828382838285,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.02498183096086326,
      "loss": 0.2293,
      "num_input_tokens_seen": 13099072,
      "step": 62070
    },
    {
      "epoch": 6.828932893289329,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.02498075601436652,
      "loss": 0.2304,
      "num_input_tokens_seen": 13100128,
      "step": 62075
    },
    {
      "epoch": 6.82948294829483,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.024979680975881942,
      "loss": 0.2309,
      "num_input_tokens_seen": 13101216,
      "step": 62080
    },
    {
      "epoch": 6.83003300330033,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.024978605845419437,
      "loss": 0.2293,
      "num_input_tokens_seen": 13102336,
      "step": 62085
    },
    {
      "epoch": 6.83058305830583,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.024977530622988908,
      "loss": 0.2283,
      "num_input_tokens_seen": 13103360,
      "step": 62090
    },
    {
      "epoch": 6.831133113311331,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02497645530860027,
      "loss": 0.2335,
      "num_input_tokens_seen": 13104384,
      "step": 62095
    },
    {
      "epoch": 6.8316831683168315,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.02497537990226342,
      "loss": 0.2284,
      "num_input_tokens_seen": 13105440,
      "step": 62100
    },
    {
      "epoch": 6.832233223322332,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.02497430440398829,
      "loss": 0.2258,
      "num_input_tokens_seen": 13106464,
      "step": 62105
    },
    {
      "epoch": 6.832783278327833,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.02497322881378478,
      "loss": 0.2311,
      "num_input_tokens_seen": 13107584,
      "step": 62110
    },
    {
      "epoch": 6.833333333333333,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.024972153131662803,
      "loss": 0.2368,
      "num_input_tokens_seen": 13108672,
      "step": 62115
    },
    {
      "epoch": 6.833883388338834,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02497107735763228,
      "loss": 0.2337,
      "num_input_tokens_seen": 13109728,
      "step": 62120
    },
    {
      "epoch": 6.834433443344334,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.02497000149170312,
      "loss": 0.231,
      "num_input_tokens_seen": 13110784,
      "step": 62125
    },
    {
      "epoch": 6.834983498349835,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.024968925533885238,
      "loss": 0.2346,
      "num_input_tokens_seen": 13111840,
      "step": 62130
    },
    {
      "epoch": 6.835533553355336,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.024967849484188554,
      "loss": 0.2325,
      "num_input_tokens_seen": 13112832,
      "step": 62135
    },
    {
      "epoch": 6.836083608360836,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.024966773342622987,
      "loss": 0.2274,
      "num_input_tokens_seen": 13113888,
      "step": 62140
    },
    {
      "epoch": 6.836633663366337,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02496569710919845,
      "loss": 0.2306,
      "num_input_tokens_seen": 13114976,
      "step": 62145
    },
    {
      "epoch": 6.837183718371837,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.024964620783924866,
      "loss": 0.2378,
      "num_input_tokens_seen": 13116000,
      "step": 62150
    },
    {
      "epoch": 6.837733773377337,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.024963544366812154,
      "loss": 0.2325,
      "num_input_tokens_seen": 13117120,
      "step": 62155
    },
    {
      "epoch": 6.838283828382838,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.024962467857870232,
      "loss": 0.2319,
      "num_input_tokens_seen": 13118208,
      "step": 62160
    },
    {
      "epoch": 6.838833883388339,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.024961391257109024,
      "loss": 0.2366,
      "num_input_tokens_seen": 13119264,
      "step": 62165
    },
    {
      "epoch": 6.83938393839384,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02496031456453846,
      "loss": 0.2329,
      "num_input_tokens_seen": 13120288,
      "step": 62170
    },
    {
      "epoch": 6.83993399339934,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.024959237780168446,
      "loss": 0.2319,
      "num_input_tokens_seen": 13121344,
      "step": 62175
    },
    {
      "epoch": 6.84048404840484,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.024958160904008923,
      "loss": 0.2314,
      "num_input_tokens_seen": 13122336,
      "step": 62180
    },
    {
      "epoch": 6.841034103410341,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.02495708393606981,
      "loss": 0.2335,
      "num_input_tokens_seen": 13123424,
      "step": 62185
    },
    {
      "epoch": 6.841584158415841,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.024956006876361025,
      "loss": 0.2309,
      "num_input_tokens_seen": 13124448,
      "step": 62190
    },
    {
      "epoch": 6.8421342134213425,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.024954929724892508,
      "loss": 0.2319,
      "num_input_tokens_seen": 13125536,
      "step": 62195
    },
    {
      "epoch": 6.842684268426843,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02495385248167418,
      "loss": 0.2319,
      "num_input_tokens_seen": 13126624,
      "step": 62200
    },
    {
      "epoch": 6.843234323432343,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.02495277514671597,
      "loss": 0.2314,
      "num_input_tokens_seen": 13127648,
      "step": 62205
    },
    {
      "epoch": 6.843784378437844,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.024951697720027803,
      "loss": 0.2314,
      "num_input_tokens_seen": 13128736,
      "step": 62210
    },
    {
      "epoch": 6.844334433443344,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.024950620201619617,
      "loss": 0.2319,
      "num_input_tokens_seen": 13129696,
      "step": 62215
    },
    {
      "epoch": 6.8448844884488445,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02494954259150134,
      "loss": 0.2314,
      "num_input_tokens_seen": 13130720,
      "step": 62220
    },
    {
      "epoch": 6.8454345434543455,
      "grad_norm": 0.00077056884765625,
      "learning_rate": 0.0249484648896829,
      "loss": 0.2324,
      "num_input_tokens_seen": 13131776,
      "step": 62225
    },
    {
      "epoch": 6.845984598459846,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.024947387096174237,
      "loss": 0.2314,
      "num_input_tokens_seen": 13132896,
      "step": 62230
    },
    {
      "epoch": 6.846534653465347,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.024946309210985278,
      "loss": 0.2293,
      "num_input_tokens_seen": 13134048,
      "step": 62235
    },
    {
      "epoch": 6.847084708470847,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02494523123412596,
      "loss": 0.2335,
      "num_input_tokens_seen": 13135104,
      "step": 62240
    },
    {
      "epoch": 6.847634763476347,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.024944153165606216,
      "loss": 0.2314,
      "num_input_tokens_seen": 13136128,
      "step": 62245
    },
    {
      "epoch": 6.848184818481848,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.024943075005435987,
      "loss": 0.2313,
      "num_input_tokens_seen": 13137184,
      "step": 62250
    },
    {
      "epoch": 6.8487348734873486,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02494199675362521,
      "loss": 0.2319,
      "num_input_tokens_seen": 13138240,
      "step": 62255
    },
    {
      "epoch": 6.84928492849285,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.024940918410183813,
      "loss": 0.2324,
      "num_input_tokens_seen": 13139328,
      "step": 62260
    },
    {
      "epoch": 6.84983498349835,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.024939839975121742,
      "loss": 0.2314,
      "num_input_tokens_seen": 13140320,
      "step": 62265
    },
    {
      "epoch": 6.85038503850385,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.024938761448448938,
      "loss": 0.2298,
      "num_input_tokens_seen": 13141344,
      "step": 62270
    },
    {
      "epoch": 6.850935093509351,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02493768283017534,
      "loss": 0.2288,
      "num_input_tokens_seen": 13142432,
      "step": 62275
    },
    {
      "epoch": 6.851485148514851,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.024936604120310884,
      "loss": 0.2283,
      "num_input_tokens_seen": 13143488,
      "step": 62280
    },
    {
      "epoch": 6.852035203520352,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.024935525318865517,
      "loss": 0.2283,
      "num_input_tokens_seen": 13144512,
      "step": 62285
    },
    {
      "epoch": 6.852585258525853,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02493444642584918,
      "loss": 0.2314,
      "num_input_tokens_seen": 13145568,
      "step": 62290
    },
    {
      "epoch": 6.853135313531353,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.02493336744127182,
      "loss": 0.2299,
      "num_input_tokens_seen": 13146560,
      "step": 62295
    },
    {
      "epoch": 6.853685368536854,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.02493228836514338,
      "loss": 0.2299,
      "num_input_tokens_seen": 13147648,
      "step": 62300
    },
    {
      "epoch": 6.854235423542354,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.0249312091974738,
      "loss": 0.2309,
      "num_input_tokens_seen": 13148672,
      "step": 62305
    },
    {
      "epoch": 6.854785478547855,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02493012993827303,
      "loss": 0.2289,
      "num_input_tokens_seen": 13149664,
      "step": 62310
    },
    {
      "epoch": 6.8553355335533555,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.02492905058755102,
      "loss": 0.2325,
      "num_input_tokens_seen": 13150752,
      "step": 62315
    },
    {
      "epoch": 6.855885588558856,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.024927971145317713,
      "loss": 0.2284,
      "num_input_tokens_seen": 13151808,
      "step": 62320
    },
    {
      "epoch": 6.856435643564357,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02492689161158306,
      "loss": 0.2315,
      "num_input_tokens_seen": 13152864,
      "step": 62325
    },
    {
      "epoch": 6.856985698569857,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.024925811986357016,
      "loss": 0.2244,
      "num_input_tokens_seen": 13153920,
      "step": 62330
    },
    {
      "epoch": 6.857535753575357,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.024924732269649523,
      "loss": 0.2324,
      "num_input_tokens_seen": 13154976,
      "step": 62335
    },
    {
      "epoch": 6.858085808580858,
      "grad_norm": 0.0084228515625,
      "learning_rate": 0.02492365246147053,
      "loss": 0.2324,
      "num_input_tokens_seen": 13156064,
      "step": 62340
    },
    {
      "epoch": 6.8586358635863585,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.02492257256183,
      "loss": 0.2318,
      "num_input_tokens_seen": 13157088,
      "step": 62345
    },
    {
      "epoch": 6.8591859185918596,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.024921492570737877,
      "loss": 0.2335,
      "num_input_tokens_seen": 13158112,
      "step": 62350
    },
    {
      "epoch": 6.85973597359736,
      "grad_norm": 0.0081787109375,
      "learning_rate": 0.024920412488204115,
      "loss": 0.2318,
      "num_input_tokens_seen": 13159168,
      "step": 62355
    },
    {
      "epoch": 6.86028602860286,
      "grad_norm": 0.007720947265625,
      "learning_rate": 0.024919332314238676,
      "loss": 0.2344,
      "num_input_tokens_seen": 13160256,
      "step": 62360
    },
    {
      "epoch": 6.860836083608361,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.024918252048851508,
      "loss": 0.2318,
      "num_input_tokens_seen": 13161344,
      "step": 62365
    },
    {
      "epoch": 6.861386138613861,
      "grad_norm": 0.007598876953125,
      "learning_rate": 0.024917171692052573,
      "loss": 0.2322,
      "num_input_tokens_seen": 13162464,
      "step": 62370
    },
    {
      "epoch": 6.861936193619362,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.024916091243851822,
      "loss": 0.2364,
      "num_input_tokens_seen": 13163584,
      "step": 62375
    },
    {
      "epoch": 6.862486248624863,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.024915010704259218,
      "loss": 0.228,
      "num_input_tokens_seen": 13164608,
      "step": 62380
    },
    {
      "epoch": 6.863036303630363,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.024913930073284723,
      "loss": 0.2399,
      "num_input_tokens_seen": 13165600,
      "step": 62385
    },
    {
      "epoch": 6.863586358635864,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.024912849350938283,
      "loss": 0.2356,
      "num_input_tokens_seen": 13166624,
      "step": 62390
    },
    {
      "epoch": 6.864136413641364,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.024911768537229872,
      "loss": 0.234,
      "num_input_tokens_seen": 13167648,
      "step": 62395
    },
    {
      "epoch": 6.864686468646864,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.024910687632169446,
      "loss": 0.2303,
      "num_input_tokens_seen": 13168704,
      "step": 62400
    },
    {
      "epoch": 6.865236523652365,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.024909606635766968,
      "loss": 0.2323,
      "num_input_tokens_seen": 13169696,
      "step": 62405
    },
    {
      "epoch": 6.865786578657866,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0249085255480324,
      "loss": 0.2303,
      "num_input_tokens_seen": 13170752,
      "step": 62410
    },
    {
      "epoch": 6.866336633663367,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.0249074443689757,
      "loss": 0.2293,
      "num_input_tokens_seen": 13171904,
      "step": 62415
    },
    {
      "epoch": 6.866886688668867,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.02490636309860685,
      "loss": 0.2334,
      "num_input_tokens_seen": 13173024,
      "step": 62420
    },
    {
      "epoch": 6.867436743674367,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.024905281736935802,
      "loss": 0.2288,
      "num_input_tokens_seen": 13174048,
      "step": 62425
    },
    {
      "epoch": 6.867986798679868,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02490420028397252,
      "loss": 0.2324,
      "num_input_tokens_seen": 13175104,
      "step": 62430
    },
    {
      "epoch": 6.868536853685368,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.024903118739726983,
      "loss": 0.2308,
      "num_input_tokens_seen": 13176256,
      "step": 62435
    },
    {
      "epoch": 6.8690869086908695,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.02490203710420915,
      "loss": 0.2334,
      "num_input_tokens_seen": 13177312,
      "step": 62440
    },
    {
      "epoch": 6.86963696369637,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.024900955377428993,
      "loss": 0.2303,
      "num_input_tokens_seen": 13178368,
      "step": 62445
    },
    {
      "epoch": 6.87018701870187,
      "grad_norm": 0.00750732421875,
      "learning_rate": 0.02489987355939648,
      "loss": 0.234,
      "num_input_tokens_seen": 13179424,
      "step": 62450
    },
    {
      "epoch": 6.870737073707371,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.024898791650121585,
      "loss": 0.233,
      "num_input_tokens_seen": 13180448,
      "step": 62455
    },
    {
      "epoch": 6.871287128712871,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.024897709649614273,
      "loss": 0.2309,
      "num_input_tokens_seen": 13181536,
      "step": 62460
    },
    {
      "epoch": 6.871837183718371,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.024896627557884528,
      "loss": 0.2319,
      "num_input_tokens_seen": 13182592,
      "step": 62465
    },
    {
      "epoch": 6.8723872387238725,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.024895545374942308,
      "loss": 0.2324,
      "num_input_tokens_seen": 13183616,
      "step": 62470
    },
    {
      "epoch": 6.872937293729373,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.024894463100797598,
      "loss": 0.2329,
      "num_input_tokens_seen": 13184672,
      "step": 62475
    },
    {
      "epoch": 6.873487348734874,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.02489338073546037,
      "loss": 0.2314,
      "num_input_tokens_seen": 13185728,
      "step": 62480
    },
    {
      "epoch": 6.874037403740374,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.024892298278940597,
      "loss": 0.2298,
      "num_input_tokens_seen": 13186784,
      "step": 62485
    },
    {
      "epoch": 6.874587458745875,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.024891215731248254,
      "loss": 0.2298,
      "num_input_tokens_seen": 13187808,
      "step": 62490
    },
    {
      "epoch": 6.875137513751375,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.024890133092393327,
      "loss": 0.2293,
      "num_input_tokens_seen": 13188864,
      "step": 62495
    },
    {
      "epoch": 6.8756875687568755,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.02488905036238579,
      "loss": 0.2324,
      "num_input_tokens_seen": 13189920,
      "step": 62500
    },
    {
      "epoch": 6.876237623762377,
      "grad_norm": 0.0074462890625,
      "learning_rate": 0.02488796754123562,
      "loss": 0.2314,
      "num_input_tokens_seen": 13190912,
      "step": 62505
    },
    {
      "epoch": 6.876787678767877,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.024886884628952794,
      "loss": 0.2319,
      "num_input_tokens_seen": 13191936,
      "step": 62510
    },
    {
      "epoch": 6.877337733773377,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.024885801625547304,
      "loss": 0.2289,
      "num_input_tokens_seen": 13193024,
      "step": 62515
    },
    {
      "epoch": 6.877887788778878,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.02488471853102912,
      "loss": 0.2335,
      "num_input_tokens_seen": 13194016,
      "step": 62520
    },
    {
      "epoch": 6.878437843784378,
      "grad_norm": 0.01556396484375,
      "learning_rate": 0.024883635345408227,
      "loss": 0.2373,
      "num_input_tokens_seen": 13195072,
      "step": 62525
    },
    {
      "epoch": 6.878987898789879,
      "grad_norm": 0.007232666015625,
      "learning_rate": 0.024882552068694608,
      "loss": 0.2299,
      "num_input_tokens_seen": 13196096,
      "step": 62530
    },
    {
      "epoch": 6.87953795379538,
      "grad_norm": 0.0074462890625,
      "learning_rate": 0.024881468700898256,
      "loss": 0.2325,
      "num_input_tokens_seen": 13197120,
      "step": 62535
    },
    {
      "epoch": 6.88008800880088,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.024880385242029145,
      "loss": 0.2293,
      "num_input_tokens_seen": 13198176,
      "step": 62540
    },
    {
      "epoch": 6.880638063806381,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.02487930169209726,
      "loss": 0.2314,
      "num_input_tokens_seen": 13199232,
      "step": 62545
    },
    {
      "epoch": 6.881188118811881,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.024878218051112597,
      "loss": 0.2356,
      "num_input_tokens_seen": 13200320,
      "step": 62550
    },
    {
      "epoch": 6.881738173817382,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.02487713431908514,
      "loss": 0.2304,
      "num_input_tokens_seen": 13201408,
      "step": 62555
    },
    {
      "epoch": 6.882288228822882,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.02487605049602487,
      "loss": 0.2319,
      "num_input_tokens_seen": 13202432,
      "step": 62560
    },
    {
      "epoch": 6.882838283828383,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.024874966581941785,
      "loss": 0.2319,
      "num_input_tokens_seen": 13203584,
      "step": 62565
    },
    {
      "epoch": 6.883388338833884,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02487388257684587,
      "loss": 0.2319,
      "num_input_tokens_seen": 13204544,
      "step": 62570
    },
    {
      "epoch": 6.883938393839384,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.024872798480747117,
      "loss": 0.233,
      "num_input_tokens_seen": 13205600,
      "step": 62575
    },
    {
      "epoch": 6.884488448844884,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.024871714293655516,
      "loss": 0.2314,
      "num_input_tokens_seen": 13206624,
      "step": 62580
    },
    {
      "epoch": 6.885038503850385,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.024870630015581065,
      "loss": 0.2277,
      "num_input_tokens_seen": 13207680,
      "step": 62585
    },
    {
      "epoch": 6.885588558855885,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.024869545646533753,
      "loss": 0.2324,
      "num_input_tokens_seen": 13208768,
      "step": 62590
    },
    {
      "epoch": 6.8861386138613865,
      "grad_norm": 0.006683349609375,
      "learning_rate": 0.024868461186523576,
      "loss": 0.2324,
      "num_input_tokens_seen": 13209824,
      "step": 62595
    },
    {
      "epoch": 6.886688668866887,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.024867376635560524,
      "loss": 0.2324,
      "num_input_tokens_seen": 13210944,
      "step": 62600
    },
    {
      "epoch": 6.887238723872387,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.024866291993654595,
      "loss": 0.2308,
      "num_input_tokens_seen": 13212000,
      "step": 62605
    },
    {
      "epoch": 6.887788778877888,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.024865207260815793,
      "loss": 0.2345,
      "num_input_tokens_seen": 13213088,
      "step": 62610
    },
    {
      "epoch": 6.888338833883388,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.024864122437054104,
      "loss": 0.2277,
      "num_input_tokens_seen": 13214144,
      "step": 62615
    },
    {
      "epoch": 6.888888888888889,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.024863037522379536,
      "loss": 0.2319,
      "num_input_tokens_seen": 13215136,
      "step": 62620
    },
    {
      "epoch": 6.8894389438943895,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02486195251680208,
      "loss": 0.2293,
      "num_input_tokens_seen": 13216192,
      "step": 62625
    },
    {
      "epoch": 6.88998899889989,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02486086742033174,
      "loss": 0.2303,
      "num_input_tokens_seen": 13217344,
      "step": 62630
    },
    {
      "epoch": 6.890539053905391,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.024859782232978522,
      "loss": 0.2303,
      "num_input_tokens_seen": 13218336,
      "step": 62635
    },
    {
      "epoch": 6.891089108910891,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.024858696954752416,
      "loss": 0.2309,
      "num_input_tokens_seen": 13219424,
      "step": 62640
    },
    {
      "epoch": 6.891639163916391,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.024857611585663433,
      "loss": 0.235,
      "num_input_tokens_seen": 13220448,
      "step": 62645
    },
    {
      "epoch": 6.892189218921892,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.024856526125721576,
      "loss": 0.2334,
      "num_input_tokens_seen": 13221472,
      "step": 62650
    },
    {
      "epoch": 6.8927392739273925,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.024855440574936844,
      "loss": 0.2308,
      "num_input_tokens_seen": 13222464,
      "step": 62655
    },
    {
      "epoch": 6.893289328932894,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.024854354933319247,
      "loss": 0.2313,
      "num_input_tokens_seen": 13223520,
      "step": 62660
    },
    {
      "epoch": 6.893839383938394,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.024853269200878794,
      "loss": 0.2297,
      "num_input_tokens_seen": 13224608,
      "step": 62665
    },
    {
      "epoch": 6.894389438943895,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02485218337762548,
      "loss": 0.2308,
      "num_input_tokens_seen": 13225728,
      "step": 62670
    },
    {
      "epoch": 6.894939493949395,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02485109746356932,
      "loss": 0.2308,
      "num_input_tokens_seen": 13226816,
      "step": 62675
    },
    {
      "epoch": 6.895489548954895,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.024850011458720327,
      "loss": 0.2334,
      "num_input_tokens_seen": 13227840,
      "step": 62680
    },
    {
      "epoch": 6.896039603960396,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0248489253630885,
      "loss": 0.2313,
      "num_input_tokens_seen": 13228864,
      "step": 62685
    },
    {
      "epoch": 6.896589658965897,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.024847839176683855,
      "loss": 0.2318,
      "num_input_tokens_seen": 13229888,
      "step": 62690
    },
    {
      "epoch": 6.897139713971397,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.024846752899516406,
      "loss": 0.2308,
      "num_input_tokens_seen": 13230912,
      "step": 62695
    },
    {
      "epoch": 6.897689768976898,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.024845666531596155,
      "loss": 0.2308,
      "num_input_tokens_seen": 13231936,
      "step": 62700
    },
    {
      "epoch": 6.898239823982398,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.024844580072933123,
      "loss": 0.2288,
      "num_input_tokens_seen": 13232992,
      "step": 62705
    },
    {
      "epoch": 6.898789878987898,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.02484349352353732,
      "loss": 0.2313,
      "num_input_tokens_seen": 13234016,
      "step": 62710
    },
    {
      "epoch": 6.899339933993399,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.024842406883418756,
      "loss": 0.2283,
      "num_input_tokens_seen": 13235104,
      "step": 62715
    },
    {
      "epoch": 6.8998899889989,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.024841320152587457,
      "loss": 0.2314,
      "num_input_tokens_seen": 13236160,
      "step": 62720
    },
    {
      "epoch": 6.900440044004401,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.024840233331053427,
      "loss": 0.2308,
      "num_input_tokens_seen": 13237184,
      "step": 62725
    },
    {
      "epoch": 6.900990099009901,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.02483914641882669,
      "loss": 0.2283,
      "num_input_tokens_seen": 13238208,
      "step": 62730
    },
    {
      "epoch": 6.901540154015402,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.02483805941591726,
      "loss": 0.2304,
      "num_input_tokens_seen": 13239200,
      "step": 62735
    },
    {
      "epoch": 6.902090209020902,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02483697232233516,
      "loss": 0.2315,
      "num_input_tokens_seen": 13240256,
      "step": 62740
    },
    {
      "epoch": 6.902640264026402,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02483588513809041,
      "loss": 0.2304,
      "num_input_tokens_seen": 13241312,
      "step": 62745
    },
    {
      "epoch": 6.9031903190319035,
      "grad_norm": 0.006805419921875,
      "learning_rate": 0.024834797863193023,
      "loss": 0.2336,
      "num_input_tokens_seen": 13242304,
      "step": 62750
    },
    {
      "epoch": 6.903740374037404,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.02483371049765302,
      "loss": 0.232,
      "num_input_tokens_seen": 13243392,
      "step": 62755
    },
    {
      "epoch": 6.904290429042904,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.02483262304148043,
      "loss": 0.2315,
      "num_input_tokens_seen": 13244384,
      "step": 62760
    },
    {
      "epoch": 6.904840484048405,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.02483153549468527,
      "loss": 0.2283,
      "num_input_tokens_seen": 13245472,
      "step": 62765
    },
    {
      "epoch": 6.905390539053905,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.024830447857277563,
      "loss": 0.2309,
      "num_input_tokens_seen": 13246496,
      "step": 62770
    },
    {
      "epoch": 6.905940594059406,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.024829360129267335,
      "loss": 0.2268,
      "num_input_tokens_seen": 13247552,
      "step": 62775
    },
    {
      "epoch": 6.9064906490649065,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.024828272310664617,
      "loss": 0.2273,
      "num_input_tokens_seen": 13248608,
      "step": 62780
    },
    {
      "epoch": 6.907040704070407,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.024827184401479425,
      "loss": 0.2265,
      "num_input_tokens_seen": 13249696,
      "step": 62785
    },
    {
      "epoch": 6.907590759075908,
      "grad_norm": 0.0037994384765625,
      "learning_rate": 0.024826096401721787,
      "loss": 0.2297,
      "num_input_tokens_seen": 13250720,
      "step": 62790
    },
    {
      "epoch": 6.908140814081408,
      "grad_norm": 0.035400390625,
      "learning_rate": 0.02482500831140174,
      "loss": 0.2334,
      "num_input_tokens_seen": 13251776,
      "step": 62795
    },
    {
      "epoch": 6.908690869086909,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.024823920130529305,
      "loss": 0.2372,
      "num_input_tokens_seen": 13252864,
      "step": 62800
    },
    {
      "epoch": 6.909240924092409,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.02482283185911451,
      "loss": 0.238,
      "num_input_tokens_seen": 13253920,
      "step": 62805
    },
    {
      "epoch": 6.9097909790979095,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.024821743497167387,
      "loss": 0.2389,
      "num_input_tokens_seen": 13254880,
      "step": 62810
    },
    {
      "epoch": 6.910341034103411,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.024820655044697968,
      "loss": 0.2278,
      "num_input_tokens_seen": 13255936,
      "step": 62815
    },
    {
      "epoch": 6.910891089108911,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.024819566501716284,
      "loss": 0.2288,
      "num_input_tokens_seen": 13256992,
      "step": 62820
    },
    {
      "epoch": 6.911441144114411,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.024818477868232366,
      "loss": 0.2277,
      "num_input_tokens_seen": 13258048,
      "step": 62825
    },
    {
      "epoch": 6.911991199119912,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.024817389144256248,
      "loss": 0.2315,
      "num_input_tokens_seen": 13259072,
      "step": 62830
    },
    {
      "epoch": 6.912541254125412,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02481630032979797,
      "loss": 0.231,
      "num_input_tokens_seen": 13260128,
      "step": 62835
    },
    {
      "epoch": 6.913091309130913,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.024815211424867558,
      "loss": 0.2274,
      "num_input_tokens_seen": 13261216,
      "step": 62840
    },
    {
      "epoch": 6.913641364136414,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.024814122429475054,
      "loss": 0.2274,
      "num_input_tokens_seen": 13262336,
      "step": 62845
    },
    {
      "epoch": 6.914191419141914,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.02481303334363049,
      "loss": 0.2342,
      "num_input_tokens_seen": 13263360,
      "step": 62850
    },
    {
      "epoch": 6.914741474147415,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.024811944167343905,
      "loss": 0.2415,
      "num_input_tokens_seen": 13264352,
      "step": 62855
    },
    {
      "epoch": 6.915291529152915,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.024810854900625345,
      "loss": 0.2278,
      "num_input_tokens_seen": 13265376,
      "step": 62860
    },
    {
      "epoch": 6.915841584158416,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.024809765543484838,
      "loss": 0.2372,
      "num_input_tokens_seen": 13266400,
      "step": 62865
    },
    {
      "epoch": 6.916391639163916,
      "grad_norm": 0.005859375,
      "learning_rate": 0.024808676095932434,
      "loss": 0.233,
      "num_input_tokens_seen": 13267520,
      "step": 62870
    },
    {
      "epoch": 6.916941694169417,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.024807586557978167,
      "loss": 0.2314,
      "num_input_tokens_seen": 13268608,
      "step": 62875
    },
    {
      "epoch": 6.917491749174918,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02480649692963208,
      "loss": 0.234,
      "num_input_tokens_seen": 13269696,
      "step": 62880
    },
    {
      "epoch": 6.918041804180418,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.024805407210904216,
      "loss": 0.2376,
      "num_input_tokens_seen": 13270720,
      "step": 62885
    },
    {
      "epoch": 6.918591859185918,
      "grad_norm": 0.009765625,
      "learning_rate": 0.024804317401804617,
      "loss": 0.2277,
      "num_input_tokens_seen": 13271776,
      "step": 62890
    },
    {
      "epoch": 6.919141914191419,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.024803227502343334,
      "loss": 0.2324,
      "num_input_tokens_seen": 13272768,
      "step": 62895
    },
    {
      "epoch": 6.919691969196919,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.024802137512530405,
      "loss": 0.2303,
      "num_input_tokens_seen": 13273888,
      "step": 62900
    },
    {
      "epoch": 6.9202420242024205,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.024801047432375878,
      "loss": 0.2287,
      "num_input_tokens_seen": 13274976,
      "step": 62905
    },
    {
      "epoch": 6.920792079207921,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.024799957261889802,
      "loss": 0.2308,
      "num_input_tokens_seen": 13276064,
      "step": 62910
    },
    {
      "epoch": 6.921342134213422,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.024798867001082216,
      "loss": 0.2308,
      "num_input_tokens_seen": 13277152,
      "step": 62915
    },
    {
      "epoch": 6.921892189218922,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02479777664996318,
      "loss": 0.2314,
      "num_input_tokens_seen": 13278240,
      "step": 62920
    },
    {
      "epoch": 6.922442244224422,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.024796686208542737,
      "loss": 0.2303,
      "num_input_tokens_seen": 13279264,
      "step": 62925
    },
    {
      "epoch": 6.922992299229923,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.024795595676830942,
      "loss": 0.2293,
      "num_input_tokens_seen": 13280288,
      "step": 62930
    },
    {
      "epoch": 6.9235423542354235,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.024794505054837834,
      "loss": 0.2329,
      "num_input_tokens_seen": 13281312,
      "step": 62935
    },
    {
      "epoch": 6.924092409240924,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.024793414342573476,
      "loss": 0.2319,
      "num_input_tokens_seen": 13282368,
      "step": 62940
    },
    {
      "epoch": 6.924642464246425,
      "grad_norm": 0.0027923583984375,
      "learning_rate": 0.024792323540047918,
      "loss": 0.2318,
      "num_input_tokens_seen": 13283424,
      "step": 62945
    },
    {
      "epoch": 6.925192519251925,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.024791232647271216,
      "loss": 0.2334,
      "num_input_tokens_seen": 13284448,
      "step": 62950
    },
    {
      "epoch": 6.925742574257426,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.024790141664253414,
      "loss": 0.2318,
      "num_input_tokens_seen": 13285536,
      "step": 62955
    },
    {
      "epoch": 6.926292629262926,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.02478905059100458,
      "loss": 0.2298,
      "num_input_tokens_seen": 13286656,
      "step": 62960
    },
    {
      "epoch": 6.9268426842684265,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.024787959427534763,
      "loss": 0.2293,
      "num_input_tokens_seen": 13287648,
      "step": 62965
    },
    {
      "epoch": 6.927392739273928,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.02478686817385402,
      "loss": 0.2319,
      "num_input_tokens_seen": 13288672,
      "step": 62970
    },
    {
      "epoch": 6.927942794279428,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02478577682997241,
      "loss": 0.235,
      "num_input_tokens_seen": 13289728,
      "step": 62975
    },
    {
      "epoch": 6.928492849284929,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02478468539589999,
      "loss": 0.2308,
      "num_input_tokens_seen": 13290752,
      "step": 62980
    },
    {
      "epoch": 6.929042904290429,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.024783593871646817,
      "loss": 0.2324,
      "num_input_tokens_seen": 13291776,
      "step": 62985
    },
    {
      "epoch": 6.929592959295929,
      "grad_norm": 0.0004253387451171875,
      "learning_rate": 0.02478250225722296,
      "loss": 0.2308,
      "num_input_tokens_seen": 13292864,
      "step": 62990
    },
    {
      "epoch": 6.93014301430143,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.024781410552638466,
      "loss": 0.2293,
      "num_input_tokens_seen": 13293888,
      "step": 62995
    },
    {
      "epoch": 6.930693069306931,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.024780318757903406,
      "loss": 0.2329,
      "num_input_tokens_seen": 13294976,
      "step": 63000
    },
    {
      "epoch": 6.931243124312431,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.024779226873027846,
      "loss": 0.2297,
      "num_input_tokens_seen": 13296000,
      "step": 63005
    },
    {
      "epoch": 6.931793179317932,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02477813489802184,
      "loss": 0.2318,
      "num_input_tokens_seen": 13297056,
      "step": 63010
    },
    {
      "epoch": 6.932343234323432,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.024777042832895464,
      "loss": 0.2313,
      "num_input_tokens_seen": 13298176,
      "step": 63015
    },
    {
      "epoch": 6.932893289328933,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02477595067765877,
      "loss": 0.2303,
      "num_input_tokens_seen": 13299328,
      "step": 63020
    },
    {
      "epoch": 6.933443344334433,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.024774858432321828,
      "loss": 0.2313,
      "num_input_tokens_seen": 13300416,
      "step": 63025
    },
    {
      "epoch": 6.933993399339934,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.024773766096894707,
      "loss": 0.2329,
      "num_input_tokens_seen": 13301440,
      "step": 63030
    },
    {
      "epoch": 6.934543454345435,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.024772673671387475,
      "loss": 0.2313,
      "num_input_tokens_seen": 13302528,
      "step": 63035
    },
    {
      "epoch": 6.935093509350935,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.024771581155810202,
      "loss": 0.2313,
      "num_input_tokens_seen": 13303520,
      "step": 63040
    },
    {
      "epoch": 6.935643564356436,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02477048855017295,
      "loss": 0.2318,
      "num_input_tokens_seen": 13304544,
      "step": 63045
    },
    {
      "epoch": 6.936193619361936,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.024769395854485794,
      "loss": 0.2313,
      "num_input_tokens_seen": 13305600,
      "step": 63050
    },
    {
      "epoch": 6.936743674367436,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.024768303068758806,
      "loss": 0.2329,
      "num_input_tokens_seen": 13306688,
      "step": 63055
    },
    {
      "epoch": 6.9372937293729375,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.024767210193002053,
      "loss": 0.2303,
      "num_input_tokens_seen": 13307680,
      "step": 63060
    },
    {
      "epoch": 6.937843784378438,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.024766117227225617,
      "loss": 0.2329,
      "num_input_tokens_seen": 13308736,
      "step": 63065
    },
    {
      "epoch": 6.938393839383938,
      "grad_norm": 0.000659942626953125,
      "learning_rate": 0.02476502417143956,
      "loss": 0.2314,
      "num_input_tokens_seen": 13309760,
      "step": 63070
    },
    {
      "epoch": 6.938943894389439,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.024763931025653958,
      "loss": 0.2324,
      "num_input_tokens_seen": 13310816,
      "step": 63075
    },
    {
      "epoch": 6.939493949394939,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02476283778987889,
      "loss": 0.2335,
      "num_input_tokens_seen": 13311904,
      "step": 63080
    },
    {
      "epoch": 6.94004400440044,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.024761744464124436,
      "loss": 0.2303,
      "num_input_tokens_seen": 13312928,
      "step": 63085
    },
    {
      "epoch": 6.9405940594059405,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.024760651048400664,
      "loss": 0.2324,
      "num_input_tokens_seen": 13313984,
      "step": 63090
    },
    {
      "epoch": 6.941144114411442,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.024759557542717654,
      "loss": 0.2303,
      "num_input_tokens_seen": 13315104,
      "step": 63095
    },
    {
      "epoch": 6.941694169416942,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.024758463947085488,
      "loss": 0.2319,
      "num_input_tokens_seen": 13316192,
      "step": 63100
    },
    {
      "epoch": 6.942244224422442,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.024757370261514238,
      "loss": 0.233,
      "num_input_tokens_seen": 13317216,
      "step": 63105
    },
    {
      "epoch": 6.942794279427943,
      "grad_norm": 0.005859375,
      "learning_rate": 0.024756276486013994,
      "loss": 0.2304,
      "num_input_tokens_seen": 13318304,
      "step": 63110
    },
    {
      "epoch": 6.943344334433443,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.024755182620594826,
      "loss": 0.2335,
      "num_input_tokens_seen": 13319360,
      "step": 63115
    },
    {
      "epoch": 6.9438943894389435,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02475408866526682,
      "loss": 0.2319,
      "num_input_tokens_seen": 13320416,
      "step": 63120
    },
    {
      "epoch": 6.944444444444445,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.024752994620040065,
      "loss": 0.2314,
      "num_input_tokens_seen": 13321504,
      "step": 63125
    },
    {
      "epoch": 6.944994499449945,
      "grad_norm": 0.000598907470703125,
      "learning_rate": 0.02475190048492464,
      "loss": 0.2298,
      "num_input_tokens_seen": 13322528,
      "step": 63130
    },
    {
      "epoch": 6.945544554455445,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02475080625993062,
      "loss": 0.2288,
      "num_input_tokens_seen": 13323584,
      "step": 63135
    },
    {
      "epoch": 6.946094609460946,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.0247497119450681,
      "loss": 0.234,
      "num_input_tokens_seen": 13324672,
      "step": 63140
    },
    {
      "epoch": 6.946644664466446,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.024748617540347163,
      "loss": 0.2314,
      "num_input_tokens_seen": 13325664,
      "step": 63145
    },
    {
      "epoch": 6.947194719471947,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0247475230457779,
      "loss": 0.2314,
      "num_input_tokens_seen": 13326752,
      "step": 63150
    },
    {
      "epoch": 6.947744774477448,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02474642846137039,
      "loss": 0.2303,
      "num_input_tokens_seen": 13327808,
      "step": 63155
    },
    {
      "epoch": 6.948294829482949,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.02474533378713473,
      "loss": 0.2329,
      "num_input_tokens_seen": 13328864,
      "step": 63160
    },
    {
      "epoch": 6.948844884488449,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.024744239023081,
      "loss": 0.234,
      "num_input_tokens_seen": 13329920,
      "step": 63165
    },
    {
      "epoch": 6.949394939493949,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.024743144169219293,
      "loss": 0.2329,
      "num_input_tokens_seen": 13331008,
      "step": 63170
    },
    {
      "epoch": 6.94994499449945,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.02474204922555971,
      "loss": 0.2314,
      "num_input_tokens_seen": 13332064,
      "step": 63175
    },
    {
      "epoch": 6.9504950495049505,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.02474095419211233,
      "loss": 0.234,
      "num_input_tokens_seen": 13333152,
      "step": 63180
    },
    {
      "epoch": 6.951045104510451,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.024739859068887246,
      "loss": 0.2314,
      "num_input_tokens_seen": 13334208,
      "step": 63185
    },
    {
      "epoch": 6.951595159515952,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.024738763855894557,
      "loss": 0.2308,
      "num_input_tokens_seen": 13335200,
      "step": 63190
    },
    {
      "epoch": 6.952145214521452,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.024737668553144358,
      "loss": 0.2314,
      "num_input_tokens_seen": 13336224,
      "step": 63195
    },
    {
      "epoch": 6.952695269526953,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.024736573160646737,
      "loss": 0.2334,
      "num_input_tokens_seen": 13337248,
      "step": 63200
    },
    {
      "epoch": 6.953245324532453,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.02473547767841179,
      "loss": 0.2272,
      "num_input_tokens_seen": 13338272,
      "step": 63205
    },
    {
      "epoch": 6.9537953795379535,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02473438210644962,
      "loss": 0.2324,
      "num_input_tokens_seen": 13339360,
      "step": 63210
    },
    {
      "epoch": 6.9543454345434546,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02473328644477032,
      "loss": 0.2335,
      "num_input_tokens_seen": 13340352,
      "step": 63215
    },
    {
      "epoch": 6.954895489548955,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.024732190693383994,
      "loss": 0.2345,
      "num_input_tokens_seen": 13341408,
      "step": 63220
    },
    {
      "epoch": 6.955445544554456,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.024731094852300732,
      "loss": 0.2309,
      "num_input_tokens_seen": 13342464,
      "step": 63225
    },
    {
      "epoch": 6.955995599559956,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.02472999892153064,
      "loss": 0.2309,
      "num_input_tokens_seen": 13343520,
      "step": 63230
    },
    {
      "epoch": 6.956545654565456,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02472890290108382,
      "loss": 0.2303,
      "num_input_tokens_seen": 13344576,
      "step": 63235
    },
    {
      "epoch": 6.957095709570957,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.024727806790970364,
      "loss": 0.2345,
      "num_input_tokens_seen": 13345600,
      "step": 63240
    },
    {
      "epoch": 6.957645764576458,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.024726710591200383,
      "loss": 0.2303,
      "num_input_tokens_seen": 13346624,
      "step": 63245
    },
    {
      "epoch": 6.958195819581958,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02472561430178398,
      "loss": 0.2324,
      "num_input_tokens_seen": 13347616,
      "step": 63250
    },
    {
      "epoch": 6.958745874587459,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.024724517922731254,
      "loss": 0.2314,
      "num_input_tokens_seen": 13348672,
      "step": 63255
    },
    {
      "epoch": 6.959295929592959,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.024723421454052312,
      "loss": 0.2329,
      "num_input_tokens_seen": 13349728,
      "step": 63260
    },
    {
      "epoch": 6.95984598459846,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.02472232489575726,
      "loss": 0.2314,
      "num_input_tokens_seen": 13350752,
      "step": 63265
    },
    {
      "epoch": 6.96039603960396,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.02472122824785621,
      "loss": 0.2351,
      "num_input_tokens_seen": 13351872,
      "step": 63270
    },
    {
      "epoch": 6.960946094609461,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.024720131510359258,
      "loss": 0.2309,
      "num_input_tokens_seen": 13352928,
      "step": 63275
    },
    {
      "epoch": 6.961496149614962,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02471903468327652,
      "loss": 0.2324,
      "num_input_tokens_seen": 13354048,
      "step": 63280
    },
    {
      "epoch": 6.962046204620462,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.024717937766618105,
      "loss": 0.2308,
      "num_input_tokens_seen": 13355136,
      "step": 63285
    },
    {
      "epoch": 6.962596259625963,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.024716840760394116,
      "loss": 0.2303,
      "num_input_tokens_seen": 13356256,
      "step": 63290
    },
    {
      "epoch": 6.963146314631463,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02471574366461467,
      "loss": 0.2308,
      "num_input_tokens_seen": 13357312,
      "step": 63295
    },
    {
      "epoch": 6.963696369636963,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.024714646479289882,
      "loss": 0.234,
      "num_input_tokens_seen": 13358368,
      "step": 63300
    },
    {
      "epoch": 6.9642464246424645,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.024713549204429853,
      "loss": 0.2324,
      "num_input_tokens_seen": 13359488,
      "step": 63305
    },
    {
      "epoch": 6.964796479647965,
      "grad_norm": 0.0004673004150390625,
      "learning_rate": 0.024712451840044704,
      "loss": 0.2309,
      "num_input_tokens_seen": 13360608,
      "step": 63310
    },
    {
      "epoch": 6.965346534653465,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.024711354386144545,
      "loss": 0.2324,
      "num_input_tokens_seen": 13361728,
      "step": 63315
    },
    {
      "epoch": 6.965896589658966,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.0247102568427395,
      "loss": 0.2313,
      "num_input_tokens_seen": 13362720,
      "step": 63320
    },
    {
      "epoch": 6.966446644664466,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.024709159209839668,
      "loss": 0.2308,
      "num_input_tokens_seen": 13363808,
      "step": 63325
    },
    {
      "epoch": 6.966996699669967,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.024708061487455177,
      "loss": 0.2308,
      "num_input_tokens_seen": 13364896,
      "step": 63330
    },
    {
      "epoch": 6.9675467546754675,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.024706963675596143,
      "loss": 0.2314,
      "num_input_tokens_seen": 13365888,
      "step": 63335
    },
    {
      "epoch": 6.968096809680969,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.024705865774272685,
      "loss": 0.2329,
      "num_input_tokens_seen": 13366880,
      "step": 63340
    },
    {
      "epoch": 6.968646864686469,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.024704767783494914,
      "loss": 0.2319,
      "num_input_tokens_seen": 13368000,
      "step": 63345
    },
    {
      "epoch": 6.969196919691969,
      "grad_norm": 0.009765625,
      "learning_rate": 0.024703669703272958,
      "loss": 0.2314,
      "num_input_tokens_seen": 13369056,
      "step": 63350
    },
    {
      "epoch": 6.96974697469747,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.024702571533616935,
      "loss": 0.2319,
      "num_input_tokens_seen": 13370112,
      "step": 63355
    },
    {
      "epoch": 6.97029702970297,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.024701473274536964,
      "loss": 0.2293,
      "num_input_tokens_seen": 13371168,
      "step": 63360
    },
    {
      "epoch": 6.9708470847084705,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02470037492604317,
      "loss": 0.2283,
      "num_input_tokens_seen": 13372256,
      "step": 63365
    },
    {
      "epoch": 6.971397139713972,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.024699276488145676,
      "loss": 0.2298,
      "num_input_tokens_seen": 13373280,
      "step": 63370
    },
    {
      "epoch": 6.971947194719472,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.024698177960854602,
      "loss": 0.2303,
      "num_input_tokens_seen": 13374400,
      "step": 63375
    },
    {
      "epoch": 6.972497249724973,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.02469707934418008,
      "loss": 0.2288,
      "num_input_tokens_seen": 13375488,
      "step": 63380
    },
    {
      "epoch": 6.973047304730473,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.024695980638132227,
      "loss": 0.2309,
      "num_input_tokens_seen": 13376544,
      "step": 63385
    },
    {
      "epoch": 6.973597359735973,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.024694881842721176,
      "loss": 0.235,
      "num_input_tokens_seen": 13377632,
      "step": 63390
    },
    {
      "epoch": 6.974147414741474,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.02469378295795705,
      "loss": 0.2293,
      "num_input_tokens_seen": 13378688,
      "step": 63395
    },
    {
      "epoch": 6.974697469746975,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.024692683983849977,
      "loss": 0.2319,
      "num_input_tokens_seen": 13379744,
      "step": 63400
    },
    {
      "epoch": 6.975247524752476,
      "grad_norm": 0.00063323974609375,
      "learning_rate": 0.024691584920410087,
      "loss": 0.2283,
      "num_input_tokens_seen": 13380736,
      "step": 63405
    },
    {
      "epoch": 6.975797579757976,
      "grad_norm": 0.00041961669921875,
      "learning_rate": 0.02469048576764751,
      "loss": 0.2324,
      "num_input_tokens_seen": 13381760,
      "step": 63410
    },
    {
      "epoch": 6.976347634763476,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.024689386525572374,
      "loss": 0.2335,
      "num_input_tokens_seen": 13382784,
      "step": 63415
    },
    {
      "epoch": 6.976897689768977,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.024688287194194816,
      "loss": 0.2288,
      "num_input_tokens_seen": 13383776,
      "step": 63420
    },
    {
      "epoch": 6.977447744774477,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.02468718777352496,
      "loss": 0.2309,
      "num_input_tokens_seen": 13384832,
      "step": 63425
    },
    {
      "epoch": 6.977997799779978,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.024686088263572943,
      "loss": 0.234,
      "num_input_tokens_seen": 13385920,
      "step": 63430
    },
    {
      "epoch": 6.978547854785479,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.024684988664348897,
      "loss": 0.2314,
      "num_input_tokens_seen": 13386944,
      "step": 63435
    },
    {
      "epoch": 6.979097909790979,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02468388897586296,
      "loss": 0.2293,
      "num_input_tokens_seen": 13388032,
      "step": 63440
    },
    {
      "epoch": 6.97964796479648,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.024682789198125266,
      "loss": 0.2345,
      "num_input_tokens_seen": 13389088,
      "step": 63445
    },
    {
      "epoch": 6.98019801980198,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02468168933114595,
      "loss": 0.2288,
      "num_input_tokens_seen": 13390080,
      "step": 63450
    },
    {
      "epoch": 6.98074807480748,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.024680589374935145,
      "loss": 0.2314,
      "num_input_tokens_seen": 13391104,
      "step": 63455
    },
    {
      "epoch": 6.9812981298129815,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.024679489329502997,
      "loss": 0.2309,
      "num_input_tokens_seen": 13392128,
      "step": 63460
    },
    {
      "epoch": 6.981848184818482,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02467838919485964,
      "loss": 0.2303,
      "num_input_tokens_seen": 13393152,
      "step": 63465
    },
    {
      "epoch": 6.982398239823983,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.024677288971015215,
      "loss": 0.2267,
      "num_input_tokens_seen": 13394144,
      "step": 63470
    },
    {
      "epoch": 6.982948294829483,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.024676188657979856,
      "loss": 0.232,
      "num_input_tokens_seen": 13395264,
      "step": 63475
    },
    {
      "epoch": 6.983498349834983,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.024675088255763714,
      "loss": 0.2324,
      "num_input_tokens_seen": 13396320,
      "step": 63480
    },
    {
      "epoch": 6.984048404840484,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.024673987764376924,
      "loss": 0.2278,
      "num_input_tokens_seen": 13397376,
      "step": 63485
    },
    {
      "epoch": 6.9845984598459845,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.024672887183829633,
      "loss": 0.232,
      "num_input_tokens_seen": 13398464,
      "step": 63490
    },
    {
      "epoch": 6.985148514851485,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02467178651413198,
      "loss": 0.2361,
      "num_input_tokens_seen": 13399456,
      "step": 63495
    },
    {
      "epoch": 6.985698569856986,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.024670685755294113,
      "loss": 0.2283,
      "num_input_tokens_seen": 13400576,
      "step": 63500
    },
    {
      "epoch": 6.986248624862486,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.024669584907326174,
      "loss": 0.2314,
      "num_input_tokens_seen": 13401632,
      "step": 63505
    },
    {
      "epoch": 6.986798679867987,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.024668483970238315,
      "loss": 0.2329,
      "num_input_tokens_seen": 13402656,
      "step": 63510
    },
    {
      "epoch": 6.987348734873487,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.024667382944040677,
      "loss": 0.2324,
      "num_input_tokens_seen": 13403680,
      "step": 63515
    },
    {
      "epoch": 6.987898789878988,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.024666281828743405,
      "loss": 0.2314,
      "num_input_tokens_seen": 13404704,
      "step": 63520
    },
    {
      "epoch": 6.988448844884489,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.024665180624356655,
      "loss": 0.2319,
      "num_input_tokens_seen": 13405728,
      "step": 63525
    },
    {
      "epoch": 6.988998899889989,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.024664079330890574,
      "loss": 0.2319,
      "num_input_tokens_seen": 13406816,
      "step": 63530
    },
    {
      "epoch": 6.98954895489549,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02466297794835531,
      "loss": 0.2335,
      "num_input_tokens_seen": 13407808,
      "step": 63535
    },
    {
      "epoch": 6.99009900990099,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.024661876476761015,
      "loss": 0.2309,
      "num_input_tokens_seen": 13408864,
      "step": 63540
    },
    {
      "epoch": 6.99064906490649,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.02466077491611784,
      "loss": 0.234,
      "num_input_tokens_seen": 13409888,
      "step": 63545
    },
    {
      "epoch": 6.991199119911991,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02465967326643594,
      "loss": 0.2293,
      "num_input_tokens_seen": 13410944,
      "step": 63550
    },
    {
      "epoch": 6.991749174917492,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.024658571527725468,
      "loss": 0.2293,
      "num_input_tokens_seen": 13411968,
      "step": 63555
    },
    {
      "epoch": 6.992299229922993,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.024657469699996572,
      "loss": 0.2314,
      "num_input_tokens_seen": 13412960,
      "step": 63560
    },
    {
      "epoch": 6.992849284928493,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.024656367783259414,
      "loss": 0.2314,
      "num_input_tokens_seen": 13414080,
      "step": 63565
    },
    {
      "epoch": 6.993399339933993,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.024655265777524152,
      "loss": 0.2319,
      "num_input_tokens_seen": 13415136,
      "step": 63570
    },
    {
      "epoch": 6.993949394939494,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.024654163682800927,
      "loss": 0.2288,
      "num_input_tokens_seen": 13416192,
      "step": 63575
    },
    {
      "epoch": 6.994499449944994,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02465306149909992,
      "loss": 0.2283,
      "num_input_tokens_seen": 13417344,
      "step": 63580
    },
    {
      "epoch": 6.9950495049504955,
      "grad_norm": 0.000736236572265625,
      "learning_rate": 0.024651959226431274,
      "loss": 0.2283,
      "num_input_tokens_seen": 13418400,
      "step": 63585
    },
    {
      "epoch": 6.995599559955996,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02465085686480515,
      "loss": 0.2288,
      "num_input_tokens_seen": 13419424,
      "step": 63590
    },
    {
      "epoch": 6.996149614961496,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.024649754414231707,
      "loss": 0.2357,
      "num_input_tokens_seen": 13420512,
      "step": 63595
    },
    {
      "epoch": 6.996699669966997,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.024648651874721112,
      "loss": 0.2347,
      "num_input_tokens_seen": 13421504,
      "step": 63600
    },
    {
      "epoch": 6.997249724972497,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02464754924628352,
      "loss": 0.2305,
      "num_input_tokens_seen": 13422560,
      "step": 63605
    },
    {
      "epoch": 6.997799779977997,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.024646446528929094,
      "loss": 0.2295,
      "num_input_tokens_seen": 13423584,
      "step": 63610
    },
    {
      "epoch": 6.9983498349834985,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.024645343722668005,
      "loss": 0.2304,
      "num_input_tokens_seen": 13424608,
      "step": 63615
    },
    {
      "epoch": 6.998899889988999,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.024644240827510405,
      "loss": 0.2274,
      "num_input_tokens_seen": 13425696,
      "step": 63620
    },
    {
      "epoch": 6.9994499449945,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02464313784346647,
      "loss": 0.2326,
      "num_input_tokens_seen": 13426720,
      "step": 63625
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.02464203477054636,
      "loss": 0.2362,
      "num_input_tokens_seen": 13427712,
      "step": 63630
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.23165836930274963,
      "eval_runtime": 60.5519,
      "eval_samples_per_second": 66.72,
      "eval_steps_per_second": 16.68,
      "num_input_tokens_seen": 13427712,
      "step": 63630
    },
    {
      "epoch": 7.0005500550055,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.024640931608760238,
      "loss": 0.2372,
      "num_input_tokens_seen": 13428736,
      "step": 63635
    },
    {
      "epoch": 7.001100110011001,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.024639828358118277,
      "loss": 0.2341,
      "num_input_tokens_seen": 13429728,
      "step": 63640
    },
    {
      "epoch": 7.0016501650165015,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.02463872501863065,
      "loss": 0.234,
      "num_input_tokens_seen": 13430816,
      "step": 63645
    },
    {
      "epoch": 7.002200220022003,
      "grad_norm": 0.000667572021484375,
      "learning_rate": 0.02463762159030751,
      "loss": 0.2319,
      "num_input_tokens_seen": 13431904,
      "step": 63650
    },
    {
      "epoch": 7.002750275027503,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.024636518073159044,
      "loss": 0.2319,
      "num_input_tokens_seen": 13433024,
      "step": 63655
    },
    {
      "epoch": 7.003300330033003,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.024635414467195412,
      "loss": 0.2298,
      "num_input_tokens_seen": 13434080,
      "step": 63660
    },
    {
      "epoch": 7.003850385038504,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.024634310772426785,
      "loss": 0.2309,
      "num_input_tokens_seen": 13435104,
      "step": 63665
    },
    {
      "epoch": 7.004400440044004,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.024633206988863343,
      "loss": 0.2335,
      "num_input_tokens_seen": 13436128,
      "step": 63670
    },
    {
      "epoch": 7.0049504950495045,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.024632103116515255,
      "loss": 0.233,
      "num_input_tokens_seen": 13437184,
      "step": 63675
    },
    {
      "epoch": 7.005500550055006,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.024630999155392692,
      "loss": 0.2298,
      "num_input_tokens_seen": 13438208,
      "step": 63680
    },
    {
      "epoch": 7.006050605060506,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.024629895105505832,
      "loss": 0.2314,
      "num_input_tokens_seen": 13439328,
      "step": 63685
    },
    {
      "epoch": 7.006600660066007,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02462879096686485,
      "loss": 0.2314,
      "num_input_tokens_seen": 13440352,
      "step": 63690
    },
    {
      "epoch": 7.007150715071507,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02462768673947992,
      "loss": 0.2309,
      "num_input_tokens_seen": 13441344,
      "step": 63695
    },
    {
      "epoch": 7.007700770077007,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02462658242336123,
      "loss": 0.2335,
      "num_input_tokens_seen": 13442400,
      "step": 63700
    },
    {
      "epoch": 7.008250825082508,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02462547801851894,
      "loss": 0.2314,
      "num_input_tokens_seen": 13443456,
      "step": 63705
    },
    {
      "epoch": 7.008800880088009,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.024624373524963238,
      "loss": 0.2335,
      "num_input_tokens_seen": 13444448,
      "step": 63710
    },
    {
      "epoch": 7.00935093509351,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.024623268942704304,
      "loss": 0.2313,
      "num_input_tokens_seen": 13445440,
      "step": 63715
    },
    {
      "epoch": 7.00990099009901,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02462216427175232,
      "loss": 0.2314,
      "num_input_tokens_seen": 13446432,
      "step": 63720
    },
    {
      "epoch": 7.01045104510451,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.024621059512117467,
      "loss": 0.2309,
      "num_input_tokens_seen": 13447456,
      "step": 63725
    },
    {
      "epoch": 7.011001100110011,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.024619954663809923,
      "loss": 0.2314,
      "num_input_tokens_seen": 13448544,
      "step": 63730
    },
    {
      "epoch": 7.011551155115511,
      "grad_norm": 0.00070953369140625,
      "learning_rate": 0.02461884972683987,
      "loss": 0.2309,
      "num_input_tokens_seen": 13449632,
      "step": 63735
    },
    {
      "epoch": 7.0121012101210125,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.024617744701217498,
      "loss": 0.2298,
      "num_input_tokens_seen": 13450688,
      "step": 63740
    },
    {
      "epoch": 7.012651265126513,
      "grad_norm": 0.000396728515625,
      "learning_rate": 0.024616639586952985,
      "loss": 0.2329,
      "num_input_tokens_seen": 13451744,
      "step": 63745
    },
    {
      "epoch": 7.013201320132013,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.024615534384056524,
      "loss": 0.2309,
      "num_input_tokens_seen": 13452736,
      "step": 63750
    },
    {
      "epoch": 7.013751375137514,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.024614429092538293,
      "loss": 0.2314,
      "num_input_tokens_seen": 13453792,
      "step": 63755
    },
    {
      "epoch": 7.014301430143014,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.024613323712408486,
      "loss": 0.2304,
      "num_input_tokens_seen": 13454848,
      "step": 63760
    },
    {
      "epoch": 7.014851485148514,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02461221824367728,
      "loss": 0.2324,
      "num_input_tokens_seen": 13455872,
      "step": 63765
    },
    {
      "epoch": 7.0154015401540155,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02461111268635488,
      "loss": 0.2303,
      "num_input_tokens_seen": 13456960,
      "step": 63770
    },
    {
      "epoch": 7.015951595159516,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02461000704045146,
      "loss": 0.2298,
      "num_input_tokens_seen": 13457952,
      "step": 63775
    },
    {
      "epoch": 7.016501650165017,
      "grad_norm": 0.000751495361328125,
      "learning_rate": 0.02460890130597722,
      "loss": 0.2319,
      "num_input_tokens_seen": 13458944,
      "step": 63780
    },
    {
      "epoch": 7.017051705170517,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.024607795482942348,
      "loss": 0.2319,
      "num_input_tokens_seen": 13460000,
      "step": 63785
    },
    {
      "epoch": 7.017601760176017,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.024606689571357032,
      "loss": 0.2314,
      "num_input_tokens_seen": 13461056,
      "step": 63790
    },
    {
      "epoch": 7.018151815181518,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.024605583571231472,
      "loss": 0.2314,
      "num_input_tokens_seen": 13462048,
      "step": 63795
    },
    {
      "epoch": 7.0187018701870185,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02460447748257585,
      "loss": 0.2319,
      "num_input_tokens_seen": 13463040,
      "step": 63800
    },
    {
      "epoch": 7.01925192519252,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.024603371305400377,
      "loss": 0.232,
      "num_input_tokens_seen": 13464096,
      "step": 63805
    },
    {
      "epoch": 7.01980198019802,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.024602265039715233,
      "loss": 0.2309,
      "num_input_tokens_seen": 13465184,
      "step": 63810
    },
    {
      "epoch": 7.02035203520352,
      "grad_norm": 0.000652313232421875,
      "learning_rate": 0.024601158685530623,
      "loss": 0.2319,
      "num_input_tokens_seen": 13466208,
      "step": 63815
    },
    {
      "epoch": 7.020902090209021,
      "grad_norm": 0.009765625,
      "learning_rate": 0.024600052242856735,
      "loss": 0.2309,
      "num_input_tokens_seen": 13467232,
      "step": 63820
    },
    {
      "epoch": 7.021452145214521,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.024598945711703776,
      "loss": 0.2309,
      "num_input_tokens_seen": 13468320,
      "step": 63825
    },
    {
      "epoch": 7.022002200220022,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.024597839092081943,
      "loss": 0.2314,
      "num_input_tokens_seen": 13469312,
      "step": 63830
    },
    {
      "epoch": 7.022552255225523,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.024596732384001428,
      "loss": 0.2324,
      "num_input_tokens_seen": 13470400,
      "step": 63835
    },
    {
      "epoch": 7.023102310231023,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.024595625587472435,
      "loss": 0.2303,
      "num_input_tokens_seen": 13471392,
      "step": 63840
    },
    {
      "epoch": 7.023652365236524,
      "grad_norm": 0.00060272216796875,
      "learning_rate": 0.024594518702505168,
      "loss": 0.2308,
      "num_input_tokens_seen": 13472416,
      "step": 63845
    },
    {
      "epoch": 7.024202420242024,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.024593411729109827,
      "loss": 0.2319,
      "num_input_tokens_seen": 13473504,
      "step": 63850
    },
    {
      "epoch": 7.024752475247524,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.024592304667296613,
      "loss": 0.2309,
      "num_input_tokens_seen": 13474592,
      "step": 63855
    },
    {
      "epoch": 7.025302530253025,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.024591197517075728,
      "loss": 0.234,
      "num_input_tokens_seen": 13475616,
      "step": 63860
    },
    {
      "epoch": 7.025852585258526,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.024590090278457377,
      "loss": 0.2272,
      "num_input_tokens_seen": 13476672,
      "step": 63865
    },
    {
      "epoch": 7.026402640264027,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.02458898295145177,
      "loss": 0.2303,
      "num_input_tokens_seen": 13477728,
      "step": 63870
    },
    {
      "epoch": 7.026952695269527,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.024587875536069104,
      "loss": 0.2309,
      "num_input_tokens_seen": 13478784,
      "step": 63875
    },
    {
      "epoch": 7.027502750275027,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.02458676803231959,
      "loss": 0.2314,
      "num_input_tokens_seen": 13479744,
      "step": 63880
    },
    {
      "epoch": 7.028052805280528,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.024585660440213437,
      "loss": 0.2293,
      "num_input_tokens_seen": 13480800,
      "step": 63885
    },
    {
      "epoch": 7.028602860286028,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.024584552759760854,
      "loss": 0.2308,
      "num_input_tokens_seen": 13481888,
      "step": 63890
    },
    {
      "epoch": 7.0291529152915295,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.02458344499097204,
      "loss": 0.2309,
      "num_input_tokens_seen": 13482976,
      "step": 63895
    },
    {
      "epoch": 7.02970297029703,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.02458233713385722,
      "loss": 0.2303,
      "num_input_tokens_seen": 13484064,
      "step": 63900
    },
    {
      "epoch": 7.03025302530253,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.024581229188426593,
      "loss": 0.2319,
      "num_input_tokens_seen": 13485184,
      "step": 63905
    },
    {
      "epoch": 7.030803080308031,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.02458012115469037,
      "loss": 0.2308,
      "num_input_tokens_seen": 13486144,
      "step": 63910
    },
    {
      "epoch": 7.031353135313531,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.024579013032658775,
      "loss": 0.2309,
      "num_input_tokens_seen": 13487168,
      "step": 63915
    },
    {
      "epoch": 7.031903190319032,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 0.024577904822342008,
      "loss": 0.2319,
      "num_input_tokens_seen": 13488192,
      "step": 63920
    },
    {
      "epoch": 7.0324532453245325,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02457679652375029,
      "loss": 0.2324,
      "num_input_tokens_seen": 13489184,
      "step": 63925
    },
    {
      "epoch": 7.033003300330033,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02457568813689384,
      "loss": 0.2314,
      "num_input_tokens_seen": 13490208,
      "step": 63930
    },
    {
      "epoch": 7.033553355335534,
      "grad_norm": 0.0033111572265625,
      "learning_rate": 0.024574579661782857,
      "loss": 0.2329,
      "num_input_tokens_seen": 13491264,
      "step": 63935
    },
    {
      "epoch": 7.034103410341034,
      "grad_norm": 0.0004024505615234375,
      "learning_rate": 0.02457347109842757,
      "loss": 0.2319,
      "num_input_tokens_seen": 13492320,
      "step": 63940
    },
    {
      "epoch": 7.034653465346534,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.024572362446838195,
      "loss": 0.2314,
      "num_input_tokens_seen": 13493472,
      "step": 63945
    },
    {
      "epoch": 7.035203520352035,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.024571253707024954,
      "loss": 0.2314,
      "num_input_tokens_seen": 13494528,
      "step": 63950
    },
    {
      "epoch": 7.0357535753575355,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02457014487899805,
      "loss": 0.2314,
      "num_input_tokens_seen": 13495616,
      "step": 63955
    },
    {
      "epoch": 7.036303630363037,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.024569035962767723,
      "loss": 0.2303,
      "num_input_tokens_seen": 13496704,
      "step": 63960
    },
    {
      "epoch": 7.036853685368537,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02456792695834418,
      "loss": 0.2314,
      "num_input_tokens_seen": 13497728,
      "step": 63965
    },
    {
      "epoch": 7.037403740374037,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02456681786573764,
      "loss": 0.234,
      "num_input_tokens_seen": 13498816,
      "step": 63970
    },
    {
      "epoch": 7.037953795379538,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.02456570868495833,
      "loss": 0.2303,
      "num_input_tokens_seen": 13499840,
      "step": 63975
    },
    {
      "epoch": 7.038503850385038,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02456459941601648,
      "loss": 0.2314,
      "num_input_tokens_seen": 13500928,
      "step": 63980
    },
    {
      "epoch": 7.039053905390539,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02456349005892231,
      "loss": 0.2345,
      "num_input_tokens_seen": 13502048,
      "step": 63985
    },
    {
      "epoch": 7.03960396039604,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02456238061368603,
      "loss": 0.2298,
      "num_input_tokens_seen": 13503040,
      "step": 63990
    },
    {
      "epoch": 7.04015401540154,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.02456127108031788,
      "loss": 0.2293,
      "num_input_tokens_seen": 13504128,
      "step": 63995
    },
    {
      "epoch": 7.040704070407041,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.024560161458828087,
      "loss": 0.2314,
      "num_input_tokens_seen": 13505184,
      "step": 64000
    },
    {
      "epoch": 7.041254125412541,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02455905174922687,
      "loss": 0.2309,
      "num_input_tokens_seen": 13506272,
      "step": 64005
    },
    {
      "epoch": 7.041804180418042,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.02455794195152446,
      "loss": 0.2309,
      "num_input_tokens_seen": 13507328,
      "step": 64010
    },
    {
      "epoch": 7.042354235423542,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.024556832065731084,
      "loss": 0.2319,
      "num_input_tokens_seen": 13508384,
      "step": 64015
    },
    {
      "epoch": 7.042904290429043,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.024555722091856975,
      "loss": 0.2309,
      "num_input_tokens_seen": 13509440,
      "step": 64020
    },
    {
      "epoch": 7.043454345434544,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.024554612029912357,
      "loss": 0.2335,
      "num_input_tokens_seen": 13510528,
      "step": 64025
    },
    {
      "epoch": 7.044004400440044,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.024553501879907465,
      "loss": 0.2299,
      "num_input_tokens_seen": 13511584,
      "step": 64030
    },
    {
      "epoch": 7.044554455445544,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02455239164185254,
      "loss": 0.233,
      "num_input_tokens_seen": 13512640,
      "step": 64035
    },
    {
      "epoch": 7.045104510451045,
      "grad_norm": 0.0003509521484375,
      "learning_rate": 0.024551281315757796,
      "loss": 0.2314,
      "num_input_tokens_seen": 13513632,
      "step": 64040
    },
    {
      "epoch": 7.0456545654565454,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02455017090163348,
      "loss": 0.2293,
      "num_input_tokens_seen": 13514656,
      "step": 64045
    },
    {
      "epoch": 7.0462046204620465,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.024549060399489814,
      "loss": 0.2304,
      "num_input_tokens_seen": 13515680,
      "step": 64050
    },
    {
      "epoch": 7.046754675467547,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.024547949809337045,
      "loss": 0.2319,
      "num_input_tokens_seen": 13516736,
      "step": 64055
    },
    {
      "epoch": 7.047304730473047,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.024546839131185404,
      "loss": 0.2298,
      "num_input_tokens_seen": 13517824,
      "step": 64060
    },
    {
      "epoch": 7.047854785478548,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02454572836504513,
      "loss": 0.2293,
      "num_input_tokens_seen": 13518848,
      "step": 64065
    },
    {
      "epoch": 7.048404840484048,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.02454461751092645,
      "loss": 0.2341,
      "num_input_tokens_seen": 13519904,
      "step": 64070
    },
    {
      "epoch": 7.048954895489549,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02454350656883962,
      "loss": 0.2298,
      "num_input_tokens_seen": 13520960,
      "step": 64075
    },
    {
      "epoch": 7.0495049504950495,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.024542395538794864,
      "loss": 0.232,
      "num_input_tokens_seen": 13522016,
      "step": 64080
    },
    {
      "epoch": 7.05005500550055,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.02454128442080243,
      "loss": 0.232,
      "num_input_tokens_seen": 13523040,
      "step": 64085
    },
    {
      "epoch": 7.050605060506051,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.024540173214872552,
      "loss": 0.2336,
      "num_input_tokens_seen": 13524064,
      "step": 64090
    },
    {
      "epoch": 7.051155115511551,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.024539061921015484,
      "loss": 0.2346,
      "num_input_tokens_seen": 13525184,
      "step": 64095
    },
    {
      "epoch": 7.051705170517051,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02453795053924145,
      "loss": 0.2299,
      "num_input_tokens_seen": 13526208,
      "step": 64100
    },
    {
      "epoch": 7.052255225522552,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.024536839069560706,
      "loss": 0.2319,
      "num_input_tokens_seen": 13527264,
      "step": 64105
    },
    {
      "epoch": 7.052805280528053,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.024535727511983494,
      "loss": 0.2304,
      "num_input_tokens_seen": 13528256,
      "step": 64110
    },
    {
      "epoch": 7.053355335533554,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.024534615866520056,
      "loss": 0.2294,
      "num_input_tokens_seen": 13529312,
      "step": 64115
    },
    {
      "epoch": 7.053905390539054,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.024533504133180634,
      "loss": 0.232,
      "num_input_tokens_seen": 13530464,
      "step": 64120
    },
    {
      "epoch": 7.054455445544554,
      "grad_norm": 0.005859375,
      "learning_rate": 0.024532392311975487,
      "loss": 0.2294,
      "num_input_tokens_seen": 13531552,
      "step": 64125
    },
    {
      "epoch": 7.055005500550055,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.024531280402914853,
      "loss": 0.234,
      "num_input_tokens_seen": 13532640,
      "step": 64130
    },
    {
      "epoch": 7.055555555555555,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.024530168406008978,
      "loss": 0.2346,
      "num_input_tokens_seen": 13533664,
      "step": 64135
    },
    {
      "epoch": 7.0561056105610565,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.024529056321268113,
      "loss": 0.2299,
      "num_input_tokens_seen": 13534752,
      "step": 64140
    },
    {
      "epoch": 7.056655665566557,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.024527944148702508,
      "loss": 0.2252,
      "num_input_tokens_seen": 13535776,
      "step": 64145
    },
    {
      "epoch": 7.057205720572057,
      "grad_norm": 0.001953125,
      "learning_rate": 0.024526831888322418,
      "loss": 0.2294,
      "num_input_tokens_seen": 13536800,
      "step": 64150
    },
    {
      "epoch": 7.057755775577558,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.02452571954013809,
      "loss": 0.2336,
      "num_input_tokens_seen": 13537888,
      "step": 64155
    },
    {
      "epoch": 7.058305830583058,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.024524607104159775,
      "loss": 0.2336,
      "num_input_tokens_seen": 13538944,
      "step": 64160
    },
    {
      "epoch": 7.058855885588559,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.024523494580397723,
      "loss": 0.2331,
      "num_input_tokens_seen": 13539936,
      "step": 64165
    },
    {
      "epoch": 7.0594059405940595,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02452238196886219,
      "loss": 0.2356,
      "num_input_tokens_seen": 13541024,
      "step": 64170
    },
    {
      "epoch": 7.05995599559956,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.02452126926956344,
      "loss": 0.2294,
      "num_input_tokens_seen": 13542016,
      "step": 64175
    },
    {
      "epoch": 7.060506050605061,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.024520156482511712,
      "loss": 0.2278,
      "num_input_tokens_seen": 13543072,
      "step": 64180
    },
    {
      "epoch": 7.061056105610561,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.024519043607717273,
      "loss": 0.2346,
      "num_input_tokens_seen": 13544160,
      "step": 64185
    },
    {
      "epoch": 7.061606160616061,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.024517930645190377,
      "loss": 0.232,
      "num_input_tokens_seen": 13545216,
      "step": 64190
    },
    {
      "epoch": 7.062156215621562,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.02451681759494128,
      "loss": 0.2294,
      "num_input_tokens_seen": 13546208,
      "step": 64195
    },
    {
      "epoch": 7.0627062706270625,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.024515704456980248,
      "loss": 0.2314,
      "num_input_tokens_seen": 13547264,
      "step": 64200
    },
    {
      "epoch": 7.063256325632564,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.024514591231317526,
      "loss": 0.2304,
      "num_input_tokens_seen": 13548352,
      "step": 64205
    },
    {
      "epoch": 7.063806380638064,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.024513477917963388,
      "loss": 0.2335,
      "num_input_tokens_seen": 13549344,
      "step": 64210
    },
    {
      "epoch": 7.064356435643564,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.024512364516928082,
      "loss": 0.2319,
      "num_input_tokens_seen": 13550400,
      "step": 64215
    },
    {
      "epoch": 7.064906490649065,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.02451125102822188,
      "loss": 0.2325,
      "num_input_tokens_seen": 13551520,
      "step": 64220
    },
    {
      "epoch": 7.065456545654565,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.024510137451855044,
      "loss": 0.2309,
      "num_input_tokens_seen": 13552544,
      "step": 64225
    },
    {
      "epoch": 7.066006600660066,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.024509023787837834,
      "loss": 0.2345,
      "num_input_tokens_seen": 13553536,
      "step": 64230
    },
    {
      "epoch": 7.066556655665567,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.024507910036180514,
      "loss": 0.2293,
      "num_input_tokens_seen": 13554592,
      "step": 64235
    },
    {
      "epoch": 7.067106710671067,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.024506796196893346,
      "loss": 0.2314,
      "num_input_tokens_seen": 13555712,
      "step": 64240
    },
    {
      "epoch": 7.067656765676568,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.024505682269986607,
      "loss": 0.2309,
      "num_input_tokens_seen": 13556768,
      "step": 64245
    },
    {
      "epoch": 7.068206820682068,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.024504568255470552,
      "loss": 0.2283,
      "num_input_tokens_seen": 13557856,
      "step": 64250
    },
    {
      "epoch": 7.068756875687569,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.024503454153355447,
      "loss": 0.2309,
      "num_input_tokens_seen": 13558880,
      "step": 64255
    },
    {
      "epoch": 7.069306930693069,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.02450233996365157,
      "loss": 0.2309,
      "num_input_tokens_seen": 13560000,
      "step": 64260
    },
    {
      "epoch": 7.06985698569857,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.024501225686369185,
      "loss": 0.2309,
      "num_input_tokens_seen": 13561120,
      "step": 64265
    },
    {
      "epoch": 7.070407040704071,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.024500111321518563,
      "loss": 0.2319,
      "num_input_tokens_seen": 13562208,
      "step": 64270
    },
    {
      "epoch": 7.070957095709571,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.02449899686910997,
      "loss": 0.2309,
      "num_input_tokens_seen": 13563328,
      "step": 64275
    },
    {
      "epoch": 7.071507150715071,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.024497882329153682,
      "loss": 0.2319,
      "num_input_tokens_seen": 13564352,
      "step": 64280
    },
    {
      "epoch": 7.072057205720572,
      "grad_norm": 0.00057220458984375,
      "learning_rate": 0.024496767701659972,
      "loss": 0.2314,
      "num_input_tokens_seen": 13565376,
      "step": 64285
    },
    {
      "epoch": 7.072607260726072,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02449565298663911,
      "loss": 0.234,
      "num_input_tokens_seen": 13566432,
      "step": 64290
    },
    {
      "epoch": 7.0731573157315735,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.02449453818410137,
      "loss": 0.2304,
      "num_input_tokens_seen": 13567488,
      "step": 64295
    },
    {
      "epoch": 7.073707370737074,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.024493423294057032,
      "loss": 0.2293,
      "num_input_tokens_seen": 13568512,
      "step": 64300
    },
    {
      "epoch": 7.074257425742574,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.024492308316516363,
      "loss": 0.232,
      "num_input_tokens_seen": 13569536,
      "step": 64305
    },
    {
      "epoch": 7.074807480748075,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.02449119325148964,
      "loss": 0.233,
      "num_input_tokens_seen": 13570560,
      "step": 64310
    },
    {
      "epoch": 7.075357535753575,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.02449007809898715,
      "loss": 0.2298,
      "num_input_tokens_seen": 13571520,
      "step": 64315
    },
    {
      "epoch": 7.075907590759076,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.02448896285901916,
      "loss": 0.2314,
      "num_input_tokens_seen": 13572576,
      "step": 64320
    },
    {
      "epoch": 7.0764576457645765,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.02448784753159595,
      "loss": 0.2304,
      "num_input_tokens_seen": 13573600,
      "step": 64325
    },
    {
      "epoch": 7.077007700770077,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.02448673211672781,
      "loss": 0.2319,
      "num_input_tokens_seen": 13574656,
      "step": 64330
    },
    {
      "epoch": 7.077557755775578,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.024485616614425006,
      "loss": 0.2314,
      "num_input_tokens_seen": 13575648,
      "step": 64335
    },
    {
      "epoch": 7.078107810781078,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.024484501024697822,
      "loss": 0.2314,
      "num_input_tokens_seen": 13576800,
      "step": 64340
    },
    {
      "epoch": 7.078657865786579,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.024483385347556553,
      "loss": 0.2335,
      "num_input_tokens_seen": 13577856,
      "step": 64345
    },
    {
      "epoch": 7.079207920792079,
      "grad_norm": 0.009765625,
      "learning_rate": 0.024482269583011466,
      "loss": 0.2309,
      "num_input_tokens_seen": 13578880,
      "step": 64350
    },
    {
      "epoch": 7.0797579757975795,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.024481153731072854,
      "loss": 0.2319,
      "num_input_tokens_seen": 13579904,
      "step": 64355
    },
    {
      "epoch": 7.080308030803081,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.02448003779175099,
      "loss": 0.2303,
      "num_input_tokens_seen": 13581024,
      "step": 64360
    },
    {
      "epoch": 7.080858085808581,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.024478921765056172,
      "loss": 0.2309,
      "num_input_tokens_seen": 13582112,
      "step": 64365
    },
    {
      "epoch": 7.081408140814081,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02447780565099868,
      "loss": 0.2309,
      "num_input_tokens_seen": 13583232,
      "step": 64370
    },
    {
      "epoch": 7.081958195819582,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.0244766894495888,
      "loss": 0.2298,
      "num_input_tokens_seen": 13584224,
      "step": 64375
    },
    {
      "epoch": 7.082508250825082,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.024475573160836823,
      "loss": 0.2298,
      "num_input_tokens_seen": 13585312,
      "step": 64380
    },
    {
      "epoch": 7.083058305830583,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.024474456784753032,
      "loss": 0.2319,
      "num_input_tokens_seen": 13586368,
      "step": 64385
    },
    {
      "epoch": 7.083608360836084,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.024473340321347726,
      "loss": 0.2304,
      "num_input_tokens_seen": 13587456,
      "step": 64390
    },
    {
      "epoch": 7.084158415841584,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.024472223770631178,
      "loss": 0.2304,
      "num_input_tokens_seen": 13588512,
      "step": 64395
    },
    {
      "epoch": 7.084708470847085,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.024471107132613697,
      "loss": 0.2372,
      "num_input_tokens_seen": 13589536,
      "step": 64400
    },
    {
      "epoch": 7.085258525852585,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02446999040730556,
      "loss": 0.232,
      "num_input_tokens_seen": 13590624,
      "step": 64405
    },
    {
      "epoch": 7.085808580858086,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.02446887359471707,
      "loss": 0.2288,
      "num_input_tokens_seen": 13591744,
      "step": 64410
    },
    {
      "epoch": 7.086358635863586,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.024467756694858515,
      "loss": 0.2314,
      "num_input_tokens_seen": 13592768,
      "step": 64415
    },
    {
      "epoch": 7.086908690869087,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.024466639707740184,
      "loss": 0.2315,
      "num_input_tokens_seen": 13593824,
      "step": 64420
    },
    {
      "epoch": 7.087458745874588,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.024465522633372382,
      "loss": 0.2298,
      "num_input_tokens_seen": 13594880,
      "step": 64425
    },
    {
      "epoch": 7.088008800880088,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.0244644054717654,
      "loss": 0.2325,
      "num_input_tokens_seen": 13595968,
      "step": 64430
    },
    {
      "epoch": 7.088558855885589,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.024463288222929532,
      "loss": 0.2314,
      "num_input_tokens_seen": 13597024,
      "step": 64435
    },
    {
      "epoch": 7.089108910891089,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.024462170886875082,
      "loss": 0.2304,
      "num_input_tokens_seen": 13598048,
      "step": 64440
    },
    {
      "epoch": 7.089658965896589,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.024461053463612343,
      "loss": 0.2314,
      "num_input_tokens_seen": 13599168,
      "step": 64445
    },
    {
      "epoch": 7.0902090209020905,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.024459935953151612,
      "loss": 0.2288,
      "num_input_tokens_seen": 13600192,
      "step": 64450
    },
    {
      "epoch": 7.090759075907591,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.024458818355503186,
      "loss": 0.2279,
      "num_input_tokens_seen": 13601216,
      "step": 64455
    },
    {
      "epoch": 7.091309130913091,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.024457700670677377,
      "loss": 0.2309,
      "num_input_tokens_seen": 13602208,
      "step": 64460
    },
    {
      "epoch": 7.091859185918592,
      "grad_norm": 0.002838134765625,
      "learning_rate": 0.02445658289868448,
      "loss": 0.2326,
      "num_input_tokens_seen": 13603264,
      "step": 64465
    },
    {
      "epoch": 7.092409240924092,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02445546503953479,
      "loss": 0.2294,
      "num_input_tokens_seen": 13604384,
      "step": 64470
    },
    {
      "epoch": 7.092959295929593,
      "grad_norm": 0.000713348388671875,
      "learning_rate": 0.02445434709323862,
      "loss": 0.227,
      "num_input_tokens_seen": 13605376,
      "step": 64475
    },
    {
      "epoch": 7.0935093509350935,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.024453229059806268,
      "loss": 0.2295,
      "num_input_tokens_seen": 13606464,
      "step": 64480
    },
    {
      "epoch": 7.094059405940594,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02445211093924804,
      "loss": 0.2193,
      "num_input_tokens_seen": 13607488,
      "step": 64485
    },
    {
      "epoch": 7.094609460946095,
      "grad_norm": 0.0034027099609375,
      "learning_rate": 0.02445099273157424,
      "loss": 0.2416,
      "num_input_tokens_seen": 13608512,
      "step": 64490
    },
    {
      "epoch": 7.095159515951595,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.024449874436795176,
      "loss": 0.2395,
      "num_input_tokens_seen": 13609568,
      "step": 64495
    },
    {
      "epoch": 7.095709570957096,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.024448756054921153,
      "loss": 0.2332,
      "num_input_tokens_seen": 13610560,
      "step": 64500
    },
    {
      "epoch": 7.096259625962596,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.02444763758596248,
      "loss": 0.2279,
      "num_input_tokens_seen": 13611648,
      "step": 64505
    },
    {
      "epoch": 7.0968096809680965,
      "grad_norm": 0.003265380859375,
      "learning_rate": 0.024446519029929465,
      "loss": 0.2346,
      "num_input_tokens_seen": 13612704,
      "step": 64510
    },
    {
      "epoch": 7.097359735973598,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.02444540038683242,
      "loss": 0.2319,
      "num_input_tokens_seen": 13613760,
      "step": 64515
    },
    {
      "epoch": 7.097909790979098,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02444428165668165,
      "loss": 0.2299,
      "num_input_tokens_seen": 13614816,
      "step": 64520
    },
    {
      "epoch": 7.098459845984599,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.024443162839487467,
      "loss": 0.2366,
      "num_input_tokens_seen": 13615808,
      "step": 64525
    },
    {
      "epoch": 7.099009900990099,
      "grad_norm": 0.01190185546875,
      "learning_rate": 0.024442043935260182,
      "loss": 0.2371,
      "num_input_tokens_seen": 13616896,
      "step": 64530
    },
    {
      "epoch": 7.099559955995599,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.02444092494401011,
      "loss": 0.2344,
      "num_input_tokens_seen": 13617920,
      "step": 64535
    },
    {
      "epoch": 7.1001100110011,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.024439805865747562,
      "loss": 0.2322,
      "num_input_tokens_seen": 13618944,
      "step": 64540
    },
    {
      "epoch": 7.100660066006601,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.024438686700482853,
      "loss": 0.2321,
      "num_input_tokens_seen": 13619968,
      "step": 64545
    },
    {
      "epoch": 7.101210121012101,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.0244375674482263,
      "loss": 0.2325,
      "num_input_tokens_seen": 13620960,
      "step": 64550
    },
    {
      "epoch": 7.101760176017602,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.024436448108988214,
      "loss": 0.2341,
      "num_input_tokens_seen": 13622048,
      "step": 64555
    },
    {
      "epoch": 7.102310231023102,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.024435328682778916,
      "loss": 0.2315,
      "num_input_tokens_seen": 13623040,
      "step": 64560
    },
    {
      "epoch": 7.102860286028603,
      "grad_norm": 0.001953125,
      "learning_rate": 0.024434209169608718,
      "loss": 0.233,
      "num_input_tokens_seen": 13624064,
      "step": 64565
    },
    {
      "epoch": 7.103410341034103,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.024433089569487944,
      "loss": 0.2304,
      "num_input_tokens_seen": 13625152,
      "step": 64570
    },
    {
      "epoch": 7.103960396039604,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.02443196988242691,
      "loss": 0.2273,
      "num_input_tokens_seen": 13626240,
      "step": 64575
    },
    {
      "epoch": 7.104510451045105,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.024430850108435934,
      "loss": 0.231,
      "num_input_tokens_seen": 13627264,
      "step": 64580
    },
    {
      "epoch": 7.105060506050605,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.024429730247525337,
      "loss": 0.2279,
      "num_input_tokens_seen": 13628288,
      "step": 64585
    },
    {
      "epoch": 7.105610561056106,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.024428610299705444,
      "loss": 0.2358,
      "num_input_tokens_seen": 13629280,
      "step": 64590
    },
    {
      "epoch": 7.106160616061606,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02442749026498657,
      "loss": 0.2331,
      "num_input_tokens_seen": 13630368,
      "step": 64595
    },
    {
      "epoch": 7.106710671067106,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.024426370143379048,
      "loss": 0.2295,
      "num_input_tokens_seen": 13631424,
      "step": 64600
    },
    {
      "epoch": 7.1072607260726075,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02442524993489319,
      "loss": 0.2346,
      "num_input_tokens_seen": 13632480,
      "step": 64605
    },
    {
      "epoch": 7.107810781078108,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.02442412963953933,
      "loss": 0.2315,
      "num_input_tokens_seen": 13633536,
      "step": 64610
    },
    {
      "epoch": 7.108360836083609,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.024423009257327783,
      "loss": 0.2373,
      "num_input_tokens_seen": 13634496,
      "step": 64615
    },
    {
      "epoch": 7.108910891089109,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02442188878826889,
      "loss": 0.2361,
      "num_input_tokens_seen": 13635584,
      "step": 64620
    },
    {
      "epoch": 7.109460946094609,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.024420768232372964,
      "loss": 0.2304,
      "num_input_tokens_seen": 13636544,
      "step": 64625
    },
    {
      "epoch": 7.11001100110011,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.024419647589650338,
      "loss": 0.2324,
      "num_input_tokens_seen": 13637600,
      "step": 64630
    },
    {
      "epoch": 7.1105610561056105,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02441852686011134,
      "loss": 0.2335,
      "num_input_tokens_seen": 13638656,
      "step": 64635
    },
    {
      "epoch": 7.111111111111111,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.024417406043766302,
      "loss": 0.2293,
      "num_input_tokens_seen": 13639680,
      "step": 64640
    },
    {
      "epoch": 7.111661166116612,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.02441628514062555,
      "loss": 0.2308,
      "num_input_tokens_seen": 13640736,
      "step": 64645
    },
    {
      "epoch": 7.112211221122112,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.024415164150699414,
      "loss": 0.234,
      "num_input_tokens_seen": 13641824,
      "step": 64650
    },
    {
      "epoch": 7.112761276127613,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02441404307399823,
      "loss": 0.2319,
      "num_input_tokens_seen": 13642880,
      "step": 64655
    },
    {
      "epoch": 7.113311331133113,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.02441292191053233,
      "loss": 0.2319,
      "num_input_tokens_seen": 13643904,
      "step": 64660
    },
    {
      "epoch": 7.1138613861386135,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.024411800660312042,
      "loss": 0.2303,
      "num_input_tokens_seen": 13644896,
      "step": 64665
    },
    {
      "epoch": 7.114411441144115,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.024410679323347704,
      "loss": 0.2298,
      "num_input_tokens_seen": 13645888,
      "step": 64670
    },
    {
      "epoch": 7.114961496149615,
      "grad_norm": 0.002899169921875,
      "learning_rate": 0.024409557899649652,
      "loss": 0.2298,
      "num_input_tokens_seen": 13646976,
      "step": 64675
    },
    {
      "epoch": 7.115511551155116,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02440843638922822,
      "loss": 0.2308,
      "num_input_tokens_seen": 13648064,
      "step": 64680
    },
    {
      "epoch": 7.116061606160616,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.024407314792093744,
      "loss": 0.2303,
      "num_input_tokens_seen": 13649152,
      "step": 64685
    },
    {
      "epoch": 7.116611661166116,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.02440619310825656,
      "loss": 0.2288,
      "num_input_tokens_seen": 13650208,
      "step": 64690
    },
    {
      "epoch": 7.117161716171617,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.024405071337727006,
      "loss": 0.234,
      "num_input_tokens_seen": 13651264,
      "step": 64695
    },
    {
      "epoch": 7.117711771177118,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02440394948051543,
      "loss": 0.232,
      "num_input_tokens_seen": 13652320,
      "step": 64700
    },
    {
      "epoch": 7.118261826182618,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02440282753663216,
      "loss": 0.2319,
      "num_input_tokens_seen": 13653376,
      "step": 64705
    },
    {
      "epoch": 7.118811881188119,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.02440170550608754,
      "loss": 0.2314,
      "num_input_tokens_seen": 13654432,
      "step": 64710
    },
    {
      "epoch": 7.119361936193619,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.02440058338889191,
      "loss": 0.2314,
      "num_input_tokens_seen": 13655488,
      "step": 64715
    },
    {
      "epoch": 7.11991199119912,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.024399461185055614,
      "loss": 0.2314,
      "num_input_tokens_seen": 13656576,
      "step": 64720
    },
    {
      "epoch": 7.12046204620462,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.024398338894589002,
      "loss": 0.2303,
      "num_input_tokens_seen": 13657728,
      "step": 64725
    },
    {
      "epoch": 7.121012101210121,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.0243972165175024,
      "loss": 0.2324,
      "num_input_tokens_seen": 13658752,
      "step": 64730
    },
    {
      "epoch": 7.121562156215622,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.024396094053806168,
      "loss": 0.2319,
      "num_input_tokens_seen": 13659872,
      "step": 64735
    },
    {
      "epoch": 7.122112211221122,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.02439497150351064,
      "loss": 0.2303,
      "num_input_tokens_seen": 13660960,
      "step": 64740
    },
    {
      "epoch": 7.122662266226623,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.024393848866626173,
      "loss": 0.2298,
      "num_input_tokens_seen": 13662048,
      "step": 64745
    },
    {
      "epoch": 7.123212321232123,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.024392726143163113,
      "loss": 0.234,
      "num_input_tokens_seen": 13663136,
      "step": 64750
    },
    {
      "epoch": 7.123762376237623,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.024391603333131798,
      "loss": 0.2288,
      "num_input_tokens_seen": 13664128,
      "step": 64755
    },
    {
      "epoch": 7.1243124312431245,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.02439048043654258,
      "loss": 0.2289,
      "num_input_tokens_seen": 13665152,
      "step": 64760
    },
    {
      "epoch": 7.124862486248625,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.024389357453405813,
      "loss": 0.2335,
      "num_input_tokens_seen": 13666176,
      "step": 64765
    },
    {
      "epoch": 7.125412541254126,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.024388234383731843,
      "loss": 0.2336,
      "num_input_tokens_seen": 13667200,
      "step": 64770
    },
    {
      "epoch": 7.125962596259626,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02438711122753102,
      "loss": 0.2299,
      "num_input_tokens_seen": 13668256,
      "step": 64775
    },
    {
      "epoch": 7.126512651265126,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.024385987984813697,
      "loss": 0.2325,
      "num_input_tokens_seen": 13669280,
      "step": 64780
    },
    {
      "epoch": 7.127062706270627,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.024384864655590227,
      "loss": 0.231,
      "num_input_tokens_seen": 13670304,
      "step": 64785
    },
    {
      "epoch": 7.1276127612761275,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.024383741239870964,
      "loss": 0.232,
      "num_input_tokens_seen": 13671296,
      "step": 64790
    },
    {
      "epoch": 7.128162816281628,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.024382617737666264,
      "loss": 0.2335,
      "num_input_tokens_seen": 13672352,
      "step": 64795
    },
    {
      "epoch": 7.128712871287129,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.024381494148986476,
      "loss": 0.233,
      "num_input_tokens_seen": 13673376,
      "step": 64800
    },
    {
      "epoch": 7.129262926292629,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.024380370473841957,
      "loss": 0.2304,
      "num_input_tokens_seen": 13674400,
      "step": 64805
    },
    {
      "epoch": 7.12981298129813,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02437924671224306,
      "loss": 0.2314,
      "num_input_tokens_seen": 13675488,
      "step": 64810
    },
    {
      "epoch": 7.13036303630363,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.02437812286420015,
      "loss": 0.2309,
      "num_input_tokens_seen": 13676512,
      "step": 64815
    },
    {
      "epoch": 7.1309130913091305,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.024376998929723583,
      "loss": 0.2314,
      "num_input_tokens_seen": 13677536,
      "step": 64820
    },
    {
      "epoch": 7.131463146314632,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02437587490882372,
      "loss": 0.2298,
      "num_input_tokens_seen": 13678592,
      "step": 64825
    },
    {
      "epoch": 7.132013201320132,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.02437475080151091,
      "loss": 0.2299,
      "num_input_tokens_seen": 13679584,
      "step": 64830
    },
    {
      "epoch": 7.132563256325633,
      "grad_norm": 0.001953125,
      "learning_rate": 0.024373626607795525,
      "loss": 0.2309,
      "num_input_tokens_seen": 13680672,
      "step": 64835
    },
    {
      "epoch": 7.133113311331133,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02437250232768792,
      "loss": 0.2304,
      "num_input_tokens_seen": 13681792,
      "step": 64840
    },
    {
      "epoch": 7.133663366336633,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.024371377961198457,
      "loss": 0.2309,
      "num_input_tokens_seen": 13682880,
      "step": 64845
    },
    {
      "epoch": 7.134213421342134,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.024370253508337502,
      "loss": 0.2341,
      "num_input_tokens_seen": 13683904,
      "step": 64850
    },
    {
      "epoch": 7.134763476347635,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.024369128969115413,
      "loss": 0.2319,
      "num_input_tokens_seen": 13684992,
      "step": 64855
    },
    {
      "epoch": 7.135313531353136,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.024368004343542565,
      "loss": 0.2335,
      "num_input_tokens_seen": 13686016,
      "step": 64860
    },
    {
      "epoch": 7.135863586358636,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.024366879631629312,
      "loss": 0.2319,
      "num_input_tokens_seen": 13687104,
      "step": 64865
    },
    {
      "epoch": 7.136413641364136,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.024365754833386023,
      "loss": 0.2329,
      "num_input_tokens_seen": 13688256,
      "step": 64870
    },
    {
      "epoch": 7.136963696369637,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.024364629948823067,
      "loss": 0.2309,
      "num_input_tokens_seen": 13689312,
      "step": 64875
    },
    {
      "epoch": 7.137513751375137,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02436350497795081,
      "loss": 0.2314,
      "num_input_tokens_seen": 13690304,
      "step": 64880
    },
    {
      "epoch": 7.138063806380638,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.024362379920779623,
      "loss": 0.2309,
      "num_input_tokens_seen": 13691360,
      "step": 64885
    },
    {
      "epoch": 7.138613861386139,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02436125477731987,
      "loss": 0.2324,
      "num_input_tokens_seen": 13692416,
      "step": 64890
    },
    {
      "epoch": 7.139163916391639,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.02436012954758193,
      "loss": 0.2308,
      "num_input_tokens_seen": 13693504,
      "step": 64895
    },
    {
      "epoch": 7.13971397139714,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.024359004231576158,
      "loss": 0.2319,
      "num_input_tokens_seen": 13694592,
      "step": 64900
    },
    {
      "epoch": 7.14026402640264,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.024357878829312937,
      "loss": 0.2324,
      "num_input_tokens_seen": 13695680,
      "step": 64905
    },
    {
      "epoch": 7.1408140814081404,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02435675334080264,
      "loss": 0.2319,
      "num_input_tokens_seen": 13696800,
      "step": 64910
    },
    {
      "epoch": 7.1413641364136415,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02435562776605564,
      "loss": 0.2304,
      "num_input_tokens_seen": 13697824,
      "step": 64915
    },
    {
      "epoch": 7.141914191419142,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.024354502105082304,
      "loss": 0.2288,
      "num_input_tokens_seen": 13698944,
      "step": 64920
    },
    {
      "epoch": 7.142464246424643,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.024353376357893013,
      "loss": 0.2309,
      "num_input_tokens_seen": 13700000,
      "step": 64925
    },
    {
      "epoch": 7.143014301430143,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.024352250524498135,
      "loss": 0.2309,
      "num_input_tokens_seen": 13701120,
      "step": 64930
    },
    {
      "epoch": 7.143564356435643,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.024351124604908053,
      "loss": 0.2294,
      "num_input_tokens_seen": 13702144,
      "step": 64935
    },
    {
      "epoch": 7.144114411441144,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.024349998599133147,
      "loss": 0.2325,
      "num_input_tokens_seen": 13703200,
      "step": 64940
    },
    {
      "epoch": 7.1446644664466445,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.024348872507183792,
      "loss": 0.2289,
      "num_input_tokens_seen": 13704320,
      "step": 64945
    },
    {
      "epoch": 7.145214521452146,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02434774632907036,
      "loss": 0.2351,
      "num_input_tokens_seen": 13705408,
      "step": 64950
    },
    {
      "epoch": 7.145764576457646,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.024346620064803235,
      "loss": 0.2315,
      "num_input_tokens_seen": 13706496,
      "step": 64955
    },
    {
      "epoch": 7.146314631463146,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.024345493714392805,
      "loss": 0.232,
      "num_input_tokens_seen": 13707584,
      "step": 64960
    },
    {
      "epoch": 7.146864686468647,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.024344367277849437,
      "loss": 0.2325,
      "num_input_tokens_seen": 13708672,
      "step": 64965
    },
    {
      "epoch": 7.147414741474147,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02434324075518352,
      "loss": 0.233,
      "num_input_tokens_seen": 13709760,
      "step": 64970
    },
    {
      "epoch": 7.1479647964796476,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.024342114146405438,
      "loss": 0.2351,
      "num_input_tokens_seen": 13710816,
      "step": 64975
    },
    {
      "epoch": 7.148514851485149,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.02434098745152557,
      "loss": 0.233,
      "num_input_tokens_seen": 13711904,
      "step": 64980
    },
    {
      "epoch": 7.149064906490649,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.024339860670554308,
      "loss": 0.2308,
      "num_input_tokens_seen": 13712960,
      "step": 64985
    },
    {
      "epoch": 7.14961496149615,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02433873380350203,
      "loss": 0.2298,
      "num_input_tokens_seen": 13713984,
      "step": 64990
    },
    {
      "epoch": 7.15016501650165,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02433760685037912,
      "loss": 0.2308,
      "num_input_tokens_seen": 13715104,
      "step": 64995
    },
    {
      "epoch": 7.15071507150715,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.024336479811195973,
      "loss": 0.2319,
      "num_input_tokens_seen": 13716096,
      "step": 65000
    },
    {
      "epoch": 7.1512651265126514,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.024335352685962968,
      "loss": 0.2308,
      "num_input_tokens_seen": 13717088,
      "step": 65005
    },
    {
      "epoch": 7.151815181518152,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.0243342254746905,
      "loss": 0.2308,
      "num_input_tokens_seen": 13718144,
      "step": 65010
    },
    {
      "epoch": 7.152365236523653,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02433309817738895,
      "loss": 0.2319,
      "num_input_tokens_seen": 13719232,
      "step": 65015
    },
    {
      "epoch": 7.152915291529153,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02433197079406872,
      "loss": 0.2324,
      "num_input_tokens_seen": 13720288,
      "step": 65020
    },
    {
      "epoch": 7.153465346534653,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.024330843324740184,
      "loss": 0.2334,
      "num_input_tokens_seen": 13721344,
      "step": 65025
    },
    {
      "epoch": 7.154015401540154,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02432971576941375,
      "loss": 0.2314,
      "num_input_tokens_seen": 13722400,
      "step": 65030
    },
    {
      "epoch": 7.1545654565456545,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.0243285881280998,
      "loss": 0.2325,
      "num_input_tokens_seen": 13723456,
      "step": 65035
    },
    {
      "epoch": 7.1551155115511555,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.02432746040080873,
      "loss": 0.2303,
      "num_input_tokens_seen": 13724544,
      "step": 65040
    },
    {
      "epoch": 7.155665566556656,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02432633258755093,
      "loss": 0.2345,
      "num_input_tokens_seen": 13725600,
      "step": 65045
    },
    {
      "epoch": 7.156215621562156,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.024325204688336804,
      "loss": 0.2308,
      "num_input_tokens_seen": 13726624,
      "step": 65050
    },
    {
      "epoch": 7.156765676567657,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02432407670317674,
      "loss": 0.2309,
      "num_input_tokens_seen": 13727680,
      "step": 65055
    },
    {
      "epoch": 7.157315731573157,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02432294863208113,
      "loss": 0.2319,
      "num_input_tokens_seen": 13728672,
      "step": 65060
    },
    {
      "epoch": 7.1578657865786575,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.024321820475060377,
      "loss": 0.232,
      "num_input_tokens_seen": 13729696,
      "step": 65065
    },
    {
      "epoch": 7.158415841584159,
      "grad_norm": 0.000514984130859375,
      "learning_rate": 0.02432069223212488,
      "loss": 0.2319,
      "num_input_tokens_seen": 13730720,
      "step": 65070
    },
    {
      "epoch": 7.158965896589659,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.024319563903285036,
      "loss": 0.2304,
      "num_input_tokens_seen": 13731840,
      "step": 65075
    },
    {
      "epoch": 7.15951595159516,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.024318435488551244,
      "loss": 0.2309,
      "num_input_tokens_seen": 13732992,
      "step": 65080
    },
    {
      "epoch": 7.16006600660066,
      "grad_norm": 0.000518798828125,
      "learning_rate": 0.024317306987933898,
      "loss": 0.2314,
      "num_input_tokens_seen": 13734048,
      "step": 65085
    },
    {
      "epoch": 7.16061606160616,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.02431617840144341,
      "loss": 0.2335,
      "num_input_tokens_seen": 13735168,
      "step": 65090
    },
    {
      "epoch": 7.161166116611661,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.024315049729090172,
      "loss": 0.2319,
      "num_input_tokens_seen": 13736224,
      "step": 65095
    },
    {
      "epoch": 7.161716171617162,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.024313920970884594,
      "loss": 0.2309,
      "num_input_tokens_seen": 13737248,
      "step": 65100
    },
    {
      "epoch": 7.162266226622663,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.024312792126837075,
      "loss": 0.2319,
      "num_input_tokens_seen": 13738272,
      "step": 65105
    },
    {
      "epoch": 7.162816281628163,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.02431166319695802,
      "loss": 0.2319,
      "num_input_tokens_seen": 13739296,
      "step": 65110
    },
    {
      "epoch": 7.163366336633663,
      "grad_norm": 0.009765625,
      "learning_rate": 0.024310534181257833,
      "loss": 0.2324,
      "num_input_tokens_seen": 13740352,
      "step": 65115
    },
    {
      "epoch": 7.163916391639164,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.02430940507974692,
      "loss": 0.2308,
      "num_input_tokens_seen": 13741408,
      "step": 65120
    },
    {
      "epoch": 7.164466446644664,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.024308275892435685,
      "loss": 0.2309,
      "num_input_tokens_seen": 13742464,
      "step": 65125
    },
    {
      "epoch": 7.165016501650165,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.02430714661933454,
      "loss": 0.2277,
      "num_input_tokens_seen": 13743520,
      "step": 65130
    },
    {
      "epoch": 7.165566556655666,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.024306017260453894,
      "loss": 0.2314,
      "num_input_tokens_seen": 13744512,
      "step": 65135
    },
    {
      "epoch": 7.166116611661166,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.024304887815804144,
      "loss": 0.2293,
      "num_input_tokens_seen": 13745568,
      "step": 65140
    },
    {
      "epoch": 7.166666666666667,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.024303758285395712,
      "loss": 0.2309,
      "num_input_tokens_seen": 13746592,
      "step": 65145
    },
    {
      "epoch": 7.167216721672167,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.024302628669239008,
      "loss": 0.2309,
      "num_input_tokens_seen": 13747616,
      "step": 65150
    },
    {
      "epoch": 7.167766776677667,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.024301498967344435,
      "loss": 0.2335,
      "num_input_tokens_seen": 13748704,
      "step": 65155
    },
    {
      "epoch": 7.1683168316831685,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02430036917972241,
      "loss": 0.2324,
      "num_input_tokens_seen": 13749760,
      "step": 65160
    },
    {
      "epoch": 7.168866886688669,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.024299239306383343,
      "loss": 0.2309,
      "num_input_tokens_seen": 13750880,
      "step": 65165
    },
    {
      "epoch": 7.16941694169417,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.024298109347337653,
      "loss": 0.2314,
      "num_input_tokens_seen": 13751936,
      "step": 65170
    },
    {
      "epoch": 7.16996699669967,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02429697930259575,
      "loss": 0.2319,
      "num_input_tokens_seen": 13753056,
      "step": 65175
    },
    {
      "epoch": 7.17051705170517,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.024295849172168052,
      "loss": 0.2273,
      "num_input_tokens_seen": 13754112,
      "step": 65180
    },
    {
      "epoch": 7.171067106710671,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02429471895606497,
      "loss": 0.2324,
      "num_input_tokens_seen": 13755168,
      "step": 65185
    },
    {
      "epoch": 7.1716171617161715,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.02429358865429692,
      "loss": 0.2298,
      "num_input_tokens_seen": 13756192,
      "step": 65190
    },
    {
      "epoch": 7.172167216721673,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02429245826687433,
      "loss": 0.2288,
      "num_input_tokens_seen": 13757216,
      "step": 65195
    },
    {
      "epoch": 7.172717271727173,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.024291327793807608,
      "loss": 0.2351,
      "num_input_tokens_seen": 13758272,
      "step": 65200
    },
    {
      "epoch": 7.173267326732673,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.024290197235107176,
      "loss": 0.2315,
      "num_input_tokens_seen": 13759328,
      "step": 65205
    },
    {
      "epoch": 7.173817381738174,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02428906659078345,
      "loss": 0.2309,
      "num_input_tokens_seen": 13760448,
      "step": 65210
    },
    {
      "epoch": 7.174367436743674,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02428793586084686,
      "loss": 0.2288,
      "num_input_tokens_seen": 13761504,
      "step": 65215
    },
    {
      "epoch": 7.174917491749175,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.024286805045307818,
      "loss": 0.2299,
      "num_input_tokens_seen": 13762560,
      "step": 65220
    },
    {
      "epoch": 7.175467546754676,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.024285674144176757,
      "loss": 0.2315,
      "num_input_tokens_seen": 13763648,
      "step": 65225
    },
    {
      "epoch": 7.176017601760176,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.02428454315746409,
      "loss": 0.2284,
      "num_input_tokens_seen": 13764736,
      "step": 65230
    },
    {
      "epoch": 7.176567656765677,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.024283412085180242,
      "loss": 0.231,
      "num_input_tokens_seen": 13765792,
      "step": 65235
    },
    {
      "epoch": 7.177117711771177,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.024282280927335635,
      "loss": 0.2336,
      "num_input_tokens_seen": 13766784,
      "step": 65240
    },
    {
      "epoch": 7.177667766776677,
      "grad_norm": 0.000701904296875,
      "learning_rate": 0.024281149683940708,
      "loss": 0.2321,
      "num_input_tokens_seen": 13767776,
      "step": 65245
    },
    {
      "epoch": 7.178217821782178,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.024280018355005876,
      "loss": 0.2336,
      "num_input_tokens_seen": 13768736,
      "step": 65250
    },
    {
      "epoch": 7.178767876787679,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.024278886940541564,
      "loss": 0.2331,
      "num_input_tokens_seen": 13769792,
      "step": 65255
    },
    {
      "epoch": 7.17931793179318,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.024277755440558206,
      "loss": 0.2321,
      "num_input_tokens_seen": 13770880,
      "step": 65260
    },
    {
      "epoch": 7.17986798679868,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.024276623855066223,
      "loss": 0.2299,
      "num_input_tokens_seen": 13772000,
      "step": 65265
    },
    {
      "epoch": 7.18041804180418,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02427549218407606,
      "loss": 0.2315,
      "num_input_tokens_seen": 13773024,
      "step": 65270
    },
    {
      "epoch": 7.180968096809681,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.024274360427598125,
      "loss": 0.2304,
      "num_input_tokens_seen": 13774112,
      "step": 65275
    },
    {
      "epoch": 7.181518151815181,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.024273228585642865,
      "loss": 0.23,
      "num_input_tokens_seen": 13775168,
      "step": 65280
    },
    {
      "epoch": 7.1820682068206825,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.024272096658220704,
      "loss": 0.2331,
      "num_input_tokens_seen": 13776224,
      "step": 65285
    },
    {
      "epoch": 7.182618261826183,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.024270964645342084,
      "loss": 0.2315,
      "num_input_tokens_seen": 13777312,
      "step": 65290
    },
    {
      "epoch": 7.183168316831683,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02426983254701743,
      "loss": 0.2326,
      "num_input_tokens_seen": 13778368,
      "step": 65295
    },
    {
      "epoch": 7.183718371837184,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02426870036325717,
      "loss": 0.2315,
      "num_input_tokens_seen": 13779424,
      "step": 65300
    },
    {
      "epoch": 7.184268426842684,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.024267568094071754,
      "loss": 0.23,
      "num_input_tokens_seen": 13780512,
      "step": 65305
    },
    {
      "epoch": 7.184818481848184,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.024266435739471605,
      "loss": 0.229,
      "num_input_tokens_seen": 13781600,
      "step": 65310
    },
    {
      "epoch": 7.1853685368536855,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.024265303299467165,
      "loss": 0.2326,
      "num_input_tokens_seen": 13782656,
      "step": 65315
    },
    {
      "epoch": 7.185918591859186,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.024264170774068877,
      "loss": 0.2332,
      "num_input_tokens_seen": 13783744,
      "step": 65320
    },
    {
      "epoch": 7.186468646864687,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02426303816328717,
      "loss": 0.229,
      "num_input_tokens_seen": 13784800,
      "step": 65325
    },
    {
      "epoch": 7.187018701870187,
      "grad_norm": 0.00457763671875,
      "learning_rate": 0.02426190546713248,
      "loss": 0.2342,
      "num_input_tokens_seen": 13785824,
      "step": 65330
    },
    {
      "epoch": 7.187568756875687,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02426077268561525,
      "loss": 0.2336,
      "num_input_tokens_seen": 13786912,
      "step": 65335
    },
    {
      "epoch": 7.188118811881188,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.024259639818745928,
      "loss": 0.232,
      "num_input_tokens_seen": 13787904,
      "step": 65340
    },
    {
      "epoch": 7.1886688668866885,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.02425850686653495,
      "loss": 0.2284,
      "num_input_tokens_seen": 13788928,
      "step": 65345
    },
    {
      "epoch": 7.18921892189219,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02425737382899275,
      "loss": 0.2331,
      "num_input_tokens_seen": 13789952,
      "step": 65350
    },
    {
      "epoch": 7.18976897689769,
      "grad_norm": 0.004486083984375,
      "learning_rate": 0.02425624070612978,
      "loss": 0.2315,
      "num_input_tokens_seen": 13791040,
      "step": 65355
    },
    {
      "epoch": 7.19031903190319,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.024255107497956483,
      "loss": 0.2295,
      "num_input_tokens_seen": 13792064,
      "step": 65360
    },
    {
      "epoch": 7.190869086908691,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.024253974204483303,
      "loss": 0.2284,
      "num_input_tokens_seen": 13793024,
      "step": 65365
    },
    {
      "epoch": 7.191419141914191,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.024252840825720677,
      "loss": 0.2289,
      "num_input_tokens_seen": 13794048,
      "step": 65370
    },
    {
      "epoch": 7.191969196919692,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.024251707361679063,
      "loss": 0.233,
      "num_input_tokens_seen": 13795072,
      "step": 65375
    },
    {
      "epoch": 7.192519251925193,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.0242505738123689,
      "loss": 0.2309,
      "num_input_tokens_seen": 13796128,
      "step": 65380
    },
    {
      "epoch": 7.193069306930693,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.024249440177800636,
      "loss": 0.2315,
      "num_input_tokens_seen": 13797216,
      "step": 65385
    },
    {
      "epoch": 7.193619361936194,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.024248306457984724,
      "loss": 0.2341,
      "num_input_tokens_seen": 13798208,
      "step": 65390
    },
    {
      "epoch": 7.194169416941694,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.024247172652931605,
      "loss": 0.2263,
      "num_input_tokens_seen": 13799200,
      "step": 65395
    },
    {
      "epoch": 7.194719471947194,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.024246038762651727,
      "loss": 0.2315,
      "num_input_tokens_seen": 13800256,
      "step": 65400
    },
    {
      "epoch": 7.195269526952695,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.024244904787155557,
      "loss": 0.2352,
      "num_input_tokens_seen": 13801312,
      "step": 65405
    },
    {
      "epoch": 7.195819581958196,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02424377072645353,
      "loss": 0.2325,
      "num_input_tokens_seen": 13802368,
      "step": 65410
    },
    {
      "epoch": 7.196369636963697,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.024242636580556108,
      "loss": 0.231,
      "num_input_tokens_seen": 13803392,
      "step": 65415
    },
    {
      "epoch": 7.196919691969197,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.024241502349473736,
      "loss": 0.2299,
      "num_input_tokens_seen": 13804448,
      "step": 65420
    },
    {
      "epoch": 7.197469746974697,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.024240368033216872,
      "loss": 0.2289,
      "num_input_tokens_seen": 13805440,
      "step": 65425
    },
    {
      "epoch": 7.198019801980198,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02423923363179597,
      "loss": 0.2315,
      "num_input_tokens_seen": 13806560,
      "step": 65430
    },
    {
      "epoch": 7.198569856985698,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.024238099145221487,
      "loss": 0.2259,
      "num_input_tokens_seen": 13807648,
      "step": 65435
    },
    {
      "epoch": 7.1991199119911995,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.024236964573503876,
      "loss": 0.2331,
      "num_input_tokens_seen": 13808704,
      "step": 65440
    },
    {
      "epoch": 7.1996699669967,
      "grad_norm": 0.00457763671875,
      "learning_rate": 0.024235829916653594,
      "loss": 0.2316,
      "num_input_tokens_seen": 13809824,
      "step": 65445
    },
    {
      "epoch": 7.2002200220022,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.024234695174681097,
      "loss": 0.2306,
      "num_input_tokens_seen": 13810880,
      "step": 65450
    },
    {
      "epoch": 7.200770077007701,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.024233560347596852,
      "loss": 0.2315,
      "num_input_tokens_seen": 13811936,
      "step": 65455
    },
    {
      "epoch": 7.201320132013201,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02423242543541131,
      "loss": 0.2305,
      "num_input_tokens_seen": 13813024,
      "step": 65460
    },
    {
      "epoch": 7.201870187018702,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.02423129043813493,
      "loss": 0.2332,
      "num_input_tokens_seen": 13814080,
      "step": 65465
    },
    {
      "epoch": 7.2024202420242025,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.024230155355778178,
      "loss": 0.2295,
      "num_input_tokens_seen": 13815136,
      "step": 65470
    },
    {
      "epoch": 7.202970297029703,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.024229020188351512,
      "loss": 0.2294,
      "num_input_tokens_seen": 13816224,
      "step": 65475
    },
    {
      "epoch": 7.203520352035204,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.024227884935865394,
      "loss": 0.2285,
      "num_input_tokens_seen": 13817184,
      "step": 65480
    },
    {
      "epoch": 7.204070407040704,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.02422674959833029,
      "loss": 0.2265,
      "num_input_tokens_seen": 13818240,
      "step": 65485
    },
    {
      "epoch": 7.204620462046204,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.024225614175756666,
      "loss": 0.2349,
      "num_input_tokens_seen": 13819264,
      "step": 65490
    },
    {
      "epoch": 7.205170517051705,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.024224478668154976,
      "loss": 0.227,
      "num_input_tokens_seen": 13820320,
      "step": 65495
    },
    {
      "epoch": 7.2057205720572055,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.024223343075535697,
      "loss": 0.2321,
      "num_input_tokens_seen": 13821408,
      "step": 65500
    },
    {
      "epoch": 7.206270627062707,
      "grad_norm": 0.00457763671875,
      "learning_rate": 0.024222207397909288,
      "loss": 0.2285,
      "num_input_tokens_seen": 13822496,
      "step": 65505
    },
    {
      "epoch": 7.206820682068207,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.024221071635286227,
      "loss": 0.2311,
      "num_input_tokens_seen": 13823584,
      "step": 65510
    },
    {
      "epoch": 7.207370737073707,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.024219935787676963,
      "loss": 0.2359,
      "num_input_tokens_seen": 13824640,
      "step": 65515
    },
    {
      "epoch": 7.207920792079208,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.024218799855091976,
      "loss": 0.2379,
      "num_input_tokens_seen": 13825728,
      "step": 65520
    },
    {
      "epoch": 7.208470847084708,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.02421766383754174,
      "loss": 0.2354,
      "num_input_tokens_seen": 13826784,
      "step": 65525
    },
    {
      "epoch": 7.209020902090209,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.024216527735036717,
      "loss": 0.2281,
      "num_input_tokens_seen": 13827840,
      "step": 65530
    },
    {
      "epoch": 7.20957095709571,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02421539154758738,
      "loss": 0.2332,
      "num_input_tokens_seen": 13828896,
      "step": 65535
    },
    {
      "epoch": 7.21012101210121,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.0242142552752042,
      "loss": 0.2312,
      "num_input_tokens_seen": 13829952,
      "step": 65540
    },
    {
      "epoch": 7.210671067106711,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.024213118917897652,
      "loss": 0.227,
      "num_input_tokens_seen": 13831072,
      "step": 65545
    },
    {
      "epoch": 7.211221122112211,
      "grad_norm": 0.00457763671875,
      "learning_rate": 0.024211982475678205,
      "loss": 0.228,
      "num_input_tokens_seen": 13832160,
      "step": 65550
    },
    {
      "epoch": 7.211771177117711,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.024210845948556343,
      "loss": 0.2326,
      "num_input_tokens_seen": 13833248,
      "step": 65555
    },
    {
      "epoch": 7.212321232123212,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.02420970933654253,
      "loss": 0.2281,
      "num_input_tokens_seen": 13834304,
      "step": 65560
    },
    {
      "epoch": 7.212871287128713,
      "grad_norm": 0.0045166015625,
      "learning_rate": 0.024208572639647243,
      "loss": 0.2352,
      "num_input_tokens_seen": 13835392,
      "step": 65565
    },
    {
      "epoch": 7.213421342134214,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.024207435857880957,
      "loss": 0.2337,
      "num_input_tokens_seen": 13836480,
      "step": 65570
    },
    {
      "epoch": 7.213971397139714,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02420629899125416,
      "loss": 0.2338,
      "num_input_tokens_seen": 13837504,
      "step": 65575
    },
    {
      "epoch": 7.214521452145214,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.02420516203977732,
      "loss": 0.2259,
      "num_input_tokens_seen": 13838560,
      "step": 65580
    },
    {
      "epoch": 7.215071507150715,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02420402500346092,
      "loss": 0.2322,
      "num_input_tokens_seen": 13839616,
      "step": 65585
    },
    {
      "epoch": 7.215621562156215,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.02420288788231544,
      "loss": 0.2342,
      "num_input_tokens_seen": 13840640,
      "step": 65590
    },
    {
      "epoch": 7.2161716171617165,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.024201750676351354,
      "loss": 0.2347,
      "num_input_tokens_seen": 13841696,
      "step": 65595
    },
    {
      "epoch": 7.216721672167217,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.024200613385579148,
      "loss": 0.2273,
      "num_input_tokens_seen": 13842720,
      "step": 65600
    },
    {
      "epoch": 7.217271727172717,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.024199476010009307,
      "loss": 0.2346,
      "num_input_tokens_seen": 13843808,
      "step": 65605
    },
    {
      "epoch": 7.217821782178218,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.024198338549652303,
      "loss": 0.231,
      "num_input_tokens_seen": 13844864,
      "step": 65610
    },
    {
      "epoch": 7.218371837183718,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.024197201004518634,
      "loss": 0.232,
      "num_input_tokens_seen": 13845952,
      "step": 65615
    },
    {
      "epoch": 7.218921892189219,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02419606337461877,
      "loss": 0.2309,
      "num_input_tokens_seen": 13847008,
      "step": 65620
    },
    {
      "epoch": 7.2194719471947195,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02419492565996321,
      "loss": 0.2351,
      "num_input_tokens_seen": 13848064,
      "step": 65625
    },
    {
      "epoch": 7.22002200220022,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.024193787860562428,
      "loss": 0.2299,
      "num_input_tokens_seen": 13849056,
      "step": 65630
    },
    {
      "epoch": 7.220572057205721,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.024192649976426915,
      "loss": 0.2293,
      "num_input_tokens_seen": 13850112,
      "step": 65635
    },
    {
      "epoch": 7.221122112211221,
      "grad_norm": 0.003082275390625,
      "learning_rate": 0.024191512007567157,
      "loss": 0.2299,
      "num_input_tokens_seen": 13851168,
      "step": 65640
    },
    {
      "epoch": 7.221672167216722,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.024190373953993642,
      "loss": 0.2289,
      "num_input_tokens_seen": 13852160,
      "step": 65645
    },
    {
      "epoch": 7.222222222222222,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.024189235815716865,
      "loss": 0.2319,
      "num_input_tokens_seen": 13853184,
      "step": 65650
    },
    {
      "epoch": 7.2227722772277225,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.024188097592747308,
      "loss": 0.2325,
      "num_input_tokens_seen": 13854304,
      "step": 65655
    },
    {
      "epoch": 7.223322332233224,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.024186959285095466,
      "loss": 0.2356,
      "num_input_tokens_seen": 13855264,
      "step": 65660
    },
    {
      "epoch": 7.223872387238724,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.024185820892771832,
      "loss": 0.2345,
      "num_input_tokens_seen": 13856256,
      "step": 65665
    },
    {
      "epoch": 7.224422442244224,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.024184682415786887,
      "loss": 0.2298,
      "num_input_tokens_seen": 13857280,
      "step": 65670
    },
    {
      "epoch": 7.224972497249725,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.024183543854151137,
      "loss": 0.2314,
      "num_input_tokens_seen": 13858336,
      "step": 65675
    },
    {
      "epoch": 7.225522552255225,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.02418240520787507,
      "loss": 0.2314,
      "num_input_tokens_seen": 13859424,
      "step": 65680
    },
    {
      "epoch": 7.226072607260726,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.024181266476969176,
      "loss": 0.2304,
      "num_input_tokens_seen": 13860512,
      "step": 65685
    },
    {
      "epoch": 7.226622662266227,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.024180127661443958,
      "loss": 0.232,
      "num_input_tokens_seen": 13861536,
      "step": 65690
    },
    {
      "epoch": 7.227172717271727,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.024178988761309902,
      "loss": 0.2324,
      "num_input_tokens_seen": 13862592,
      "step": 65695
    },
    {
      "epoch": 7.227722772277228,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.02417784977657752,
      "loss": 0.2324,
      "num_input_tokens_seen": 13863616,
      "step": 65700
    },
    {
      "epoch": 7.228272827282728,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.024176710707257295,
      "loss": 0.2329,
      "num_input_tokens_seen": 13864736,
      "step": 65705
    },
    {
      "epoch": 7.228822882288229,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.024175571553359736,
      "loss": 0.2319,
      "num_input_tokens_seen": 13865792,
      "step": 65710
    },
    {
      "epoch": 7.229372937293729,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.02417443231489533,
      "loss": 0.2324,
      "num_input_tokens_seen": 13866848,
      "step": 65715
    },
    {
      "epoch": 7.22992299229923,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.02417329299187459,
      "loss": 0.2319,
      "num_input_tokens_seen": 13867872,
      "step": 65720
    },
    {
      "epoch": 7.230473047304731,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.024172153584308007,
      "loss": 0.2324,
      "num_input_tokens_seen": 13868992,
      "step": 65725
    },
    {
      "epoch": 7.231023102310231,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.024171014092206084,
      "loss": 0.2329,
      "num_input_tokens_seen": 13870080,
      "step": 65730
    },
    {
      "epoch": 7.231573157315731,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.024169874515579327,
      "loss": 0.2309,
      "num_input_tokens_seen": 13871200,
      "step": 65735
    },
    {
      "epoch": 7.232123212321232,
      "grad_norm": 0.009765625,
      "learning_rate": 0.024168734854438236,
      "loss": 0.2298,
      "num_input_tokens_seen": 13872288,
      "step": 65740
    },
    {
      "epoch": 7.232673267326732,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.024167595108793318,
      "loss": 0.2309,
      "num_input_tokens_seen": 13873344,
      "step": 65745
    },
    {
      "epoch": 7.2332233223322335,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.02416645527865507,
      "loss": 0.2298,
      "num_input_tokens_seen": 13874400,
      "step": 65750
    },
    {
      "epoch": 7.233773377337734,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.024165315364034008,
      "loss": 0.2324,
      "num_input_tokens_seen": 13875488,
      "step": 65755
    },
    {
      "epoch": 7.234323432343234,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.02416417536494063,
      "loss": 0.2298,
      "num_input_tokens_seen": 13876544,
      "step": 65760
    },
    {
      "epoch": 7.234873487348735,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.024163035281385442,
      "loss": 0.2324,
      "num_input_tokens_seen": 13877568,
      "step": 65765
    },
    {
      "epoch": 7.235423542354235,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02416189511337896,
      "loss": 0.2283,
      "num_input_tokens_seen": 13878656,
      "step": 65770
    },
    {
      "epoch": 7.235973597359736,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.024160754860931687,
      "loss": 0.2293,
      "num_input_tokens_seen": 13879776,
      "step": 65775
    },
    {
      "epoch": 7.2365236523652365,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.024159614524054126,
      "loss": 0.2304,
      "num_input_tokens_seen": 13880832,
      "step": 65780
    },
    {
      "epoch": 7.237073707370737,
      "grad_norm": 0.00457763671875,
      "learning_rate": 0.024158474102756792,
      "loss": 0.2315,
      "num_input_tokens_seen": 13881952,
      "step": 65785
    },
    {
      "epoch": 7.237623762376238,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.024157333597050204,
      "loss": 0.2332,
      "num_input_tokens_seen": 13883040,
      "step": 65790
    },
    {
      "epoch": 7.238173817381738,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.024156193006944864,
      "loss": 0.2326,
      "num_input_tokens_seen": 13884032,
      "step": 65795
    },
    {
      "epoch": 7.238723872387239,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.024155052332451285,
      "loss": 0.2385,
      "num_input_tokens_seen": 13885056,
      "step": 65800
    },
    {
      "epoch": 7.239273927392739,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.024153911573579982,
      "loss": 0.2311,
      "num_input_tokens_seen": 13886208,
      "step": 65805
    },
    {
      "epoch": 7.2398239823982395,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.024152770730341472,
      "loss": 0.2301,
      "num_input_tokens_seen": 13887200,
      "step": 65810
    },
    {
      "epoch": 7.240374037403741,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.024151629802746266,
      "loss": 0.2331,
      "num_input_tokens_seen": 13888224,
      "step": 65815
    },
    {
      "epoch": 7.240924092409241,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.024150488790804874,
      "loss": 0.2322,
      "num_input_tokens_seen": 13889312,
      "step": 65820
    },
    {
      "epoch": 7.241474147414741,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.02414934769452782,
      "loss": 0.2295,
      "num_input_tokens_seen": 13890272,
      "step": 65825
    },
    {
      "epoch": 7.242024202420242,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.02414820651392562,
      "loss": 0.2315,
      "num_input_tokens_seen": 13891424,
      "step": 65830
    },
    {
      "epoch": 7.242574257425742,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.02414706524900879,
      "loss": 0.2352,
      "num_input_tokens_seen": 13892512,
      "step": 65835
    },
    {
      "epoch": 7.243124312431243,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.024145923899787848,
      "loss": 0.232,
      "num_input_tokens_seen": 13893600,
      "step": 65840
    },
    {
      "epoch": 7.243674367436744,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02414478246627331,
      "loss": 0.2309,
      "num_input_tokens_seen": 13894720,
      "step": 65845
    },
    {
      "epoch": 7.244224422442244,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.02414364094847571,
      "loss": 0.2299,
      "num_input_tokens_seen": 13895840,
      "step": 65850
    },
    {
      "epoch": 7.244774477447745,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.02414249934640555,
      "loss": 0.2325,
      "num_input_tokens_seen": 13896896,
      "step": 65855
    },
    {
      "epoch": 7.245324532453245,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.024141357660073366,
      "loss": 0.2314,
      "num_input_tokens_seen": 13897952,
      "step": 65860
    },
    {
      "epoch": 7.245874587458746,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.024140215889489672,
      "loss": 0.2335,
      "num_input_tokens_seen": 13899008,
      "step": 65865
    },
    {
      "epoch": 7.2464246424642464,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.024139074034664997,
      "loss": 0.2351,
      "num_input_tokens_seen": 13900032,
      "step": 65870
    },
    {
      "epoch": 7.246974697469747,
      "grad_norm": 0.01251220703125,
      "learning_rate": 0.02413793209560986,
      "loss": 0.2308,
      "num_input_tokens_seen": 13901088,
      "step": 65875
    },
    {
      "epoch": 7.247524752475248,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02413679007233479,
      "loss": 0.2329,
      "num_input_tokens_seen": 13902176,
      "step": 65880
    },
    {
      "epoch": 7.248074807480748,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.024135647964850303,
      "loss": 0.2303,
      "num_input_tokens_seen": 13903232,
      "step": 65885
    },
    {
      "epoch": 7.248624862486249,
      "grad_norm": 0.007476806640625,
      "learning_rate": 0.02413450577316694,
      "loss": 0.2313,
      "num_input_tokens_seen": 13904256,
      "step": 65890
    },
    {
      "epoch": 7.249174917491749,
      "grad_norm": 0.007049560546875,
      "learning_rate": 0.024133363497295217,
      "loss": 0.2329,
      "num_input_tokens_seen": 13905312,
      "step": 65895
    },
    {
      "epoch": 7.2497249724972495,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.02413222113724567,
      "loss": 0.2334,
      "num_input_tokens_seen": 13906400,
      "step": 65900
    },
    {
      "epoch": 7.2502750275027505,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.02413107869302881,
      "loss": 0.2313,
      "num_input_tokens_seen": 13907392,
      "step": 65905
    },
    {
      "epoch": 7.250825082508251,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.024129936164655193,
      "loss": 0.2318,
      "num_input_tokens_seen": 13908416,
      "step": 65910
    },
    {
      "epoch": 7.251375137513751,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.02412879355213533,
      "loss": 0.2334,
      "num_input_tokens_seen": 13909536,
      "step": 65915
    },
    {
      "epoch": 7.251925192519252,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.024127650855479757,
      "loss": 0.2324,
      "num_input_tokens_seen": 13910560,
      "step": 65920
    },
    {
      "epoch": 7.252475247524752,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.024126508074699005,
      "loss": 0.2303,
      "num_input_tokens_seen": 13911616,
      "step": 65925
    },
    {
      "epoch": 7.253025302530253,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.024125365209803608,
      "loss": 0.2329,
      "num_input_tokens_seen": 13912672,
      "step": 65930
    },
    {
      "epoch": 7.2535753575357536,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0241242222608041,
      "loss": 0.2313,
      "num_input_tokens_seen": 13913728,
      "step": 65935
    },
    {
      "epoch": 7.254125412541254,
      "grad_norm": 0.0118408203125,
      "learning_rate": 0.024123079227711015,
      "loss": 0.2313,
      "num_input_tokens_seen": 13914752,
      "step": 65940
    },
    {
      "epoch": 7.254675467546755,
      "grad_norm": 0.006927490234375,
      "learning_rate": 0.024121936110534883,
      "loss": 0.2329,
      "num_input_tokens_seen": 13915808,
      "step": 65945
    },
    {
      "epoch": 7.255225522552255,
      "grad_norm": 0.013916015625,
      "learning_rate": 0.024120792909286242,
      "loss": 0.2319,
      "num_input_tokens_seen": 13916832,
      "step": 65950
    },
    {
      "epoch": 7.255775577557756,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.024119649623975635,
      "loss": 0.2313,
      "num_input_tokens_seen": 13917952,
      "step": 65955
    },
    {
      "epoch": 7.256325632563256,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.024118506254613592,
      "loss": 0.2309,
      "num_input_tokens_seen": 13919008,
      "step": 65960
    },
    {
      "epoch": 7.256875687568757,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.02411736280121065,
      "loss": 0.2335,
      "num_input_tokens_seen": 13920064,
      "step": 65965
    },
    {
      "epoch": 7.257425742574258,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.02411621926377735,
      "loss": 0.2319,
      "num_input_tokens_seen": 13921152,
      "step": 65970
    },
    {
      "epoch": 7.257975797579758,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02411507564232423,
      "loss": 0.2309,
      "num_input_tokens_seen": 13922176,
      "step": 65975
    },
    {
      "epoch": 7.258525852585258,
      "grad_norm": 0.0262451171875,
      "learning_rate": 0.024113931936861836,
      "loss": 0.2329,
      "num_input_tokens_seen": 13923168,
      "step": 65980
    },
    {
      "epoch": 7.259075907590759,
      "grad_norm": 0.0125732421875,
      "learning_rate": 0.024112788147400706,
      "loss": 0.2356,
      "num_input_tokens_seen": 13924128,
      "step": 65985
    },
    {
      "epoch": 7.259625962596259,
      "grad_norm": 0.0032958984375,
      "learning_rate": 0.024111644273951374,
      "loss": 0.2324,
      "num_input_tokens_seen": 13925216,
      "step": 65990
    },
    {
      "epoch": 7.2601760176017605,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.02411050031652439,
      "loss": 0.2303,
      "num_input_tokens_seen": 13926208,
      "step": 65995
    },
    {
      "epoch": 7.260726072607261,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0241093562751303,
      "loss": 0.2319,
      "num_input_tokens_seen": 13927232,
      "step": 66000
    },
    {
      "epoch": 7.261276127612761,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.024108212149779644,
      "loss": 0.2324,
      "num_input_tokens_seen": 13928320,
      "step": 66005
    },
    {
      "epoch": 7.261826182618262,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.02410706794048296,
      "loss": 0.2314,
      "num_input_tokens_seen": 13929376,
      "step": 66010
    },
    {
      "epoch": 7.262376237623762,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.02410592364725081,
      "loss": 0.2303,
      "num_input_tokens_seen": 13930368,
      "step": 66015
    },
    {
      "epoch": 7.262926292629263,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.024104779270093725,
      "loss": 0.2303,
      "num_input_tokens_seen": 13931392,
      "step": 66020
    },
    {
      "epoch": 7.2634763476347635,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02410363480902226,
      "loss": 0.2303,
      "num_input_tokens_seen": 13932448,
      "step": 66025
    },
    {
      "epoch": 7.264026402640264,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.024102490264046962,
      "loss": 0.2303,
      "num_input_tokens_seen": 13933536,
      "step": 66030
    },
    {
      "epoch": 7.264576457645765,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.024101345635178382,
      "loss": 0.2319,
      "num_input_tokens_seen": 13934592,
      "step": 66035
    },
    {
      "epoch": 7.265126512651265,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02410020092242706,
      "loss": 0.2303,
      "num_input_tokens_seen": 13935680,
      "step": 66040
    },
    {
      "epoch": 7.265676567656766,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.024099056125803563,
      "loss": 0.2319,
      "num_input_tokens_seen": 13936768,
      "step": 66045
    },
    {
      "epoch": 7.266226622662266,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.024097911245318418,
      "loss": 0.2308,
      "num_input_tokens_seen": 13937824,
      "step": 66050
    },
    {
      "epoch": 7.2667766776677665,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.024096766280982205,
      "loss": 0.2309,
      "num_input_tokens_seen": 13938848,
      "step": 66055
    },
    {
      "epoch": 7.267326732673268,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.02409562123280545,
      "loss": 0.2298,
      "num_input_tokens_seen": 13939936,
      "step": 66060
    },
    {
      "epoch": 7.267876787678768,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.024094476100798727,
      "loss": 0.2324,
      "num_input_tokens_seen": 13940992,
      "step": 66065
    },
    {
      "epoch": 7.268426842684269,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.024093330884972574,
      "loss": 0.232,
      "num_input_tokens_seen": 13941984,
      "step": 66070
    },
    {
      "epoch": 7.268976897689769,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.024092185585337562,
      "loss": 0.2314,
      "num_input_tokens_seen": 13943072,
      "step": 66075
    },
    {
      "epoch": 7.269526952695269,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.024091040201904237,
      "loss": 0.2324,
      "num_input_tokens_seen": 13944160,
      "step": 66080
    },
    {
      "epoch": 7.27007700770077,
      "grad_norm": 0.0004119873046875,
      "learning_rate": 0.02408989473468315,
      "loss": 0.2308,
      "num_input_tokens_seen": 13945216,
      "step": 66085
    },
    {
      "epoch": 7.270627062706271,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.024088749183684866,
      "loss": 0.2314,
      "num_input_tokens_seen": 13946272,
      "step": 66090
    },
    {
      "epoch": 7.271177117711771,
      "grad_norm": 0.000713348388671875,
      "learning_rate": 0.024087603548919947,
      "loss": 0.2329,
      "num_input_tokens_seen": 13947296,
      "step": 66095
    },
    {
      "epoch": 7.271727172717272,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02408645783039895,
      "loss": 0.2314,
      "num_input_tokens_seen": 13948352,
      "step": 66100
    },
    {
      "epoch": 7.272277227722772,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.024085312028132425,
      "loss": 0.2298,
      "num_input_tokens_seen": 13949440,
      "step": 66105
    },
    {
      "epoch": 7.272827282728273,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.024084166142130936,
      "loss": 0.2329,
      "num_input_tokens_seen": 13950528,
      "step": 66110
    },
    {
      "epoch": 7.273377337733773,
      "grad_norm": 0.00063323974609375,
      "learning_rate": 0.024083020172405052,
      "loss": 0.2319,
      "num_input_tokens_seen": 13951520,
      "step": 66115
    },
    {
      "epoch": 7.273927392739274,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02408187411896533,
      "loss": 0.2303,
      "num_input_tokens_seen": 13952608,
      "step": 66120
    },
    {
      "epoch": 7.274477447744775,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.024080727981822328,
      "loss": 0.2319,
      "num_input_tokens_seen": 13953696,
      "step": 66125
    },
    {
      "epoch": 7.275027502750275,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.024079581760986617,
      "loss": 0.2319,
      "num_input_tokens_seen": 13954688,
      "step": 66130
    },
    {
      "epoch": 7.275577557755776,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.024078435456468754,
      "loss": 0.2298,
      "num_input_tokens_seen": 13955776,
      "step": 66135
    },
    {
      "epoch": 7.276127612761276,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.02407728906827931,
      "loss": 0.2303,
      "num_input_tokens_seen": 13956832,
      "step": 66140
    },
    {
      "epoch": 7.276677667766776,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02407614259642885,
      "loss": 0.233,
      "num_input_tokens_seen": 13957888,
      "step": 66145
    },
    {
      "epoch": 7.2772277227722775,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.024074996040927936,
      "loss": 0.2314,
      "num_input_tokens_seen": 13958912,
      "step": 66150
    },
    {
      "epoch": 7.277777777777778,
      "grad_norm": 0.00061798095703125,
      "learning_rate": 0.02407384940178714,
      "loss": 0.2319,
      "num_input_tokens_seen": 13959936,
      "step": 66155
    },
    {
      "epoch": 7.278327832783278,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.024072702679017028,
      "loss": 0.2278,
      "num_input_tokens_seen": 13960960,
      "step": 66160
    },
    {
      "epoch": 7.278877887788779,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.024071555872628166,
      "loss": 0.2309,
      "num_input_tokens_seen": 13962080,
      "step": 66165
    },
    {
      "epoch": 7.279427942794279,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.02407040898263113,
      "loss": 0.2304,
      "num_input_tokens_seen": 13963136,
      "step": 66170
    },
    {
      "epoch": 7.27997799779978,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.024069262009036488,
      "loss": 0.2283,
      "num_input_tokens_seen": 13964192,
      "step": 66175
    },
    {
      "epoch": 7.2805280528052805,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.02406811495185481,
      "loss": 0.233,
      "num_input_tokens_seen": 13965248,
      "step": 66180
    },
    {
      "epoch": 7.281078107810781,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.02406696781109666,
      "loss": 0.2299,
      "num_input_tokens_seen": 13966368,
      "step": 66185
    },
    {
      "epoch": 7.281628162816282,
      "grad_norm": 0.00677490234375,
      "learning_rate": 0.024065820586772625,
      "loss": 0.2336,
      "num_input_tokens_seen": 13967424,
      "step": 66190
    },
    {
      "epoch": 7.282178217821782,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.02406467327889327,
      "loss": 0.2347,
      "num_input_tokens_seen": 13968512,
      "step": 66195
    },
    {
      "epoch": 7.282728272827283,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.024063525887469173,
      "loss": 0.2309,
      "num_input_tokens_seen": 13969600,
      "step": 66200
    },
    {
      "epoch": 7.283278327832783,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.024062378412510906,
      "loss": 0.2315,
      "num_input_tokens_seen": 13970720,
      "step": 66205
    },
    {
      "epoch": 7.2838283828382835,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.02406123085402905,
      "loss": 0.2319,
      "num_input_tokens_seen": 13971712,
      "step": 66210
    },
    {
      "epoch": 7.284378437843785,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.024060083212034167,
      "loss": 0.2278,
      "num_input_tokens_seen": 13972768,
      "step": 66215
    },
    {
      "epoch": 7.284928492849285,
      "grad_norm": 0.0120849609375,
      "learning_rate": 0.02405893548653685,
      "loss": 0.2283,
      "num_input_tokens_seen": 13973792,
      "step": 66220
    },
    {
      "epoch": 7.285478547854786,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.024057787677547675,
      "loss": 0.2301,
      "num_input_tokens_seen": 13974912,
      "step": 66225
    },
    {
      "epoch": 7.286028602860286,
      "grad_norm": 0.007293701171875,
      "learning_rate": 0.02405663978507721,
      "loss": 0.2311,
      "num_input_tokens_seen": 13976032,
      "step": 66230
    },
    {
      "epoch": 7.286578657865786,
      "grad_norm": 0.007476806640625,
      "learning_rate": 0.024055491809136048,
      "loss": 0.2342,
      "num_input_tokens_seen": 13977024,
      "step": 66235
    },
    {
      "epoch": 7.287128712871287,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.02405434374973476,
      "loss": 0.2378,
      "num_input_tokens_seen": 13978112,
      "step": 66240
    },
    {
      "epoch": 7.287678767876788,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02405319560688393,
      "loss": 0.2309,
      "num_input_tokens_seen": 13979168,
      "step": 66245
    },
    {
      "epoch": 7.288228822882289,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.024052047380594143,
      "loss": 0.232,
      "num_input_tokens_seen": 13980192,
      "step": 66250
    },
    {
      "epoch": 7.288778877887789,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.02405089907087598,
      "loss": 0.232,
      "num_input_tokens_seen": 13981280,
      "step": 66255
    },
    {
      "epoch": 7.289328932893289,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.02404975067774002,
      "loss": 0.2314,
      "num_input_tokens_seen": 13982336,
      "step": 66260
    },
    {
      "epoch": 7.28987898789879,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.02404860220119685,
      "loss": 0.2346,
      "num_input_tokens_seen": 13983392,
      "step": 66265
    },
    {
      "epoch": 7.29042904290429,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.024047453641257058,
      "loss": 0.2314,
      "num_input_tokens_seen": 13984480,
      "step": 66270
    },
    {
      "epoch": 7.290979097909791,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.024046304997931225,
      "loss": 0.2324,
      "num_input_tokens_seen": 13985568,
      "step": 66275
    },
    {
      "epoch": 7.291529152915292,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.024045156271229943,
      "loss": 0.2293,
      "num_input_tokens_seen": 13986720,
      "step": 66280
    },
    {
      "epoch": 7.292079207920792,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.024044007461163794,
      "loss": 0.2288,
      "num_input_tokens_seen": 13987872,
      "step": 66285
    },
    {
      "epoch": 7.292629262926293,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.024042858567743363,
      "loss": 0.232,
      "num_input_tokens_seen": 13988928,
      "step": 66290
    },
    {
      "epoch": 7.293179317931793,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.024041709590979253,
      "loss": 0.2314,
      "num_input_tokens_seen": 13989984,
      "step": 66295
    },
    {
      "epoch": 7.293729372937293,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02404056053088204,
      "loss": 0.2314,
      "num_input_tokens_seen": 13991104,
      "step": 66300
    },
    {
      "epoch": 7.2942794279427945,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02403941138746232,
      "loss": 0.2304,
      "num_input_tokens_seen": 13992160,
      "step": 66305
    },
    {
      "epoch": 7.294829482948295,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.024038262160730677,
      "loss": 0.2324,
      "num_input_tokens_seen": 13993248,
      "step": 66310
    },
    {
      "epoch": 7.295379537953796,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.024037112850697716,
      "loss": 0.2325,
      "num_input_tokens_seen": 13994304,
      "step": 66315
    },
    {
      "epoch": 7.295929592959296,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.024035963457374018,
      "loss": 0.232,
      "num_input_tokens_seen": 13995360,
      "step": 66320
    },
    {
      "epoch": 7.296479647964796,
      "grad_norm": 0.0004634857177734375,
      "learning_rate": 0.02403481398077018,
      "loss": 0.2325,
      "num_input_tokens_seen": 13996416,
      "step": 66325
    },
    {
      "epoch": 7.297029702970297,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.024033664420896796,
      "loss": 0.2304,
      "num_input_tokens_seen": 13997472,
      "step": 66330
    },
    {
      "epoch": 7.2975797579757975,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.024032514777764465,
      "loss": 0.232,
      "num_input_tokens_seen": 13998560,
      "step": 66335
    },
    {
      "epoch": 7.298129812981298,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.02403136505138378,
      "loss": 0.2309,
      "num_input_tokens_seen": 13999648,
      "step": 66340
    },
    {
      "epoch": 7.298679867986799,
      "grad_norm": 0.00079345703125,
      "learning_rate": 0.024030215241765338,
      "loss": 0.2325,
      "num_input_tokens_seen": 14000608,
      "step": 66345
    },
    {
      "epoch": 7.299229922992299,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.024029065348919732,
      "loss": 0.2294,
      "num_input_tokens_seen": 14001696,
      "step": 66350
    },
    {
      "epoch": 7.2997799779978,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.024027915372857565,
      "loss": 0.2351,
      "num_input_tokens_seen": 14002816,
      "step": 66355
    },
    {
      "epoch": 7.3003300330033,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02402676531358943,
      "loss": 0.2361,
      "num_input_tokens_seen": 14003872,
      "step": 66360
    },
    {
      "epoch": 7.3008800880088005,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.024025615171125937,
      "loss": 0.2304,
      "num_input_tokens_seen": 14004896,
      "step": 66365
    },
    {
      "epoch": 7.301430143014302,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.024024464945477678,
      "loss": 0.2309,
      "num_input_tokens_seen": 14005984,
      "step": 66370
    },
    {
      "epoch": 7.301980198019802,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.024023314636655253,
      "loss": 0.2325,
      "num_input_tokens_seen": 14006976,
      "step": 66375
    },
    {
      "epoch": 7.302530253025303,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.02402216424466927,
      "loss": 0.2351,
      "num_input_tokens_seen": 14008032,
      "step": 66380
    },
    {
      "epoch": 7.303080308030803,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.02402101376953033,
      "loss": 0.2325,
      "num_input_tokens_seen": 14009120,
      "step": 66385
    },
    {
      "epoch": 7.303630363036303,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.02401986321124903,
      "loss": 0.2293,
      "num_input_tokens_seen": 14010240,
      "step": 66390
    },
    {
      "epoch": 7.304180418041804,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02401871256983598,
      "loss": 0.2288,
      "num_input_tokens_seen": 14011200,
      "step": 66395
    },
    {
      "epoch": 7.304730473047305,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.024017561845301787,
      "loss": 0.2319,
      "num_input_tokens_seen": 14012256,
      "step": 66400
    },
    {
      "epoch": 7.305280528052805,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02401641103765705,
      "loss": 0.2309,
      "num_input_tokens_seen": 14013344,
      "step": 66405
    },
    {
      "epoch": 7.305830583058306,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02401526014691238,
      "loss": 0.2319,
      "num_input_tokens_seen": 14014400,
      "step": 66410
    },
    {
      "epoch": 7.306380638063806,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.024014109173078384,
      "loss": 0.2308,
      "num_input_tokens_seen": 14015456,
      "step": 66415
    },
    {
      "epoch": 7.306930693069307,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.024012958116165665,
      "loss": 0.2303,
      "num_input_tokens_seen": 14016512,
      "step": 66420
    },
    {
      "epoch": 7.307480748074807,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02401180697618484,
      "loss": 0.2298,
      "num_input_tokens_seen": 14017536,
      "step": 66425
    },
    {
      "epoch": 7.3080308030803085,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.024010655753146516,
      "loss": 0.2278,
      "num_input_tokens_seen": 14018624,
      "step": 66430
    },
    {
      "epoch": 7.308580858085809,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.024009504447061302,
      "loss": 0.2298,
      "num_input_tokens_seen": 14019648,
      "step": 66435
    },
    {
      "epoch": 7.309130913091309,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.024008353057939802,
      "loss": 0.2319,
      "num_input_tokens_seen": 14020672,
      "step": 66440
    },
    {
      "epoch": 7.30968096809681,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02400720158579264,
      "loss": 0.2304,
      "num_input_tokens_seen": 14021696,
      "step": 66445
    },
    {
      "epoch": 7.31023102310231,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02400605003063042,
      "loss": 0.2325,
      "num_input_tokens_seen": 14022784,
      "step": 66450
    },
    {
      "epoch": 7.31078107810781,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.024004898392463763,
      "loss": 0.2361,
      "num_input_tokens_seen": 14023840,
      "step": 66455
    },
    {
      "epoch": 7.3113311331133115,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.024003746671303273,
      "loss": 0.233,
      "num_input_tokens_seen": 14024992,
      "step": 66460
    },
    {
      "epoch": 7.311881188118812,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.024002594867159572,
      "loss": 0.2329,
      "num_input_tokens_seen": 14026048,
      "step": 66465
    },
    {
      "epoch": 7.312431243124313,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.024001442980043274,
      "loss": 0.2298,
      "num_input_tokens_seen": 14027040,
      "step": 66470
    },
    {
      "epoch": 7.312981298129813,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.024000291009965,
      "loss": 0.2324,
      "num_input_tokens_seen": 14028160,
      "step": 66475
    },
    {
      "epoch": 7.313531353135313,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.023999138956935357,
      "loss": 0.2309,
      "num_input_tokens_seen": 14029184,
      "step": 66480
    },
    {
      "epoch": 7.314081408140814,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.023997986820964972,
      "loss": 0.2293,
      "num_input_tokens_seen": 14030304,
      "step": 66485
    },
    {
      "epoch": 7.3146314631463145,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.023996834602064455,
      "loss": 0.2304,
      "num_input_tokens_seen": 14031392,
      "step": 66490
    },
    {
      "epoch": 7.315181518151816,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.023995682300244438,
      "loss": 0.2329,
      "num_input_tokens_seen": 14032480,
      "step": 66495
    },
    {
      "epoch": 7.315731573157316,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.023994529915515525,
      "loss": 0.2309,
      "num_input_tokens_seen": 14033504,
      "step": 66500
    },
    {
      "epoch": 7.316281628162816,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02399337744788835,
      "loss": 0.2319,
      "num_input_tokens_seen": 14034528,
      "step": 66505
    },
    {
      "epoch": 7.316831683168317,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.023992224897373533,
      "loss": 0.2314,
      "num_input_tokens_seen": 14035584,
      "step": 66510
    },
    {
      "epoch": 7.317381738173817,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023991072263981693,
      "loss": 0.2324,
      "num_input_tokens_seen": 14036640,
      "step": 66515
    },
    {
      "epoch": 7.3179317931793175,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023989919547723455,
      "loss": 0.2314,
      "num_input_tokens_seen": 14037664,
      "step": 66520
    },
    {
      "epoch": 7.318481848184819,
      "grad_norm": 0.00040435791015625,
      "learning_rate": 0.02398876674860944,
      "loss": 0.2314,
      "num_input_tokens_seen": 14038688,
      "step": 66525
    },
    {
      "epoch": 7.319031903190319,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.023987613866650274,
      "loss": 0.2329,
      "num_input_tokens_seen": 14039776,
      "step": 66530
    },
    {
      "epoch": 7.31958195819582,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.023986460901856587,
      "loss": 0.2298,
      "num_input_tokens_seen": 14040864,
      "step": 66535
    },
    {
      "epoch": 7.32013201320132,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.023985307854238996,
      "loss": 0.2314,
      "num_input_tokens_seen": 14041952,
      "step": 66540
    },
    {
      "epoch": 7.32068206820682,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.023984154723808142,
      "loss": 0.2309,
      "num_input_tokens_seen": 14043008,
      "step": 66545
    },
    {
      "epoch": 7.321232123212321,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023983001510574642,
      "loss": 0.2314,
      "num_input_tokens_seen": 14044032,
      "step": 66550
    },
    {
      "epoch": 7.321782178217822,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.023981848214549125,
      "loss": 0.2298,
      "num_input_tokens_seen": 14045056,
      "step": 66555
    },
    {
      "epoch": 7.322332233223323,
      "grad_norm": 0.003082275390625,
      "learning_rate": 0.023980694835742226,
      "loss": 0.2319,
      "num_input_tokens_seen": 14046208,
      "step": 66560
    },
    {
      "epoch": 7.322882288228823,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.023979541374164567,
      "loss": 0.2293,
      "num_input_tokens_seen": 14047296,
      "step": 66565
    },
    {
      "epoch": 7.323432343234323,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02397838782982679,
      "loss": 0.2324,
      "num_input_tokens_seen": 14048320,
      "step": 66570
    },
    {
      "epoch": 7.323982398239824,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02397723420273952,
      "loss": 0.2303,
      "num_input_tokens_seen": 14049408,
      "step": 66575
    },
    {
      "epoch": 7.324532453245324,
      "grad_norm": 0.009765625,
      "learning_rate": 0.023976080492913387,
      "loss": 0.2299,
      "num_input_tokens_seen": 14050528,
      "step": 66580
    },
    {
      "epoch": 7.325082508250825,
      "grad_norm": 0.0003490447998046875,
      "learning_rate": 0.023974926700359025,
      "loss": 0.2309,
      "num_input_tokens_seen": 14051552,
      "step": 66585
    },
    {
      "epoch": 7.325632563256326,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.023973772825087076,
      "loss": 0.2304,
      "num_input_tokens_seen": 14052608,
      "step": 66590
    },
    {
      "epoch": 7.326182618261826,
      "grad_norm": 0.000606536865234375,
      "learning_rate": 0.023972618867108168,
      "loss": 0.2303,
      "num_input_tokens_seen": 14053600,
      "step": 66595
    },
    {
      "epoch": 7.326732673267327,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.023971464826432937,
      "loss": 0.2325,
      "num_input_tokens_seen": 14054656,
      "step": 66600
    },
    {
      "epoch": 7.327282728272827,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02397031070307202,
      "loss": 0.2278,
      "num_input_tokens_seen": 14055712,
      "step": 66605
    },
    {
      "epoch": 7.327832783278327,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.02396915649703605,
      "loss": 0.2356,
      "num_input_tokens_seen": 14056800,
      "step": 66610
    },
    {
      "epoch": 7.3283828382838285,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.02396800220833567,
      "loss": 0.2247,
      "num_input_tokens_seen": 14057888,
      "step": 66615
    },
    {
      "epoch": 7.328932893289329,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.02396684783698152,
      "loss": 0.2309,
      "num_input_tokens_seen": 14058944,
      "step": 66620
    },
    {
      "epoch": 7.32948294829483,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.023965693382984234,
      "loss": 0.2309,
      "num_input_tokens_seen": 14060032,
      "step": 66625
    },
    {
      "epoch": 7.33003300330033,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.023964538846354456,
      "loss": 0.233,
      "num_input_tokens_seen": 14061056,
      "step": 66630
    },
    {
      "epoch": 7.33058305830583,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02396338422710282,
      "loss": 0.2325,
      "num_input_tokens_seen": 14062112,
      "step": 66635
    },
    {
      "epoch": 7.331133113311331,
      "grad_norm": 0.001953125,
      "learning_rate": 0.023962229525239975,
      "loss": 0.233,
      "num_input_tokens_seen": 14063072,
      "step": 66640
    },
    {
      "epoch": 7.3316831683168315,
      "grad_norm": 0.000705718994140625,
      "learning_rate": 0.02396107474077656,
      "loss": 0.2314,
      "num_input_tokens_seen": 14064064,
      "step": 66645
    },
    {
      "epoch": 7.332233223322333,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.023959919873723225,
      "loss": 0.2325,
      "num_input_tokens_seen": 14065120,
      "step": 66650
    },
    {
      "epoch": 7.332783278327833,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.023958764924090603,
      "loss": 0.2319,
      "num_input_tokens_seen": 14066208,
      "step": 66655
    },
    {
      "epoch": 7.333333333333333,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.023957609891889345,
      "loss": 0.2298,
      "num_input_tokens_seen": 14067296,
      "step": 66660
    },
    {
      "epoch": 7.333883388338834,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02395645477713009,
      "loss": 0.2298,
      "num_input_tokens_seen": 14068384,
      "step": 66665
    },
    {
      "epoch": 7.334433443344334,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.023955299579823494,
      "loss": 0.2304,
      "num_input_tokens_seen": 14069472,
      "step": 66670
    },
    {
      "epoch": 7.334983498349835,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.0239541442999802,
      "loss": 0.2324,
      "num_input_tokens_seen": 14070528,
      "step": 66675
    },
    {
      "epoch": 7.335533553355336,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.023952988937610855,
      "loss": 0.2314,
      "num_input_tokens_seen": 14071520,
      "step": 66680
    },
    {
      "epoch": 7.336083608360836,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.023951833492726105,
      "loss": 0.2314,
      "num_input_tokens_seen": 14072544,
      "step": 66685
    },
    {
      "epoch": 7.336633663366337,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.023950677965336595,
      "loss": 0.2329,
      "num_input_tokens_seen": 14073568,
      "step": 66690
    },
    {
      "epoch": 7.337183718371837,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02394952235545299,
      "loss": 0.2345,
      "num_input_tokens_seen": 14074560,
      "step": 66695
    },
    {
      "epoch": 7.337733773377337,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.02394836666308593,
      "loss": 0.2314,
      "num_input_tokens_seen": 14075584,
      "step": 66700
    },
    {
      "epoch": 7.338283828382838,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.023947210888246067,
      "loss": 0.2314,
      "num_input_tokens_seen": 14076672,
      "step": 66705
    },
    {
      "epoch": 7.338833883388339,
      "grad_norm": 0.00048828125,
      "learning_rate": 0.023946055030944054,
      "loss": 0.2319,
      "num_input_tokens_seen": 14077696,
      "step": 66710
    },
    {
      "epoch": 7.33938393839384,
      "grad_norm": 0.0003604888916015625,
      "learning_rate": 0.023944899091190543,
      "loss": 0.2324,
      "num_input_tokens_seen": 14078720,
      "step": 66715
    },
    {
      "epoch": 7.33993399339934,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.023943743068996194,
      "loss": 0.2298,
      "num_input_tokens_seen": 14079744,
      "step": 66720
    },
    {
      "epoch": 7.34048404840484,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023942586964371654,
      "loss": 0.2303,
      "num_input_tokens_seen": 14080896,
      "step": 66725
    },
    {
      "epoch": 7.341034103410341,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.023941430777327584,
      "loss": 0.2319,
      "num_input_tokens_seen": 14081920,
      "step": 66730
    },
    {
      "epoch": 7.341584158415841,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.023940274507874632,
      "loss": 0.2309,
      "num_input_tokens_seen": 14083072,
      "step": 66735
    },
    {
      "epoch": 7.3421342134213425,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.023939118156023464,
      "loss": 0.2314,
      "num_input_tokens_seen": 14084128,
      "step": 66740
    },
    {
      "epoch": 7.342684268426843,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02393796172178473,
      "loss": 0.2319,
      "num_input_tokens_seen": 14085184,
      "step": 66745
    },
    {
      "epoch": 7.343234323432343,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.023936805205169095,
      "loss": 0.2308,
      "num_input_tokens_seen": 14086208,
      "step": 66750
    },
    {
      "epoch": 7.343784378437844,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.023935648606187212,
      "loss": 0.2324,
      "num_input_tokens_seen": 14087360,
      "step": 66755
    },
    {
      "epoch": 7.344334433443344,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.023934491924849744,
      "loss": 0.2298,
      "num_input_tokens_seen": 14088416,
      "step": 66760
    },
    {
      "epoch": 7.3448844884488445,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023933335161167354,
      "loss": 0.2293,
      "num_input_tokens_seen": 14089504,
      "step": 66765
    },
    {
      "epoch": 7.3454345434543455,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.0239321783151507,
      "loss": 0.2335,
      "num_input_tokens_seen": 14090592,
      "step": 66770
    },
    {
      "epoch": 7.345984598459846,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.023931021386810438,
      "loss": 0.2314,
      "num_input_tokens_seen": 14091712,
      "step": 66775
    },
    {
      "epoch": 7.346534653465347,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023929864376157243,
      "loss": 0.2314,
      "num_input_tokens_seen": 14092768,
      "step": 66780
    },
    {
      "epoch": 7.347084708470847,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.02392870728320177,
      "loss": 0.2314,
      "num_input_tokens_seen": 14093792,
      "step": 66785
    },
    {
      "epoch": 7.347634763476347,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02392755010795469,
      "loss": 0.2309,
      "num_input_tokens_seen": 14094848,
      "step": 66790
    },
    {
      "epoch": 7.348184818481848,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.023926392850426664,
      "loss": 0.2324,
      "num_input_tokens_seen": 14095872,
      "step": 66795
    },
    {
      "epoch": 7.3487348734873486,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02392523551062836,
      "loss": 0.233,
      "num_input_tokens_seen": 14096864,
      "step": 66800
    },
    {
      "epoch": 7.34928492849285,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.023924078088570437,
      "loss": 0.2303,
      "num_input_tokens_seen": 14097952,
      "step": 66805
    },
    {
      "epoch": 7.34983498349835,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02392292058426357,
      "loss": 0.2298,
      "num_input_tokens_seen": 14098976,
      "step": 66810
    },
    {
      "epoch": 7.35038503850385,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.023921762997718428,
      "loss": 0.2303,
      "num_input_tokens_seen": 14100096,
      "step": 66815
    },
    {
      "epoch": 7.350935093509351,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.023920605328945677,
      "loss": 0.2319,
      "num_input_tokens_seen": 14101152,
      "step": 66820
    },
    {
      "epoch": 7.351485148514851,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02391944757795599,
      "loss": 0.2314,
      "num_input_tokens_seen": 14102176,
      "step": 66825
    },
    {
      "epoch": 7.3520352035203524,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.02391828974476003,
      "loss": 0.2293,
      "num_input_tokens_seen": 14103200,
      "step": 66830
    },
    {
      "epoch": 7.352585258525853,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.023917131829368474,
      "loss": 0.2324,
      "num_input_tokens_seen": 14104320,
      "step": 66835
    },
    {
      "epoch": 7.353135313531353,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.023915973831791996,
      "loss": 0.2345,
      "num_input_tokens_seen": 14105344,
      "step": 66840
    },
    {
      "epoch": 7.353685368536854,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 0.02391481575204126,
      "loss": 0.2335,
      "num_input_tokens_seen": 14106368,
      "step": 66845
    },
    {
      "epoch": 7.354235423542354,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.02391365759012695,
      "loss": 0.2309,
      "num_input_tokens_seen": 14107392,
      "step": 66850
    },
    {
      "epoch": 7.354785478547855,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.023912499346059733,
      "loss": 0.2319,
      "num_input_tokens_seen": 14108448,
      "step": 66855
    },
    {
      "epoch": 7.3553355335533555,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.023911341019850284,
      "loss": 0.2314,
      "num_input_tokens_seen": 14109472,
      "step": 66860
    },
    {
      "epoch": 7.355885588558856,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.023910182611509283,
      "loss": 0.2329,
      "num_input_tokens_seen": 14110496,
      "step": 66865
    },
    {
      "epoch": 7.356435643564357,
      "grad_norm": 0.009765625,
      "learning_rate": 0.023909024121047406,
      "loss": 0.2308,
      "num_input_tokens_seen": 14111488,
      "step": 66870
    },
    {
      "epoch": 7.356985698569857,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.023907865548475328,
      "loss": 0.2314,
      "num_input_tokens_seen": 14112576,
      "step": 66875
    },
    {
      "epoch": 7.357535753575357,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023906706893803725,
      "loss": 0.2314,
      "num_input_tokens_seen": 14113632,
      "step": 66880
    },
    {
      "epoch": 7.358085808580858,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02390554815704328,
      "loss": 0.2293,
      "num_input_tokens_seen": 14114816,
      "step": 66885
    },
    {
      "epoch": 7.3586358635863585,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.023904389338204674,
      "loss": 0.2303,
      "num_input_tokens_seen": 14115808,
      "step": 66890
    },
    {
      "epoch": 7.3591859185918596,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02390323043729858,
      "loss": 0.2319,
      "num_input_tokens_seen": 14116896,
      "step": 66895
    },
    {
      "epoch": 7.35973597359736,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.023902071454335688,
      "loss": 0.2319,
      "num_input_tokens_seen": 14117952,
      "step": 66900
    },
    {
      "epoch": 7.36028602860286,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.023900912389326668,
      "loss": 0.2309,
      "num_input_tokens_seen": 14118976,
      "step": 66905
    },
    {
      "epoch": 7.360836083608361,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02389975324228221,
      "loss": 0.2298,
      "num_input_tokens_seen": 14120064,
      "step": 66910
    },
    {
      "epoch": 7.361386138613861,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.023898594013213002,
      "loss": 0.2298,
      "num_input_tokens_seen": 14121088,
      "step": 66915
    },
    {
      "epoch": 7.361936193619362,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.023897434702129725,
      "loss": 0.2309,
      "num_input_tokens_seen": 14122240,
      "step": 66920
    },
    {
      "epoch": 7.362486248624863,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023896275309043057,
      "loss": 0.2309,
      "num_input_tokens_seen": 14123264,
      "step": 66925
    },
    {
      "epoch": 7.363036303630363,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.023895115833963688,
      "loss": 0.2303,
      "num_input_tokens_seen": 14124288,
      "step": 66930
    },
    {
      "epoch": 7.363586358635864,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023893956276902308,
      "loss": 0.2283,
      "num_input_tokens_seen": 14125376,
      "step": 66935
    },
    {
      "epoch": 7.364136413641364,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0238927966378696,
      "loss": 0.2309,
      "num_input_tokens_seen": 14126464,
      "step": 66940
    },
    {
      "epoch": 7.364686468646864,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.023891636916876246,
      "loss": 0.2324,
      "num_input_tokens_seen": 14127456,
      "step": 66945
    },
    {
      "epoch": 7.365236523652365,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.023890477113932947,
      "loss": 0.2293,
      "num_input_tokens_seen": 14128480,
      "step": 66950
    },
    {
      "epoch": 7.365786578657866,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.023889317229050386,
      "loss": 0.2309,
      "num_input_tokens_seen": 14129536,
      "step": 66955
    },
    {
      "epoch": 7.366336633663367,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.023888157262239258,
      "loss": 0.2304,
      "num_input_tokens_seen": 14130592,
      "step": 66960
    },
    {
      "epoch": 7.366886688668867,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.02388699721351024,
      "loss": 0.2309,
      "num_input_tokens_seen": 14131552,
      "step": 66965
    },
    {
      "epoch": 7.367436743674367,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023885837082874038,
      "loss": 0.2319,
      "num_input_tokens_seen": 14132640,
      "step": 66970
    },
    {
      "epoch": 7.367986798679868,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02388467687034134,
      "loss": 0.2314,
      "num_input_tokens_seen": 14133760,
      "step": 66975
    },
    {
      "epoch": 7.368536853685368,
      "grad_norm": 0.0028839111328125,
      "learning_rate": 0.023883516575922838,
      "loss": 0.2309,
      "num_input_tokens_seen": 14134848,
      "step": 66980
    },
    {
      "epoch": 7.3690869086908695,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.023882356199629227,
      "loss": 0.2314,
      "num_input_tokens_seen": 14135904,
      "step": 66985
    },
    {
      "epoch": 7.36963696369637,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.023881195741471198,
      "loss": 0.234,
      "num_input_tokens_seen": 14136960,
      "step": 66990
    },
    {
      "epoch": 7.37018701870187,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.023880035201459454,
      "loss": 0.2309,
      "num_input_tokens_seen": 14138048,
      "step": 66995
    },
    {
      "epoch": 7.370737073707371,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.023878874579604685,
      "loss": 0.2314,
      "num_input_tokens_seen": 14139104,
      "step": 67000
    },
    {
      "epoch": 7.371287128712871,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.023877713875917587,
      "loss": 0.2314,
      "num_input_tokens_seen": 14140160,
      "step": 67005
    },
    {
      "epoch": 7.371837183718371,
      "grad_norm": 0.000499725341796875,
      "learning_rate": 0.023876553090408863,
      "loss": 0.2308,
      "num_input_tokens_seen": 14141184,
      "step": 67010
    },
    {
      "epoch": 7.3723872387238725,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.02387539222308921,
      "loss": 0.2324,
      "num_input_tokens_seen": 14142208,
      "step": 67015
    },
    {
      "epoch": 7.372937293729373,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.02387423127396932,
      "loss": 0.2325,
      "num_input_tokens_seen": 14143328,
      "step": 67020
    },
    {
      "epoch": 7.373487348734874,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.023873070243059902,
      "loss": 0.2324,
      "num_input_tokens_seen": 14144384,
      "step": 67025
    },
    {
      "epoch": 7.374037403740374,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.023871909130371655,
      "loss": 0.2319,
      "num_input_tokens_seen": 14145376,
      "step": 67030
    },
    {
      "epoch": 7.374587458745874,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.023870747935915277,
      "loss": 0.233,
      "num_input_tokens_seen": 14146496,
      "step": 67035
    },
    {
      "epoch": 7.375137513751375,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.023869586659701472,
      "loss": 0.2314,
      "num_input_tokens_seen": 14147552,
      "step": 67040
    },
    {
      "epoch": 7.3756875687568755,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02386842530174094,
      "loss": 0.2303,
      "num_input_tokens_seen": 14148576,
      "step": 67045
    },
    {
      "epoch": 7.376237623762377,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02386726386204439,
      "loss": 0.2309,
      "num_input_tokens_seen": 14149632,
      "step": 67050
    },
    {
      "epoch": 7.376787678767877,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.023866102340622523,
      "loss": 0.2303,
      "num_input_tokens_seen": 14150656,
      "step": 67055
    },
    {
      "epoch": 7.377337733773377,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02386494073748605,
      "loss": 0.2319,
      "num_input_tokens_seen": 14151680,
      "step": 67060
    },
    {
      "epoch": 7.377887788778878,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.023863779052645667,
      "loss": 0.2319,
      "num_input_tokens_seen": 14152768,
      "step": 67065
    },
    {
      "epoch": 7.378437843784378,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.023862617286112088,
      "loss": 0.2309,
      "num_input_tokens_seen": 14153824,
      "step": 67070
    },
    {
      "epoch": 7.378987898789879,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.023861455437896018,
      "loss": 0.233,
      "num_input_tokens_seen": 14154912,
      "step": 67075
    },
    {
      "epoch": 7.37953795379538,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023860293508008164,
      "loss": 0.2314,
      "num_input_tokens_seen": 14156064,
      "step": 67080
    },
    {
      "epoch": 7.38008800880088,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.023859131496459237,
      "loss": 0.2298,
      "num_input_tokens_seen": 14157120,
      "step": 67085
    },
    {
      "epoch": 7.380638063806381,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.023857969403259946,
      "loss": 0.2304,
      "num_input_tokens_seen": 14158272,
      "step": 67090
    },
    {
      "epoch": 7.381188118811881,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.023856807228420998,
      "loss": 0.2288,
      "num_input_tokens_seen": 14159360,
      "step": 67095
    },
    {
      "epoch": 7.381738173817382,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.023855644971953114,
      "loss": 0.2278,
      "num_input_tokens_seen": 14160384,
      "step": 67100
    },
    {
      "epoch": 7.382288228822882,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.023854482633866993,
      "loss": 0.2283,
      "num_input_tokens_seen": 14161440,
      "step": 67105
    },
    {
      "epoch": 7.382838283828383,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.02385332021417336,
      "loss": 0.233,
      "num_input_tokens_seen": 14162592,
      "step": 67110
    },
    {
      "epoch": 7.383388338833884,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.02385215771288292,
      "loss": 0.2324,
      "num_input_tokens_seen": 14163616,
      "step": 67115
    },
    {
      "epoch": 7.383938393839384,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.02385099513000639,
      "loss": 0.2314,
      "num_input_tokens_seen": 14164704,
      "step": 67120
    },
    {
      "epoch": 7.384488448844884,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.02384983246555449,
      "loss": 0.2324,
      "num_input_tokens_seen": 14165728,
      "step": 67125
    },
    {
      "epoch": 7.385038503850385,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.023848669719537925,
      "loss": 0.2299,
      "num_input_tokens_seen": 14166752,
      "step": 67130
    },
    {
      "epoch": 7.385588558855885,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.023847506891967418,
      "loss": 0.2293,
      "num_input_tokens_seen": 14167840,
      "step": 67135
    },
    {
      "epoch": 7.3861386138613865,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.023846343982853686,
      "loss": 0.2299,
      "num_input_tokens_seen": 14168928,
      "step": 67140
    },
    {
      "epoch": 7.386688668866887,
      "grad_norm": 0.000782012939453125,
      "learning_rate": 0.023845180992207445,
      "loss": 0.2288,
      "num_input_tokens_seen": 14169984,
      "step": 67145
    },
    {
      "epoch": 7.387238723872387,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.023844017920039418,
      "loss": 0.234,
      "num_input_tokens_seen": 14171008,
      "step": 67150
    },
    {
      "epoch": 7.387788778877888,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.023842854766360323,
      "loss": 0.2346,
      "num_input_tokens_seen": 14172096,
      "step": 67155
    },
    {
      "epoch": 7.388338833883388,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.023841691531180872,
      "loss": 0.2319,
      "num_input_tokens_seen": 14173152,
      "step": 67160
    },
    {
      "epoch": 7.388888888888889,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0238405282145118,
      "loss": 0.2299,
      "num_input_tokens_seen": 14174304,
      "step": 67165
    },
    {
      "epoch": 7.3894389438943895,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02383936481636382,
      "loss": 0.2319,
      "num_input_tokens_seen": 14175328,
      "step": 67170
    },
    {
      "epoch": 7.38998899889989,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.023838201336747654,
      "loss": 0.2293,
      "num_input_tokens_seen": 14176416,
      "step": 67175
    },
    {
      "epoch": 7.390539053905391,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.023837037775674027,
      "loss": 0.2298,
      "num_input_tokens_seen": 14177440,
      "step": 67180
    },
    {
      "epoch": 7.391089108910891,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.02383587413315366,
      "loss": 0.2319,
      "num_input_tokens_seen": 14178496,
      "step": 67185
    },
    {
      "epoch": 7.391639163916391,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02383471040919729,
      "loss": 0.2298,
      "num_input_tokens_seen": 14179552,
      "step": 67190
    },
    {
      "epoch": 7.392189218921892,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.02383354660381563,
      "loss": 0.2351,
      "num_input_tokens_seen": 14180576,
      "step": 67195
    },
    {
      "epoch": 7.3927392739273925,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02383238271701941,
      "loss": 0.2335,
      "num_input_tokens_seen": 14181632,
      "step": 67200
    },
    {
      "epoch": 7.393289328932894,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02383121874881935,
      "loss": 0.233,
      "num_input_tokens_seen": 14182720,
      "step": 67205
    },
    {
      "epoch": 7.393839383938394,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.023830054699226187,
      "loss": 0.234,
      "num_input_tokens_seen": 14183808,
      "step": 67210
    },
    {
      "epoch": 7.394389438943894,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.02382889056825065,
      "loss": 0.2335,
      "num_input_tokens_seen": 14184832,
      "step": 67215
    },
    {
      "epoch": 7.394939493949395,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.023827726355903467,
      "loss": 0.2314,
      "num_input_tokens_seen": 14185920,
      "step": 67220
    },
    {
      "epoch": 7.395489548954895,
      "grad_norm": 0.000682830810546875,
      "learning_rate": 0.023826562062195362,
      "loss": 0.2298,
      "num_input_tokens_seen": 14186976,
      "step": 67225
    },
    {
      "epoch": 7.396039603960396,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.023825397687137075,
      "loss": 0.2304,
      "num_input_tokens_seen": 14188064,
      "step": 67230
    },
    {
      "epoch": 7.396589658965897,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.023824233230739324,
      "loss": 0.2351,
      "num_input_tokens_seen": 14189152,
      "step": 67235
    },
    {
      "epoch": 7.397139713971397,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.023823068693012855,
      "loss": 0.233,
      "num_input_tokens_seen": 14190176,
      "step": 67240
    },
    {
      "epoch": 7.397689768976898,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.023821904073968395,
      "loss": 0.2304,
      "num_input_tokens_seen": 14191264,
      "step": 67245
    },
    {
      "epoch": 7.398239823982398,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.02382073937361668,
      "loss": 0.2299,
      "num_input_tokens_seen": 14192320,
      "step": 67250
    },
    {
      "epoch": 7.398789878987899,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02381957459196844,
      "loss": 0.2309,
      "num_input_tokens_seen": 14193344,
      "step": 67255
    },
    {
      "epoch": 7.399339933993399,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.023818409729034415,
      "loss": 0.2335,
      "num_input_tokens_seen": 14194400,
      "step": 67260
    },
    {
      "epoch": 7.3998899889989,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.023817244784825338,
      "loss": 0.2283,
      "num_input_tokens_seen": 14195456,
      "step": 67265
    },
    {
      "epoch": 7.400440044004401,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.023816079759351943,
      "loss": 0.2325,
      "num_input_tokens_seen": 14196608,
      "step": 67270
    },
    {
      "epoch": 7.400990099009901,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.023814914652624978,
      "loss": 0.2304,
      "num_input_tokens_seen": 14197696,
      "step": 67275
    },
    {
      "epoch": 7.401540154015402,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02381374946465517,
      "loss": 0.2314,
      "num_input_tokens_seen": 14198752,
      "step": 67280
    },
    {
      "epoch": 7.402090209020902,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.023812584195453262,
      "loss": 0.233,
      "num_input_tokens_seen": 14199776,
      "step": 67285
    },
    {
      "epoch": 7.402640264026402,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02381141884503,
      "loss": 0.232,
      "num_input_tokens_seen": 14200800,
      "step": 67290
    },
    {
      "epoch": 7.4031903190319035,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.023810253413396116,
      "loss": 0.2325,
      "num_input_tokens_seen": 14201888,
      "step": 67295
    },
    {
      "epoch": 7.403740374037404,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02380908790056235,
      "loss": 0.232,
      "num_input_tokens_seen": 14202976,
      "step": 67300
    },
    {
      "epoch": 7.404290429042904,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.02380792230653945,
      "loss": 0.2319,
      "num_input_tokens_seen": 14204000,
      "step": 67305
    },
    {
      "epoch": 7.404840484048405,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.023806756631338157,
      "loss": 0.2309,
      "num_input_tokens_seen": 14205056,
      "step": 67310
    },
    {
      "epoch": 7.405390539053905,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023805590874969217,
      "loss": 0.2325,
      "num_input_tokens_seen": 14206112,
      "step": 67315
    },
    {
      "epoch": 7.405940594059406,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.023804425037443365,
      "loss": 0.2304,
      "num_input_tokens_seen": 14207168,
      "step": 67320
    },
    {
      "epoch": 7.4064906490649065,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.023803259118771354,
      "loss": 0.2335,
      "num_input_tokens_seen": 14208192,
      "step": 67325
    },
    {
      "epoch": 7.407040704070407,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02380209311896393,
      "loss": 0.2314,
      "num_input_tokens_seen": 14209248,
      "step": 67330
    },
    {
      "epoch": 7.407590759075908,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.023800927038031834,
      "loss": 0.234,
      "num_input_tokens_seen": 14210368,
      "step": 67335
    },
    {
      "epoch": 7.408140814081408,
      "grad_norm": 0.00058746337890625,
      "learning_rate": 0.023799760875985822,
      "loss": 0.2324,
      "num_input_tokens_seen": 14211360,
      "step": 67340
    },
    {
      "epoch": 7.408690869086909,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.023798594632836637,
      "loss": 0.2309,
      "num_input_tokens_seen": 14212416,
      "step": 67345
    },
    {
      "epoch": 7.409240924092409,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.02379742830859502,
      "loss": 0.2309,
      "num_input_tokens_seen": 14213440,
      "step": 67350
    },
    {
      "epoch": 7.4097909790979095,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.023796261903271734,
      "loss": 0.2309,
      "num_input_tokens_seen": 14214528,
      "step": 67355
    },
    {
      "epoch": 7.410341034103411,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.02379509541687752,
      "loss": 0.2304,
      "num_input_tokens_seen": 14215552,
      "step": 67360
    },
    {
      "epoch": 7.410891089108911,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02379392884942313,
      "loss": 0.2335,
      "num_input_tokens_seen": 14216704,
      "step": 67365
    },
    {
      "epoch": 7.411441144114411,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.023792762200919325,
      "loss": 0.2314,
      "num_input_tokens_seen": 14217760,
      "step": 67370
    },
    {
      "epoch": 7.411991199119912,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.023791595471376846,
      "loss": 0.2304,
      "num_input_tokens_seen": 14218816,
      "step": 67375
    },
    {
      "epoch": 7.412541254125412,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.023790428660806443,
      "loss": 0.2293,
      "num_input_tokens_seen": 14219904,
      "step": 67380
    },
    {
      "epoch": 7.413091309130913,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.023789261769218886,
      "loss": 0.2319,
      "num_input_tokens_seen": 14220992,
      "step": 67385
    },
    {
      "epoch": 7.413641364136414,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02378809479662492,
      "loss": 0.2324,
      "num_input_tokens_seen": 14221984,
      "step": 67390
    },
    {
      "epoch": 7.414191419141914,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0237869277430353,
      "loss": 0.2288,
      "num_input_tokens_seen": 14223008,
      "step": 67395
    },
    {
      "epoch": 7.414741474147415,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.023785760608460778,
      "loss": 0.2314,
      "num_input_tokens_seen": 14224064,
      "step": 67400
    },
    {
      "epoch": 7.415291529152915,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.023784593392912126,
      "loss": 0.2314,
      "num_input_tokens_seen": 14225088,
      "step": 67405
    },
    {
      "epoch": 7.415841584158416,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.023783426096400086,
      "loss": 0.2288,
      "num_input_tokens_seen": 14226112,
      "step": 67410
    },
    {
      "epoch": 7.416391639163916,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02378225871893542,
      "loss": 0.2304,
      "num_input_tokens_seen": 14227136,
      "step": 67415
    },
    {
      "epoch": 7.416941694169417,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.02378109126052889,
      "loss": 0.2319,
      "num_input_tokens_seen": 14228160,
      "step": 67420
    },
    {
      "epoch": 7.417491749174918,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02377992372119126,
      "loss": 0.2314,
      "num_input_tokens_seen": 14229152,
      "step": 67425
    },
    {
      "epoch": 7.418041804180418,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.023778756100933284,
      "loss": 0.2335,
      "num_input_tokens_seen": 14230240,
      "step": 67430
    },
    {
      "epoch": 7.418591859185918,
      "grad_norm": 0.000736236572265625,
      "learning_rate": 0.02377758839976573,
      "loss": 0.2303,
      "num_input_tokens_seen": 14231296,
      "step": 67435
    },
    {
      "epoch": 7.419141914191419,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.02377642061769935,
      "loss": 0.2314,
      "num_input_tokens_seen": 14232384,
      "step": 67440
    },
    {
      "epoch": 7.419691969196919,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.023775252754744913,
      "loss": 0.2309,
      "num_input_tokens_seen": 14233472,
      "step": 67445
    },
    {
      "epoch": 7.4202420242024205,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02377408481091318,
      "loss": 0.2304,
      "num_input_tokens_seen": 14234496,
      "step": 67450
    },
    {
      "epoch": 7.420792079207921,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.023772916786214925,
      "loss": 0.2304,
      "num_input_tokens_seen": 14235680,
      "step": 67455
    },
    {
      "epoch": 7.421342134213421,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023771748680660898,
      "loss": 0.2283,
      "num_input_tokens_seen": 14236672,
      "step": 67460
    },
    {
      "epoch": 7.421892189218922,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.023770580494261877,
      "loss": 0.2309,
      "num_input_tokens_seen": 14237664,
      "step": 67465
    },
    {
      "epoch": 7.422442244224422,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.023769412227028625,
      "loss": 0.2298,
      "num_input_tokens_seen": 14238656,
      "step": 67470
    },
    {
      "epoch": 7.422992299229923,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.023768243878971904,
      "loss": 0.2299,
      "num_input_tokens_seen": 14239712,
      "step": 67475
    },
    {
      "epoch": 7.4235423542354235,
      "grad_norm": 0.000888824462890625,
      "learning_rate": 0.023767075450102492,
      "loss": 0.2288,
      "num_input_tokens_seen": 14240832,
      "step": 67480
    },
    {
      "epoch": 7.424092409240924,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.023765906940431146,
      "loss": 0.2314,
      "num_input_tokens_seen": 14241952,
      "step": 67485
    },
    {
      "epoch": 7.424642464246425,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.023764738349968645,
      "loss": 0.2304,
      "num_input_tokens_seen": 14243072,
      "step": 67490
    },
    {
      "epoch": 7.425192519251925,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.023763569678725754,
      "loss": 0.2294,
      "num_input_tokens_seen": 14244192,
      "step": 67495
    },
    {
      "epoch": 7.425742574257426,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.02376240092671325,
      "loss": 0.2335,
      "num_input_tokens_seen": 14245216,
      "step": 67500
    },
    {
      "epoch": 7.426292629262926,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.023761232093941903,
      "loss": 0.2314,
      "num_input_tokens_seen": 14246272,
      "step": 67505
    },
    {
      "epoch": 7.4268426842684265,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02376006318042248,
      "loss": 0.2315,
      "num_input_tokens_seen": 14247360,
      "step": 67510
    },
    {
      "epoch": 7.427392739273928,
      "grad_norm": 0.0008392333984375,
      "learning_rate": 0.02375889418616576,
      "loss": 0.233,
      "num_input_tokens_seen": 14248384,
      "step": 67515
    },
    {
      "epoch": 7.427942794279428,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02375772511118251,
      "loss": 0.2283,
      "num_input_tokens_seen": 14249440,
      "step": 67520
    },
    {
      "epoch": 7.428492849284929,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.023756555955483517,
      "loss": 0.2288,
      "num_input_tokens_seen": 14250528,
      "step": 67525
    },
    {
      "epoch": 7.429042904290429,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.023755386719079546,
      "loss": 0.2346,
      "num_input_tokens_seen": 14251584,
      "step": 67530
    },
    {
      "epoch": 7.429592959295929,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02375421740198138,
      "loss": 0.2345,
      "num_input_tokens_seen": 14252640,
      "step": 67535
    },
    {
      "epoch": 7.43014301430143,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02375304800419979,
      "loss": 0.2325,
      "num_input_tokens_seen": 14253696,
      "step": 67540
    },
    {
      "epoch": 7.430693069306931,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02375187852574556,
      "loss": 0.2294,
      "num_input_tokens_seen": 14254784,
      "step": 67545
    },
    {
      "epoch": 7.431243124312431,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023750708966629463,
      "loss": 0.2335,
      "num_input_tokens_seen": 14255840,
      "step": 67550
    },
    {
      "epoch": 7.431793179317932,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.023749539326862277,
      "loss": 0.2309,
      "num_input_tokens_seen": 14256896,
      "step": 67555
    },
    {
      "epoch": 7.432343234323432,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02374836960645479,
      "loss": 0.2304,
      "num_input_tokens_seen": 14258016,
      "step": 67560
    },
    {
      "epoch": 7.432893289328933,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.023747199805417778,
      "loss": 0.2309,
      "num_input_tokens_seen": 14259072,
      "step": 67565
    },
    {
      "epoch": 7.433443344334433,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02374602992376202,
      "loss": 0.2324,
      "num_input_tokens_seen": 14260096,
      "step": 67570
    },
    {
      "epoch": 7.433993399339934,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.023744859961498305,
      "loss": 0.2293,
      "num_input_tokens_seen": 14261216,
      "step": 67575
    },
    {
      "epoch": 7.434543454345435,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023743689918637408,
      "loss": 0.233,
      "num_input_tokens_seen": 14262272,
      "step": 67580
    },
    {
      "epoch": 7.435093509350935,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.02374251979519012,
      "loss": 0.2319,
      "num_input_tokens_seen": 14263328,
      "step": 67585
    },
    {
      "epoch": 7.435643564356436,
      "grad_norm": 0.00048828125,
      "learning_rate": 0.023741349591167218,
      "loss": 0.2324,
      "num_input_tokens_seen": 14264352,
      "step": 67590
    },
    {
      "epoch": 7.436193619361936,
      "grad_norm": 0.000518798828125,
      "learning_rate": 0.023740179306579495,
      "loss": 0.2319,
      "num_input_tokens_seen": 14265440,
      "step": 67595
    },
    {
      "epoch": 7.436743674367436,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02373900894143773,
      "loss": 0.2319,
      "num_input_tokens_seen": 14266464,
      "step": 67600
    },
    {
      "epoch": 7.4372937293729375,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023737838495752714,
      "loss": 0.2309,
      "num_input_tokens_seen": 14267520,
      "step": 67605
    },
    {
      "epoch": 7.437843784378438,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.023736667969535233,
      "loss": 0.2309,
      "num_input_tokens_seen": 14268608,
      "step": 67610
    },
    {
      "epoch": 7.438393839383938,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02373549736279608,
      "loss": 0.2324,
      "num_input_tokens_seen": 14269632,
      "step": 67615
    },
    {
      "epoch": 7.438943894389439,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.023734326675546032,
      "loss": 0.2335,
      "num_input_tokens_seen": 14270656,
      "step": 67620
    },
    {
      "epoch": 7.439493949394939,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.023733155907795893,
      "loss": 0.233,
      "num_input_tokens_seen": 14271776,
      "step": 67625
    },
    {
      "epoch": 7.44004400440044,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02373198505955644,
      "loss": 0.2314,
      "num_input_tokens_seen": 14272864,
      "step": 67630
    },
    {
      "epoch": 7.4405940594059405,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02373081413083848,
      "loss": 0.2324,
      "num_input_tokens_seen": 14273952,
      "step": 67635
    },
    {
      "epoch": 7.441144114411441,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02372964312165279,
      "loss": 0.2309,
      "num_input_tokens_seen": 14275040,
      "step": 67640
    },
    {
      "epoch": 7.441694169416942,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02372847203201017,
      "loss": 0.2308,
      "num_input_tokens_seen": 14276096,
      "step": 67645
    },
    {
      "epoch": 7.442244224422442,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.023727300861921406,
      "loss": 0.2314,
      "num_input_tokens_seen": 14277280,
      "step": 67650
    },
    {
      "epoch": 7.442794279427943,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.023726129611397306,
      "loss": 0.2324,
      "num_input_tokens_seen": 14278336,
      "step": 67655
    },
    {
      "epoch": 7.443344334433443,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02372495828044865,
      "loss": 0.2304,
      "num_input_tokens_seen": 14279456,
      "step": 67660
    },
    {
      "epoch": 7.4438943894389435,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02372378686908624,
      "loss": 0.232,
      "num_input_tokens_seen": 14280512,
      "step": 67665
    },
    {
      "epoch": 7.444444444444445,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02372261537732088,
      "loss": 0.2309,
      "num_input_tokens_seen": 14281568,
      "step": 67670
    },
    {
      "epoch": 7.444994499449945,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.02372144380516335,
      "loss": 0.2304,
      "num_input_tokens_seen": 14282688,
      "step": 67675
    },
    {
      "epoch": 7.445544554455446,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.023720272152624466,
      "loss": 0.2314,
      "num_input_tokens_seen": 14283680,
      "step": 67680
    },
    {
      "epoch": 7.446094609460946,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.023719100419715013,
      "loss": 0.233,
      "num_input_tokens_seen": 14284704,
      "step": 67685
    },
    {
      "epoch": 7.446644664466446,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.023717928606445796,
      "loss": 0.233,
      "num_input_tokens_seen": 14285728,
      "step": 67690
    },
    {
      "epoch": 7.447194719471947,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.023716756712827615,
      "loss": 0.2319,
      "num_input_tokens_seen": 14286720,
      "step": 67695
    },
    {
      "epoch": 7.447744774477448,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.023715584738871268,
      "loss": 0.2303,
      "num_input_tokens_seen": 14287712,
      "step": 67700
    },
    {
      "epoch": 7.448294829482949,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.02371441268458756,
      "loss": 0.2299,
      "num_input_tokens_seen": 14288832,
      "step": 67705
    },
    {
      "epoch": 7.448844884488449,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.023713240549987295,
      "loss": 0.2303,
      "num_input_tokens_seen": 14289856,
      "step": 67710
    },
    {
      "epoch": 7.449394939493949,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.023712068335081264,
      "loss": 0.2314,
      "num_input_tokens_seen": 14290912,
      "step": 67715
    },
    {
      "epoch": 7.44994499449945,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023710896039880292,
      "loss": 0.2314,
      "num_input_tokens_seen": 14291968,
      "step": 67720
    },
    {
      "epoch": 7.4504950495049505,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02370972366439516,
      "loss": 0.233,
      "num_input_tokens_seen": 14293056,
      "step": 67725
    },
    {
      "epoch": 7.451045104510451,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023708551208636693,
      "loss": 0.2319,
      "num_input_tokens_seen": 14294080,
      "step": 67730
    },
    {
      "epoch": 7.451595159515952,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.023707378672615683,
      "loss": 0.2314,
      "num_input_tokens_seen": 14295168,
      "step": 67735
    },
    {
      "epoch": 7.452145214521452,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.023706206056342945,
      "loss": 0.2314,
      "num_input_tokens_seen": 14296224,
      "step": 67740
    },
    {
      "epoch": 7.452695269526953,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02370503335982928,
      "loss": 0.2319,
      "num_input_tokens_seen": 14297248,
      "step": 67745
    },
    {
      "epoch": 7.453245324532453,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.0237038605830855,
      "loss": 0.2324,
      "num_input_tokens_seen": 14298368,
      "step": 67750
    },
    {
      "epoch": 7.4537953795379535,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.023702687726122416,
      "loss": 0.2319,
      "num_input_tokens_seen": 14299424,
      "step": 67755
    },
    {
      "epoch": 7.4543454345434546,
      "grad_norm": 0.0004482269287109375,
      "learning_rate": 0.023701514788950835,
      "loss": 0.2324,
      "num_input_tokens_seen": 14300480,
      "step": 67760
    },
    {
      "epoch": 7.454895489548955,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.02370034177158157,
      "loss": 0.2324,
      "num_input_tokens_seen": 14301472,
      "step": 67765
    },
    {
      "epoch": 7.455445544554456,
      "grad_norm": 0.00070953369140625,
      "learning_rate": 0.023699168674025425,
      "loss": 0.2303,
      "num_input_tokens_seen": 14302496,
      "step": 67770
    },
    {
      "epoch": 7.455995599559956,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.023697995496293214,
      "loss": 0.2308,
      "num_input_tokens_seen": 14303520,
      "step": 67775
    },
    {
      "epoch": 7.456545654565456,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.023696822238395757,
      "loss": 0.2298,
      "num_input_tokens_seen": 14304608,
      "step": 67780
    },
    {
      "epoch": 7.457095709570957,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02369564890034386,
      "loss": 0.233,
      "num_input_tokens_seen": 14305664,
      "step": 67785
    },
    {
      "epoch": 7.457645764576458,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.023694475482148345,
      "loss": 0.2314,
      "num_input_tokens_seen": 14306720,
      "step": 67790
    },
    {
      "epoch": 7.458195819581958,
      "grad_norm": 0.000732421875,
      "learning_rate": 0.02369330198382002,
      "loss": 0.2304,
      "num_input_tokens_seen": 14307712,
      "step": 67795
    },
    {
      "epoch": 7.458745874587459,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.0236921284053697,
      "loss": 0.2324,
      "num_input_tokens_seen": 14308768,
      "step": 67800
    },
    {
      "epoch": 7.459295929592959,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.023690954746808202,
      "loss": 0.2304,
      "num_input_tokens_seen": 14309760,
      "step": 67805
    },
    {
      "epoch": 7.45984598459846,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.023689781008146345,
      "loss": 0.2319,
      "num_input_tokens_seen": 14310752,
      "step": 67810
    },
    {
      "epoch": 7.46039603960396,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.023688607189394945,
      "loss": 0.2288,
      "num_input_tokens_seen": 14311872,
      "step": 67815
    },
    {
      "epoch": 7.460946094609461,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.023687433290564826,
      "loss": 0.2309,
      "num_input_tokens_seen": 14312928,
      "step": 67820
    },
    {
      "epoch": 7.461496149614962,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0236862593116668,
      "loss": 0.2335,
      "num_input_tokens_seen": 14314048,
      "step": 67825
    },
    {
      "epoch": 7.462046204620462,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.023685085252711693,
      "loss": 0.2325,
      "num_input_tokens_seen": 14315136,
      "step": 67830
    },
    {
      "epoch": 7.462596259625963,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023683911113710317,
      "loss": 0.234,
      "num_input_tokens_seen": 14316128,
      "step": 67835
    },
    {
      "epoch": 7.463146314631463,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.023682736894673505,
      "loss": 0.2314,
      "num_input_tokens_seen": 14317184,
      "step": 67840
    },
    {
      "epoch": 7.463696369636963,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.023681562595612073,
      "loss": 0.2314,
      "num_input_tokens_seen": 14318208,
      "step": 67845
    },
    {
      "epoch": 7.4642464246424645,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02368038821653684,
      "loss": 0.2319,
      "num_input_tokens_seen": 14319296,
      "step": 67850
    },
    {
      "epoch": 7.464796479647965,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.023679213757458635,
      "loss": 0.2304,
      "num_input_tokens_seen": 14320352,
      "step": 67855
    },
    {
      "epoch": 7.465346534653466,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.023678039218388285,
      "loss": 0.2314,
      "num_input_tokens_seen": 14321440,
      "step": 67860
    },
    {
      "epoch": 7.465896589658966,
      "grad_norm": 0.000499725341796875,
      "learning_rate": 0.023676864599336607,
      "loss": 0.2324,
      "num_input_tokens_seen": 14322464,
      "step": 67865
    },
    {
      "epoch": 7.466446644664466,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.023675689900314434,
      "loss": 0.2314,
      "num_input_tokens_seen": 14323520,
      "step": 67870
    },
    {
      "epoch": 7.466996699669967,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023674515121332593,
      "loss": 0.2303,
      "num_input_tokens_seen": 14324544,
      "step": 67875
    },
    {
      "epoch": 7.4675467546754675,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.023673340262401903,
      "loss": 0.2309,
      "num_input_tokens_seen": 14325600,
      "step": 67880
    },
    {
      "epoch": 7.468096809680969,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.023672165323533202,
      "loss": 0.2314,
      "num_input_tokens_seen": 14326720,
      "step": 67885
    },
    {
      "epoch": 7.468646864686469,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.023670990304737318,
      "loss": 0.2329,
      "num_input_tokens_seen": 14327808,
      "step": 67890
    },
    {
      "epoch": 7.469196919691969,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02366981520602507,
      "loss": 0.2314,
      "num_input_tokens_seen": 14328832,
      "step": 67895
    },
    {
      "epoch": 7.46974697469747,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023668640027407296,
      "loss": 0.233,
      "num_input_tokens_seen": 14329952,
      "step": 67900
    },
    {
      "epoch": 7.47029702970297,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02366746476889483,
      "loss": 0.2314,
      "num_input_tokens_seen": 14331040,
      "step": 67905
    },
    {
      "epoch": 7.4708470847084705,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.023666289430498498,
      "loss": 0.2314,
      "num_input_tokens_seen": 14332064,
      "step": 67910
    },
    {
      "epoch": 7.471397139713972,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.023665114012229135,
      "loss": 0.2309,
      "num_input_tokens_seen": 14333120,
      "step": 67915
    },
    {
      "epoch": 7.471947194719472,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.023663938514097576,
      "loss": 0.2309,
      "num_input_tokens_seen": 14334144,
      "step": 67920
    },
    {
      "epoch": 7.472497249724973,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.023662762936114648,
      "loss": 0.233,
      "num_input_tokens_seen": 14335232,
      "step": 67925
    },
    {
      "epoch": 7.473047304730473,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.023661587278291197,
      "loss": 0.2314,
      "num_input_tokens_seen": 14336288,
      "step": 67930
    },
    {
      "epoch": 7.473597359735973,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023660411540638047,
      "loss": 0.2309,
      "num_input_tokens_seen": 14337344,
      "step": 67935
    },
    {
      "epoch": 7.474147414741474,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023659235723166045,
      "loss": 0.2304,
      "num_input_tokens_seen": 14338336,
      "step": 67940
    },
    {
      "epoch": 7.474697469746975,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.023658059825886018,
      "loss": 0.2309,
      "num_input_tokens_seen": 14339360,
      "step": 67945
    },
    {
      "epoch": 7.475247524752476,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.023656883848808808,
      "loss": 0.2304,
      "num_input_tokens_seen": 14340384,
      "step": 67950
    },
    {
      "epoch": 7.475797579757976,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.023655707791945257,
      "loss": 0.2298,
      "num_input_tokens_seen": 14341408,
      "step": 67955
    },
    {
      "epoch": 7.476347634763476,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.023654531655306198,
      "loss": 0.2303,
      "num_input_tokens_seen": 14342464,
      "step": 67960
    },
    {
      "epoch": 7.476897689768977,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.023653355438902478,
      "loss": 0.2309,
      "num_input_tokens_seen": 14343520,
      "step": 67965
    },
    {
      "epoch": 7.477447744774477,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.023652179142744922,
      "loss": 0.2314,
      "num_input_tokens_seen": 14344640,
      "step": 67970
    },
    {
      "epoch": 7.477997799779978,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.02365100276684439,
      "loss": 0.2351,
      "num_input_tokens_seen": 14345664,
      "step": 67975
    },
    {
      "epoch": 7.478547854785479,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.023649826311211722,
      "loss": 0.2309,
      "num_input_tokens_seen": 14346720,
      "step": 67980
    },
    {
      "epoch": 7.479097909790979,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.023648649775857754,
      "loss": 0.2319,
      "num_input_tokens_seen": 14347808,
      "step": 67985
    },
    {
      "epoch": 7.47964796479648,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.023647473160793322,
      "loss": 0.2298,
      "num_input_tokens_seen": 14348896,
      "step": 67990
    },
    {
      "epoch": 7.48019801980198,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.023646296466029285,
      "loss": 0.2293,
      "num_input_tokens_seen": 14350016,
      "step": 67995
    },
    {
      "epoch": 7.48074807480748,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.023645119691576483,
      "loss": 0.23,
      "num_input_tokens_seen": 14351008,
      "step": 68000
    },
    {
      "epoch": 7.4812981298129815,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023643942837445763,
      "loss": 0.2341,
      "num_input_tokens_seen": 14352096,
      "step": 68005
    },
    {
      "epoch": 7.481848184818482,
      "grad_norm": 0.0005035400390625,
      "learning_rate": 0.023642765903647965,
      "loss": 0.2299,
      "num_input_tokens_seen": 14353152,
      "step": 68010
    },
    {
      "epoch": 7.482398239823983,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02364158889019394,
      "loss": 0.2325,
      "num_input_tokens_seen": 14354240,
      "step": 68015
    },
    {
      "epoch": 7.482948294829483,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02364041179709454,
      "loss": 0.2336,
      "num_input_tokens_seen": 14355328,
      "step": 68020
    },
    {
      "epoch": 7.483498349834983,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02363923462436061,
      "loss": 0.2325,
      "num_input_tokens_seen": 14356416,
      "step": 68025
    },
    {
      "epoch": 7.484048404840484,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.023638057372003002,
      "loss": 0.231,
      "num_input_tokens_seen": 14357472,
      "step": 68030
    },
    {
      "epoch": 7.4845984598459845,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02363688004003256,
      "loss": 0.2299,
      "num_input_tokens_seen": 14358528,
      "step": 68035
    },
    {
      "epoch": 7.485148514851485,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.023635702628460143,
      "loss": 0.231,
      "num_input_tokens_seen": 14359552,
      "step": 68040
    },
    {
      "epoch": 7.485698569856986,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.023634525137296594,
      "loss": 0.2289,
      "num_input_tokens_seen": 14360672,
      "step": 68045
    },
    {
      "epoch": 7.486248624862486,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.023633347566552776,
      "loss": 0.2331,
      "num_input_tokens_seen": 14361696,
      "step": 68050
    },
    {
      "epoch": 7.486798679867987,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.023632169916239534,
      "loss": 0.2367,
      "num_input_tokens_seen": 14362752,
      "step": 68055
    },
    {
      "epoch": 7.487348734873487,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.02363099218636772,
      "loss": 0.231,
      "num_input_tokens_seen": 14363840,
      "step": 68060
    },
    {
      "epoch": 7.4878987898789875,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023629814376948197,
      "loss": 0.2284,
      "num_input_tokens_seen": 14364864,
      "step": 68065
    },
    {
      "epoch": 7.488448844884489,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.023628636487991815,
      "loss": 0.2352,
      "num_input_tokens_seen": 14365920,
      "step": 68070
    },
    {
      "epoch": 7.488998899889989,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.023627458519509432,
      "loss": 0.2309,
      "num_input_tokens_seen": 14366944,
      "step": 68075
    },
    {
      "epoch": 7.48954895489549,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.023626280471511902,
      "loss": 0.2289,
      "num_input_tokens_seen": 14367968,
      "step": 68080
    },
    {
      "epoch": 7.49009900990099,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.023625102344010083,
      "loss": 0.233,
      "num_input_tokens_seen": 14369056,
      "step": 68085
    },
    {
      "epoch": 7.49064906490649,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.023623924137014835,
      "loss": 0.2346,
      "num_input_tokens_seen": 14370080,
      "step": 68090
    },
    {
      "epoch": 7.491199119911991,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02362274585053702,
      "loss": 0.234,
      "num_input_tokens_seen": 14371136,
      "step": 68095
    },
    {
      "epoch": 7.491749174917492,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02362156748458749,
      "loss": 0.2288,
      "num_input_tokens_seen": 14372224,
      "step": 68100
    },
    {
      "epoch": 7.492299229922993,
      "grad_norm": 0.009765625,
      "learning_rate": 0.023620389039177108,
      "loss": 0.2263,
      "num_input_tokens_seen": 14373312,
      "step": 68105
    },
    {
      "epoch": 7.492849284928493,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02361921051431674,
      "loss": 0.2357,
      "num_input_tokens_seen": 14374368,
      "step": 68110
    },
    {
      "epoch": 7.493399339933993,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.023618031910017243,
      "loss": 0.2315,
      "num_input_tokens_seen": 14375360,
      "step": 68115
    },
    {
      "epoch": 7.493949394939494,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.023616853226289484,
      "loss": 0.2335,
      "num_input_tokens_seen": 14376448,
      "step": 68120
    },
    {
      "epoch": 7.494499449944994,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.023615674463144325,
      "loss": 0.2305,
      "num_input_tokens_seen": 14377440,
      "step": 68125
    },
    {
      "epoch": 7.4950495049504955,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02361449562059262,
      "loss": 0.2304,
      "num_input_tokens_seen": 14378464,
      "step": 68130
    },
    {
      "epoch": 7.495599559955996,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.02361331669864525,
      "loss": 0.2284,
      "num_input_tokens_seen": 14379488,
      "step": 68135
    },
    {
      "epoch": 7.496149614961496,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02361213769731307,
      "loss": 0.2315,
      "num_input_tokens_seen": 14380576,
      "step": 68140
    },
    {
      "epoch": 7.496699669966997,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.023610958616606954,
      "loss": 0.233,
      "num_input_tokens_seen": 14381696,
      "step": 68145
    },
    {
      "epoch": 7.497249724972497,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.023609779456537763,
      "loss": 0.2326,
      "num_input_tokens_seen": 14382784,
      "step": 68150
    },
    {
      "epoch": 7.497799779977997,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.023608600217116364,
      "loss": 0.2305,
      "num_input_tokens_seen": 14383840,
      "step": 68155
    },
    {
      "epoch": 7.4983498349834985,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.023607420898353628,
      "loss": 0.2315,
      "num_input_tokens_seen": 14384864,
      "step": 68160
    },
    {
      "epoch": 7.498899889988999,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02360624150026042,
      "loss": 0.232,
      "num_input_tokens_seen": 14385984,
      "step": 68165
    },
    {
      "epoch": 7.4994499449945,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02360506202284762,
      "loss": 0.232,
      "num_input_tokens_seen": 14387104,
      "step": 68170
    },
    {
      "epoch": 7.5,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.02360388246612609,
      "loss": 0.234,
      "num_input_tokens_seen": 14388192,
      "step": 68175
    },
    {
      "epoch": 7.5005500550055,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.023602702830106702,
      "loss": 0.233,
      "num_input_tokens_seen": 14389280,
      "step": 68180
    },
    {
      "epoch": 7.501100110011001,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.023601523114800328,
      "loss": 0.2304,
      "num_input_tokens_seen": 14390368,
      "step": 68185
    },
    {
      "epoch": 7.5016501650165015,
      "grad_norm": 0.000728607177734375,
      "learning_rate": 0.023600343320217848,
      "loss": 0.233,
      "num_input_tokens_seen": 14391392,
      "step": 68190
    },
    {
      "epoch": 7.502200220022003,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02359916344637013,
      "loss": 0.2309,
      "num_input_tokens_seen": 14392480,
      "step": 68195
    },
    {
      "epoch": 7.502750275027503,
      "grad_norm": 0.0004825592041015625,
      "learning_rate": 0.02359798349326805,
      "loss": 0.233,
      "num_input_tokens_seen": 14393568,
      "step": 68200
    },
    {
      "epoch": 7.503300330033003,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.023596803460922477,
      "loss": 0.2309,
      "num_input_tokens_seen": 14394560,
      "step": 68205
    },
    {
      "epoch": 7.503850385038504,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.023595623349344293,
      "loss": 0.234,
      "num_input_tokens_seen": 14395552,
      "step": 68210
    },
    {
      "epoch": 7.504400440044004,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.023594443158544372,
      "loss": 0.2356,
      "num_input_tokens_seen": 14396640,
      "step": 68215
    },
    {
      "epoch": 7.5049504950495045,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.0235932628885336,
      "loss": 0.233,
      "num_input_tokens_seen": 14397696,
      "step": 68220
    },
    {
      "epoch": 7.505500550055006,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.023592082539322844,
      "loss": 0.2288,
      "num_input_tokens_seen": 14398752,
      "step": 68225
    },
    {
      "epoch": 7.506050605060506,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023590902110922983,
      "loss": 0.2324,
      "num_input_tokens_seen": 14399840,
      "step": 68230
    },
    {
      "epoch": 7.506600660066007,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.023589721603344896,
      "loss": 0.2325,
      "num_input_tokens_seen": 14400864,
      "step": 68235
    },
    {
      "epoch": 7.507150715071507,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.023588541016599473,
      "loss": 0.2304,
      "num_input_tokens_seen": 14401856,
      "step": 68240
    },
    {
      "epoch": 7.507700770077007,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.023587360350697587,
      "loss": 0.234,
      "num_input_tokens_seen": 14402848,
      "step": 68245
    },
    {
      "epoch": 7.508250825082508,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023586179605650123,
      "loss": 0.233,
      "num_input_tokens_seen": 14403936,
      "step": 68250
    },
    {
      "epoch": 7.508800880088009,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02358499878146796,
      "loss": 0.2319,
      "num_input_tokens_seen": 14404928,
      "step": 68255
    },
    {
      "epoch": 7.50935093509351,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.023583817878161984,
      "loss": 0.233,
      "num_input_tokens_seen": 14405984,
      "step": 68260
    },
    {
      "epoch": 7.50990099009901,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.023582636895743074,
      "loss": 0.2293,
      "num_input_tokens_seen": 14407072,
      "step": 68265
    },
    {
      "epoch": 7.51045104510451,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.023581455834222123,
      "loss": 0.2324,
      "num_input_tokens_seen": 14408128,
      "step": 68270
    },
    {
      "epoch": 7.511001100110011,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.023580274693610005,
      "loss": 0.2324,
      "num_input_tokens_seen": 14409216,
      "step": 68275
    },
    {
      "epoch": 7.511551155115511,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.023579093473917616,
      "loss": 0.2298,
      "num_input_tokens_seen": 14410240,
      "step": 68280
    },
    {
      "epoch": 7.512101210121012,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.02357791217515584,
      "loss": 0.2303,
      "num_input_tokens_seen": 14411264,
      "step": 68285
    },
    {
      "epoch": 7.512651265126513,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.023576730797335563,
      "loss": 0.2314,
      "num_input_tokens_seen": 14412352,
      "step": 68290
    },
    {
      "epoch": 7.513201320132013,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.023575549340467676,
      "loss": 0.2319,
      "num_input_tokens_seen": 14413376,
      "step": 68295
    },
    {
      "epoch": 7.513751375137514,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02357436780456306,
      "loss": 0.2314,
      "num_input_tokens_seen": 14414432,
      "step": 68300
    },
    {
      "epoch": 7.514301430143014,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.02357318618963261,
      "loss": 0.2324,
      "num_input_tokens_seen": 14415488,
      "step": 68305
    },
    {
      "epoch": 7.514851485148515,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.023572004495687224,
      "loss": 0.2314,
      "num_input_tokens_seen": 14416544,
      "step": 68310
    },
    {
      "epoch": 7.5154015401540155,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.023570822722737785,
      "loss": 0.2319,
      "num_input_tokens_seen": 14417600,
      "step": 68315
    },
    {
      "epoch": 7.515951595159516,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.023569640870795182,
      "loss": 0.2329,
      "num_input_tokens_seen": 14418688,
      "step": 68320
    },
    {
      "epoch": 7.516501650165017,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.023568458939870307,
      "loss": 0.2324,
      "num_input_tokens_seen": 14419712,
      "step": 68325
    },
    {
      "epoch": 7.517051705170517,
      "grad_norm": 0.009765625,
      "learning_rate": 0.023567276929974068,
      "loss": 0.2303,
      "num_input_tokens_seen": 14420736,
      "step": 68330
    },
    {
      "epoch": 7.517601760176017,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02356609484111734,
      "loss": 0.2298,
      "num_input_tokens_seen": 14421760,
      "step": 68335
    },
    {
      "epoch": 7.518151815181518,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023564912673311032,
      "loss": 0.2293,
      "num_input_tokens_seen": 14422816,
      "step": 68340
    },
    {
      "epoch": 7.5187018701870185,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.023563730426566027,
      "loss": 0.2308,
      "num_input_tokens_seen": 14423840,
      "step": 68345
    },
    {
      "epoch": 7.51925192519252,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.023562548100893234,
      "loss": 0.2345,
      "num_input_tokens_seen": 14424864,
      "step": 68350
    },
    {
      "epoch": 7.51980198019802,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02356136569630354,
      "loss": 0.2324,
      "num_input_tokens_seen": 14425888,
      "step": 68355
    },
    {
      "epoch": 7.52035203520352,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02356018321280785,
      "loss": 0.2329,
      "num_input_tokens_seen": 14426880,
      "step": 68360
    },
    {
      "epoch": 7.520902090209021,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.023559000650417057,
      "loss": 0.2309,
      "num_input_tokens_seen": 14427872,
      "step": 68365
    },
    {
      "epoch": 7.521452145214521,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.023557818009142063,
      "loss": 0.2319,
      "num_input_tokens_seen": 14428896,
      "step": 68370
    },
    {
      "epoch": 7.522002200220022,
      "grad_norm": 0.009765625,
      "learning_rate": 0.023556635288993762,
      "loss": 0.2325,
      "num_input_tokens_seen": 14429952,
      "step": 68375
    },
    {
      "epoch": 7.522552255225523,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.023555452489983066,
      "loss": 0.2298,
      "num_input_tokens_seen": 14431008,
      "step": 68380
    },
    {
      "epoch": 7.523102310231023,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023554269612120864,
      "loss": 0.2324,
      "num_input_tokens_seen": 14432096,
      "step": 68385
    },
    {
      "epoch": 7.523652365236524,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02355308665541807,
      "loss": 0.2319,
      "num_input_tokens_seen": 14433152,
      "step": 68390
    },
    {
      "epoch": 7.524202420242024,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.023551903619885576,
      "loss": 0.2303,
      "num_input_tokens_seen": 14434240,
      "step": 68395
    },
    {
      "epoch": 7.524752475247524,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023550720505534292,
      "loss": 0.2319,
      "num_input_tokens_seen": 14435232,
      "step": 68400
    },
    {
      "epoch": 7.525302530253025,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02354953731237512,
      "loss": 0.2314,
      "num_input_tokens_seen": 14436256,
      "step": 68405
    },
    {
      "epoch": 7.525852585258526,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.023548354040418964,
      "loss": 0.2293,
      "num_input_tokens_seen": 14437280,
      "step": 68410
    },
    {
      "epoch": 7.526402640264027,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02354717068967673,
      "loss": 0.2308,
      "num_input_tokens_seen": 14438304,
      "step": 68415
    },
    {
      "epoch": 7.526952695269527,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02354598726015933,
      "loss": 0.2329,
      "num_input_tokens_seen": 14439392,
      "step": 68420
    },
    {
      "epoch": 7.527502750275027,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.023544803751877657,
      "loss": 0.2309,
      "num_input_tokens_seen": 14440448,
      "step": 68425
    },
    {
      "epoch": 7.528052805280528,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.023543620164842637,
      "loss": 0.2319,
      "num_input_tokens_seen": 14441472,
      "step": 68430
    },
    {
      "epoch": 7.528602860286028,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.023542436499065168,
      "loss": 0.2278,
      "num_input_tokens_seen": 14442464,
      "step": 68435
    },
    {
      "epoch": 7.5291529152915295,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.023541252754556156,
      "loss": 0.2293,
      "num_input_tokens_seen": 14443488,
      "step": 68440
    },
    {
      "epoch": 7.52970297029703,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.023540068931326517,
      "loss": 0.2324,
      "num_input_tokens_seen": 14444576,
      "step": 68445
    },
    {
      "epoch": 7.53025302530253,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.023538885029387165,
      "loss": 0.2314,
      "num_input_tokens_seen": 14445568,
      "step": 68450
    },
    {
      "epoch": 7.530803080308031,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.023537701048749003,
      "loss": 0.2319,
      "num_input_tokens_seen": 14446656,
      "step": 68455
    },
    {
      "epoch": 7.531353135313531,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02353651698942295,
      "loss": 0.2314,
      "num_input_tokens_seen": 14447712,
      "step": 68460
    },
    {
      "epoch": 7.531903190319031,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.023535332851419916,
      "loss": 0.2272,
      "num_input_tokens_seen": 14448768,
      "step": 68465
    },
    {
      "epoch": 7.5324532453245325,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.023534148634750812,
      "loss": 0.2299,
      "num_input_tokens_seen": 14449792,
      "step": 68470
    },
    {
      "epoch": 7.533003300330033,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.023532964339426553,
      "loss": 0.2299,
      "num_input_tokens_seen": 14450880,
      "step": 68475
    },
    {
      "epoch": 7.533553355335534,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.023531779965458067,
      "loss": 0.2314,
      "num_input_tokens_seen": 14452000,
      "step": 68480
    },
    {
      "epoch": 7.534103410341034,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.023530595512856245,
      "loss": 0.2299,
      "num_input_tokens_seen": 14453056,
      "step": 68485
    },
    {
      "epoch": 7.534653465346535,
      "grad_norm": 0.000522613525390625,
      "learning_rate": 0.023529410981632026,
      "loss": 0.2294,
      "num_input_tokens_seen": 14454080,
      "step": 68490
    },
    {
      "epoch": 7.535203520352035,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02352822637179632,
      "loss": 0.2284,
      "num_input_tokens_seen": 14455072,
      "step": 68495
    },
    {
      "epoch": 7.5357535753575355,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.023527041683360037,
      "loss": 0.2341,
      "num_input_tokens_seen": 14456064,
      "step": 68500
    },
    {
      "epoch": 7.536303630363037,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02352585691633411,
      "loss": 0.2326,
      "num_input_tokens_seen": 14457120,
      "step": 68505
    },
    {
      "epoch": 7.536853685368537,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.023524672070729443,
      "loss": 0.2331,
      "num_input_tokens_seen": 14458208,
      "step": 68510
    },
    {
      "epoch": 7.537403740374037,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.02352348714655697,
      "loss": 0.2325,
      "num_input_tokens_seen": 14459296,
      "step": 68515
    },
    {
      "epoch": 7.537953795379538,
      "grad_norm": 0.005859375,
      "learning_rate": 0.023522302143827604,
      "loss": 0.2356,
      "num_input_tokens_seen": 14460384,
      "step": 68520
    },
    {
      "epoch": 7.538503850385038,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02352111706255227,
      "loss": 0.2315,
      "num_input_tokens_seen": 14461440,
      "step": 68525
    },
    {
      "epoch": 7.539053905390539,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02351993190274189,
      "loss": 0.2346,
      "num_input_tokens_seen": 14462560,
      "step": 68530
    },
    {
      "epoch": 7.53960396039604,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.023518746664407383,
      "loss": 0.2345,
      "num_input_tokens_seen": 14463616,
      "step": 68535
    },
    {
      "epoch": 7.54015401540154,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.02351756134755967,
      "loss": 0.2283,
      "num_input_tokens_seen": 14464640,
      "step": 68540
    },
    {
      "epoch": 7.540704070407041,
      "grad_norm": 0.0045166015625,
      "learning_rate": 0.023516375952209692,
      "loss": 0.2325,
      "num_input_tokens_seen": 14465728,
      "step": 68545
    },
    {
      "epoch": 7.541254125412541,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.023515190478368357,
      "loss": 0.234,
      "num_input_tokens_seen": 14466784,
      "step": 68550
    },
    {
      "epoch": 7.541804180418042,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.0235140049260466,
      "loss": 0.2283,
      "num_input_tokens_seen": 14467840,
      "step": 68555
    },
    {
      "epoch": 7.542354235423542,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023512819295255345,
      "loss": 0.2324,
      "num_input_tokens_seen": 14468992,
      "step": 68560
    },
    {
      "epoch": 7.542904290429043,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023511633586005515,
      "loss": 0.2319,
      "num_input_tokens_seen": 14469984,
      "step": 68565
    },
    {
      "epoch": 7.543454345434544,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02351044779830805,
      "loss": 0.2304,
      "num_input_tokens_seen": 14471072,
      "step": 68570
    },
    {
      "epoch": 7.544004400440044,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.023509261932173866,
      "loss": 0.2304,
      "num_input_tokens_seen": 14472032,
      "step": 68575
    },
    {
      "epoch": 7.544554455445544,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.023508075987613904,
      "loss": 0.2324,
      "num_input_tokens_seen": 14473056,
      "step": 68580
    },
    {
      "epoch": 7.545104510451045,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.02350688996463908,
      "loss": 0.2309,
      "num_input_tokens_seen": 14474144,
      "step": 68585
    },
    {
      "epoch": 7.5456545654565454,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.023505703863260334,
      "loss": 0.2304,
      "num_input_tokens_seen": 14475200,
      "step": 68590
    },
    {
      "epoch": 7.5462046204620465,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.023504517683488598,
      "loss": 0.2309,
      "num_input_tokens_seen": 14476224,
      "step": 68595
    },
    {
      "epoch": 7.546754675467547,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.02350333142533481,
      "loss": 0.2319,
      "num_input_tokens_seen": 14477344,
      "step": 68600
    },
    {
      "epoch": 7.547304730473047,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02350214508880989,
      "loss": 0.2308,
      "num_input_tokens_seen": 14478400,
      "step": 68605
    },
    {
      "epoch": 7.547854785478548,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.02350095867392478,
      "loss": 0.2304,
      "num_input_tokens_seen": 14479456,
      "step": 68610
    },
    {
      "epoch": 7.548404840484048,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.023499772180690413,
      "loss": 0.2293,
      "num_input_tokens_seen": 14480448,
      "step": 68615
    },
    {
      "epoch": 7.548954895489549,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.023498585609117727,
      "loss": 0.2309,
      "num_input_tokens_seen": 14481472,
      "step": 68620
    },
    {
      "epoch": 7.5495049504950495,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.023497398959217652,
      "loss": 0.2298,
      "num_input_tokens_seen": 14482496,
      "step": 68625
    },
    {
      "epoch": 7.55005500550055,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02349621223100113,
      "loss": 0.232,
      "num_input_tokens_seen": 14483648,
      "step": 68630
    },
    {
      "epoch": 7.550605060506051,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.02349502542447909,
      "loss": 0.2283,
      "num_input_tokens_seen": 14484736,
      "step": 68635
    },
    {
      "epoch": 7.551155115511551,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.023493838539662488,
      "loss": 0.233,
      "num_input_tokens_seen": 14485792,
      "step": 68640
    },
    {
      "epoch": 7.551705170517051,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02349265157656225,
      "loss": 0.2293,
      "num_input_tokens_seen": 14486880,
      "step": 68645
    },
    {
      "epoch": 7.552255225522552,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.023491464535189313,
      "loss": 0.233,
      "num_input_tokens_seen": 14487936,
      "step": 68650
    },
    {
      "epoch": 7.552805280528053,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02349027741555462,
      "loss": 0.2288,
      "num_input_tokens_seen": 14489024,
      "step": 68655
    },
    {
      "epoch": 7.553355335533554,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.02348909021766912,
      "loss": 0.232,
      "num_input_tokens_seen": 14490016,
      "step": 68660
    },
    {
      "epoch": 7.553905390539054,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.023487902941543744,
      "loss": 0.2309,
      "num_input_tokens_seen": 14491104,
      "step": 68665
    },
    {
      "epoch": 7.554455445544555,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02348671558718944,
      "loss": 0.2325,
      "num_input_tokens_seen": 14492224,
      "step": 68670
    },
    {
      "epoch": 7.555005500550055,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.023485528154617158,
      "loss": 0.2283,
      "num_input_tokens_seen": 14493248,
      "step": 68675
    },
    {
      "epoch": 7.555555555555555,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02348434064383783,
      "loss": 0.2283,
      "num_input_tokens_seen": 14494336,
      "step": 68680
    },
    {
      "epoch": 7.5561056105610565,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.023483153054862402,
      "loss": 0.2273,
      "num_input_tokens_seen": 14495360,
      "step": 68685
    },
    {
      "epoch": 7.556655665566557,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.023481965387701823,
      "loss": 0.2212,
      "num_input_tokens_seen": 14496384,
      "step": 68690
    },
    {
      "epoch": 7.557205720572057,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.023480777642367046,
      "loss": 0.2277,
      "num_input_tokens_seen": 14497472,
      "step": 68695
    },
    {
      "epoch": 7.557755775577558,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.023479589818869002,
      "loss": 0.245,
      "num_input_tokens_seen": 14498528,
      "step": 68700
    },
    {
      "epoch": 7.558305830583058,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.023478401917218655,
      "loss": 0.2393,
      "num_input_tokens_seen": 14499552,
      "step": 68705
    },
    {
      "epoch": 7.558855885588558,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.023477213937426942,
      "loss": 0.2334,
      "num_input_tokens_seen": 14500576,
      "step": 68710
    },
    {
      "epoch": 7.5594059405940595,
      "grad_norm": 0.004547119140625,
      "learning_rate": 0.023476025879504817,
      "loss": 0.2297,
      "num_input_tokens_seen": 14501600,
      "step": 68715
    },
    {
      "epoch": 7.55995599559956,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.02347483774346323,
      "loss": 0.2281,
      "num_input_tokens_seen": 14502688,
      "step": 68720
    },
    {
      "epoch": 7.560506050605061,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.02347364952931313,
      "loss": 0.2317,
      "num_input_tokens_seen": 14503712,
      "step": 68725
    },
    {
      "epoch": 7.561056105610561,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.023472461237065468,
      "loss": 0.2302,
      "num_input_tokens_seen": 14504704,
      "step": 68730
    },
    {
      "epoch": 7.561606160616062,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.023471272866731195,
      "loss": 0.2254,
      "num_input_tokens_seen": 14505728,
      "step": 68735
    },
    {
      "epoch": 7.562156215621562,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.023470084418321265,
      "loss": 0.2261,
      "num_input_tokens_seen": 14506848,
      "step": 68740
    },
    {
      "epoch": 7.5627062706270625,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02346889589184663,
      "loss": 0.2428,
      "num_input_tokens_seen": 14507872,
      "step": 68745
    },
    {
      "epoch": 7.563256325632564,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.023467707287318246,
      "loss": 0.2261,
      "num_input_tokens_seen": 14508864,
      "step": 68750
    },
    {
      "epoch": 7.563806380638064,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02346651860474707,
      "loss": 0.2329,
      "num_input_tokens_seen": 14509920,
      "step": 68755
    },
    {
      "epoch": 7.564356435643564,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.02346532984414405,
      "loss": 0.226,
      "num_input_tokens_seen": 14510880,
      "step": 68760
    },
    {
      "epoch": 7.564906490649065,
      "grad_norm": 0.004547119140625,
      "learning_rate": 0.023464141005520155,
      "loss": 0.234,
      "num_input_tokens_seen": 14511904,
      "step": 68765
    },
    {
      "epoch": 7.565456545654565,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.023462952088886328,
      "loss": 0.2308,
      "num_input_tokens_seen": 14512992,
      "step": 68770
    },
    {
      "epoch": 7.566006600660066,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.023461763094253532,
      "loss": 0.2308,
      "num_input_tokens_seen": 14514112,
      "step": 68775
    },
    {
      "epoch": 7.566556655665567,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.023460574021632728,
      "loss": 0.2319,
      "num_input_tokens_seen": 14515136,
      "step": 68780
    },
    {
      "epoch": 7.567106710671067,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.02345938487103487,
      "loss": 0.2287,
      "num_input_tokens_seen": 14516256,
      "step": 68785
    },
    {
      "epoch": 7.567656765676568,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.023458195642470925,
      "loss": 0.2282,
      "num_input_tokens_seen": 14517280,
      "step": 68790
    },
    {
      "epoch": 7.568206820682068,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.023457006335951845,
      "loss": 0.2355,
      "num_input_tokens_seen": 14518272,
      "step": 68795
    },
    {
      "epoch": 7.568756875687569,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.023455816951488597,
      "loss": 0.2281,
      "num_input_tokens_seen": 14519328,
      "step": 68800
    },
    {
      "epoch": 7.569306930693069,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02345462748909214,
      "loss": 0.2329,
      "num_input_tokens_seen": 14520352,
      "step": 68805
    },
    {
      "epoch": 7.56985698569857,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.023453437948773446,
      "loss": 0.2297,
      "num_input_tokens_seen": 14521408,
      "step": 68810
    },
    {
      "epoch": 7.570407040704071,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.023452248330543463,
      "loss": 0.234,
      "num_input_tokens_seen": 14522496,
      "step": 68815
    },
    {
      "epoch": 7.570957095709571,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.023451058634413165,
      "loss": 0.2324,
      "num_input_tokens_seen": 14523520,
      "step": 68820
    },
    {
      "epoch": 7.571507150715071,
      "grad_norm": 0.004547119140625,
      "learning_rate": 0.023449868860393512,
      "loss": 0.2251,
      "num_input_tokens_seen": 14524576,
      "step": 68825
    },
    {
      "epoch": 7.572057205720572,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.02344867900849548,
      "loss": 0.2324,
      "num_input_tokens_seen": 14525568,
      "step": 68830
    },
    {
      "epoch": 7.572607260726072,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.023447489078730024,
      "loss": 0.2313,
      "num_input_tokens_seen": 14526560,
      "step": 68835
    },
    {
      "epoch": 7.5731573157315735,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.023446299071108113,
      "loss": 0.2299,
      "num_input_tokens_seen": 14527648,
      "step": 68840
    },
    {
      "epoch": 7.573707370737074,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.023445108985640718,
      "loss": 0.2328,
      "num_input_tokens_seen": 14528704,
      "step": 68845
    },
    {
      "epoch": 7.574257425742574,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.023443918822338806,
      "loss": 0.2325,
      "num_input_tokens_seen": 14529792,
      "step": 68850
    },
    {
      "epoch": 7.574807480748075,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.023442728581213344,
      "loss": 0.2308,
      "num_input_tokens_seen": 14530784,
      "step": 68855
    },
    {
      "epoch": 7.575357535753575,
      "grad_norm": 0.005859375,
      "learning_rate": 0.023441538262275307,
      "loss": 0.2423,
      "num_input_tokens_seen": 14531808,
      "step": 68860
    },
    {
      "epoch": 7.575907590759076,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.02344034786553566,
      "loss": 0.2313,
      "num_input_tokens_seen": 14532864,
      "step": 68865
    },
    {
      "epoch": 7.5764576457645765,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.023439157391005376,
      "loss": 0.2369,
      "num_input_tokens_seen": 14534016,
      "step": 68870
    },
    {
      "epoch": 7.577007700770077,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.023437966838695427,
      "loss": 0.2348,
      "num_input_tokens_seen": 14534976,
      "step": 68875
    },
    {
      "epoch": 7.577557755775578,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02343677620861679,
      "loss": 0.2343,
      "num_input_tokens_seen": 14536032,
      "step": 68880
    },
    {
      "epoch": 7.578107810781078,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.023435585500780436,
      "loss": 0.2253,
      "num_input_tokens_seen": 14537120,
      "step": 68885
    },
    {
      "epoch": 7.578657865786578,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.023434394715197336,
      "loss": 0.2327,
      "num_input_tokens_seen": 14538176,
      "step": 68890
    },
    {
      "epoch": 7.579207920792079,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.023433203851878467,
      "loss": 0.2333,
      "num_input_tokens_seen": 14539232,
      "step": 68895
    },
    {
      "epoch": 7.5797579757975795,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.023432012910834805,
      "loss": 0.2331,
      "num_input_tokens_seen": 14540288,
      "step": 68900
    },
    {
      "epoch": 7.580308030803081,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.023430821892077326,
      "loss": 0.2338,
      "num_input_tokens_seen": 14541408,
      "step": 68905
    },
    {
      "epoch": 7.580858085808581,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.023429630795617006,
      "loss": 0.2274,
      "num_input_tokens_seen": 14542464,
      "step": 68910
    },
    {
      "epoch": 7.581408140814082,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02342843962146483,
      "loss": 0.2326,
      "num_input_tokens_seen": 14543552,
      "step": 68915
    },
    {
      "epoch": 7.581958195819582,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.02342724836963176,
      "loss": 0.2274,
      "num_input_tokens_seen": 14544544,
      "step": 68920
    },
    {
      "epoch": 7.582508250825082,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.02342605704012879,
      "loss": 0.2295,
      "num_input_tokens_seen": 14545600,
      "step": 68925
    },
    {
      "epoch": 7.583058305830583,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.023424865632966897,
      "loss": 0.229,
      "num_input_tokens_seen": 14546592,
      "step": 68930
    },
    {
      "epoch": 7.583608360836084,
      "grad_norm": 0.0045166015625,
      "learning_rate": 0.02342367414815706,
      "loss": 0.2296,
      "num_input_tokens_seen": 14547712,
      "step": 68935
    },
    {
      "epoch": 7.584158415841584,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.02342248258571026,
      "loss": 0.2281,
      "num_input_tokens_seen": 14548768,
      "step": 68940
    },
    {
      "epoch": 7.584708470847085,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.02342129094563748,
      "loss": 0.2301,
      "num_input_tokens_seen": 14549856,
      "step": 68945
    },
    {
      "epoch": 7.585258525852585,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0234200992279497,
      "loss": 0.2323,
      "num_input_tokens_seen": 14550976,
      "step": 68950
    },
    {
      "epoch": 7.585808580858086,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.023418907432657912,
      "loss": 0.2276,
      "num_input_tokens_seen": 14552096,
      "step": 68955
    },
    {
      "epoch": 7.586358635863586,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.023417715559773088,
      "loss": 0.2254,
      "num_input_tokens_seen": 14553152,
      "step": 68960
    },
    {
      "epoch": 7.586908690869087,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.023416523609306218,
      "loss": 0.2291,
      "num_input_tokens_seen": 14554208,
      "step": 68965
    },
    {
      "epoch": 7.587458745874588,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.023415331581268295,
      "loss": 0.2338,
      "num_input_tokens_seen": 14555264,
      "step": 68970
    },
    {
      "epoch": 7.588008800880088,
      "grad_norm": 0.001953125,
      "learning_rate": 0.023414139475670292,
      "loss": 0.2265,
      "num_input_tokens_seen": 14556320,
      "step": 68975
    },
    {
      "epoch": 7.588558855885589,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02341294729252321,
      "loss": 0.2322,
      "num_input_tokens_seen": 14557344,
      "step": 68980
    },
    {
      "epoch": 7.589108910891089,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.023411755031838025,
      "loss": 0.2364,
      "num_input_tokens_seen": 14558400,
      "step": 68985
    },
    {
      "epoch": 7.589658965896589,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.023410562693625726,
      "loss": 0.2296,
      "num_input_tokens_seen": 14559488,
      "step": 68990
    },
    {
      "epoch": 7.5902090209020905,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.02340937027789731,
      "loss": 0.2308,
      "num_input_tokens_seen": 14560512,
      "step": 68995
    },
    {
      "epoch": 7.590759075907591,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.02340817778466377,
      "loss": 0.2312,
      "num_input_tokens_seen": 14561632,
      "step": 69000
    },
    {
      "epoch": 7.591309130913091,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.023406985213936082,
      "loss": 0.2322,
      "num_input_tokens_seen": 14562592,
      "step": 69005
    },
    {
      "epoch": 7.591859185918592,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02340579256572525,
      "loss": 0.2255,
      "num_input_tokens_seen": 14563616,
      "step": 69010
    },
    {
      "epoch": 7.592409240924092,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.023404599840042256,
      "loss": 0.2292,
      "num_input_tokens_seen": 14564704,
      "step": 69015
    },
    {
      "epoch": 7.592959295929593,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.0234034070368981,
      "loss": 0.2343,
      "num_input_tokens_seen": 14565792,
      "step": 69020
    },
    {
      "epoch": 7.5935093509350935,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.023402214156303777,
      "loss": 0.2276,
      "num_input_tokens_seen": 14566816,
      "step": 69025
    },
    {
      "epoch": 7.594059405940594,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.02340102119827028,
      "loss": 0.2298,
      "num_input_tokens_seen": 14567904,
      "step": 69030
    },
    {
      "epoch": 7.594609460946095,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.023399828162808597,
      "loss": 0.2303,
      "num_input_tokens_seen": 14568960,
      "step": 69035
    },
    {
      "epoch": 7.595159515951595,
      "grad_norm": 0.0045166015625,
      "learning_rate": 0.023398635049929728,
      "loss": 0.2318,
      "num_input_tokens_seen": 14570080,
      "step": 69040
    },
    {
      "epoch": 7.595709570957096,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.02339744185964467,
      "loss": 0.2292,
      "num_input_tokens_seen": 14571136,
      "step": 69045
    },
    {
      "epoch": 7.596259625962596,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.02339624859196442,
      "loss": 0.2345,
      "num_input_tokens_seen": 14572160,
      "step": 69050
    },
    {
      "epoch": 7.5968096809680965,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.023395055246899978,
      "loss": 0.2303,
      "num_input_tokens_seen": 14573216,
      "step": 69055
    },
    {
      "epoch": 7.597359735973598,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02339386182446234,
      "loss": 0.2255,
      "num_input_tokens_seen": 14574272,
      "step": 69060
    },
    {
      "epoch": 7.597909790979098,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023392668324662504,
      "loss": 0.2339,
      "num_input_tokens_seen": 14575296,
      "step": 69065
    },
    {
      "epoch": 7.598459845984598,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.023391474747511472,
      "loss": 0.237,
      "num_input_tokens_seen": 14576288,
      "step": 69070
    },
    {
      "epoch": 7.599009900990099,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02339028109302024,
      "loss": 0.222,
      "num_input_tokens_seen": 14577344,
      "step": 69075
    },
    {
      "epoch": 7.599559955995599,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02338908736119982,
      "loss": 0.2302,
      "num_input_tokens_seen": 14578368,
      "step": 69080
    },
    {
      "epoch": 7.6001100110011,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.023387893552061202,
      "loss": 0.2286,
      "num_input_tokens_seen": 14579392,
      "step": 69085
    },
    {
      "epoch": 7.600660066006601,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.023386699665615397,
      "loss": 0.2314,
      "num_input_tokens_seen": 14580480,
      "step": 69090
    },
    {
      "epoch": 7.601210121012102,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.023385505701873404,
      "loss": 0.2298,
      "num_input_tokens_seen": 14581568,
      "step": 69095
    },
    {
      "epoch": 7.601760176017602,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02338431166084623,
      "loss": 0.2339,
      "num_input_tokens_seen": 14582560,
      "step": 69100
    },
    {
      "epoch": 7.602310231023102,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.023383117542544875,
      "loss": 0.2339,
      "num_input_tokens_seen": 14583616,
      "step": 69105
    },
    {
      "epoch": 7.602860286028603,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.02338192334698035,
      "loss": 0.2335,
      "num_input_tokens_seen": 14584672,
      "step": 69110
    },
    {
      "epoch": 7.603410341034103,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02338072907416366,
      "loss": 0.2387,
      "num_input_tokens_seen": 14585664,
      "step": 69115
    },
    {
      "epoch": 7.603960396039604,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.023379534724105817,
      "loss": 0.2286,
      "num_input_tokens_seen": 14586784,
      "step": 69120
    },
    {
      "epoch": 7.604510451045105,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.023378340296817814,
      "loss": 0.2343,
      "num_input_tokens_seen": 14587808,
      "step": 69125
    },
    {
      "epoch": 7.605060506050605,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.023377145792310674,
      "loss": 0.2297,
      "num_input_tokens_seen": 14588864,
      "step": 69130
    },
    {
      "epoch": 7.605610561056105,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.0233759512105954,
      "loss": 0.2334,
      "num_input_tokens_seen": 14589984,
      "step": 69135
    },
    {
      "epoch": 7.606160616061606,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.023374756551683005,
      "loss": 0.2348,
      "num_input_tokens_seen": 14590976,
      "step": 69140
    },
    {
      "epoch": 7.606710671067106,
      "grad_norm": 0.00457763671875,
      "learning_rate": 0.023373561815584493,
      "loss": 0.2302,
      "num_input_tokens_seen": 14592032,
      "step": 69145
    },
    {
      "epoch": 7.6072607260726075,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.023372367002310886,
      "loss": 0.2301,
      "num_input_tokens_seen": 14593088,
      "step": 69150
    },
    {
      "epoch": 7.607810781078108,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.023371172111873183,
      "loss": 0.2353,
      "num_input_tokens_seen": 14594144,
      "step": 69155
    },
    {
      "epoch": 7.608360836083609,
      "grad_norm": 0.004547119140625,
      "learning_rate": 0.023369977144282406,
      "loss": 0.2265,
      "num_input_tokens_seen": 14595232,
      "step": 69160
    },
    {
      "epoch": 7.608910891089109,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.023368782099549568,
      "loss": 0.239,
      "num_input_tokens_seen": 14596256,
      "step": 69165
    },
    {
      "epoch": 7.609460946094609,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.023367586977685678,
      "loss": 0.2279,
      "num_input_tokens_seen": 14597280,
      "step": 69170
    },
    {
      "epoch": 7.61001100110011,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.023366391778701756,
      "loss": 0.2332,
      "num_input_tokens_seen": 14598304,
      "step": 69175
    },
    {
      "epoch": 7.6105610561056105,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.023365196502608813,
      "loss": 0.2363,
      "num_input_tokens_seen": 14599424,
      "step": 69180
    },
    {
      "epoch": 7.611111111111111,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.023364001149417873,
      "loss": 0.2321,
      "num_input_tokens_seen": 14600448,
      "step": 69185
    },
    {
      "epoch": 7.611661166116612,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.023362805719139946,
      "loss": 0.2331,
      "num_input_tokens_seen": 14601536,
      "step": 69190
    },
    {
      "epoch": 7.612211221122112,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.02336161021178605,
      "loss": 0.2316,
      "num_input_tokens_seen": 14602656,
      "step": 69195
    },
    {
      "epoch": 7.612761276127613,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.023360414627367203,
      "loss": 0.2331,
      "num_input_tokens_seen": 14603648,
      "step": 69200
    },
    {
      "epoch": 7.613311331133113,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.023359218965894428,
      "loss": 0.2368,
      "num_input_tokens_seen": 14604736,
      "step": 69205
    },
    {
      "epoch": 7.6138613861386135,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.023358023227378746,
      "loss": 0.2346,
      "num_input_tokens_seen": 14605856,
      "step": 69210
    },
    {
      "epoch": 7.614411441144115,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.02335682741183117,
      "loss": 0.2314,
      "num_input_tokens_seen": 14606880,
      "step": 69215
    },
    {
      "epoch": 7.614961496149615,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.023355631519262728,
      "loss": 0.2299,
      "num_input_tokens_seen": 14607936,
      "step": 69220
    },
    {
      "epoch": 7.615511551155116,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.023354435549684437,
      "loss": 0.2319,
      "num_input_tokens_seen": 14608960,
      "step": 69225
    },
    {
      "epoch": 7.616061606160616,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.023353239503107324,
      "loss": 0.2319,
      "num_input_tokens_seen": 14609952,
      "step": 69230
    },
    {
      "epoch": 7.616611661166116,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.023352043379542414,
      "loss": 0.2304,
      "num_input_tokens_seen": 14611008,
      "step": 69235
    },
    {
      "epoch": 7.617161716171617,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.023350847179000725,
      "loss": 0.2283,
      "num_input_tokens_seen": 14612032,
      "step": 69240
    },
    {
      "epoch": 7.617711771177118,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.023349650901493282,
      "loss": 0.233,
      "num_input_tokens_seen": 14613056,
      "step": 69245
    },
    {
      "epoch": 7.618261826182618,
      "grad_norm": 0.000560760498046875,
      "learning_rate": 0.02334845454703112,
      "loss": 0.2309,
      "num_input_tokens_seen": 14614144,
      "step": 69250
    },
    {
      "epoch": 7.618811881188119,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.02334725811562525,
      "loss": 0.233,
      "num_input_tokens_seen": 14615136,
      "step": 69255
    },
    {
      "epoch": 7.619361936193619,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.023346061607286715,
      "loss": 0.234,
      "num_input_tokens_seen": 14616128,
      "step": 69260
    },
    {
      "epoch": 7.61991199119912,
      "grad_norm": 0.00069427490234375,
      "learning_rate": 0.023344865022026533,
      "loss": 0.2325,
      "num_input_tokens_seen": 14617120,
      "step": 69265
    },
    {
      "epoch": 7.62046204620462,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.023343668359855733,
      "loss": 0.232,
      "num_input_tokens_seen": 14618208,
      "step": 69270
    },
    {
      "epoch": 7.621012101210121,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023342471620785343,
      "loss": 0.2314,
      "num_input_tokens_seen": 14619168,
      "step": 69275
    },
    {
      "epoch": 7.621562156215622,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.023341274804826402,
      "loss": 0.233,
      "num_input_tokens_seen": 14620256,
      "step": 69280
    },
    {
      "epoch": 7.622112211221122,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.02334007791198993,
      "loss": 0.2304,
      "num_input_tokens_seen": 14621312,
      "step": 69285
    },
    {
      "epoch": 7.622662266226623,
      "grad_norm": 0.0004558563232421875,
      "learning_rate": 0.02333888094228696,
      "loss": 0.232,
      "num_input_tokens_seen": 14622368,
      "step": 69290
    },
    {
      "epoch": 7.623212321232123,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023337683895728524,
      "loss": 0.2288,
      "num_input_tokens_seen": 14623424,
      "step": 69295
    },
    {
      "epoch": 7.623762376237623,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.023336486772325662,
      "loss": 0.234,
      "num_input_tokens_seen": 14624416,
      "step": 69300
    },
    {
      "epoch": 7.6243124312431245,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.0233352895720894,
      "loss": 0.2356,
      "num_input_tokens_seen": 14625504,
      "step": 69305
    },
    {
      "epoch": 7.624862486248625,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.02333409229503077,
      "loss": 0.2283,
      "num_input_tokens_seen": 14626624,
      "step": 69310
    },
    {
      "epoch": 7.625412541254125,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.023332894941160816,
      "loss": 0.2309,
      "num_input_tokens_seen": 14627680,
      "step": 69315
    },
    {
      "epoch": 7.625962596259626,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.023331697510490565,
      "loss": 0.2299,
      "num_input_tokens_seen": 14628672,
      "step": 69320
    },
    {
      "epoch": 7.626512651265126,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02333050000303105,
      "loss": 0.2293,
      "num_input_tokens_seen": 14629696,
      "step": 69325
    },
    {
      "epoch": 7.627062706270627,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.023329302418793323,
      "loss": 0.2314,
      "num_input_tokens_seen": 14630688,
      "step": 69330
    },
    {
      "epoch": 7.6276127612761275,
      "grad_norm": 0.000888824462890625,
      "learning_rate": 0.02332810475778841,
      "loss": 0.2304,
      "num_input_tokens_seen": 14631680,
      "step": 69335
    },
    {
      "epoch": 7.628162816281629,
      "grad_norm": 0.000606536865234375,
      "learning_rate": 0.023326907020027352,
      "loss": 0.2304,
      "num_input_tokens_seen": 14632768,
      "step": 69340
    },
    {
      "epoch": 7.628712871287129,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.023325709205521183,
      "loss": 0.2304,
      "num_input_tokens_seen": 14633856,
      "step": 69345
    },
    {
      "epoch": 7.629262926292629,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.023324511314280955,
      "loss": 0.2314,
      "num_input_tokens_seen": 14634880,
      "step": 69350
    },
    {
      "epoch": 7.62981298129813,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.023323313346317695,
      "loss": 0.2315,
      "num_input_tokens_seen": 14635968,
      "step": 69355
    },
    {
      "epoch": 7.63036303630363,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.02332211530164245,
      "loss": 0.232,
      "num_input_tokens_seen": 14636992,
      "step": 69360
    },
    {
      "epoch": 7.6309130913091305,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.023320917180266264,
      "loss": 0.2294,
      "num_input_tokens_seen": 14638080,
      "step": 69365
    },
    {
      "epoch": 7.631463146314632,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.02331971898220018,
      "loss": 0.2335,
      "num_input_tokens_seen": 14639104,
      "step": 69370
    },
    {
      "epoch": 7.632013201320132,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.02331852070745523,
      "loss": 0.2299,
      "num_input_tokens_seen": 14640128,
      "step": 69375
    },
    {
      "epoch": 7.632563256325633,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.023317322356042475,
      "loss": 0.2299,
      "num_input_tokens_seen": 14641152,
      "step": 69380
    },
    {
      "epoch": 7.633113311331133,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.023316123927972946,
      "loss": 0.2341,
      "num_input_tokens_seen": 14642240,
      "step": 69385
    },
    {
      "epoch": 7.633663366336633,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.02331492542325769,
      "loss": 0.2341,
      "num_input_tokens_seen": 14643264,
      "step": 69390
    },
    {
      "epoch": 7.634213421342134,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.023313726841907766,
      "loss": 0.2309,
      "num_input_tokens_seen": 14644288,
      "step": 69395
    },
    {
      "epoch": 7.634763476347635,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.0233125281839342,
      "loss": 0.233,
      "num_input_tokens_seen": 14645344,
      "step": 69400
    },
    {
      "epoch": 7.635313531353136,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.023311329449348057,
      "loss": 0.2293,
      "num_input_tokens_seen": 14646432,
      "step": 69405
    },
    {
      "epoch": 7.635863586358636,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.023310130638160377,
      "loss": 0.232,
      "num_input_tokens_seen": 14647520,
      "step": 69410
    },
    {
      "epoch": 7.636413641364136,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02330893175038221,
      "loss": 0.2324,
      "num_input_tokens_seen": 14648512,
      "step": 69415
    },
    {
      "epoch": 7.636963696369637,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.023307732786024604,
      "loss": 0.232,
      "num_input_tokens_seen": 14649536,
      "step": 69420
    },
    {
      "epoch": 7.637513751375137,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.023306533745098616,
      "loss": 0.2325,
      "num_input_tokens_seen": 14650592,
      "step": 69425
    },
    {
      "epoch": 7.638063806380638,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.023305334627615294,
      "loss": 0.2304,
      "num_input_tokens_seen": 14651680,
      "step": 69430
    },
    {
      "epoch": 7.638613861386139,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.023304135433585685,
      "loss": 0.232,
      "num_input_tokens_seen": 14652768,
      "step": 69435
    },
    {
      "epoch": 7.639163916391639,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.02330293616302084,
      "loss": 0.2351,
      "num_input_tokens_seen": 14653760,
      "step": 69440
    },
    {
      "epoch": 7.63971397139714,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02330173681593182,
      "loss": 0.2314,
      "num_input_tokens_seen": 14654848,
      "step": 69445
    },
    {
      "epoch": 7.64026402640264,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.023300537392329672,
      "loss": 0.2299,
      "num_input_tokens_seen": 14655904,
      "step": 69450
    },
    {
      "epoch": 7.6408140814081404,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.023299337892225463,
      "loss": 0.2309,
      "num_input_tokens_seen": 14656928,
      "step": 69455
    },
    {
      "epoch": 7.6413641364136415,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.023298138315630233,
      "loss": 0.2325,
      "num_input_tokens_seen": 14657952,
      "step": 69460
    },
    {
      "epoch": 7.641914191419142,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.023296938662555042,
      "loss": 0.2304,
      "num_input_tokens_seen": 14659072,
      "step": 69465
    },
    {
      "epoch": 7.642464246424643,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.02329573893301095,
      "loss": 0.2314,
      "num_input_tokens_seen": 14660160,
      "step": 69470
    },
    {
      "epoch": 7.643014301430143,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.023294539127009014,
      "loss": 0.232,
      "num_input_tokens_seen": 14661184,
      "step": 69475
    },
    {
      "epoch": 7.643564356435643,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.023293339244560295,
      "loss": 0.233,
      "num_input_tokens_seen": 14662208,
      "step": 69480
    },
    {
      "epoch": 7.644114411441144,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.023292139285675844,
      "loss": 0.2278,
      "num_input_tokens_seen": 14663264,
      "step": 69485
    },
    {
      "epoch": 7.6446644664466445,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02329093925036672,
      "loss": 0.233,
      "num_input_tokens_seen": 14664384,
      "step": 69490
    },
    {
      "epoch": 7.645214521452145,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023289739138643996,
      "loss": 0.233,
      "num_input_tokens_seen": 14665536,
      "step": 69495
    },
    {
      "epoch": 7.645764576457646,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02328853895051872,
      "loss": 0.233,
      "num_input_tokens_seen": 14666560,
      "step": 69500
    },
    {
      "epoch": 7.646314631463146,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.023287338686001958,
      "loss": 0.232,
      "num_input_tokens_seen": 14667680,
      "step": 69505
    },
    {
      "epoch": 7.646864686468647,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02328613834510477,
      "loss": 0.2325,
      "num_input_tokens_seen": 14668704,
      "step": 69510
    },
    {
      "epoch": 7.647414741474147,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.023284937927838218,
      "loss": 0.2335,
      "num_input_tokens_seen": 14669728,
      "step": 69515
    },
    {
      "epoch": 7.647964796479648,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.023283737434213372,
      "loss": 0.2346,
      "num_input_tokens_seen": 14670784,
      "step": 69520
    },
    {
      "epoch": 7.648514851485149,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023282536864241296,
      "loss": 0.2324,
      "num_input_tokens_seen": 14671840,
      "step": 69525
    },
    {
      "epoch": 7.649064906490649,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02328133621793305,
      "loss": 0.2309,
      "num_input_tokens_seen": 14672864,
      "step": 69530
    },
    {
      "epoch": 7.64961496149615,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.0232801354952997,
      "loss": 0.2314,
      "num_input_tokens_seen": 14673952,
      "step": 69535
    },
    {
      "epoch": 7.65016501650165,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.023278934696352317,
      "loss": 0.2303,
      "num_input_tokens_seen": 14675008,
      "step": 69540
    },
    {
      "epoch": 7.65071507150715,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.023277733821101963,
      "loss": 0.2298,
      "num_input_tokens_seen": 14676160,
      "step": 69545
    },
    {
      "epoch": 7.6512651265126514,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02327653286955971,
      "loss": 0.2319,
      "num_input_tokens_seen": 14677248,
      "step": 69550
    },
    {
      "epoch": 7.651815181518152,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.023275331841736628,
      "loss": 0.2309,
      "num_input_tokens_seen": 14678336,
      "step": 69555
    },
    {
      "epoch": 7.652365236523653,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.023274130737643776,
      "loss": 0.2309,
      "num_input_tokens_seen": 14679456,
      "step": 69560
    },
    {
      "epoch": 7.652915291529153,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.023272929557292238,
      "loss": 0.2319,
      "num_input_tokens_seen": 14680512,
      "step": 69565
    },
    {
      "epoch": 7.653465346534653,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023271728300693073,
      "loss": 0.2319,
      "num_input_tokens_seen": 14681472,
      "step": 69570
    },
    {
      "epoch": 7.654015401540154,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.023270526967857363,
      "loss": 0.2314,
      "num_input_tokens_seen": 14682528,
      "step": 69575
    },
    {
      "epoch": 7.6545654565456545,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02326932555879617,
      "loss": 0.2314,
      "num_input_tokens_seen": 14683584,
      "step": 69580
    },
    {
      "epoch": 7.6551155115511555,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.023268124073520573,
      "loss": 0.2319,
      "num_input_tokens_seen": 14684640,
      "step": 69585
    },
    {
      "epoch": 7.655665566556656,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.023266922512041644,
      "loss": 0.2303,
      "num_input_tokens_seen": 14685664,
      "step": 69590
    },
    {
      "epoch": 7.656215621562156,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.023265720874370458,
      "loss": 0.2293,
      "num_input_tokens_seen": 14686688,
      "step": 69595
    },
    {
      "epoch": 7.656765676567657,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.023264519160518086,
      "loss": 0.2314,
      "num_input_tokens_seen": 14687776,
      "step": 69600
    },
    {
      "epoch": 7.657315731573157,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.023263317370495612,
      "loss": 0.2325,
      "num_input_tokens_seen": 14688864,
      "step": 69605
    },
    {
      "epoch": 7.6578657865786575,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0232621155043141,
      "loss": 0.233,
      "num_input_tokens_seen": 14689920,
      "step": 69610
    },
    {
      "epoch": 7.658415841584159,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.023260913561984637,
      "loss": 0.2325,
      "num_input_tokens_seen": 14691040,
      "step": 69615
    },
    {
      "epoch": 7.658965896589659,
      "grad_norm": 0.0004634857177734375,
      "learning_rate": 0.0232597115435183,
      "loss": 0.2304,
      "num_input_tokens_seen": 14692192,
      "step": 69620
    },
    {
      "epoch": 7.65951595159516,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02325850944892616,
      "loss": 0.2319,
      "num_input_tokens_seen": 14693216,
      "step": 69625
    },
    {
      "epoch": 7.66006600660066,
      "grad_norm": 0.005859375,
      "learning_rate": 0.023257307278219307,
      "loss": 0.2319,
      "num_input_tokens_seen": 14694336,
      "step": 69630
    },
    {
      "epoch": 7.66061606160616,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.023256105031408812,
      "loss": 0.2309,
      "num_input_tokens_seen": 14695328,
      "step": 69635
    },
    {
      "epoch": 7.661166116611661,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.023254902708505755,
      "loss": 0.2283,
      "num_input_tokens_seen": 14696352,
      "step": 69640
    },
    {
      "epoch": 7.661716171617162,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02325370030952123,
      "loss": 0.2309,
      "num_input_tokens_seen": 14697408,
      "step": 69645
    },
    {
      "epoch": 7.662266226622663,
      "grad_norm": 0.0006866455078125,
      "learning_rate": 0.023252497834466302,
      "loss": 0.2314,
      "num_input_tokens_seen": 14698400,
      "step": 69650
    },
    {
      "epoch": 7.662816281628163,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023251295283352062,
      "loss": 0.2319,
      "num_input_tokens_seen": 14699424,
      "step": 69655
    },
    {
      "epoch": 7.663366336633663,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.023250092656189596,
      "loss": 0.2314,
      "num_input_tokens_seen": 14700448,
      "step": 69660
    },
    {
      "epoch": 7.663916391639164,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02324888995298998,
      "loss": 0.2325,
      "num_input_tokens_seen": 14701568,
      "step": 69665
    },
    {
      "epoch": 7.664466446644664,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.02324768717376431,
      "loss": 0.2314,
      "num_input_tokens_seen": 14702560,
      "step": 69670
    },
    {
      "epoch": 7.665016501650165,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.023246484318523664,
      "loss": 0.2314,
      "num_input_tokens_seen": 14703616,
      "step": 69675
    },
    {
      "epoch": 7.665566556655666,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02324528138727912,
      "loss": 0.2298,
      "num_input_tokens_seen": 14704608,
      "step": 69680
    },
    {
      "epoch": 7.666116611661166,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.023244078380041784,
      "loss": 0.2335,
      "num_input_tokens_seen": 14705728,
      "step": 69685
    },
    {
      "epoch": 7.666666666666667,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.023242875296822733,
      "loss": 0.232,
      "num_input_tokens_seen": 14706720,
      "step": 69690
    },
    {
      "epoch": 7.667216721672167,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.023241672137633052,
      "loss": 0.2309,
      "num_input_tokens_seen": 14707744,
      "step": 69695
    },
    {
      "epoch": 7.667766776677668,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02324046890248383,
      "loss": 0.2309,
      "num_input_tokens_seen": 14708864,
      "step": 69700
    },
    {
      "epoch": 7.6683168316831685,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.023239265591386168,
      "loss": 0.2304,
      "num_input_tokens_seen": 14709920,
      "step": 69705
    },
    {
      "epoch": 7.668866886688669,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.023238062204351144,
      "loss": 0.2299,
      "num_input_tokens_seen": 14710944,
      "step": 69710
    },
    {
      "epoch": 7.66941694169417,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.023236858741389855,
      "loss": 0.2324,
      "num_input_tokens_seen": 14712096,
      "step": 69715
    },
    {
      "epoch": 7.66996699669967,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.023235655202513396,
      "loss": 0.2309,
      "num_input_tokens_seen": 14713120,
      "step": 69720
    },
    {
      "epoch": 7.67051705170517,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.023234451587732845,
      "loss": 0.2314,
      "num_input_tokens_seen": 14714144,
      "step": 69725
    },
    {
      "epoch": 7.671067106710671,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02323324789705931,
      "loss": 0.2289,
      "num_input_tokens_seen": 14715200,
      "step": 69730
    },
    {
      "epoch": 7.6716171617161715,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.023232044130503882,
      "loss": 0.234,
      "num_input_tokens_seen": 14716224,
      "step": 69735
    },
    {
      "epoch": 7.672167216721672,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02323084028807765,
      "loss": 0.2309,
      "num_input_tokens_seen": 14717248,
      "step": 69740
    },
    {
      "epoch": 7.672717271727173,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.023229636369791716,
      "loss": 0.2294,
      "num_input_tokens_seen": 14718336,
      "step": 69745
    },
    {
      "epoch": 7.673267326732673,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02322843237565717,
      "loss": 0.2294,
      "num_input_tokens_seen": 14719456,
      "step": 69750
    },
    {
      "epoch": 7.673817381738174,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.02322722830568511,
      "loss": 0.2325,
      "num_input_tokens_seen": 14720544,
      "step": 69755
    },
    {
      "epoch": 7.674367436743674,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.023226024159886637,
      "loss": 0.2283,
      "num_input_tokens_seen": 14721600,
      "step": 69760
    },
    {
      "epoch": 7.674917491749175,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02322481993827285,
      "loss": 0.2295,
      "num_input_tokens_seen": 14722560,
      "step": 69765
    },
    {
      "epoch": 7.675467546754676,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.02322361564085484,
      "loss": 0.2316,
      "num_input_tokens_seen": 14723584,
      "step": 69770
    },
    {
      "epoch": 7.676017601760176,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.023222411267643708,
      "loss": 0.2338,
      "num_input_tokens_seen": 14724640,
      "step": 69775
    },
    {
      "epoch": 7.676567656765677,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.02322120681865056,
      "loss": 0.2337,
      "num_input_tokens_seen": 14725664,
      "step": 69780
    },
    {
      "epoch": 7.677117711771177,
      "grad_norm": 0.0126953125,
      "learning_rate": 0.023220002293886495,
      "loss": 0.2347,
      "num_input_tokens_seen": 14726720,
      "step": 69785
    },
    {
      "epoch": 7.677667766776677,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.02321879769336261,
      "loss": 0.2309,
      "num_input_tokens_seen": 14727712,
      "step": 69790
    },
    {
      "epoch": 7.678217821782178,
      "grad_norm": 0.01214599609375,
      "learning_rate": 0.023217593017090015,
      "loss": 0.2362,
      "num_input_tokens_seen": 14728736,
      "step": 69795
    },
    {
      "epoch": 7.678767876787679,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.023216388265079803,
      "loss": 0.2304,
      "num_input_tokens_seen": 14729760,
      "step": 69800
    },
    {
      "epoch": 7.67931793179318,
      "grad_norm": 0.005859375,
      "learning_rate": 0.023215183437343086,
      "loss": 0.2309,
      "num_input_tokens_seen": 14730816,
      "step": 69805
    },
    {
      "epoch": 7.67986798679868,
      "grad_norm": 0.00051116943359375,
      "learning_rate": 0.023213978533890962,
      "loss": 0.2325,
      "num_input_tokens_seen": 14731808,
      "step": 69810
    },
    {
      "epoch": 7.68041804180418,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.02321277355473455,
      "loss": 0.233,
      "num_input_tokens_seen": 14732928,
      "step": 69815
    },
    {
      "epoch": 7.680968096809681,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.02321156849988493,
      "loss": 0.2319,
      "num_input_tokens_seen": 14734048,
      "step": 69820
    },
    {
      "epoch": 7.681518151815181,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.023210363369353233,
      "loss": 0.2309,
      "num_input_tokens_seen": 14735168,
      "step": 69825
    },
    {
      "epoch": 7.6820682068206825,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.02320915816315055,
      "loss": 0.2293,
      "num_input_tokens_seen": 14736256,
      "step": 69830
    },
    {
      "epoch": 7.682618261826183,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.023207952881288,
      "loss": 0.235,
      "num_input_tokens_seen": 14737248,
      "step": 69835
    },
    {
      "epoch": 7.683168316831683,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.023206747523776685,
      "loss": 0.2313,
      "num_input_tokens_seen": 14738272,
      "step": 69840
    },
    {
      "epoch": 7.683718371837184,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02320554209062772,
      "loss": 0.2313,
      "num_input_tokens_seen": 14739328,
      "step": 69845
    },
    {
      "epoch": 7.684268426842684,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.02320433658185221,
      "loss": 0.2319,
      "num_input_tokens_seen": 14740416,
      "step": 69850
    },
    {
      "epoch": 7.684818481848184,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.023203130997461262,
      "loss": 0.2303,
      "num_input_tokens_seen": 14741408,
      "step": 69855
    },
    {
      "epoch": 7.6853685368536855,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.023201925337466,
      "loss": 0.2293,
      "num_input_tokens_seen": 14742496,
      "step": 69860
    },
    {
      "epoch": 7.685918591859186,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02320071960187752,
      "loss": 0.2313,
      "num_input_tokens_seen": 14743552,
      "step": 69865
    },
    {
      "epoch": 7.686468646864687,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.023199513790706942,
      "loss": 0.2319,
      "num_input_tokens_seen": 14744576,
      "step": 69870
    },
    {
      "epoch": 7.687018701870187,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.023198307903965387,
      "loss": 0.2334,
      "num_input_tokens_seen": 14745664,
      "step": 69875
    },
    {
      "epoch": 7.687568756875687,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02319710194166396,
      "loss": 0.2319,
      "num_input_tokens_seen": 14746784,
      "step": 69880
    },
    {
      "epoch": 7.688118811881188,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02319589590381377,
      "loss": 0.2303,
      "num_input_tokens_seen": 14747840,
      "step": 69885
    },
    {
      "epoch": 7.6886688668866885,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.023194689790425948,
      "loss": 0.2272,
      "num_input_tokens_seen": 14748928,
      "step": 69890
    },
    {
      "epoch": 7.68921892189219,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.0231934836015116,
      "loss": 0.2308,
      "num_input_tokens_seen": 14749984,
      "step": 69895
    },
    {
      "epoch": 7.68976897689769,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.023192277337081842,
      "loss": 0.2319,
      "num_input_tokens_seen": 14751040,
      "step": 69900
    },
    {
      "epoch": 7.69031903190319,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.0231910709971478,
      "loss": 0.2319,
      "num_input_tokens_seen": 14752064,
      "step": 69905
    },
    {
      "epoch": 7.690869086908691,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.023189864581720586,
      "loss": 0.2318,
      "num_input_tokens_seen": 14753152,
      "step": 69910
    },
    {
      "epoch": 7.691419141914191,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023188658090811318,
      "loss": 0.2298,
      "num_input_tokens_seen": 14754208,
      "step": 69915
    },
    {
      "epoch": 7.6919691969196915,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.02318745152443112,
      "loss": 0.2314,
      "num_input_tokens_seen": 14755264,
      "step": 69920
    },
    {
      "epoch": 7.692519251925193,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.02318624488259111,
      "loss": 0.2324,
      "num_input_tokens_seen": 14756384,
      "step": 69925
    },
    {
      "epoch": 7.693069306930693,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02318503816530241,
      "loss": 0.2319,
      "num_input_tokens_seen": 14757376,
      "step": 69930
    },
    {
      "epoch": 7.693619361936194,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.023183831372576135,
      "loss": 0.2329,
      "num_input_tokens_seen": 14758368,
      "step": 69935
    },
    {
      "epoch": 7.694169416941694,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.02318262450442342,
      "loss": 0.2319,
      "num_input_tokens_seen": 14759520,
      "step": 69940
    },
    {
      "epoch": 7.694719471947195,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.023181417560855375,
      "loss": 0.2324,
      "num_input_tokens_seen": 14760640,
      "step": 69945
    },
    {
      "epoch": 7.695269526952695,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.023180210541883132,
      "loss": 0.2309,
      "num_input_tokens_seen": 14761696,
      "step": 69950
    },
    {
      "epoch": 7.695819581958196,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.023179003447517818,
      "loss": 0.2324,
      "num_input_tokens_seen": 14762752,
      "step": 69955
    },
    {
      "epoch": 7.696369636963697,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.02317779627777055,
      "loss": 0.2324,
      "num_input_tokens_seen": 14763744,
      "step": 69960
    },
    {
      "epoch": 7.696919691969197,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.023176589032652458,
      "loss": 0.2329,
      "num_input_tokens_seen": 14764800,
      "step": 69965
    },
    {
      "epoch": 7.697469746974697,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.023175381712174672,
      "loss": 0.2324,
      "num_input_tokens_seen": 14765952,
      "step": 69970
    },
    {
      "epoch": 7.698019801980198,
      "grad_norm": 0.00311279296875,
      "learning_rate": 0.023174174316348314,
      "loss": 0.2303,
      "num_input_tokens_seen": 14767008,
      "step": 69975
    },
    {
      "epoch": 7.698569856985698,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.023172966845184512,
      "loss": 0.2308,
      "num_input_tokens_seen": 14768064,
      "step": 69980
    },
    {
      "epoch": 7.6991199119911995,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.0231717592986944,
      "loss": 0.2298,
      "num_input_tokens_seen": 14769120,
      "step": 69985
    },
    {
      "epoch": 7.6996699669967,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.023170551676889098,
      "loss": 0.2277,
      "num_input_tokens_seen": 14770208,
      "step": 69990
    },
    {
      "epoch": 7.7002200220022,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.023169343979779745,
      "loss": 0.2298,
      "num_input_tokens_seen": 14771296,
      "step": 69995
    },
    {
      "epoch": 7.700770077007701,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.023168136207377468,
      "loss": 0.234,
      "num_input_tokens_seen": 14772384,
      "step": 70000
    },
    {
      "epoch": 7.701320132013201,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.0231669283596934,
      "loss": 0.234,
      "num_input_tokens_seen": 14773408,
      "step": 70005
    },
    {
      "epoch": 7.701870187018702,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02316572043673867,
      "loss": 0.2303,
      "num_input_tokens_seen": 14774432,
      "step": 70010
    },
    {
      "epoch": 7.7024202420242025,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.023164512438524413,
      "loss": 0.2329,
      "num_input_tokens_seen": 14775488,
      "step": 70015
    },
    {
      "epoch": 7.702970297029703,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.023163304365061768,
      "loss": 0.2314,
      "num_input_tokens_seen": 14776544,
      "step": 70020
    },
    {
      "epoch": 7.703520352035204,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02316209621636186,
      "loss": 0.2334,
      "num_input_tokens_seen": 14777568,
      "step": 70025
    },
    {
      "epoch": 7.704070407040704,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023160887992435826,
      "loss": 0.2345,
      "num_input_tokens_seen": 14778720,
      "step": 70030
    },
    {
      "epoch": 7.704620462046204,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.023159679693294806,
      "loss": 0.2303,
      "num_input_tokens_seen": 14779744,
      "step": 70035
    },
    {
      "epoch": 7.705170517051705,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.023158471318949932,
      "loss": 0.2309,
      "num_input_tokens_seen": 14780800,
      "step": 70040
    },
    {
      "epoch": 7.7057205720572055,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.023157262869412342,
      "loss": 0.2303,
      "num_input_tokens_seen": 14781888,
      "step": 70045
    },
    {
      "epoch": 7.706270627062707,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02315605434469318,
      "loss": 0.2334,
      "num_input_tokens_seen": 14782880,
      "step": 70050
    },
    {
      "epoch": 7.706820682068207,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02315484574480357,
      "loss": 0.2324,
      "num_input_tokens_seen": 14783936,
      "step": 70055
    },
    {
      "epoch": 7.707370737073707,
      "grad_norm": 0.000766754150390625,
      "learning_rate": 0.023153637069754665,
      "loss": 0.2319,
      "num_input_tokens_seen": 14784960,
      "step": 70060
    },
    {
      "epoch": 7.707920792079208,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.0231524283195576,
      "loss": 0.2308,
      "num_input_tokens_seen": 14785984,
      "step": 70065
    },
    {
      "epoch": 7.708470847084708,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.023151219494223513,
      "loss": 0.2319,
      "num_input_tokens_seen": 14787008,
      "step": 70070
    },
    {
      "epoch": 7.709020902090209,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.023150010593763547,
      "loss": 0.2324,
      "num_input_tokens_seen": 14788032,
      "step": 70075
    },
    {
      "epoch": 7.70957095709571,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023148801618188847,
      "loss": 0.2314,
      "num_input_tokens_seen": 14789152,
      "step": 70080
    },
    {
      "epoch": 7.71012101210121,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02314759256751055,
      "loss": 0.2319,
      "num_input_tokens_seen": 14790208,
      "step": 70085
    },
    {
      "epoch": 7.710671067106711,
      "grad_norm": 0.00081634521484375,
      "learning_rate": 0.0231463834417398,
      "loss": 0.2329,
      "num_input_tokens_seen": 14791264,
      "step": 70090
    },
    {
      "epoch": 7.711221122112211,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.023145174240887748,
      "loss": 0.2303,
      "num_input_tokens_seen": 14792352,
      "step": 70095
    },
    {
      "epoch": 7.711771177117711,
      "grad_norm": 0.00067901611328125,
      "learning_rate": 0.023143964964965534,
      "loss": 0.2324,
      "num_input_tokens_seen": 14793472,
      "step": 70100
    },
    {
      "epoch": 7.712321232123212,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.0231427556139843,
      "loss": 0.2314,
      "num_input_tokens_seen": 14794496,
      "step": 70105
    },
    {
      "epoch": 7.712871287128713,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.023141546187955192,
      "loss": 0.2324,
      "num_input_tokens_seen": 14795520,
      "step": 70110
    },
    {
      "epoch": 7.713421342134214,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.023140336686889362,
      "loss": 0.2345,
      "num_input_tokens_seen": 14796608,
      "step": 70115
    },
    {
      "epoch": 7.713971397139714,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.023139127110797957,
      "loss": 0.233,
      "num_input_tokens_seen": 14797600,
      "step": 70120
    },
    {
      "epoch": 7.714521452145215,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02313791745969212,
      "loss": 0.2324,
      "num_input_tokens_seen": 14798592,
      "step": 70125
    },
    {
      "epoch": 7.715071507150715,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.023136707733583003,
      "loss": 0.2324,
      "num_input_tokens_seen": 14799648,
      "step": 70130
    },
    {
      "epoch": 7.715621562156215,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.023135497932481756,
      "loss": 0.234,
      "num_input_tokens_seen": 14800736,
      "step": 70135
    },
    {
      "epoch": 7.7161716171617165,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02313428805639953,
      "loss": 0.2298,
      "num_input_tokens_seen": 14801856,
      "step": 70140
    },
    {
      "epoch": 7.716721672167217,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02313307810534747,
      "loss": 0.2319,
      "num_input_tokens_seen": 14802944,
      "step": 70145
    },
    {
      "epoch": 7.717271727172717,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.023131868079336735,
      "loss": 0.2303,
      "num_input_tokens_seen": 14804032,
      "step": 70150
    },
    {
      "epoch": 7.717821782178218,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.02313065797837847,
      "loss": 0.2345,
      "num_input_tokens_seen": 14805120,
      "step": 70155
    },
    {
      "epoch": 7.718371837183718,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.023129447802483837,
      "loss": 0.2319,
      "num_input_tokens_seen": 14806208,
      "step": 70160
    },
    {
      "epoch": 7.718921892189218,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02312823755166398,
      "loss": 0.2334,
      "num_input_tokens_seen": 14807264,
      "step": 70165
    },
    {
      "epoch": 7.7194719471947195,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023127027225930068,
      "loss": 0.2309,
      "num_input_tokens_seen": 14808320,
      "step": 70170
    },
    {
      "epoch": 7.72002200220022,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.023125816825293233,
      "loss": 0.2314,
      "num_input_tokens_seen": 14809440,
      "step": 70175
    },
    {
      "epoch": 7.720572057205721,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.023124606349764647,
      "loss": 0.2319,
      "num_input_tokens_seen": 14810496,
      "step": 70180
    },
    {
      "epoch": 7.721122112211221,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02312339579935547,
      "loss": 0.2309,
      "num_input_tokens_seen": 14811584,
      "step": 70185
    },
    {
      "epoch": 7.721672167216722,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.023122185174076847,
      "loss": 0.2308,
      "num_input_tokens_seen": 14812608,
      "step": 70190
    },
    {
      "epoch": 7.722222222222222,
      "grad_norm": 0.000469207763671875,
      "learning_rate": 0.02312097447393994,
      "loss": 0.2298,
      "num_input_tokens_seen": 14813664,
      "step": 70195
    },
    {
      "epoch": 7.7227722772277225,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.023119763698955905,
      "loss": 0.2308,
      "num_input_tokens_seen": 14814784,
      "step": 70200
    },
    {
      "epoch": 7.723322332233224,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.023118552849135908,
      "loss": 0.2318,
      "num_input_tokens_seen": 14815904,
      "step": 70205
    },
    {
      "epoch": 7.723872387238724,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023117341924491105,
      "loss": 0.2282,
      "num_input_tokens_seen": 14816960,
      "step": 70210
    },
    {
      "epoch": 7.724422442244224,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.023116130925032657,
      "loss": 0.2324,
      "num_input_tokens_seen": 14818048,
      "step": 70215
    },
    {
      "epoch": 7.724972497249725,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.023114919850771725,
      "loss": 0.2324,
      "num_input_tokens_seen": 14819168,
      "step": 70220
    },
    {
      "epoch": 7.725522552255225,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.023113708701719467,
      "loss": 0.2329,
      "num_input_tokens_seen": 14820224,
      "step": 70225
    },
    {
      "epoch": 7.726072607260726,
      "grad_norm": 0.0006866455078125,
      "learning_rate": 0.02311249747788705,
      "loss": 0.2303,
      "num_input_tokens_seen": 14821344,
      "step": 70230
    },
    {
      "epoch": 7.726622662266227,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.023111286179285637,
      "loss": 0.2314,
      "num_input_tokens_seen": 14822464,
      "step": 70235
    },
    {
      "epoch": 7.727172717271727,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.023110074805926396,
      "loss": 0.2319,
      "num_input_tokens_seen": 14823616,
      "step": 70240
    },
    {
      "epoch": 7.727722772277228,
      "grad_norm": 0.00054168701171875,
      "learning_rate": 0.023108863357820482,
      "loss": 0.2308,
      "num_input_tokens_seen": 14824640,
      "step": 70245
    },
    {
      "epoch": 7.728272827282728,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.023107651834979065,
      "loss": 0.2324,
      "num_input_tokens_seen": 14825728,
      "step": 70250
    },
    {
      "epoch": 7.728822882288229,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023106440237413317,
      "loss": 0.2319,
      "num_input_tokens_seen": 14826784,
      "step": 70255
    },
    {
      "epoch": 7.729372937293729,
      "grad_norm": 0.00070953369140625,
      "learning_rate": 0.023105228565134396,
      "loss": 0.2319,
      "num_input_tokens_seen": 14827840,
      "step": 70260
    },
    {
      "epoch": 7.72992299229923,
      "grad_norm": 0.00063323974609375,
      "learning_rate": 0.023104016818153474,
      "loss": 0.2298,
      "num_input_tokens_seen": 14828896,
      "step": 70265
    },
    {
      "epoch": 7.730473047304731,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.023102804996481714,
      "loss": 0.2303,
      "num_input_tokens_seen": 14829920,
      "step": 70270
    },
    {
      "epoch": 7.731023102310231,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02310159310013029,
      "loss": 0.2309,
      "num_input_tokens_seen": 14831008,
      "step": 70275
    },
    {
      "epoch": 7.731573157315731,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02310038112911037,
      "loss": 0.2314,
      "num_input_tokens_seen": 14832064,
      "step": 70280
    },
    {
      "epoch": 7.732123212321232,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.023099169083433124,
      "loss": 0.2298,
      "num_input_tokens_seen": 14833120,
      "step": 70285
    },
    {
      "epoch": 7.732673267326732,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.023097956963109723,
      "loss": 0.2308,
      "num_input_tokens_seen": 14834080,
      "step": 70290
    },
    {
      "epoch": 7.7332233223322335,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.023096744768151337,
      "loss": 0.2303,
      "num_input_tokens_seen": 14835072,
      "step": 70295
    },
    {
      "epoch": 7.733773377337734,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.023095532498569146,
      "loss": 0.2319,
      "num_input_tokens_seen": 14836160,
      "step": 70300
    },
    {
      "epoch": 7.734323432343234,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02309432015437431,
      "loss": 0.2335,
      "num_input_tokens_seen": 14837216,
      "step": 70305
    },
    {
      "epoch": 7.734873487348735,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.023093107735578013,
      "loss": 0.2324,
      "num_input_tokens_seen": 14838272,
      "step": 70310
    },
    {
      "epoch": 7.735423542354235,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02309189524219142,
      "loss": 0.2319,
      "num_input_tokens_seen": 14839264,
      "step": 70315
    },
    {
      "epoch": 7.735973597359736,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.023090682674225718,
      "loss": 0.2314,
      "num_input_tokens_seen": 14840352,
      "step": 70320
    },
    {
      "epoch": 7.7365236523652365,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.023089470031692073,
      "loss": 0.2303,
      "num_input_tokens_seen": 14841408,
      "step": 70325
    },
    {
      "epoch": 7.737073707370737,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.023088257314601665,
      "loss": 0.2314,
      "num_input_tokens_seen": 14842496,
      "step": 70330
    },
    {
      "epoch": 7.737623762376238,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02308704452296567,
      "loss": 0.2308,
      "num_input_tokens_seen": 14843520,
      "step": 70335
    },
    {
      "epoch": 7.738173817381738,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.023085831656795265,
      "loss": 0.2308,
      "num_input_tokens_seen": 14844544,
      "step": 70340
    },
    {
      "epoch": 7.738723872387238,
      "grad_norm": 0.0007171630859375,
      "learning_rate": 0.023084618716101627,
      "loss": 0.2308,
      "num_input_tokens_seen": 14845600,
      "step": 70345
    },
    {
      "epoch": 7.739273927392739,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.023083405700895947,
      "loss": 0.2329,
      "num_input_tokens_seen": 14846624,
      "step": 70350
    },
    {
      "epoch": 7.7398239823982395,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02308219261118939,
      "loss": 0.2314,
      "num_input_tokens_seen": 14847712,
      "step": 70355
    },
    {
      "epoch": 7.740374037403741,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02308097944699314,
      "loss": 0.2324,
      "num_input_tokens_seen": 14848736,
      "step": 70360
    },
    {
      "epoch": 7.740924092409241,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02307976620831838,
      "loss": 0.2324,
      "num_input_tokens_seen": 14849792,
      "step": 70365
    },
    {
      "epoch": 7.741474147414742,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.023078552895176296,
      "loss": 0.2319,
      "num_input_tokens_seen": 14850848,
      "step": 70370
    },
    {
      "epoch": 7.742024202420242,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.023077339507578058,
      "loss": 0.2309,
      "num_input_tokens_seen": 14851904,
      "step": 70375
    },
    {
      "epoch": 7.742574257425742,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.023076126045534873,
      "loss": 0.2324,
      "num_input_tokens_seen": 14852928,
      "step": 70380
    },
    {
      "epoch": 7.743124312431243,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.0230749125090579,
      "loss": 0.2294,
      "num_input_tokens_seen": 14853984,
      "step": 70385
    },
    {
      "epoch": 7.743674367436744,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02307369889815833,
      "loss": 0.2314,
      "num_input_tokens_seen": 14854944,
      "step": 70390
    },
    {
      "epoch": 7.744224422442244,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.023072485212847358,
      "loss": 0.2324,
      "num_input_tokens_seen": 14856000,
      "step": 70395
    },
    {
      "epoch": 7.744774477447745,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.023071271453136158,
      "loss": 0.2309,
      "num_input_tokens_seen": 14857056,
      "step": 70400
    },
    {
      "epoch": 7.745324532453245,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.023070057619035925,
      "loss": 0.2319,
      "num_input_tokens_seen": 14858176,
      "step": 70405
    },
    {
      "epoch": 7.745874587458746,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02306884371055784,
      "loss": 0.2324,
      "num_input_tokens_seen": 14859200,
      "step": 70410
    },
    {
      "epoch": 7.7464246424642464,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.023067629727713097,
      "loss": 0.2313,
      "num_input_tokens_seen": 14860288,
      "step": 70415
    },
    {
      "epoch": 7.746974697469747,
      "grad_norm": 0.0004215240478515625,
      "learning_rate": 0.02306641567051288,
      "loss": 0.2319,
      "num_input_tokens_seen": 14861344,
      "step": 70420
    },
    {
      "epoch": 7.747524752475248,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.023065201538968384,
      "loss": 0.2308,
      "num_input_tokens_seen": 14862304,
      "step": 70425
    },
    {
      "epoch": 7.748074807480748,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.023063987333090796,
      "loss": 0.2309,
      "num_input_tokens_seen": 14863392,
      "step": 70430
    },
    {
      "epoch": 7.748624862486249,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.023062773052891304,
      "loss": 0.2303,
      "num_input_tokens_seen": 14864384,
      "step": 70435
    },
    {
      "epoch": 7.749174917491749,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.023061558698381102,
      "loss": 0.2314,
      "num_input_tokens_seen": 14865472,
      "step": 70440
    },
    {
      "epoch": 7.7497249724972495,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.023060344269571383,
      "loss": 0.2309,
      "num_input_tokens_seen": 14866496,
      "step": 70445
    },
    {
      "epoch": 7.7502750275027505,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023059129766473337,
      "loss": 0.2314,
      "num_input_tokens_seen": 14867456,
      "step": 70450
    },
    {
      "epoch": 7.750825082508251,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.023057915189098156,
      "loss": 0.2314,
      "num_input_tokens_seen": 14868480,
      "step": 70455
    },
    {
      "epoch": 7.751375137513751,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02305670053745704,
      "loss": 0.2314,
      "num_input_tokens_seen": 14869536,
      "step": 70460
    },
    {
      "epoch": 7.751925192519252,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.023055485811561183,
      "loss": 0.2314,
      "num_input_tokens_seen": 14870624,
      "step": 70465
    },
    {
      "epoch": 7.752475247524752,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.023054271011421774,
      "loss": 0.2304,
      "num_input_tokens_seen": 14871680,
      "step": 70470
    },
    {
      "epoch": 7.753025302530253,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02305305613705002,
      "loss": 0.2314,
      "num_input_tokens_seen": 14872672,
      "step": 70475
    },
    {
      "epoch": 7.7535753575357536,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.023051841188457108,
      "loss": 0.2319,
      "num_input_tokens_seen": 14873728,
      "step": 70480
    },
    {
      "epoch": 7.754125412541254,
      "grad_norm": 0.00048065185546875,
      "learning_rate": 0.02305062616565424,
      "loss": 0.2314,
      "num_input_tokens_seen": 14874816,
      "step": 70485
    },
    {
      "epoch": 7.754675467546755,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.023049411068652617,
      "loss": 0.2314,
      "num_input_tokens_seen": 14875872,
      "step": 70490
    },
    {
      "epoch": 7.755225522552255,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.02304819589746343,
      "loss": 0.2325,
      "num_input_tokens_seen": 14876896,
      "step": 70495
    },
    {
      "epoch": 7.755775577557756,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.023046980652097888,
      "loss": 0.2324,
      "num_input_tokens_seen": 14877888,
      "step": 70500
    },
    {
      "epoch": 7.756325632563256,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.02304576533256718,
      "loss": 0.2314,
      "num_input_tokens_seen": 14878912,
      "step": 70505
    },
    {
      "epoch": 7.756875687568757,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.023044549938882518,
      "loss": 0.2324,
      "num_input_tokens_seen": 14879968,
      "step": 70510
    },
    {
      "epoch": 7.757425742574258,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.0230433344710551,
      "loss": 0.233,
      "num_input_tokens_seen": 14880960,
      "step": 70515
    },
    {
      "epoch": 7.757975797579758,
      "grad_norm": 0.000476837158203125,
      "learning_rate": 0.023042118929096127,
      "loss": 0.2319,
      "num_input_tokens_seen": 14882048,
      "step": 70520
    },
    {
      "epoch": 7.758525852585258,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.0230409033130168,
      "loss": 0.2324,
      "num_input_tokens_seen": 14883104,
      "step": 70525
    },
    {
      "epoch": 7.759075907590759,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02303968762282833,
      "loss": 0.2314,
      "num_input_tokens_seen": 14884128,
      "step": 70530
    },
    {
      "epoch": 7.759625962596259,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.023038471858541913,
      "loss": 0.2309,
      "num_input_tokens_seen": 14885184,
      "step": 70535
    },
    {
      "epoch": 7.7601760176017605,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.023037256020168758,
      "loss": 0.2309,
      "num_input_tokens_seen": 14886240,
      "step": 70540
    },
    {
      "epoch": 7.760726072607261,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.02303604010772008,
      "loss": 0.2294,
      "num_input_tokens_seen": 14887296,
      "step": 70545
    },
    {
      "epoch": 7.761276127612762,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.023034824121207064,
      "loss": 0.2309,
      "num_input_tokens_seen": 14888352,
      "step": 70550
    },
    {
      "epoch": 7.761826182618262,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02303360806064094,
      "loss": 0.2304,
      "num_input_tokens_seen": 14889376,
      "step": 70555
    },
    {
      "epoch": 7.762376237623762,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.023032391926032895,
      "loss": 0.2314,
      "num_input_tokens_seen": 14890400,
      "step": 70560
    },
    {
      "epoch": 7.762926292629263,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.023031175717394152,
      "loss": 0.2304,
      "num_input_tokens_seen": 14891456,
      "step": 70565
    },
    {
      "epoch": 7.7634763476347635,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02302995943473592,
      "loss": 0.2309,
      "num_input_tokens_seen": 14892512,
      "step": 70570
    },
    {
      "epoch": 7.764026402640264,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.023028743078069402,
      "loss": 0.2314,
      "num_input_tokens_seen": 14893568,
      "step": 70575
    },
    {
      "epoch": 7.764576457645765,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.023027526647405813,
      "loss": 0.233,
      "num_input_tokens_seen": 14894624,
      "step": 70580
    },
    {
      "epoch": 7.765126512651265,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.02302631014275636,
      "loss": 0.2309,
      "num_input_tokens_seen": 14895648,
      "step": 70585
    },
    {
      "epoch": 7.765676567656766,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02302509356413226,
      "loss": 0.2319,
      "num_input_tokens_seen": 14896736,
      "step": 70590
    },
    {
      "epoch": 7.766226622662266,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.023023876911544727,
      "loss": 0.2299,
      "num_input_tokens_seen": 14897792,
      "step": 70595
    },
    {
      "epoch": 7.7667766776677665,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.023022660185004967,
      "loss": 0.2325,
      "num_input_tokens_seen": 14898816,
      "step": 70600
    },
    {
      "epoch": 7.767326732673268,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.023021443384524196,
      "loss": 0.2283,
      "num_input_tokens_seen": 14899872,
      "step": 70605
    },
    {
      "epoch": 7.767876787678768,
      "grad_norm": 0.000667572021484375,
      "learning_rate": 0.023020226510113626,
      "loss": 0.2299,
      "num_input_tokens_seen": 14900864,
      "step": 70610
    },
    {
      "epoch": 7.768426842684269,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.023019009561784487,
      "loss": 0.2309,
      "num_input_tokens_seen": 14901920,
      "step": 70615
    },
    {
      "epoch": 7.768976897689769,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.023017792539547978,
      "loss": 0.2341,
      "num_input_tokens_seen": 14902944,
      "step": 70620
    },
    {
      "epoch": 7.769526952695269,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.023016575443415326,
      "loss": 0.2289,
      "num_input_tokens_seen": 14904000,
      "step": 70625
    },
    {
      "epoch": 7.77007700770077,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.023015358273397745,
      "loss": 0.232,
      "num_input_tokens_seen": 14905088,
      "step": 70630
    },
    {
      "epoch": 7.770627062706271,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.02301414102950645,
      "loss": 0.232,
      "num_input_tokens_seen": 14906176,
      "step": 70635
    },
    {
      "epoch": 7.771177117711771,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02301292371175267,
      "loss": 0.231,
      "num_input_tokens_seen": 14907168,
      "step": 70640
    },
    {
      "epoch": 7.771727172717272,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.02301170632014761,
      "loss": 0.2299,
      "num_input_tokens_seen": 14908288,
      "step": 70645
    },
    {
      "epoch": 7.772277227722772,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.023010488854702495,
      "loss": 0.233,
      "num_input_tokens_seen": 14909408,
      "step": 70650
    },
    {
      "epoch": 7.772827282728273,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.023009271315428553,
      "loss": 0.2298,
      "num_input_tokens_seen": 14910400,
      "step": 70655
    },
    {
      "epoch": 7.773377337733773,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.023008053702337,
      "loss": 0.2309,
      "num_input_tokens_seen": 14911456,
      "step": 70660
    },
    {
      "epoch": 7.773927392739274,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02300683601543906,
      "loss": 0.2325,
      "num_input_tokens_seen": 14912480,
      "step": 70665
    },
    {
      "epoch": 7.774477447744775,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02300561825474595,
      "loss": 0.2309,
      "num_input_tokens_seen": 14913536,
      "step": 70670
    },
    {
      "epoch": 7.775027502750275,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.023004400420268904,
      "loss": 0.2314,
      "num_input_tokens_seen": 14914592,
      "step": 70675
    },
    {
      "epoch": 7.775577557755776,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.023003182512019132,
      "loss": 0.233,
      "num_input_tokens_seen": 14915616,
      "step": 70680
    },
    {
      "epoch": 7.776127612761276,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.023001964530007874,
      "loss": 0.2293,
      "num_input_tokens_seen": 14916640,
      "step": 70685
    },
    {
      "epoch": 7.776677667766776,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.023000746474246346,
      "loss": 0.2314,
      "num_input_tokens_seen": 14917664,
      "step": 70690
    },
    {
      "epoch": 7.7772277227722775,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.02299952834474578,
      "loss": 0.232,
      "num_input_tokens_seen": 14918752,
      "step": 70695
    },
    {
      "epoch": 7.777777777777778,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.022998310141517393,
      "loss": 0.2299,
      "num_input_tokens_seen": 14919744,
      "step": 70700
    },
    {
      "epoch": 7.778327832783278,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.02299709186457243,
      "loss": 0.2268,
      "num_input_tokens_seen": 14920800,
      "step": 70705
    },
    {
      "epoch": 7.778877887788779,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.022995873513922092,
      "loss": 0.2331,
      "num_input_tokens_seen": 14921856,
      "step": 70710
    },
    {
      "epoch": 7.779427942794279,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.022994655089577643,
      "loss": 0.2279,
      "num_input_tokens_seen": 14922880,
      "step": 70715
    },
    {
      "epoch": 7.77997799779978,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.022993436591550283,
      "loss": 0.229,
      "num_input_tokens_seen": 14923840,
      "step": 70720
    },
    {
      "epoch": 7.7805280528052805,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.022992218019851258,
      "loss": 0.2295,
      "num_input_tokens_seen": 14924896,
      "step": 70725
    },
    {
      "epoch": 7.781078107810782,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.022990999374491793,
      "loss": 0.2337,
      "num_input_tokens_seen": 14925984,
      "step": 70730
    },
    {
      "epoch": 7.781628162816282,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.02298978065548312,
      "loss": 0.23,
      "num_input_tokens_seen": 14927040,
      "step": 70735
    },
    {
      "epoch": 7.782178217821782,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.022988561862836477,
      "loss": 0.2337,
      "num_input_tokens_seen": 14928064,
      "step": 70740
    },
    {
      "epoch": 7.782728272827283,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.022987342996563086,
      "loss": 0.2269,
      "num_input_tokens_seen": 14929120,
      "step": 70745
    },
    {
      "epoch": 7.783278327832783,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.022986124056674193,
      "loss": 0.231,
      "num_input_tokens_seen": 14930176,
      "step": 70750
    },
    {
      "epoch": 7.7838283828382835,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02298490504318102,
      "loss": 0.2321,
      "num_input_tokens_seen": 14931232,
      "step": 70755
    },
    {
      "epoch": 7.784378437843785,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.022983685956094818,
      "loss": 0.2295,
      "num_input_tokens_seen": 14932256,
      "step": 70760
    },
    {
      "epoch": 7.784928492849285,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.022982466795426808,
      "loss": 0.2296,
      "num_input_tokens_seen": 14933280,
      "step": 70765
    },
    {
      "epoch": 7.785478547854785,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.022981247561188228,
      "loss": 0.2373,
      "num_input_tokens_seen": 14934368,
      "step": 70770
    },
    {
      "epoch": 7.786028602860286,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.022980028253390326,
      "loss": 0.2285,
      "num_input_tokens_seen": 14935488,
      "step": 70775
    },
    {
      "epoch": 7.786578657865786,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.022978808872044332,
      "loss": 0.23,
      "num_input_tokens_seen": 14936512,
      "step": 70780
    },
    {
      "epoch": 7.787128712871287,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.022977589417161486,
      "loss": 0.226,
      "num_input_tokens_seen": 14937504,
      "step": 70785
    },
    {
      "epoch": 7.787678767876788,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02297636988875302,
      "loss": 0.2316,
      "num_input_tokens_seen": 14938624,
      "step": 70790
    },
    {
      "epoch": 7.788228822882289,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.022975150286830188,
      "loss": 0.2338,
      "num_input_tokens_seen": 14939712,
      "step": 70795
    },
    {
      "epoch": 7.788778877887789,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.022973930611404215,
      "loss": 0.2337,
      "num_input_tokens_seen": 14940800,
      "step": 70800
    },
    {
      "epoch": 7.789328932893289,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.022972710862486357,
      "loss": 0.2327,
      "num_input_tokens_seen": 14941888,
      "step": 70805
    },
    {
      "epoch": 7.78987898789879,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.022971491040087846,
      "loss": 0.2322,
      "num_input_tokens_seen": 14942912,
      "step": 70810
    },
    {
      "epoch": 7.79042904290429,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.022970271144219925,
      "loss": 0.2274,
      "num_input_tokens_seen": 14943936,
      "step": 70815
    },
    {
      "epoch": 7.790979097909791,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02296905117489384,
      "loss": 0.2323,
      "num_input_tokens_seen": 14944992,
      "step": 70820
    },
    {
      "epoch": 7.791529152915292,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.022967831132120833,
      "loss": 0.2254,
      "num_input_tokens_seen": 14946080,
      "step": 70825
    },
    {
      "epoch": 7.792079207920792,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.022966611015912155,
      "loss": 0.2302,
      "num_input_tokens_seen": 14947168,
      "step": 70830
    },
    {
      "epoch": 7.792629262926293,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02296539082627904,
      "loss": 0.235,
      "num_input_tokens_seen": 14948224,
      "step": 70835
    },
    {
      "epoch": 7.793179317931793,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.022964170563232744,
      "loss": 0.2283,
      "num_input_tokens_seen": 14949248,
      "step": 70840
    },
    {
      "epoch": 7.793729372937293,
      "grad_norm": 0.007080078125,
      "learning_rate": 0.02296295022678451,
      "loss": 0.2324,
      "num_input_tokens_seen": 14950336,
      "step": 70845
    },
    {
      "epoch": 7.7942794279427945,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.02296172981694558,
      "loss": 0.233,
      "num_input_tokens_seen": 14951360,
      "step": 70850
    },
    {
      "epoch": 7.794829482948295,
      "grad_norm": 0.01324462890625,
      "learning_rate": 0.02296050933372721,
      "loss": 0.2377,
      "num_input_tokens_seen": 14952384,
      "step": 70855
    },
    {
      "epoch": 7.795379537953796,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02295928877714065,
      "loss": 0.2371,
      "num_input_tokens_seen": 14953440,
      "step": 70860
    },
    {
      "epoch": 7.795929592959296,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.022958068147197138,
      "loss": 0.2338,
      "num_input_tokens_seen": 14954528,
      "step": 70865
    },
    {
      "epoch": 7.796479647964796,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.02295684744390793,
      "loss": 0.2353,
      "num_input_tokens_seen": 14955584,
      "step": 70870
    },
    {
      "epoch": 7.797029702970297,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02295562666728428,
      "loss": 0.2311,
      "num_input_tokens_seen": 14956736,
      "step": 70875
    },
    {
      "epoch": 7.7975797579757975,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.022954405817337438,
      "loss": 0.2378,
      "num_input_tokens_seen": 14957760,
      "step": 70880
    },
    {
      "epoch": 7.798129812981298,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.02295318489407865,
      "loss": 0.2346,
      "num_input_tokens_seen": 14958848,
      "step": 70885
    },
    {
      "epoch": 7.798679867986799,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.02295196389751918,
      "loss": 0.2325,
      "num_input_tokens_seen": 14959840,
      "step": 70890
    },
    {
      "epoch": 7.799229922992299,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.022950742827670265,
      "loss": 0.2304,
      "num_input_tokens_seen": 14960928,
      "step": 70895
    },
    {
      "epoch": 7.7997799779978,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.022949521684543173,
      "loss": 0.2293,
      "num_input_tokens_seen": 14961952,
      "step": 70900
    },
    {
      "epoch": 7.8003300330033,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.022948300468149156,
      "loss": 0.2293,
      "num_input_tokens_seen": 14963104,
      "step": 70905
    },
    {
      "epoch": 7.8008800880088005,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.022947079178499467,
      "loss": 0.235,
      "num_input_tokens_seen": 14964096,
      "step": 70910
    },
    {
      "epoch": 7.801430143014302,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.022945857815605358,
      "loss": 0.2308,
      "num_input_tokens_seen": 14965152,
      "step": 70915
    },
    {
      "epoch": 7.801980198019802,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0229446363794781,
      "loss": 0.2319,
      "num_input_tokens_seen": 14966176,
      "step": 70920
    },
    {
      "epoch": 7.802530253025303,
      "grad_norm": 0.000701904296875,
      "learning_rate": 0.02294341487012893,
      "loss": 0.2313,
      "num_input_tokens_seen": 14967264,
      "step": 70925
    },
    {
      "epoch": 7.803080308030803,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.022942193287569124,
      "loss": 0.2309,
      "num_input_tokens_seen": 14968352,
      "step": 70930
    },
    {
      "epoch": 7.803630363036303,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.022940971631809932,
      "loss": 0.2298,
      "num_input_tokens_seen": 14969408,
      "step": 70935
    },
    {
      "epoch": 7.804180418041804,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.02293974990286261,
      "loss": 0.2334,
      "num_input_tokens_seen": 14970432,
      "step": 70940
    },
    {
      "epoch": 7.804730473047305,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.022938528100738428,
      "loss": 0.2318,
      "num_input_tokens_seen": 14971520,
      "step": 70945
    },
    {
      "epoch": 7.805280528052805,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.022937306225448643,
      "loss": 0.2308,
      "num_input_tokens_seen": 14972576,
      "step": 70950
    },
    {
      "epoch": 7.805830583058306,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.022936084277004512,
      "loss": 0.2318,
      "num_input_tokens_seen": 14973568,
      "step": 70955
    },
    {
      "epoch": 7.806380638063806,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.022934862255417302,
      "loss": 0.2308,
      "num_input_tokens_seen": 14974656,
      "step": 70960
    },
    {
      "epoch": 7.806930693069307,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.022933640160698272,
      "loss": 0.2319,
      "num_input_tokens_seen": 14975680,
      "step": 70965
    },
    {
      "epoch": 7.807480748074807,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02293241799285869,
      "loss": 0.2308,
      "num_input_tokens_seen": 14976768,
      "step": 70970
    },
    {
      "epoch": 7.8080308030803085,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.022931195751909817,
      "loss": 0.2287,
      "num_input_tokens_seen": 14977856,
      "step": 70975
    },
    {
      "epoch": 7.808580858085809,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02292997343786292,
      "loss": 0.2303,
      "num_input_tokens_seen": 14978944,
      "step": 70980
    },
    {
      "epoch": 7.809130913091309,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.02292875105072926,
      "loss": 0.2319,
      "num_input_tokens_seen": 14979936,
      "step": 70985
    },
    {
      "epoch": 7.80968096809681,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.022927528590520108,
      "loss": 0.2335,
      "num_input_tokens_seen": 14980992,
      "step": 70990
    },
    {
      "epoch": 7.81023102310231,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.022926306057246725,
      "loss": 0.2303,
      "num_input_tokens_seen": 14982080,
      "step": 70995
    },
    {
      "epoch": 7.81078107810781,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.022925083450920387,
      "loss": 0.2324,
      "num_input_tokens_seen": 14983168,
      "step": 71000
    },
    {
      "epoch": 7.8113311331133115,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.022923860771552354,
      "loss": 0.2303,
      "num_input_tokens_seen": 14984256,
      "step": 71005
    },
    {
      "epoch": 7.811881188118812,
      "grad_norm": 0.00078582763671875,
      "learning_rate": 0.022922638019153903,
      "loss": 0.2308,
      "num_input_tokens_seen": 14985312,
      "step": 71010
    },
    {
      "epoch": 7.812431243124313,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.0229214151937363,
      "loss": 0.2324,
      "num_input_tokens_seen": 14986336,
      "step": 71015
    },
    {
      "epoch": 7.812981298129813,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.022920192295310806,
      "loss": 0.2303,
      "num_input_tokens_seen": 14987456,
      "step": 71020
    },
    {
      "epoch": 7.813531353135313,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02291896932388871,
      "loss": 0.2318,
      "num_input_tokens_seen": 14988608,
      "step": 71025
    },
    {
      "epoch": 7.814081408140814,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.022917746279481264,
      "loss": 0.2314,
      "num_input_tokens_seen": 14989664,
      "step": 71030
    },
    {
      "epoch": 7.8146314631463145,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.022916523162099755,
      "loss": 0.2329,
      "num_input_tokens_seen": 14990688,
      "step": 71035
    },
    {
      "epoch": 7.815181518151816,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.022915299971755454,
      "loss": 0.2324,
      "num_input_tokens_seen": 14991712,
      "step": 71040
    },
    {
      "epoch": 7.815731573157316,
      "grad_norm": 0.009765625,
      "learning_rate": 0.022914076708459626,
      "loss": 0.2298,
      "num_input_tokens_seen": 14992800,
      "step": 71045
    },
    {
      "epoch": 7.816281628162816,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.022912853372223554,
      "loss": 0.2324,
      "num_input_tokens_seen": 14993824,
      "step": 71050
    },
    {
      "epoch": 7.816831683168317,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.022911629963058507,
      "loss": 0.2308,
      "num_input_tokens_seen": 14994912,
      "step": 71055
    },
    {
      "epoch": 7.817381738173817,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.022910406480975764,
      "loss": 0.2324,
      "num_input_tokens_seen": 14995968,
      "step": 71060
    },
    {
      "epoch": 7.8179317931793175,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.0229091829259866,
      "loss": 0.2313,
      "num_input_tokens_seen": 14997024,
      "step": 71065
    },
    {
      "epoch": 7.818481848184819,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02290795929810229,
      "loss": 0.2314,
      "num_input_tokens_seen": 14998048,
      "step": 71070
    },
    {
      "epoch": 7.819031903190319,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02290673559733412,
      "loss": 0.2329,
      "num_input_tokens_seen": 14999136,
      "step": 71075
    },
    {
      "epoch": 7.81958195819582,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.022905511823693357,
      "loss": 0.2298,
      "num_input_tokens_seen": 15000160,
      "step": 71080
    },
    {
      "epoch": 7.82013201320132,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.022904287977191285,
      "loss": 0.2303,
      "num_input_tokens_seen": 15001216,
      "step": 71085
    },
    {
      "epoch": 7.82068206820682,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.022903064057839182,
      "loss": 0.2371,
      "num_input_tokens_seen": 15002304,
      "step": 71090
    },
    {
      "epoch": 7.821232123212321,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.022901840065648337,
      "loss": 0.2314,
      "num_input_tokens_seen": 15003360,
      "step": 71095
    },
    {
      "epoch": 7.821782178217822,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.022900616000630012,
      "loss": 0.2308,
      "num_input_tokens_seen": 15004416,
      "step": 71100
    },
    {
      "epoch": 7.822332233223323,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02289939186279551,
      "loss": 0.2314,
      "num_input_tokens_seen": 15005472,
      "step": 71105
    },
    {
      "epoch": 7.822882288228823,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.022898167652156104,
      "loss": 0.2304,
      "num_input_tokens_seen": 15006560,
      "step": 71110
    },
    {
      "epoch": 7.823432343234323,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.022896943368723076,
      "loss": 0.2309,
      "num_input_tokens_seen": 15007648,
      "step": 71115
    },
    {
      "epoch": 7.823982398239824,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.022895719012507705,
      "loss": 0.2309,
      "num_input_tokens_seen": 15008672,
      "step": 71120
    },
    {
      "epoch": 7.824532453245324,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.022894494583521288,
      "loss": 0.2293,
      "num_input_tokens_seen": 15009760,
      "step": 71125
    },
    {
      "epoch": 7.825082508250825,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.022893270081775094,
      "loss": 0.2335,
      "num_input_tokens_seen": 15010816,
      "step": 71130
    },
    {
      "epoch": 7.825632563256326,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.02289204550728042,
      "loss": 0.2304,
      "num_input_tokens_seen": 15011872,
      "step": 71135
    },
    {
      "epoch": 7.826182618261826,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.022890820860048556,
      "loss": 0.234,
      "num_input_tokens_seen": 15012928,
      "step": 71140
    },
    {
      "epoch": 7.826732673267327,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.022889596140090774,
      "loss": 0.2362,
      "num_input_tokens_seen": 15013952,
      "step": 71145
    },
    {
      "epoch": 7.827282728272827,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.022888371347418373,
      "loss": 0.2314,
      "num_input_tokens_seen": 15015008,
      "step": 71150
    },
    {
      "epoch": 7.827832783278328,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02288714648204264,
      "loss": 0.2314,
      "num_input_tokens_seen": 15016032,
      "step": 71155
    },
    {
      "epoch": 7.8283828382838285,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02288592154397486,
      "loss": 0.2314,
      "num_input_tokens_seen": 15017120,
      "step": 71160
    },
    {
      "epoch": 7.828932893289329,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02288469653322633,
      "loss": 0.234,
      "num_input_tokens_seen": 15018176,
      "step": 71165
    },
    {
      "epoch": 7.82948294829483,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.022883471449808332,
      "loss": 0.232,
      "num_input_tokens_seen": 15019232,
      "step": 71170
    },
    {
      "epoch": 7.83003300330033,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02288224629373216,
      "loss": 0.234,
      "num_input_tokens_seen": 15020288,
      "step": 71175
    },
    {
      "epoch": 7.83058305830583,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.022881021065009107,
      "loss": 0.2319,
      "num_input_tokens_seen": 15021344,
      "step": 71180
    },
    {
      "epoch": 7.831133113311331,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.022879795763650464,
      "loss": 0.2288,
      "num_input_tokens_seen": 15022368,
      "step": 71185
    },
    {
      "epoch": 7.8316831683168315,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.022878570389667525,
      "loss": 0.2325,
      "num_input_tokens_seen": 15023456,
      "step": 71190
    },
    {
      "epoch": 7.832233223322332,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.022877344943071583,
      "loss": 0.2299,
      "num_input_tokens_seen": 15024512,
      "step": 71195
    },
    {
      "epoch": 7.832783278327833,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.022876119423873933,
      "loss": 0.2299,
      "num_input_tokens_seen": 15025600,
      "step": 71200
    },
    {
      "epoch": 7.833333333333333,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.022874893832085873,
      "loss": 0.2299,
      "num_input_tokens_seen": 15026688,
      "step": 71205
    },
    {
      "epoch": 7.833883388338834,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02287366816771869,
      "loss": 0.2294,
      "num_input_tokens_seen": 15027680,
      "step": 71210
    },
    {
      "epoch": 7.834433443344334,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.022872442430783686,
      "loss": 0.2342,
      "num_input_tokens_seen": 15028832,
      "step": 71215
    },
    {
      "epoch": 7.834983498349835,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02287121662129216,
      "loss": 0.2269,
      "num_input_tokens_seen": 15029856,
      "step": 71220
    },
    {
      "epoch": 7.835533553355336,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.022869990739255407,
      "loss": 0.2341,
      "num_input_tokens_seen": 15030880,
      "step": 71225
    },
    {
      "epoch": 7.836083608360836,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.022868764784684727,
      "loss": 0.2285,
      "num_input_tokens_seen": 15031936,
      "step": 71230
    },
    {
      "epoch": 7.836633663366337,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.022867538757591416,
      "loss": 0.2305,
      "num_input_tokens_seen": 15032960,
      "step": 71235
    },
    {
      "epoch": 7.837183718371837,
      "grad_norm": 0.000659942626953125,
      "learning_rate": 0.022866312657986772,
      "loss": 0.227,
      "num_input_tokens_seen": 15034016,
      "step": 71240
    },
    {
      "epoch": 7.837733773377337,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.022865086485882105,
      "loss": 0.2331,
      "num_input_tokens_seen": 15035072,
      "step": 71245
    },
    {
      "epoch": 7.838283828382838,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.022863860241288704,
      "loss": 0.2316,
      "num_input_tokens_seen": 15036160,
      "step": 71250
    },
    {
      "epoch": 7.838833883388339,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.022862633924217883,
      "loss": 0.229,
      "num_input_tokens_seen": 15037216,
      "step": 71255
    },
    {
      "epoch": 7.83938393839384,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 0.022861407534680935,
      "loss": 0.2322,
      "num_input_tokens_seen": 15038176,
      "step": 71260
    },
    {
      "epoch": 7.83993399339934,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.022860181072689164,
      "loss": 0.2337,
      "num_input_tokens_seen": 15039264,
      "step": 71265
    },
    {
      "epoch": 7.84048404840484,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.02285895453825388,
      "loss": 0.2274,
      "num_input_tokens_seen": 15040320,
      "step": 71270
    },
    {
      "epoch": 7.841034103410341,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.022857727931386383,
      "loss": 0.2285,
      "num_input_tokens_seen": 15041344,
      "step": 71275
    },
    {
      "epoch": 7.841584158415841,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02285650125209797,
      "loss": 0.2321,
      "num_input_tokens_seen": 15042464,
      "step": 71280
    },
    {
      "epoch": 7.8421342134213425,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.022855274500399963,
      "loss": 0.2322,
      "num_input_tokens_seen": 15043552,
      "step": 71285
    },
    {
      "epoch": 7.842684268426843,
      "grad_norm": 0.00067138671875,
      "learning_rate": 0.022854047676303658,
      "loss": 0.2321,
      "num_input_tokens_seen": 15044544,
      "step": 71290
    },
    {
      "epoch": 7.843234323432343,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.022852820779820364,
      "loss": 0.2327,
      "num_input_tokens_seen": 15045664,
      "step": 71295
    },
    {
      "epoch": 7.843784378437844,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.022851593810961388,
      "loss": 0.2331,
      "num_input_tokens_seen": 15046848,
      "step": 71300
    },
    {
      "epoch": 7.844334433443344,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.02285036676973804,
      "loss": 0.2315,
      "num_input_tokens_seen": 15047968,
      "step": 71305
    },
    {
      "epoch": 7.8448844884488445,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.022849139656161633,
      "loss": 0.2337,
      "num_input_tokens_seen": 15049024,
      "step": 71310
    },
    {
      "epoch": 7.8454345434543455,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.022847912470243467,
      "loss": 0.2295,
      "num_input_tokens_seen": 15050080,
      "step": 71315
    },
    {
      "epoch": 7.845984598459846,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.022846685211994858,
      "loss": 0.2238,
      "num_input_tokens_seen": 15051072,
      "step": 71320
    },
    {
      "epoch": 7.846534653465347,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02284545788142712,
      "loss": 0.2363,
      "num_input_tokens_seen": 15052064,
      "step": 71325
    },
    {
      "epoch": 7.847084708470847,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02284423047855156,
      "loss": 0.2321,
      "num_input_tokens_seen": 15053120,
      "step": 71330
    },
    {
      "epoch": 7.847634763476347,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.022843003003379485,
      "loss": 0.2343,
      "num_input_tokens_seen": 15054208,
      "step": 71335
    },
    {
      "epoch": 7.848184818481848,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.022841775455922222,
      "loss": 0.2305,
      "num_input_tokens_seen": 15055200,
      "step": 71340
    },
    {
      "epoch": 7.8487348734873486,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.02284054783619108,
      "loss": 0.2363,
      "num_input_tokens_seen": 15056256,
      "step": 71345
    },
    {
      "epoch": 7.84928492849285,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.022839320144197362,
      "loss": 0.2279,
      "num_input_tokens_seen": 15057280,
      "step": 71350
    },
    {
      "epoch": 7.84983498349835,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0228380923799524,
      "loss": 0.2311,
      "num_input_tokens_seen": 15058336,
      "step": 71355
    },
    {
      "epoch": 7.85038503850385,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.0228368645434675,
      "loss": 0.2295,
      "num_input_tokens_seen": 15059424,
      "step": 71360
    },
    {
      "epoch": 7.850935093509351,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.022835636634753978,
      "loss": 0.2305,
      "num_input_tokens_seen": 15060512,
      "step": 71365
    },
    {
      "epoch": 7.851485148514851,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.02283440865382315,
      "loss": 0.2321,
      "num_input_tokens_seen": 15061632,
      "step": 71370
    },
    {
      "epoch": 7.852035203520352,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.022833180600686342,
      "loss": 0.2306,
      "num_input_tokens_seen": 15062720,
      "step": 71375
    },
    {
      "epoch": 7.852585258525853,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.022831952475354862,
      "loss": 0.2294,
      "num_input_tokens_seen": 15063776,
      "step": 71380
    },
    {
      "epoch": 7.853135313531353,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.02283072427784004,
      "loss": 0.2336,
      "num_input_tokens_seen": 15064864,
      "step": 71385
    },
    {
      "epoch": 7.853685368536854,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022829496008153188,
      "loss": 0.2305,
      "num_input_tokens_seen": 15065920,
      "step": 71390
    },
    {
      "epoch": 7.854235423542354,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.022828267666305625,
      "loss": 0.2274,
      "num_input_tokens_seen": 15067008,
      "step": 71395
    },
    {
      "epoch": 7.854785478547855,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.022827039252308678,
      "loss": 0.2331,
      "num_input_tokens_seen": 15068032,
      "step": 71400
    },
    {
      "epoch": 7.8553355335533555,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.022825810766173665,
      "loss": 0.2301,
      "num_input_tokens_seen": 15069120,
      "step": 71405
    },
    {
      "epoch": 7.855885588558856,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.022824582207911913,
      "loss": 0.2357,
      "num_input_tokens_seen": 15070176,
      "step": 71410
    },
    {
      "epoch": 7.856435643564357,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.022823353577534733,
      "loss": 0.2394,
      "num_input_tokens_seen": 15071200,
      "step": 71415
    },
    {
      "epoch": 7.856985698569857,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.022822124875053466,
      "loss": 0.2325,
      "num_input_tokens_seen": 15072256,
      "step": 71420
    },
    {
      "epoch": 7.857535753575357,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.022820896100479424,
      "loss": 0.2315,
      "num_input_tokens_seen": 15073280,
      "step": 71425
    },
    {
      "epoch": 7.858085808580858,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02281966725382394,
      "loss": 0.233,
      "num_input_tokens_seen": 15074368,
      "step": 71430
    },
    {
      "epoch": 7.8586358635863585,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.022818438335098324,
      "loss": 0.2329,
      "num_input_tokens_seen": 15075424,
      "step": 71435
    },
    {
      "epoch": 7.8591859185918596,
      "grad_norm": 0.000606536865234375,
      "learning_rate": 0.022817209344313923,
      "loss": 0.2314,
      "num_input_tokens_seen": 15076480,
      "step": 71440
    },
    {
      "epoch": 7.85973597359736,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.022815980281482053,
      "loss": 0.2324,
      "num_input_tokens_seen": 15077536,
      "step": 71445
    },
    {
      "epoch": 7.86028602860286,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.02281475114661404,
      "loss": 0.2308,
      "num_input_tokens_seen": 15078592,
      "step": 71450
    },
    {
      "epoch": 7.860836083608361,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.022813521939721217,
      "loss": 0.2324,
      "num_input_tokens_seen": 15079616,
      "step": 71455
    },
    {
      "epoch": 7.861386138613861,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.022812292660814915,
      "loss": 0.2288,
      "num_input_tokens_seen": 15080672,
      "step": 71460
    },
    {
      "epoch": 7.861936193619362,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02281106330990645,
      "loss": 0.2313,
      "num_input_tokens_seen": 15081664,
      "step": 71465
    },
    {
      "epoch": 7.862486248624863,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.02280983388700717,
      "loss": 0.2319,
      "num_input_tokens_seen": 15082720,
      "step": 71470
    },
    {
      "epoch": 7.863036303630363,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.022808604392128402,
      "loss": 0.2314,
      "num_input_tokens_seen": 15083744,
      "step": 71475
    },
    {
      "epoch": 7.863586358635864,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.02280737482528147,
      "loss": 0.2298,
      "num_input_tokens_seen": 15084768,
      "step": 71480
    },
    {
      "epoch": 7.864136413641364,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.02280614518647771,
      "loss": 0.2282,
      "num_input_tokens_seen": 15085792,
      "step": 71485
    },
    {
      "epoch": 7.864686468646864,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.02280491547572846,
      "loss": 0.2309,
      "num_input_tokens_seen": 15086848,
      "step": 71490
    },
    {
      "epoch": 7.865236523652365,
      "grad_norm": 0.00726318359375,
      "learning_rate": 0.022803685693045042,
      "loss": 0.2341,
      "num_input_tokens_seen": 15087936,
      "step": 71495
    },
    {
      "epoch": 7.865786578657866,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.022802455838438807,
      "loss": 0.2309,
      "num_input_tokens_seen": 15088992,
      "step": 71500
    },
    {
      "epoch": 7.866336633663367,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.022801225911921075,
      "loss": 0.2299,
      "num_input_tokens_seen": 15090048,
      "step": 71505
    },
    {
      "epoch": 7.866886688668867,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.022799995913503186,
      "loss": 0.231,
      "num_input_tokens_seen": 15091104,
      "step": 71510
    },
    {
      "epoch": 7.867436743674367,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.02279876584319648,
      "loss": 0.2294,
      "num_input_tokens_seen": 15092288,
      "step": 71515
    },
    {
      "epoch": 7.867986798679868,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.02279753570101229,
      "loss": 0.2336,
      "num_input_tokens_seen": 15093312,
      "step": 71520
    },
    {
      "epoch": 7.868536853685368,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.022796305486961953,
      "loss": 0.2273,
      "num_input_tokens_seen": 15094400,
      "step": 71525
    },
    {
      "epoch": 7.8690869086908695,
      "grad_norm": 0.0133056640625,
      "learning_rate": 0.022795075201056815,
      "loss": 0.2305,
      "num_input_tokens_seen": 15095456,
      "step": 71530
    },
    {
      "epoch": 7.86963696369637,
      "grad_norm": 0.007354736328125,
      "learning_rate": 0.022793844843308205,
      "loss": 0.2301,
      "num_input_tokens_seen": 15096544,
      "step": 71535
    },
    {
      "epoch": 7.87018701870187,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.02279261441372746,
      "loss": 0.2316,
      "num_input_tokens_seen": 15097536,
      "step": 71540
    },
    {
      "epoch": 7.870737073707371,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.022791383912325942,
      "loss": 0.2315,
      "num_input_tokens_seen": 15098624,
      "step": 71545
    },
    {
      "epoch": 7.871287128712871,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.022790153339114973,
      "loss": 0.2342,
      "num_input_tokens_seen": 15099744,
      "step": 71550
    },
    {
      "epoch": 7.871837183718371,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.02278892269410589,
      "loss": 0.2311,
      "num_input_tokens_seen": 15100768,
      "step": 71555
    },
    {
      "epoch": 7.8723872387238725,
      "grad_norm": 0.007080078125,
      "learning_rate": 0.022787691977310055,
      "loss": 0.2331,
      "num_input_tokens_seen": 15101888,
      "step": 71560
    },
    {
      "epoch": 7.872937293729373,
      "grad_norm": 0.01318359375,
      "learning_rate": 0.022786461188738793,
      "loss": 0.2362,
      "num_input_tokens_seen": 15102912,
      "step": 71565
    },
    {
      "epoch": 7.873487348734874,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.022785230328403457,
      "loss": 0.2315,
      "num_input_tokens_seen": 15103904,
      "step": 71570
    },
    {
      "epoch": 7.874037403740374,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.022783999396315386,
      "loss": 0.2336,
      "num_input_tokens_seen": 15104928,
      "step": 71575
    },
    {
      "epoch": 7.874587458745875,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.022782768392485934,
      "loss": 0.2305,
      "num_input_tokens_seen": 15106048,
      "step": 71580
    },
    {
      "epoch": 7.875137513751375,
      "grad_norm": 0.006591796875,
      "learning_rate": 0.022781537316926434,
      "loss": 0.2305,
      "num_input_tokens_seen": 15107072,
      "step": 71585
    },
    {
      "epoch": 7.8756875687568755,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02278030616964824,
      "loss": 0.2284,
      "num_input_tokens_seen": 15108096,
      "step": 71590
    },
    {
      "epoch": 7.876237623762377,
      "grad_norm": 0.0076904296875,
      "learning_rate": 0.022779074950662698,
      "loss": 0.2331,
      "num_input_tokens_seen": 15109152,
      "step": 71595
    },
    {
      "epoch": 7.876787678767877,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.022777843659981156,
      "loss": 0.2362,
      "num_input_tokens_seen": 15110272,
      "step": 71600
    },
    {
      "epoch": 7.877337733773377,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.02277661229761496,
      "loss": 0.232,
      "num_input_tokens_seen": 15111264,
      "step": 71605
    },
    {
      "epoch": 7.877887788778878,
      "grad_norm": 0.005859375,
      "learning_rate": 0.022775380863575456,
      "loss": 0.2356,
      "num_input_tokens_seen": 15112352,
      "step": 71610
    },
    {
      "epoch": 7.878437843784378,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.022774149357874005,
      "loss": 0.2298,
      "num_input_tokens_seen": 15113408,
      "step": 71615
    },
    {
      "epoch": 7.878987898789879,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.02277291778052195,
      "loss": 0.2314,
      "num_input_tokens_seen": 15114400,
      "step": 71620
    },
    {
      "epoch": 7.87953795379538,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.022771686131530633,
      "loss": 0.2324,
      "num_input_tokens_seen": 15115488,
      "step": 71625
    },
    {
      "epoch": 7.88008800880088,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.022770454410911422,
      "loss": 0.2298,
      "num_input_tokens_seen": 15116544,
      "step": 71630
    },
    {
      "epoch": 7.880638063806381,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.022769222618675658,
      "loss": 0.2319,
      "num_input_tokens_seen": 15117600,
      "step": 71635
    },
    {
      "epoch": 7.881188118811881,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.022767990754834697,
      "loss": 0.2303,
      "num_input_tokens_seen": 15118720,
      "step": 71640
    },
    {
      "epoch": 7.881738173817382,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.022766758819399897,
      "loss": 0.2314,
      "num_input_tokens_seen": 15119808,
      "step": 71645
    },
    {
      "epoch": 7.882288228822882,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.0227655268123826,
      "loss": 0.2314,
      "num_input_tokens_seen": 15120864,
      "step": 71650
    },
    {
      "epoch": 7.882838283828383,
      "grad_norm": 0.0004673004150390625,
      "learning_rate": 0.022764294733794176,
      "loss": 0.2313,
      "num_input_tokens_seen": 15121920,
      "step": 71655
    },
    {
      "epoch": 7.883388338833884,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02276306258364597,
      "loss": 0.2303,
      "num_input_tokens_seen": 15122944,
      "step": 71660
    },
    {
      "epoch": 7.883938393839384,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.022761830361949342,
      "loss": 0.2308,
      "num_input_tokens_seen": 15124032,
      "step": 71665
    },
    {
      "epoch": 7.884488448844884,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.02276059806871564,
      "loss": 0.2314,
      "num_input_tokens_seen": 15125056,
      "step": 71670
    },
    {
      "epoch": 7.885038503850385,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02275936570395624,
      "loss": 0.2308,
      "num_input_tokens_seen": 15126144,
      "step": 71675
    },
    {
      "epoch": 7.885588558855885,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.022758133267682487,
      "loss": 0.2319,
      "num_input_tokens_seen": 15127168,
      "step": 71680
    },
    {
      "epoch": 7.8861386138613865,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.022756900759905743,
      "loss": 0.2308,
      "num_input_tokens_seen": 15128192,
      "step": 71685
    },
    {
      "epoch": 7.886688668866887,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.022755668180637366,
      "loss": 0.2308,
      "num_input_tokens_seen": 15129280,
      "step": 71690
    },
    {
      "epoch": 7.887238723872387,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.02275443552988872,
      "loss": 0.2314,
      "num_input_tokens_seen": 15130336,
      "step": 71695
    },
    {
      "epoch": 7.887788778877888,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.022753202807671156,
      "loss": 0.2303,
      "num_input_tokens_seen": 15131360,
      "step": 71700
    },
    {
      "epoch": 7.888338833883388,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.022751970013996044,
      "loss": 0.2314,
      "num_input_tokens_seen": 15132416,
      "step": 71705
    },
    {
      "epoch": 7.888888888888889,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.022750737148874746,
      "loss": 0.2314,
      "num_input_tokens_seen": 15133440,
      "step": 71710
    },
    {
      "epoch": 7.8894389438943895,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02274950421231862,
      "loss": 0.2345,
      "num_input_tokens_seen": 15134496,
      "step": 71715
    },
    {
      "epoch": 7.88998899889989,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.02274827120433904,
      "loss": 0.2334,
      "num_input_tokens_seen": 15135520,
      "step": 71720
    },
    {
      "epoch": 7.890539053905391,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.022747038124947355,
      "loss": 0.2319,
      "num_input_tokens_seen": 15136576,
      "step": 71725
    },
    {
      "epoch": 7.891089108910891,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.022745804974154935,
      "loss": 0.2329,
      "num_input_tokens_seen": 15137632,
      "step": 71730
    },
    {
      "epoch": 7.891639163916391,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.022744571751973156,
      "loss": 0.2324,
      "num_input_tokens_seen": 15138688,
      "step": 71735
    },
    {
      "epoch": 7.892189218921892,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.022743338458413368,
      "loss": 0.2308,
      "num_input_tokens_seen": 15139776,
      "step": 71740
    },
    {
      "epoch": 7.8927392739273925,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.022742105093486946,
      "loss": 0.2303,
      "num_input_tokens_seen": 15140832,
      "step": 71745
    },
    {
      "epoch": 7.893289328932894,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.022740871657205262,
      "loss": 0.2319,
      "num_input_tokens_seen": 15141920,
      "step": 71750
    },
    {
      "epoch": 7.893839383938394,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02273963814957967,
      "loss": 0.2303,
      "num_input_tokens_seen": 15142944,
      "step": 71755
    },
    {
      "epoch": 7.894389438943895,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.022738404570621554,
      "loss": 0.2319,
      "num_input_tokens_seen": 15144032,
      "step": 71760
    },
    {
      "epoch": 7.894939493949395,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.022737170920342272,
      "loss": 0.2303,
      "num_input_tokens_seen": 15145184,
      "step": 71765
    },
    {
      "epoch": 7.895489548954895,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.022735937198753196,
      "loss": 0.2313,
      "num_input_tokens_seen": 15146240,
      "step": 71770
    },
    {
      "epoch": 7.896039603960396,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.022734703405865708,
      "loss": 0.2298,
      "num_input_tokens_seen": 15147296,
      "step": 71775
    },
    {
      "epoch": 7.896589658965897,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.022733469541691163,
      "loss": 0.2303,
      "num_input_tokens_seen": 15148352,
      "step": 71780
    },
    {
      "epoch": 7.897139713971397,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.022732235606240943,
      "loss": 0.2303,
      "num_input_tokens_seen": 15149376,
      "step": 71785
    },
    {
      "epoch": 7.897689768976898,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.02273100159952642,
      "loss": 0.2308,
      "num_input_tokens_seen": 15150432,
      "step": 71790
    },
    {
      "epoch": 7.898239823982398,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02272976752155896,
      "loss": 0.2318,
      "num_input_tokens_seen": 15151456,
      "step": 71795
    },
    {
      "epoch": 7.898789878987898,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02272853337234994,
      "loss": 0.2308,
      "num_input_tokens_seen": 15152512,
      "step": 71800
    },
    {
      "epoch": 7.899339933993399,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.022727299151910744,
      "loss": 0.2319,
      "num_input_tokens_seen": 15153632,
      "step": 71805
    },
    {
      "epoch": 7.8998899889989,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.022726064860252738,
      "loss": 0.2303,
      "num_input_tokens_seen": 15154720,
      "step": 71810
    },
    {
      "epoch": 7.900440044004401,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.022724830497387292,
      "loss": 0.2335,
      "num_input_tokens_seen": 15155776,
      "step": 71815
    },
    {
      "epoch": 7.900990099009901,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.022723596063325798,
      "loss": 0.2319,
      "num_input_tokens_seen": 15156832,
      "step": 71820
    },
    {
      "epoch": 7.901540154015402,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02272236155807962,
      "loss": 0.2298,
      "num_input_tokens_seen": 15157888,
      "step": 71825
    },
    {
      "epoch": 7.902090209020902,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.022721126981660143,
      "loss": 0.2293,
      "num_input_tokens_seen": 15159008,
      "step": 71830
    },
    {
      "epoch": 7.902640264026402,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022719892334078746,
      "loss": 0.2329,
      "num_input_tokens_seen": 15160000,
      "step": 71835
    },
    {
      "epoch": 7.9031903190319035,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.022718657615346805,
      "loss": 0.2329,
      "num_input_tokens_seen": 15161088,
      "step": 71840
    },
    {
      "epoch": 7.903740374037404,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.022717422825475696,
      "loss": 0.2313,
      "num_input_tokens_seen": 15162176,
      "step": 71845
    },
    {
      "epoch": 7.904290429042904,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.02271618796447681,
      "loss": 0.2329,
      "num_input_tokens_seen": 15163232,
      "step": 71850
    },
    {
      "epoch": 7.904840484048405,
      "grad_norm": 0.009765625,
      "learning_rate": 0.022714953032361516,
      "loss": 0.2319,
      "num_input_tokens_seen": 15164256,
      "step": 71855
    },
    {
      "epoch": 7.905390539053905,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.022713718029141208,
      "loss": 0.2303,
      "num_input_tokens_seen": 15165280,
      "step": 71860
    },
    {
      "epoch": 7.905940594059406,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.022712482954827258,
      "loss": 0.2319,
      "num_input_tokens_seen": 15166400,
      "step": 71865
    },
    {
      "epoch": 7.9064906490649065,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.022711247809431057,
      "loss": 0.2319,
      "num_input_tokens_seen": 15167392,
      "step": 71870
    },
    {
      "epoch": 7.907040704070407,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.02271001259296398,
      "loss": 0.2319,
      "num_input_tokens_seen": 15168448,
      "step": 71875
    },
    {
      "epoch": 7.907590759075908,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.022708777305437425,
      "loss": 0.2308,
      "num_input_tokens_seen": 15169600,
      "step": 71880
    },
    {
      "epoch": 7.908140814081408,
      "grad_norm": 0.000797271728515625,
      "learning_rate": 0.02270754194686276,
      "loss": 0.2298,
      "num_input_tokens_seen": 15170656,
      "step": 71885
    },
    {
      "epoch": 7.908690869086909,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.022706306517251383,
      "loss": 0.2309,
      "num_input_tokens_seen": 15171712,
      "step": 71890
    },
    {
      "epoch": 7.909240924092409,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02270507101661468,
      "loss": 0.2314,
      "num_input_tokens_seen": 15172736,
      "step": 71895
    },
    {
      "epoch": 7.9097909790979095,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02270383544496403,
      "loss": 0.2319,
      "num_input_tokens_seen": 15173824,
      "step": 71900
    },
    {
      "epoch": 7.910341034103411,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.022702599802310826,
      "loss": 0.2309,
      "num_input_tokens_seen": 15174880,
      "step": 71905
    },
    {
      "epoch": 7.910891089108911,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.022701364088666458,
      "loss": 0.2314,
      "num_input_tokens_seen": 15175904,
      "step": 71910
    },
    {
      "epoch": 7.911441144114411,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02270012830404231,
      "loss": 0.2303,
      "num_input_tokens_seen": 15176960,
      "step": 71915
    },
    {
      "epoch": 7.911991199119912,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.022698892448449772,
      "loss": 0.2314,
      "num_input_tokens_seen": 15178080,
      "step": 71920
    },
    {
      "epoch": 7.912541254125412,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.022697656521900242,
      "loss": 0.2314,
      "num_input_tokens_seen": 15179200,
      "step": 71925
    },
    {
      "epoch": 7.913091309130913,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.0226964205244051,
      "loss": 0.2314,
      "num_input_tokens_seen": 15180256,
      "step": 71930
    },
    {
      "epoch": 7.913641364136414,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.022695184455975747,
      "loss": 0.2319,
      "num_input_tokens_seen": 15181280,
      "step": 71935
    },
    {
      "epoch": 7.914191419141914,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.022693948316623573,
      "loss": 0.2298,
      "num_input_tokens_seen": 15182368,
      "step": 71940
    },
    {
      "epoch": 7.914741474147415,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.022692712106359963,
      "loss": 0.2319,
      "num_input_tokens_seen": 15183552,
      "step": 71945
    },
    {
      "epoch": 7.915291529152915,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.022691475825196327,
      "loss": 0.2314,
      "num_input_tokens_seen": 15184640,
      "step": 71950
    },
    {
      "epoch": 7.915841584158416,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02269023947314404,
      "loss": 0.2313,
      "num_input_tokens_seen": 15185664,
      "step": 71955
    },
    {
      "epoch": 7.916391639163916,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.022689003050214505,
      "loss": 0.2313,
      "num_input_tokens_seen": 15186688,
      "step": 71960
    },
    {
      "epoch": 7.916941694169417,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.022687766556419124,
      "loss": 0.2324,
      "num_input_tokens_seen": 15187808,
      "step": 71965
    },
    {
      "epoch": 7.917491749174918,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.02268652999176929,
      "loss": 0.2324,
      "num_input_tokens_seen": 15188864,
      "step": 71970
    },
    {
      "epoch": 7.918041804180418,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.02268529335627639,
      "loss": 0.2319,
      "num_input_tokens_seen": 15189920,
      "step": 71975
    },
    {
      "epoch": 7.918591859185918,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.022684056649951836,
      "loss": 0.2319,
      "num_input_tokens_seen": 15191040,
      "step": 71980
    },
    {
      "epoch": 7.919141914191419,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02268281987280701,
      "loss": 0.2298,
      "num_input_tokens_seen": 15192064,
      "step": 71985
    },
    {
      "epoch": 7.919691969196919,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02268158302485333,
      "loss": 0.233,
      "num_input_tokens_seen": 15193120,
      "step": 71990
    },
    {
      "epoch": 7.9202420242024205,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.022680346106102176,
      "loss": 0.233,
      "num_input_tokens_seen": 15194176,
      "step": 71995
    },
    {
      "epoch": 7.920792079207921,
      "grad_norm": 0.009765625,
      "learning_rate": 0.022679109116564962,
      "loss": 0.2314,
      "num_input_tokens_seen": 15195200,
      "step": 72000
    },
    {
      "epoch": 7.921342134213422,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.022677872056253084,
      "loss": 0.2324,
      "num_input_tokens_seen": 15196192,
      "step": 72005
    },
    {
      "epoch": 7.921892189218922,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.022676634925177942,
      "loss": 0.2309,
      "num_input_tokens_seen": 15197248,
      "step": 72010
    },
    {
      "epoch": 7.922442244224422,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02267539772335094,
      "loss": 0.2324,
      "num_input_tokens_seen": 15198304,
      "step": 72015
    },
    {
      "epoch": 7.922992299229923,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.022674160450783486,
      "loss": 0.2309,
      "num_input_tokens_seen": 15199328,
      "step": 72020
    },
    {
      "epoch": 7.9235423542354235,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02267292310748697,
      "loss": 0.233,
      "num_input_tokens_seen": 15200416,
      "step": 72025
    },
    {
      "epoch": 7.924092409240924,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.022671685693472807,
      "loss": 0.2329,
      "num_input_tokens_seen": 15201472,
      "step": 72030
    },
    {
      "epoch": 7.924642464246425,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.022670448208752393,
      "loss": 0.2308,
      "num_input_tokens_seen": 15202496,
      "step": 72035
    },
    {
      "epoch": 7.925192519251925,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.022669210653337143,
      "loss": 0.2309,
      "num_input_tokens_seen": 15203552,
      "step": 72040
    },
    {
      "epoch": 7.925742574257426,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02266797302723846,
      "loss": 0.2293,
      "num_input_tokens_seen": 15204576,
      "step": 72045
    },
    {
      "epoch": 7.926292629262926,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02266673533046775,
      "loss": 0.2319,
      "num_input_tokens_seen": 15205728,
      "step": 72050
    },
    {
      "epoch": 7.9268426842684265,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.022665497563036412,
      "loss": 0.2314,
      "num_input_tokens_seen": 15206752,
      "step": 72055
    },
    {
      "epoch": 7.927392739273928,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.02266425972495587,
      "loss": 0.2309,
      "num_input_tokens_seen": 15207776,
      "step": 72060
    },
    {
      "epoch": 7.927942794279428,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.022663021816237517,
      "loss": 0.232,
      "num_input_tokens_seen": 15208768,
      "step": 72065
    },
    {
      "epoch": 7.928492849284929,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.022661783836892775,
      "loss": 0.2325,
      "num_input_tokens_seen": 15209824,
      "step": 72070
    },
    {
      "epoch": 7.929042904290429,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02266054578693304,
      "loss": 0.234,
      "num_input_tokens_seen": 15210880,
      "step": 72075
    },
    {
      "epoch": 7.929592959295929,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.022659307666369737,
      "loss": 0.2315,
      "num_input_tokens_seen": 15211872,
      "step": 72080
    },
    {
      "epoch": 7.93014301430143,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.022658069475214267,
      "loss": 0.2346,
      "num_input_tokens_seen": 15212896,
      "step": 72085
    },
    {
      "epoch": 7.930693069306931,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.022656831213478047,
      "loss": 0.2273,
      "num_input_tokens_seen": 15213952,
      "step": 72090
    },
    {
      "epoch": 7.931243124312431,
      "grad_norm": 0.009765625,
      "learning_rate": 0.022655592881172484,
      "loss": 0.2324,
      "num_input_tokens_seen": 15215072,
      "step": 72095
    },
    {
      "epoch": 7.931793179317932,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.022654354478309,
      "loss": 0.2324,
      "num_input_tokens_seen": 15216096,
      "step": 72100
    },
    {
      "epoch": 7.932343234323432,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.022653116004899002,
      "loss": 0.2304,
      "num_input_tokens_seen": 15217120,
      "step": 72105
    },
    {
      "epoch": 7.932893289328933,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.022651877460953904,
      "loss": 0.2309,
      "num_input_tokens_seen": 15218176,
      "step": 72110
    },
    {
      "epoch": 7.933443344334433,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.02265063884648513,
      "loss": 0.2304,
      "num_input_tokens_seen": 15219168,
      "step": 72115
    },
    {
      "epoch": 7.933993399339934,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02264940016150408,
      "loss": 0.233,
      "num_input_tokens_seen": 15220256,
      "step": 72120
    },
    {
      "epoch": 7.934543454345435,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.022648161406022185,
      "loss": 0.233,
      "num_input_tokens_seen": 15221280,
      "step": 72125
    },
    {
      "epoch": 7.935093509350935,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.022646922580050858,
      "loss": 0.2319,
      "num_input_tokens_seen": 15222368,
      "step": 72130
    },
    {
      "epoch": 7.935643564356436,
      "grad_norm": 0.000812530517578125,
      "learning_rate": 0.02264568368360151,
      "loss": 0.2319,
      "num_input_tokens_seen": 15223424,
      "step": 72135
    },
    {
      "epoch": 7.936193619361936,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.022644444716685568,
      "loss": 0.2319,
      "num_input_tokens_seen": 15224480,
      "step": 72140
    },
    {
      "epoch": 7.936743674367436,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.022643205679314445,
      "loss": 0.2324,
      "num_input_tokens_seen": 15225504,
      "step": 72145
    },
    {
      "epoch": 7.9372937293729375,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.02264196657149956,
      "loss": 0.2319,
      "num_input_tokens_seen": 15226496,
      "step": 72150
    },
    {
      "epoch": 7.937843784378438,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02264072739325234,
      "loss": 0.2314,
      "num_input_tokens_seen": 15227520,
      "step": 72155
    },
    {
      "epoch": 7.938393839383938,
      "grad_norm": 0.00063323974609375,
      "learning_rate": 0.022639488144584198,
      "loss": 0.2314,
      "num_input_tokens_seen": 15228608,
      "step": 72160
    },
    {
      "epoch": 7.938943894389439,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.022638248825506562,
      "loss": 0.2298,
      "num_input_tokens_seen": 15229728,
      "step": 72165
    },
    {
      "epoch": 7.939493949394939,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.022637009436030853,
      "loss": 0.2324,
      "num_input_tokens_seen": 15230816,
      "step": 72170
    },
    {
      "epoch": 7.94004400440044,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.022635769976168497,
      "loss": 0.2335,
      "num_input_tokens_seen": 15231936,
      "step": 72175
    },
    {
      "epoch": 7.9405940594059405,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.022634530445930903,
      "loss": 0.2298,
      "num_input_tokens_seen": 15232992,
      "step": 72180
    },
    {
      "epoch": 7.941144114411442,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.02263329084532951,
      "loss": 0.2308,
      "num_input_tokens_seen": 15234016,
      "step": 72185
    },
    {
      "epoch": 7.941694169416942,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02263205117437574,
      "loss": 0.2319,
      "num_input_tokens_seen": 15235104,
      "step": 72190
    },
    {
      "epoch": 7.942244224422442,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.022630811433081013,
      "loss": 0.2319,
      "num_input_tokens_seen": 15236192,
      "step": 72195
    },
    {
      "epoch": 7.942794279427943,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02262957162145676,
      "loss": 0.2308,
      "num_input_tokens_seen": 15237248,
      "step": 72200
    },
    {
      "epoch": 7.943344334433443,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.02262833173951441,
      "loss": 0.2324,
      "num_input_tokens_seen": 15238304,
      "step": 72205
    },
    {
      "epoch": 7.9438943894389435,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.022627091787265383,
      "loss": 0.2303,
      "num_input_tokens_seen": 15239328,
      "step": 72210
    },
    {
      "epoch": 7.944444444444445,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.022625851764721115,
      "loss": 0.2313,
      "num_input_tokens_seen": 15240384,
      "step": 72215
    },
    {
      "epoch": 7.944994499449945,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.022624611671893023,
      "loss": 0.2314,
      "num_input_tokens_seen": 15241440,
      "step": 72220
    },
    {
      "epoch": 7.945544554455445,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02262337150879255,
      "loss": 0.2334,
      "num_input_tokens_seen": 15242464,
      "step": 72225
    },
    {
      "epoch": 7.946094609460946,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.022622131275431116,
      "loss": 0.2319,
      "num_input_tokens_seen": 15243488,
      "step": 72230
    },
    {
      "epoch": 7.946644664466446,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.022620890971820153,
      "loss": 0.2334,
      "num_input_tokens_seen": 15244512,
      "step": 72235
    },
    {
      "epoch": 7.947194719471947,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.0226196505979711,
      "loss": 0.2324,
      "num_input_tokens_seen": 15245568,
      "step": 72240
    },
    {
      "epoch": 7.947744774477448,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.022618410153895388,
      "loss": 0.2308,
      "num_input_tokens_seen": 15246656,
      "step": 72245
    },
    {
      "epoch": 7.948294829482949,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.02261716963960444,
      "loss": 0.2329,
      "num_input_tokens_seen": 15247744,
      "step": 72250
    },
    {
      "epoch": 7.948844884488449,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.02261592905510969,
      "loss": 0.2283,
      "num_input_tokens_seen": 15248864,
      "step": 72255
    },
    {
      "epoch": 7.949394939493949,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.022614688400422585,
      "loss": 0.2319,
      "num_input_tokens_seen": 15249856,
      "step": 72260
    },
    {
      "epoch": 7.94994499449945,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.022613447675554548,
      "loss": 0.2303,
      "num_input_tokens_seen": 15250880,
      "step": 72265
    },
    {
      "epoch": 7.9504950495049505,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02261220688051701,
      "loss": 0.2303,
      "num_input_tokens_seen": 15251904,
      "step": 72270
    },
    {
      "epoch": 7.951045104510451,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02261096601532142,
      "loss": 0.2319,
      "num_input_tokens_seen": 15252960,
      "step": 72275
    },
    {
      "epoch": 7.951595159515952,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02260972507997921,
      "loss": 0.2313,
      "num_input_tokens_seen": 15254048,
      "step": 72280
    },
    {
      "epoch": 7.952145214521452,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022608484074501816,
      "loss": 0.2319,
      "num_input_tokens_seen": 15255136,
      "step": 72285
    },
    {
      "epoch": 7.952695269526953,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.022607242998900674,
      "loss": 0.2288,
      "num_input_tokens_seen": 15256256,
      "step": 72290
    },
    {
      "epoch": 7.953245324532453,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02260600185318722,
      "loss": 0.2308,
      "num_input_tokens_seen": 15257344,
      "step": 72295
    },
    {
      "epoch": 7.9537953795379535,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.022604760637372896,
      "loss": 0.2313,
      "num_input_tokens_seen": 15258368,
      "step": 72300
    },
    {
      "epoch": 7.9543454345434546,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02260351935146915,
      "loss": 0.2324,
      "num_input_tokens_seen": 15259424,
      "step": 72305
    },
    {
      "epoch": 7.954895489548955,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.022602277995487408,
      "loss": 0.2324,
      "num_input_tokens_seen": 15260544,
      "step": 72310
    },
    {
      "epoch": 7.955445544554456,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02260103656943912,
      "loss": 0.2324,
      "num_input_tokens_seen": 15261632,
      "step": 72315
    },
    {
      "epoch": 7.955995599559956,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.022599795073335725,
      "loss": 0.2308,
      "num_input_tokens_seen": 15262656,
      "step": 72320
    },
    {
      "epoch": 7.956545654565456,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.022598553507188666,
      "loss": 0.2319,
      "num_input_tokens_seen": 15263680,
      "step": 72325
    },
    {
      "epoch": 7.957095709570957,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.022597311871009385,
      "loss": 0.2308,
      "num_input_tokens_seen": 15264736,
      "step": 72330
    },
    {
      "epoch": 7.957645764576458,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.022596070164809325,
      "loss": 0.2293,
      "num_input_tokens_seen": 15265856,
      "step": 72335
    },
    {
      "epoch": 7.958195819581958,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.02259482838859993,
      "loss": 0.2308,
      "num_input_tokens_seen": 15266912,
      "step": 72340
    },
    {
      "epoch": 7.958745874587459,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02259358654239265,
      "loss": 0.2308,
      "num_input_tokens_seen": 15267936,
      "step": 72345
    },
    {
      "epoch": 7.959295929592959,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.022592344626198923,
      "loss": 0.2319,
      "num_input_tokens_seen": 15268896,
      "step": 72350
    },
    {
      "epoch": 7.95984598459846,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.022591102640030204,
      "loss": 0.2324,
      "num_input_tokens_seen": 15269984,
      "step": 72355
    },
    {
      "epoch": 7.96039603960396,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02258986058389793,
      "loss": 0.2313,
      "num_input_tokens_seen": 15271040,
      "step": 72360
    },
    {
      "epoch": 7.960946094609461,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.022588618457813556,
      "loss": 0.2298,
      "num_input_tokens_seen": 15272128,
      "step": 72365
    },
    {
      "epoch": 7.961496149614962,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.02258737626178852,
      "loss": 0.2319,
      "num_input_tokens_seen": 15273184,
      "step": 72370
    },
    {
      "epoch": 7.962046204620462,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022586133995834284,
      "loss": 0.2324,
      "num_input_tokens_seen": 15274272,
      "step": 72375
    },
    {
      "epoch": 7.962596259625963,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02258489165996229,
      "loss": 0.2314,
      "num_input_tokens_seen": 15275360,
      "step": 72380
    },
    {
      "epoch": 7.963146314631463,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02258364925418399,
      "loss": 0.2303,
      "num_input_tokens_seen": 15276384,
      "step": 72385
    },
    {
      "epoch": 7.963696369636963,
      "grad_norm": 0.000423431396484375,
      "learning_rate": 0.02258240677851083,
      "loss": 0.2314,
      "num_input_tokens_seen": 15277440,
      "step": 72390
    },
    {
      "epoch": 7.9642464246424645,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.022581164232954275,
      "loss": 0.2324,
      "num_input_tokens_seen": 15278560,
      "step": 72395
    },
    {
      "epoch": 7.964796479647965,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02257992161752576,
      "loss": 0.233,
      "num_input_tokens_seen": 15279552,
      "step": 72400
    },
    {
      "epoch": 7.965346534653465,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.02257867893223674,
      "loss": 0.2314,
      "num_input_tokens_seen": 15280608,
      "step": 72405
    },
    {
      "epoch": 7.965896589658966,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.022577436177098675,
      "loss": 0.2308,
      "num_input_tokens_seen": 15281600,
      "step": 72410
    },
    {
      "epoch": 7.966446644664466,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.022576193352123023,
      "loss": 0.2324,
      "num_input_tokens_seen": 15282688,
      "step": 72415
    },
    {
      "epoch": 7.966996699669967,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02257495045732123,
      "loss": 0.2314,
      "num_input_tokens_seen": 15283776,
      "step": 72420
    },
    {
      "epoch": 7.9675467546754675,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.022573707492704754,
      "loss": 0.2293,
      "num_input_tokens_seen": 15284864,
      "step": 72425
    },
    {
      "epoch": 7.968096809680969,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.022572464458285042,
      "loss": 0.2334,
      "num_input_tokens_seen": 15285920,
      "step": 72430
    },
    {
      "epoch": 7.968646864686469,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.022571221354073568,
      "loss": 0.2309,
      "num_input_tokens_seen": 15287008,
      "step": 72435
    },
    {
      "epoch": 7.969196919691969,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.022569978180081777,
      "loss": 0.2319,
      "num_input_tokens_seen": 15288064,
      "step": 72440
    },
    {
      "epoch": 7.96974697469747,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02256873493632113,
      "loss": 0.2319,
      "num_input_tokens_seen": 15289088,
      "step": 72445
    },
    {
      "epoch": 7.97029702970297,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.022567491622803085,
      "loss": 0.2313,
      "num_input_tokens_seen": 15290176,
      "step": 72450
    },
    {
      "epoch": 7.9708470847084705,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0225662482395391,
      "loss": 0.2324,
      "num_input_tokens_seen": 15291200,
      "step": 72455
    },
    {
      "epoch": 7.971397139713972,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.022565004786540634,
      "loss": 0.2313,
      "num_input_tokens_seen": 15292256,
      "step": 72460
    },
    {
      "epoch": 7.971947194719472,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022563761263819156,
      "loss": 0.2319,
      "num_input_tokens_seen": 15293376,
      "step": 72465
    },
    {
      "epoch": 7.972497249724973,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02256251767138611,
      "loss": 0.2319,
      "num_input_tokens_seen": 15294400,
      "step": 72470
    },
    {
      "epoch": 7.973047304730473,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.022561274009252973,
      "loss": 0.2314,
      "num_input_tokens_seen": 15295392,
      "step": 72475
    },
    {
      "epoch": 7.973597359735973,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.0225600302774312,
      "loss": 0.2309,
      "num_input_tokens_seen": 15296480,
      "step": 72480
    },
    {
      "epoch": 7.974147414741474,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02255878647593226,
      "loss": 0.2319,
      "num_input_tokens_seen": 15297536,
      "step": 72485
    },
    {
      "epoch": 7.974697469746975,
      "grad_norm": 0.009765625,
      "learning_rate": 0.022557542604767607,
      "loss": 0.2324,
      "num_input_tokens_seen": 15298592,
      "step": 72490
    },
    {
      "epoch": 7.975247524752476,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.02255629866394871,
      "loss": 0.2319,
      "num_input_tokens_seen": 15299680,
      "step": 72495
    },
    {
      "epoch": 7.975797579757976,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.022555054653487037,
      "loss": 0.2298,
      "num_input_tokens_seen": 15300672,
      "step": 72500
    },
    {
      "epoch": 7.976347634763476,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02255381057339405,
      "loss": 0.2304,
      "num_input_tokens_seen": 15301760,
      "step": 72505
    },
    {
      "epoch": 7.976897689768977,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.022552566423681218,
      "loss": 0.2314,
      "num_input_tokens_seen": 15302720,
      "step": 72510
    },
    {
      "epoch": 7.977447744774477,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02255132220436,
      "loss": 0.2319,
      "num_input_tokens_seen": 15303808,
      "step": 72515
    },
    {
      "epoch": 7.977997799779978,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.022550077915441873,
      "loss": 0.2309,
      "num_input_tokens_seen": 15304800,
      "step": 72520
    },
    {
      "epoch": 7.978547854785479,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.022548833556938295,
      "loss": 0.2304,
      "num_input_tokens_seen": 15305856,
      "step": 72525
    },
    {
      "epoch": 7.979097909790979,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022547589128860745,
      "loss": 0.2303,
      "num_input_tokens_seen": 15306880,
      "step": 72530
    },
    {
      "epoch": 7.97964796479648,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02254634463122069,
      "loss": 0.2329,
      "num_input_tokens_seen": 15307904,
      "step": 72535
    },
    {
      "epoch": 7.98019801980198,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.022545100064029593,
      "loss": 0.2325,
      "num_input_tokens_seen": 15308960,
      "step": 72540
    },
    {
      "epoch": 7.98074807480748,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.02254385542729893,
      "loss": 0.2308,
      "num_input_tokens_seen": 15310016,
      "step": 72545
    },
    {
      "epoch": 7.9812981298129815,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.022542610721040173,
      "loss": 0.2309,
      "num_input_tokens_seen": 15311040,
      "step": 72550
    },
    {
      "epoch": 7.981848184818482,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.02254136594526479,
      "loss": 0.2308,
      "num_input_tokens_seen": 15312064,
      "step": 72555
    },
    {
      "epoch": 7.982398239823983,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02254012109998426,
      "loss": 0.2319,
      "num_input_tokens_seen": 15313152,
      "step": 72560
    },
    {
      "epoch": 7.982948294829483,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02253887618521005,
      "loss": 0.2329,
      "num_input_tokens_seen": 15314240,
      "step": 72565
    },
    {
      "epoch": 7.983498349834983,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.022537631200953635,
      "loss": 0.2314,
      "num_input_tokens_seen": 15315296,
      "step": 72570
    },
    {
      "epoch": 7.984048404840484,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.02253638614722649,
      "loss": 0.2309,
      "num_input_tokens_seen": 15316320,
      "step": 72575
    },
    {
      "epoch": 7.9845984598459845,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.022535141024040087,
      "loss": 0.2324,
      "num_input_tokens_seen": 15317440,
      "step": 72580
    },
    {
      "epoch": 7.985148514851485,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.022533895831405908,
      "loss": 0.2303,
      "num_input_tokens_seen": 15318496,
      "step": 72585
    },
    {
      "epoch": 7.985698569856986,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02253265056933543,
      "loss": 0.2303,
      "num_input_tokens_seen": 15319456,
      "step": 72590
    },
    {
      "epoch": 7.986248624862486,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.022531405237840123,
      "loss": 0.2314,
      "num_input_tokens_seen": 15320512,
      "step": 72595
    },
    {
      "epoch": 7.986798679867987,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.022530159836931466,
      "loss": 0.2314,
      "num_input_tokens_seen": 15321536,
      "step": 72600
    },
    {
      "epoch": 7.987348734873487,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.022528914366620943,
      "loss": 0.2314,
      "num_input_tokens_seen": 15322560,
      "step": 72605
    },
    {
      "epoch": 7.987898789878988,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.022527668826920023,
      "loss": 0.2314,
      "num_input_tokens_seen": 15323616,
      "step": 72610
    },
    {
      "epoch": 7.988448844884489,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.022526423217840198,
      "loss": 0.2309,
      "num_input_tokens_seen": 15324640,
      "step": 72615
    },
    {
      "epoch": 7.988998899889989,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.022525177539392937,
      "loss": 0.2309,
      "num_input_tokens_seen": 15325728,
      "step": 72620
    },
    {
      "epoch": 7.98954895489549,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.02252393179158973,
      "loss": 0.2314,
      "num_input_tokens_seen": 15326784,
      "step": 72625
    },
    {
      "epoch": 7.99009900990099,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.022522685974442045,
      "loss": 0.2298,
      "num_input_tokens_seen": 15327808,
      "step": 72630
    },
    {
      "epoch": 7.99064906490649,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.022521440087961385,
      "loss": 0.2309,
      "num_input_tokens_seen": 15328928,
      "step": 72635
    },
    {
      "epoch": 7.991199119911991,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.02252019413215921,
      "loss": 0.2319,
      "num_input_tokens_seen": 15329952,
      "step": 72640
    },
    {
      "epoch": 7.991749174917492,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.02251894810704702,
      "loss": 0.2304,
      "num_input_tokens_seen": 15331008,
      "step": 72645
    },
    {
      "epoch": 7.992299229922993,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.022517702012636295,
      "loss": 0.2314,
      "num_input_tokens_seen": 15332000,
      "step": 72650
    },
    {
      "epoch": 7.992849284928493,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.022516455848938518,
      "loss": 0.2314,
      "num_input_tokens_seen": 15333024,
      "step": 72655
    },
    {
      "epoch": 7.993399339933993,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02251520961596517,
      "loss": 0.2324,
      "num_input_tokens_seen": 15334016,
      "step": 72660
    },
    {
      "epoch": 7.993949394939494,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022513963313727743,
      "loss": 0.2324,
      "num_input_tokens_seen": 15335072,
      "step": 72665
    },
    {
      "epoch": 7.994499449944994,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02251271694223772,
      "loss": 0.2324,
      "num_input_tokens_seen": 15336160,
      "step": 72670
    },
    {
      "epoch": 7.9950495049504955,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.022511470501506592,
      "loss": 0.233,
      "num_input_tokens_seen": 15337184,
      "step": 72675
    },
    {
      "epoch": 7.995599559955996,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.022510223991545845,
      "loss": 0.2324,
      "num_input_tokens_seen": 15338272,
      "step": 72680
    },
    {
      "epoch": 7.996149614961496,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.022508977412366966,
      "loss": 0.2304,
      "num_input_tokens_seen": 15339360,
      "step": 72685
    },
    {
      "epoch": 7.996699669966997,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.022507730763981448,
      "loss": 0.233,
      "num_input_tokens_seen": 15340384,
      "step": 72690
    },
    {
      "epoch": 7.997249724972497,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.022506484046400776,
      "loss": 0.2314,
      "num_input_tokens_seen": 15341440,
      "step": 72695
    },
    {
      "epoch": 7.997799779977997,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.02250523725963644,
      "loss": 0.2314,
      "num_input_tokens_seen": 15342496,
      "step": 72700
    },
    {
      "epoch": 7.9983498349834985,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.022503990403699936,
      "loss": 0.2309,
      "num_input_tokens_seen": 15343552,
      "step": 72705
    },
    {
      "epoch": 7.998899889988999,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.022502743478602757,
      "loss": 0.2303,
      "num_input_tokens_seen": 15344608,
      "step": 72710
    },
    {
      "epoch": 7.9994499449945,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.022501496484356388,
      "loss": 0.2309,
      "num_input_tokens_seen": 15345696,
      "step": 72715
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.022500249420972323,
      "loss": 0.2314,
      "num_input_tokens_seen": 15346672,
      "step": 72720
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.2313498854637146,
      "eval_runtime": 60.5656,
      "eval_samples_per_second": 66.704,
      "eval_steps_per_second": 16.676,
      "num_input_tokens_seen": 15346672,
      "step": 72720
    },
    {
      "epoch": 8.000550055005501,
      "grad_norm": 0.0030364990234375,
      "learning_rate": 0.022499002288462064,
      "loss": 0.2324,
      "num_input_tokens_seen": 15347728,
      "step": 72725
    },
    {
      "epoch": 8.001100110011,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02249775508683709,
      "loss": 0.2314,
      "num_input_tokens_seen": 15348784,
      "step": 72730
    },
    {
      "epoch": 8.001650165016502,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.022496507816108915,
      "loss": 0.2303,
      "num_input_tokens_seen": 15349840,
      "step": 72735
    },
    {
      "epoch": 8.002200220022003,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02249526047628902,
      "loss": 0.2303,
      "num_input_tokens_seen": 15350928,
      "step": 72740
    },
    {
      "epoch": 8.002750275027502,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.022494013067388904,
      "loss": 0.2303,
      "num_input_tokens_seen": 15352016,
      "step": 72745
    },
    {
      "epoch": 8.003300330033003,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.022492765589420066,
      "loss": 0.2293,
      "num_input_tokens_seen": 15353040,
      "step": 72750
    },
    {
      "epoch": 8.003850385038504,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02249151804239401,
      "loss": 0.2324,
      "num_input_tokens_seen": 15354032,
      "step": 72755
    },
    {
      "epoch": 8.004400440044005,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02249027042632222,
      "loss": 0.2303,
      "num_input_tokens_seen": 15355088,
      "step": 72760
    },
    {
      "epoch": 8.004950495049505,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.022489022741216204,
      "loss": 0.2314,
      "num_input_tokens_seen": 15356144,
      "step": 72765
    },
    {
      "epoch": 8.005500550055006,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02248777498708746,
      "loss": 0.2324,
      "num_input_tokens_seen": 15357168,
      "step": 72770
    },
    {
      "epoch": 8.006050605060507,
      "grad_norm": 0.009765625,
      "learning_rate": 0.022486527163947483,
      "loss": 0.2319,
      "num_input_tokens_seen": 15358192,
      "step": 72775
    },
    {
      "epoch": 8.006600660066006,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02248527927180778,
      "loss": 0.2308,
      "num_input_tokens_seen": 15359248,
      "step": 72780
    },
    {
      "epoch": 8.007150715071507,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02248403131067985,
      "loss": 0.2324,
      "num_input_tokens_seen": 15360304,
      "step": 72785
    },
    {
      "epoch": 8.007700770077008,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.02248278328057519,
      "loss": 0.2329,
      "num_input_tokens_seen": 15361392,
      "step": 72790
    },
    {
      "epoch": 8.008250825082508,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.022481535181505315,
      "loss": 0.2329,
      "num_input_tokens_seen": 15362448,
      "step": 72795
    },
    {
      "epoch": 8.008800880088009,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.022480287013481717,
      "loss": 0.2319,
      "num_input_tokens_seen": 15363504,
      "step": 72800
    },
    {
      "epoch": 8.00935093509351,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.0224790387765159,
      "loss": 0.2319,
      "num_input_tokens_seen": 15364528,
      "step": 72805
    },
    {
      "epoch": 8.009900990099009,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.022477790470619374,
      "loss": 0.2303,
      "num_input_tokens_seen": 15365584,
      "step": 72810
    },
    {
      "epoch": 8.01045104510451,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02247654209580364,
      "loss": 0.2303,
      "num_input_tokens_seen": 15366576,
      "step": 72815
    },
    {
      "epoch": 8.011001100110011,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.022475293652080206,
      "loss": 0.2308,
      "num_input_tokens_seen": 15367536,
      "step": 72820
    },
    {
      "epoch": 8.011551155115512,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.022474045139460577,
      "loss": 0.2304,
      "num_input_tokens_seen": 15368624,
      "step": 72825
    },
    {
      "epoch": 8.012101210121012,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.022472796557956264,
      "loss": 0.2319,
      "num_input_tokens_seen": 15369584,
      "step": 72830
    },
    {
      "epoch": 8.012651265126513,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.022471547907578767,
      "loss": 0.2293,
      "num_input_tokens_seen": 15370672,
      "step": 72835
    },
    {
      "epoch": 8.013201320132014,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.0224702991883396,
      "loss": 0.2309,
      "num_input_tokens_seen": 15371728,
      "step": 72840
    },
    {
      "epoch": 8.013751375137513,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.022469050400250268,
      "loss": 0.2324,
      "num_input_tokens_seen": 15372752,
      "step": 72845
    },
    {
      "epoch": 8.014301430143014,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.022467801543322282,
      "loss": 0.2356,
      "num_input_tokens_seen": 15373776,
      "step": 72850
    },
    {
      "epoch": 8.014851485148515,
      "grad_norm": 0.00921630859375,
      "learning_rate": 0.022466552617567158,
      "loss": 0.2273,
      "num_input_tokens_seen": 15374800,
      "step": 72855
    },
    {
      "epoch": 8.015401540154015,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.022465303622996393,
      "loss": 0.2319,
      "num_input_tokens_seen": 15375888,
      "step": 72860
    },
    {
      "epoch": 8.015951595159516,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.022464054559621513,
      "loss": 0.2325,
      "num_input_tokens_seen": 15376944,
      "step": 72865
    },
    {
      "epoch": 8.016501650165017,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.022462805427454022,
      "loss": 0.2319,
      "num_input_tokens_seen": 15377936,
      "step": 72870
    },
    {
      "epoch": 8.017051705170518,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.022461556226505437,
      "loss": 0.234,
      "num_input_tokens_seen": 15378992,
      "step": 72875
    },
    {
      "epoch": 8.017601760176017,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.022460306956787267,
      "loss": 0.2319,
      "num_input_tokens_seen": 15380048,
      "step": 72880
    },
    {
      "epoch": 8.018151815181518,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02245905761831102,
      "loss": 0.2314,
      "num_input_tokens_seen": 15381072,
      "step": 72885
    },
    {
      "epoch": 8.01870187018702,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.022457808211088226,
      "loss": 0.2329,
      "num_input_tokens_seen": 15382160,
      "step": 72890
    },
    {
      "epoch": 8.019251925192519,
      "grad_norm": 0.004547119140625,
      "learning_rate": 0.02245655873513039,
      "loss": 0.2293,
      "num_input_tokens_seen": 15383184,
      "step": 72895
    },
    {
      "epoch": 8.01980198019802,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.022455309190449035,
      "loss": 0.234,
      "num_input_tokens_seen": 15384272,
      "step": 72900
    },
    {
      "epoch": 8.020352035203521,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.022454059577055666,
      "loss": 0.2314,
      "num_input_tokens_seen": 15385328,
      "step": 72905
    },
    {
      "epoch": 8.02090209020902,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02245280989496181,
      "loss": 0.2309,
      "num_input_tokens_seen": 15386352,
      "step": 72910
    },
    {
      "epoch": 8.021452145214521,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02245156014417898,
      "loss": 0.2303,
      "num_input_tokens_seen": 15387440,
      "step": 72915
    },
    {
      "epoch": 8.022002200220022,
      "grad_norm": 0.00074005126953125,
      "learning_rate": 0.022450310324718697,
      "loss": 0.2319,
      "num_input_tokens_seen": 15388496,
      "step": 72920
    },
    {
      "epoch": 8.022552255225522,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02244906043659248,
      "loss": 0.2314,
      "num_input_tokens_seen": 15389584,
      "step": 72925
    },
    {
      "epoch": 8.023102310231023,
      "grad_norm": 0.009765625,
      "learning_rate": 0.022447810479811845,
      "loss": 0.2314,
      "num_input_tokens_seen": 15390640,
      "step": 72930
    },
    {
      "epoch": 8.023652365236524,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.022446560454388313,
      "loss": 0.2334,
      "num_input_tokens_seen": 15391696,
      "step": 72935
    },
    {
      "epoch": 8.024202420242025,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.022445310360333412,
      "loss": 0.2319,
      "num_input_tokens_seen": 15392816,
      "step": 72940
    },
    {
      "epoch": 8.024752475247524,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.02244406019765865,
      "loss": 0.2309,
      "num_input_tokens_seen": 15393840,
      "step": 72945
    },
    {
      "epoch": 8.025302530253025,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.022442809966375564,
      "loss": 0.2293,
      "num_input_tokens_seen": 15394928,
      "step": 72950
    },
    {
      "epoch": 8.025852585258527,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02244155966649567,
      "loss": 0.233,
      "num_input_tokens_seen": 15396016,
      "step": 72955
    },
    {
      "epoch": 8.026402640264026,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02244030929803049,
      "loss": 0.2324,
      "num_input_tokens_seen": 15397040,
      "step": 72960
    },
    {
      "epoch": 8.026952695269527,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.022439058860991547,
      "loss": 0.2298,
      "num_input_tokens_seen": 15398096,
      "step": 72965
    },
    {
      "epoch": 8.027502750275028,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.02243780835539037,
      "loss": 0.2314,
      "num_input_tokens_seen": 15399184,
      "step": 72970
    },
    {
      "epoch": 8.028052805280527,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02243655778123848,
      "loss": 0.2324,
      "num_input_tokens_seen": 15400272,
      "step": 72975
    },
    {
      "epoch": 8.028602860286028,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.022435307138547408,
      "loss": 0.2319,
      "num_input_tokens_seen": 15401328,
      "step": 72980
    },
    {
      "epoch": 8.02915291529153,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.02243405642732868,
      "loss": 0.2309,
      "num_input_tokens_seen": 15402384,
      "step": 72985
    },
    {
      "epoch": 8.029702970297029,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02243280564759382,
      "loss": 0.2309,
      "num_input_tokens_seen": 15403408,
      "step": 72990
    },
    {
      "epoch": 8.03025302530253,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.022431554799354353,
      "loss": 0.2319,
      "num_input_tokens_seen": 15404528,
      "step": 72995
    },
    {
      "epoch": 8.030803080308031,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.022430303882621815,
      "loss": 0.2335,
      "num_input_tokens_seen": 15405584,
      "step": 73000
    },
    {
      "epoch": 8.031353135313532,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.02242905289740773,
      "loss": 0.2335,
      "num_input_tokens_seen": 15406672,
      "step": 73005
    },
    {
      "epoch": 8.031903190319031,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.022427801843723626,
      "loss": 0.2314,
      "num_input_tokens_seen": 15407728,
      "step": 73010
    },
    {
      "epoch": 8.032453245324533,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.02242655072158104,
      "loss": 0.234,
      "num_input_tokens_seen": 15408784,
      "step": 73015
    },
    {
      "epoch": 8.033003300330034,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.022425299530991503,
      "loss": 0.2309,
      "num_input_tokens_seen": 15409840,
      "step": 73020
    },
    {
      "epoch": 8.033553355335533,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.022424048271966538,
      "loss": 0.2314,
      "num_input_tokens_seen": 15410896,
      "step": 73025
    },
    {
      "epoch": 8.034103410341034,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.022422796944517685,
      "loss": 0.2309,
      "num_input_tokens_seen": 15411920,
      "step": 73030
    },
    {
      "epoch": 8.034653465346535,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.022421545548656472,
      "loss": 0.2319,
      "num_input_tokens_seen": 15413008,
      "step": 73035
    },
    {
      "epoch": 8.035203520352034,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.022420294084394436,
      "loss": 0.2309,
      "num_input_tokens_seen": 15414032,
      "step": 73040
    },
    {
      "epoch": 8.035753575357536,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.022419042551743117,
      "loss": 0.2319,
      "num_input_tokens_seen": 15415024,
      "step": 73045
    },
    {
      "epoch": 8.036303630363037,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022417790950714032,
      "loss": 0.233,
      "num_input_tokens_seen": 15416112,
      "step": 73050
    },
    {
      "epoch": 8.036853685368538,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02241653928131873,
      "loss": 0.2324,
      "num_input_tokens_seen": 15417168,
      "step": 73055
    },
    {
      "epoch": 8.037403740374037,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02241528754356875,
      "loss": 0.2308,
      "num_input_tokens_seen": 15418288,
      "step": 73060
    },
    {
      "epoch": 8.037953795379538,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.02241403573747562,
      "loss": 0.2277,
      "num_input_tokens_seen": 15419280,
      "step": 73065
    },
    {
      "epoch": 8.03850385038504,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.022412783863050876,
      "loss": 0.2324,
      "num_input_tokens_seen": 15420336,
      "step": 73070
    },
    {
      "epoch": 8.039053905390539,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.022411531920306066,
      "loss": 0.2324,
      "num_input_tokens_seen": 15421424,
      "step": 73075
    },
    {
      "epoch": 8.03960396039604,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02241027990925272,
      "loss": 0.2319,
      "num_input_tokens_seen": 15422480,
      "step": 73080
    },
    {
      "epoch": 8.04015401540154,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.022409027829902377,
      "loss": 0.2325,
      "num_input_tokens_seen": 15423536,
      "step": 73085
    },
    {
      "epoch": 8.04070407040704,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.02240777568226659,
      "loss": 0.2303,
      "num_input_tokens_seen": 15424624,
      "step": 73090
    },
    {
      "epoch": 8.041254125412541,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.022406523466356878,
      "loss": 0.2293,
      "num_input_tokens_seen": 15425616,
      "step": 73095
    },
    {
      "epoch": 8.041804180418042,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.022405271182184796,
      "loss": 0.2314,
      "num_input_tokens_seen": 15426704,
      "step": 73100
    },
    {
      "epoch": 8.042354235423542,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.022404018829761885,
      "loss": 0.2304,
      "num_input_tokens_seen": 15427792,
      "step": 73105
    },
    {
      "epoch": 8.042904290429043,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.022402766409099682,
      "loss": 0.2303,
      "num_input_tokens_seen": 15428880,
      "step": 73110
    },
    {
      "epoch": 8.043454345434544,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02240151392020974,
      "loss": 0.234,
      "num_input_tokens_seen": 15429872,
      "step": 73115
    },
    {
      "epoch": 8.044004400440045,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.02240026136310359,
      "loss": 0.2319,
      "num_input_tokens_seen": 15430960,
      "step": 73120
    },
    {
      "epoch": 8.044554455445544,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02239900873779278,
      "loss": 0.2314,
      "num_input_tokens_seen": 15432016,
      "step": 73125
    },
    {
      "epoch": 8.045104510451045,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.02239775604428886,
      "loss": 0.2314,
      "num_input_tokens_seen": 15433040,
      "step": 73130
    },
    {
      "epoch": 8.045654565456546,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.022396503282603373,
      "loss": 0.2278,
      "num_input_tokens_seen": 15434096,
      "step": 73135
    },
    {
      "epoch": 8.046204620462046,
      "grad_norm": 0.0004825592041015625,
      "learning_rate": 0.022395250452747865,
      "loss": 0.2319,
      "num_input_tokens_seen": 15435088,
      "step": 73140
    },
    {
      "epoch": 8.046754675467547,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.022393997554733883,
      "loss": 0.2314,
      "num_input_tokens_seen": 15436144,
      "step": 73145
    },
    {
      "epoch": 8.047304730473048,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.02239274458857297,
      "loss": 0.2324,
      "num_input_tokens_seen": 15437200,
      "step": 73150
    },
    {
      "epoch": 8.047854785478547,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.022391491554276673,
      "loss": 0.2308,
      "num_input_tokens_seen": 15438192,
      "step": 73155
    },
    {
      "epoch": 8.048404840484048,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.022390238451856555,
      "loss": 0.2324,
      "num_input_tokens_seen": 15439216,
      "step": 73160
    },
    {
      "epoch": 8.04895489548955,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.02238898528132415,
      "loss": 0.2324,
      "num_input_tokens_seen": 15440208,
      "step": 73165
    },
    {
      "epoch": 8.049504950495049,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.022387732042691018,
      "loss": 0.2314,
      "num_input_tokens_seen": 15441296,
      "step": 73170
    },
    {
      "epoch": 8.05005500550055,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.0223864787359687,
      "loss": 0.2278,
      "num_input_tokens_seen": 15442320,
      "step": 73175
    },
    {
      "epoch": 8.05060506050605,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.022385225361168753,
      "loss": 0.2293,
      "num_input_tokens_seen": 15443344,
      "step": 73180
    },
    {
      "epoch": 8.051155115511552,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.022383971918302728,
      "loss": 0.233,
      "num_input_tokens_seen": 15444400,
      "step": 73185
    },
    {
      "epoch": 8.051705170517051,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.022382718407382176,
      "loss": 0.2293,
      "num_input_tokens_seen": 15445488,
      "step": 73190
    },
    {
      "epoch": 8.052255225522552,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.02238146482841865,
      "loss": 0.2324,
      "num_input_tokens_seen": 15446576,
      "step": 73195
    },
    {
      "epoch": 8.052805280528053,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022380211181423706,
      "loss": 0.2288,
      "num_input_tokens_seen": 15447632,
      "step": 73200
    },
    {
      "epoch": 8.053355335533553,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.0223789574664089,
      "loss": 0.233,
      "num_input_tokens_seen": 15448688,
      "step": 73205
    },
    {
      "epoch": 8.053905390539054,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.02237770368338578,
      "loss": 0.2293,
      "num_input_tokens_seen": 15449808,
      "step": 73210
    },
    {
      "epoch": 8.054455445544555,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02237644983236591,
      "loss": 0.2319,
      "num_input_tokens_seen": 15450864,
      "step": 73215
    },
    {
      "epoch": 8.055005500550054,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.022375195913360835,
      "loss": 0.2345,
      "num_input_tokens_seen": 15451888,
      "step": 73220
    },
    {
      "epoch": 8.055555555555555,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.022373941926382118,
      "loss": 0.234,
      "num_input_tokens_seen": 15452944,
      "step": 73225
    },
    {
      "epoch": 8.056105610561056,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02237268787144132,
      "loss": 0.2314,
      "num_input_tokens_seen": 15453968,
      "step": 73230
    },
    {
      "epoch": 8.056655665566556,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02237143374855,
      "loss": 0.2329,
      "num_input_tokens_seen": 15455024,
      "step": 73235
    },
    {
      "epoch": 8.057205720572057,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.022370179557719708,
      "loss": 0.2319,
      "num_input_tokens_seen": 15456112,
      "step": 73240
    },
    {
      "epoch": 8.057755775577558,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.022368925298962006,
      "loss": 0.2298,
      "num_input_tokens_seen": 15457168,
      "step": 73245
    },
    {
      "epoch": 8.058305830583059,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.022367670972288457,
      "loss": 0.2319,
      "num_input_tokens_seen": 15458224,
      "step": 73250
    },
    {
      "epoch": 8.058855885588558,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.022366416577710622,
      "loss": 0.2309,
      "num_input_tokens_seen": 15459280,
      "step": 73255
    },
    {
      "epoch": 8.05940594059406,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.022365162115240055,
      "loss": 0.2293,
      "num_input_tokens_seen": 15460432,
      "step": 73260
    },
    {
      "epoch": 8.05995599559956,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.022363907584888328,
      "loss": 0.2298,
      "num_input_tokens_seen": 15461456,
      "step": 73265
    },
    {
      "epoch": 8.06050605060506,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02236265298666699,
      "loss": 0.2324,
      "num_input_tokens_seen": 15462544,
      "step": 73270
    },
    {
      "epoch": 8.061056105610561,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02236139832058762,
      "loss": 0.2293,
      "num_input_tokens_seen": 15463664,
      "step": 73275
    },
    {
      "epoch": 8.061606160616062,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022360143586661774,
      "loss": 0.2314,
      "num_input_tokens_seen": 15464656,
      "step": 73280
    },
    {
      "epoch": 8.062156215621561,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.02235888878490102,
      "loss": 0.2319,
      "num_input_tokens_seen": 15465712,
      "step": 73285
    },
    {
      "epoch": 8.062706270627062,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.02235763391531691,
      "loss": 0.2303,
      "num_input_tokens_seen": 15466704,
      "step": 73290
    },
    {
      "epoch": 8.063256325632564,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.022356378977921025,
      "loss": 0.2314,
      "num_input_tokens_seen": 15467728,
      "step": 73295
    },
    {
      "epoch": 8.063806380638065,
      "grad_norm": 0.000354766845703125,
      "learning_rate": 0.022355123972724922,
      "loss": 0.2319,
      "num_input_tokens_seen": 15468752,
      "step": 73300
    },
    {
      "epoch": 8.064356435643564,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.022353868899740174,
      "loss": 0.2309,
      "num_input_tokens_seen": 15469808,
      "step": 73305
    },
    {
      "epoch": 8.064906490649065,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02235261375897834,
      "loss": 0.2303,
      "num_input_tokens_seen": 15470864,
      "step": 73310
    },
    {
      "epoch": 8.065456545654566,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.022351358550450996,
      "loss": 0.2329,
      "num_input_tokens_seen": 15471952,
      "step": 73315
    },
    {
      "epoch": 8.066006600660065,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.022350103274169704,
      "loss": 0.2319,
      "num_input_tokens_seen": 15473008,
      "step": 73320
    },
    {
      "epoch": 8.066556655665567,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02234884793014604,
      "loss": 0.2319,
      "num_input_tokens_seen": 15474000,
      "step": 73325
    },
    {
      "epoch": 8.067106710671068,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.022347592518391572,
      "loss": 0.2324,
      "num_input_tokens_seen": 15475088,
      "step": 73330
    },
    {
      "epoch": 8.067656765676567,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.02234633703891787,
      "loss": 0.2293,
      "num_input_tokens_seen": 15476144,
      "step": 73335
    },
    {
      "epoch": 8.068206820682068,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.0223450814917365,
      "loss": 0.2324,
      "num_input_tokens_seen": 15477168,
      "step": 73340
    },
    {
      "epoch": 8.06875687568757,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02234382587685905,
      "loss": 0.2293,
      "num_input_tokens_seen": 15478192,
      "step": 73345
    },
    {
      "epoch": 8.069306930693068,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02234257019429707,
      "loss": 0.2319,
      "num_input_tokens_seen": 15479280,
      "step": 73350
    },
    {
      "epoch": 8.06985698569857,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.022341314444062146,
      "loss": 0.2319,
      "num_input_tokens_seen": 15480272,
      "step": 73355
    },
    {
      "epoch": 8.07040704070407,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02234005862616585,
      "loss": 0.2309,
      "num_input_tokens_seen": 15481296,
      "step": 73360
    },
    {
      "epoch": 8.070957095709572,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.022338802740619758,
      "loss": 0.2309,
      "num_input_tokens_seen": 15482320,
      "step": 73365
    },
    {
      "epoch": 8.071507150715071,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.022337546787435436,
      "loss": 0.233,
      "num_input_tokens_seen": 15483376,
      "step": 73370
    },
    {
      "epoch": 8.072057205720572,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.022336290766624474,
      "loss": 0.2324,
      "num_input_tokens_seen": 15484496,
      "step": 73375
    },
    {
      "epoch": 8.072607260726073,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.02233503467819844,
      "loss": 0.2298,
      "num_input_tokens_seen": 15485552,
      "step": 73380
    },
    {
      "epoch": 8.073157315731573,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.022333778522168904,
      "loss": 0.234,
      "num_input_tokens_seen": 15486608,
      "step": 73385
    },
    {
      "epoch": 8.073707370737074,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.022332522298547455,
      "loss": 0.2346,
      "num_input_tokens_seen": 15487696,
      "step": 73390
    },
    {
      "epoch": 8.074257425742575,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.022331266007345665,
      "loss": 0.2319,
      "num_input_tokens_seen": 15488784,
      "step": 73395
    },
    {
      "epoch": 8.074807480748074,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.022330009648575115,
      "loss": 0.2303,
      "num_input_tokens_seen": 15489840,
      "step": 73400
    },
    {
      "epoch": 8.075357535753575,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.022328753222247387,
      "loss": 0.2304,
      "num_input_tokens_seen": 15490928,
      "step": 73405
    },
    {
      "epoch": 8.075907590759076,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.02232749672837405,
      "loss": 0.2314,
      "num_input_tokens_seen": 15492016,
      "step": 73410
    },
    {
      "epoch": 8.076457645764576,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.02232624016696669,
      "loss": 0.2309,
      "num_input_tokens_seen": 15493136,
      "step": 73415
    },
    {
      "epoch": 8.077007700770077,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0223249835380369,
      "loss": 0.2288,
      "num_input_tokens_seen": 15494192,
      "step": 73420
    },
    {
      "epoch": 8.077557755775578,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.02232372684159625,
      "loss": 0.2309,
      "num_input_tokens_seen": 15495248,
      "step": 73425
    },
    {
      "epoch": 8.078107810781079,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.02232247007765632,
      "loss": 0.2304,
      "num_input_tokens_seen": 15496272,
      "step": 73430
    },
    {
      "epoch": 8.078657865786578,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.022321213246228693,
      "loss": 0.2273,
      "num_input_tokens_seen": 15497328,
      "step": 73435
    },
    {
      "epoch": 8.07920792079208,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.022319956347324966,
      "loss": 0.2278,
      "num_input_tokens_seen": 15498416,
      "step": 73440
    },
    {
      "epoch": 8.07975797579758,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.022318699380956708,
      "loss": 0.231,
      "num_input_tokens_seen": 15499536,
      "step": 73445
    },
    {
      "epoch": 8.08030803080308,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.02231744234713551,
      "loss": 0.2357,
      "num_input_tokens_seen": 15500592,
      "step": 73450
    },
    {
      "epoch": 8.08085808580858,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.022316185245872956,
      "loss": 0.2341,
      "num_input_tokens_seen": 15501616,
      "step": 73455
    },
    {
      "epoch": 8.081408140814082,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.022314928077180635,
      "loss": 0.2314,
      "num_input_tokens_seen": 15502608,
      "step": 73460
    },
    {
      "epoch": 8.081958195819581,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.022313670841070134,
      "loss": 0.2304,
      "num_input_tokens_seen": 15503728,
      "step": 73465
    },
    {
      "epoch": 8.082508250825082,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.022312413537553033,
      "loss": 0.2325,
      "num_input_tokens_seen": 15504848,
      "step": 73470
    },
    {
      "epoch": 8.083058305830583,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.022311156166640932,
      "loss": 0.2325,
      "num_input_tokens_seen": 15505872,
      "step": 73475
    },
    {
      "epoch": 8.083608360836084,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02230989872834541,
      "loss": 0.2314,
      "num_input_tokens_seen": 15506928,
      "step": 73480
    },
    {
      "epoch": 8.084158415841584,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.022308641222678054,
      "loss": 0.2299,
      "num_input_tokens_seen": 15507920,
      "step": 73485
    },
    {
      "epoch": 8.084708470847085,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.022307383649650465,
      "loss": 0.2325,
      "num_input_tokens_seen": 15509104,
      "step": 73490
    },
    {
      "epoch": 8.085258525852586,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02230612600927423,
      "loss": 0.2314,
      "num_input_tokens_seen": 15510096,
      "step": 73495
    },
    {
      "epoch": 8.085808580858085,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.02230486830156093,
      "loss": 0.2309,
      "num_input_tokens_seen": 15511152,
      "step": 73500
    },
    {
      "epoch": 8.086358635863586,
      "grad_norm": 0.0004711151123046875,
      "learning_rate": 0.022303610526522164,
      "loss": 0.2335,
      "num_input_tokens_seen": 15512144,
      "step": 73505
    },
    {
      "epoch": 8.086908690869087,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02230235268416953,
      "loss": 0.2304,
      "num_input_tokens_seen": 15513200,
      "step": 73510
    },
    {
      "epoch": 8.087458745874587,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02230109477451461,
      "loss": 0.2298,
      "num_input_tokens_seen": 15514288,
      "step": 73515
    },
    {
      "epoch": 8.088008800880088,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.02229983679756901,
      "loss": 0.2308,
      "num_input_tokens_seen": 15515312,
      "step": 73520
    },
    {
      "epoch": 8.088558855885589,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.022298578753344313,
      "loss": 0.2314,
      "num_input_tokens_seen": 15516432,
      "step": 73525
    },
    {
      "epoch": 8.089108910891088,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.022297320641852116,
      "loss": 0.2335,
      "num_input_tokens_seen": 15517488,
      "step": 73530
    },
    {
      "epoch": 8.08965896589659,
      "grad_norm": 0.0004825592041015625,
      "learning_rate": 0.02229606246310402,
      "loss": 0.2319,
      "num_input_tokens_seen": 15518544,
      "step": 73535
    },
    {
      "epoch": 8.09020902090209,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.02229480421711162,
      "loss": 0.2309,
      "num_input_tokens_seen": 15519632,
      "step": 73540
    },
    {
      "epoch": 8.090759075907592,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0222935459038865,
      "loss": 0.2319,
      "num_input_tokens_seen": 15520624,
      "step": 73545
    },
    {
      "epoch": 8.091309130913091,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.02229228752344027,
      "loss": 0.234,
      "num_input_tokens_seen": 15521680,
      "step": 73550
    },
    {
      "epoch": 8.091859185918592,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02229102907578453,
      "loss": 0.2293,
      "num_input_tokens_seen": 15522704,
      "step": 73555
    },
    {
      "epoch": 8.092409240924093,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.022289770560930865,
      "loss": 0.2314,
      "num_input_tokens_seen": 15523760,
      "step": 73560
    },
    {
      "epoch": 8.092959295929592,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.022288511978890896,
      "loss": 0.2313,
      "num_input_tokens_seen": 15524816,
      "step": 73565
    },
    {
      "epoch": 8.093509350935093,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.0222872533296762,
      "loss": 0.2314,
      "num_input_tokens_seen": 15525776,
      "step": 73570
    },
    {
      "epoch": 8.094059405940595,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.022285994613298384,
      "loss": 0.2314,
      "num_input_tokens_seen": 15526832,
      "step": 73575
    },
    {
      "epoch": 8.094609460946094,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02228473582976906,
      "loss": 0.2329,
      "num_input_tokens_seen": 15527920,
      "step": 73580
    },
    {
      "epoch": 8.095159515951595,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02228347697909982,
      "loss": 0.2329,
      "num_input_tokens_seen": 15528976,
      "step": 73585
    },
    {
      "epoch": 8.095709570957096,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.022282218061302265,
      "loss": 0.2303,
      "num_input_tokens_seen": 15529968,
      "step": 73590
    },
    {
      "epoch": 8.096259625962595,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.022280959076388,
      "loss": 0.2313,
      "num_input_tokens_seen": 15531056,
      "step": 73595
    },
    {
      "epoch": 8.096809680968097,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.02227970002436863,
      "loss": 0.2319,
      "num_input_tokens_seen": 15532176,
      "step": 73600
    },
    {
      "epoch": 8.097359735973598,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.022278440905255756,
      "loss": 0.2319,
      "num_input_tokens_seen": 15533200,
      "step": 73605
    },
    {
      "epoch": 8.097909790979099,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022277181719060988,
      "loss": 0.2308,
      "num_input_tokens_seen": 15534256,
      "step": 73610
    },
    {
      "epoch": 8.098459845984598,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.022275922465795926,
      "loss": 0.2308,
      "num_input_tokens_seen": 15535344,
      "step": 73615
    },
    {
      "epoch": 8.099009900990099,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02227466314547218,
      "loss": 0.2319,
      "num_input_tokens_seen": 15536400,
      "step": 73620
    },
    {
      "epoch": 8.0995599559956,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.022273403758101353,
      "loss": 0.2319,
      "num_input_tokens_seen": 15537488,
      "step": 73625
    },
    {
      "epoch": 8.1001100110011,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.022272144303695056,
      "loss": 0.2314,
      "num_input_tokens_seen": 15538544,
      "step": 73630
    },
    {
      "epoch": 8.1006600660066,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.022270884782264893,
      "loss": 0.2314,
      "num_input_tokens_seen": 15539600,
      "step": 73635
    },
    {
      "epoch": 8.101210121012102,
      "grad_norm": 0.0004787445068359375,
      "learning_rate": 0.022269625193822473,
      "loss": 0.2298,
      "num_input_tokens_seen": 15540592,
      "step": 73640
    },
    {
      "epoch": 8.101760176017601,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.022268365538379402,
      "loss": 0.2314,
      "num_input_tokens_seen": 15541712,
      "step": 73645
    },
    {
      "epoch": 8.102310231023102,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.022267105815947298,
      "loss": 0.2319,
      "num_input_tokens_seen": 15542736,
      "step": 73650
    },
    {
      "epoch": 8.102860286028603,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.022265846026537767,
      "loss": 0.2345,
      "num_input_tokens_seen": 15543824,
      "step": 73655
    },
    {
      "epoch": 8.103410341034103,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.022264586170162422,
      "loss": 0.2319,
      "num_input_tokens_seen": 15544848,
      "step": 73660
    },
    {
      "epoch": 8.103960396039604,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.022263326246832867,
      "loss": 0.2309,
      "num_input_tokens_seen": 15545904,
      "step": 73665
    },
    {
      "epoch": 8.104510451045105,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.022262066256560725,
      "loss": 0.2304,
      "num_input_tokens_seen": 15546960,
      "step": 73670
    },
    {
      "epoch": 8.105060506050606,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.022260806199357597,
      "loss": 0.233,
      "num_input_tokens_seen": 15548016,
      "step": 73675
    },
    {
      "epoch": 8.105610561056105,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.022259546075235108,
      "loss": 0.233,
      "num_input_tokens_seen": 15549104,
      "step": 73680
    },
    {
      "epoch": 8.106160616061606,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.022258285884204865,
      "loss": 0.2309,
      "num_input_tokens_seen": 15550192,
      "step": 73685
    },
    {
      "epoch": 8.106710671067107,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02225702562627848,
      "loss": 0.2314,
      "num_input_tokens_seen": 15551248,
      "step": 73690
    },
    {
      "epoch": 8.107260726072607,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.022255765301467573,
      "loss": 0.2324,
      "num_input_tokens_seen": 15552336,
      "step": 73695
    },
    {
      "epoch": 8.107810781078108,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.022254504909783764,
      "loss": 0.2314,
      "num_input_tokens_seen": 15553296,
      "step": 73700
    },
    {
      "epoch": 8.108360836083609,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.022253244451238656,
      "loss": 0.2309,
      "num_input_tokens_seen": 15554320,
      "step": 73705
    },
    {
      "epoch": 8.108910891089108,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.02225198392584388,
      "loss": 0.233,
      "num_input_tokens_seen": 15555376,
      "step": 73710
    },
    {
      "epoch": 8.10946094609461,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.02225072333361105,
      "loss": 0.2319,
      "num_input_tokens_seen": 15556432,
      "step": 73715
    },
    {
      "epoch": 8.11001100110011,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.022249462674551773,
      "loss": 0.2325,
      "num_input_tokens_seen": 15557456,
      "step": 73720
    },
    {
      "epoch": 8.110561056105611,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.022248201948677684,
      "loss": 0.2324,
      "num_input_tokens_seen": 15558608,
      "step": 73725
    },
    {
      "epoch": 8.11111111111111,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.022246941156000392,
      "loss": 0.2314,
      "num_input_tokens_seen": 15559728,
      "step": 73730
    },
    {
      "epoch": 8.111661166116612,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.02224568029653153,
      "loss": 0.233,
      "num_input_tokens_seen": 15560848,
      "step": 73735
    },
    {
      "epoch": 8.112211221122113,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.0222444193702827,
      "loss": 0.2278,
      "num_input_tokens_seen": 15561904,
      "step": 73740
    },
    {
      "epoch": 8.112761276127612,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.022243158377265537,
      "loss": 0.2329,
      "num_input_tokens_seen": 15562928,
      "step": 73745
    },
    {
      "epoch": 8.113311331133113,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.022241897317491655,
      "loss": 0.2309,
      "num_input_tokens_seen": 15564016,
      "step": 73750
    },
    {
      "epoch": 8.113861386138614,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.022240636190972687,
      "loss": 0.2325,
      "num_input_tokens_seen": 15565072,
      "step": 73755
    },
    {
      "epoch": 8.114411441144114,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.022239374997720242,
      "loss": 0.2288,
      "num_input_tokens_seen": 15566128,
      "step": 73760
    },
    {
      "epoch": 8.114961496149615,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.022238113737745955,
      "loss": 0.2345,
      "num_input_tokens_seen": 15567184,
      "step": 73765
    },
    {
      "epoch": 8.115511551155116,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.02223685241106145,
      "loss": 0.2314,
      "num_input_tokens_seen": 15568240,
      "step": 73770
    },
    {
      "epoch": 8.116061606160615,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.022235591017678347,
      "loss": 0.2335,
      "num_input_tokens_seen": 15569264,
      "step": 73775
    },
    {
      "epoch": 8.116611661166116,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022234329557608273,
      "loss": 0.2319,
      "num_input_tokens_seen": 15570384,
      "step": 73780
    },
    {
      "epoch": 8.117161716171617,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.022233068030862856,
      "loss": 0.2319,
      "num_input_tokens_seen": 15571472,
      "step": 73785
    },
    {
      "epoch": 8.117711771177119,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02223180643745372,
      "loss": 0.2319,
      "num_input_tokens_seen": 15572528,
      "step": 73790
    },
    {
      "epoch": 8.118261826182618,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.022230544777392492,
      "loss": 0.2303,
      "num_input_tokens_seen": 15573648,
      "step": 73795
    },
    {
      "epoch": 8.118811881188119,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.022229283050690805,
      "loss": 0.2313,
      "num_input_tokens_seen": 15574768,
      "step": 73800
    },
    {
      "epoch": 8.11936193619362,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.022228021257360283,
      "loss": 0.2308,
      "num_input_tokens_seen": 15575792,
      "step": 73805
    },
    {
      "epoch": 8.11991199119912,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.022226759397412563,
      "loss": 0.2324,
      "num_input_tokens_seen": 15576816,
      "step": 73810
    },
    {
      "epoch": 8.12046204620462,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02222549747085926,
      "loss": 0.2313,
      "num_input_tokens_seen": 15577840,
      "step": 73815
    },
    {
      "epoch": 8.121012101210122,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.02222423547771202,
      "loss": 0.2314,
      "num_input_tokens_seen": 15578832,
      "step": 73820
    },
    {
      "epoch": 8.12156215621562,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.02222297341798247,
      "loss": 0.2309,
      "num_input_tokens_seen": 15579856,
      "step": 73825
    },
    {
      "epoch": 8.122112211221122,
      "grad_norm": 0.000606536865234375,
      "learning_rate": 0.02222171129168224,
      "loss": 0.2324,
      "num_input_tokens_seen": 15580976,
      "step": 73830
    },
    {
      "epoch": 8.122662266226623,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.022220449098822957,
      "loss": 0.2319,
      "num_input_tokens_seen": 15582000,
      "step": 73835
    },
    {
      "epoch": 8.123212321232122,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02221918683941626,
      "loss": 0.2303,
      "num_input_tokens_seen": 15583088,
      "step": 73840
    },
    {
      "epoch": 8.123762376237623,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.022217924513473788,
      "loss": 0.2293,
      "num_input_tokens_seen": 15584144,
      "step": 73845
    },
    {
      "epoch": 8.124312431243125,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.022216662121007168,
      "loss": 0.2308,
      "num_input_tokens_seen": 15585200,
      "step": 73850
    },
    {
      "epoch": 8.124862486248626,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.022215399662028033,
      "loss": 0.2314,
      "num_input_tokens_seen": 15586192,
      "step": 73855
    },
    {
      "epoch": 8.125412541254125,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.02221413713654802,
      "loss": 0.2314,
      "num_input_tokens_seen": 15587280,
      "step": 73860
    },
    {
      "epoch": 8.125962596259626,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.022212874544578768,
      "loss": 0.2308,
      "num_input_tokens_seen": 15588368,
      "step": 73865
    },
    {
      "epoch": 8.126512651265127,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.022211611886131914,
      "loss": 0.2313,
      "num_input_tokens_seen": 15589392,
      "step": 73870
    },
    {
      "epoch": 8.127062706270626,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.022210349161219094,
      "loss": 0.2309,
      "num_input_tokens_seen": 15590384,
      "step": 73875
    },
    {
      "epoch": 8.127612761276128,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.022209086369851944,
      "loss": 0.2309,
      "num_input_tokens_seen": 15591408,
      "step": 73880
    },
    {
      "epoch": 8.128162816281629,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.022207823512042108,
      "loss": 0.2335,
      "num_input_tokens_seen": 15592496,
      "step": 73885
    },
    {
      "epoch": 8.128712871287128,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.022206560587801218,
      "loss": 0.2319,
      "num_input_tokens_seen": 15593648,
      "step": 73890
    },
    {
      "epoch": 8.129262926292629,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.022205297597140915,
      "loss": 0.2308,
      "num_input_tokens_seen": 15594704,
      "step": 73895
    },
    {
      "epoch": 8.12981298129813,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.02220403454007285,
      "loss": 0.2314,
      "num_input_tokens_seen": 15595696,
      "step": 73900
    },
    {
      "epoch": 8.130363036303631,
      "grad_norm": 0.000766754150390625,
      "learning_rate": 0.022202771416608647,
      "loss": 0.2303,
      "num_input_tokens_seen": 15596784,
      "step": 73905
    },
    {
      "epoch": 8.13091309130913,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.022201508226759956,
      "loss": 0.2313,
      "num_input_tokens_seen": 15597936,
      "step": 73910
    },
    {
      "epoch": 8.131463146314632,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02220024497053842,
      "loss": 0.2324,
      "num_input_tokens_seen": 15598992,
      "step": 73915
    },
    {
      "epoch": 8.132013201320133,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.022198981647955685,
      "loss": 0.2319,
      "num_input_tokens_seen": 15600176,
      "step": 73920
    },
    {
      "epoch": 8.132563256325632,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.022197718259023388,
      "loss": 0.2298,
      "num_input_tokens_seen": 15601232,
      "step": 73925
    },
    {
      "epoch": 8.133113311331133,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.022196454803753176,
      "loss": 0.2329,
      "num_input_tokens_seen": 15602320,
      "step": 73930
    },
    {
      "epoch": 8.133663366336634,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.022195191282156693,
      "loss": 0.2334,
      "num_input_tokens_seen": 15603344,
      "step": 73935
    },
    {
      "epoch": 8.134213421342134,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.022193927694245583,
      "loss": 0.2314,
      "num_input_tokens_seen": 15604368,
      "step": 73940
    },
    {
      "epoch": 8.134763476347635,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.0221926640400315,
      "loss": 0.2314,
      "num_input_tokens_seen": 15605392,
      "step": 73945
    },
    {
      "epoch": 8.135313531353136,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.022191400319526076,
      "loss": 0.2335,
      "num_input_tokens_seen": 15606384,
      "step": 73950
    },
    {
      "epoch": 8.135863586358635,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.02219013653274097,
      "loss": 0.2319,
      "num_input_tokens_seen": 15607408,
      "step": 73955
    },
    {
      "epoch": 8.136413641364136,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.022188872679687825,
      "loss": 0.2298,
      "num_input_tokens_seen": 15608432,
      "step": 73960
    },
    {
      "epoch": 8.136963696369637,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.02218760876037829,
      "loss": 0.2314,
      "num_input_tokens_seen": 15609520,
      "step": 73965
    },
    {
      "epoch": 8.137513751375138,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.022186344774824014,
      "loss": 0.2314,
      "num_input_tokens_seen": 15610640,
      "step": 73970
    },
    {
      "epoch": 8.138063806380638,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.022185080723036647,
      "loss": 0.2298,
      "num_input_tokens_seen": 15611664,
      "step": 73975
    },
    {
      "epoch": 8.138613861386139,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.022183816605027835,
      "loss": 0.2314,
      "num_input_tokens_seen": 15612688,
      "step": 73980
    },
    {
      "epoch": 8.13916391639164,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.022182552420809236,
      "loss": 0.2309,
      "num_input_tokens_seen": 15613680,
      "step": 73985
    },
    {
      "epoch": 8.13971397139714,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.0221812881703925,
      "loss": 0.2298,
      "num_input_tokens_seen": 15614800,
      "step": 73990
    },
    {
      "epoch": 8.14026402640264,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.022180023853789275,
      "loss": 0.2308,
      "num_input_tokens_seen": 15615856,
      "step": 73995
    },
    {
      "epoch": 8.140814081408141,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.022178759471011215,
      "loss": 0.2304,
      "num_input_tokens_seen": 15616912,
      "step": 74000
    },
    {
      "epoch": 8.14136413641364,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.022177495022069973,
      "loss": 0.233,
      "num_input_tokens_seen": 15617936,
      "step": 74005
    },
    {
      "epoch": 8.141914191419142,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.0221762305069772,
      "loss": 0.2324,
      "num_input_tokens_seen": 15618928,
      "step": 74010
    },
    {
      "epoch": 8.142464246424643,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.022174965925744566,
      "loss": 0.2329,
      "num_input_tokens_seen": 15619920,
      "step": 74015
    },
    {
      "epoch": 8.143014301430142,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.022173701278383702,
      "loss": 0.2319,
      "num_input_tokens_seen": 15620944,
      "step": 74020
    },
    {
      "epoch": 8.143564356435643,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.022172436564906278,
      "loss": 0.2288,
      "num_input_tokens_seen": 15622032,
      "step": 74025
    },
    {
      "epoch": 8.144114411441144,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.022171171785323948,
      "loss": 0.2308,
      "num_input_tokens_seen": 15623024,
      "step": 74030
    },
    {
      "epoch": 8.144664466446645,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.02216990693964837,
      "loss": 0.2293,
      "num_input_tokens_seen": 15624112,
      "step": 74035
    },
    {
      "epoch": 8.145214521452145,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.022168642027891202,
      "loss": 0.2335,
      "num_input_tokens_seen": 15625168,
      "step": 74040
    },
    {
      "epoch": 8.145764576457646,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.0221673770500641,
      "loss": 0.2314,
      "num_input_tokens_seen": 15626192,
      "step": 74045
    },
    {
      "epoch": 8.146314631463147,
      "grad_norm": 0.0027923583984375,
      "learning_rate": 0.022166112006178722,
      "loss": 0.2335,
      "num_input_tokens_seen": 15627312,
      "step": 74050
    },
    {
      "epoch": 8.146864686468646,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.022164846896246725,
      "loss": 0.2293,
      "num_input_tokens_seen": 15628368,
      "step": 74055
    },
    {
      "epoch": 8.147414741474147,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.022163581720279775,
      "loss": 0.2314,
      "num_input_tokens_seen": 15629392,
      "step": 74060
    },
    {
      "epoch": 8.147964796479648,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02216231647828953,
      "loss": 0.2324,
      "num_input_tokens_seen": 15630448,
      "step": 74065
    },
    {
      "epoch": 8.148514851485148,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.02216105117028765,
      "loss": 0.2319,
      "num_input_tokens_seen": 15631472,
      "step": 74070
    },
    {
      "epoch": 8.149064906490649,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.0221597857962858,
      "loss": 0.2319,
      "num_input_tokens_seen": 15632560,
      "step": 74075
    },
    {
      "epoch": 8.14961496149615,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.02215852035629564,
      "loss": 0.2314,
      "num_input_tokens_seen": 15633648,
      "step": 74080
    },
    {
      "epoch": 8.150165016501651,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.022157254850328826,
      "loss": 0.2309,
      "num_input_tokens_seen": 15634640,
      "step": 74085
    },
    {
      "epoch": 8.15071507150715,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02215598927839704,
      "loss": 0.2303,
      "num_input_tokens_seen": 15635664,
      "step": 74090
    },
    {
      "epoch": 8.151265126512651,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022154723640511922,
      "loss": 0.2319,
      "num_input_tokens_seen": 15636784,
      "step": 74095
    },
    {
      "epoch": 8.151815181518153,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.022153457936685154,
      "loss": 0.2309,
      "num_input_tokens_seen": 15637808,
      "step": 74100
    },
    {
      "epoch": 8.152365236523652,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.0221521921669284,
      "loss": 0.2314,
      "num_input_tokens_seen": 15638896,
      "step": 74105
    },
    {
      "epoch": 8.152915291529153,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.02215092633125332,
      "loss": 0.2309,
      "num_input_tokens_seen": 15639920,
      "step": 74110
    },
    {
      "epoch": 8.153465346534654,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.022149660429671585,
      "loss": 0.2314,
      "num_input_tokens_seen": 15641040,
      "step": 74115
    },
    {
      "epoch": 8.154015401540153,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.02214839446219486,
      "loss": 0.2309,
      "num_input_tokens_seen": 15642096,
      "step": 74120
    },
    {
      "epoch": 8.154565456545654,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02214712842883481,
      "loss": 0.233,
      "num_input_tokens_seen": 15643152,
      "step": 74125
    },
    {
      "epoch": 8.155115511551156,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02214586232960311,
      "loss": 0.2304,
      "num_input_tokens_seen": 15644208,
      "step": 74130
    },
    {
      "epoch": 8.155665566556655,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.02214459616451143,
      "loss": 0.2314,
      "num_input_tokens_seen": 15645232,
      "step": 74135
    },
    {
      "epoch": 8.156215621562156,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.02214332993357143,
      "loss": 0.2309,
      "num_input_tokens_seen": 15646288,
      "step": 74140
    },
    {
      "epoch": 8.156765676567657,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.022142063636794787,
      "loss": 0.2309,
      "num_input_tokens_seen": 15647280,
      "step": 74145
    },
    {
      "epoch": 8.157315731573158,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.022140797274193172,
      "loss": 0.2288,
      "num_input_tokens_seen": 15648336,
      "step": 74150
    },
    {
      "epoch": 8.157865786578657,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02213953084577825,
      "loss": 0.2288,
      "num_input_tokens_seen": 15649392,
      "step": 74155
    },
    {
      "epoch": 8.158415841584159,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.022138264351561708,
      "loss": 0.2345,
      "num_input_tokens_seen": 15650416,
      "step": 74160
    },
    {
      "epoch": 8.15896589658966,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0221369977915552,
      "loss": 0.2298,
      "num_input_tokens_seen": 15651408,
      "step": 74165
    },
    {
      "epoch": 8.159515951595159,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02213573116577041,
      "loss": 0.2319,
      "num_input_tokens_seen": 15652464,
      "step": 74170
    },
    {
      "epoch": 8.16006600660066,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.022134464474219007,
      "loss": 0.2308,
      "num_input_tokens_seen": 15653488,
      "step": 74175
    },
    {
      "epoch": 8.160616061606161,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.022133197716912672,
      "loss": 0.2335,
      "num_input_tokens_seen": 15654608,
      "step": 74180
    },
    {
      "epoch": 8.16116611661166,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.02213193089386308,
      "loss": 0.2309,
      "num_input_tokens_seen": 15655696,
      "step": 74185
    },
    {
      "epoch": 8.161716171617162,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.022130664005081898,
      "loss": 0.2324,
      "num_input_tokens_seen": 15656784,
      "step": 74190
    },
    {
      "epoch": 8.162266226622663,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022129397050580805,
      "loss": 0.2308,
      "num_input_tokens_seen": 15657776,
      "step": 74195
    },
    {
      "epoch": 8.162816281628162,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.022128130030371482,
      "loss": 0.2314,
      "num_input_tokens_seen": 15658832,
      "step": 74200
    },
    {
      "epoch": 8.163366336633663,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.02212686294446561,
      "loss": 0.2267,
      "num_input_tokens_seen": 15659888,
      "step": 74205
    },
    {
      "epoch": 8.163916391639164,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.022125595792874864,
      "loss": 0.2314,
      "num_input_tokens_seen": 15660976,
      "step": 74210
    },
    {
      "epoch": 8.164466446644665,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.022124328575610912,
      "loss": 0.2314,
      "num_input_tokens_seen": 15662064,
      "step": 74215
    },
    {
      "epoch": 8.165016501650165,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.022123061292685445,
      "loss": 0.2324,
      "num_input_tokens_seen": 15663152,
      "step": 74220
    },
    {
      "epoch": 8.165566556655666,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.022121793944110145,
      "loss": 0.2319,
      "num_input_tokens_seen": 15664240,
      "step": 74225
    },
    {
      "epoch": 8.166116611661167,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.02212052652989668,
      "loss": 0.2308,
      "num_input_tokens_seen": 15665264,
      "step": 74230
    },
    {
      "epoch": 8.166666666666666,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.022119259050056742,
      "loss": 0.2314,
      "num_input_tokens_seen": 15666288,
      "step": 74235
    },
    {
      "epoch": 8.167216721672167,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02211799150460201,
      "loss": 0.2324,
      "num_input_tokens_seen": 15667376,
      "step": 74240
    },
    {
      "epoch": 8.167766776677668,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.022116723893544167,
      "loss": 0.2314,
      "num_input_tokens_seen": 15668464,
      "step": 74245
    },
    {
      "epoch": 8.168316831683168,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02211545621689489,
      "loss": 0.2293,
      "num_input_tokens_seen": 15669552,
      "step": 74250
    },
    {
      "epoch": 8.168866886688669,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.022114188474665872,
      "loss": 0.2314,
      "num_input_tokens_seen": 15670608,
      "step": 74255
    },
    {
      "epoch": 8.16941694169417,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02211292066686879,
      "loss": 0.234,
      "num_input_tokens_seen": 15671696,
      "step": 74260
    },
    {
      "epoch": 8.16996699669967,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.022111652793515336,
      "loss": 0.2324,
      "num_input_tokens_seen": 15672816,
      "step": 74265
    },
    {
      "epoch": 8.17051705170517,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.022110384854617183,
      "loss": 0.2324,
      "num_input_tokens_seen": 15673904,
      "step": 74270
    },
    {
      "epoch": 8.171067106710671,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02210911685018603,
      "loss": 0.2324,
      "num_input_tokens_seen": 15674992,
      "step": 74275
    },
    {
      "epoch": 8.171617161716172,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02210784878023356,
      "loss": 0.2319,
      "num_input_tokens_seen": 15676048,
      "step": 74280
    },
    {
      "epoch": 8.172167216721672,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.022106580644771452,
      "loss": 0.2309,
      "num_input_tokens_seen": 15677072,
      "step": 74285
    },
    {
      "epoch": 8.172717271727173,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.022105312443811406,
      "loss": 0.2309,
      "num_input_tokens_seen": 15678160,
      "step": 74290
    },
    {
      "epoch": 8.173267326732674,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.022104044177365105,
      "loss": 0.2298,
      "num_input_tokens_seen": 15679216,
      "step": 74295
    },
    {
      "epoch": 8.173817381738173,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.022102775845444237,
      "loss": 0.234,
      "num_input_tokens_seen": 15680240,
      "step": 74300
    },
    {
      "epoch": 8.174367436743674,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.022101507448060487,
      "loss": 0.2319,
      "num_input_tokens_seen": 15681264,
      "step": 74305
    },
    {
      "epoch": 8.174917491749175,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02210023898522556,
      "loss": 0.2293,
      "num_input_tokens_seen": 15682320,
      "step": 74310
    },
    {
      "epoch": 8.175467546754675,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.022098970456951125,
      "loss": 0.2298,
      "num_input_tokens_seen": 15683344,
      "step": 74315
    },
    {
      "epoch": 8.176017601760176,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.022097701863248893,
      "loss": 0.2324,
      "num_input_tokens_seen": 15684336,
      "step": 74320
    },
    {
      "epoch": 8.176567656765677,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02209643320413055,
      "loss": 0.2319,
      "num_input_tokens_seen": 15685360,
      "step": 74325
    },
    {
      "epoch": 8.177117711771178,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.022095164479607785,
      "loss": 0.2335,
      "num_input_tokens_seen": 15686416,
      "step": 74330
    },
    {
      "epoch": 8.177667766776677,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02209389568969229,
      "loss": 0.2329,
      "num_input_tokens_seen": 15687440,
      "step": 74335
    },
    {
      "epoch": 8.178217821782178,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.022092626834395772,
      "loss": 0.2308,
      "num_input_tokens_seen": 15688464,
      "step": 74340
    },
    {
      "epoch": 8.17876787678768,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.022091357913729907,
      "loss": 0.2314,
      "num_input_tokens_seen": 15689488,
      "step": 74345
    },
    {
      "epoch": 8.179317931793179,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.022090088927706406,
      "loss": 0.2304,
      "num_input_tokens_seen": 15690512,
      "step": 74350
    },
    {
      "epoch": 8.17986798679868,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.022088819876336954,
      "loss": 0.2314,
      "num_input_tokens_seen": 15691536,
      "step": 74355
    },
    {
      "epoch": 8.180418041804181,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.02208755075963325,
      "loss": 0.2319,
      "num_input_tokens_seen": 15692592,
      "step": 74360
    },
    {
      "epoch": 8.18096809680968,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02208628157760699,
      "loss": 0.2309,
      "num_input_tokens_seen": 15693616,
      "step": 74365
    },
    {
      "epoch": 8.181518151815181,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.022085012330269875,
      "loss": 0.2319,
      "num_input_tokens_seen": 15694640,
      "step": 74370
    },
    {
      "epoch": 8.182068206820682,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.022083743017633602,
      "loss": 0.2308,
      "num_input_tokens_seen": 15695632,
      "step": 74375
    },
    {
      "epoch": 8.182618261826182,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.022082473639709867,
      "loss": 0.2309,
      "num_input_tokens_seen": 15696688,
      "step": 74380
    },
    {
      "epoch": 8.183168316831683,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.022081204196510374,
      "loss": 0.2283,
      "num_input_tokens_seen": 15697712,
      "step": 74385
    },
    {
      "epoch": 8.183718371837184,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.022079934688046813,
      "loss": 0.2319,
      "num_input_tokens_seen": 15698832,
      "step": 74390
    },
    {
      "epoch": 8.184268426842685,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.022078665114330896,
      "loss": 0.2309,
      "num_input_tokens_seen": 15699856,
      "step": 74395
    },
    {
      "epoch": 8.184818481848184,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02207739547537432,
      "loss": 0.2319,
      "num_input_tokens_seen": 15700912,
      "step": 74400
    },
    {
      "epoch": 8.185368536853685,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.02207612577118878,
      "loss": 0.2335,
      "num_input_tokens_seen": 15701936,
      "step": 74405
    },
    {
      "epoch": 8.185918591859187,
      "grad_norm": 0.002777099609375,
      "learning_rate": 0.022074856001785992,
      "loss": 0.2314,
      "num_input_tokens_seen": 15702992,
      "step": 74410
    },
    {
      "epoch": 8.186468646864686,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02207358616717765,
      "loss": 0.2335,
      "num_input_tokens_seen": 15703984,
      "step": 74415
    },
    {
      "epoch": 8.187018701870187,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.022072316267375457,
      "loss": 0.2283,
      "num_input_tokens_seen": 15705040,
      "step": 74420
    },
    {
      "epoch": 8.187568756875688,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.02207104630239112,
      "loss": 0.2319,
      "num_input_tokens_seen": 15706160,
      "step": 74425
    },
    {
      "epoch": 8.188118811881187,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.022069776272236337,
      "loss": 0.2309,
      "num_input_tokens_seen": 15707248,
      "step": 74430
    },
    {
      "epoch": 8.188668866886688,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.022068506176922822,
      "loss": 0.2314,
      "num_input_tokens_seen": 15708304,
      "step": 74435
    },
    {
      "epoch": 8.18921892189219,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02206723601646228,
      "loss": 0.2314,
      "num_input_tokens_seen": 15709424,
      "step": 74440
    },
    {
      "epoch": 8.189768976897689,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.022065965790866414,
      "loss": 0.2314,
      "num_input_tokens_seen": 15710448,
      "step": 74445
    },
    {
      "epoch": 8.19031903190319,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02206469550014693,
      "loss": 0.2314,
      "num_input_tokens_seen": 15711504,
      "step": 74450
    },
    {
      "epoch": 8.190869086908691,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.022063425144315536,
      "loss": 0.2324,
      "num_input_tokens_seen": 15712528,
      "step": 74455
    },
    {
      "epoch": 8.191419141914192,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.022062154723383944,
      "loss": 0.2324,
      "num_input_tokens_seen": 15713584,
      "step": 74460
    },
    {
      "epoch": 8.191969196919691,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.022060884237363857,
      "loss": 0.2314,
      "num_input_tokens_seen": 15714640,
      "step": 74465
    },
    {
      "epoch": 8.192519251925193,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.022059613686266997,
      "loss": 0.2324,
      "num_input_tokens_seen": 15715696,
      "step": 74470
    },
    {
      "epoch": 8.193069306930694,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.022058343070105058,
      "loss": 0.2309,
      "num_input_tokens_seen": 15716688,
      "step": 74475
    },
    {
      "epoch": 8.193619361936193,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.022057072388889758,
      "loss": 0.2288,
      "num_input_tokens_seen": 15717840,
      "step": 74480
    },
    {
      "epoch": 8.194169416941694,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.02205580164263281,
      "loss": 0.2324,
      "num_input_tokens_seen": 15718928,
      "step": 74485
    },
    {
      "epoch": 8.194719471947195,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.022054530831345922,
      "loss": 0.2314,
      "num_input_tokens_seen": 15719952,
      "step": 74490
    },
    {
      "epoch": 8.195269526952695,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.022053259955040812,
      "loss": 0.2309,
      "num_input_tokens_seen": 15721040,
      "step": 74495
    },
    {
      "epoch": 8.195819581958196,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02205198901372919,
      "loss": 0.2314,
      "num_input_tokens_seen": 15722096,
      "step": 74500
    },
    {
      "epoch": 8.196369636963697,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.022050718007422764,
      "loss": 0.2314,
      "num_input_tokens_seen": 15723088,
      "step": 74505
    },
    {
      "epoch": 8.196919691969198,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.022049446936133255,
      "loss": 0.235,
      "num_input_tokens_seen": 15724080,
      "step": 74510
    },
    {
      "epoch": 8.197469746974697,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.02204817579987238,
      "loss": 0.2314,
      "num_input_tokens_seen": 15725136,
      "step": 74515
    },
    {
      "epoch": 8.198019801980198,
      "grad_norm": 0.000797271728515625,
      "learning_rate": 0.02204690459865185,
      "loss": 0.2309,
      "num_input_tokens_seen": 15726192,
      "step": 74520
    },
    {
      "epoch": 8.1985698569857,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.022045633332483376,
      "loss": 0.2309,
      "num_input_tokens_seen": 15727280,
      "step": 74525
    },
    {
      "epoch": 8.199119911991199,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022044362001378683,
      "loss": 0.2309,
      "num_input_tokens_seen": 15728336,
      "step": 74530
    },
    {
      "epoch": 8.1996699669967,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.022043090605349487,
      "loss": 0.2319,
      "num_input_tokens_seen": 15729392,
      "step": 74535
    },
    {
      "epoch": 8.2002200220022,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.022041819144407507,
      "loss": 0.2303,
      "num_input_tokens_seen": 15730384,
      "step": 74540
    },
    {
      "epoch": 8.2007700770077,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02204054761856445,
      "loss": 0.2299,
      "num_input_tokens_seen": 15731472,
      "step": 74545
    },
    {
      "epoch": 8.201320132013201,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.022039276027832056,
      "loss": 0.2298,
      "num_input_tokens_seen": 15732432,
      "step": 74550
    },
    {
      "epoch": 8.201870187018702,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.022038004372222024,
      "loss": 0.2319,
      "num_input_tokens_seen": 15733520,
      "step": 74555
    },
    {
      "epoch": 8.202420242024202,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.022036732651746084,
      "loss": 0.2335,
      "num_input_tokens_seen": 15734608,
      "step": 74560
    },
    {
      "epoch": 8.202970297029703,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02203546086641596,
      "loss": 0.2298,
      "num_input_tokens_seen": 15735760,
      "step": 74565
    },
    {
      "epoch": 8.203520352035204,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.022034189016243367,
      "loss": 0.233,
      "num_input_tokens_seen": 15736816,
      "step": 74570
    },
    {
      "epoch": 8.204070407040705,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.022032917101240026,
      "loss": 0.2319,
      "num_input_tokens_seen": 15737840,
      "step": 74575
    },
    {
      "epoch": 8.204620462046204,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.022031645121417663,
      "loss": 0.2304,
      "num_input_tokens_seen": 15738864,
      "step": 74580
    },
    {
      "epoch": 8.205170517051705,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.022030373076788005,
      "loss": 0.2304,
      "num_input_tokens_seen": 15739984,
      "step": 74585
    },
    {
      "epoch": 8.205720572057206,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.022029100967362773,
      "loss": 0.233,
      "num_input_tokens_seen": 15741008,
      "step": 74590
    },
    {
      "epoch": 8.206270627062706,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.022027828793153688,
      "loss": 0.2298,
      "num_input_tokens_seen": 15742064,
      "step": 74595
    },
    {
      "epoch": 8.206820682068207,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.022026556554172478,
      "loss": 0.2325,
      "num_input_tokens_seen": 15743120,
      "step": 74600
    },
    {
      "epoch": 8.207370737073708,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.022025284250430863,
      "loss": 0.233,
      "num_input_tokens_seen": 15744080,
      "step": 74605
    },
    {
      "epoch": 8.207920792079207,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.02202401188194058,
      "loss": 0.2299,
      "num_input_tokens_seen": 15745104,
      "step": 74610
    },
    {
      "epoch": 8.208470847084708,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02202273944871335,
      "loss": 0.2345,
      "num_input_tokens_seen": 15746128,
      "step": 74615
    },
    {
      "epoch": 8.20902090209021,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.022021466950760894,
      "loss": 0.233,
      "num_input_tokens_seen": 15747216,
      "step": 74620
    },
    {
      "epoch": 8.209570957095709,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.022020194388094952,
      "loss": 0.2309,
      "num_input_tokens_seen": 15748208,
      "step": 74625
    },
    {
      "epoch": 8.21012101210121,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.022018921760727245,
      "loss": 0.2298,
      "num_input_tokens_seen": 15749296,
      "step": 74630
    },
    {
      "epoch": 8.210671067106711,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.022017649068669504,
      "loss": 0.2314,
      "num_input_tokens_seen": 15750320,
      "step": 74635
    },
    {
      "epoch": 8.211221122112212,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02201637631193346,
      "loss": 0.2319,
      "num_input_tokens_seen": 15751344,
      "step": 74640
    },
    {
      "epoch": 8.211771177117711,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.022015103490530843,
      "loss": 0.2319,
      "num_input_tokens_seen": 15752400,
      "step": 74645
    },
    {
      "epoch": 8.212321232123212,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02201383060447338,
      "loss": 0.2293,
      "num_input_tokens_seen": 15753520,
      "step": 74650
    },
    {
      "epoch": 8.212871287128714,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.02201255765377281,
      "loss": 0.2319,
      "num_input_tokens_seen": 15754608,
      "step": 74655
    },
    {
      "epoch": 8.213421342134213,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02201128463844086,
      "loss": 0.2298,
      "num_input_tokens_seen": 15755696,
      "step": 74660
    },
    {
      "epoch": 8.213971397139714,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.022010011558489258,
      "loss": 0.234,
      "num_input_tokens_seen": 15756720,
      "step": 74665
    },
    {
      "epoch": 8.214521452145215,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02200873841392975,
      "loss": 0.2336,
      "num_input_tokens_seen": 15757712,
      "step": 74670
    },
    {
      "epoch": 8.215071507150714,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.022007465204774063,
      "loss": 0.2325,
      "num_input_tokens_seen": 15758768,
      "step": 74675
    },
    {
      "epoch": 8.215621562156215,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.022006191931033925,
      "loss": 0.2325,
      "num_input_tokens_seen": 15759792,
      "step": 74680
    },
    {
      "epoch": 8.216171617161717,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02200491859272109,
      "loss": 0.2319,
      "num_input_tokens_seen": 15760816,
      "step": 74685
    },
    {
      "epoch": 8.216721672167218,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.022003645189847272,
      "loss": 0.2278,
      "num_input_tokens_seen": 15761904,
      "step": 74690
    },
    {
      "epoch": 8.217271727172717,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.022002371722424217,
      "loss": 0.2309,
      "num_input_tokens_seen": 15762864,
      "step": 74695
    },
    {
      "epoch": 8.217821782178218,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.022001098190463662,
      "loss": 0.2314,
      "num_input_tokens_seen": 15763920,
      "step": 74700
    },
    {
      "epoch": 8.218371837183719,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02199982459397735,
      "loss": 0.234,
      "num_input_tokens_seen": 15764912,
      "step": 74705
    },
    {
      "epoch": 8.218921892189218,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.02199855093297701,
      "loss": 0.2335,
      "num_input_tokens_seen": 15765968,
      "step": 74710
    },
    {
      "epoch": 8.21947194719472,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.021997277207474387,
      "loss": 0.2309,
      "num_input_tokens_seen": 15767024,
      "step": 74715
    },
    {
      "epoch": 8.22002200220022,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.021996003417481213,
      "loss": 0.2314,
      "num_input_tokens_seen": 15768048,
      "step": 74720
    },
    {
      "epoch": 8.22057205720572,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.021994729563009234,
      "loss": 0.2319,
      "num_input_tokens_seen": 15769072,
      "step": 74725
    },
    {
      "epoch": 8.221122112211221,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.02199345564407019,
      "loss": 0.2294,
      "num_input_tokens_seen": 15770128,
      "step": 74730
    },
    {
      "epoch": 8.221672167216722,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02199218166067582,
      "loss": 0.233,
      "num_input_tokens_seen": 15771184,
      "step": 74735
    },
    {
      "epoch": 8.222222222222221,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.021990907612837868,
      "loss": 0.2304,
      "num_input_tokens_seen": 15772272,
      "step": 74740
    },
    {
      "epoch": 8.222772277227723,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.021989633500568075,
      "loss": 0.2319,
      "num_input_tokens_seen": 15773392,
      "step": 74745
    },
    {
      "epoch": 8.223322332233224,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.021988359323878184,
      "loss": 0.2314,
      "num_input_tokens_seen": 15774416,
      "step": 74750
    },
    {
      "epoch": 8.223872387238725,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.02198708508277994,
      "loss": 0.2319,
      "num_input_tokens_seen": 15775440,
      "step": 74755
    },
    {
      "epoch": 8.224422442244224,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.021985810777285085,
      "loss": 0.234,
      "num_input_tokens_seen": 15776400,
      "step": 74760
    },
    {
      "epoch": 8.224972497249725,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.02198453640740536,
      "loss": 0.2293,
      "num_input_tokens_seen": 15777456,
      "step": 74765
    },
    {
      "epoch": 8.225522552255226,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.021983261973152517,
      "loss": 0.2324,
      "num_input_tokens_seen": 15778544,
      "step": 74770
    },
    {
      "epoch": 8.226072607260726,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.0219819874745383,
      "loss": 0.2319,
      "num_input_tokens_seen": 15779536,
      "step": 74775
    },
    {
      "epoch": 8.226622662266227,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.021980712911574454,
      "loss": 0.2319,
      "num_input_tokens_seen": 15780560,
      "step": 74780
    },
    {
      "epoch": 8.227172717271728,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.021979438284272727,
      "loss": 0.2309,
      "num_input_tokens_seen": 15781584,
      "step": 74785
    },
    {
      "epoch": 8.227722772277227,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.021978163592644866,
      "loss": 0.2303,
      "num_input_tokens_seen": 15782672,
      "step": 74790
    },
    {
      "epoch": 8.228272827282728,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02197688883670262,
      "loss": 0.2324,
      "num_input_tokens_seen": 15783728,
      "step": 74795
    },
    {
      "epoch": 8.22882288228823,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02197561401645774,
      "loss": 0.2309,
      "num_input_tokens_seen": 15784720,
      "step": 74800
    },
    {
      "epoch": 8.229372937293729,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.021974339131921972,
      "loss": 0.2309,
      "num_input_tokens_seen": 15785776,
      "step": 74805
    },
    {
      "epoch": 8.22992299229923,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02197306418310706,
      "loss": 0.2319,
      "num_input_tokens_seen": 15786864,
      "step": 74810
    },
    {
      "epoch": 8.23047304730473,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.02197178917002477,
      "loss": 0.2283,
      "num_input_tokens_seen": 15787920,
      "step": 74815
    },
    {
      "epoch": 8.231023102310232,
      "grad_norm": 0.00058746337890625,
      "learning_rate": 0.021970514092686837,
      "loss": 0.233,
      "num_input_tokens_seen": 15788976,
      "step": 74820
    },
    {
      "epoch": 8.231573157315731,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.021969238951105027,
      "loss": 0.2298,
      "num_input_tokens_seen": 15790032,
      "step": 74825
    },
    {
      "epoch": 8.232123212321232,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.021967963745291082,
      "loss": 0.2293,
      "num_input_tokens_seen": 15791088,
      "step": 74830
    },
    {
      "epoch": 8.232673267326733,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.021966688475256763,
      "loss": 0.2309,
      "num_input_tokens_seen": 15792144,
      "step": 74835
    },
    {
      "epoch": 8.233223322332233,
      "grad_norm": 0.009765625,
      "learning_rate": 0.021965413141013815,
      "loss": 0.2324,
      "num_input_tokens_seen": 15793296,
      "step": 74840
    },
    {
      "epoch": 8.233773377337734,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.021964137742573997,
      "loss": 0.2303,
      "num_input_tokens_seen": 15794416,
      "step": 74845
    },
    {
      "epoch": 8.234323432343235,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021962862279949064,
      "loss": 0.2304,
      "num_input_tokens_seen": 15795440,
      "step": 74850
    },
    {
      "epoch": 8.234873487348734,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.021961586753150773,
      "loss": 0.2314,
      "num_input_tokens_seen": 15796496,
      "step": 74855
    },
    {
      "epoch": 8.235423542354235,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02196031116219087,
      "loss": 0.2319,
      "num_input_tokens_seen": 15797584,
      "step": 74860
    },
    {
      "epoch": 8.235973597359736,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.02195903550708113,
      "loss": 0.232,
      "num_input_tokens_seen": 15798608,
      "step": 74865
    },
    {
      "epoch": 8.236523652365236,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.02195775978783329,
      "loss": 0.2325,
      "num_input_tokens_seen": 15799728,
      "step": 74870
    },
    {
      "epoch": 8.237073707370737,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.02195648400445912,
      "loss": 0.2309,
      "num_input_tokens_seen": 15800720,
      "step": 74875
    },
    {
      "epoch": 8.237623762376238,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.021955208156970375,
      "loss": 0.233,
      "num_input_tokens_seen": 15801808,
      "step": 74880
    },
    {
      "epoch": 8.238173817381739,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.02195393224537881,
      "loss": 0.2308,
      "num_input_tokens_seen": 15802864,
      "step": 74885
    },
    {
      "epoch": 8.238723872387238,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.021952656269696193,
      "loss": 0.2309,
      "num_input_tokens_seen": 15803920,
      "step": 74890
    },
    {
      "epoch": 8.23927392739274,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.02195138022993428,
      "loss": 0.2304,
      "num_input_tokens_seen": 15804944,
      "step": 74895
    },
    {
      "epoch": 8.23982398239824,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.021950104126104826,
      "loss": 0.2283,
      "num_input_tokens_seen": 15806000,
      "step": 74900
    },
    {
      "epoch": 8.24037403740374,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.0219488279582196,
      "loss": 0.2304,
      "num_input_tokens_seen": 15806960,
      "step": 74905
    },
    {
      "epoch": 8.24092409240924,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.02194755172629036,
      "loss": 0.2309,
      "num_input_tokens_seen": 15808016,
      "step": 74910
    },
    {
      "epoch": 8.241474147414742,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02194627543032887,
      "loss": 0.2273,
      "num_input_tokens_seen": 15809104,
      "step": 74915
    },
    {
      "epoch": 8.242024202420241,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.021944999070346898,
      "loss": 0.2278,
      "num_input_tokens_seen": 15810160,
      "step": 74920
    },
    {
      "epoch": 8.242574257425742,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.021943722646356197,
      "loss": 0.2325,
      "num_input_tokens_seen": 15811184,
      "step": 74925
    },
    {
      "epoch": 8.243124312431243,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.021942446158368534,
      "loss": 0.2304,
      "num_input_tokens_seen": 15812208,
      "step": 74930
    },
    {
      "epoch": 8.243674367436745,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.02194116960639568,
      "loss": 0.234,
      "num_input_tokens_seen": 15813296,
      "step": 74935
    },
    {
      "epoch": 8.244224422442244,
      "grad_norm": 0.0032806396484375,
      "learning_rate": 0.021939892990449395,
      "loss": 0.2319,
      "num_input_tokens_seen": 15814384,
      "step": 74940
    },
    {
      "epoch": 8.244774477447745,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.021938616310541444,
      "loss": 0.2319,
      "num_input_tokens_seen": 15815440,
      "step": 74945
    },
    {
      "epoch": 8.245324532453246,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.021937339566683603,
      "loss": 0.2293,
      "num_input_tokens_seen": 15816496,
      "step": 74950
    },
    {
      "epoch": 8.245874587458745,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.021936062758887622,
      "loss": 0.2335,
      "num_input_tokens_seen": 15817552,
      "step": 74955
    },
    {
      "epoch": 8.246424642464246,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.021934785887165287,
      "loss": 0.2346,
      "num_input_tokens_seen": 15818576,
      "step": 74960
    },
    {
      "epoch": 8.246974697469748,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.021933508951528357,
      "loss": 0.2309,
      "num_input_tokens_seen": 15819632,
      "step": 74965
    },
    {
      "epoch": 8.247524752475247,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.0219322319519886,
      "loss": 0.2346,
      "num_input_tokens_seen": 15820688,
      "step": 74970
    },
    {
      "epoch": 8.248074807480748,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.021930954888557785,
      "loss": 0.2309,
      "num_input_tokens_seen": 15821712,
      "step": 74975
    },
    {
      "epoch": 8.248624862486249,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.02192967776124769,
      "loss": 0.2304,
      "num_input_tokens_seen": 15822800,
      "step": 74980
    },
    {
      "epoch": 8.249174917491748,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.021928400570070075,
      "loss": 0.2294,
      "num_input_tokens_seen": 15823824,
      "step": 74985
    },
    {
      "epoch": 8.24972497249725,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.021927123315036726,
      "loss": 0.2289,
      "num_input_tokens_seen": 15824912,
      "step": 74990
    },
    {
      "epoch": 8.25027502750275,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.021925845996159397,
      "loss": 0.2315,
      "num_input_tokens_seen": 15825936,
      "step": 74995
    },
    {
      "epoch": 8.250825082508252,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.02192456861344987,
      "loss": 0.231,
      "num_input_tokens_seen": 15827024,
      "step": 75000
    },
    {
      "epoch": 8.251375137513751,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.021923291166919913,
      "loss": 0.2325,
      "num_input_tokens_seen": 15828048,
      "step": 75005
    },
    {
      "epoch": 8.251925192519252,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.021922013656581307,
      "loss": 0.2304,
      "num_input_tokens_seen": 15829136,
      "step": 75010
    },
    {
      "epoch": 8.252475247524753,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.021920736082445824,
      "loss": 0.2325,
      "num_input_tokens_seen": 15830224,
      "step": 75015
    },
    {
      "epoch": 8.253025302530252,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02191945844452524,
      "loss": 0.2346,
      "num_input_tokens_seen": 15831248,
      "step": 75020
    },
    {
      "epoch": 8.253575357535754,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.021918180742831327,
      "loss": 0.2294,
      "num_input_tokens_seen": 15832336,
      "step": 75025
    },
    {
      "epoch": 8.254125412541255,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.021916902977375856,
      "loss": 0.231,
      "num_input_tokens_seen": 15833392,
      "step": 75030
    },
    {
      "epoch": 8.254675467546754,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02191562514817061,
      "loss": 0.2336,
      "num_input_tokens_seen": 15834448,
      "step": 75035
    },
    {
      "epoch": 8.255225522552255,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02191434725522737,
      "loss": 0.2309,
      "num_input_tokens_seen": 15835536,
      "step": 75040
    },
    {
      "epoch": 8.255775577557756,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.021913069298557908,
      "loss": 0.2299,
      "num_input_tokens_seen": 15836656,
      "step": 75045
    },
    {
      "epoch": 8.256325632563255,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.021911791278174,
      "loss": 0.2367,
      "num_input_tokens_seen": 15837712,
      "step": 75050
    },
    {
      "epoch": 8.256875687568757,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.02191051319408743,
      "loss": 0.2309,
      "num_input_tokens_seen": 15838800,
      "step": 75055
    },
    {
      "epoch": 8.257425742574258,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.02190923504630998,
      "loss": 0.2304,
      "num_input_tokens_seen": 15839856,
      "step": 75060
    },
    {
      "epoch": 8.257975797579759,
      "grad_norm": 0.009765625,
      "learning_rate": 0.021907956834853425,
      "loss": 0.2257,
      "num_input_tokens_seen": 15840976,
      "step": 75065
    },
    {
      "epoch": 8.258525852585258,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.021906678559729543,
      "loss": 0.2314,
      "num_input_tokens_seen": 15842032,
      "step": 75070
    },
    {
      "epoch": 8.25907590759076,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.021905400220950118,
      "loss": 0.231,
      "num_input_tokens_seen": 15843120,
      "step": 75075
    },
    {
      "epoch": 8.25962596259626,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.021904121818526936,
      "loss": 0.2294,
      "num_input_tokens_seen": 15844112,
      "step": 75080
    },
    {
      "epoch": 8.26017601760176,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.021902843352471776,
      "loss": 0.2346,
      "num_input_tokens_seen": 15845200,
      "step": 75085
    },
    {
      "epoch": 8.26072607260726,
      "grad_norm": 0.00396728515625,
      "learning_rate": 0.021901564822796423,
      "loss": 0.2331,
      "num_input_tokens_seen": 15846256,
      "step": 75090
    },
    {
      "epoch": 8.261276127612762,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.021900286229512657,
      "loss": 0.2325,
      "num_input_tokens_seen": 15847344,
      "step": 75095
    },
    {
      "epoch": 8.261826182618261,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.02189900757263226,
      "loss": 0.2325,
      "num_input_tokens_seen": 15848336,
      "step": 75100
    },
    {
      "epoch": 8.262376237623762,
      "grad_norm": 0.00408935546875,
      "learning_rate": 0.021897728852167023,
      "loss": 0.2309,
      "num_input_tokens_seen": 15849360,
      "step": 75105
    },
    {
      "epoch": 8.262926292629263,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.02189645006812873,
      "loss": 0.2341,
      "num_input_tokens_seen": 15850448,
      "step": 75110
    },
    {
      "epoch": 8.263476347634764,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.021895171220529166,
      "loss": 0.232,
      "num_input_tokens_seen": 15851472,
      "step": 75115
    },
    {
      "epoch": 8.264026402640264,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 0.021893892309380117,
      "loss": 0.2309,
      "num_input_tokens_seen": 15852496,
      "step": 75120
    },
    {
      "epoch": 8.264576457645765,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.021892613334693372,
      "loss": 0.2293,
      "num_input_tokens_seen": 15853584,
      "step": 75125
    },
    {
      "epoch": 8.265126512651266,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.021891334296480713,
      "loss": 0.2314,
      "num_input_tokens_seen": 15854672,
      "step": 75130
    },
    {
      "epoch": 8.265676567656765,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02189005519475394,
      "loss": 0.2319,
      "num_input_tokens_seen": 15855760,
      "step": 75135
    },
    {
      "epoch": 8.266226622662266,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02188877602952483,
      "loss": 0.233,
      "num_input_tokens_seen": 15856752,
      "step": 75140
    },
    {
      "epoch": 8.266776677667767,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.021887496800805175,
      "loss": 0.2314,
      "num_input_tokens_seen": 15857840,
      "step": 75145
    },
    {
      "epoch": 8.267326732673267,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.021886217508606767,
      "loss": 0.2298,
      "num_input_tokens_seen": 15858960,
      "step": 75150
    },
    {
      "epoch": 8.267876787678768,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0218849381529414,
      "loss": 0.233,
      "num_input_tokens_seen": 15860048,
      "step": 75155
    },
    {
      "epoch": 8.268426842684269,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.021883658733820862,
      "loss": 0.2324,
      "num_input_tokens_seen": 15861104,
      "step": 75160
    },
    {
      "epoch": 8.268976897689768,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.02188237925125694,
      "loss": 0.2298,
      "num_input_tokens_seen": 15862192,
      "step": 75165
    },
    {
      "epoch": 8.26952695269527,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.021881099705261434,
      "loss": 0.2314,
      "num_input_tokens_seen": 15863216,
      "step": 75170
    },
    {
      "epoch": 8.27007700770077,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.021879820095846134,
      "loss": 0.2293,
      "num_input_tokens_seen": 15864240,
      "step": 75175
    },
    {
      "epoch": 8.270627062706271,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.021878540423022834,
      "loss": 0.2309,
      "num_input_tokens_seen": 15865328,
      "step": 75180
    },
    {
      "epoch": 8.27117711771177,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.021877260686803325,
      "loss": 0.2309,
      "num_input_tokens_seen": 15866416,
      "step": 75185
    },
    {
      "epoch": 8.271727172717272,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.021875980887199407,
      "loss": 0.2304,
      "num_input_tokens_seen": 15867536,
      "step": 75190
    },
    {
      "epoch": 8.272277227722773,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.021874701024222874,
      "loss": 0.2325,
      "num_input_tokens_seen": 15868592,
      "step": 75195
    },
    {
      "epoch": 8.272827282728272,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.021873421097885516,
      "loss": 0.2293,
      "num_input_tokens_seen": 15869712,
      "step": 75200
    },
    {
      "epoch": 8.273377337733773,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.021872141108199135,
      "loss": 0.2304,
      "num_input_tokens_seen": 15870768,
      "step": 75205
    },
    {
      "epoch": 8.273927392739274,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.02187086105517553,
      "loss": 0.2294,
      "num_input_tokens_seen": 15871856,
      "step": 75210
    },
    {
      "epoch": 8.274477447744774,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.02186958093882649,
      "loss": 0.2305,
      "num_input_tokens_seen": 15872976,
      "step": 75215
    },
    {
      "epoch": 8.275027502750275,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.021868300759163822,
      "loss": 0.2304,
      "num_input_tokens_seen": 15874000,
      "step": 75220
    },
    {
      "epoch": 8.275577557755776,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02186702051619932,
      "loss": 0.2331,
      "num_input_tokens_seen": 15875024,
      "step": 75225
    },
    {
      "epoch": 8.276127612761275,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.02186574020994479,
      "loss": 0.2294,
      "num_input_tokens_seen": 15876144,
      "step": 75230
    },
    {
      "epoch": 8.276677667766776,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.021864459840412024,
      "loss": 0.2315,
      "num_input_tokens_seen": 15877232,
      "step": 75235
    },
    {
      "epoch": 8.277227722772277,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.021863179407612825,
      "loss": 0.2336,
      "num_input_tokens_seen": 15878256,
      "step": 75240
    },
    {
      "epoch": 8.277777777777779,
      "grad_norm": 0.003326416015625,
      "learning_rate": 0.021861898911558993,
      "loss": 0.2299,
      "num_input_tokens_seen": 15879344,
      "step": 75245
    },
    {
      "epoch": 8.278327832783278,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.021860618352262332,
      "loss": 0.232,
      "num_input_tokens_seen": 15880368,
      "step": 75250
    },
    {
      "epoch": 8.278877887788779,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.021859337729734647,
      "loss": 0.2258,
      "num_input_tokens_seen": 15881456,
      "step": 75255
    },
    {
      "epoch": 8.27942794279428,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.021858057043987737,
      "loss": 0.2279,
      "num_input_tokens_seen": 15882512,
      "step": 75260
    },
    {
      "epoch": 8.27997799779978,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.021856776295033405,
      "loss": 0.2322,
      "num_input_tokens_seen": 15883600,
      "step": 75265
    },
    {
      "epoch": 8.28052805280528,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.021855495482883455,
      "loss": 0.2285,
      "num_input_tokens_seen": 15884688,
      "step": 75270
    },
    {
      "epoch": 8.281078107810782,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.021854214607549695,
      "loss": 0.2322,
      "num_input_tokens_seen": 15885744,
      "step": 75275
    },
    {
      "epoch": 8.281628162816281,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.021852933669043926,
      "loss": 0.2306,
      "num_input_tokens_seen": 15886832,
      "step": 75280
    },
    {
      "epoch": 8.282178217821782,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.021851652667377958,
      "loss": 0.2281,
      "num_input_tokens_seen": 15887888,
      "step": 75285
    },
    {
      "epoch": 8.282728272827283,
      "grad_norm": 0.0028533935546875,
      "learning_rate": 0.02185037160256359,
      "loss": 0.2386,
      "num_input_tokens_seen": 15888944,
      "step": 75290
    },
    {
      "epoch": 8.283278327832782,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.02184909047461264,
      "loss": 0.2353,
      "num_input_tokens_seen": 15890064,
      "step": 75295
    },
    {
      "epoch": 8.283828382838283,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.021847809283536905,
      "loss": 0.2302,
      "num_input_tokens_seen": 15891120,
      "step": 75300
    },
    {
      "epoch": 8.284378437843785,
      "grad_norm": 0.003936767578125,
      "learning_rate": 0.0218465280293482,
      "loss": 0.2349,
      "num_input_tokens_seen": 15892144,
      "step": 75305
    },
    {
      "epoch": 8.284928492849286,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.021845246712058335,
      "loss": 0.2281,
      "num_input_tokens_seen": 15893232,
      "step": 75310
    },
    {
      "epoch": 8.285478547854785,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02184396533167911,
      "loss": 0.2316,
      "num_input_tokens_seen": 15894288,
      "step": 75315
    },
    {
      "epoch": 8.286028602860286,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.021842683888222338,
      "loss": 0.2296,
      "num_input_tokens_seen": 15895312,
      "step": 75320
    },
    {
      "epoch": 8.286578657865787,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.02184140238169984,
      "loss": 0.2327,
      "num_input_tokens_seen": 15896336,
      "step": 75325
    },
    {
      "epoch": 8.287128712871286,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.021840120812123415,
      "loss": 0.2347,
      "num_input_tokens_seen": 15897424,
      "step": 75330
    },
    {
      "epoch": 8.287678767876788,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02183883917950488,
      "loss": 0.228,
      "num_input_tokens_seen": 15898448,
      "step": 75335
    },
    {
      "epoch": 8.288228822882289,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.021837557483856043,
      "loss": 0.2321,
      "num_input_tokens_seen": 15899536,
      "step": 75340
    },
    {
      "epoch": 8.288778877887788,
      "grad_norm": 0.003021240234375,
      "learning_rate": 0.021836275725188724,
      "loss": 0.2326,
      "num_input_tokens_seen": 15900592,
      "step": 75345
    },
    {
      "epoch": 8.289328932893289,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02183499390351473,
      "loss": 0.2321,
      "num_input_tokens_seen": 15901680,
      "step": 75350
    },
    {
      "epoch": 8.28987898789879,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.021833712018845874,
      "loss": 0.2332,
      "num_input_tokens_seen": 15902800,
      "step": 75355
    },
    {
      "epoch": 8.290429042904291,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.021832430071193978,
      "loss": 0.2304,
      "num_input_tokens_seen": 15903824,
      "step": 75360
    },
    {
      "epoch": 8.29097909790979,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02183114806057085,
      "loss": 0.2316,
      "num_input_tokens_seen": 15904880,
      "step": 75365
    },
    {
      "epoch": 8.291529152915292,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.02182986598698831,
      "loss": 0.2295,
      "num_input_tokens_seen": 15905904,
      "step": 75370
    },
    {
      "epoch": 8.292079207920793,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.021828583850458175,
      "loss": 0.2316,
      "num_input_tokens_seen": 15906928,
      "step": 75375
    },
    {
      "epoch": 8.292629262926292,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.02182730165099225,
      "loss": 0.2279,
      "num_input_tokens_seen": 15907952,
      "step": 75380
    },
    {
      "epoch": 8.293179317931793,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.02182601938860237,
      "loss": 0.2336,
      "num_input_tokens_seen": 15909040,
      "step": 75385
    },
    {
      "epoch": 8.293729372937294,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.02182473706330034,
      "loss": 0.2347,
      "num_input_tokens_seen": 15910064,
      "step": 75390
    },
    {
      "epoch": 8.294279427942794,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02182345467509799,
      "loss": 0.2289,
      "num_input_tokens_seen": 15911152,
      "step": 75395
    },
    {
      "epoch": 8.294829482948295,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.02182217222400713,
      "loss": 0.2315,
      "num_input_tokens_seen": 15912208,
      "step": 75400
    },
    {
      "epoch": 8.295379537953796,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.021820889710039575,
      "loss": 0.2347,
      "num_input_tokens_seen": 15913264,
      "step": 75405
    },
    {
      "epoch": 8.295929592959295,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.02181960713320716,
      "loss": 0.2315,
      "num_input_tokens_seen": 15914320,
      "step": 75410
    },
    {
      "epoch": 8.296479647964796,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.021818324493521692,
      "loss": 0.2311,
      "num_input_tokens_seen": 15915312,
      "step": 75415
    },
    {
      "epoch": 8.297029702970297,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.021817041790995004,
      "loss": 0.23,
      "num_input_tokens_seen": 15916400,
      "step": 75420
    },
    {
      "epoch": 8.297579757975798,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.021815759025638915,
      "loss": 0.2326,
      "num_input_tokens_seen": 15917456,
      "step": 75425
    },
    {
      "epoch": 8.298129812981298,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02181447619746524,
      "loss": 0.2314,
      "num_input_tokens_seen": 15918544,
      "step": 75430
    },
    {
      "epoch": 8.298679867986799,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.021813193306485814,
      "loss": 0.231,
      "num_input_tokens_seen": 15919568,
      "step": 75435
    },
    {
      "epoch": 8.2992299229923,
      "grad_norm": 0.000507354736328125,
      "learning_rate": 0.02181191035271245,
      "loss": 0.2336,
      "num_input_tokens_seen": 15920560,
      "step": 75440
    },
    {
      "epoch": 8.2997799779978,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.02181062733615698,
      "loss": 0.2305,
      "num_input_tokens_seen": 15921616,
      "step": 75445
    },
    {
      "epoch": 8.3003300330033,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.02180934425683122,
      "loss": 0.2316,
      "num_input_tokens_seen": 15922736,
      "step": 75450
    },
    {
      "epoch": 8.300880088008801,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.021808061114747004,
      "loss": 0.231,
      "num_input_tokens_seen": 15923792,
      "step": 75455
    },
    {
      "epoch": 8.3014301430143,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.02180677790991616,
      "loss": 0.2352,
      "num_input_tokens_seen": 15924848,
      "step": 75460
    },
    {
      "epoch": 8.301980198019802,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.021805494642350508,
      "loss": 0.2342,
      "num_input_tokens_seen": 15925904,
      "step": 75465
    },
    {
      "epoch": 8.302530253025303,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.021804211312061877,
      "loss": 0.2336,
      "num_input_tokens_seen": 15926992,
      "step": 75470
    },
    {
      "epoch": 8.303080308030804,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.021802927919062094,
      "loss": 0.2336,
      "num_input_tokens_seen": 15928048,
      "step": 75475
    },
    {
      "epoch": 8.303630363036303,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.02180164446336299,
      "loss": 0.2295,
      "num_input_tokens_seen": 15929104,
      "step": 75480
    },
    {
      "epoch": 8.304180418041804,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.02180036094497639,
      "loss": 0.2351,
      "num_input_tokens_seen": 15930160,
      "step": 75485
    },
    {
      "epoch": 8.304730473047305,
      "grad_norm": 0.003326416015625,
      "learning_rate": 0.02179907736391413,
      "loss": 0.2263,
      "num_input_tokens_seen": 15931184,
      "step": 75490
    },
    {
      "epoch": 8.305280528052805,
      "grad_norm": 0.00390625,
      "learning_rate": 0.021797793720188034,
      "loss": 0.2335,
      "num_input_tokens_seen": 15932240,
      "step": 75495
    },
    {
      "epoch": 8.305830583058306,
      "grad_norm": 0.009765625,
      "learning_rate": 0.021796510013809937,
      "loss": 0.2309,
      "num_input_tokens_seen": 15933328,
      "step": 75500
    },
    {
      "epoch": 8.306380638063807,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.021795226244791667,
      "loss": 0.2325,
      "num_input_tokens_seen": 15934384,
      "step": 75505
    },
    {
      "epoch": 8.306930693069306,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02179394241314506,
      "loss": 0.2335,
      "num_input_tokens_seen": 15935504,
      "step": 75510
    },
    {
      "epoch": 8.307480748074807,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.021792658518881944,
      "loss": 0.2293,
      "num_input_tokens_seen": 15936528,
      "step": 75515
    },
    {
      "epoch": 8.308030803080309,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.021791374562014153,
      "loss": 0.2345,
      "num_input_tokens_seen": 15937552,
      "step": 75520
    },
    {
      "epoch": 8.308580858085808,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02179009054255352,
      "loss": 0.2308,
      "num_input_tokens_seen": 15938640,
      "step": 75525
    },
    {
      "epoch": 8.309130913091309,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.021788806460511886,
      "loss": 0.2309,
      "num_input_tokens_seen": 15939600,
      "step": 75530
    },
    {
      "epoch": 8.30968096809681,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.02178752231590108,
      "loss": 0.235,
      "num_input_tokens_seen": 15940656,
      "step": 75535
    },
    {
      "epoch": 8.310231023102311,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.021786238108732932,
      "loss": 0.2319,
      "num_input_tokens_seen": 15941744,
      "step": 75540
    },
    {
      "epoch": 8.31078107810781,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.021784953839019287,
      "loss": 0.2324,
      "num_input_tokens_seen": 15942864,
      "step": 75545
    },
    {
      "epoch": 8.311331133113312,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.021783669506771977,
      "loss": 0.2308,
      "num_input_tokens_seen": 15943856,
      "step": 75550
    },
    {
      "epoch": 8.311881188118813,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.02178238511200284,
      "loss": 0.2298,
      "num_input_tokens_seen": 15944912,
      "step": 75555
    },
    {
      "epoch": 8.312431243124312,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02178110065472372,
      "loss": 0.2314,
      "num_input_tokens_seen": 15946000,
      "step": 75560
    },
    {
      "epoch": 8.312981298129813,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021779816134946443,
      "loss": 0.2298,
      "num_input_tokens_seen": 15947056,
      "step": 75565
    },
    {
      "epoch": 8.313531353135314,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.021778531552682855,
      "loss": 0.2319,
      "num_input_tokens_seen": 15948048,
      "step": 75570
    },
    {
      "epoch": 8.314081408140813,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.021777246907944796,
      "loss": 0.2319,
      "num_input_tokens_seen": 15949104,
      "step": 75575
    },
    {
      "epoch": 8.314631463146315,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0217759622007441,
      "loss": 0.2303,
      "num_input_tokens_seen": 15950160,
      "step": 75580
    },
    {
      "epoch": 8.315181518151816,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02177467743109262,
      "loss": 0.2335,
      "num_input_tokens_seen": 15951216,
      "step": 75585
    },
    {
      "epoch": 8.315731573157315,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02177339259900218,
      "loss": 0.2303,
      "num_input_tokens_seen": 15952336,
      "step": 75590
    },
    {
      "epoch": 8.316281628162816,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.021772107704484633,
      "loss": 0.2329,
      "num_input_tokens_seen": 15953360,
      "step": 75595
    },
    {
      "epoch": 8.316831683168317,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02177082274755182,
      "loss": 0.2313,
      "num_input_tokens_seen": 15954384,
      "step": 75600
    },
    {
      "epoch": 8.317381738173818,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.021769537728215584,
      "loss": 0.2308,
      "num_input_tokens_seen": 15955408,
      "step": 75605
    },
    {
      "epoch": 8.317931793179318,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.021768252646487768,
      "loss": 0.2308,
      "num_input_tokens_seen": 15956432,
      "step": 75610
    },
    {
      "epoch": 8.318481848184819,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.021766967502380212,
      "loss": 0.2334,
      "num_input_tokens_seen": 15957488,
      "step": 75615
    },
    {
      "epoch": 8.31903190319032,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.021765682295904764,
      "loss": 0.235,
      "num_input_tokens_seen": 15958544,
      "step": 75620
    },
    {
      "epoch": 8.319581958195819,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02176439702707327,
      "loss": 0.2319,
      "num_input_tokens_seen": 15959632,
      "step": 75625
    },
    {
      "epoch": 8.32013201320132,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.02176311169589757,
      "loss": 0.2318,
      "num_input_tokens_seen": 15960720,
      "step": 75630
    },
    {
      "epoch": 8.320682068206821,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02176182630238952,
      "loss": 0.2313,
      "num_input_tokens_seen": 15961744,
      "step": 75635
    },
    {
      "epoch": 8.32123212321232,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.021760540846560962,
      "loss": 0.2313,
      "num_input_tokens_seen": 15962832,
      "step": 75640
    },
    {
      "epoch": 8.321782178217822,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02175925532842374,
      "loss": 0.2324,
      "num_input_tokens_seen": 15963888,
      "step": 75645
    },
    {
      "epoch": 8.322332233223323,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.021757969747989707,
      "loss": 0.2319,
      "num_input_tokens_seen": 15964944,
      "step": 75650
    },
    {
      "epoch": 8.322882288228822,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.021756684105270714,
      "loss": 0.2303,
      "num_input_tokens_seen": 15965968,
      "step": 75655
    },
    {
      "epoch": 8.323432343234323,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.021755398400278597,
      "loss": 0.2319,
      "num_input_tokens_seen": 15967120,
      "step": 75660
    },
    {
      "epoch": 8.323982398239824,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.021754112633025218,
      "loss": 0.2303,
      "num_input_tokens_seen": 15968176,
      "step": 75665
    },
    {
      "epoch": 8.324532453245325,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02175282680352242,
      "loss": 0.2324,
      "num_input_tokens_seen": 15969232,
      "step": 75670
    },
    {
      "epoch": 8.325082508250825,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.021751540911782066,
      "loss": 0.2308,
      "num_input_tokens_seen": 15970224,
      "step": 75675
    },
    {
      "epoch": 8.325632563256326,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.021750254957815993,
      "loss": 0.2308,
      "num_input_tokens_seen": 15971216,
      "step": 75680
    },
    {
      "epoch": 8.326182618261827,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.021748968941636065,
      "loss": 0.2319,
      "num_input_tokens_seen": 15972240,
      "step": 75685
    },
    {
      "epoch": 8.326732673267326,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.021747682863254125,
      "loss": 0.2324,
      "num_input_tokens_seen": 15973296,
      "step": 75690
    },
    {
      "epoch": 8.327282728272827,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.021746396722682027,
      "loss": 0.2313,
      "num_input_tokens_seen": 15974352,
      "step": 75695
    },
    {
      "epoch": 8.327832783278328,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02174511051993163,
      "loss": 0.2288,
      "num_input_tokens_seen": 15975376,
      "step": 75700
    },
    {
      "epoch": 8.328382838283828,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.021743824255014792,
      "loss": 0.233,
      "num_input_tokens_seen": 15976368,
      "step": 75705
    },
    {
      "epoch": 8.328932893289329,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.021742537927943353,
      "loss": 0.2303,
      "num_input_tokens_seen": 15977488,
      "step": 75710
    },
    {
      "epoch": 8.32948294829483,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02174125153872918,
      "loss": 0.2314,
      "num_input_tokens_seen": 15978544,
      "step": 75715
    },
    {
      "epoch": 8.33003300330033,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.021739965087384124,
      "loss": 0.234,
      "num_input_tokens_seen": 15979632,
      "step": 75720
    },
    {
      "epoch": 8.33058305830583,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.021738678573920044,
      "loss": 0.233,
      "num_input_tokens_seen": 15980624,
      "step": 75725
    },
    {
      "epoch": 8.331133113311331,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.021737391998348805,
      "loss": 0.2304,
      "num_input_tokens_seen": 15981648,
      "step": 75730
    },
    {
      "epoch": 8.331683168316832,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.021736105360682254,
      "loss": 0.2298,
      "num_input_tokens_seen": 15982672,
      "step": 75735
    },
    {
      "epoch": 8.332233223322332,
      "grad_norm": 0.0028839111328125,
      "learning_rate": 0.021734818660932246,
      "loss": 0.2309,
      "num_input_tokens_seen": 15983728,
      "step": 75740
    },
    {
      "epoch": 8.332783278327833,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02173353189911065,
      "loss": 0.2303,
      "num_input_tokens_seen": 15984752,
      "step": 75745
    },
    {
      "epoch": 8.333333333333334,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.021732245075229323,
      "loss": 0.2319,
      "num_input_tokens_seen": 15985840,
      "step": 75750
    },
    {
      "epoch": 8.333883388338833,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.021730958189300122,
      "loss": 0.2324,
      "num_input_tokens_seen": 15986928,
      "step": 75755
    },
    {
      "epoch": 8.334433443344334,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02172967124133491,
      "loss": 0.2335,
      "num_input_tokens_seen": 15987984,
      "step": 75760
    },
    {
      "epoch": 8.334983498349835,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.021728384231345543,
      "loss": 0.2314,
      "num_input_tokens_seen": 15989040,
      "step": 75765
    },
    {
      "epoch": 8.335533553355335,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.021727097159343893,
      "loss": 0.2293,
      "num_input_tokens_seen": 15990160,
      "step": 75770
    },
    {
      "epoch": 8.336083608360836,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.021725810025341817,
      "loss": 0.2309,
      "num_input_tokens_seen": 15991216,
      "step": 75775
    },
    {
      "epoch": 8.336633663366337,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.021724522829351174,
      "loss": 0.2324,
      "num_input_tokens_seen": 15992272,
      "step": 75780
    },
    {
      "epoch": 8.337183718371838,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.02172323557138383,
      "loss": 0.2303,
      "num_input_tokens_seen": 15993360,
      "step": 75785
    },
    {
      "epoch": 8.337733773377337,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.021721948251451653,
      "loss": 0.2304,
      "num_input_tokens_seen": 15994352,
      "step": 75790
    },
    {
      "epoch": 8.338283828382838,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.021720660869566498,
      "loss": 0.2303,
      "num_input_tokens_seen": 15995376,
      "step": 75795
    },
    {
      "epoch": 8.33883388338834,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.021719373425740245,
      "loss": 0.2325,
      "num_input_tokens_seen": 15996336,
      "step": 75800
    },
    {
      "epoch": 8.339383938393839,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02171808591998475,
      "loss": 0.2309,
      "num_input_tokens_seen": 15997456,
      "step": 75805
    },
    {
      "epoch": 8.33993399339934,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.021716798352311877,
      "loss": 0.2314,
      "num_input_tokens_seen": 15998544,
      "step": 75810
    },
    {
      "epoch": 8.340484048404841,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0217155107227335,
      "loss": 0.2303,
      "num_input_tokens_seen": 15999664,
      "step": 75815
    },
    {
      "epoch": 8.34103410341034,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.021714223031261477,
      "loss": 0.233,
      "num_input_tokens_seen": 16000688,
      "step": 75820
    },
    {
      "epoch": 8.341584158415841,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.021712935277907686,
      "loss": 0.2293,
      "num_input_tokens_seen": 16001712,
      "step": 75825
    },
    {
      "epoch": 8.342134213421343,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.021711647462683992,
      "loss": 0.2314,
      "num_input_tokens_seen": 16002832,
      "step": 75830
    },
    {
      "epoch": 8.342684268426842,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.021710359585602267,
      "loss": 0.2324,
      "num_input_tokens_seen": 16003920,
      "step": 75835
    },
    {
      "epoch": 8.343234323432343,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02170907164667437,
      "loss": 0.2308,
      "num_input_tokens_seen": 16004944,
      "step": 75840
    },
    {
      "epoch": 8.343784378437844,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.021707783645912186,
      "loss": 0.233,
      "num_input_tokens_seen": 16005968,
      "step": 75845
    },
    {
      "epoch": 8.344334433443345,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.02170649558332758,
      "loss": 0.2308,
      "num_input_tokens_seen": 16007088,
      "step": 75850
    },
    {
      "epoch": 8.344884488448844,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.021705207458932414,
      "loss": 0.2319,
      "num_input_tokens_seen": 16008176,
      "step": 75855
    },
    {
      "epoch": 8.345434543454346,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.021703919272738573,
      "loss": 0.233,
      "num_input_tokens_seen": 16009168,
      "step": 75860
    },
    {
      "epoch": 8.345984598459847,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.02170263102475793,
      "loss": 0.2293,
      "num_input_tokens_seen": 16010160,
      "step": 75865
    },
    {
      "epoch": 8.346534653465346,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.021701342715002344,
      "loss": 0.2314,
      "num_input_tokens_seen": 16011216,
      "step": 75870
    },
    {
      "epoch": 8.347084708470847,
      "grad_norm": 0.0004062652587890625,
      "learning_rate": 0.021700054343483703,
      "loss": 0.2324,
      "num_input_tokens_seen": 16012272,
      "step": 75875
    },
    {
      "epoch": 8.347634763476348,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.021698765910213875,
      "loss": 0.2329,
      "num_input_tokens_seen": 16013360,
      "step": 75880
    },
    {
      "epoch": 8.348184818481847,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.021697477415204735,
      "loss": 0.2319,
      "num_input_tokens_seen": 16014448,
      "step": 75885
    },
    {
      "epoch": 8.348734873487349,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02169618885846816,
      "loss": 0.2308,
      "num_input_tokens_seen": 16015472,
      "step": 75890
    },
    {
      "epoch": 8.34928492849285,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021694900240016027,
      "loss": 0.2334,
      "num_input_tokens_seen": 16016528,
      "step": 75895
    },
    {
      "epoch": 8.34983498349835,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02169361155986021,
      "loss": 0.2298,
      "num_input_tokens_seen": 16017520,
      "step": 75900
    },
    {
      "epoch": 8.35038503850385,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021692322818012585,
      "loss": 0.2313,
      "num_input_tokens_seen": 16018608,
      "step": 75905
    },
    {
      "epoch": 8.350935093509351,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.021691034014485036,
      "loss": 0.2293,
      "num_input_tokens_seen": 16019632,
      "step": 75910
    },
    {
      "epoch": 8.351485148514852,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.021689745149289435,
      "loss": 0.2319,
      "num_input_tokens_seen": 16020720,
      "step": 75915
    },
    {
      "epoch": 8.352035203520352,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021688456222437664,
      "loss": 0.2298,
      "num_input_tokens_seen": 16021808,
      "step": 75920
    },
    {
      "epoch": 8.352585258525853,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.0216871672339416,
      "loss": 0.2303,
      "num_input_tokens_seen": 16022864,
      "step": 75925
    },
    {
      "epoch": 8.353135313531354,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.021685878183813123,
      "loss": 0.234,
      "num_input_tokens_seen": 16024048,
      "step": 75930
    },
    {
      "epoch": 8.353685368536853,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.021684589072064116,
      "loss": 0.2329,
      "num_input_tokens_seen": 16025072,
      "step": 75935
    },
    {
      "epoch": 8.354235423542354,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.02168329989870646,
      "loss": 0.2324,
      "num_input_tokens_seen": 16026128,
      "step": 75940
    },
    {
      "epoch": 8.354785478547855,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.021682010663752038,
      "loss": 0.2308,
      "num_input_tokens_seen": 16027216,
      "step": 75945
    },
    {
      "epoch": 8.355335533553355,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.021680721367212726,
      "loss": 0.2314,
      "num_input_tokens_seen": 16028304,
      "step": 75950
    },
    {
      "epoch": 8.355885588558856,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.021679432009100417,
      "loss": 0.2303,
      "num_input_tokens_seen": 16029328,
      "step": 75955
    },
    {
      "epoch": 8.356435643564357,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.02167814258942698,
      "loss": 0.2324,
      "num_input_tokens_seen": 16030352,
      "step": 75960
    },
    {
      "epoch": 8.356985698569858,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.021676853108204316,
      "loss": 0.2329,
      "num_input_tokens_seen": 16031408,
      "step": 75965
    },
    {
      "epoch": 8.357535753575357,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.02167556356544429,
      "loss": 0.2303,
      "num_input_tokens_seen": 16032432,
      "step": 75970
    },
    {
      "epoch": 8.358085808580858,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.021674273961158806,
      "loss": 0.2293,
      "num_input_tokens_seen": 16033488,
      "step": 75975
    },
    {
      "epoch": 8.35863586358636,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02167298429535974,
      "loss": 0.2313,
      "num_input_tokens_seen": 16034544,
      "step": 75980
    },
    {
      "epoch": 8.359185918591859,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.021671694568058977,
      "loss": 0.2324,
      "num_input_tokens_seen": 16035504,
      "step": 75985
    },
    {
      "epoch": 8.35973597359736,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.02167040477926841,
      "loss": 0.2288,
      "num_input_tokens_seen": 16036592,
      "step": 75990
    },
    {
      "epoch": 8.36028602860286,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.021669114928999925,
      "loss": 0.2298,
      "num_input_tokens_seen": 16037584,
      "step": 75995
    },
    {
      "epoch": 8.36083608360836,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021667825017265403,
      "loss": 0.2319,
      "num_input_tokens_seen": 16038672,
      "step": 76000
    },
    {
      "epoch": 8.361386138613861,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.021666535044076736,
      "loss": 0.2298,
      "num_input_tokens_seen": 16039696,
      "step": 76005
    },
    {
      "epoch": 8.361936193619362,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.021665245009445815,
      "loss": 0.2288,
      "num_input_tokens_seen": 16040688,
      "step": 76010
    },
    {
      "epoch": 8.362486248624862,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.021663954913384537,
      "loss": 0.2299,
      "num_input_tokens_seen": 16041712,
      "step": 76015
    },
    {
      "epoch": 8.363036303630363,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.021662664755904778,
      "loss": 0.2319,
      "num_input_tokens_seen": 16042704,
      "step": 76020
    },
    {
      "epoch": 8.363586358635864,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.021661374537018434,
      "loss": 0.2319,
      "num_input_tokens_seen": 16043792,
      "step": 76025
    },
    {
      "epoch": 8.364136413641365,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.0216600842567374,
      "loss": 0.2288,
      "num_input_tokens_seen": 16044848,
      "step": 76030
    },
    {
      "epoch": 8.364686468646864,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.02165879391507356,
      "loss": 0.2367,
      "num_input_tokens_seen": 16045904,
      "step": 76035
    },
    {
      "epoch": 8.365236523652365,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.021657503512038823,
      "loss": 0.233,
      "num_input_tokens_seen": 16046992,
      "step": 76040
    },
    {
      "epoch": 8.365786578657866,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.02165621304764506,
      "loss": 0.2335,
      "num_input_tokens_seen": 16048016,
      "step": 76045
    },
    {
      "epoch": 8.366336633663366,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.021654922521904176,
      "loss": 0.2319,
      "num_input_tokens_seen": 16049104,
      "step": 76050
    },
    {
      "epoch": 8.366886688668867,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.02165363193482807,
      "loss": 0.2304,
      "num_input_tokens_seen": 16050160,
      "step": 76055
    },
    {
      "epoch": 8.367436743674368,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.021652341286428626,
      "loss": 0.2299,
      "num_input_tokens_seen": 16051152,
      "step": 76060
    },
    {
      "epoch": 8.367986798679867,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02165105057671775,
      "loss": 0.2309,
      "num_input_tokens_seen": 16052272,
      "step": 76065
    },
    {
      "epoch": 8.368536853685368,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.021649759805707332,
      "loss": 0.2303,
      "num_input_tokens_seen": 16053264,
      "step": 76070
    },
    {
      "epoch": 8.36908690869087,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02164846897340926,
      "loss": 0.2324,
      "num_input_tokens_seen": 16054288,
      "step": 76075
    },
    {
      "epoch": 8.369636963696369,
      "grad_norm": 0.0031890869140625,
      "learning_rate": 0.021647178079835447,
      "loss": 0.234,
      "num_input_tokens_seen": 16055408,
      "step": 76080
    },
    {
      "epoch": 8.37018701870187,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.021645887124997782,
      "loss": 0.2319,
      "num_input_tokens_seen": 16056464,
      "step": 76085
    },
    {
      "epoch": 8.370737073707371,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021644596108908166,
      "loss": 0.2319,
      "num_input_tokens_seen": 16057520,
      "step": 76090
    },
    {
      "epoch": 8.371287128712872,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.02164330503157849,
      "loss": 0.2303,
      "num_input_tokens_seen": 16058544,
      "step": 76095
    },
    {
      "epoch": 8.371837183718371,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.021642013893020666,
      "loss": 0.2303,
      "num_input_tokens_seen": 16059568,
      "step": 76100
    },
    {
      "epoch": 8.372387238723872,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.021640722693246583,
      "loss": 0.2319,
      "num_input_tokens_seen": 16060560,
      "step": 76105
    },
    {
      "epoch": 8.372937293729374,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.021639431432268152,
      "loss": 0.2303,
      "num_input_tokens_seen": 16061584,
      "step": 76110
    },
    {
      "epoch": 8.373487348734873,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.021638140110097258,
      "loss": 0.2303,
      "num_input_tokens_seen": 16062672,
      "step": 76115
    },
    {
      "epoch": 8.374037403740374,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.021636848726745816,
      "loss": 0.2319,
      "num_input_tokens_seen": 16063696,
      "step": 76120
    },
    {
      "epoch": 8.374587458745875,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.02163555728222573,
      "loss": 0.2319,
      "num_input_tokens_seen": 16064752,
      "step": 76125
    },
    {
      "epoch": 8.375137513751374,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.021634265776548887,
      "loss": 0.2282,
      "num_input_tokens_seen": 16065872,
      "step": 76130
    },
    {
      "epoch": 8.375687568756875,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.021632974209727204,
      "loss": 0.2319,
      "num_input_tokens_seen": 16066928,
      "step": 76135
    },
    {
      "epoch": 8.376237623762377,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02163168258177258,
      "loss": 0.2319,
      "num_input_tokens_seen": 16067952,
      "step": 76140
    },
    {
      "epoch": 8.376787678767876,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02163039089269692,
      "loss": 0.2308,
      "num_input_tokens_seen": 16068976,
      "step": 76145
    },
    {
      "epoch": 8.377337733773377,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02162909914251213,
      "loss": 0.234,
      "num_input_tokens_seen": 16070000,
      "step": 76150
    },
    {
      "epoch": 8.377887788778878,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02162780733123012,
      "loss": 0.2303,
      "num_input_tokens_seen": 16071056,
      "step": 76155
    },
    {
      "epoch": 8.37843784378438,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.021626515458862777,
      "loss": 0.2335,
      "num_input_tokens_seen": 16072144,
      "step": 76160
    },
    {
      "epoch": 8.378987898789878,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.021625223525422028,
      "loss": 0.2309,
      "num_input_tokens_seen": 16073264,
      "step": 76165
    },
    {
      "epoch": 8.37953795379538,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.02162393153091977,
      "loss": 0.233,
      "num_input_tokens_seen": 16074288,
      "step": 76170
    },
    {
      "epoch": 8.38008800880088,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.021622639475367914,
      "loss": 0.2324,
      "num_input_tokens_seen": 16075280,
      "step": 76175
    },
    {
      "epoch": 8.38063806380638,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.02162134735877837,
      "loss": 0.2304,
      "num_input_tokens_seen": 16076368,
      "step": 76180
    },
    {
      "epoch": 8.381188118811881,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.021620055181163046,
      "loss": 0.2319,
      "num_input_tokens_seen": 16077360,
      "step": 76185
    },
    {
      "epoch": 8.381738173817382,
      "grad_norm": 0.000736236572265625,
      "learning_rate": 0.021618762942533846,
      "loss": 0.2319,
      "num_input_tokens_seen": 16078352,
      "step": 76190
    },
    {
      "epoch": 8.382288228822881,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.021617470642902686,
      "loss": 0.2314,
      "num_input_tokens_seen": 16079376,
      "step": 76195
    },
    {
      "epoch": 8.382838283828383,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.021616178282281476,
      "loss": 0.2319,
      "num_input_tokens_seen": 16080400,
      "step": 76200
    },
    {
      "epoch": 8.383388338833884,
      "grad_norm": 0.00701904296875,
      "learning_rate": 0.02161488586068212,
      "loss": 0.2314,
      "num_input_tokens_seen": 16081424,
      "step": 76205
    },
    {
      "epoch": 8.383938393839385,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02161359337811654,
      "loss": 0.233,
      "num_input_tokens_seen": 16082416,
      "step": 76210
    },
    {
      "epoch": 8.384488448844884,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.021612300834596643,
      "loss": 0.2303,
      "num_input_tokens_seen": 16083408,
      "step": 76215
    },
    {
      "epoch": 8.385038503850385,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.021611008230134336,
      "loss": 0.2303,
      "num_input_tokens_seen": 16084464,
      "step": 76220
    },
    {
      "epoch": 8.385588558855886,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.021609715564741547,
      "loss": 0.233,
      "num_input_tokens_seen": 16085456,
      "step": 76225
    },
    {
      "epoch": 8.386138613861386,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02160842283843018,
      "loss": 0.2288,
      "num_input_tokens_seen": 16086576,
      "step": 76230
    },
    {
      "epoch": 8.386688668866887,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.021607130051212148,
      "loss": 0.2303,
      "num_input_tokens_seen": 16087664,
      "step": 76235
    },
    {
      "epoch": 8.387238723872388,
      "grad_norm": 0.0026397705078125,
      "learning_rate": 0.02160583720309937,
      "loss": 0.2293,
      "num_input_tokens_seen": 16088656,
      "step": 76240
    },
    {
      "epoch": 8.387788778877887,
      "grad_norm": 0.000762939453125,
      "learning_rate": 0.02160454429410376,
      "loss": 0.2314,
      "num_input_tokens_seen": 16089744,
      "step": 76245
    },
    {
      "epoch": 8.388338833883388,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02160325132423723,
      "loss": 0.2313,
      "num_input_tokens_seen": 16090864,
      "step": 76250
    },
    {
      "epoch": 8.38888888888889,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02160195829351171,
      "loss": 0.2314,
      "num_input_tokens_seen": 16091952,
      "step": 76255
    },
    {
      "epoch": 8.389438943894389,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.021600665201939102,
      "loss": 0.2309,
      "num_input_tokens_seen": 16093008,
      "step": 76260
    },
    {
      "epoch": 8.38998899889989,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.021599372049531333,
      "loss": 0.2319,
      "num_input_tokens_seen": 16094064,
      "step": 76265
    },
    {
      "epoch": 8.39053905390539,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.021598078836300313,
      "loss": 0.2313,
      "num_input_tokens_seen": 16095120,
      "step": 76270
    },
    {
      "epoch": 8.391089108910892,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.021596785562257973,
      "loss": 0.2293,
      "num_input_tokens_seen": 16096176,
      "step": 76275
    },
    {
      "epoch": 8.391639163916391,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.021595492227416226,
      "loss": 0.2309,
      "num_input_tokens_seen": 16097232,
      "step": 76280
    },
    {
      "epoch": 8.392189218921892,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02159419883178699,
      "loss": 0.233,
      "num_input_tokens_seen": 16098288,
      "step": 76285
    },
    {
      "epoch": 8.392739273927393,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.021592905375382186,
      "loss": 0.2329,
      "num_input_tokens_seen": 16099376,
      "step": 76290
    },
    {
      "epoch": 8.393289328932893,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.021591611858213733,
      "loss": 0.2319,
      "num_input_tokens_seen": 16100432,
      "step": 76295
    },
    {
      "epoch": 8.393839383938394,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.021590318280293567,
      "loss": 0.2309,
      "num_input_tokens_seen": 16101584,
      "step": 76300
    },
    {
      "epoch": 8.394389438943895,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02158902464163359,
      "loss": 0.233,
      "num_input_tokens_seen": 16102608,
      "step": 76305
    },
    {
      "epoch": 8.394939493949394,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.02158773094224574,
      "loss": 0.2319,
      "num_input_tokens_seen": 16103728,
      "step": 76310
    },
    {
      "epoch": 8.395489548954895,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.021586437182141933,
      "loss": 0.2298,
      "num_input_tokens_seen": 16104752,
      "step": 76315
    },
    {
      "epoch": 8.396039603960396,
      "grad_norm": 0.00372314453125,
      "learning_rate": 0.021585143361334096,
      "loss": 0.2319,
      "num_input_tokens_seen": 16105808,
      "step": 76320
    },
    {
      "epoch": 8.396589658965897,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.021583849479834154,
      "loss": 0.2319,
      "num_input_tokens_seen": 16106896,
      "step": 76325
    },
    {
      "epoch": 8.397139713971397,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02158255553765403,
      "loss": 0.2324,
      "num_input_tokens_seen": 16107984,
      "step": 76330
    },
    {
      "epoch": 8.397689768976898,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.021581261534805645,
      "loss": 0.2298,
      "num_input_tokens_seen": 16109040,
      "step": 76335
    },
    {
      "epoch": 8.398239823982399,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.021579967471300934,
      "loss": 0.2324,
      "num_input_tokens_seen": 16110160,
      "step": 76340
    },
    {
      "epoch": 8.398789878987898,
      "grad_norm": 0.0067138671875,
      "learning_rate": 0.021578673347151824,
      "loss": 0.2303,
      "num_input_tokens_seen": 16111280,
      "step": 76345
    },
    {
      "epoch": 8.3993399339934,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.021577379162370236,
      "loss": 0.2309,
      "num_input_tokens_seen": 16112368,
      "step": 76350
    },
    {
      "epoch": 8.3998899889989,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.021576084916968098,
      "loss": 0.2324,
      "num_input_tokens_seen": 16113392,
      "step": 76355
    },
    {
      "epoch": 8.4004400440044,
      "grad_norm": 0.003204345703125,
      "learning_rate": 0.021574790610957344,
      "loss": 0.2303,
      "num_input_tokens_seen": 16114512,
      "step": 76360
    },
    {
      "epoch": 8.400990099009901,
      "grad_norm": 0.000732421875,
      "learning_rate": 0.021573496244349892,
      "loss": 0.2303,
      "num_input_tokens_seen": 16115600,
      "step": 76365
    },
    {
      "epoch": 8.401540154015402,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02157220181715769,
      "loss": 0.2304,
      "num_input_tokens_seen": 16116688,
      "step": 76370
    },
    {
      "epoch": 8.402090209020901,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.021570907329392652,
      "loss": 0.2319,
      "num_input_tokens_seen": 16117680,
      "step": 76375
    },
    {
      "epoch": 8.402640264026402,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.021569612781066713,
      "loss": 0.2324,
      "num_input_tokens_seen": 16118736,
      "step": 76380
    },
    {
      "epoch": 8.403190319031903,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02156831817219181,
      "loss": 0.2314,
      "num_input_tokens_seen": 16119792,
      "step": 76385
    },
    {
      "epoch": 8.403740374037405,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.021567023502779867,
      "loss": 0.2319,
      "num_input_tokens_seen": 16120848,
      "step": 76390
    },
    {
      "epoch": 8.404290429042904,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02156572877284282,
      "loss": 0.2319,
      "num_input_tokens_seen": 16121904,
      "step": 76395
    },
    {
      "epoch": 8.404840484048405,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.021564433982392606,
      "loss": 0.2314,
      "num_input_tokens_seen": 16122864,
      "step": 76400
    },
    {
      "epoch": 8.405390539053906,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.021563139131441148,
      "loss": 0.235,
      "num_input_tokens_seen": 16123920,
      "step": 76405
    },
    {
      "epoch": 8.405940594059405,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.021561844220000383,
      "loss": 0.2319,
      "num_input_tokens_seen": 16125008,
      "step": 76410
    },
    {
      "epoch": 8.406490649064907,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.021560549248082254,
      "loss": 0.2298,
      "num_input_tokens_seen": 16126064,
      "step": 76415
    },
    {
      "epoch": 8.407040704070408,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.02155925421569869,
      "loss": 0.2335,
      "num_input_tokens_seen": 16127088,
      "step": 76420
    },
    {
      "epoch": 8.407590759075907,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.02155795912286163,
      "loss": 0.232,
      "num_input_tokens_seen": 16128144,
      "step": 76425
    },
    {
      "epoch": 8.408140814081408,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.021556663969583004,
      "loss": 0.2319,
      "num_input_tokens_seen": 16129296,
      "step": 76430
    },
    {
      "epoch": 8.408690869086909,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.021555368755874754,
      "loss": 0.2288,
      "num_input_tokens_seen": 16130384,
      "step": 76435
    },
    {
      "epoch": 8.409240924092408,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.021554073481748814,
      "loss": 0.2325,
      "num_input_tokens_seen": 16131440,
      "step": 76440
    },
    {
      "epoch": 8.40979097909791,
      "grad_norm": 0.0005645751953125,
      "learning_rate": 0.021552778147217126,
      "loss": 0.2319,
      "num_input_tokens_seen": 16132528,
      "step": 76445
    },
    {
      "epoch": 8.41034103410341,
      "grad_norm": 0.00079345703125,
      "learning_rate": 0.02155148275229162,
      "loss": 0.2288,
      "num_input_tokens_seen": 16133584,
      "step": 76450
    },
    {
      "epoch": 8.410891089108912,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.021550187296984245,
      "loss": 0.233,
      "num_input_tokens_seen": 16134672,
      "step": 76455
    },
    {
      "epoch": 8.411441144114411,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02154889178130694,
      "loss": 0.2324,
      "num_input_tokens_seen": 16135728,
      "step": 76460
    },
    {
      "epoch": 8.411991199119912,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.02154759620527164,
      "loss": 0.2309,
      "num_input_tokens_seen": 16136816,
      "step": 76465
    },
    {
      "epoch": 8.412541254125413,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.021546300568890286,
      "loss": 0.2304,
      "num_input_tokens_seen": 16137808,
      "step": 76470
    },
    {
      "epoch": 8.413091309130913,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.021545004872174825,
      "loss": 0.2324,
      "num_input_tokens_seen": 16138832,
      "step": 76475
    },
    {
      "epoch": 8.413641364136414,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02154370911513719,
      "loss": 0.2319,
      "num_input_tokens_seen": 16139856,
      "step": 76480
    },
    {
      "epoch": 8.414191419141915,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.021542413297789328,
      "loss": 0.2325,
      "num_input_tokens_seen": 16140880,
      "step": 76485
    },
    {
      "epoch": 8.414741474147414,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.021541117420143186,
      "loss": 0.2309,
      "num_input_tokens_seen": 16141872,
      "step": 76490
    },
    {
      "epoch": 8.415291529152915,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.0215398214822107,
      "loss": 0.2299,
      "num_input_tokens_seen": 16142896,
      "step": 76495
    },
    {
      "epoch": 8.415841584158416,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02153852548400382,
      "loss": 0.2335,
      "num_input_tokens_seen": 16143952,
      "step": 76500
    },
    {
      "epoch": 8.416391639163916,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.021537229425534484,
      "loss": 0.2324,
      "num_input_tokens_seen": 16145008,
      "step": 76505
    },
    {
      "epoch": 8.416941694169417,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.021535933306814643,
      "loss": 0.2314,
      "num_input_tokens_seen": 16146096,
      "step": 76510
    },
    {
      "epoch": 8.417491749174918,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.021534637127856245,
      "loss": 0.2314,
      "num_input_tokens_seen": 16147120,
      "step": 76515
    },
    {
      "epoch": 8.418041804180419,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02153334088867123,
      "loss": 0.233,
      "num_input_tokens_seen": 16148176,
      "step": 76520
    },
    {
      "epoch": 8.418591859185918,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.021532044589271545,
      "loss": 0.2319,
      "num_input_tokens_seen": 16149168,
      "step": 76525
    },
    {
      "epoch": 8.41914191419142,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.02153074822966914,
      "loss": 0.2308,
      "num_input_tokens_seen": 16150256,
      "step": 76530
    },
    {
      "epoch": 8.41969196919692,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.021529451809875966,
      "loss": 0.2324,
      "num_input_tokens_seen": 16151280,
      "step": 76535
    },
    {
      "epoch": 8.42024202420242,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.021528155329903962,
      "loss": 0.2324,
      "num_input_tokens_seen": 16152368,
      "step": 76540
    },
    {
      "epoch": 8.42079207920792,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.02152685878976509,
      "loss": 0.2309,
      "num_input_tokens_seen": 16153424,
      "step": 76545
    },
    {
      "epoch": 8.421342134213422,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.021525562189471287,
      "loss": 0.2298,
      "num_input_tokens_seen": 16154448,
      "step": 76550
    },
    {
      "epoch": 8.421892189218921,
      "grad_norm": 0.0027313232421875,
      "learning_rate": 0.021524265529034508,
      "loss": 0.2319,
      "num_input_tokens_seen": 16155472,
      "step": 76555
    },
    {
      "epoch": 8.422442244224422,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.021522968808466707,
      "loss": 0.2308,
      "num_input_tokens_seen": 16156464,
      "step": 76560
    },
    {
      "epoch": 8.422992299229923,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.021521672027779833,
      "loss": 0.2308,
      "num_input_tokens_seen": 16157520,
      "step": 76565
    },
    {
      "epoch": 8.423542354235423,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021520375186985834,
      "loss": 0.2319,
      "num_input_tokens_seen": 16158544,
      "step": 76570
    },
    {
      "epoch": 8.424092409240924,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.021519078286096664,
      "loss": 0.2303,
      "num_input_tokens_seen": 16159600,
      "step": 76575
    },
    {
      "epoch": 8.424642464246425,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02151778132512428,
      "loss": 0.2319,
      "num_input_tokens_seen": 16160656,
      "step": 76580
    },
    {
      "epoch": 8.425192519251926,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.021516484304080638,
      "loss": 0.2324,
      "num_input_tokens_seen": 16161712,
      "step": 76585
    },
    {
      "epoch": 8.425742574257425,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.021515187222977683,
      "loss": 0.2319,
      "num_input_tokens_seen": 16162736,
      "step": 76590
    },
    {
      "epoch": 8.426292629262926,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.021513890081827376,
      "loss": 0.2303,
      "num_input_tokens_seen": 16163792,
      "step": 76595
    },
    {
      "epoch": 8.426842684268427,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.02151259288064166,
      "loss": 0.2324,
      "num_input_tokens_seen": 16164784,
      "step": 76600
    },
    {
      "epoch": 8.427392739273927,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.021511295619432512,
      "loss": 0.2309,
      "num_input_tokens_seen": 16165872,
      "step": 76605
    },
    {
      "epoch": 8.427942794279428,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.02150999829821187,
      "loss": 0.2313,
      "num_input_tokens_seen": 16166832,
      "step": 76610
    },
    {
      "epoch": 8.428492849284929,
      "grad_norm": 0.000629425048828125,
      "learning_rate": 0.021508700916991703,
      "loss": 0.2308,
      "num_input_tokens_seen": 16167856,
      "step": 76615
    },
    {
      "epoch": 8.429042904290428,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02150740347578396,
      "loss": 0.2308,
      "num_input_tokens_seen": 16168848,
      "step": 76620
    },
    {
      "epoch": 8.42959295929593,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.0215061059746006,
      "loss": 0.2324,
      "num_input_tokens_seen": 16169904,
      "step": 76625
    },
    {
      "epoch": 8.43014301430143,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.021504808413453588,
      "loss": 0.2329,
      "num_input_tokens_seen": 16170992,
      "step": 76630
    },
    {
      "epoch": 8.430693069306932,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.021503510792354876,
      "loss": 0.2314,
      "num_input_tokens_seen": 16172048,
      "step": 76635
    },
    {
      "epoch": 8.43124312431243,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.021502213111316424,
      "loss": 0.2303,
      "num_input_tokens_seen": 16173072,
      "step": 76640
    },
    {
      "epoch": 8.431793179317932,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02150091537035019,
      "loss": 0.2329,
      "num_input_tokens_seen": 16174224,
      "step": 76645
    },
    {
      "epoch": 8.432343234323433,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.02149961756946815,
      "loss": 0.2329,
      "num_input_tokens_seen": 16175280,
      "step": 76650
    },
    {
      "epoch": 8.432893289328932,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.021498319708682243,
      "loss": 0.2298,
      "num_input_tokens_seen": 16176304,
      "step": 76655
    },
    {
      "epoch": 8.433443344334433,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.021497021788004445,
      "loss": 0.2324,
      "num_input_tokens_seen": 16177360,
      "step": 76660
    },
    {
      "epoch": 8.433993399339935,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.021495723807446715,
      "loss": 0.2319,
      "num_input_tokens_seen": 16178480,
      "step": 76665
    },
    {
      "epoch": 8.434543454345434,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.021494425767021014,
      "loss": 0.2324,
      "num_input_tokens_seen": 16179536,
      "step": 76670
    },
    {
      "epoch": 8.435093509350935,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02149312766673931,
      "loss": 0.2324,
      "num_input_tokens_seen": 16180656,
      "step": 76675
    },
    {
      "epoch": 8.435643564356436,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.021491829506613564,
      "loss": 0.2319,
      "num_input_tokens_seen": 16181744,
      "step": 76680
    },
    {
      "epoch": 8.436193619361935,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.021490531286655738,
      "loss": 0.2313,
      "num_input_tokens_seen": 16182832,
      "step": 76685
    },
    {
      "epoch": 8.436743674367436,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.0214892330068778,
      "loss": 0.2298,
      "num_input_tokens_seen": 16183920,
      "step": 76690
    },
    {
      "epoch": 8.437293729372938,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02148793466729172,
      "loss": 0.2335,
      "num_input_tokens_seen": 16184944,
      "step": 76695
    },
    {
      "epoch": 8.437843784378439,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02148663626790945,
      "loss": 0.2319,
      "num_input_tokens_seen": 16186064,
      "step": 76700
    },
    {
      "epoch": 8.438393839383938,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.021485337808742976,
      "loss": 0.2314,
      "num_input_tokens_seen": 16187184,
      "step": 76705
    },
    {
      "epoch": 8.438943894389439,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.021484039289804246,
      "loss": 0.2319,
      "num_input_tokens_seen": 16188240,
      "step": 76710
    },
    {
      "epoch": 8.43949394939494,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.021482740711105244,
      "loss": 0.2314,
      "num_input_tokens_seen": 16189328,
      "step": 76715
    },
    {
      "epoch": 8.44004400440044,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.021481442072657928,
      "loss": 0.234,
      "num_input_tokens_seen": 16190352,
      "step": 76720
    },
    {
      "epoch": 8.44059405940594,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.021480143374474275,
      "loss": 0.2329,
      "num_input_tokens_seen": 16191440,
      "step": 76725
    },
    {
      "epoch": 8.441144114411442,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.021478844616566246,
      "loss": 0.2303,
      "num_input_tokens_seen": 16192496,
      "step": 76730
    },
    {
      "epoch": 8.441694169416941,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.021477545798945816,
      "loss": 0.2308,
      "num_input_tokens_seen": 16193552,
      "step": 76735
    },
    {
      "epoch": 8.442244224422442,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.021476246921624954,
      "loss": 0.2324,
      "num_input_tokens_seen": 16194704,
      "step": 76740
    },
    {
      "epoch": 8.442794279427943,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021474947984615633,
      "loss": 0.2308,
      "num_input_tokens_seen": 16195728,
      "step": 76745
    },
    {
      "epoch": 8.443344334433444,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.021473648987929826,
      "loss": 0.2313,
      "num_input_tokens_seen": 16196784,
      "step": 76750
    },
    {
      "epoch": 8.443894389438944,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.021472349931579496,
      "loss": 0.2324,
      "num_input_tokens_seen": 16197808,
      "step": 76755
    },
    {
      "epoch": 8.444444444444445,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.021471050815576626,
      "loss": 0.2308,
      "num_input_tokens_seen": 16198800,
      "step": 76760
    },
    {
      "epoch": 8.444994499449946,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.02146975163993319,
      "loss": 0.2303,
      "num_input_tokens_seen": 16199792,
      "step": 76765
    },
    {
      "epoch": 8.445544554455445,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02146845240466115,
      "loss": 0.2329,
      "num_input_tokens_seen": 16200848,
      "step": 76770
    },
    {
      "epoch": 8.446094609460946,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021467153109772492,
      "loss": 0.2293,
      "num_input_tokens_seen": 16201904,
      "step": 76775
    },
    {
      "epoch": 8.446644664466447,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021465853755279186,
      "loss": 0.2324,
      "num_input_tokens_seen": 16202960,
      "step": 76780
    },
    {
      "epoch": 8.447194719471947,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02146455434119321,
      "loss": 0.2314,
      "num_input_tokens_seen": 16203984,
      "step": 76785
    },
    {
      "epoch": 8.447744774477448,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.021463254867526538,
      "loss": 0.2308,
      "num_input_tokens_seen": 16205008,
      "step": 76790
    },
    {
      "epoch": 8.448294829482949,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.021461955334291145,
      "loss": 0.2298,
      "num_input_tokens_seen": 16206000,
      "step": 76795
    },
    {
      "epoch": 8.448844884488448,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.021460655741499014,
      "loss": 0.2288,
      "num_input_tokens_seen": 16207024,
      "step": 76800
    },
    {
      "epoch": 8.44939493949395,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.021459356089162118,
      "loss": 0.2319,
      "num_input_tokens_seen": 16208144,
      "step": 76805
    },
    {
      "epoch": 8.44994499449945,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021458056377292437,
      "loss": 0.2303,
      "num_input_tokens_seen": 16209200,
      "step": 76810
    },
    {
      "epoch": 8.450495049504951,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.021456756605901944,
      "loss": 0.2309,
      "num_input_tokens_seen": 16210192,
      "step": 76815
    },
    {
      "epoch": 8.45104510451045,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02145545677500263,
      "loss": 0.2313,
      "num_input_tokens_seen": 16211216,
      "step": 76820
    },
    {
      "epoch": 8.451595159515952,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.021454156884606466,
      "loss": 0.2324,
      "num_input_tokens_seen": 16212272,
      "step": 76825
    },
    {
      "epoch": 8.452145214521453,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021452856934725435,
      "loss": 0.2324,
      "num_input_tokens_seen": 16213264,
      "step": 76830
    },
    {
      "epoch": 8.452695269526952,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.02145155692537152,
      "loss": 0.2308,
      "num_input_tokens_seen": 16214320,
      "step": 76835
    },
    {
      "epoch": 8.453245324532453,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.021450256856556695,
      "loss": 0.2319,
      "num_input_tokens_seen": 16215376,
      "step": 76840
    },
    {
      "epoch": 8.453795379537954,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.021448956728292955,
      "loss": 0.2314,
      "num_input_tokens_seen": 16216464,
      "step": 76845
    },
    {
      "epoch": 8.454345434543454,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.02144765654059227,
      "loss": 0.2288,
      "num_input_tokens_seen": 16217552,
      "step": 76850
    },
    {
      "epoch": 8.454895489548955,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021446356293466633,
      "loss": 0.2329,
      "num_input_tokens_seen": 16218608,
      "step": 76855
    },
    {
      "epoch": 8.455445544554456,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.021445055986928016,
      "loss": 0.2303,
      "num_input_tokens_seen": 16219664,
      "step": 76860
    },
    {
      "epoch": 8.455995599559955,
      "grad_norm": 0.009765625,
      "learning_rate": 0.021443755620988415,
      "loss": 0.2309,
      "num_input_tokens_seen": 16220720,
      "step": 76865
    },
    {
      "epoch": 8.456545654565456,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.021442455195659814,
      "loss": 0.2324,
      "num_input_tokens_seen": 16221808,
      "step": 76870
    },
    {
      "epoch": 8.457095709570957,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.021441154710954192,
      "loss": 0.2324,
      "num_input_tokens_seen": 16222864,
      "step": 76875
    },
    {
      "epoch": 8.457645764576458,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.021439854166883536,
      "loss": 0.233,
      "num_input_tokens_seen": 16223856,
      "step": 76880
    },
    {
      "epoch": 8.458195819581958,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.021438553563459837,
      "loss": 0.2314,
      "num_input_tokens_seen": 16224880,
      "step": 76885
    },
    {
      "epoch": 8.458745874587459,
      "grad_norm": 0.0004558563232421875,
      "learning_rate": 0.021437252900695074,
      "loss": 0.2319,
      "num_input_tokens_seen": 16225904,
      "step": 76890
    },
    {
      "epoch": 8.45929592959296,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02143595217860125,
      "loss": 0.2324,
      "num_input_tokens_seen": 16226896,
      "step": 76895
    },
    {
      "epoch": 8.45984598459846,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02143465139719033,
      "loss": 0.2319,
      "num_input_tokens_seen": 16227984,
      "step": 76900
    },
    {
      "epoch": 8.46039603960396,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.021433350556474325,
      "loss": 0.2314,
      "num_input_tokens_seen": 16229072,
      "step": 76905
    },
    {
      "epoch": 8.460946094609461,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.021432049656465214,
      "loss": 0.2314,
      "num_input_tokens_seen": 16230064,
      "step": 76910
    },
    {
      "epoch": 8.46149614961496,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021430748697174987,
      "loss": 0.2308,
      "num_input_tokens_seen": 16231184,
      "step": 76915
    },
    {
      "epoch": 8.462046204620462,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.021429447678615637,
      "loss": 0.2309,
      "num_input_tokens_seen": 16232208,
      "step": 76920
    },
    {
      "epoch": 8.462596259625963,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02142814660079915,
      "loss": 0.233,
      "num_input_tokens_seen": 16233264,
      "step": 76925
    },
    {
      "epoch": 8.463146314631462,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.021426845463737517,
      "loss": 0.2303,
      "num_input_tokens_seen": 16234320,
      "step": 76930
    },
    {
      "epoch": 8.463696369636963,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.021425544267442736,
      "loss": 0.2308,
      "num_input_tokens_seen": 16235408,
      "step": 76935
    },
    {
      "epoch": 8.464246424642464,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.021424243011926802,
      "loss": 0.2309,
      "num_input_tokens_seen": 16236528,
      "step": 76940
    },
    {
      "epoch": 8.464796479647966,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.021422941697201697,
      "loss": 0.2319,
      "num_input_tokens_seen": 16237584,
      "step": 76945
    },
    {
      "epoch": 8.465346534653465,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02142164032327942,
      "loss": 0.2314,
      "num_input_tokens_seen": 16238576,
      "step": 76950
    },
    {
      "epoch": 8.465896589658966,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021420338890171968,
      "loss": 0.2319,
      "num_input_tokens_seen": 16239568,
      "step": 76955
    },
    {
      "epoch": 8.466446644664467,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02141903739789133,
      "loss": 0.2303,
      "num_input_tokens_seen": 16240592,
      "step": 76960
    },
    {
      "epoch": 8.466996699669966,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.02141773584644951,
      "loss": 0.2314,
      "num_input_tokens_seen": 16241648,
      "step": 76965
    },
    {
      "epoch": 8.467546754675467,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.021416434235858497,
      "loss": 0.2329,
      "num_input_tokens_seen": 16242672,
      "step": 76970
    },
    {
      "epoch": 8.468096809680969,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.021415132566130285,
      "loss": 0.2303,
      "num_input_tokens_seen": 16243696,
      "step": 76975
    },
    {
      "epoch": 8.468646864686468,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021413830837276875,
      "loss": 0.2314,
      "num_input_tokens_seen": 16244720,
      "step": 76980
    },
    {
      "epoch": 8.469196919691969,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.021412529049310267,
      "loss": 0.2303,
      "num_input_tokens_seen": 16245808,
      "step": 76985
    },
    {
      "epoch": 8.46974697469747,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021411227202242453,
      "loss": 0.2309,
      "num_input_tokens_seen": 16246832,
      "step": 76990
    },
    {
      "epoch": 8.47029702970297,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.021409925296085434,
      "loss": 0.2329,
      "num_input_tokens_seen": 16247888,
      "step": 76995
    },
    {
      "epoch": 8.47084708470847,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.021408623330851216,
      "loss": 0.2314,
      "num_input_tokens_seen": 16248944,
      "step": 77000
    },
    {
      "epoch": 8.471397139713972,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.02140732130655178,
      "loss": 0.2308,
      "num_input_tokens_seen": 16250000,
      "step": 77005
    },
    {
      "epoch": 8.471947194719473,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.02140601922319915,
      "loss": 0.2324,
      "num_input_tokens_seen": 16251088,
      "step": 77010
    },
    {
      "epoch": 8.472497249724972,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.021404717080805307,
      "loss": 0.2309,
      "num_input_tokens_seen": 16252208,
      "step": 77015
    },
    {
      "epoch": 8.473047304730473,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.021403414879382264,
      "loss": 0.2288,
      "num_input_tokens_seen": 16253264,
      "step": 77020
    },
    {
      "epoch": 8.473597359735974,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.021402112618942013,
      "loss": 0.2308,
      "num_input_tokens_seen": 16254320,
      "step": 77025
    },
    {
      "epoch": 8.474147414741473,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.02140081029949657,
      "loss": 0.2298,
      "num_input_tokens_seen": 16255344,
      "step": 77030
    },
    {
      "epoch": 8.474697469746975,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.021399507921057923,
      "loss": 0.2319,
      "num_input_tokens_seen": 16256400,
      "step": 77035
    },
    {
      "epoch": 8.475247524752476,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.021398205483638084,
      "loss": 0.2319,
      "num_input_tokens_seen": 16257488,
      "step": 77040
    },
    {
      "epoch": 8.475797579757975,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.02139690298724906,
      "loss": 0.2319,
      "num_input_tokens_seen": 16258544,
      "step": 77045
    },
    {
      "epoch": 8.476347634763476,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.021395600431902844,
      "loss": 0.2309,
      "num_input_tokens_seen": 16259632,
      "step": 77050
    },
    {
      "epoch": 8.476897689768977,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.02139429781761145,
      "loss": 0.234,
      "num_input_tokens_seen": 16260752,
      "step": 77055
    },
    {
      "epoch": 8.477447744774478,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.021392995144386882,
      "loss": 0.2314,
      "num_input_tokens_seen": 16261808,
      "step": 77060
    },
    {
      "epoch": 8.477997799779978,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.021391692412241144,
      "loss": 0.2324,
      "num_input_tokens_seen": 16262832,
      "step": 77065
    },
    {
      "epoch": 8.478547854785479,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.021390389621186248,
      "loss": 0.2308,
      "num_input_tokens_seen": 16263952,
      "step": 77070
    },
    {
      "epoch": 8.47909790979098,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.021389086771234196,
      "loss": 0.2303,
      "num_input_tokens_seen": 16265008,
      "step": 77075
    },
    {
      "epoch": 8.479647964796479,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.021387783862396995,
      "loss": 0.2314,
      "num_input_tokens_seen": 16266064,
      "step": 77080
    },
    {
      "epoch": 8.48019801980198,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.021386480894686657,
      "loss": 0.2314,
      "num_input_tokens_seen": 16267120,
      "step": 77085
    },
    {
      "epoch": 8.480748074807481,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.021385177868115186,
      "loss": 0.2303,
      "num_input_tokens_seen": 16268240,
      "step": 77090
    },
    {
      "epoch": 8.48129812981298,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.021383874782694597,
      "loss": 0.2309,
      "num_input_tokens_seen": 16269296,
      "step": 77095
    },
    {
      "epoch": 8.481848184818482,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.021382571638436897,
      "loss": 0.233,
      "num_input_tokens_seen": 16270352,
      "step": 77100
    },
    {
      "epoch": 8.482398239823983,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.0213812684353541,
      "loss": 0.2309,
      "num_input_tokens_seen": 16271376,
      "step": 77105
    },
    {
      "epoch": 8.482948294829482,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.02137996517345821,
      "loss": 0.2319,
      "num_input_tokens_seen": 16272368,
      "step": 77110
    },
    {
      "epoch": 8.483498349834983,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.021378661852761244,
      "loss": 0.2324,
      "num_input_tokens_seen": 16273424,
      "step": 77115
    },
    {
      "epoch": 8.484048404840484,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.021377358473275206,
      "loss": 0.2319,
      "num_input_tokens_seen": 16274512,
      "step": 77120
    },
    {
      "epoch": 8.484598459845985,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.021376055035012123,
      "loss": 0.233,
      "num_input_tokens_seen": 16275536,
      "step": 77125
    },
    {
      "epoch": 8.485148514851485,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.021374751537984002,
      "loss": 0.2319,
      "num_input_tokens_seen": 16276560,
      "step": 77130
    },
    {
      "epoch": 8.485698569856986,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.021373447982202847,
      "loss": 0.2319,
      "num_input_tokens_seen": 16277584,
      "step": 77135
    },
    {
      "epoch": 8.486248624862487,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021372144367680687,
      "loss": 0.2309,
      "num_input_tokens_seen": 16278608,
      "step": 77140
    },
    {
      "epoch": 8.486798679867986,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.021370840694429526,
      "loss": 0.2335,
      "num_input_tokens_seen": 16279568,
      "step": 77145
    },
    {
      "epoch": 8.487348734873487,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.021369536962461383,
      "loss": 0.2288,
      "num_input_tokens_seen": 16280624,
      "step": 77150
    },
    {
      "epoch": 8.487898789878988,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02136823317178828,
      "loss": 0.2314,
      "num_input_tokens_seen": 16281680,
      "step": 77155
    },
    {
      "epoch": 8.488448844884488,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02136692932242222,
      "loss": 0.2329,
      "num_input_tokens_seen": 16282736,
      "step": 77160
    },
    {
      "epoch": 8.488998899889989,
      "grad_norm": 0.009765625,
      "learning_rate": 0.021365625414375228,
      "loss": 0.233,
      "num_input_tokens_seen": 16283824,
      "step": 77165
    },
    {
      "epoch": 8.48954895489549,
      "grad_norm": 0.00072479248046875,
      "learning_rate": 0.02136432144765933,
      "loss": 0.2314,
      "num_input_tokens_seen": 16284880,
      "step": 77170
    },
    {
      "epoch": 8.490099009900991,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.021363017422286523,
      "loss": 0.2309,
      "num_input_tokens_seen": 16285936,
      "step": 77175
    },
    {
      "epoch": 8.49064906490649,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021361713338268848,
      "loss": 0.2308,
      "num_input_tokens_seen": 16286960,
      "step": 77180
    },
    {
      "epoch": 8.491199119911991,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.02136040919561831,
      "loss": 0.2314,
      "num_input_tokens_seen": 16288048,
      "step": 77185
    },
    {
      "epoch": 8.491749174917492,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02135910499434693,
      "loss": 0.2309,
      "num_input_tokens_seen": 16289104,
      "step": 77190
    },
    {
      "epoch": 8.492299229922992,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02135780073446673,
      "loss": 0.2309,
      "num_input_tokens_seen": 16290160,
      "step": 77195
    },
    {
      "epoch": 8.492849284928493,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02135649641598974,
      "loss": 0.2319,
      "num_input_tokens_seen": 16291184,
      "step": 77200
    },
    {
      "epoch": 8.493399339933994,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.021355192038927964,
      "loss": 0.2324,
      "num_input_tokens_seen": 16292272,
      "step": 77205
    },
    {
      "epoch": 8.493949394939493,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.021353887603293438,
      "loss": 0.2304,
      "num_input_tokens_seen": 16293328,
      "step": 77210
    },
    {
      "epoch": 8.494499449944994,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.021352583109098176,
      "loss": 0.2303,
      "num_input_tokens_seen": 16294416,
      "step": 77215
    },
    {
      "epoch": 8.495049504950495,
      "grad_norm": 0.000453948974609375,
      "learning_rate": 0.021351278556354204,
      "loss": 0.2288,
      "num_input_tokens_seen": 16295440,
      "step": 77220
    },
    {
      "epoch": 8.495599559955995,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021349973945073545,
      "loss": 0.2298,
      "num_input_tokens_seen": 16296592,
      "step": 77225
    },
    {
      "epoch": 8.496149614961496,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.021348669275268226,
      "loss": 0.2314,
      "num_input_tokens_seen": 16297648,
      "step": 77230
    },
    {
      "epoch": 8.496699669966997,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02134736454695026,
      "loss": 0.2299,
      "num_input_tokens_seen": 16298640,
      "step": 77235
    },
    {
      "epoch": 8.497249724972498,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02134605976013169,
      "loss": 0.2288,
      "num_input_tokens_seen": 16299632,
      "step": 77240
    },
    {
      "epoch": 8.497799779977997,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02134475491482453,
      "loss": 0.2304,
      "num_input_tokens_seen": 16300688,
      "step": 77245
    },
    {
      "epoch": 8.498349834983498,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.021343450011040804,
      "loss": 0.2314,
      "num_input_tokens_seen": 16301744,
      "step": 77250
    },
    {
      "epoch": 8.498899889989,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02134214504879255,
      "loss": 0.2288,
      "num_input_tokens_seen": 16302768,
      "step": 77255
    },
    {
      "epoch": 8.499449944994499,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021340840028091786,
      "loss": 0.2325,
      "num_input_tokens_seen": 16303760,
      "step": 77260
    },
    {
      "epoch": 8.5,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.021339534948950536,
      "loss": 0.2303,
      "num_input_tokens_seen": 16304880,
      "step": 77265
    },
    {
      "epoch": 8.500550055005501,
      "grad_norm": 0.000774383544921875,
      "learning_rate": 0.02133822981138084,
      "loss": 0.233,
      "num_input_tokens_seen": 16305936,
      "step": 77270
    },
    {
      "epoch": 8.501100110011,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.021336924615394725,
      "loss": 0.2324,
      "num_input_tokens_seen": 16307024,
      "step": 77275
    },
    {
      "epoch": 8.501650165016502,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.021335619361004213,
      "loss": 0.2324,
      "num_input_tokens_seen": 16308048,
      "step": 77280
    },
    {
      "epoch": 8.502200220022003,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.021334314048221337,
      "loss": 0.2314,
      "num_input_tokens_seen": 16309104,
      "step": 77285
    },
    {
      "epoch": 8.502750275027502,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.021333008677058125,
      "loss": 0.2319,
      "num_input_tokens_seen": 16310160,
      "step": 77290
    },
    {
      "epoch": 8.503300330033003,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.021331703247526614,
      "loss": 0.233,
      "num_input_tokens_seen": 16311312,
      "step": 77295
    },
    {
      "epoch": 8.503850385038504,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.021330397759638835,
      "loss": 0.2288,
      "num_input_tokens_seen": 16312336,
      "step": 77300
    },
    {
      "epoch": 8.504400440044005,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02132909221340681,
      "loss": 0.2314,
      "num_input_tokens_seen": 16313360,
      "step": 77305
    },
    {
      "epoch": 8.504950495049505,
      "grad_norm": 0.000843048095703125,
      "learning_rate": 0.021327786608842585,
      "loss": 0.2304,
      "num_input_tokens_seen": 16314384,
      "step": 77310
    },
    {
      "epoch": 8.505500550055006,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.021326480945958185,
      "loss": 0.2304,
      "num_input_tokens_seen": 16315376,
      "step": 77315
    },
    {
      "epoch": 8.506050605060507,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.02132517522476565,
      "loss": 0.2324,
      "num_input_tokens_seen": 16316464,
      "step": 77320
    },
    {
      "epoch": 8.506600660066006,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021323869445277,
      "loss": 0.2314,
      "num_input_tokens_seen": 16317488,
      "step": 77325
    },
    {
      "epoch": 8.507150715071507,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.02132256360750429,
      "loss": 0.2294,
      "num_input_tokens_seen": 16318544,
      "step": 77330
    },
    {
      "epoch": 8.507700770077008,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.02132125771145954,
      "loss": 0.2304,
      "num_input_tokens_seen": 16319632,
      "step": 77335
    },
    {
      "epoch": 8.508250825082508,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.021319951757154794,
      "loss": 0.2315,
      "num_input_tokens_seen": 16320656,
      "step": 77340
    },
    {
      "epoch": 8.508800880088009,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.021318645744602086,
      "loss": 0.2278,
      "num_input_tokens_seen": 16321744,
      "step": 77345
    },
    {
      "epoch": 8.50935093509351,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02131733967381345,
      "loss": 0.233,
      "num_input_tokens_seen": 16322832,
      "step": 77350
    },
    {
      "epoch": 8.509900990099009,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.021316033544800925,
      "loss": 0.2341,
      "num_input_tokens_seen": 16323888,
      "step": 77355
    },
    {
      "epoch": 8.51045104510451,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02131472735757655,
      "loss": 0.232,
      "num_input_tokens_seen": 16324976,
      "step": 77360
    },
    {
      "epoch": 8.511001100110011,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.021313421112152363,
      "loss": 0.2331,
      "num_input_tokens_seen": 16326032,
      "step": 77365
    },
    {
      "epoch": 8.511551155115512,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.0213121148085404,
      "loss": 0.2346,
      "num_input_tokens_seen": 16327088,
      "step": 77370
    },
    {
      "epoch": 8.512101210121012,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.02131080844675271,
      "loss": 0.2283,
      "num_input_tokens_seen": 16328176,
      "step": 77375
    },
    {
      "epoch": 8.512651265126513,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.021309502026801323,
      "loss": 0.2304,
      "num_input_tokens_seen": 16329232,
      "step": 77380
    },
    {
      "epoch": 8.513201320132014,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02130819554869828,
      "loss": 0.2346,
      "num_input_tokens_seen": 16330352,
      "step": 77385
    },
    {
      "epoch": 8.513751375137513,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.021306889012455636,
      "loss": 0.232,
      "num_input_tokens_seen": 16331376,
      "step": 77390
    },
    {
      "epoch": 8.514301430143014,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02130558241808541,
      "loss": 0.2325,
      "num_input_tokens_seen": 16332400,
      "step": 77395
    },
    {
      "epoch": 8.514851485148515,
      "grad_norm": 0.0005950927734375,
      "learning_rate": 0.021304275765599668,
      "loss": 0.2309,
      "num_input_tokens_seen": 16333424,
      "step": 77400
    },
    {
      "epoch": 8.515401540154015,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.021302969055010435,
      "loss": 0.2309,
      "num_input_tokens_seen": 16334544,
      "step": 77405
    },
    {
      "epoch": 8.515951595159516,
      "grad_norm": 0.009765625,
      "learning_rate": 0.021301662286329758,
      "loss": 0.2319,
      "num_input_tokens_seen": 16335536,
      "step": 77410
    },
    {
      "epoch": 8.516501650165017,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02130035545956969,
      "loss": 0.2293,
      "num_input_tokens_seen": 16336528,
      "step": 77415
    },
    {
      "epoch": 8.517051705170516,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.021299048574742265,
      "loss": 0.2345,
      "num_input_tokens_seen": 16337520,
      "step": 77420
    },
    {
      "epoch": 8.517601760176017,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02129774163185953,
      "loss": 0.233,
      "num_input_tokens_seen": 16338640,
      "step": 77425
    },
    {
      "epoch": 8.518151815181518,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.021296434630933534,
      "loss": 0.233,
      "num_input_tokens_seen": 16339728,
      "step": 77430
    },
    {
      "epoch": 8.51870187018702,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021295127571976324,
      "loss": 0.2324,
      "num_input_tokens_seen": 16340784,
      "step": 77435
    },
    {
      "epoch": 8.519251925192519,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.021293820454999945,
      "loss": 0.2309,
      "num_input_tokens_seen": 16341808,
      "step": 77440
    },
    {
      "epoch": 8.51980198019802,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.021292513280016443,
      "loss": 0.2298,
      "num_input_tokens_seen": 16342896,
      "step": 77445
    },
    {
      "epoch": 8.520352035203521,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02129120604703786,
      "loss": 0.2314,
      "num_input_tokens_seen": 16343952,
      "step": 77450
    },
    {
      "epoch": 8.52090209020902,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.021289898756076252,
      "loss": 0.2324,
      "num_input_tokens_seen": 16345104,
      "step": 77455
    },
    {
      "epoch": 8.521452145214521,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.021288591407143665,
      "loss": 0.2303,
      "num_input_tokens_seen": 16346096,
      "step": 77460
    },
    {
      "epoch": 8.522002200220022,
      "grad_norm": 0.000782012939453125,
      "learning_rate": 0.021287284000252153,
      "loss": 0.2314,
      "num_input_tokens_seen": 16347120,
      "step": 77465
    },
    {
      "epoch": 8.522552255225522,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.021285976535413757,
      "loss": 0.2335,
      "num_input_tokens_seen": 16348208,
      "step": 77470
    },
    {
      "epoch": 8.523102310231023,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02128466901264053,
      "loss": 0.2319,
      "num_input_tokens_seen": 16349264,
      "step": 77475
    },
    {
      "epoch": 8.523652365236524,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.021283361431944526,
      "loss": 0.2314,
      "num_input_tokens_seen": 16350288,
      "step": 77480
    },
    {
      "epoch": 8.524202420242025,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.021282053793337797,
      "loss": 0.2319,
      "num_input_tokens_seen": 16351312,
      "step": 77485
    },
    {
      "epoch": 8.524752475247524,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.021280746096832392,
      "loss": 0.2324,
      "num_input_tokens_seen": 16352336,
      "step": 77490
    },
    {
      "epoch": 8.525302530253025,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.021279438342440362,
      "loss": 0.2298,
      "num_input_tokens_seen": 16353360,
      "step": 77495
    },
    {
      "epoch": 8.525852585258527,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.021278130530173765,
      "loss": 0.2313,
      "num_input_tokens_seen": 16354416,
      "step": 77500
    },
    {
      "epoch": 8.526402640264026,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.021276822660044648,
      "loss": 0.2303,
      "num_input_tokens_seen": 16355408,
      "step": 77505
    },
    {
      "epoch": 8.526952695269527,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.021275514732065067,
      "loss": 0.2303,
      "num_input_tokens_seen": 16356432,
      "step": 77510
    },
    {
      "epoch": 8.527502750275028,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.021274206746247082,
      "loss": 0.2309,
      "num_input_tokens_seen": 16357520,
      "step": 77515
    },
    {
      "epoch": 8.528052805280527,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.021272898702602745,
      "loss": 0.2324,
      "num_input_tokens_seen": 16358576,
      "step": 77520
    },
    {
      "epoch": 8.528602860286028,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.021271590601144107,
      "loss": 0.2293,
      "num_input_tokens_seen": 16359664,
      "step": 77525
    },
    {
      "epoch": 8.52915291529153,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02127028244188323,
      "loss": 0.2303,
      "num_input_tokens_seen": 16360720,
      "step": 77530
    },
    {
      "epoch": 8.52970297029703,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02126897422483217,
      "loss": 0.2288,
      "num_input_tokens_seen": 16361776,
      "step": 77535
    },
    {
      "epoch": 8.53025302530253,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.021267665950002977,
      "loss": 0.2314,
      "num_input_tokens_seen": 16362800,
      "step": 77540
    },
    {
      "epoch": 8.530803080308031,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02126635761740772,
      "loss": 0.2314,
      "num_input_tokens_seen": 16363824,
      "step": 77545
    },
    {
      "epoch": 8.531353135313532,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.021265049227058453,
      "loss": 0.2308,
      "num_input_tokens_seen": 16364912,
      "step": 77550
    },
    {
      "epoch": 8.531903190319031,
      "grad_norm": 0.00081634521484375,
      "learning_rate": 0.02126374077896723,
      "loss": 0.2314,
      "num_input_tokens_seen": 16365904,
      "step": 77555
    },
    {
      "epoch": 8.532453245324533,
      "grad_norm": 0.000667572021484375,
      "learning_rate": 0.021262432273146115,
      "loss": 0.2308,
      "num_input_tokens_seen": 16366960,
      "step": 77560
    },
    {
      "epoch": 8.533003300330034,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.021261123709607165,
      "loss": 0.2319,
      "num_input_tokens_seen": 16367952,
      "step": 77565
    },
    {
      "epoch": 8.533553355335533,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02125981508836244,
      "loss": 0.2298,
      "num_input_tokens_seen": 16368944,
      "step": 77570
    },
    {
      "epoch": 8.534103410341034,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.021258506409424005,
      "loss": 0.2319,
      "num_input_tokens_seen": 16370000,
      "step": 77575
    },
    {
      "epoch": 8.534653465346535,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.021257197672803922,
      "loss": 0.2308,
      "num_input_tokens_seen": 16371024,
      "step": 77580
    },
    {
      "epoch": 8.535203520352034,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.02125588887851425,
      "loss": 0.2314,
      "num_input_tokens_seen": 16372016,
      "step": 77585
    },
    {
      "epoch": 8.535753575357536,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021254580026567053,
      "loss": 0.2314,
      "num_input_tokens_seen": 16373008,
      "step": 77590
    },
    {
      "epoch": 8.536303630363037,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02125327111697439,
      "loss": 0.2329,
      "num_input_tokens_seen": 16374000,
      "step": 77595
    },
    {
      "epoch": 8.536853685368538,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.021251962149748328,
      "loss": 0.2319,
      "num_input_tokens_seen": 16375088,
      "step": 77600
    },
    {
      "epoch": 8.537403740374037,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.021250653124900935,
      "loss": 0.2314,
      "num_input_tokens_seen": 16376176,
      "step": 77605
    },
    {
      "epoch": 8.537953795379538,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.02124934404244427,
      "loss": 0.2313,
      "num_input_tokens_seen": 16377296,
      "step": 77610
    },
    {
      "epoch": 8.53850385038504,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.021248034902390394,
      "loss": 0.2324,
      "num_input_tokens_seen": 16378384,
      "step": 77615
    },
    {
      "epoch": 8.539053905390539,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.021246725704751383,
      "loss": 0.2298,
      "num_input_tokens_seen": 16379440,
      "step": 77620
    },
    {
      "epoch": 8.53960396039604,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.0212454164495393,
      "loss": 0.2324,
      "num_input_tokens_seen": 16380560,
      "step": 77625
    },
    {
      "epoch": 8.54015401540154,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02124410713676621,
      "loss": 0.2314,
      "num_input_tokens_seen": 16381680,
      "step": 77630
    },
    {
      "epoch": 8.54070407040704,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02124279776644418,
      "loss": 0.2303,
      "num_input_tokens_seen": 16382800,
      "step": 77635
    },
    {
      "epoch": 8.541254125412541,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.021241488338585277,
      "loss": 0.2329,
      "num_input_tokens_seen": 16383856,
      "step": 77640
    },
    {
      "epoch": 8.541804180418042,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.02124017885320157,
      "loss": 0.2319,
      "num_input_tokens_seen": 16384944,
      "step": 77645
    },
    {
      "epoch": 8.542354235423542,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021238869310305133,
      "loss": 0.2308,
      "num_input_tokens_seen": 16386000,
      "step": 77650
    },
    {
      "epoch": 8.542904290429043,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02123755970990803,
      "loss": 0.2303,
      "num_input_tokens_seen": 16387024,
      "step": 77655
    },
    {
      "epoch": 8.543454345434544,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02123625005202233,
      "loss": 0.2314,
      "num_input_tokens_seen": 16388080,
      "step": 77660
    },
    {
      "epoch": 8.544004400440045,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02123494033666011,
      "loss": 0.2334,
      "num_input_tokens_seen": 16389104,
      "step": 77665
    },
    {
      "epoch": 8.544554455445544,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.021233630563833435,
      "loss": 0.2298,
      "num_input_tokens_seen": 16390192,
      "step": 77670
    },
    {
      "epoch": 8.545104510451045,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.021232320733554372,
      "loss": 0.2324,
      "num_input_tokens_seen": 16391280,
      "step": 77675
    },
    {
      "epoch": 8.545654565456546,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.021231010845835008,
      "loss": 0.2308,
      "num_input_tokens_seen": 16392336,
      "step": 77680
    },
    {
      "epoch": 8.546204620462046,
      "grad_norm": 0.001953125,
      "learning_rate": 0.0212297009006874,
      "loss": 0.2304,
      "num_input_tokens_seen": 16393456,
      "step": 77685
    },
    {
      "epoch": 8.546754675467547,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.021228390898123634,
      "loss": 0.2304,
      "num_input_tokens_seen": 16394544,
      "step": 77690
    },
    {
      "epoch": 8.547304730473048,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.021227080838155776,
      "loss": 0.2308,
      "num_input_tokens_seen": 16395568,
      "step": 77695
    },
    {
      "epoch": 8.547854785478547,
      "grad_norm": 0.0006256103515625,
      "learning_rate": 0.0212257707207959,
      "loss": 0.2304,
      "num_input_tokens_seen": 16396688,
      "step": 77700
    },
    {
      "epoch": 8.548404840484048,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.021224460546056083,
      "loss": 0.2319,
      "num_input_tokens_seen": 16397808,
      "step": 77705
    },
    {
      "epoch": 8.54895489548955,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.0212231503139484,
      "loss": 0.2324,
      "num_input_tokens_seen": 16398896,
      "step": 77710
    },
    {
      "epoch": 8.549504950495049,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.021221840024484927,
      "loss": 0.2293,
      "num_input_tokens_seen": 16399952,
      "step": 77715
    },
    {
      "epoch": 8.55005500550055,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.021220529677677742,
      "loss": 0.2288,
      "num_input_tokens_seen": 16401072,
      "step": 77720
    },
    {
      "epoch": 8.55060506050605,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021219219273538918,
      "loss": 0.2308,
      "num_input_tokens_seen": 16402096,
      "step": 77725
    },
    {
      "epoch": 8.551155115511552,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.021217908812080535,
      "loss": 0.2319,
      "num_input_tokens_seen": 16403184,
      "step": 77730
    },
    {
      "epoch": 8.551705170517051,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02121659829331467,
      "loss": 0.2314,
      "num_input_tokens_seen": 16404272,
      "step": 77735
    },
    {
      "epoch": 8.552255225522552,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.021215287717253402,
      "loss": 0.2293,
      "num_input_tokens_seen": 16405296,
      "step": 77740
    },
    {
      "epoch": 8.552805280528053,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.021213977083908805,
      "loss": 0.2308,
      "num_input_tokens_seen": 16406352,
      "step": 77745
    },
    {
      "epoch": 8.553355335533553,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.021212666393292973,
      "loss": 0.2314,
      "num_input_tokens_seen": 16407408,
      "step": 77750
    },
    {
      "epoch": 8.553905390539054,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02121135564541797,
      "loss": 0.2293,
      "num_input_tokens_seen": 16408400,
      "step": 77755
    },
    {
      "epoch": 8.554455445544555,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02121004484029588,
      "loss": 0.2277,
      "num_input_tokens_seen": 16409456,
      "step": 77760
    },
    {
      "epoch": 8.555005500550054,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.02120873397793879,
      "loss": 0.2324,
      "num_input_tokens_seen": 16410576,
      "step": 77765
    },
    {
      "epoch": 8.555555555555555,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.021207423058358776,
      "loss": 0.2298,
      "num_input_tokens_seen": 16411568,
      "step": 77770
    },
    {
      "epoch": 8.556105610561056,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.021206112081567925,
      "loss": 0.2319,
      "num_input_tokens_seen": 16412592,
      "step": 77775
    },
    {
      "epoch": 8.556655665566556,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.021204801047578315,
      "loss": 0.2298,
      "num_input_tokens_seen": 16413680,
      "step": 77780
    },
    {
      "epoch": 8.557205720572057,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.021203489956402033,
      "loss": 0.2324,
      "num_input_tokens_seen": 16414768,
      "step": 77785
    },
    {
      "epoch": 8.557755775577558,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.021202178808051156,
      "loss": 0.2319,
      "num_input_tokens_seen": 16415728,
      "step": 77790
    },
    {
      "epoch": 8.558305830583059,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02120086760253778,
      "loss": 0.2325,
      "num_input_tokens_seen": 16416880,
      "step": 77795
    },
    {
      "epoch": 8.558855885588558,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.021199556339873978,
      "loss": 0.2304,
      "num_input_tokens_seen": 16417968,
      "step": 77800
    },
    {
      "epoch": 8.55940594059406,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02119824502007184,
      "loss": 0.2304,
      "num_input_tokens_seen": 16419056,
      "step": 77805
    },
    {
      "epoch": 8.55995599559956,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.021196933643143453,
      "loss": 0.2304,
      "num_input_tokens_seen": 16420112,
      "step": 77810
    },
    {
      "epoch": 8.56050605060506,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.0211956222091009,
      "loss": 0.233,
      "num_input_tokens_seen": 16421136,
      "step": 77815
    },
    {
      "epoch": 8.561056105610561,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.021194310717956275,
      "loss": 0.2309,
      "num_input_tokens_seen": 16422224,
      "step": 77820
    },
    {
      "epoch": 8.561606160616062,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.021192999169721657,
      "loss": 0.2304,
      "num_input_tokens_seen": 16423344,
      "step": 77825
    },
    {
      "epoch": 8.562156215621561,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.021191687564409138,
      "loss": 0.234,
      "num_input_tokens_seen": 16424336,
      "step": 77830
    },
    {
      "epoch": 8.562706270627062,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.0211903759020308,
      "loss": 0.2293,
      "num_input_tokens_seen": 16425424,
      "step": 77835
    },
    {
      "epoch": 8.563256325632564,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.021189064182598746,
      "loss": 0.2324,
      "num_input_tokens_seen": 16426480,
      "step": 77840
    },
    {
      "epoch": 8.563806380638063,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.021187752406125058,
      "loss": 0.2319,
      "num_input_tokens_seen": 16427568,
      "step": 77845
    },
    {
      "epoch": 8.564356435643564,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021186440572621822,
      "loss": 0.2314,
      "num_input_tokens_seen": 16428592,
      "step": 77850
    },
    {
      "epoch": 8.564906490649065,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.021185128682101134,
      "loss": 0.2293,
      "num_input_tokens_seen": 16429680,
      "step": 77855
    },
    {
      "epoch": 8.565456545654566,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02118381673457508,
      "loss": 0.2324,
      "num_input_tokens_seen": 16430704,
      "step": 77860
    },
    {
      "epoch": 8.566006600660065,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02118250473005575,
      "loss": 0.2324,
      "num_input_tokens_seen": 16431792,
      "step": 77865
    },
    {
      "epoch": 8.566556655665567,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.021181192668555247,
      "loss": 0.2288,
      "num_input_tokens_seen": 16432848,
      "step": 77870
    },
    {
      "epoch": 8.567106710671068,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.021179880550085656,
      "loss": 0.2314,
      "num_input_tokens_seen": 16433936,
      "step": 77875
    },
    {
      "epoch": 8.567656765676567,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.02117856837465907,
      "loss": 0.2303,
      "num_input_tokens_seen": 16435024,
      "step": 77880
    },
    {
      "epoch": 8.568206820682068,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.021177256142287586,
      "loss": 0.2319,
      "num_input_tokens_seen": 16436080,
      "step": 77885
    },
    {
      "epoch": 8.56875687568757,
      "grad_norm": 0.009765625,
      "learning_rate": 0.021175943852983294,
      "loss": 0.2298,
      "num_input_tokens_seen": 16437200,
      "step": 77890
    },
    {
      "epoch": 8.569306930693068,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.021174631506758294,
      "loss": 0.2314,
      "num_input_tokens_seen": 16438192,
      "step": 77895
    },
    {
      "epoch": 8.56985698569857,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02117331910362468,
      "loss": 0.2309,
      "num_input_tokens_seen": 16439152,
      "step": 77900
    },
    {
      "epoch": 8.57040704070407,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.02117200664359454,
      "loss": 0.2314,
      "num_input_tokens_seen": 16440240,
      "step": 77905
    },
    {
      "epoch": 8.570957095709572,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02117069412667998,
      "loss": 0.232,
      "num_input_tokens_seen": 16441264,
      "step": 77910
    },
    {
      "epoch": 8.571507150715071,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.021169381552893093,
      "loss": 0.2346,
      "num_input_tokens_seen": 16442320,
      "step": 77915
    },
    {
      "epoch": 8.572057205720572,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.021168068922245976,
      "loss": 0.2309,
      "num_input_tokens_seen": 16443408,
      "step": 77920
    },
    {
      "epoch": 8.572607260726073,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021166756234750728,
      "loss": 0.2319,
      "num_input_tokens_seen": 16444464,
      "step": 77925
    },
    {
      "epoch": 8.573157315731573,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.021165443490419444,
      "loss": 0.2293,
      "num_input_tokens_seen": 16445520,
      "step": 77930
    },
    {
      "epoch": 8.573707370737074,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021164130689264225,
      "loss": 0.2308,
      "num_input_tokens_seen": 16446576,
      "step": 77935
    },
    {
      "epoch": 8.574257425742575,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.02116281783129718,
      "loss": 0.2335,
      "num_input_tokens_seen": 16447696,
      "step": 77940
    },
    {
      "epoch": 8.574807480748074,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.021161504916530398,
      "loss": 0.2319,
      "num_input_tokens_seen": 16448720,
      "step": 77945
    },
    {
      "epoch": 8.575357535753575,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021160191944975974,
      "loss": 0.2309,
      "num_input_tokens_seen": 16449808,
      "step": 77950
    },
    {
      "epoch": 8.575907590759076,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.021158878916646022,
      "loss": 0.2319,
      "num_input_tokens_seen": 16450864,
      "step": 77955
    },
    {
      "epoch": 8.576457645764577,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.021157565831552635,
      "loss": 0.2303,
      "num_input_tokens_seen": 16451952,
      "step": 77960
    },
    {
      "epoch": 8.577007700770077,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.021156252689707923,
      "loss": 0.2314,
      "num_input_tokens_seen": 16453040,
      "step": 77965
    },
    {
      "epoch": 8.577557755775578,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.021154939491123983,
      "loss": 0.2313,
      "num_input_tokens_seen": 16454000,
      "step": 77970
    },
    {
      "epoch": 8.578107810781079,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.021153626235812916,
      "loss": 0.2314,
      "num_input_tokens_seen": 16455024,
      "step": 77975
    },
    {
      "epoch": 8.578657865786578,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02115231292378683,
      "loss": 0.2303,
      "num_input_tokens_seen": 16456112,
      "step": 77980
    },
    {
      "epoch": 8.57920792079208,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02115099955505783,
      "loss": 0.2324,
      "num_input_tokens_seen": 16457168,
      "step": 77985
    },
    {
      "epoch": 8.57975797579758,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.021149686129638016,
      "loss": 0.2314,
      "num_input_tokens_seen": 16458224,
      "step": 77990
    },
    {
      "epoch": 8.58030803080308,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.0211483726475395,
      "loss": 0.2319,
      "num_input_tokens_seen": 16459312,
      "step": 77995
    },
    {
      "epoch": 8.58085808580858,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02114705910877438,
      "loss": 0.2303,
      "num_input_tokens_seen": 16460368,
      "step": 78000
    },
    {
      "epoch": 8.581408140814082,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02114574551335477,
      "loss": 0.2309,
      "num_input_tokens_seen": 16461488,
      "step": 78005
    },
    {
      "epoch": 8.581958195819581,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.021144431861292768,
      "loss": 0.2329,
      "num_input_tokens_seen": 16462512,
      "step": 78010
    },
    {
      "epoch": 8.582508250825082,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02114311815260049,
      "loss": 0.2303,
      "num_input_tokens_seen": 16463568,
      "step": 78015
    },
    {
      "epoch": 8.583058305830583,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.021141804387290034,
      "loss": 0.2308,
      "num_input_tokens_seen": 16464656,
      "step": 78020
    },
    {
      "epoch": 8.583608360836084,
      "grad_norm": 0.00058746337890625,
      "learning_rate": 0.021140490565373517,
      "loss": 0.2313,
      "num_input_tokens_seen": 16465776,
      "step": 78025
    },
    {
      "epoch": 8.584158415841584,
      "grad_norm": 0.000415802001953125,
      "learning_rate": 0.02113917668686304,
      "loss": 0.2319,
      "num_input_tokens_seen": 16466864,
      "step": 78030
    },
    {
      "epoch": 8.584708470847085,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02113786275177073,
      "loss": 0.2329,
      "num_input_tokens_seen": 16467888,
      "step": 78035
    },
    {
      "epoch": 8.585258525852586,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.021136548760108672,
      "loss": 0.2314,
      "num_input_tokens_seen": 16468912,
      "step": 78040
    },
    {
      "epoch": 8.585808580858085,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.021135234711888995,
      "loss": 0.2298,
      "num_input_tokens_seen": 16470032,
      "step": 78045
    },
    {
      "epoch": 8.586358635863586,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.0211339206071238,
      "loss": 0.2324,
      "num_input_tokens_seen": 16471088,
      "step": 78050
    },
    {
      "epoch": 8.586908690869087,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.021132606445825205,
      "loss": 0.2303,
      "num_input_tokens_seen": 16472144,
      "step": 78055
    },
    {
      "epoch": 8.587458745874587,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.021131292228005322,
      "loss": 0.2303,
      "num_input_tokens_seen": 16473264,
      "step": 78060
    },
    {
      "epoch": 8.588008800880088,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.021129977953676254,
      "loss": 0.2324,
      "num_input_tokens_seen": 16474288,
      "step": 78065
    },
    {
      "epoch": 8.588558855885589,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02112866362285012,
      "loss": 0.2308,
      "num_input_tokens_seen": 16475312,
      "step": 78070
    },
    {
      "epoch": 8.589108910891088,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.021127349235539043,
      "loss": 0.2324,
      "num_input_tokens_seen": 16476336,
      "step": 78075
    },
    {
      "epoch": 8.58965896589659,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02112603479175512,
      "loss": 0.2308,
      "num_input_tokens_seen": 16477392,
      "step": 78080
    },
    {
      "epoch": 8.59020902090209,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.021124720291510482,
      "loss": 0.2298,
      "num_input_tokens_seen": 16478448,
      "step": 78085
    },
    {
      "epoch": 8.590759075907592,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02112340573481723,
      "loss": 0.2308,
      "num_input_tokens_seen": 16479536,
      "step": 78090
    },
    {
      "epoch": 8.591309130913091,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.021122091121687482,
      "loss": 0.2308,
      "num_input_tokens_seen": 16480592,
      "step": 78095
    },
    {
      "epoch": 8.591859185918592,
      "grad_norm": 0.009765625,
      "learning_rate": 0.021120776452133363,
      "loss": 0.2314,
      "num_input_tokens_seen": 16481712,
      "step": 78100
    },
    {
      "epoch": 8.592409240924093,
      "grad_norm": 0.000675201416015625,
      "learning_rate": 0.021119461726166987,
      "loss": 0.2314,
      "num_input_tokens_seen": 16482768,
      "step": 78105
    },
    {
      "epoch": 8.592959295929592,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.021118146943800462,
      "loss": 0.2319,
      "num_input_tokens_seen": 16483920,
      "step": 78110
    },
    {
      "epoch": 8.593509350935093,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02111683210504592,
      "loss": 0.2314,
      "num_input_tokens_seen": 16485008,
      "step": 78115
    },
    {
      "epoch": 8.594059405940595,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.021115517209915464,
      "loss": 0.2309,
      "num_input_tokens_seen": 16486128,
      "step": 78120
    },
    {
      "epoch": 8.594609460946094,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.021114202258421224,
      "loss": 0.2309,
      "num_input_tokens_seen": 16487152,
      "step": 78125
    },
    {
      "epoch": 8.595159515951595,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.021112887250575318,
      "loss": 0.2329,
      "num_input_tokens_seen": 16488176,
      "step": 78130
    },
    {
      "epoch": 8.595709570957096,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.02111157218638986,
      "loss": 0.234,
      "num_input_tokens_seen": 16489200,
      "step": 78135
    },
    {
      "epoch": 8.596259625962595,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.021110257065876974,
      "loss": 0.2329,
      "num_input_tokens_seen": 16490288,
      "step": 78140
    },
    {
      "epoch": 8.596809680968097,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02110894188904878,
      "loss": 0.2303,
      "num_input_tokens_seen": 16491344,
      "step": 78145
    },
    {
      "epoch": 8.597359735973598,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.021107626655917405,
      "loss": 0.2298,
      "num_input_tokens_seen": 16492432,
      "step": 78150
    },
    {
      "epoch": 8.597909790979099,
      "grad_norm": 0.000782012939453125,
      "learning_rate": 0.02110631136649496,
      "loss": 0.2351,
      "num_input_tokens_seen": 16493552,
      "step": 78155
    },
    {
      "epoch": 8.598459845984598,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021104996020793576,
      "loss": 0.2314,
      "num_input_tokens_seen": 16494576,
      "step": 78160
    },
    {
      "epoch": 8.599009900990099,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021103680618825373,
      "loss": 0.2324,
      "num_input_tokens_seen": 16495568,
      "step": 78165
    },
    {
      "epoch": 8.5995599559956,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.021102365160602474,
      "loss": 0.2303,
      "num_input_tokens_seen": 16496656,
      "step": 78170
    },
    {
      "epoch": 8.6001100110011,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.021101049646137005,
      "loss": 0.2314,
      "num_input_tokens_seen": 16497712,
      "step": 78175
    },
    {
      "epoch": 8.6006600660066,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.021099734075441086,
      "loss": 0.2314,
      "num_input_tokens_seen": 16498800,
      "step": 78180
    },
    {
      "epoch": 8.601210121012102,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.021098418448526845,
      "loss": 0.2319,
      "num_input_tokens_seen": 16499792,
      "step": 78185
    },
    {
      "epoch": 8.601760176017601,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02109710276540641,
      "loss": 0.2309,
      "num_input_tokens_seen": 16500848,
      "step": 78190
    },
    {
      "epoch": 8.602310231023102,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.021095787026091904,
      "loss": 0.2298,
      "num_input_tokens_seen": 16501872,
      "step": 78195
    },
    {
      "epoch": 8.602860286028603,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021094471230595456,
      "loss": 0.2319,
      "num_input_tokens_seen": 16502992,
      "step": 78200
    },
    {
      "epoch": 8.603410341034103,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.02109315537892919,
      "loss": 0.2324,
      "num_input_tokens_seen": 16504048,
      "step": 78205
    },
    {
      "epoch": 8.603960396039604,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02109183947110523,
      "loss": 0.2324,
      "num_input_tokens_seen": 16505104,
      "step": 78210
    },
    {
      "epoch": 8.604510451045105,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.02109052350713571,
      "loss": 0.2319,
      "num_input_tokens_seen": 16506160,
      "step": 78215
    },
    {
      "epoch": 8.605060506050606,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.02108920748703276,
      "loss": 0.2319,
      "num_input_tokens_seen": 16507248,
      "step": 78220
    },
    {
      "epoch": 8.605610561056105,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.021087891410808506,
      "loss": 0.2308,
      "num_input_tokens_seen": 16508304,
      "step": 78225
    },
    {
      "epoch": 8.606160616061606,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02108657527847508,
      "loss": 0.2314,
      "num_input_tokens_seen": 16509328,
      "step": 78230
    },
    {
      "epoch": 8.606710671067107,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.021085259090044606,
      "loss": 0.2324,
      "num_input_tokens_seen": 16510416,
      "step": 78235
    },
    {
      "epoch": 8.607260726072607,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.021083942845529217,
      "loss": 0.2298,
      "num_input_tokens_seen": 16511472,
      "step": 78240
    },
    {
      "epoch": 8.607810781078108,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.021082626544941054,
      "loss": 0.2308,
      "num_input_tokens_seen": 16512528,
      "step": 78245
    },
    {
      "epoch": 8.608360836083609,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.021081310188292242,
      "loss": 0.2314,
      "num_input_tokens_seen": 16513520,
      "step": 78250
    },
    {
      "epoch": 8.608910891089108,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.021079993775594906,
      "loss": 0.2303,
      "num_input_tokens_seen": 16514640,
      "step": 78255
    },
    {
      "epoch": 8.60946094609461,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.021078677306861186,
      "loss": 0.2298,
      "num_input_tokens_seen": 16515664,
      "step": 78260
    },
    {
      "epoch": 8.61001100110011,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.021077360782103212,
      "loss": 0.2324,
      "num_input_tokens_seen": 16516784,
      "step": 78265
    },
    {
      "epoch": 8.61056105610561,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.021076044201333122,
      "loss": 0.2319,
      "num_input_tokens_seen": 16517872,
      "step": 78270
    },
    {
      "epoch": 8.61111111111111,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02107472756456305,
      "loss": 0.2319,
      "num_input_tokens_seen": 16518896,
      "step": 78275
    },
    {
      "epoch": 8.611661166116612,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02107341087180513,
      "loss": 0.2308,
      "num_input_tokens_seen": 16519952,
      "step": 78280
    },
    {
      "epoch": 8.612211221122113,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.021072094123071487,
      "loss": 0.2303,
      "num_input_tokens_seen": 16521008,
      "step": 78285
    },
    {
      "epoch": 8.612761276127612,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.021070777318374274,
      "loss": 0.2329,
      "num_input_tokens_seen": 16522064,
      "step": 78290
    },
    {
      "epoch": 8.613311331133113,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02106946045772562,
      "loss": 0.2303,
      "num_input_tokens_seen": 16523152,
      "step": 78295
    },
    {
      "epoch": 8.613861386138614,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.02106814354113766,
      "loss": 0.2303,
      "num_input_tokens_seen": 16524240,
      "step": 78300
    },
    {
      "epoch": 8.614411441144114,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.02106682656862253,
      "loss": 0.2309,
      "num_input_tokens_seen": 16525296,
      "step": 78305
    },
    {
      "epoch": 8.614961496149615,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.021065509540192377,
      "loss": 0.2303,
      "num_input_tokens_seen": 16526320,
      "step": 78310
    },
    {
      "epoch": 8.615511551155116,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.021064192455859325,
      "loss": 0.2303,
      "num_input_tokens_seen": 16527344,
      "step": 78315
    },
    {
      "epoch": 8.616061606160617,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.021062875315635526,
      "loss": 0.2298,
      "num_input_tokens_seen": 16528432,
      "step": 78320
    },
    {
      "epoch": 8.616611661166116,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.021061558119533114,
      "loss": 0.2303,
      "num_input_tokens_seen": 16529488,
      "step": 78325
    },
    {
      "epoch": 8.617161716171617,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02106024086756423,
      "loss": 0.2298,
      "num_input_tokens_seen": 16530544,
      "step": 78330
    },
    {
      "epoch": 8.617711771177119,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02105892355974101,
      "loss": 0.2293,
      "num_input_tokens_seen": 16531664,
      "step": 78335
    },
    {
      "epoch": 8.618261826182618,
      "grad_norm": 0.009765625,
      "learning_rate": 0.021057606196075603,
      "loss": 0.233,
      "num_input_tokens_seen": 16532816,
      "step": 78340
    },
    {
      "epoch": 8.618811881188119,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.021056288776580147,
      "loss": 0.2314,
      "num_input_tokens_seen": 16533936,
      "step": 78345
    },
    {
      "epoch": 8.61936193619362,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.02105497130126678,
      "loss": 0.2314,
      "num_input_tokens_seen": 16534928,
      "step": 78350
    },
    {
      "epoch": 8.61991199119912,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02105365377014765,
      "loss": 0.2314,
      "num_input_tokens_seen": 16535952,
      "step": 78355
    },
    {
      "epoch": 8.62046204620462,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.021052336183234897,
      "loss": 0.2308,
      "num_input_tokens_seen": 16536976,
      "step": 78360
    },
    {
      "epoch": 8.621012101210122,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02105101854054067,
      "loss": 0.2298,
      "num_input_tokens_seen": 16538064,
      "step": 78365
    },
    {
      "epoch": 8.62156215621562,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0210497008420771,
      "loss": 0.2324,
      "num_input_tokens_seen": 16539088,
      "step": 78370
    },
    {
      "epoch": 8.622112211221122,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021048383087856348,
      "loss": 0.2309,
      "num_input_tokens_seen": 16540144,
      "step": 78375
    },
    {
      "epoch": 8.622662266226623,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021047065277890548,
      "loss": 0.2309,
      "num_input_tokens_seen": 16541168,
      "step": 78380
    },
    {
      "epoch": 8.623212321232124,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.021045747412191847,
      "loss": 0.2314,
      "num_input_tokens_seen": 16542224,
      "step": 78385
    },
    {
      "epoch": 8.623762376237623,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.0210444294907724,
      "loss": 0.2314,
      "num_input_tokens_seen": 16543312,
      "step": 78390
    },
    {
      "epoch": 8.624312431243125,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.021043111513644342,
      "loss": 0.2314,
      "num_input_tokens_seen": 16544400,
      "step": 78395
    },
    {
      "epoch": 8.624862486248626,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.021041793480819825,
      "loss": 0.2319,
      "num_input_tokens_seen": 16545488,
      "step": 78400
    },
    {
      "epoch": 8.625412541254125,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.021040475392310996,
      "loss": 0.2309,
      "num_input_tokens_seen": 16546576,
      "step": 78405
    },
    {
      "epoch": 8.625962596259626,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.021039157248130005,
      "loss": 0.2293,
      "num_input_tokens_seen": 16547568,
      "step": 78410
    },
    {
      "epoch": 8.626512651265127,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.021037839048289,
      "loss": 0.2303,
      "num_input_tokens_seen": 16548592,
      "step": 78415
    },
    {
      "epoch": 8.627062706270626,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.021036520792800127,
      "loss": 0.2324,
      "num_input_tokens_seen": 16549680,
      "step": 78420
    },
    {
      "epoch": 8.627612761276128,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.021035202481675543,
      "loss": 0.2324,
      "num_input_tokens_seen": 16550736,
      "step": 78425
    },
    {
      "epoch": 8.628162816281629,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.02103388411492739,
      "loss": 0.2324,
      "num_input_tokens_seen": 16551888,
      "step": 78430
    },
    {
      "epoch": 8.628712871287128,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.021032565692567822,
      "loss": 0.2283,
      "num_input_tokens_seen": 16552912,
      "step": 78435
    },
    {
      "epoch": 8.629262926292629,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.021031247214608997,
      "loss": 0.2324,
      "num_input_tokens_seen": 16553968,
      "step": 78440
    },
    {
      "epoch": 8.62981298129813,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.021029928681063053,
      "loss": 0.2278,
      "num_input_tokens_seen": 16554992,
      "step": 78445
    },
    {
      "epoch": 8.630363036303631,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.021028610091942155,
      "loss": 0.2303,
      "num_input_tokens_seen": 16556048,
      "step": 78450
    },
    {
      "epoch": 8.63091309130913,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02102729144725845,
      "loss": 0.2304,
      "num_input_tokens_seen": 16557168,
      "step": 78455
    },
    {
      "epoch": 8.631463146314632,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.021025972747024085,
      "loss": 0.2356,
      "num_input_tokens_seen": 16558192,
      "step": 78460
    },
    {
      "epoch": 8.632013201320133,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.021024653991251228,
      "loss": 0.2304,
      "num_input_tokens_seen": 16559280,
      "step": 78465
    },
    {
      "epoch": 8.632563256325632,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.021023335179952023,
      "loss": 0.2309,
      "num_input_tokens_seen": 16560272,
      "step": 78470
    },
    {
      "epoch": 8.633113311331133,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.021022016313138627,
      "loss": 0.2324,
      "num_input_tokens_seen": 16561264,
      "step": 78475
    },
    {
      "epoch": 8.633663366336634,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.021020697390823198,
      "loss": 0.2325,
      "num_input_tokens_seen": 16562288,
      "step": 78480
    },
    {
      "epoch": 8.634213421342134,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.021019378413017892,
      "loss": 0.234,
      "num_input_tokens_seen": 16563376,
      "step": 78485
    },
    {
      "epoch": 8.634763476347635,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.021018059379734864,
      "loss": 0.2324,
      "num_input_tokens_seen": 16564336,
      "step": 78490
    },
    {
      "epoch": 8.635313531353136,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02101674029098627,
      "loss": 0.2308,
      "num_input_tokens_seen": 16565296,
      "step": 78495
    },
    {
      "epoch": 8.635863586358635,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.02101542114678426,
      "loss": 0.2309,
      "num_input_tokens_seen": 16566320,
      "step": 78500
    },
    {
      "epoch": 8.636413641364136,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02101410194714101,
      "loss": 0.2293,
      "num_input_tokens_seen": 16567376,
      "step": 78505
    },
    {
      "epoch": 8.636963696369637,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.021012782692068667,
      "loss": 0.233,
      "num_input_tokens_seen": 16568400,
      "step": 78510
    },
    {
      "epoch": 8.637513751375138,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.021011463381579384,
      "loss": 0.233,
      "num_input_tokens_seen": 16569488,
      "step": 78515
    },
    {
      "epoch": 8.638063806380638,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.021010144015685333,
      "loss": 0.2283,
      "num_input_tokens_seen": 16570608,
      "step": 78520
    },
    {
      "epoch": 8.638613861386139,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.021008824594398668,
      "loss": 0.2324,
      "num_input_tokens_seen": 16571664,
      "step": 78525
    },
    {
      "epoch": 8.63916391639164,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.021007505117731546,
      "loss": 0.2293,
      "num_input_tokens_seen": 16572688,
      "step": 78530
    },
    {
      "epoch": 8.63971397139714,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.02100618558569614,
      "loss": 0.2314,
      "num_input_tokens_seen": 16573776,
      "step": 78535
    },
    {
      "epoch": 8.64026402640264,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.021004865998304596,
      "loss": 0.2319,
      "num_input_tokens_seen": 16574896,
      "step": 78540
    },
    {
      "epoch": 8.640814081408141,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.021003546355569086,
      "loss": 0.2293,
      "num_input_tokens_seen": 16575952,
      "step": 78545
    },
    {
      "epoch": 8.64136413641364,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.02100222665750177,
      "loss": 0.2346,
      "num_input_tokens_seen": 16577008,
      "step": 78550
    },
    {
      "epoch": 8.641914191419142,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.021000906904114817,
      "loss": 0.2304,
      "num_input_tokens_seen": 16578032,
      "step": 78555
    },
    {
      "epoch": 8.642464246424643,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.020999587095420377,
      "loss": 0.2324,
      "num_input_tokens_seen": 16579056,
      "step": 78560
    },
    {
      "epoch": 8.643014301430142,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.020998267231430622,
      "loss": 0.2299,
      "num_input_tokens_seen": 16580016,
      "step": 78565
    },
    {
      "epoch": 8.643564356435643,
      "grad_norm": 0.002899169921875,
      "learning_rate": 0.020996947312157718,
      "loss": 0.2298,
      "num_input_tokens_seen": 16581104,
      "step": 78570
    },
    {
      "epoch": 8.644114411441144,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02099562733761383,
      "loss": 0.233,
      "num_input_tokens_seen": 16582192,
      "step": 78575
    },
    {
      "epoch": 8.644664466446645,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02099430730781112,
      "loss": 0.2309,
      "num_input_tokens_seen": 16583248,
      "step": 78580
    },
    {
      "epoch": 8.645214521452145,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.020992987222761758,
      "loss": 0.2299,
      "num_input_tokens_seen": 16584272,
      "step": 78585
    },
    {
      "epoch": 8.645764576457646,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.0209916670824779,
      "loss": 0.2309,
      "num_input_tokens_seen": 16585360,
      "step": 78590
    },
    {
      "epoch": 8.646314631463147,
      "grad_norm": 0.0030059814453125,
      "learning_rate": 0.020990346886971732,
      "loss": 0.2335,
      "num_input_tokens_seen": 16586416,
      "step": 78595
    },
    {
      "epoch": 8.646864686468646,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020989026636255407,
      "loss": 0.234,
      "num_input_tokens_seen": 16587472,
      "step": 78600
    },
    {
      "epoch": 8.647414741474147,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0209877063303411,
      "loss": 0.232,
      "num_input_tokens_seen": 16588592,
      "step": 78605
    },
    {
      "epoch": 8.647964796479648,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.020986385969240975,
      "loss": 0.2293,
      "num_input_tokens_seen": 16589680,
      "step": 78610
    },
    {
      "epoch": 8.648514851485148,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.0209850655529672,
      "loss": 0.2324,
      "num_input_tokens_seen": 16590736,
      "step": 78615
    },
    {
      "epoch": 8.649064906490649,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.02098374508153195,
      "loss": 0.2314,
      "num_input_tokens_seen": 16591792,
      "step": 78620
    },
    {
      "epoch": 8.64961496149615,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02098242455494739,
      "loss": 0.2314,
      "num_input_tokens_seen": 16592848,
      "step": 78625
    },
    {
      "epoch": 8.65016501650165,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0209811039732257,
      "loss": 0.2314,
      "num_input_tokens_seen": 16593840,
      "step": 78630
    },
    {
      "epoch": 8.65071507150715,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.02097978333637904,
      "loss": 0.232,
      "num_input_tokens_seen": 16594928,
      "step": 78635
    },
    {
      "epoch": 8.651265126512651,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.020978462644419587,
      "loss": 0.2335,
      "num_input_tokens_seen": 16595952,
      "step": 78640
    },
    {
      "epoch": 8.651815181518153,
      "grad_norm": 0.0006256103515625,
      "learning_rate": 0.020977141897359518,
      "loss": 0.2299,
      "num_input_tokens_seen": 16596976,
      "step": 78645
    },
    {
      "epoch": 8.652365236523652,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.020975821095210994,
      "loss": 0.232,
      "num_input_tokens_seen": 16598000,
      "step": 78650
    },
    {
      "epoch": 8.652915291529153,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.020974500237986193,
      "loss": 0.2325,
      "num_input_tokens_seen": 16599056,
      "step": 78655
    },
    {
      "epoch": 8.653465346534654,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020973179325697294,
      "loss": 0.2319,
      "num_input_tokens_seen": 16600208,
      "step": 78660
    },
    {
      "epoch": 8.654015401540153,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.020971858358356466,
      "loss": 0.234,
      "num_input_tokens_seen": 16601264,
      "step": 78665
    },
    {
      "epoch": 8.654565456545654,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.020970537335975887,
      "loss": 0.2283,
      "num_input_tokens_seen": 16602352,
      "step": 78670
    },
    {
      "epoch": 8.655115511551156,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02096921625856773,
      "loss": 0.2319,
      "num_input_tokens_seen": 16603472,
      "step": 78675
    },
    {
      "epoch": 8.655665566556655,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.02096789512614417,
      "loss": 0.2314,
      "num_input_tokens_seen": 16604560,
      "step": 78680
    },
    {
      "epoch": 8.656215621562156,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020966573938717384,
      "loss": 0.2314,
      "num_input_tokens_seen": 16605584,
      "step": 78685
    },
    {
      "epoch": 8.656765676567657,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.020965252696299546,
      "loss": 0.233,
      "num_input_tokens_seen": 16606608,
      "step": 78690
    },
    {
      "epoch": 8.657315731573158,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02096393139890284,
      "loss": 0.2329,
      "num_input_tokens_seen": 16607664,
      "step": 78695
    },
    {
      "epoch": 8.657865786578657,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.020962610046539442,
      "loss": 0.2298,
      "num_input_tokens_seen": 16608720,
      "step": 78700
    },
    {
      "epoch": 8.658415841584159,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020961288639221524,
      "loss": 0.2309,
      "num_input_tokens_seen": 16609744,
      "step": 78705
    },
    {
      "epoch": 8.65896589658966,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02095996717696127,
      "loss": 0.2335,
      "num_input_tokens_seen": 16610800,
      "step": 78710
    },
    {
      "epoch": 8.659515951595159,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.02095864565977086,
      "loss": 0.2324,
      "num_input_tokens_seen": 16611792,
      "step": 78715
    },
    {
      "epoch": 8.66006600660066,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02095732408766247,
      "loss": 0.2335,
      "num_input_tokens_seen": 16612784,
      "step": 78720
    },
    {
      "epoch": 8.660616061606161,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.020956002460648288,
      "loss": 0.2309,
      "num_input_tokens_seen": 16613840,
      "step": 78725
    },
    {
      "epoch": 8.66116611661166,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.020954680778740485,
      "loss": 0.2314,
      "num_input_tokens_seen": 16614864,
      "step": 78730
    },
    {
      "epoch": 8.661716171617162,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.020953359041951242,
      "loss": 0.2314,
      "num_input_tokens_seen": 16615920,
      "step": 78735
    },
    {
      "epoch": 8.662266226622663,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.020952037250292755,
      "loss": 0.2309,
      "num_input_tokens_seen": 16616976,
      "step": 78740
    },
    {
      "epoch": 8.662816281628164,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02095071540377719,
      "loss": 0.2319,
      "num_input_tokens_seen": 16618000,
      "step": 78745
    },
    {
      "epoch": 8.663366336633663,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02094939350241674,
      "loss": 0.2314,
      "num_input_tokens_seen": 16619024,
      "step": 78750
    },
    {
      "epoch": 8.663916391639164,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.020948071546223582,
      "loss": 0.2309,
      "num_input_tokens_seen": 16620048,
      "step": 78755
    },
    {
      "epoch": 8.664466446644665,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.020946749535209905,
      "loss": 0.2298,
      "num_input_tokens_seen": 16621136,
      "step": 78760
    },
    {
      "epoch": 8.665016501650165,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.020945427469387892,
      "loss": 0.2319,
      "num_input_tokens_seen": 16622192,
      "step": 78765
    },
    {
      "epoch": 8.665566556655666,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.020944105348769726,
      "loss": 0.2298,
      "num_input_tokens_seen": 16623152,
      "step": 78770
    },
    {
      "epoch": 8.666116611661167,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.020942783173367587,
      "loss": 0.2283,
      "num_input_tokens_seen": 16624240,
      "step": 78775
    },
    {
      "epoch": 8.666666666666666,
      "grad_norm": 0.002838134765625,
      "learning_rate": 0.020941460943193674,
      "loss": 0.2309,
      "num_input_tokens_seen": 16625296,
      "step": 78780
    },
    {
      "epoch": 8.667216721672167,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.02094013865826017,
      "loss": 0.2299,
      "num_input_tokens_seen": 16626448,
      "step": 78785
    },
    {
      "epoch": 8.667766776677668,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.020938816318579252,
      "loss": 0.234,
      "num_input_tokens_seen": 16627408,
      "step": 78790
    },
    {
      "epoch": 8.668316831683168,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.020937493924163116,
      "loss": 0.2315,
      "num_input_tokens_seen": 16628496,
      "step": 78795
    },
    {
      "epoch": 8.668866886688669,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.020936171475023944,
      "loss": 0.2293,
      "num_input_tokens_seen": 16629488,
      "step": 78800
    },
    {
      "epoch": 8.66941694169417,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.02093484897117393,
      "loss": 0.2268,
      "num_input_tokens_seen": 16630576,
      "step": 78805
    },
    {
      "epoch": 8.66996699669967,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02093352641262526,
      "loss": 0.234,
      "num_input_tokens_seen": 16631632,
      "step": 78810
    },
    {
      "epoch": 8.67051705170517,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02093220379939013,
      "loss": 0.2309,
      "num_input_tokens_seen": 16632720,
      "step": 78815
    },
    {
      "epoch": 8.671067106710671,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.020930881131480717,
      "loss": 0.2294,
      "num_input_tokens_seen": 16633808,
      "step": 78820
    },
    {
      "epoch": 8.671617161716172,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.020929558408909223,
      "loss": 0.2315,
      "num_input_tokens_seen": 16634864,
      "step": 78825
    },
    {
      "epoch": 8.672167216721672,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.020928235631687834,
      "loss": 0.2299,
      "num_input_tokens_seen": 16635920,
      "step": 78830
    },
    {
      "epoch": 8.672717271727173,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.02092691279982874,
      "loss": 0.2299,
      "num_input_tokens_seen": 16637008,
      "step": 78835
    },
    {
      "epoch": 8.673267326732674,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.020925589913344138,
      "loss": 0.2304,
      "num_input_tokens_seen": 16638096,
      "step": 78840
    },
    {
      "epoch": 8.673817381738173,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.020924266972246214,
      "loss": 0.2335,
      "num_input_tokens_seen": 16639152,
      "step": 78845
    },
    {
      "epoch": 8.674367436743674,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.020922943976547162,
      "loss": 0.2324,
      "num_input_tokens_seen": 16640208,
      "step": 78850
    },
    {
      "epoch": 8.674917491749175,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02092162092625918,
      "loss": 0.232,
      "num_input_tokens_seen": 16641232,
      "step": 78855
    },
    {
      "epoch": 8.675467546754675,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02092029782139446,
      "loss": 0.232,
      "num_input_tokens_seen": 16642352,
      "step": 78860
    },
    {
      "epoch": 8.676017601760176,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.020918974661965198,
      "loss": 0.2319,
      "num_input_tokens_seen": 16643376,
      "step": 78865
    },
    {
      "epoch": 8.676567656765677,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020917651447983583,
      "loss": 0.2299,
      "num_input_tokens_seen": 16644400,
      "step": 78870
    },
    {
      "epoch": 8.677117711771178,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020916328179461815,
      "loss": 0.2309,
      "num_input_tokens_seen": 16645456,
      "step": 78875
    },
    {
      "epoch": 8.677667766776677,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.020915004856412088,
      "loss": 0.2341,
      "num_input_tokens_seen": 16646544,
      "step": 78880
    },
    {
      "epoch": 8.678217821782178,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.0209136814788466,
      "loss": 0.2325,
      "num_input_tokens_seen": 16647600,
      "step": 78885
    },
    {
      "epoch": 8.67876787678768,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02091235804677755,
      "loss": 0.2325,
      "num_input_tokens_seen": 16648720,
      "step": 78890
    },
    {
      "epoch": 8.679317931793179,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.020911034560217128,
      "loss": 0.2304,
      "num_input_tokens_seen": 16649744,
      "step": 78895
    },
    {
      "epoch": 8.67986798679868,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.020909711019177542,
      "loss": 0.2288,
      "num_input_tokens_seen": 16650768,
      "step": 78900
    },
    {
      "epoch": 8.680418041804181,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.020908387423670984,
      "loss": 0.2273,
      "num_input_tokens_seen": 16651888,
      "step": 78905
    },
    {
      "epoch": 8.68096809680968,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.020907063773709652,
      "loss": 0.2278,
      "num_input_tokens_seen": 16652880,
      "step": 78910
    },
    {
      "epoch": 8.681518151815181,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.020905740069305755,
      "loss": 0.233,
      "num_input_tokens_seen": 16653968,
      "step": 78915
    },
    {
      "epoch": 8.682068206820682,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.020904416310471477,
      "loss": 0.2303,
      "num_input_tokens_seen": 16655024,
      "step": 78920
    },
    {
      "epoch": 8.682618261826182,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.020903092497219028,
      "loss": 0.2304,
      "num_input_tokens_seen": 16656080,
      "step": 78925
    },
    {
      "epoch": 8.683168316831683,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02090176862956061,
      "loss": 0.2325,
      "num_input_tokens_seen": 16657136,
      "step": 78930
    },
    {
      "epoch": 8.683718371837184,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.020900444707508424,
      "loss": 0.2361,
      "num_input_tokens_seen": 16658256,
      "step": 78935
    },
    {
      "epoch": 8.684268426842685,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.02089912073107467,
      "loss": 0.2304,
      "num_input_tokens_seen": 16659280,
      "step": 78940
    },
    {
      "epoch": 8.684818481848184,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.020897796700271552,
      "loss": 0.235,
      "num_input_tokens_seen": 16660400,
      "step": 78945
    },
    {
      "epoch": 8.685368536853685,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02089647261511127,
      "loss": 0.2298,
      "num_input_tokens_seen": 16661456,
      "step": 78950
    },
    {
      "epoch": 8.685918591859187,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.020895148475606026,
      "loss": 0.2345,
      "num_input_tokens_seen": 16662480,
      "step": 78955
    },
    {
      "epoch": 8.686468646864686,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.02089382428176803,
      "loss": 0.2314,
      "num_input_tokens_seen": 16663568,
      "step": 78960
    },
    {
      "epoch": 8.687018701870187,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02089250003360948,
      "loss": 0.2335,
      "num_input_tokens_seen": 16664688,
      "step": 78965
    },
    {
      "epoch": 8.687568756875688,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.020891175731142596,
      "loss": 0.2309,
      "num_input_tokens_seen": 16665808,
      "step": 78970
    },
    {
      "epoch": 8.688118811881187,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.020889851374379563,
      "loss": 0.2319,
      "num_input_tokens_seen": 16666896,
      "step": 78975
    },
    {
      "epoch": 8.688668866886688,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.020888526963332595,
      "loss": 0.2345,
      "num_input_tokens_seen": 16667888,
      "step": 78980
    },
    {
      "epoch": 8.68921892189219,
      "grad_norm": 0.00067138671875,
      "learning_rate": 0.020887202498013905,
      "loss": 0.2304,
      "num_input_tokens_seen": 16668912,
      "step": 78985
    },
    {
      "epoch": 8.689768976897689,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02088587797843569,
      "loss": 0.2314,
      "num_input_tokens_seen": 16670000,
      "step": 78990
    },
    {
      "epoch": 8.69031903190319,
      "grad_norm": 0.002960205078125,
      "learning_rate": 0.020884553404610165,
      "loss": 0.2335,
      "num_input_tokens_seen": 16671152,
      "step": 78995
    },
    {
      "epoch": 8.690869086908691,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.020883228776549537,
      "loss": 0.2298,
      "num_input_tokens_seen": 16672272,
      "step": 79000
    },
    {
      "epoch": 8.691419141914192,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.020881904094266008,
      "loss": 0.233,
      "num_input_tokens_seen": 16673328,
      "step": 79005
    },
    {
      "epoch": 8.691969196919691,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02088057935777179,
      "loss": 0.2309,
      "num_input_tokens_seen": 16674416,
      "step": 79010
    },
    {
      "epoch": 8.692519251925193,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.0208792545670791,
      "loss": 0.2314,
      "num_input_tokens_seen": 16675504,
      "step": 79015
    },
    {
      "epoch": 8.693069306930694,
      "grad_norm": 0.002960205078125,
      "learning_rate": 0.02087792972220014,
      "loss": 0.2314,
      "num_input_tokens_seen": 16676528,
      "step": 79020
    },
    {
      "epoch": 8.693619361936193,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.020876604823147116,
      "loss": 0.2309,
      "num_input_tokens_seen": 16677552,
      "step": 79025
    },
    {
      "epoch": 8.694169416941694,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.020875279869932255,
      "loss": 0.2329,
      "num_input_tokens_seen": 16678608,
      "step": 79030
    },
    {
      "epoch": 8.694719471947195,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.020873954862567753,
      "loss": 0.2293,
      "num_input_tokens_seen": 16679600,
      "step": 79035
    },
    {
      "epoch": 8.695269526952695,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.020872629801065825,
      "loss": 0.2308,
      "num_input_tokens_seen": 16680720,
      "step": 79040
    },
    {
      "epoch": 8.695819581958196,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020871304685438692,
      "loss": 0.2299,
      "num_input_tokens_seen": 16681840,
      "step": 79045
    },
    {
      "epoch": 8.696369636963697,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.020869979515698557,
      "loss": 0.235,
      "num_input_tokens_seen": 16682896,
      "step": 79050
    },
    {
      "epoch": 8.696919691969196,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02086865429185764,
      "loss": 0.233,
      "num_input_tokens_seen": 16683984,
      "step": 79055
    },
    {
      "epoch": 8.697469746974697,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020867329013928147,
      "loss": 0.2324,
      "num_input_tokens_seen": 16685040,
      "step": 79060
    },
    {
      "epoch": 8.698019801980198,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.020866003681922308,
      "loss": 0.2314,
      "num_input_tokens_seen": 16686096,
      "step": 79065
    },
    {
      "epoch": 8.6985698569857,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02086467829585232,
      "loss": 0.2309,
      "num_input_tokens_seen": 16687216,
      "step": 79070
    },
    {
      "epoch": 8.699119911991199,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.02086335285573041,
      "loss": 0.2303,
      "num_input_tokens_seen": 16688272,
      "step": 79075
    },
    {
      "epoch": 8.6996699669967,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.020862027361568784,
      "loss": 0.2324,
      "num_input_tokens_seen": 16689392,
      "step": 79080
    },
    {
      "epoch": 8.7002200220022,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02086070181337967,
      "loss": 0.2319,
      "num_input_tokens_seen": 16690480,
      "step": 79085
    },
    {
      "epoch": 8.7007700770077,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.020859376211175278,
      "loss": 0.2304,
      "num_input_tokens_seen": 16691472,
      "step": 79090
    },
    {
      "epoch": 8.701320132013201,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.020858050554967828,
      "loss": 0.2309,
      "num_input_tokens_seen": 16692528,
      "step": 79095
    },
    {
      "epoch": 8.701870187018702,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.020856724844769536,
      "loss": 0.2309,
      "num_input_tokens_seen": 16693552,
      "step": 79100
    },
    {
      "epoch": 8.702420242024202,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.020855399080592626,
      "loss": 0.2309,
      "num_input_tokens_seen": 16694608,
      "step": 79105
    },
    {
      "epoch": 8.702970297029703,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.020854073262449306,
      "loss": 0.2324,
      "num_input_tokens_seen": 16695760,
      "step": 79110
    },
    {
      "epoch": 8.703520352035204,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.020852747390351804,
      "loss": 0.2303,
      "num_input_tokens_seen": 16696944,
      "step": 79115
    },
    {
      "epoch": 8.704070407040705,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020851421464312337,
      "loss": 0.2329,
      "num_input_tokens_seen": 16697968,
      "step": 79120
    },
    {
      "epoch": 8.704620462046204,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.02085009548434313,
      "loss": 0.2314,
      "num_input_tokens_seen": 16699120,
      "step": 79125
    },
    {
      "epoch": 8.705170517051705,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.020848769450456398,
      "loss": 0.234,
      "num_input_tokens_seen": 16700144,
      "step": 79130
    },
    {
      "epoch": 8.705720572057206,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.020847443362664364,
      "loss": 0.2314,
      "num_input_tokens_seen": 16701200,
      "step": 79135
    },
    {
      "epoch": 8.706270627062706,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02084611722097925,
      "loss": 0.2298,
      "num_input_tokens_seen": 16702288,
      "step": 79140
    },
    {
      "epoch": 8.706820682068207,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02084479102541328,
      "loss": 0.2329,
      "num_input_tokens_seen": 16703312,
      "step": 79145
    },
    {
      "epoch": 8.707370737073708,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.020843464775978676,
      "loss": 0.2324,
      "num_input_tokens_seen": 16704400,
      "step": 79150
    },
    {
      "epoch": 8.707920792079207,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.02084213847268766,
      "loss": 0.2293,
      "num_input_tokens_seen": 16705456,
      "step": 79155
    },
    {
      "epoch": 8.708470847084708,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.020840812115552457,
      "loss": 0.2304,
      "num_input_tokens_seen": 16706512,
      "step": 79160
    },
    {
      "epoch": 8.70902090209021,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02083948570458529,
      "loss": 0.2304,
      "num_input_tokens_seen": 16707568,
      "step": 79165
    },
    {
      "epoch": 8.70957095709571,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.020838159239798384,
      "loss": 0.2293,
      "num_input_tokens_seen": 16708688,
      "step": 79170
    },
    {
      "epoch": 8.71012101210121,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02083683272120397,
      "loss": 0.2267,
      "num_input_tokens_seen": 16709712,
      "step": 79175
    },
    {
      "epoch": 8.710671067106711,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02083550614881427,
      "loss": 0.2304,
      "num_input_tokens_seen": 16710736,
      "step": 79180
    },
    {
      "epoch": 8.711221122112212,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.020834179522641504,
      "loss": 0.2288,
      "num_input_tokens_seen": 16711824,
      "step": 79185
    },
    {
      "epoch": 8.711771177117711,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020832852842697914,
      "loss": 0.2331,
      "num_input_tokens_seen": 16712848,
      "step": 79190
    },
    {
      "epoch": 8.712321232123212,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.02083152610899571,
      "loss": 0.2326,
      "num_input_tokens_seen": 16713936,
      "step": 79195
    },
    {
      "epoch": 8.712871287128714,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.02083019932154713,
      "loss": 0.2331,
      "num_input_tokens_seen": 16714992,
      "step": 79200
    },
    {
      "epoch": 8.713421342134213,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.020828872480364406,
      "loss": 0.2326,
      "num_input_tokens_seen": 16715984,
      "step": 79205
    },
    {
      "epoch": 8.713971397139714,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.020827545585459756,
      "loss": 0.2335,
      "num_input_tokens_seen": 16717040,
      "step": 79210
    },
    {
      "epoch": 8.714521452145215,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02082621863684541,
      "loss": 0.2341,
      "num_input_tokens_seen": 16718096,
      "step": 79215
    },
    {
      "epoch": 8.715071507150714,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.02082489163453361,
      "loss": 0.2294,
      "num_input_tokens_seen": 16719184,
      "step": 79220
    },
    {
      "epoch": 8.715621562156215,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020823564578536582,
      "loss": 0.2319,
      "num_input_tokens_seen": 16720208,
      "step": 79225
    },
    {
      "epoch": 8.716171617161717,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.020822237468866547,
      "loss": 0.2289,
      "num_input_tokens_seen": 16721200,
      "step": 79230
    },
    {
      "epoch": 8.716721672167218,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.020820910305535744,
      "loss": 0.2309,
      "num_input_tokens_seen": 16722224,
      "step": 79235
    },
    {
      "epoch": 8.717271727172717,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.020819583088556408,
      "loss": 0.2335,
      "num_input_tokens_seen": 16723280,
      "step": 79240
    },
    {
      "epoch": 8.717821782178218,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.02081825581794076,
      "loss": 0.2304,
      "num_input_tokens_seen": 16724400,
      "step": 79245
    },
    {
      "epoch": 8.718371837183719,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.020816928493701047,
      "loss": 0.232,
      "num_input_tokens_seen": 16725456,
      "step": 79250
    },
    {
      "epoch": 8.718921892189218,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.02081560111584949,
      "loss": 0.2319,
      "num_input_tokens_seen": 16726576,
      "step": 79255
    },
    {
      "epoch": 8.71947194719472,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.02081427368439833,
      "loss": 0.2341,
      "num_input_tokens_seen": 16727696,
      "step": 79260
    },
    {
      "epoch": 8.72002200220022,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.0208129461993598,
      "loss": 0.234,
      "num_input_tokens_seen": 16728720,
      "step": 79265
    },
    {
      "epoch": 8.72057205720572,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.020811618660746135,
      "loss": 0.2314,
      "num_input_tokens_seen": 16729808,
      "step": 79270
    },
    {
      "epoch": 8.721122112211221,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.020810291068569565,
      "loss": 0.2304,
      "num_input_tokens_seen": 16730800,
      "step": 79275
    },
    {
      "epoch": 8.721672167216722,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020808963422842337,
      "loss": 0.2319,
      "num_input_tokens_seen": 16731856,
      "step": 79280
    },
    {
      "epoch": 8.722222222222221,
      "grad_norm": 0.00286865234375,
      "learning_rate": 0.020807635723576672,
      "loss": 0.2335,
      "num_input_tokens_seen": 16732880,
      "step": 79285
    },
    {
      "epoch": 8.722772277227723,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020806307970784824,
      "loss": 0.2325,
      "num_input_tokens_seen": 16733968,
      "step": 79290
    },
    {
      "epoch": 8.723322332233224,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02080498016447902,
      "loss": 0.2314,
      "num_input_tokens_seen": 16735024,
      "step": 79295
    },
    {
      "epoch": 8.723872387238725,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.020803652304671495,
      "loss": 0.2319,
      "num_input_tokens_seen": 16735984,
      "step": 79300
    },
    {
      "epoch": 8.724422442244224,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.020802324391374496,
      "loss": 0.2314,
      "num_input_tokens_seen": 16736976,
      "step": 79305
    },
    {
      "epoch": 8.724972497249725,
      "grad_norm": 0.00323486328125,
      "learning_rate": 0.020800996424600253,
      "loss": 0.2293,
      "num_input_tokens_seen": 16738128,
      "step": 79310
    },
    {
      "epoch": 8.725522552255226,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02079966840436101,
      "loss": 0.2288,
      "num_input_tokens_seen": 16739280,
      "step": 79315
    },
    {
      "epoch": 8.726072607260726,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.02079834033066901,
      "loss": 0.2283,
      "num_input_tokens_seen": 16740304,
      "step": 79320
    },
    {
      "epoch": 8.726622662266227,
      "grad_norm": 0.00077056884765625,
      "learning_rate": 0.02079701220353649,
      "loss": 0.2293,
      "num_input_tokens_seen": 16741360,
      "step": 79325
    },
    {
      "epoch": 8.727172717271728,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.020795684022975686,
      "loss": 0.234,
      "num_input_tokens_seen": 16742448,
      "step": 79330
    },
    {
      "epoch": 8.727722772277227,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.020794355788998842,
      "loss": 0.234,
      "num_input_tokens_seen": 16743440,
      "step": 79335
    },
    {
      "epoch": 8.728272827282728,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.02079302750161821,
      "loss": 0.2304,
      "num_input_tokens_seen": 16744464,
      "step": 79340
    },
    {
      "epoch": 8.72882288228823,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.020791699160846014,
      "loss": 0.2319,
      "num_input_tokens_seen": 16745552,
      "step": 79345
    },
    {
      "epoch": 8.729372937293729,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.02079037076669451,
      "loss": 0.233,
      "num_input_tokens_seen": 16746576,
      "step": 79350
    },
    {
      "epoch": 8.72992299229923,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.020789042319175936,
      "loss": 0.233,
      "num_input_tokens_seen": 16747600,
      "step": 79355
    },
    {
      "epoch": 8.73047304730473,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.020787713818302536,
      "loss": 0.234,
      "num_input_tokens_seen": 16748656,
      "step": 79360
    },
    {
      "epoch": 8.731023102310232,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.020786385264086563,
      "loss": 0.2304,
      "num_input_tokens_seen": 16749744,
      "step": 79365
    },
    {
      "epoch": 8.731573157315731,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.020785056656540245,
      "loss": 0.2324,
      "num_input_tokens_seen": 16750800,
      "step": 79370
    },
    {
      "epoch": 8.732123212321232,
      "grad_norm": 0.002777099609375,
      "learning_rate": 0.020783727995675835,
      "loss": 0.2314,
      "num_input_tokens_seen": 16751856,
      "step": 79375
    },
    {
      "epoch": 8.732673267326733,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.02078239928150558,
      "loss": 0.2314,
      "num_input_tokens_seen": 16752944,
      "step": 79380
    },
    {
      "epoch": 8.733223322332233,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02078107051404173,
      "loss": 0.2319,
      "num_input_tokens_seen": 16754000,
      "step": 79385
    },
    {
      "epoch": 8.733773377337734,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.020779741693296525,
      "loss": 0.2319,
      "num_input_tokens_seen": 16754960,
      "step": 79390
    },
    {
      "epoch": 8.734323432343235,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.020778412819282215,
      "loss": 0.2314,
      "num_input_tokens_seen": 16756048,
      "step": 79395
    },
    {
      "epoch": 8.734873487348734,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020777083892011047,
      "loss": 0.2325,
      "num_input_tokens_seen": 16757136,
      "step": 79400
    },
    {
      "epoch": 8.735423542354235,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.020775754911495265,
      "loss": 0.2303,
      "num_input_tokens_seen": 16758224,
      "step": 79405
    },
    {
      "epoch": 8.735973597359736,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.020774425877747124,
      "loss": 0.2319,
      "num_input_tokens_seen": 16759312,
      "step": 79410
    },
    {
      "epoch": 8.736523652365236,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.020773096790778867,
      "loss": 0.2324,
      "num_input_tokens_seen": 16760336,
      "step": 79415
    },
    {
      "epoch": 8.737073707370737,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.020771767650602756,
      "loss": 0.233,
      "num_input_tokens_seen": 16761424,
      "step": 79420
    },
    {
      "epoch": 8.737623762376238,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.020770438457231026,
      "loss": 0.2303,
      "num_input_tokens_seen": 16762480,
      "step": 79425
    },
    {
      "epoch": 8.738173817381739,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.020769109210675932,
      "loss": 0.2298,
      "num_input_tokens_seen": 16763568,
      "step": 79430
    },
    {
      "epoch": 8.738723872387238,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.020767779910949732,
      "loss": 0.2303,
      "num_input_tokens_seen": 16764624,
      "step": 79435
    },
    {
      "epoch": 8.73927392739274,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.020766450558064666,
      "loss": 0.2325,
      "num_input_tokens_seen": 16765712,
      "step": 79440
    },
    {
      "epoch": 8.73982398239824,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.020765121152032995,
      "loss": 0.2314,
      "num_input_tokens_seen": 16766800,
      "step": 79445
    },
    {
      "epoch": 8.74037403740374,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.020763791692866972,
      "loss": 0.2314,
      "num_input_tokens_seen": 16767824,
      "step": 79450
    },
    {
      "epoch": 8.74092409240924,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.020762462180578842,
      "loss": 0.2319,
      "num_input_tokens_seen": 16768848,
      "step": 79455
    },
    {
      "epoch": 8.741474147414742,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.020761132615180866,
      "loss": 0.2303,
      "num_input_tokens_seen": 16769840,
      "step": 79460
    },
    {
      "epoch": 8.742024202420241,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.020759802996685298,
      "loss": 0.2335,
      "num_input_tokens_seen": 16770864,
      "step": 79465
    },
    {
      "epoch": 8.742574257425742,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.02075847332510438,
      "loss": 0.2335,
      "num_input_tokens_seen": 16771952,
      "step": 79470
    },
    {
      "epoch": 8.743124312431243,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.020757143600450383,
      "loss": 0.233,
      "num_input_tokens_seen": 16773040,
      "step": 79475
    },
    {
      "epoch": 8.743674367436743,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.020755813822735555,
      "loss": 0.2304,
      "num_input_tokens_seen": 16774160,
      "step": 79480
    },
    {
      "epoch": 8.744224422442244,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.020754483991972158,
      "loss": 0.2324,
      "num_input_tokens_seen": 16775184,
      "step": 79485
    },
    {
      "epoch": 8.744774477447745,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.020753154108172436,
      "loss": 0.2314,
      "num_input_tokens_seen": 16776208,
      "step": 79490
    },
    {
      "epoch": 8.745324532453246,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.020751824171348653,
      "loss": 0.2335,
      "num_input_tokens_seen": 16777296,
      "step": 79495
    },
    {
      "epoch": 8.745874587458745,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020750494181513073,
      "loss": 0.2309,
      "num_input_tokens_seen": 16778384,
      "step": 79500
    },
    {
      "epoch": 8.746424642464246,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.02074916413867794,
      "loss": 0.2335,
      "num_input_tokens_seen": 16779472,
      "step": 79505
    },
    {
      "epoch": 8.746974697469748,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.020747834042855524,
      "loss": 0.2314,
      "num_input_tokens_seen": 16780560,
      "step": 79510
    },
    {
      "epoch": 8.747524752475247,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.020746503894058084,
      "loss": 0.2309,
      "num_input_tokens_seen": 16781648,
      "step": 79515
    },
    {
      "epoch": 8.748074807480748,
      "grad_norm": 0.0004119873046875,
      "learning_rate": 0.020745173692297867,
      "loss": 0.2309,
      "num_input_tokens_seen": 16782640,
      "step": 79520
    },
    {
      "epoch": 8.748624862486249,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.020743843437587145,
      "loss": 0.2293,
      "num_input_tokens_seen": 16783664,
      "step": 79525
    },
    {
      "epoch": 8.749174917491748,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.020742513129938172,
      "loss": 0.233,
      "num_input_tokens_seen": 16784688,
      "step": 79530
    },
    {
      "epoch": 8.74972497249725,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.020741182769363216,
      "loss": 0.2303,
      "num_input_tokens_seen": 16785744,
      "step": 79535
    },
    {
      "epoch": 8.75027502750275,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.020739852355874527,
      "loss": 0.2319,
      "num_input_tokens_seen": 16786768,
      "step": 79540
    },
    {
      "epoch": 8.750825082508252,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.020738521889484378,
      "loss": 0.2324,
      "num_input_tokens_seen": 16787824,
      "step": 79545
    },
    {
      "epoch": 8.751375137513751,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02073719137020502,
      "loss": 0.2304,
      "num_input_tokens_seen": 16788880,
      "step": 79550
    },
    {
      "epoch": 8.751925192519252,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.02073586079804873,
      "loss": 0.233,
      "num_input_tokens_seen": 16789936,
      "step": 79555
    },
    {
      "epoch": 8.752475247524753,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02073453017302776,
      "loss": 0.2319,
      "num_input_tokens_seen": 16790992,
      "step": 79560
    },
    {
      "epoch": 8.753025302530252,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.020733199495154377,
      "loss": 0.2303,
      "num_input_tokens_seen": 16792016,
      "step": 79565
    },
    {
      "epoch": 8.753575357535754,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.020731868764440844,
      "loss": 0.2309,
      "num_input_tokens_seen": 16793104,
      "step": 79570
    },
    {
      "epoch": 8.754125412541255,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.02073053798089943,
      "loss": 0.2298,
      "num_input_tokens_seen": 16794128,
      "step": 79575
    },
    {
      "epoch": 8.754675467546754,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0207292071445424,
      "loss": 0.2329,
      "num_input_tokens_seen": 16795216,
      "step": 79580
    },
    {
      "epoch": 8.755225522552255,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.02072787625538201,
      "loss": 0.2308,
      "num_input_tokens_seen": 16796272,
      "step": 79585
    },
    {
      "epoch": 8.755775577557756,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.020726545313430534,
      "loss": 0.2319,
      "num_input_tokens_seen": 16797456,
      "step": 79590
    },
    {
      "epoch": 8.756325632563257,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.020725214318700235,
      "loss": 0.233,
      "num_input_tokens_seen": 16798416,
      "step": 79595
    },
    {
      "epoch": 8.756875687568757,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.020723883271203388,
      "loss": 0.2314,
      "num_input_tokens_seen": 16799472,
      "step": 79600
    },
    {
      "epoch": 8.757425742574258,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.020722552170952258,
      "loss": 0.2324,
      "num_input_tokens_seen": 16800528,
      "step": 79605
    },
    {
      "epoch": 8.757975797579759,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.020721221017959108,
      "loss": 0.2324,
      "num_input_tokens_seen": 16801552,
      "step": 79610
    },
    {
      "epoch": 8.758525852585258,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.020719889812236207,
      "loss": 0.2319,
      "num_input_tokens_seen": 16802672,
      "step": 79615
    },
    {
      "epoch": 8.75907590759076,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.020718558553795823,
      "loss": 0.2309,
      "num_input_tokens_seen": 16803760,
      "step": 79620
    },
    {
      "epoch": 8.75962596259626,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020717227242650238,
      "loss": 0.233,
      "num_input_tokens_seen": 16804848,
      "step": 79625
    },
    {
      "epoch": 8.76017601760176,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02071589587881171,
      "loss": 0.233,
      "num_input_tokens_seen": 16805872,
      "step": 79630
    },
    {
      "epoch": 8.76072607260726,
      "grad_norm": 0.001953125,
      "learning_rate": 0.020714564462292508,
      "loss": 0.2303,
      "num_input_tokens_seen": 16806960,
      "step": 79635
    },
    {
      "epoch": 8.761276127612762,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02071323299310491,
      "loss": 0.2319,
      "num_input_tokens_seen": 16807984,
      "step": 79640
    },
    {
      "epoch": 8.761826182618261,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020711901471261184,
      "loss": 0.2309,
      "num_input_tokens_seen": 16809008,
      "step": 79645
    },
    {
      "epoch": 8.762376237623762,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.020710569896773605,
      "loss": 0.2273,
      "num_input_tokens_seen": 16810064,
      "step": 79650
    },
    {
      "epoch": 8.762926292629263,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.020709238269654442,
      "loss": 0.2309,
      "num_input_tokens_seen": 16811088,
      "step": 79655
    },
    {
      "epoch": 8.763476347634764,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02070790658991597,
      "loss": 0.234,
      "num_input_tokens_seen": 16812208,
      "step": 79660
    },
    {
      "epoch": 8.764026402640264,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.020706574857570458,
      "loss": 0.2325,
      "num_input_tokens_seen": 16813168,
      "step": 79665
    },
    {
      "epoch": 8.764576457645765,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.02070524307263019,
      "loss": 0.2325,
      "num_input_tokens_seen": 16814256,
      "step": 79670
    },
    {
      "epoch": 8.765126512651266,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.020703911235107435,
      "loss": 0.2324,
      "num_input_tokens_seen": 16815280,
      "step": 79675
    },
    {
      "epoch": 8.765676567656765,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.020702579345014464,
      "loss": 0.2319,
      "num_input_tokens_seen": 16816336,
      "step": 79680
    },
    {
      "epoch": 8.766226622662266,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020701247402363555,
      "loss": 0.2314,
      "num_input_tokens_seen": 16817392,
      "step": 79685
    },
    {
      "epoch": 8.766776677667767,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.020699915407166987,
      "loss": 0.2314,
      "num_input_tokens_seen": 16818480,
      "step": 79690
    },
    {
      "epoch": 8.767326732673267,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.02069858335943703,
      "loss": 0.2314,
      "num_input_tokens_seen": 16819472,
      "step": 79695
    },
    {
      "epoch": 8.767876787678768,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.020697251259185964,
      "loss": 0.2319,
      "num_input_tokens_seen": 16820496,
      "step": 79700
    },
    {
      "epoch": 8.768426842684269,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.020695919106426073,
      "loss": 0.233,
      "num_input_tokens_seen": 16821552,
      "step": 79705
    },
    {
      "epoch": 8.768976897689768,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020694586901169622,
      "loss": 0.2324,
      "num_input_tokens_seen": 16822640,
      "step": 79710
    },
    {
      "epoch": 8.76952695269527,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.020693254643428898,
      "loss": 0.2293,
      "num_input_tokens_seen": 16823728,
      "step": 79715
    },
    {
      "epoch": 8.77007700770077,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.020691922333216182,
      "loss": 0.2319,
      "num_input_tokens_seen": 16824720,
      "step": 79720
    },
    {
      "epoch": 8.770627062706271,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020690589970543745,
      "loss": 0.2298,
      "num_input_tokens_seen": 16825776,
      "step": 79725
    },
    {
      "epoch": 8.77117711771177,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.020689257555423873,
      "loss": 0.2324,
      "num_input_tokens_seen": 16826832,
      "step": 79730
    },
    {
      "epoch": 8.771727172717272,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.020687925087868844,
      "loss": 0.2288,
      "num_input_tokens_seen": 16827920,
      "step": 79735
    },
    {
      "epoch": 8.772277227722773,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020686592567890936,
      "loss": 0.2319,
      "num_input_tokens_seen": 16828912,
      "step": 79740
    },
    {
      "epoch": 8.772827282728272,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.020685259995502435,
      "loss": 0.2309,
      "num_input_tokens_seen": 16829936,
      "step": 79745
    },
    {
      "epoch": 8.773377337733773,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.02068392737071562,
      "loss": 0.2345,
      "num_input_tokens_seen": 16830960,
      "step": 79750
    },
    {
      "epoch": 8.773927392739274,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.020682594693542774,
      "loss": 0.2314,
      "num_input_tokens_seen": 16832016,
      "step": 79755
    },
    {
      "epoch": 8.774477447744774,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02068126196399618,
      "loss": 0.233,
      "num_input_tokens_seen": 16833072,
      "step": 79760
    },
    {
      "epoch": 8.775027502750275,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020679929182088122,
      "loss": 0.2298,
      "num_input_tokens_seen": 16834160,
      "step": 79765
    },
    {
      "epoch": 8.775577557755776,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.02067859634783088,
      "loss": 0.2304,
      "num_input_tokens_seen": 16835216,
      "step": 79770
    },
    {
      "epoch": 8.776127612761275,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.02067726346123674,
      "loss": 0.233,
      "num_input_tokens_seen": 16836336,
      "step": 79775
    },
    {
      "epoch": 8.776677667766776,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.020675930522317986,
      "loss": 0.234,
      "num_input_tokens_seen": 16837456,
      "step": 79780
    },
    {
      "epoch": 8.777227722772277,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.020674597531086907,
      "loss": 0.2314,
      "num_input_tokens_seen": 16838512,
      "step": 79785
    },
    {
      "epoch": 8.777777777777779,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.020673264487555784,
      "loss": 0.2304,
      "num_input_tokens_seen": 16839536,
      "step": 79790
    },
    {
      "epoch": 8.778327832783278,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.020671931391736908,
      "loss": 0.2303,
      "num_input_tokens_seen": 16840624,
      "step": 79795
    },
    {
      "epoch": 8.778877887788779,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02067059824364256,
      "loss": 0.2309,
      "num_input_tokens_seen": 16841712,
      "step": 79800
    },
    {
      "epoch": 8.77942794279428,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.020669265043285028,
      "loss": 0.2335,
      "num_input_tokens_seen": 16842800,
      "step": 79805
    },
    {
      "epoch": 8.77997799779978,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.020667931790676594,
      "loss": 0.2303,
      "num_input_tokens_seen": 16843824,
      "step": 79810
    },
    {
      "epoch": 8.78052805280528,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.020666598485829562,
      "loss": 0.2314,
      "num_input_tokens_seen": 16844848,
      "step": 79815
    },
    {
      "epoch": 8.781078107810782,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.02066526512875621,
      "loss": 0.2319,
      "num_input_tokens_seen": 16845872,
      "step": 79820
    },
    {
      "epoch": 8.781628162816281,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.02066393171946882,
      "loss": 0.2319,
      "num_input_tokens_seen": 16846928,
      "step": 79825
    },
    {
      "epoch": 8.782178217821782,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.020662598257979697,
      "loss": 0.2293,
      "num_input_tokens_seen": 16848048,
      "step": 79830
    },
    {
      "epoch": 8.782728272827283,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02066126474430112,
      "loss": 0.2329,
      "num_input_tokens_seen": 16849104,
      "step": 79835
    },
    {
      "epoch": 8.783278327832782,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020659931178445383,
      "loss": 0.2309,
      "num_input_tokens_seen": 16850160,
      "step": 79840
    },
    {
      "epoch": 8.783828382838283,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.020658597560424777,
      "loss": 0.2288,
      "num_input_tokens_seen": 16851120,
      "step": 79845
    },
    {
      "epoch": 8.784378437843785,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.020657263890251593,
      "loss": 0.2314,
      "num_input_tokens_seen": 16852144,
      "step": 79850
    },
    {
      "epoch": 8.784928492849286,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.020655930167938118,
      "loss": 0.2304,
      "num_input_tokens_seen": 16853232,
      "step": 79855
    },
    {
      "epoch": 8.785478547854785,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020654596393496653,
      "loss": 0.2314,
      "num_input_tokens_seen": 16854320,
      "step": 79860
    },
    {
      "epoch": 8.786028602860286,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.020653262566939487,
      "loss": 0.2314,
      "num_input_tokens_seen": 16855440,
      "step": 79865
    },
    {
      "epoch": 8.786578657865787,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02065192868827891,
      "loss": 0.2293,
      "num_input_tokens_seen": 16856496,
      "step": 79870
    },
    {
      "epoch": 8.787128712871286,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02065059475752722,
      "loss": 0.2336,
      "num_input_tokens_seen": 16857584,
      "step": 79875
    },
    {
      "epoch": 8.787678767876788,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.020649260774696705,
      "loss": 0.2293,
      "num_input_tokens_seen": 16858672,
      "step": 79880
    },
    {
      "epoch": 8.788228822882289,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.020647926739799666,
      "loss": 0.2314,
      "num_input_tokens_seen": 16859664,
      "step": 79885
    },
    {
      "epoch": 8.788778877887788,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.020646592652848402,
      "loss": 0.2309,
      "num_input_tokens_seen": 16860688,
      "step": 79890
    },
    {
      "epoch": 8.789328932893289,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.0206452585138552,
      "loss": 0.232,
      "num_input_tokens_seen": 16861744,
      "step": 79895
    },
    {
      "epoch": 8.78987898789879,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.02064392432283236,
      "loss": 0.2319,
      "num_input_tokens_seen": 16862832,
      "step": 79900
    },
    {
      "epoch": 8.79042904290429,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.02064259007979217,
      "loss": 0.2335,
      "num_input_tokens_seen": 16863920,
      "step": 79905
    },
    {
      "epoch": 8.79097909790979,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.020641255784746945,
      "loss": 0.231,
      "num_input_tokens_seen": 16864944,
      "step": 79910
    },
    {
      "epoch": 8.791529152915292,
      "grad_norm": 0.000774383544921875,
      "learning_rate": 0.02063992143770897,
      "loss": 0.2288,
      "num_input_tokens_seen": 16866000,
      "step": 79915
    },
    {
      "epoch": 8.792079207920793,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.020638587038690544,
      "loss": 0.2335,
      "num_input_tokens_seen": 16866992,
      "step": 79920
    },
    {
      "epoch": 8.792629262926292,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.020637252587703964,
      "loss": 0.232,
      "num_input_tokens_seen": 16868080,
      "step": 79925
    },
    {
      "epoch": 8.793179317931793,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.02063591808476153,
      "loss": 0.2315,
      "num_input_tokens_seen": 16869104,
      "step": 79930
    },
    {
      "epoch": 8.793729372937294,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.020634583529875555,
      "loss": 0.232,
      "num_input_tokens_seen": 16870160,
      "step": 79935
    },
    {
      "epoch": 8.794279427942794,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.020633248923058317,
      "loss": 0.2356,
      "num_input_tokens_seen": 16871248,
      "step": 79940
    },
    {
      "epoch": 8.794829482948295,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02063191426432213,
      "loss": 0.2366,
      "num_input_tokens_seen": 16872304,
      "step": 79945
    },
    {
      "epoch": 8.795379537953796,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020630579553679293,
      "loss": 0.2304,
      "num_input_tokens_seen": 16873360,
      "step": 79950
    },
    {
      "epoch": 8.795929592959295,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.0206292447911421,
      "loss": 0.2309,
      "num_input_tokens_seen": 16874384,
      "step": 79955
    },
    {
      "epoch": 8.796479647964796,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.02062790997672287,
      "loss": 0.233,
      "num_input_tokens_seen": 16875408,
      "step": 79960
    },
    {
      "epoch": 8.797029702970297,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02062657511043389,
      "loss": 0.2319,
      "num_input_tokens_seen": 16876496,
      "step": 79965
    },
    {
      "epoch": 8.797579757975798,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02062524019228746,
      "loss": 0.2324,
      "num_input_tokens_seen": 16877584,
      "step": 79970
    },
    {
      "epoch": 8.798129812981298,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.020623905222295895,
      "loss": 0.2314,
      "num_input_tokens_seen": 16878672,
      "step": 79975
    },
    {
      "epoch": 8.798679867986799,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.020622570200471497,
      "loss": 0.2303,
      "num_input_tokens_seen": 16879664,
      "step": 79980
    },
    {
      "epoch": 8.7992299229923,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020621235126826565,
      "loss": 0.2303,
      "num_input_tokens_seen": 16880816,
      "step": 79985
    },
    {
      "epoch": 8.7997799779978,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02061990000137341,
      "loss": 0.2319,
      "num_input_tokens_seen": 16881808,
      "step": 79990
    },
    {
      "epoch": 8.8003300330033,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020618564824124327,
      "loss": 0.2298,
      "num_input_tokens_seen": 16882896,
      "step": 79995
    },
    {
      "epoch": 8.800880088008801,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.020617229595091626,
      "loss": 0.2314,
      "num_input_tokens_seen": 16884016,
      "step": 80000
    },
    {
      "epoch": 8.8014301430143,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.020615894314287626,
      "loss": 0.2309,
      "num_input_tokens_seen": 16885072,
      "step": 80005
    },
    {
      "epoch": 8.801980198019802,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.020614558981724616,
      "loss": 0.2314,
      "num_input_tokens_seen": 16886064,
      "step": 80010
    },
    {
      "epoch": 8.802530253025303,
      "grad_norm": 0.000621795654296875,
      "learning_rate": 0.020613223597414906,
      "loss": 0.233,
      "num_input_tokens_seen": 16887120,
      "step": 80015
    },
    {
      "epoch": 8.803080308030804,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020611888161370818,
      "loss": 0.2293,
      "num_input_tokens_seen": 16888112,
      "step": 80020
    },
    {
      "epoch": 8.803630363036303,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.02061055267360464,
      "loss": 0.2329,
      "num_input_tokens_seen": 16889168,
      "step": 80025
    },
    {
      "epoch": 8.804180418041804,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.020609217134128693,
      "loss": 0.2303,
      "num_input_tokens_seen": 16890224,
      "step": 80030
    },
    {
      "epoch": 8.804730473047305,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.020607881542955286,
      "loss": 0.2303,
      "num_input_tokens_seen": 16891280,
      "step": 80035
    },
    {
      "epoch": 8.805280528052805,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02060654590009672,
      "loss": 0.2314,
      "num_input_tokens_seen": 16892304,
      "step": 80040
    },
    {
      "epoch": 8.805830583058306,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02060521020556532,
      "loss": 0.233,
      "num_input_tokens_seen": 16893328,
      "step": 80045
    },
    {
      "epoch": 8.806380638063807,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02060387445937338,
      "loss": 0.2304,
      "num_input_tokens_seen": 16894416,
      "step": 80050
    },
    {
      "epoch": 8.806930693069306,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.02060253866153322,
      "loss": 0.2319,
      "num_input_tokens_seen": 16895504,
      "step": 80055
    },
    {
      "epoch": 8.807480748074807,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02060120281205715,
      "loss": 0.2314,
      "num_input_tokens_seen": 16896496,
      "step": 80060
    },
    {
      "epoch": 8.808030803080309,
      "grad_norm": 0.000568389892578125,
      "learning_rate": 0.02059986691095748,
      "loss": 0.2314,
      "num_input_tokens_seen": 16897488,
      "step": 80065
    },
    {
      "epoch": 8.808580858085808,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.020598530958246524,
      "loss": 0.2309,
      "num_input_tokens_seen": 16898512,
      "step": 80070
    },
    {
      "epoch": 8.809130913091309,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.020597194953936596,
      "loss": 0.2303,
      "num_input_tokens_seen": 16899568,
      "step": 80075
    },
    {
      "epoch": 8.80968096809681,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.020595858898040012,
      "loss": 0.2319,
      "num_input_tokens_seen": 16900624,
      "step": 80080
    },
    {
      "epoch": 8.810231023102311,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.020594522790569076,
      "loss": 0.2314,
      "num_input_tokens_seen": 16901616,
      "step": 80085
    },
    {
      "epoch": 8.81078107810781,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.020593186631536108,
      "loss": 0.2314,
      "num_input_tokens_seen": 16902672,
      "step": 80090
    },
    {
      "epoch": 8.811331133113312,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.020591850420953423,
      "loss": 0.2314,
      "num_input_tokens_seen": 16903696,
      "step": 80095
    },
    {
      "epoch": 8.811881188118813,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.02059051415883334,
      "loss": 0.2329,
      "num_input_tokens_seen": 16904688,
      "step": 80100
    },
    {
      "epoch": 8.812431243124312,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.020589177845188166,
      "loss": 0.233,
      "num_input_tokens_seen": 16905744,
      "step": 80105
    },
    {
      "epoch": 8.812981298129813,
      "grad_norm": 0.002960205078125,
      "learning_rate": 0.020587841480030223,
      "loss": 0.2324,
      "num_input_tokens_seen": 16906736,
      "step": 80110
    },
    {
      "epoch": 8.813531353135314,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.020586505063371827,
      "loss": 0.233,
      "num_input_tokens_seen": 16907792,
      "step": 80115
    },
    {
      "epoch": 8.814081408140813,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.02058516859522529,
      "loss": 0.2288,
      "num_input_tokens_seen": 16908848,
      "step": 80120
    },
    {
      "epoch": 8.814631463146315,
      "grad_norm": 0.0027313232421875,
      "learning_rate": 0.020583832075602944,
      "loss": 0.233,
      "num_input_tokens_seen": 16909872,
      "step": 80125
    },
    {
      "epoch": 8.815181518151816,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.020582495504517087,
      "loss": 0.2303,
      "num_input_tokens_seen": 16910928,
      "step": 80130
    },
    {
      "epoch": 8.815731573157315,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.020581158881980054,
      "loss": 0.2324,
      "num_input_tokens_seen": 16911952,
      "step": 80135
    },
    {
      "epoch": 8.816281628162816,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.020579822208004157,
      "loss": 0.2314,
      "num_input_tokens_seen": 16913040,
      "step": 80140
    },
    {
      "epoch": 8.816831683168317,
      "grad_norm": 0.009765625,
      "learning_rate": 0.020578485482601713,
      "loss": 0.2329,
      "num_input_tokens_seen": 16914096,
      "step": 80145
    },
    {
      "epoch": 8.817381738173818,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.02057714870578505,
      "loss": 0.2319,
      "num_input_tokens_seen": 16915152,
      "step": 80150
    },
    {
      "epoch": 8.817931793179318,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.020575811877566477,
      "loss": 0.2314,
      "num_input_tokens_seen": 16916208,
      "step": 80155
    },
    {
      "epoch": 8.818481848184819,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020574474997958326,
      "loss": 0.2309,
      "num_input_tokens_seen": 16917200,
      "step": 80160
    },
    {
      "epoch": 8.81903190319032,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.02057313806697291,
      "loss": 0.2303,
      "num_input_tokens_seen": 16918192,
      "step": 80165
    },
    {
      "epoch": 8.819581958195819,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020571801084622556,
      "loss": 0.2324,
      "num_input_tokens_seen": 16919216,
      "step": 80170
    },
    {
      "epoch": 8.82013201320132,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.020570464050919588,
      "loss": 0.2309,
      "num_input_tokens_seen": 16920272,
      "step": 80175
    },
    {
      "epoch": 8.820682068206821,
      "grad_norm": 0.00079345703125,
      "learning_rate": 0.020569126965876323,
      "loss": 0.2324,
      "num_input_tokens_seen": 16921296,
      "step": 80180
    },
    {
      "epoch": 8.82123212321232,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.020567789829505093,
      "loss": 0.2303,
      "num_input_tokens_seen": 16922448,
      "step": 80185
    },
    {
      "epoch": 8.821782178217822,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.020566452641818205,
      "loss": 0.2319,
      "num_input_tokens_seen": 16923504,
      "step": 80190
    },
    {
      "epoch": 8.822332233223323,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.020565115402828002,
      "loss": 0.2314,
      "num_input_tokens_seen": 16924592,
      "step": 80195
    },
    {
      "epoch": 8.822882288228822,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0205637781125468,
      "loss": 0.2303,
      "num_input_tokens_seen": 16925648,
      "step": 80200
    },
    {
      "epoch": 8.823432343234323,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02056244077098692,
      "loss": 0.2298,
      "num_input_tokens_seen": 16926704,
      "step": 80205
    },
    {
      "epoch": 8.823982398239824,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.020561103378160698,
      "loss": 0.233,
      "num_input_tokens_seen": 16927728,
      "step": 80210
    },
    {
      "epoch": 8.824532453245325,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.020559765934080453,
      "loss": 0.2303,
      "num_input_tokens_seen": 16928816,
      "step": 80215
    },
    {
      "epoch": 8.825082508250825,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.020558428438758516,
      "loss": 0.2304,
      "num_input_tokens_seen": 16929840,
      "step": 80220
    },
    {
      "epoch": 8.825632563256326,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020557090892207208,
      "loss": 0.2319,
      "num_input_tokens_seen": 16930928,
      "step": 80225
    },
    {
      "epoch": 8.826182618261827,
      "grad_norm": 0.0007476806640625,
      "learning_rate": 0.020555753294438858,
      "loss": 0.233,
      "num_input_tokens_seen": 16931952,
      "step": 80230
    },
    {
      "epoch": 8.826732673267326,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.0205544156454658,
      "loss": 0.2324,
      "num_input_tokens_seen": 16932976,
      "step": 80235
    },
    {
      "epoch": 8.827282728272827,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.020553077945300355,
      "loss": 0.2308,
      "num_input_tokens_seen": 16933968,
      "step": 80240
    },
    {
      "epoch": 8.827832783278328,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.020551740193954858,
      "loss": 0.2325,
      "num_input_tokens_seen": 16935152,
      "step": 80245
    },
    {
      "epoch": 8.828382838283828,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.02055040239144163,
      "loss": 0.2304,
      "num_input_tokens_seen": 16936208,
      "step": 80250
    },
    {
      "epoch": 8.828932893289329,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.020549064537773012,
      "loss": 0.2319,
      "num_input_tokens_seen": 16937264,
      "step": 80255
    },
    {
      "epoch": 8.82948294829483,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.020547726632961325,
      "loss": 0.2293,
      "num_input_tokens_seen": 16938288,
      "step": 80260
    },
    {
      "epoch": 8.83003300330033,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.02054638867701891,
      "loss": 0.2325,
      "num_input_tokens_seen": 16939376,
      "step": 80265
    },
    {
      "epoch": 8.83058305830583,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.02054505066995809,
      "loss": 0.2324,
      "num_input_tokens_seen": 16940400,
      "step": 80270
    },
    {
      "epoch": 8.831133113311331,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.02054371261179119,
      "loss": 0.2288,
      "num_input_tokens_seen": 16941456,
      "step": 80275
    },
    {
      "epoch": 8.831683168316832,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02054237450253056,
      "loss": 0.2324,
      "num_input_tokens_seen": 16942512,
      "step": 80280
    },
    {
      "epoch": 8.832233223322332,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020541036342188523,
      "loss": 0.233,
      "num_input_tokens_seen": 16943568,
      "step": 80285
    },
    {
      "epoch": 8.832783278327833,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.02053969813077741,
      "loss": 0.2319,
      "num_input_tokens_seen": 16944656,
      "step": 80290
    },
    {
      "epoch": 8.833333333333334,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.020538359868309562,
      "loss": 0.2298,
      "num_input_tokens_seen": 16945744,
      "step": 80295
    },
    {
      "epoch": 8.833883388338833,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.020537021554797304,
      "loss": 0.2303,
      "num_input_tokens_seen": 16946800,
      "step": 80300
    },
    {
      "epoch": 8.834433443344334,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.020535683190252976,
      "loss": 0.2329,
      "num_input_tokens_seen": 16947920,
      "step": 80305
    },
    {
      "epoch": 8.834983498349835,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020534344774688913,
      "loss": 0.2293,
      "num_input_tokens_seen": 16949040,
      "step": 80310
    },
    {
      "epoch": 8.835533553355335,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.02053300630811745,
      "loss": 0.2324,
      "num_input_tokens_seen": 16950160,
      "step": 80315
    },
    {
      "epoch": 8.836083608360836,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.020531667790550923,
      "loss": 0.234,
      "num_input_tokens_seen": 16951248,
      "step": 80320
    },
    {
      "epoch": 8.836633663366337,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02053032922200167,
      "loss": 0.2345,
      "num_input_tokens_seen": 16952208,
      "step": 80325
    },
    {
      "epoch": 8.837183718371836,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020528990602482023,
      "loss": 0.2319,
      "num_input_tokens_seen": 16953264,
      "step": 80330
    },
    {
      "epoch": 8.837733773377337,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.020527651932004325,
      "loss": 0.2319,
      "num_input_tokens_seen": 16954352,
      "step": 80335
    },
    {
      "epoch": 8.838283828382838,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02052631321058091,
      "loss": 0.2298,
      "num_input_tokens_seen": 16955376,
      "step": 80340
    },
    {
      "epoch": 8.83883388338834,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.020524974438224118,
      "loss": 0.2324,
      "num_input_tokens_seen": 16956496,
      "step": 80345
    },
    {
      "epoch": 8.839383938393839,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.020523635614946287,
      "loss": 0.2303,
      "num_input_tokens_seen": 16957552,
      "step": 80350
    },
    {
      "epoch": 8.83993399339934,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.02052229674075976,
      "loss": 0.2319,
      "num_input_tokens_seen": 16958672,
      "step": 80355
    },
    {
      "epoch": 8.840484048404841,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020520957815676873,
      "loss": 0.2314,
      "num_input_tokens_seen": 16959696,
      "step": 80360
    },
    {
      "epoch": 8.84103410341034,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.020519618839709968,
      "loss": 0.2314,
      "num_input_tokens_seen": 16960752,
      "step": 80365
    },
    {
      "epoch": 8.841584158415841,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02051827981287138,
      "loss": 0.2319,
      "num_input_tokens_seen": 16961872,
      "step": 80370
    },
    {
      "epoch": 8.842134213421343,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.020516940735173458,
      "loss": 0.2314,
      "num_input_tokens_seen": 16962896,
      "step": 80375
    },
    {
      "epoch": 8.842684268426842,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02051560160662854,
      "loss": 0.2309,
      "num_input_tokens_seen": 16963984,
      "step": 80380
    },
    {
      "epoch": 8.843234323432343,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.020514262427248974,
      "loss": 0.2309,
      "num_input_tokens_seen": 16965008,
      "step": 80385
    },
    {
      "epoch": 8.843784378437844,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.020512923197047086,
      "loss": 0.2335,
      "num_input_tokens_seen": 16966096,
      "step": 80390
    },
    {
      "epoch": 8.844334433443345,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02051158391603524,
      "loss": 0.2314,
      "num_input_tokens_seen": 16967152,
      "step": 80395
    },
    {
      "epoch": 8.844884488448844,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.020510244584225765,
      "loss": 0.233,
      "num_input_tokens_seen": 16968112,
      "step": 80400
    },
    {
      "epoch": 8.845434543454346,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.020508905201631008,
      "loss": 0.2308,
      "num_input_tokens_seen": 16969200,
      "step": 80405
    },
    {
      "epoch": 8.845984598459847,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02050756576826332,
      "loss": 0.2329,
      "num_input_tokens_seen": 16970288,
      "step": 80410
    },
    {
      "epoch": 8.846534653465346,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02050622628413504,
      "loss": 0.2303,
      "num_input_tokens_seen": 16971248,
      "step": 80415
    },
    {
      "epoch": 8.847084708470847,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.020504886749258513,
      "loss": 0.2314,
      "num_input_tokens_seen": 16972272,
      "step": 80420
    },
    {
      "epoch": 8.847634763476348,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020503547163646087,
      "loss": 0.2319,
      "num_input_tokens_seen": 16973296,
      "step": 80425
    },
    {
      "epoch": 8.848184818481847,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.02050220752731011,
      "loss": 0.2319,
      "num_input_tokens_seen": 16974320,
      "step": 80430
    },
    {
      "epoch": 8.848734873487349,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.02050086784026292,
      "loss": 0.2319,
      "num_input_tokens_seen": 16975312,
      "step": 80435
    },
    {
      "epoch": 8.84928492849285,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.020499528102516876,
      "loss": 0.2303,
      "num_input_tokens_seen": 16976368,
      "step": 80440
    },
    {
      "epoch": 8.84983498349835,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02049818831408432,
      "loss": 0.2308,
      "num_input_tokens_seen": 16977392,
      "step": 80445
    },
    {
      "epoch": 8.85038503850385,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020496848474977597,
      "loss": 0.2308,
      "num_input_tokens_seen": 16978448,
      "step": 80450
    },
    {
      "epoch": 8.850935093509351,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020495508585209064,
      "loss": 0.2309,
      "num_input_tokens_seen": 16979472,
      "step": 80455
    },
    {
      "epoch": 8.851485148514852,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.020494168644791064,
      "loss": 0.2314,
      "num_input_tokens_seen": 16980528,
      "step": 80460
    },
    {
      "epoch": 8.852035203520352,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02049282865373595,
      "loss": 0.2309,
      "num_input_tokens_seen": 16981616,
      "step": 80465
    },
    {
      "epoch": 8.852585258525853,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020491488612056068,
      "loss": 0.2314,
      "num_input_tokens_seen": 16982640,
      "step": 80470
    },
    {
      "epoch": 8.853135313531354,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02049014851976377,
      "loss": 0.2314,
      "num_input_tokens_seen": 16983664,
      "step": 80475
    },
    {
      "epoch": 8.853685368536853,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020488808376871406,
      "loss": 0.2298,
      "num_input_tokens_seen": 16984720,
      "step": 80480
    },
    {
      "epoch": 8.854235423542354,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.02048746818339133,
      "loss": 0.2314,
      "num_input_tokens_seen": 16985776,
      "step": 80485
    },
    {
      "epoch": 8.854785478547855,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.020486127939335894,
      "loss": 0.2298,
      "num_input_tokens_seen": 16986832,
      "step": 80490
    },
    {
      "epoch": 8.855335533553355,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.020484787644717447,
      "loss": 0.2314,
      "num_input_tokens_seen": 16987888,
      "step": 80495
    },
    {
      "epoch": 8.855885588558856,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020483447299548346,
      "loss": 0.2314,
      "num_input_tokens_seen": 16988912,
      "step": 80500
    },
    {
      "epoch": 8.856435643564357,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.020482106903840943,
      "loss": 0.2288,
      "num_input_tokens_seen": 16989968,
      "step": 80505
    },
    {
      "epoch": 8.856985698569858,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02048076645760759,
      "loss": 0.2314,
      "num_input_tokens_seen": 16991024,
      "step": 80510
    },
    {
      "epoch": 8.857535753575357,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020479425960860644,
      "loss": 0.2329,
      "num_input_tokens_seen": 16992080,
      "step": 80515
    },
    {
      "epoch": 8.858085808580858,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.020478085413612457,
      "loss": 0.2319,
      "num_input_tokens_seen": 16993200,
      "step": 80520
    },
    {
      "epoch": 8.85863586358636,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020476744815875383,
      "loss": 0.2319,
      "num_input_tokens_seen": 16994288,
      "step": 80525
    },
    {
      "epoch": 8.859185918591859,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.020475404167661786,
      "loss": 0.2329,
      "num_input_tokens_seen": 16995344,
      "step": 80530
    },
    {
      "epoch": 8.85973597359736,
      "grad_norm": 0.009765625,
      "learning_rate": 0.020474063468984012,
      "loss": 0.2309,
      "num_input_tokens_seen": 16996368,
      "step": 80535
    },
    {
      "epoch": 8.86028602860286,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02047272271985442,
      "loss": 0.2319,
      "num_input_tokens_seen": 16997456,
      "step": 80540
    },
    {
      "epoch": 8.86083608360836,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.020471381920285373,
      "loss": 0.2319,
      "num_input_tokens_seen": 16998448,
      "step": 80545
    },
    {
      "epoch": 8.861386138613861,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.020470041070289224,
      "loss": 0.2319,
      "num_input_tokens_seen": 16999568,
      "step": 80550
    },
    {
      "epoch": 8.861936193619362,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02046870016987833,
      "loss": 0.2324,
      "num_input_tokens_seen": 17000656,
      "step": 80555
    },
    {
      "epoch": 8.862486248624862,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020467359219065052,
      "loss": 0.2308,
      "num_input_tokens_seen": 17001744,
      "step": 80560
    },
    {
      "epoch": 8.863036303630363,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.02046601821786174,
      "loss": 0.2325,
      "num_input_tokens_seen": 17002800,
      "step": 80565
    },
    {
      "epoch": 8.863586358635864,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020464677166280767,
      "loss": 0.2309,
      "num_input_tokens_seen": 17003824,
      "step": 80570
    },
    {
      "epoch": 8.864136413641365,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.02046333606433449,
      "loss": 0.2314,
      "num_input_tokens_seen": 17004880,
      "step": 80575
    },
    {
      "epoch": 8.864686468646864,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02046199491203526,
      "loss": 0.2314,
      "num_input_tokens_seen": 17005904,
      "step": 80580
    },
    {
      "epoch": 8.865236523652365,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02046065370939545,
      "loss": 0.2299,
      "num_input_tokens_seen": 17006928,
      "step": 80585
    },
    {
      "epoch": 8.865786578657866,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02045931245642741,
      "loss": 0.2314,
      "num_input_tokens_seen": 17008048,
      "step": 80590
    },
    {
      "epoch": 8.866336633663366,
      "grad_norm": 0.000774383544921875,
      "learning_rate": 0.020457971153143507,
      "loss": 0.2324,
      "num_input_tokens_seen": 17009072,
      "step": 80595
    },
    {
      "epoch": 8.866886688668867,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.020456629799556106,
      "loss": 0.2314,
      "num_input_tokens_seen": 17010064,
      "step": 80600
    },
    {
      "epoch": 8.867436743674368,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.020455288395677562,
      "loss": 0.2303,
      "num_input_tokens_seen": 17011120,
      "step": 80605
    },
    {
      "epoch": 8.867986798679867,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.020453946941520244,
      "loss": 0.2314,
      "num_input_tokens_seen": 17012112,
      "step": 80610
    },
    {
      "epoch": 8.868536853685368,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.020452605437096517,
      "loss": 0.233,
      "num_input_tokens_seen": 17013168,
      "step": 80615
    },
    {
      "epoch": 8.86908690869087,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.020451263882418743,
      "loss": 0.2319,
      "num_input_tokens_seen": 17014288,
      "step": 80620
    },
    {
      "epoch": 8.869636963696369,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.020449922277499283,
      "loss": 0.2314,
      "num_input_tokens_seen": 17015344,
      "step": 80625
    },
    {
      "epoch": 8.87018701870187,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020448580622350505,
      "loss": 0.2309,
      "num_input_tokens_seen": 17016400,
      "step": 80630
    },
    {
      "epoch": 8.870737073707371,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.020447238916984772,
      "loss": 0.2314,
      "num_input_tokens_seen": 17017424,
      "step": 80635
    },
    {
      "epoch": 8.871287128712872,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.020445897161414452,
      "loss": 0.2293,
      "num_input_tokens_seen": 17018480,
      "step": 80640
    },
    {
      "epoch": 8.871837183718371,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.02044455535565191,
      "loss": 0.2309,
      "num_input_tokens_seen": 17019536,
      "step": 80645
    },
    {
      "epoch": 8.872387238723872,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02044321349970952,
      "loss": 0.2303,
      "num_input_tokens_seen": 17020688,
      "step": 80650
    },
    {
      "epoch": 8.872937293729374,
      "grad_norm": 0.005859375,
      "learning_rate": 0.02044187159359964,
      "loss": 0.2309,
      "num_input_tokens_seen": 17021776,
      "step": 80655
    },
    {
      "epoch": 8.873487348734873,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.020440529637334635,
      "loss": 0.2329,
      "num_input_tokens_seen": 17022768,
      "step": 80660
    },
    {
      "epoch": 8.874037403740374,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.020439187630926888,
      "loss": 0.2329,
      "num_input_tokens_seen": 17023792,
      "step": 80665
    },
    {
      "epoch": 8.874587458745875,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.020437845574388755,
      "loss": 0.2314,
      "num_input_tokens_seen": 17024816,
      "step": 80670
    },
    {
      "epoch": 8.875137513751374,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.020436503467732608,
      "loss": 0.2329,
      "num_input_tokens_seen": 17025872,
      "step": 80675
    },
    {
      "epoch": 8.875687568756875,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.020435161310970815,
      "loss": 0.2314,
      "num_input_tokens_seen": 17026960,
      "step": 80680
    },
    {
      "epoch": 8.876237623762377,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.02043381910411575,
      "loss": 0.2314,
      "num_input_tokens_seen": 17028048,
      "step": 80685
    },
    {
      "epoch": 8.876787678767876,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.020432476847179783,
      "loss": 0.2309,
      "num_input_tokens_seen": 17029104,
      "step": 80690
    },
    {
      "epoch": 8.877337733773377,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.02043113454017528,
      "loss": 0.2319,
      "num_input_tokens_seen": 17030192,
      "step": 80695
    },
    {
      "epoch": 8.877887788778878,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.02042979218311462,
      "loss": 0.2335,
      "num_input_tokens_seen": 17031280,
      "step": 80700
    },
    {
      "epoch": 8.87843784378438,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.02042844977601017,
      "loss": 0.2288,
      "num_input_tokens_seen": 17032272,
      "step": 80705
    },
    {
      "epoch": 8.878987898789878,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.020427107318874302,
      "loss": 0.2319,
      "num_input_tokens_seen": 17033296,
      "step": 80710
    },
    {
      "epoch": 8.87953795379538,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.020425764811719392,
      "loss": 0.2293,
      "num_input_tokens_seen": 17034352,
      "step": 80715
    },
    {
      "epoch": 8.88008800880088,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020424422254557813,
      "loss": 0.234,
      "num_input_tokens_seen": 17035440,
      "step": 80720
    },
    {
      "epoch": 8.88063806380638,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02042307964740193,
      "loss": 0.2356,
      "num_input_tokens_seen": 17036560,
      "step": 80725
    },
    {
      "epoch": 8.881188118811881,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02042173699026413,
      "loss": 0.2314,
      "num_input_tokens_seen": 17037616,
      "step": 80730
    },
    {
      "epoch": 8.881738173817382,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02042039428315678,
      "loss": 0.234,
      "num_input_tokens_seen": 17038576,
      "step": 80735
    },
    {
      "epoch": 8.882288228822881,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020419051526092254,
      "loss": 0.2324,
      "num_input_tokens_seen": 17039632,
      "step": 80740
    },
    {
      "epoch": 8.882838283828383,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.02041770871908294,
      "loss": 0.2314,
      "num_input_tokens_seen": 17040688,
      "step": 80745
    },
    {
      "epoch": 8.883388338833884,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.020416365862141198,
      "loss": 0.2308,
      "num_input_tokens_seen": 17041744,
      "step": 80750
    },
    {
      "epoch": 8.883938393839383,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020415022955279406,
      "loss": 0.2309,
      "num_input_tokens_seen": 17042896,
      "step": 80755
    },
    {
      "epoch": 8.884488448844884,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.020413679998509952,
      "loss": 0.2319,
      "num_input_tokens_seen": 17044080,
      "step": 80760
    },
    {
      "epoch": 8.885038503850385,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.020412336991845213,
      "loss": 0.2303,
      "num_input_tokens_seen": 17045104,
      "step": 80765
    },
    {
      "epoch": 8.885588558855886,
      "grad_norm": 0.0027313232421875,
      "learning_rate": 0.02041099393529755,
      "loss": 0.2293,
      "num_input_tokens_seen": 17046224,
      "step": 80770
    },
    {
      "epoch": 8.886138613861386,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.020409650828879355,
      "loss": 0.2324,
      "num_input_tokens_seen": 17047280,
      "step": 80775
    },
    {
      "epoch": 8.886688668866887,
      "grad_norm": 0.0006256103515625,
      "learning_rate": 0.020408307672603005,
      "loss": 0.2335,
      "num_input_tokens_seen": 17048304,
      "step": 80780
    },
    {
      "epoch": 8.887238723872388,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020406964466480878,
      "loss": 0.2319,
      "num_input_tokens_seen": 17049360,
      "step": 80785
    },
    {
      "epoch": 8.887788778877887,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020405621210525356,
      "loss": 0.2308,
      "num_input_tokens_seen": 17050448,
      "step": 80790
    },
    {
      "epoch": 8.888338833883388,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.020404277904748816,
      "loss": 0.2324,
      "num_input_tokens_seen": 17051472,
      "step": 80795
    },
    {
      "epoch": 8.88888888888889,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.02040293454916364,
      "loss": 0.2314,
      "num_input_tokens_seen": 17052496,
      "step": 80800
    },
    {
      "epoch": 8.88943894389439,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020401591143782205,
      "loss": 0.2314,
      "num_input_tokens_seen": 17053488,
      "step": 80805
    },
    {
      "epoch": 8.88998899889989,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.0204002476886169,
      "loss": 0.2324,
      "num_input_tokens_seen": 17054608,
      "step": 80810
    },
    {
      "epoch": 8.89053905390539,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.020398904183680105,
      "loss": 0.2319,
      "num_input_tokens_seen": 17055600,
      "step": 80815
    },
    {
      "epoch": 8.891089108910892,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.0203975606289842,
      "loss": 0.2319,
      "num_input_tokens_seen": 17056688,
      "step": 80820
    },
    {
      "epoch": 8.891639163916391,
      "grad_norm": 0.0029144287109375,
      "learning_rate": 0.020396217024541568,
      "loss": 0.2319,
      "num_input_tokens_seen": 17057712,
      "step": 80825
    },
    {
      "epoch": 8.892189218921892,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02039487337036459,
      "loss": 0.2329,
      "num_input_tokens_seen": 17058736,
      "step": 80830
    },
    {
      "epoch": 8.892739273927393,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.020393529666465654,
      "loss": 0.2329,
      "num_input_tokens_seen": 17059728,
      "step": 80835
    },
    {
      "epoch": 8.893289328932893,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.020392185912857148,
      "loss": 0.234,
      "num_input_tokens_seen": 17060816,
      "step": 80840
    },
    {
      "epoch": 8.893839383938394,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020390842109551447,
      "loss": 0.2324,
      "num_input_tokens_seen": 17061840,
      "step": 80845
    },
    {
      "epoch": 8.894389438943895,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02038949825656094,
      "loss": 0.2309,
      "num_input_tokens_seen": 17062992,
      "step": 80850
    },
    {
      "epoch": 8.894939493949394,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.020388154353898014,
      "loss": 0.2309,
      "num_input_tokens_seen": 17064048,
      "step": 80855
    },
    {
      "epoch": 8.895489548954895,
      "grad_norm": 0.0007476806640625,
      "learning_rate": 0.02038681040157506,
      "loss": 0.2309,
      "num_input_tokens_seen": 17065104,
      "step": 80860
    },
    {
      "epoch": 8.896039603960396,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02038546639960446,
      "loss": 0.2314,
      "num_input_tokens_seen": 17066160,
      "step": 80865
    },
    {
      "epoch": 8.896589658965897,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.020384122347998596,
      "loss": 0.2319,
      "num_input_tokens_seen": 17067184,
      "step": 80870
    },
    {
      "epoch": 8.897139713971397,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.020382778246769862,
      "loss": 0.2324,
      "num_input_tokens_seen": 17068272,
      "step": 80875
    },
    {
      "epoch": 8.897689768976898,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.020381434095930642,
      "loss": 0.2293,
      "num_input_tokens_seen": 17069328,
      "step": 80880
    },
    {
      "epoch": 8.898239823982399,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.020380089895493326,
      "loss": 0.2325,
      "num_input_tokens_seen": 17070384,
      "step": 80885
    },
    {
      "epoch": 8.898789878987898,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.020378745645470306,
      "loss": 0.2309,
      "num_input_tokens_seen": 17071440,
      "step": 80890
    },
    {
      "epoch": 8.8993399339934,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.020377401345873965,
      "loss": 0.2319,
      "num_input_tokens_seen": 17072464,
      "step": 80895
    },
    {
      "epoch": 8.8998899889989,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.020376056996716692,
      "loss": 0.2309,
      "num_input_tokens_seen": 17073488,
      "step": 80900
    },
    {
      "epoch": 8.9004400440044,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02037471259801089,
      "loss": 0.2309,
      "num_input_tokens_seen": 17074512,
      "step": 80905
    },
    {
      "epoch": 8.900990099009901,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02037336814976894,
      "loss": 0.2351,
      "num_input_tokens_seen": 17075568,
      "step": 80910
    },
    {
      "epoch": 8.901540154015402,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02037202365200323,
      "loss": 0.2314,
      "num_input_tokens_seen": 17076624,
      "step": 80915
    },
    {
      "epoch": 8.902090209020901,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.020370679104726157,
      "loss": 0.2304,
      "num_input_tokens_seen": 17077680,
      "step": 80920
    },
    {
      "epoch": 8.902640264026402,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020369334507950112,
      "loss": 0.233,
      "num_input_tokens_seen": 17078640,
      "step": 80925
    },
    {
      "epoch": 8.903190319031903,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.020367989861687485,
      "loss": 0.2309,
      "num_input_tokens_seen": 17079664,
      "step": 80930
    },
    {
      "epoch": 8.903740374037405,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.02036664516595068,
      "loss": 0.2298,
      "num_input_tokens_seen": 17080656,
      "step": 80935
    },
    {
      "epoch": 8.904290429042904,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02036530042075207,
      "loss": 0.2319,
      "num_input_tokens_seen": 17081712,
      "step": 80940
    },
    {
      "epoch": 8.904840484048405,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.020363955626104067,
      "loss": 0.2324,
      "num_input_tokens_seen": 17082800,
      "step": 80945
    },
    {
      "epoch": 8.905390539053906,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.02036261078201906,
      "loss": 0.2314,
      "num_input_tokens_seen": 17083856,
      "step": 80950
    },
    {
      "epoch": 8.905940594059405,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02036126588850944,
      "loss": 0.2304,
      "num_input_tokens_seen": 17084912,
      "step": 80955
    },
    {
      "epoch": 8.906490649064907,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020359920945587605,
      "loss": 0.2308,
      "num_input_tokens_seen": 17085936,
      "step": 80960
    },
    {
      "epoch": 8.907040704070408,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.020358575953265953,
      "loss": 0.2303,
      "num_input_tokens_seen": 17087024,
      "step": 80965
    },
    {
      "epoch": 8.907590759075907,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020357230911556876,
      "loss": 0.2314,
      "num_input_tokens_seen": 17088112,
      "step": 80970
    },
    {
      "epoch": 8.908140814081408,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.02035588582047277,
      "loss": 0.2324,
      "num_input_tokens_seen": 17089168,
      "step": 80975
    },
    {
      "epoch": 8.908690869086909,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02035454068002604,
      "loss": 0.2309,
      "num_input_tokens_seen": 17090192,
      "step": 80980
    },
    {
      "epoch": 8.909240924092408,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.02035319549022907,
      "loss": 0.2319,
      "num_input_tokens_seen": 17091280,
      "step": 80985
    },
    {
      "epoch": 8.90979097909791,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02035185025109427,
      "loss": 0.2329,
      "num_input_tokens_seen": 17092304,
      "step": 80990
    },
    {
      "epoch": 8.91034103410341,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020350504962634037,
      "loss": 0.2309,
      "num_input_tokens_seen": 17093360,
      "step": 80995
    },
    {
      "epoch": 8.910891089108912,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.020349159624860766,
      "loss": 0.2309,
      "num_input_tokens_seen": 17094416,
      "step": 81000
    },
    {
      "epoch": 8.911441144114411,
      "grad_norm": 0.0003414154052734375,
      "learning_rate": 0.020347814237786855,
      "loss": 0.2303,
      "num_input_tokens_seen": 17095376,
      "step": 81005
    },
    {
      "epoch": 8.911991199119912,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.02034646880142471,
      "loss": 0.2308,
      "num_input_tokens_seen": 17096400,
      "step": 81010
    },
    {
      "epoch": 8.912541254125413,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020345123315786723,
      "loss": 0.2319,
      "num_input_tokens_seen": 17097488,
      "step": 81015
    },
    {
      "epoch": 8.913091309130913,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.0203437777808853,
      "loss": 0.2298,
      "num_input_tokens_seen": 17098576,
      "step": 81020
    },
    {
      "epoch": 8.913641364136414,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020342432196732842,
      "loss": 0.2319,
      "num_input_tokens_seen": 17099664,
      "step": 81025
    },
    {
      "epoch": 8.914191419141915,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.020341086563341753,
      "loss": 0.2288,
      "num_input_tokens_seen": 17100752,
      "step": 81030
    },
    {
      "epoch": 8.914741474147414,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02033974088072443,
      "loss": 0.2319,
      "num_input_tokens_seen": 17101808,
      "step": 81035
    },
    {
      "epoch": 8.915291529152915,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.02033839514889328,
      "loss": 0.2298,
      "num_input_tokens_seen": 17102864,
      "step": 81040
    },
    {
      "epoch": 8.915841584158416,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.020337049367860696,
      "loss": 0.2335,
      "num_input_tokens_seen": 17103952,
      "step": 81045
    },
    {
      "epoch": 8.916391639163916,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020335703537639093,
      "loss": 0.2304,
      "num_input_tokens_seen": 17105040,
      "step": 81050
    },
    {
      "epoch": 8.916941694169417,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.020334357658240876,
      "loss": 0.2329,
      "num_input_tokens_seen": 17106128,
      "step": 81055
    },
    {
      "epoch": 8.917491749174918,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.020333011729678437,
      "loss": 0.2314,
      "num_input_tokens_seen": 17107152,
      "step": 81060
    },
    {
      "epoch": 8.918041804180419,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02033166575196419,
      "loss": 0.2303,
      "num_input_tokens_seen": 17108144,
      "step": 81065
    },
    {
      "epoch": 8.918591859185918,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.020330319725110537,
      "loss": 0.2293,
      "num_input_tokens_seen": 17109200,
      "step": 81070
    },
    {
      "epoch": 8.91914191419142,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.020328973649129892,
      "loss": 0.233,
      "num_input_tokens_seen": 17110224,
      "step": 81075
    },
    {
      "epoch": 8.91969196919692,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.020327627524034646,
      "loss": 0.2293,
      "num_input_tokens_seen": 17111248,
      "step": 81080
    },
    {
      "epoch": 8.92024202420242,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.02032628134983722,
      "loss": 0.233,
      "num_input_tokens_seen": 17112272,
      "step": 81085
    },
    {
      "epoch": 8.92079207920792,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.020324935126550005,
      "loss": 0.2314,
      "num_input_tokens_seen": 17113328,
      "step": 81090
    },
    {
      "epoch": 8.921342134213422,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.020323588854185424,
      "loss": 0.2309,
      "num_input_tokens_seen": 17114448,
      "step": 81095
    },
    {
      "epoch": 8.921892189218921,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020322242532755882,
      "loss": 0.2304,
      "num_input_tokens_seen": 17115472,
      "step": 81100
    },
    {
      "epoch": 8.922442244224422,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.020320896162273783,
      "loss": 0.2303,
      "num_input_tokens_seen": 17116464,
      "step": 81105
    },
    {
      "epoch": 8.922992299229923,
      "grad_norm": 0.00286865234375,
      "learning_rate": 0.020319549742751533,
      "loss": 0.2309,
      "num_input_tokens_seen": 17117520,
      "step": 81110
    },
    {
      "epoch": 8.923542354235423,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02031820327420155,
      "loss": 0.2304,
      "num_input_tokens_seen": 17118608,
      "step": 81115
    },
    {
      "epoch": 8.924092409240924,
      "grad_norm": 0.005859375,
      "learning_rate": 0.020316856756636237,
      "loss": 0.2335,
      "num_input_tokens_seen": 17119728,
      "step": 81120
    },
    {
      "epoch": 8.924642464246425,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02031551019006801,
      "loss": 0.2345,
      "num_input_tokens_seen": 17120816,
      "step": 81125
    },
    {
      "epoch": 8.925192519251926,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.020314163574509266,
      "loss": 0.2283,
      "num_input_tokens_seen": 17121904,
      "step": 81130
    },
    {
      "epoch": 8.925742574257425,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02031281690997244,
      "loss": 0.2304,
      "num_input_tokens_seen": 17122960,
      "step": 81135
    },
    {
      "epoch": 8.926292629262926,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02031147019646992,
      "loss": 0.2314,
      "num_input_tokens_seen": 17124016,
      "step": 81140
    },
    {
      "epoch": 8.926842684268427,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.020310123434014133,
      "loss": 0.2309,
      "num_input_tokens_seen": 17125136,
      "step": 81145
    },
    {
      "epoch": 8.927392739273927,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.020308776622617482,
      "loss": 0.2308,
      "num_input_tokens_seen": 17126192,
      "step": 81150
    },
    {
      "epoch": 8.927942794279428,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.02030742976229239,
      "loss": 0.2319,
      "num_input_tokens_seen": 17127248,
      "step": 81155
    },
    {
      "epoch": 8.928492849284929,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.02030608285305126,
      "loss": 0.2304,
      "num_input_tokens_seen": 17128208,
      "step": 81160
    },
    {
      "epoch": 8.929042904290428,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.020304735894906514,
      "loss": 0.233,
      "num_input_tokens_seen": 17129232,
      "step": 81165
    },
    {
      "epoch": 8.92959295929593,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.02030338888787056,
      "loss": 0.2319,
      "num_input_tokens_seen": 17130256,
      "step": 81170
    },
    {
      "epoch": 8.93014301430143,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.020302041831955813,
      "loss": 0.233,
      "num_input_tokens_seen": 17131312,
      "step": 81175
    },
    {
      "epoch": 8.930693069306932,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.020300694727174697,
      "loss": 0.2319,
      "num_input_tokens_seen": 17132464,
      "step": 81180
    },
    {
      "epoch": 8.93124312431243,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.020299347573539617,
      "loss": 0.2314,
      "num_input_tokens_seen": 17133520,
      "step": 81185
    },
    {
      "epoch": 8.931793179317932,
      "grad_norm": 0.0003986358642578125,
      "learning_rate": 0.020298000371062996,
      "loss": 0.2308,
      "num_input_tokens_seen": 17134544,
      "step": 81190
    },
    {
      "epoch": 8.932343234323433,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.020296653119757247,
      "loss": 0.2324,
      "num_input_tokens_seen": 17135536,
      "step": 81195
    },
    {
      "epoch": 8.932893289328932,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.020295305819634787,
      "loss": 0.2319,
      "num_input_tokens_seen": 17136624,
      "step": 81200
    },
    {
      "epoch": 8.933443344334433,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02029395847070803,
      "loss": 0.2329,
      "num_input_tokens_seen": 17137616,
      "step": 81205
    },
    {
      "epoch": 8.933993399339935,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.020292611072989405,
      "loss": 0.2314,
      "num_input_tokens_seen": 17138672,
      "step": 81210
    },
    {
      "epoch": 8.934543454345434,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02029126362649132,
      "loss": 0.2288,
      "num_input_tokens_seen": 17139728,
      "step": 81215
    },
    {
      "epoch": 8.935093509350935,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.0202899161312262,
      "loss": 0.2309,
      "num_input_tokens_seen": 17140848,
      "step": 81220
    },
    {
      "epoch": 8.935643564356436,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02028856858720646,
      "loss": 0.2325,
      "num_input_tokens_seen": 17141936,
      "step": 81225
    },
    {
      "epoch": 8.936193619361937,
      "grad_norm": 0.009765625,
      "learning_rate": 0.02028722099444452,
      "loss": 0.2314,
      "num_input_tokens_seen": 17142992,
      "step": 81230
    },
    {
      "epoch": 8.936743674367436,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.0202858733529528,
      "loss": 0.2309,
      "num_input_tokens_seen": 17144048,
      "step": 81235
    },
    {
      "epoch": 8.937293729372938,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020284525662743722,
      "loss": 0.233,
      "num_input_tokens_seen": 17145072,
      "step": 81240
    },
    {
      "epoch": 8.937843784378439,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02028317792382971,
      "loss": 0.2309,
      "num_input_tokens_seen": 17146128,
      "step": 81245
    },
    {
      "epoch": 8.938393839383938,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02028183013622318,
      "loss": 0.2309,
      "num_input_tokens_seen": 17147216,
      "step": 81250
    },
    {
      "epoch": 8.938943894389439,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.020280482299936557,
      "loss": 0.2324,
      "num_input_tokens_seen": 17148304,
      "step": 81255
    },
    {
      "epoch": 8.93949394939494,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020279134414982258,
      "loss": 0.2329,
      "num_input_tokens_seen": 17149360,
      "step": 81260
    },
    {
      "epoch": 8.94004400440044,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020277786481372715,
      "loss": 0.2325,
      "num_input_tokens_seen": 17150448,
      "step": 81265
    },
    {
      "epoch": 8.94059405940594,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020276438499120347,
      "loss": 0.2324,
      "num_input_tokens_seen": 17151440,
      "step": 81270
    },
    {
      "epoch": 8.941144114411442,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.020275090468237576,
      "loss": 0.2319,
      "num_input_tokens_seen": 17152496,
      "step": 81275
    },
    {
      "epoch": 8.941694169416941,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.020273742388736825,
      "loss": 0.2319,
      "num_input_tokens_seen": 17153520,
      "step": 81280
    },
    {
      "epoch": 8.942244224422442,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.02027239426063052,
      "loss": 0.2319,
      "num_input_tokens_seen": 17154608,
      "step": 81285
    },
    {
      "epoch": 8.942794279427943,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020271046083931092,
      "loss": 0.2319,
      "num_input_tokens_seen": 17155632,
      "step": 81290
    },
    {
      "epoch": 8.943344334433444,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.020269697858650963,
      "loss": 0.2319,
      "num_input_tokens_seen": 17156656,
      "step": 81295
    },
    {
      "epoch": 8.943894389438944,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.02026834958480255,
      "loss": 0.2309,
      "num_input_tokens_seen": 17157744,
      "step": 81300
    },
    {
      "epoch": 8.944444444444445,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.020267001262398294,
      "loss": 0.2329,
      "num_input_tokens_seen": 17158832,
      "step": 81305
    },
    {
      "epoch": 8.944994499449946,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02026565289145061,
      "loss": 0.2314,
      "num_input_tokens_seen": 17159888,
      "step": 81310
    },
    {
      "epoch": 8.945544554455445,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020264304471971934,
      "loss": 0.2309,
      "num_input_tokens_seen": 17160976,
      "step": 81315
    },
    {
      "epoch": 8.946094609460946,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.020262956003974684,
      "loss": 0.2329,
      "num_input_tokens_seen": 17162032,
      "step": 81320
    },
    {
      "epoch": 8.946644664466447,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.0202616074874713,
      "loss": 0.2308,
      "num_input_tokens_seen": 17163120,
      "step": 81325
    },
    {
      "epoch": 8.947194719471947,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.0202602589224742,
      "loss": 0.2314,
      "num_input_tokens_seen": 17164176,
      "step": 81330
    },
    {
      "epoch": 8.947744774477448,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.02025891030899582,
      "loss": 0.2314,
      "num_input_tokens_seen": 17165232,
      "step": 81335
    },
    {
      "epoch": 8.948294829482949,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.02025756164704859,
      "loss": 0.233,
      "num_input_tokens_seen": 17166192,
      "step": 81340
    },
    {
      "epoch": 8.948844884488448,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02025621293664493,
      "loss": 0.2298,
      "num_input_tokens_seen": 17167248,
      "step": 81345
    },
    {
      "epoch": 8.94939493949395,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020254864177797278,
      "loss": 0.2329,
      "num_input_tokens_seen": 17168336,
      "step": 81350
    },
    {
      "epoch": 8.94994499449945,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020253515370518067,
      "loss": 0.2309,
      "num_input_tokens_seen": 17169424,
      "step": 81355
    },
    {
      "epoch": 8.950495049504951,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.020252166514819734,
      "loss": 0.2314,
      "num_input_tokens_seen": 17170480,
      "step": 81360
    },
    {
      "epoch": 8.95104510451045,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.02025081761071469,
      "loss": 0.2293,
      "num_input_tokens_seen": 17171536,
      "step": 81365
    },
    {
      "epoch": 8.951595159515952,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.020249468658215388,
      "loss": 0.2314,
      "num_input_tokens_seen": 17172624,
      "step": 81370
    },
    {
      "epoch": 8.952145214521453,
      "grad_norm": 0.001953125,
      "learning_rate": 0.020248119657334248,
      "loss": 0.2303,
      "num_input_tokens_seen": 17173680,
      "step": 81375
    },
    {
      "epoch": 8.952695269526952,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.020246770608083706,
      "loss": 0.2309,
      "num_input_tokens_seen": 17174704,
      "step": 81380
    },
    {
      "epoch": 8.953245324532453,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.0202454215104762,
      "loss": 0.2304,
      "num_input_tokens_seen": 17175792,
      "step": 81385
    },
    {
      "epoch": 8.953795379537954,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02024407236452416,
      "loss": 0.2309,
      "num_input_tokens_seen": 17176880,
      "step": 81390
    },
    {
      "epoch": 8.954345434543454,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.02024272317024002,
      "loss": 0.2314,
      "num_input_tokens_seen": 17177968,
      "step": 81395
    },
    {
      "epoch": 8.954895489548955,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.02024137392763622,
      "loss": 0.2298,
      "num_input_tokens_seen": 17179056,
      "step": 81400
    },
    {
      "epoch": 8.955445544554456,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02024002463672519,
      "loss": 0.2319,
      "num_input_tokens_seen": 17180144,
      "step": 81405
    },
    {
      "epoch": 8.955995599559955,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.020238675297519364,
      "loss": 0.2314,
      "num_input_tokens_seen": 17181168,
      "step": 81410
    },
    {
      "epoch": 8.956545654565456,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.020237325910031186,
      "loss": 0.2335,
      "num_input_tokens_seen": 17182288,
      "step": 81415
    },
    {
      "epoch": 8.957095709570957,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.02023597647427309,
      "loss": 0.2303,
      "num_input_tokens_seen": 17183344,
      "step": 81420
    },
    {
      "epoch": 8.957645764576458,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020234626990257503,
      "loss": 0.2303,
      "num_input_tokens_seen": 17184400,
      "step": 81425
    },
    {
      "epoch": 8.958195819581958,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.020233277457996874,
      "loss": 0.2309,
      "num_input_tokens_seen": 17185424,
      "step": 81430
    },
    {
      "epoch": 8.958745874587459,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.02023192787750364,
      "loss": 0.2298,
      "num_input_tokens_seen": 17186448,
      "step": 81435
    },
    {
      "epoch": 8.95929592959296,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02023057824879024,
      "loss": 0.2329,
      "num_input_tokens_seen": 17187440,
      "step": 81440
    },
    {
      "epoch": 8.95984598459846,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020229228571869106,
      "loss": 0.2319,
      "num_input_tokens_seen": 17188624,
      "step": 81445
    },
    {
      "epoch": 8.96039603960396,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.020227878846752683,
      "loss": 0.2293,
      "num_input_tokens_seen": 17189744,
      "step": 81450
    },
    {
      "epoch": 8.960946094609461,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.020226529073453407,
      "loss": 0.2325,
      "num_input_tokens_seen": 17190800,
      "step": 81455
    },
    {
      "epoch": 8.96149614961496,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.02022517925198372,
      "loss": 0.2314,
      "num_input_tokens_seen": 17191824,
      "step": 81460
    },
    {
      "epoch": 8.962046204620462,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.020223829382356066,
      "loss": 0.2298,
      "num_input_tokens_seen": 17192880,
      "step": 81465
    },
    {
      "epoch": 8.962596259625963,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.020222479464582884,
      "loss": 0.2314,
      "num_input_tokens_seen": 17193936,
      "step": 81470
    },
    {
      "epoch": 8.963146314631462,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.02022112949867661,
      "loss": 0.2335,
      "num_input_tokens_seen": 17194960,
      "step": 81475
    },
    {
      "epoch": 8.963696369636963,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.0202197794846497,
      "loss": 0.2314,
      "num_input_tokens_seen": 17195984,
      "step": 81480
    },
    {
      "epoch": 8.964246424642464,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.02021842942251458,
      "loss": 0.234,
      "num_input_tokens_seen": 17197040,
      "step": 81485
    },
    {
      "epoch": 8.964796479647966,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.020217079312283698,
      "loss": 0.2324,
      "num_input_tokens_seen": 17198096,
      "step": 81490
    },
    {
      "epoch": 8.965346534653465,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.020215729153969502,
      "loss": 0.2304,
      "num_input_tokens_seen": 17199120,
      "step": 81495
    },
    {
      "epoch": 8.965896589658966,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.020214378947584433,
      "loss": 0.2298,
      "num_input_tokens_seen": 17200112,
      "step": 81500
    },
    {
      "epoch": 8.966446644664467,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02021302869314094,
      "loss": 0.2324,
      "num_input_tokens_seen": 17201200,
      "step": 81505
    },
    {
      "epoch": 8.966996699669966,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.020211678390651456,
      "loss": 0.2335,
      "num_input_tokens_seen": 17202224,
      "step": 81510
    },
    {
      "epoch": 8.967546754675467,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.020210328040128435,
      "loss": 0.2309,
      "num_input_tokens_seen": 17203312,
      "step": 81515
    },
    {
      "epoch": 8.968096809680969,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.020208977641584323,
      "loss": 0.2314,
      "num_input_tokens_seen": 17204368,
      "step": 81520
    },
    {
      "epoch": 8.968646864686468,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.02020762719503156,
      "loss": 0.2319,
      "num_input_tokens_seen": 17205392,
      "step": 81525
    },
    {
      "epoch": 8.969196919691969,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.0202062767004826,
      "loss": 0.2324,
      "num_input_tokens_seen": 17206416,
      "step": 81530
    },
    {
      "epoch": 8.96974697469747,
      "grad_norm": 0.000782012939453125,
      "learning_rate": 0.020204926157949884,
      "loss": 0.2314,
      "num_input_tokens_seen": 17207440,
      "step": 81535
    },
    {
      "epoch": 8.97029702970297,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.02020357556744586,
      "loss": 0.233,
      "num_input_tokens_seen": 17208432,
      "step": 81540
    },
    {
      "epoch": 8.97084708470847,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02020222492898298,
      "loss": 0.2319,
      "num_input_tokens_seen": 17209520,
      "step": 81545
    },
    {
      "epoch": 8.971397139713972,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.02020087424257369,
      "loss": 0.2319,
      "num_input_tokens_seen": 17210544,
      "step": 81550
    },
    {
      "epoch": 8.971947194719473,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02019952350823043,
      "loss": 0.2319,
      "num_input_tokens_seen": 17211632,
      "step": 81555
    },
    {
      "epoch": 8.972497249724972,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020198172725965663,
      "loss": 0.2288,
      "num_input_tokens_seen": 17212688,
      "step": 81560
    },
    {
      "epoch": 8.973047304730473,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.020196821895791833,
      "loss": 0.2314,
      "num_input_tokens_seen": 17213744,
      "step": 81565
    },
    {
      "epoch": 8.973597359735974,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.020195471017721386,
      "loss": 0.2319,
      "num_input_tokens_seen": 17214736,
      "step": 81570
    },
    {
      "epoch": 8.974147414741473,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.02019412009176678,
      "loss": 0.2309,
      "num_input_tokens_seen": 17215824,
      "step": 81575
    },
    {
      "epoch": 8.974697469746975,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.02019276911794046,
      "loss": 0.2293,
      "num_input_tokens_seen": 17216912,
      "step": 81580
    },
    {
      "epoch": 8.975247524752476,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020191418096254875,
      "loss": 0.2345,
      "num_input_tokens_seen": 17217904,
      "step": 81585
    },
    {
      "epoch": 8.975797579757975,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020190067026722484,
      "loss": 0.2304,
      "num_input_tokens_seen": 17218928,
      "step": 81590
    },
    {
      "epoch": 8.976347634763476,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.020188715909355733,
      "loss": 0.2304,
      "num_input_tokens_seen": 17220016,
      "step": 81595
    },
    {
      "epoch": 8.976897689768977,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.020187364744167078,
      "loss": 0.2346,
      "num_input_tokens_seen": 17221008,
      "step": 81600
    },
    {
      "epoch": 8.977447744774478,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.020186013531168973,
      "loss": 0.2309,
      "num_input_tokens_seen": 17222032,
      "step": 81605
    },
    {
      "epoch": 8.977997799779978,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.020184662270373868,
      "loss": 0.2304,
      "num_input_tokens_seen": 17223056,
      "step": 81610
    },
    {
      "epoch": 8.978547854785479,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.02018331096179422,
      "loss": 0.2304,
      "num_input_tokens_seen": 17224176,
      "step": 81615
    },
    {
      "epoch": 8.97909790979098,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02018195960544248,
      "loss": 0.233,
      "num_input_tokens_seen": 17225232,
      "step": 81620
    },
    {
      "epoch": 8.979647964796479,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.020180608201331107,
      "loss": 0.2341,
      "num_input_tokens_seen": 17226256,
      "step": 81625
    },
    {
      "epoch": 8.98019801980198,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.020179256749472553,
      "loss": 0.2314,
      "num_input_tokens_seen": 17227280,
      "step": 81630
    },
    {
      "epoch": 8.980748074807481,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.020177905249879276,
      "loss": 0.2319,
      "num_input_tokens_seen": 17228368,
      "step": 81635
    },
    {
      "epoch": 8.98129812981298,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.02017655370256373,
      "loss": 0.2309,
      "num_input_tokens_seen": 17229424,
      "step": 81640
    },
    {
      "epoch": 8.981848184818482,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.020175202107538374,
      "loss": 0.2309,
      "num_input_tokens_seen": 17230544,
      "step": 81645
    },
    {
      "epoch": 8.982398239823983,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.020173850464815665,
      "loss": 0.2325,
      "num_input_tokens_seen": 17231664,
      "step": 81650
    },
    {
      "epoch": 8.982948294829484,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.020172498774408052,
      "loss": 0.2304,
      "num_input_tokens_seen": 17232720,
      "step": 81655
    },
    {
      "epoch": 8.983498349834983,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.020171147036328003,
      "loss": 0.2309,
      "num_input_tokens_seen": 17233712,
      "step": 81660
    },
    {
      "epoch": 8.984048404840484,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.020169795250587975,
      "loss": 0.2299,
      "num_input_tokens_seen": 17234768,
      "step": 81665
    },
    {
      "epoch": 8.984598459845985,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.020168443417200425,
      "loss": 0.232,
      "num_input_tokens_seen": 17235824,
      "step": 81670
    },
    {
      "epoch": 8.985148514851485,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.020167091536177812,
      "loss": 0.2346,
      "num_input_tokens_seen": 17236880,
      "step": 81675
    },
    {
      "epoch": 8.985698569856986,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.020165739607532596,
      "loss": 0.2346,
      "num_input_tokens_seen": 17237936,
      "step": 81680
    },
    {
      "epoch": 8.986248624862487,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02016438763127723,
      "loss": 0.2329,
      "num_input_tokens_seen": 17238992,
      "step": 81685
    },
    {
      "epoch": 8.986798679867986,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.020163035607424187,
      "loss": 0.2329,
      "num_input_tokens_seen": 17239984,
      "step": 81690
    },
    {
      "epoch": 8.987348734873487,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.020161683535985925,
      "loss": 0.2324,
      "num_input_tokens_seen": 17241104,
      "step": 81695
    },
    {
      "epoch": 8.987898789878988,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.0201603314169749,
      "loss": 0.2319,
      "num_input_tokens_seen": 17242128,
      "step": 81700
    },
    {
      "epoch": 8.988448844884488,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.020158979250403577,
      "loss": 0.2309,
      "num_input_tokens_seen": 17243216,
      "step": 81705
    },
    {
      "epoch": 8.988998899889989,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.020157627036284417,
      "loss": 0.2298,
      "num_input_tokens_seen": 17244368,
      "step": 81710
    },
    {
      "epoch": 8.98954895489549,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.02015627477462988,
      "loss": 0.2324,
      "num_input_tokens_seen": 17245456,
      "step": 81715
    },
    {
      "epoch": 8.990099009900991,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.02015492246545244,
      "loss": 0.2293,
      "num_input_tokens_seen": 17246480,
      "step": 81720
    },
    {
      "epoch": 8.99064906490649,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02015357010876455,
      "loss": 0.2293,
      "num_input_tokens_seen": 17247504,
      "step": 81725
    },
    {
      "epoch": 8.991199119911991,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.020152217704578673,
      "loss": 0.2319,
      "num_input_tokens_seen": 17248560,
      "step": 81730
    },
    {
      "epoch": 8.991749174917492,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.020150865252907282,
      "loss": 0.2325,
      "num_input_tokens_seen": 17249616,
      "step": 81735
    },
    {
      "epoch": 8.992299229922992,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.020149512753762838,
      "loss": 0.2288,
      "num_input_tokens_seen": 17250704,
      "step": 81740
    },
    {
      "epoch": 8.992849284928493,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.020148160207157807,
      "loss": 0.2319,
      "num_input_tokens_seen": 17251856,
      "step": 81745
    },
    {
      "epoch": 8.993399339933994,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02014680761310465,
      "loss": 0.2314,
      "num_input_tokens_seen": 17252912,
      "step": 81750
    },
    {
      "epoch": 8.993949394939493,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.020145454971615837,
      "loss": 0.2319,
      "num_input_tokens_seen": 17254064,
      "step": 81755
    },
    {
      "epoch": 8.994499449944994,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.020144102282703833,
      "loss": 0.2309,
      "num_input_tokens_seen": 17255088,
      "step": 81760
    },
    {
      "epoch": 8.995049504950495,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.020142749546381115,
      "loss": 0.2314,
      "num_input_tokens_seen": 17256112,
      "step": 81765
    },
    {
      "epoch": 8.995599559955995,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.020141396762660132,
      "loss": 0.2319,
      "num_input_tokens_seen": 17257104,
      "step": 81770
    },
    {
      "epoch": 8.996149614961496,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.02014004393155337,
      "loss": 0.2329,
      "num_input_tokens_seen": 17258096,
      "step": 81775
    },
    {
      "epoch": 8.996699669966997,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.020138691053073286,
      "loss": 0.2324,
      "num_input_tokens_seen": 17259216,
      "step": 81780
    },
    {
      "epoch": 8.997249724972498,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020137338127232354,
      "loss": 0.2351,
      "num_input_tokens_seen": 17260336,
      "step": 81785
    },
    {
      "epoch": 8.997799779977997,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.02013598515404304,
      "loss": 0.2329,
      "num_input_tokens_seen": 17261328,
      "step": 81790
    },
    {
      "epoch": 8.998349834983498,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020134632133517814,
      "loss": 0.2298,
      "num_input_tokens_seen": 17262320,
      "step": 81795
    },
    {
      "epoch": 8.998899889989,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.020133279065669148,
      "loss": 0.2298,
      "num_input_tokens_seen": 17263376,
      "step": 81800
    },
    {
      "epoch": 8.999449944994499,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.020131925950509513,
      "loss": 0.2303,
      "num_input_tokens_seen": 17264432,
      "step": 81805
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.02013057278805138,
      "loss": 0.2329,
      "num_input_tokens_seen": 17265344,
      "step": 81810
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.23148299753665924,
      "eval_runtime": 60.5719,
      "eval_samples_per_second": 66.698,
      "eval_steps_per_second": 16.674,
      "num_input_tokens_seen": 17265344,
      "step": 81810
    },
    {
      "epoch": 9.000550055005501,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02012921957830722,
      "loss": 0.2314,
      "num_input_tokens_seen": 17266336,
      "step": 81815
    },
    {
      "epoch": 9.001100110011,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.020127866321289503,
      "loss": 0.2319,
      "num_input_tokens_seen": 17267392,
      "step": 81820
    },
    {
      "epoch": 9.001650165016502,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020126513017010704,
      "loss": 0.2303,
      "num_input_tokens_seen": 17268416,
      "step": 81825
    },
    {
      "epoch": 9.002200220022003,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02012515966548329,
      "loss": 0.2303,
      "num_input_tokens_seen": 17269376,
      "step": 81830
    },
    {
      "epoch": 9.002750275027502,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.020123806266719745,
      "loss": 0.2314,
      "num_input_tokens_seen": 17270368,
      "step": 81835
    },
    {
      "epoch": 9.003300330033003,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.020122452820732533,
      "loss": 0.2319,
      "num_input_tokens_seen": 17271392,
      "step": 81840
    },
    {
      "epoch": 9.003850385038504,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.020121099327534128,
      "loss": 0.2313,
      "num_input_tokens_seen": 17272480,
      "step": 81845
    },
    {
      "epoch": 9.004400440044005,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.020119745787137013,
      "loss": 0.2314,
      "num_input_tokens_seen": 17273600,
      "step": 81850
    },
    {
      "epoch": 9.004950495049505,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.020118392199553657,
      "loss": 0.2309,
      "num_input_tokens_seen": 17274720,
      "step": 81855
    },
    {
      "epoch": 9.005500550055006,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.020117038564796536,
      "loss": 0.2309,
      "num_input_tokens_seen": 17275776,
      "step": 81860
    },
    {
      "epoch": 9.006050605060507,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.020115684882878127,
      "loss": 0.2314,
      "num_input_tokens_seen": 17276832,
      "step": 81865
    },
    {
      "epoch": 9.006600660066006,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.0201143311538109,
      "loss": 0.2314,
      "num_input_tokens_seen": 17277888,
      "step": 81870
    },
    {
      "epoch": 9.007150715071507,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02011297737760734,
      "loss": 0.2309,
      "num_input_tokens_seen": 17279040,
      "step": 81875
    },
    {
      "epoch": 9.007700770077008,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.02011162355427992,
      "loss": 0.2313,
      "num_input_tokens_seen": 17280096,
      "step": 81880
    },
    {
      "epoch": 9.008250825082508,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.020110269683841123,
      "loss": 0.2313,
      "num_input_tokens_seen": 17281184,
      "step": 81885
    },
    {
      "epoch": 9.008800880088009,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020108915766303418,
      "loss": 0.2319,
      "num_input_tokens_seen": 17282176,
      "step": 81890
    },
    {
      "epoch": 9.00935093509351,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02010756180167929,
      "loss": 0.2308,
      "num_input_tokens_seen": 17283232,
      "step": 81895
    },
    {
      "epoch": 9.009900990099009,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.020106207789981213,
      "loss": 0.2335,
      "num_input_tokens_seen": 17284288,
      "step": 81900
    },
    {
      "epoch": 9.01045104510451,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02010485373122167,
      "loss": 0.2313,
      "num_input_tokens_seen": 17285344,
      "step": 81905
    },
    {
      "epoch": 9.011001100110011,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.020103499625413142,
      "loss": 0.2335,
      "num_input_tokens_seen": 17286400,
      "step": 81910
    },
    {
      "epoch": 9.011551155115512,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.0201021454725681,
      "loss": 0.2298,
      "num_input_tokens_seen": 17287424,
      "step": 81915
    },
    {
      "epoch": 9.012101210121012,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.02010079127269904,
      "loss": 0.2329,
      "num_input_tokens_seen": 17288480,
      "step": 81920
    },
    {
      "epoch": 9.012651265126513,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.020099437025818432,
      "loss": 0.2303,
      "num_input_tokens_seen": 17289536,
      "step": 81925
    },
    {
      "epoch": 9.013201320132014,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.02009808273193876,
      "loss": 0.2335,
      "num_input_tokens_seen": 17290592,
      "step": 81930
    },
    {
      "epoch": 9.013751375137513,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020096728391072505,
      "loss": 0.2308,
      "num_input_tokens_seen": 17291680,
      "step": 81935
    },
    {
      "epoch": 9.014301430143014,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02009537400323215,
      "loss": 0.2345,
      "num_input_tokens_seen": 17292704,
      "step": 81940
    },
    {
      "epoch": 9.014851485148515,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.020094019568430176,
      "loss": 0.2298,
      "num_input_tokens_seen": 17293760,
      "step": 81945
    },
    {
      "epoch": 9.015401540154015,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.02009266508667907,
      "loss": 0.2319,
      "num_input_tokens_seen": 17294752,
      "step": 81950
    },
    {
      "epoch": 9.015951595159516,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.020091310557991314,
      "loss": 0.2319,
      "num_input_tokens_seen": 17295776,
      "step": 81955
    },
    {
      "epoch": 9.016501650165017,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.02008995598237939,
      "loss": 0.2314,
      "num_input_tokens_seen": 17296832,
      "step": 81960
    },
    {
      "epoch": 9.017051705170518,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.02008860135985579,
      "loss": 0.2309,
      "num_input_tokens_seen": 17297952,
      "step": 81965
    },
    {
      "epoch": 9.017601760176017,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.020087246690432988,
      "loss": 0.2319,
      "num_input_tokens_seen": 17299040,
      "step": 81970
    },
    {
      "epoch": 9.018151815181518,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.020085891974123467,
      "loss": 0.2303,
      "num_input_tokens_seen": 17300160,
      "step": 81975
    },
    {
      "epoch": 9.01870187018702,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02008453721093973,
      "loss": 0.2303,
      "num_input_tokens_seen": 17301216,
      "step": 81980
    },
    {
      "epoch": 9.019251925192519,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.020083182400894253,
      "loss": 0.2325,
      "num_input_tokens_seen": 17302272,
      "step": 81985
    },
    {
      "epoch": 9.01980198019802,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.02008182754399952,
      "loss": 0.2314,
      "num_input_tokens_seen": 17303328,
      "step": 81990
    },
    {
      "epoch": 9.020352035203521,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.02008047264026802,
      "loss": 0.2309,
      "num_input_tokens_seen": 17304416,
      "step": 81995
    },
    {
      "epoch": 9.02090209020902,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.020079117689712246,
      "loss": 0.2304,
      "num_input_tokens_seen": 17305440,
      "step": 82000
    },
    {
      "epoch": 9.021452145214521,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.02007776269234468,
      "loss": 0.2324,
      "num_input_tokens_seen": 17306528,
      "step": 82005
    },
    {
      "epoch": 9.022002200220022,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.02007640764817781,
      "loss": 0.2314,
      "num_input_tokens_seen": 17307616,
      "step": 82010
    },
    {
      "epoch": 9.022552255225522,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020075052557224127,
      "loss": 0.2335,
      "num_input_tokens_seen": 17308640,
      "step": 82015
    },
    {
      "epoch": 9.023102310231023,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.020073697419496116,
      "loss": 0.2304,
      "num_input_tokens_seen": 17309728,
      "step": 82020
    },
    {
      "epoch": 9.023652365236524,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.020072342235006282,
      "loss": 0.2325,
      "num_input_tokens_seen": 17310752,
      "step": 82025
    },
    {
      "epoch": 9.024202420242025,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.020070987003767093,
      "loss": 0.2299,
      "num_input_tokens_seen": 17311840,
      "step": 82030
    },
    {
      "epoch": 9.024752475247524,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.020069631725791053,
      "loss": 0.2304,
      "num_input_tokens_seen": 17312896,
      "step": 82035
    },
    {
      "epoch": 9.025302530253025,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.02006827640109065,
      "loss": 0.233,
      "num_input_tokens_seen": 17313920,
      "step": 82040
    },
    {
      "epoch": 9.025852585258527,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.020066921029678376,
      "loss": 0.2309,
      "num_input_tokens_seen": 17314976,
      "step": 82045
    },
    {
      "epoch": 9.026402640264026,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.02006556561156672,
      "loss": 0.2314,
      "num_input_tokens_seen": 17316064,
      "step": 82050
    },
    {
      "epoch": 9.026952695269527,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.02006421014676818,
      "loss": 0.2314,
      "num_input_tokens_seen": 17317152,
      "step": 82055
    },
    {
      "epoch": 9.027502750275028,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.020062854635295245,
      "loss": 0.2298,
      "num_input_tokens_seen": 17318208,
      "step": 82060
    },
    {
      "epoch": 9.028052805280527,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.020061499077160403,
      "loss": 0.2324,
      "num_input_tokens_seen": 17319296,
      "step": 82065
    },
    {
      "epoch": 9.028602860286028,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.020060143472376155,
      "loss": 0.2303,
      "num_input_tokens_seen": 17320320,
      "step": 82070
    },
    {
      "epoch": 9.02915291529153,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.020058787820954993,
      "loss": 0.2303,
      "num_input_tokens_seen": 17321408,
      "step": 82075
    },
    {
      "epoch": 9.029702970297029,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.020057432122909415,
      "loss": 0.2314,
      "num_input_tokens_seen": 17322432,
      "step": 82080
    },
    {
      "epoch": 9.03025302530253,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.02005607637825191,
      "loss": 0.2309,
      "num_input_tokens_seen": 17323424,
      "step": 82085
    },
    {
      "epoch": 9.030803080308031,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.020054720586994975,
      "loss": 0.233,
      "num_input_tokens_seen": 17324512,
      "step": 82090
    },
    {
      "epoch": 9.031353135313532,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.020053364749151103,
      "loss": 0.2309,
      "num_input_tokens_seen": 17325600,
      "step": 82095
    },
    {
      "epoch": 9.031903190319031,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.020052008864732795,
      "loss": 0.2346,
      "num_input_tokens_seen": 17326656,
      "step": 82100
    },
    {
      "epoch": 9.032453245324533,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.020050652933752546,
      "loss": 0.2298,
      "num_input_tokens_seen": 17327776,
      "step": 82105
    },
    {
      "epoch": 9.033003300330034,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.020049296956222852,
      "loss": 0.2319,
      "num_input_tokens_seen": 17328864,
      "step": 82110
    },
    {
      "epoch": 9.033553355335533,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.02004794093215621,
      "loss": 0.2314,
      "num_input_tokens_seen": 17329920,
      "step": 82115
    },
    {
      "epoch": 9.034103410341034,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.02004658486156512,
      "loss": 0.233,
      "num_input_tokens_seen": 17330912,
      "step": 82120
    },
    {
      "epoch": 9.034653465346535,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.02004522874446208,
      "loss": 0.2314,
      "num_input_tokens_seen": 17331968,
      "step": 82125
    },
    {
      "epoch": 9.035203520352034,
      "grad_norm": 0.004547119140625,
      "learning_rate": 0.020043872580859583,
      "loss": 0.2314,
      "num_input_tokens_seen": 17332960,
      "step": 82130
    },
    {
      "epoch": 9.035753575357536,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.020042516370770137,
      "loss": 0.2319,
      "num_input_tokens_seen": 17333984,
      "step": 82135
    },
    {
      "epoch": 9.036303630363037,
      "grad_norm": 0.000843048095703125,
      "learning_rate": 0.020041160114206237,
      "loss": 0.233,
      "num_input_tokens_seen": 17335104,
      "step": 82140
    },
    {
      "epoch": 9.036853685368538,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.020039803811180385,
      "loss": 0.233,
      "num_input_tokens_seen": 17336096,
      "step": 82145
    },
    {
      "epoch": 9.037403740374037,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.020038447461705074,
      "loss": 0.2293,
      "num_input_tokens_seen": 17337152,
      "step": 82150
    },
    {
      "epoch": 9.037953795379538,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.020037091065792816,
      "loss": 0.2346,
      "num_input_tokens_seen": 17338240,
      "step": 82155
    },
    {
      "epoch": 9.03850385038504,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.020035734623456102,
      "loss": 0.233,
      "num_input_tokens_seen": 17339264,
      "step": 82160
    },
    {
      "epoch": 9.039053905390539,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.020034378134707438,
      "loss": 0.2293,
      "num_input_tokens_seen": 17340320,
      "step": 82165
    },
    {
      "epoch": 9.03960396039604,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.02003302159955933,
      "loss": 0.2299,
      "num_input_tokens_seen": 17341376,
      "step": 82170
    },
    {
      "epoch": 9.04015401540154,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.020031665018024277,
      "loss": 0.2324,
      "num_input_tokens_seen": 17342496,
      "step": 82175
    },
    {
      "epoch": 9.04070407040704,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.020030308390114777,
      "loss": 0.2335,
      "num_input_tokens_seen": 17343584,
      "step": 82180
    },
    {
      "epoch": 9.041254125412541,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.020028951715843346,
      "loss": 0.2314,
      "num_input_tokens_seen": 17344672,
      "step": 82185
    },
    {
      "epoch": 9.041804180418042,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.020027594995222473,
      "loss": 0.2293,
      "num_input_tokens_seen": 17345792,
      "step": 82190
    },
    {
      "epoch": 9.042354235423542,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.020026238228264673,
      "loss": 0.2324,
      "num_input_tokens_seen": 17346784,
      "step": 82195
    },
    {
      "epoch": 9.042904290429043,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.020024881414982446,
      "loss": 0.233,
      "num_input_tokens_seen": 17347808,
      "step": 82200
    },
    {
      "epoch": 9.043454345434544,
      "grad_norm": 0.00054168701171875,
      "learning_rate": 0.020023524555388302,
      "loss": 0.2319,
      "num_input_tokens_seen": 17348896,
      "step": 82205
    },
    {
      "epoch": 9.044004400440045,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.020022167649494736,
      "loss": 0.233,
      "num_input_tokens_seen": 17349952,
      "step": 82210
    },
    {
      "epoch": 9.044554455445544,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.02002081069731427,
      "loss": 0.2309,
      "num_input_tokens_seen": 17351008,
      "step": 82215
    },
    {
      "epoch": 9.045104510451045,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.02001945369885939,
      "loss": 0.2319,
      "num_input_tokens_seen": 17352032,
      "step": 82220
    },
    {
      "epoch": 9.045654565456546,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.020018096654142618,
      "loss": 0.2304,
      "num_input_tokens_seen": 17353088,
      "step": 82225
    },
    {
      "epoch": 9.046204620462046,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.02001673956317646,
      "loss": 0.2304,
      "num_input_tokens_seen": 17354112,
      "step": 82230
    },
    {
      "epoch": 9.046754675467547,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.020015382425973418,
      "loss": 0.2324,
      "num_input_tokens_seen": 17355232,
      "step": 82235
    },
    {
      "epoch": 9.047304730473048,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.020014025242546005,
      "loss": 0.2325,
      "num_input_tokens_seen": 17356320,
      "step": 82240
    },
    {
      "epoch": 9.047854785478547,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.02001266801290673,
      "loss": 0.2324,
      "num_input_tokens_seen": 17357344,
      "step": 82245
    },
    {
      "epoch": 9.048404840484048,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.02001131073706809,
      "loss": 0.2293,
      "num_input_tokens_seen": 17358400,
      "step": 82250
    },
    {
      "epoch": 9.04895489548955,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.02000995341504261,
      "loss": 0.2309,
      "num_input_tokens_seen": 17359456,
      "step": 82255
    },
    {
      "epoch": 9.049504950495049,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.02000859604684279,
      "loss": 0.2298,
      "num_input_tokens_seen": 17360576,
      "step": 82260
    },
    {
      "epoch": 9.05005500550055,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.020007238632481147,
      "loss": 0.2325,
      "num_input_tokens_seen": 17361632,
      "step": 82265
    },
    {
      "epoch": 9.05060506050605,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.020005881171970188,
      "loss": 0.2325,
      "num_input_tokens_seen": 17362720,
      "step": 82270
    },
    {
      "epoch": 9.051155115511552,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.020004523665322422,
      "loss": 0.2335,
      "num_input_tokens_seen": 17363744,
      "step": 82275
    },
    {
      "epoch": 9.051705170517051,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.020003166112550364,
      "loss": 0.2309,
      "num_input_tokens_seen": 17364800,
      "step": 82280
    },
    {
      "epoch": 9.052255225522552,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.020001808513666525,
      "loss": 0.2335,
      "num_input_tokens_seen": 17365856,
      "step": 82285
    },
    {
      "epoch": 9.052805280528053,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.02000045086868342,
      "loss": 0.2325,
      "num_input_tokens_seen": 17366848,
      "step": 82290
    },
    {
      "epoch": 9.053355335533553,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.019999093177613555,
      "loss": 0.233,
      "num_input_tokens_seen": 17367904,
      "step": 82295
    },
    {
      "epoch": 9.053905390539054,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.019997735440469445,
      "loss": 0.2324,
      "num_input_tokens_seen": 17368992,
      "step": 82300
    },
    {
      "epoch": 9.054455445544555,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019996377657263607,
      "loss": 0.2293,
      "num_input_tokens_seen": 17370080,
      "step": 82305
    },
    {
      "epoch": 9.055005500550054,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.019995019828008557,
      "loss": 0.2303,
      "num_input_tokens_seen": 17371104,
      "step": 82310
    },
    {
      "epoch": 9.055555555555555,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.019993661952716804,
      "loss": 0.234,
      "num_input_tokens_seen": 17372192,
      "step": 82315
    },
    {
      "epoch": 9.056105610561056,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.019992304031400863,
      "loss": 0.2308,
      "num_input_tokens_seen": 17373248,
      "step": 82320
    },
    {
      "epoch": 9.056655665566556,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019990946064073253,
      "loss": 0.2319,
      "num_input_tokens_seen": 17374336,
      "step": 82325
    },
    {
      "epoch": 9.057205720572057,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019989588050746486,
      "loss": 0.2319,
      "num_input_tokens_seen": 17375360,
      "step": 82330
    },
    {
      "epoch": 9.057755775577558,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.019988229991433088,
      "loss": 0.2319,
      "num_input_tokens_seen": 17376416,
      "step": 82335
    },
    {
      "epoch": 9.058305830583059,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.01998687188614556,
      "loss": 0.2298,
      "num_input_tokens_seen": 17377440,
      "step": 82340
    },
    {
      "epoch": 9.058855885588558,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01998551373489643,
      "loss": 0.234,
      "num_input_tokens_seen": 17378432,
      "step": 82345
    },
    {
      "epoch": 9.05940594059406,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01998415553769821,
      "loss": 0.2319,
      "num_input_tokens_seen": 17379488,
      "step": 82350
    },
    {
      "epoch": 9.05995599559956,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01998279729456342,
      "loss": 0.2309,
      "num_input_tokens_seen": 17380576,
      "step": 82355
    },
    {
      "epoch": 9.06050605060506,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01998143900550458,
      "loss": 0.2335,
      "num_input_tokens_seen": 17381664,
      "step": 82360
    },
    {
      "epoch": 9.061056105610561,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.019980080670534205,
      "loss": 0.2324,
      "num_input_tokens_seen": 17382784,
      "step": 82365
    },
    {
      "epoch": 9.061606160616062,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.019978722289664817,
      "loss": 0.2303,
      "num_input_tokens_seen": 17383776,
      "step": 82370
    },
    {
      "epoch": 9.062156215621561,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.019977363862908933,
      "loss": 0.2309,
      "num_input_tokens_seen": 17384832,
      "step": 82375
    },
    {
      "epoch": 9.062706270627062,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.01997600539027908,
      "loss": 0.2308,
      "num_input_tokens_seen": 17385920,
      "step": 82380
    },
    {
      "epoch": 9.063256325632564,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01997464687178777,
      "loss": 0.2319,
      "num_input_tokens_seen": 17386976,
      "step": 82385
    },
    {
      "epoch": 9.063806380638065,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.019973288307447528,
      "loss": 0.2319,
      "num_input_tokens_seen": 17388064,
      "step": 82390
    },
    {
      "epoch": 9.064356435643564,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.019971929697270872,
      "loss": 0.2319,
      "num_input_tokens_seen": 17389056,
      "step": 82395
    },
    {
      "epoch": 9.064906490649065,
      "grad_norm": 0.009765625,
      "learning_rate": 0.01997057104127032,
      "loss": 0.2329,
      "num_input_tokens_seen": 17390112,
      "step": 82400
    },
    {
      "epoch": 9.065456545654566,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.019969212339458416,
      "loss": 0.2324,
      "num_input_tokens_seen": 17391168,
      "step": 82405
    },
    {
      "epoch": 9.066006600660065,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.01996785359184765,
      "loss": 0.2319,
      "num_input_tokens_seen": 17392256,
      "step": 82410
    },
    {
      "epoch": 9.066556655665567,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.01996649479845057,
      "loss": 0.2299,
      "num_input_tokens_seen": 17393312,
      "step": 82415
    },
    {
      "epoch": 9.067106710671068,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.019965135959279685,
      "loss": 0.2303,
      "num_input_tokens_seen": 17394400,
      "step": 82420
    },
    {
      "epoch": 9.067656765676567,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01996377707434753,
      "loss": 0.2309,
      "num_input_tokens_seen": 17395488,
      "step": 82425
    },
    {
      "epoch": 9.068206820682068,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019962418143666623,
      "loss": 0.2314,
      "num_input_tokens_seen": 17396544,
      "step": 82430
    },
    {
      "epoch": 9.06875687568757,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.01996105916724949,
      "loss": 0.2324,
      "num_input_tokens_seen": 17397664,
      "step": 82435
    },
    {
      "epoch": 9.069306930693068,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.019959700145108648,
      "loss": 0.2293,
      "num_input_tokens_seen": 17398752,
      "step": 82440
    },
    {
      "epoch": 9.06985698569857,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.019958341077256633,
      "loss": 0.2278,
      "num_input_tokens_seen": 17399808,
      "step": 82445
    },
    {
      "epoch": 9.07040704070407,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01995698196370597,
      "loss": 0.2309,
      "num_input_tokens_seen": 17400832,
      "step": 82450
    },
    {
      "epoch": 9.070957095709572,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.019955622804469185,
      "loss": 0.2314,
      "num_input_tokens_seen": 17401888,
      "step": 82455
    },
    {
      "epoch": 9.071507150715071,
      "grad_norm": 0.00457763671875,
      "learning_rate": 0.0199542635995588,
      "loss": 0.2314,
      "num_input_tokens_seen": 17402944,
      "step": 82460
    },
    {
      "epoch": 9.072057205720572,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.019952904348987343,
      "loss": 0.2335,
      "num_input_tokens_seen": 17404032,
      "step": 82465
    },
    {
      "epoch": 9.072607260726073,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01995154505276734,
      "loss": 0.2319,
      "num_input_tokens_seen": 17405088,
      "step": 82470
    },
    {
      "epoch": 9.073157315731573,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01995018571091133,
      "loss": 0.2293,
      "num_input_tokens_seen": 17406080,
      "step": 82475
    },
    {
      "epoch": 9.073707370737074,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.019948826323431833,
      "loss": 0.2345,
      "num_input_tokens_seen": 17407168,
      "step": 82480
    },
    {
      "epoch": 9.074257425742575,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019947466890341374,
      "loss": 0.2309,
      "num_input_tokens_seen": 17408256,
      "step": 82485
    },
    {
      "epoch": 9.074807480748074,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.019946107411652488,
      "loss": 0.2309,
      "num_input_tokens_seen": 17409376,
      "step": 82490
    },
    {
      "epoch": 9.075357535753575,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.0199447478873777,
      "loss": 0.2314,
      "num_input_tokens_seen": 17410432,
      "step": 82495
    },
    {
      "epoch": 9.075907590759076,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019943388317529547,
      "loss": 0.233,
      "num_input_tokens_seen": 17411520,
      "step": 82500
    },
    {
      "epoch": 9.076457645764576,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.01994202870212056,
      "loss": 0.2314,
      "num_input_tokens_seen": 17412544,
      "step": 82505
    },
    {
      "epoch": 9.077007700770077,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01994066904116326,
      "loss": 0.2324,
      "num_input_tokens_seen": 17413600,
      "step": 82510
    },
    {
      "epoch": 9.077557755775578,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019939309334670186,
      "loss": 0.2314,
      "num_input_tokens_seen": 17414656,
      "step": 82515
    },
    {
      "epoch": 9.078107810781079,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.019937949582653866,
      "loss": 0.2309,
      "num_input_tokens_seen": 17415648,
      "step": 82520
    },
    {
      "epoch": 9.078657865786578,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01993658978512684,
      "loss": 0.2304,
      "num_input_tokens_seen": 17416736,
      "step": 82525
    },
    {
      "epoch": 9.07920792079208,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01993522994210163,
      "loss": 0.2288,
      "num_input_tokens_seen": 17417760,
      "step": 82530
    },
    {
      "epoch": 9.07975797579758,
      "grad_norm": 0.000720977783203125,
      "learning_rate": 0.019933870053590777,
      "loss": 0.2314,
      "num_input_tokens_seen": 17418848,
      "step": 82535
    },
    {
      "epoch": 9.08030803080308,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.019932510119606808,
      "loss": 0.233,
      "num_input_tokens_seen": 17419936,
      "step": 82540
    },
    {
      "epoch": 9.08085808580858,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01993115014016226,
      "loss": 0.2314,
      "num_input_tokens_seen": 17420992,
      "step": 82545
    },
    {
      "epoch": 9.081408140814082,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01992979011526967,
      "loss": 0.2324,
      "num_input_tokens_seen": 17422048,
      "step": 82550
    },
    {
      "epoch": 9.081958195819581,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01992843004494157,
      "loss": 0.2335,
      "num_input_tokens_seen": 17423136,
      "step": 82555
    },
    {
      "epoch": 9.082508250825082,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.01992706992919049,
      "loss": 0.2319,
      "num_input_tokens_seen": 17424192,
      "step": 82560
    },
    {
      "epoch": 9.083058305830583,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.019925709768028974,
      "loss": 0.2314,
      "num_input_tokens_seen": 17425184,
      "step": 82565
    },
    {
      "epoch": 9.083608360836084,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019924349561469556,
      "loss": 0.2319,
      "num_input_tokens_seen": 17426272,
      "step": 82570
    },
    {
      "epoch": 9.084158415841584,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019922989309524772,
      "loss": 0.2314,
      "num_input_tokens_seen": 17427360,
      "step": 82575
    },
    {
      "epoch": 9.084708470847085,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019921629012207155,
      "loss": 0.2304,
      "num_input_tokens_seen": 17428384,
      "step": 82580
    },
    {
      "epoch": 9.085258525852586,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019920268669529242,
      "loss": 0.2324,
      "num_input_tokens_seen": 17429472,
      "step": 82585
    },
    {
      "epoch": 9.085808580858085,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01991890828150358,
      "loss": 0.2309,
      "num_input_tokens_seen": 17430560,
      "step": 82590
    },
    {
      "epoch": 9.086358635863586,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0199175478481427,
      "loss": 0.2319,
      "num_input_tokens_seen": 17431648,
      "step": 82595
    },
    {
      "epoch": 9.086908690869087,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.01991618736945914,
      "loss": 0.2319,
      "num_input_tokens_seen": 17432736,
      "step": 82600
    },
    {
      "epoch": 9.087458745874587,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.019914826845465437,
      "loss": 0.2309,
      "num_input_tokens_seen": 17433760,
      "step": 82605
    },
    {
      "epoch": 9.088008800880088,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.019913466276174137,
      "loss": 0.2319,
      "num_input_tokens_seen": 17434752,
      "step": 82610
    },
    {
      "epoch": 9.088558855885589,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.019912105661597772,
      "loss": 0.2319,
      "num_input_tokens_seen": 17435776,
      "step": 82615
    },
    {
      "epoch": 9.089108910891088,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.019910745001748893,
      "loss": 0.2298,
      "num_input_tokens_seen": 17436864,
      "step": 82620
    },
    {
      "epoch": 9.08965896589659,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019909384296640027,
      "loss": 0.2314,
      "num_input_tokens_seen": 17438016,
      "step": 82625
    },
    {
      "epoch": 9.09020902090209,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019908023546283724,
      "loss": 0.2319,
      "num_input_tokens_seen": 17438976,
      "step": 82630
    },
    {
      "epoch": 9.090759075907592,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.019906662750692524,
      "loss": 0.2319,
      "num_input_tokens_seen": 17439968,
      "step": 82635
    },
    {
      "epoch": 9.091309130913091,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.019905301909878968,
      "loss": 0.2309,
      "num_input_tokens_seen": 17440960,
      "step": 82640
    },
    {
      "epoch": 9.091859185918592,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019903941023855596,
      "loss": 0.2308,
      "num_input_tokens_seen": 17442080,
      "step": 82645
    },
    {
      "epoch": 9.092409240924093,
      "grad_norm": 0.000843048095703125,
      "learning_rate": 0.019902580092634955,
      "loss": 0.2288,
      "num_input_tokens_seen": 17443136,
      "step": 82650
    },
    {
      "epoch": 9.092959295929592,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019901219116229586,
      "loss": 0.2314,
      "num_input_tokens_seen": 17444128,
      "step": 82655
    },
    {
      "epoch": 9.093509350935093,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019899858094652025,
      "loss": 0.2309,
      "num_input_tokens_seen": 17445184,
      "step": 82660
    },
    {
      "epoch": 9.094059405940595,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.01989849702791483,
      "loss": 0.2303,
      "num_input_tokens_seen": 17446304,
      "step": 82665
    },
    {
      "epoch": 9.094609460946094,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.019897135916030543,
      "loss": 0.2299,
      "num_input_tokens_seen": 17447296,
      "step": 82670
    },
    {
      "epoch": 9.095159515951595,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.0198957747590117,
      "loss": 0.2293,
      "num_input_tokens_seen": 17448352,
      "step": 82675
    },
    {
      "epoch": 9.095709570957096,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.019894413556870848,
      "loss": 0.2304,
      "num_input_tokens_seen": 17449408,
      "step": 82680
    },
    {
      "epoch": 9.096259625962595,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.019893052309620535,
      "loss": 0.2319,
      "num_input_tokens_seen": 17450496,
      "step": 82685
    },
    {
      "epoch": 9.096809680968097,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.019891691017273307,
      "loss": 0.2335,
      "num_input_tokens_seen": 17451488,
      "step": 82690
    },
    {
      "epoch": 9.097359735973598,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019890329679841712,
      "loss": 0.2309,
      "num_input_tokens_seen": 17452448,
      "step": 82695
    },
    {
      "epoch": 9.097909790979099,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019888968297338293,
      "loss": 0.2314,
      "num_input_tokens_seen": 17453472,
      "step": 82700
    },
    {
      "epoch": 9.098459845984598,
      "grad_norm": 0.00079345703125,
      "learning_rate": 0.0198876068697756,
      "loss": 0.2325,
      "num_input_tokens_seen": 17454464,
      "step": 82705
    },
    {
      "epoch": 9.099009900990099,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01988624539716618,
      "loss": 0.2324,
      "num_input_tokens_seen": 17455552,
      "step": 82710
    },
    {
      "epoch": 9.0995599559956,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019884883879522582,
      "loss": 0.2324,
      "num_input_tokens_seen": 17456576,
      "step": 82715
    },
    {
      "epoch": 9.1001100110011,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.01988352231685735,
      "loss": 0.2298,
      "num_input_tokens_seen": 17457632,
      "step": 82720
    },
    {
      "epoch": 9.1006600660066,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.019882160709183038,
      "loss": 0.2314,
      "num_input_tokens_seen": 17458624,
      "step": 82725
    },
    {
      "epoch": 9.101210121012102,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.019880799056512192,
      "loss": 0.2314,
      "num_input_tokens_seen": 17459712,
      "step": 82730
    },
    {
      "epoch": 9.101760176017601,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019879437358857364,
      "loss": 0.2304,
      "num_input_tokens_seen": 17460768,
      "step": 82735
    },
    {
      "epoch": 9.102310231023102,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.01987807561623111,
      "loss": 0.2308,
      "num_input_tokens_seen": 17461792,
      "step": 82740
    },
    {
      "epoch": 9.102860286028603,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.01987671382864596,
      "loss": 0.2283,
      "num_input_tokens_seen": 17462848,
      "step": 82745
    },
    {
      "epoch": 9.103410341034103,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.01987535199611449,
      "loss": 0.2319,
      "num_input_tokens_seen": 17463872,
      "step": 82750
    },
    {
      "epoch": 9.103960396039604,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.019873990118649237,
      "loss": 0.2319,
      "num_input_tokens_seen": 17464992,
      "step": 82755
    },
    {
      "epoch": 9.104510451045105,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.019872628196262752,
      "loss": 0.2314,
      "num_input_tokens_seen": 17466016,
      "step": 82760
    },
    {
      "epoch": 9.105060506050606,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.019871266228967597,
      "loss": 0.2309,
      "num_input_tokens_seen": 17467104,
      "step": 82765
    },
    {
      "epoch": 9.105610561056105,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.019869904216776316,
      "loss": 0.2314,
      "num_input_tokens_seen": 17468192,
      "step": 82770
    },
    {
      "epoch": 9.106160616061606,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.019868542159701463,
      "loss": 0.2314,
      "num_input_tokens_seen": 17469184,
      "step": 82775
    },
    {
      "epoch": 9.106710671067107,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.019867180057755594,
      "loss": 0.2299,
      "num_input_tokens_seen": 17470272,
      "step": 82780
    },
    {
      "epoch": 9.107260726072607,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.019865817910951263,
      "loss": 0.2324,
      "num_input_tokens_seen": 17471392,
      "step": 82785
    },
    {
      "epoch": 9.107810781078108,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01986445571930102,
      "loss": 0.2313,
      "num_input_tokens_seen": 17472448,
      "step": 82790
    },
    {
      "epoch": 9.108360836083609,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01986309348281743,
      "loss": 0.233,
      "num_input_tokens_seen": 17473440,
      "step": 82795
    },
    {
      "epoch": 9.108910891089108,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019861731201513036,
      "loss": 0.233,
      "num_input_tokens_seen": 17474528,
      "step": 82800
    },
    {
      "epoch": 9.10946094609461,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.019860368875400394,
      "loss": 0.2309,
      "num_input_tokens_seen": 17475616,
      "step": 82805
    },
    {
      "epoch": 9.11001100110011,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019859006504492075,
      "loss": 0.2319,
      "num_input_tokens_seen": 17476704,
      "step": 82810
    },
    {
      "epoch": 9.110561056105611,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.019857644088800617,
      "loss": 0.2308,
      "num_input_tokens_seen": 17477760,
      "step": 82815
    },
    {
      "epoch": 9.11111111111111,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.019856281628338582,
      "loss": 0.2319,
      "num_input_tokens_seen": 17478752,
      "step": 82820
    },
    {
      "epoch": 9.111661166116612,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.019854919123118535,
      "loss": 0.2335,
      "num_input_tokens_seen": 17479776,
      "step": 82825
    },
    {
      "epoch": 9.112211221122113,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.01985355657315303,
      "loss": 0.2329,
      "num_input_tokens_seen": 17480832,
      "step": 82830
    },
    {
      "epoch": 9.112761276127612,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.019852193978454617,
      "loss": 0.2308,
      "num_input_tokens_seen": 17481952,
      "step": 82835
    },
    {
      "epoch": 9.113311331133113,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.019850831339035865,
      "loss": 0.2314,
      "num_input_tokens_seen": 17482976,
      "step": 82840
    },
    {
      "epoch": 9.113861386138614,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.019849468654909327,
      "loss": 0.2314,
      "num_input_tokens_seen": 17484032,
      "step": 82845
    },
    {
      "epoch": 9.114411441144114,
      "grad_norm": 0.009765625,
      "learning_rate": 0.01984810592608756,
      "loss": 0.2329,
      "num_input_tokens_seen": 17485056,
      "step": 82850
    },
    {
      "epoch": 9.114961496149615,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.01984674315258313,
      "loss": 0.2314,
      "num_input_tokens_seen": 17486048,
      "step": 82855
    },
    {
      "epoch": 9.115511551155116,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.019845380334408594,
      "loss": 0.2314,
      "num_input_tokens_seen": 17487168,
      "step": 82860
    },
    {
      "epoch": 9.116061606160615,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.019844017471576512,
      "loss": 0.2309,
      "num_input_tokens_seen": 17488224,
      "step": 82865
    },
    {
      "epoch": 9.116611661166116,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.01984265456409945,
      "loss": 0.2304,
      "num_input_tokens_seen": 17489312,
      "step": 82870
    },
    {
      "epoch": 9.117161716171617,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.01984129161198996,
      "loss": 0.2303,
      "num_input_tokens_seen": 17490400,
      "step": 82875
    },
    {
      "epoch": 9.117711771177119,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.019839928615260603,
      "loss": 0.2298,
      "num_input_tokens_seen": 17491456,
      "step": 82880
    },
    {
      "epoch": 9.118261826182618,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019838565573923956,
      "loss": 0.2319,
      "num_input_tokens_seen": 17492480,
      "step": 82885
    },
    {
      "epoch": 9.118811881188119,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019837202487992565,
      "loss": 0.2335,
      "num_input_tokens_seen": 17493536,
      "step": 82890
    },
    {
      "epoch": 9.11936193619362,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019835839357479002,
      "loss": 0.2314,
      "num_input_tokens_seen": 17494656,
      "step": 82895
    },
    {
      "epoch": 9.11991199119912,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.01983447618239583,
      "loss": 0.2298,
      "num_input_tokens_seen": 17495744,
      "step": 82900
    },
    {
      "epoch": 9.12046204620462,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.019833112962755608,
      "loss": 0.2314,
      "num_input_tokens_seen": 17496768,
      "step": 82905
    },
    {
      "epoch": 9.121012101210122,
      "grad_norm": 0.009765625,
      "learning_rate": 0.019831749698570905,
      "loss": 0.2309,
      "num_input_tokens_seen": 17497856,
      "step": 82910
    },
    {
      "epoch": 9.12156215621562,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.019830386389854286,
      "loss": 0.2314,
      "num_input_tokens_seen": 17498880,
      "step": 82915
    },
    {
      "epoch": 9.122112211221122,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.019829023036618307,
      "loss": 0.2293,
      "num_input_tokens_seen": 17500000,
      "step": 82920
    },
    {
      "epoch": 9.122662266226623,
      "grad_norm": 0.004638671875,
      "learning_rate": 0.019827659638875544,
      "loss": 0.2298,
      "num_input_tokens_seen": 17501024,
      "step": 82925
    },
    {
      "epoch": 9.123212321232122,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.01982629619663856,
      "loss": 0.2319,
      "num_input_tokens_seen": 17502080,
      "step": 82930
    },
    {
      "epoch": 9.123762376237623,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.019824932709919914,
      "loss": 0.2298,
      "num_input_tokens_seen": 17503168,
      "step": 82935
    },
    {
      "epoch": 9.124312431243125,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.019823569178732185,
      "loss": 0.2324,
      "num_input_tokens_seen": 17504256,
      "step": 82940
    },
    {
      "epoch": 9.124862486248626,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.01982220560308793,
      "loss": 0.2314,
      "num_input_tokens_seen": 17505312,
      "step": 82945
    },
    {
      "epoch": 9.125412541254125,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.019820841982999715,
      "loss": 0.233,
      "num_input_tokens_seen": 17506400,
      "step": 82950
    },
    {
      "epoch": 9.125962596259626,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019819478318480126,
      "loss": 0.2303,
      "num_input_tokens_seen": 17507392,
      "step": 82955
    },
    {
      "epoch": 9.126512651265127,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.019818114609541714,
      "loss": 0.2319,
      "num_input_tokens_seen": 17508448,
      "step": 82960
    },
    {
      "epoch": 9.127062706270626,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01981675085619704,
      "loss": 0.2309,
      "num_input_tokens_seen": 17509568,
      "step": 82965
    },
    {
      "epoch": 9.127612761276128,
      "grad_norm": 0.00074005126953125,
      "learning_rate": 0.019815387058458693,
      "loss": 0.2298,
      "num_input_tokens_seen": 17510624,
      "step": 82970
    },
    {
      "epoch": 9.128162816281629,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019814023216339235,
      "loss": 0.2324,
      "num_input_tokens_seen": 17511712,
      "step": 82975
    },
    {
      "epoch": 9.128712871287128,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019812659329851236,
      "loss": 0.2309,
      "num_input_tokens_seen": 17512832,
      "step": 82980
    },
    {
      "epoch": 9.129262926292629,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01981129539900726,
      "loss": 0.2298,
      "num_input_tokens_seen": 17513920,
      "step": 82985
    },
    {
      "epoch": 9.12981298129813,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01980993142381989,
      "loss": 0.2314,
      "num_input_tokens_seen": 17514944,
      "step": 82990
    },
    {
      "epoch": 9.130363036303631,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.019808567404301678,
      "loss": 0.2314,
      "num_input_tokens_seen": 17515968,
      "step": 82995
    },
    {
      "epoch": 9.13091309130913,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.01980720334046522,
      "loss": 0.233,
      "num_input_tokens_seen": 17517056,
      "step": 83000
    },
    {
      "epoch": 9.131463146314632,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.019805839232323073,
      "loss": 0.2319,
      "num_input_tokens_seen": 17518080,
      "step": 83005
    },
    {
      "epoch": 9.132013201320133,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.019804475079887807,
      "loss": 0.2314,
      "num_input_tokens_seen": 17519168,
      "step": 83010
    },
    {
      "epoch": 9.132563256325632,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019803110883172003,
      "loss": 0.2319,
      "num_input_tokens_seen": 17520288,
      "step": 83015
    },
    {
      "epoch": 9.133113311331133,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.01980174664218823,
      "loss": 0.2319,
      "num_input_tokens_seen": 17521408,
      "step": 83020
    },
    {
      "epoch": 9.133663366336634,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.019800382356949066,
      "loss": 0.2303,
      "num_input_tokens_seen": 17522496,
      "step": 83025
    },
    {
      "epoch": 9.134213421342134,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.019799018027467075,
      "loss": 0.2303,
      "num_input_tokens_seen": 17523616,
      "step": 83030
    },
    {
      "epoch": 9.134763476347635,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.01979765365375484,
      "loss": 0.2319,
      "num_input_tokens_seen": 17524640,
      "step": 83035
    },
    {
      "epoch": 9.135313531353136,
      "grad_norm": 0.00058746337890625,
      "learning_rate": 0.019796289235824934,
      "loss": 0.2319,
      "num_input_tokens_seen": 17525728,
      "step": 83040
    },
    {
      "epoch": 9.135863586358635,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019794924773689934,
      "loss": 0.2314,
      "num_input_tokens_seen": 17526784,
      "step": 83045
    },
    {
      "epoch": 9.136413641364136,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019793560267362408,
      "loss": 0.2293,
      "num_input_tokens_seen": 17527840,
      "step": 83050
    },
    {
      "epoch": 9.136963696369637,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019792195716854942,
      "loss": 0.2303,
      "num_input_tokens_seen": 17528832,
      "step": 83055
    },
    {
      "epoch": 9.137513751375138,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.019790831122180105,
      "loss": 0.2324,
      "num_input_tokens_seen": 17529920,
      "step": 83060
    },
    {
      "epoch": 9.138063806380638,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.019789466483350476,
      "loss": 0.2303,
      "num_input_tokens_seen": 17530976,
      "step": 83065
    },
    {
      "epoch": 9.138613861386139,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.01978810180037863,
      "loss": 0.2298,
      "num_input_tokens_seen": 17532000,
      "step": 83070
    },
    {
      "epoch": 9.13916391639164,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.019786737073277157,
      "loss": 0.2314,
      "num_input_tokens_seen": 17533152,
      "step": 83075
    },
    {
      "epoch": 9.13971397139714,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.019785372302058617,
      "loss": 0.2293,
      "num_input_tokens_seen": 17534208,
      "step": 83080
    },
    {
      "epoch": 9.14026402640264,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0197840074867356,
      "loss": 0.2324,
      "num_input_tokens_seen": 17535232,
      "step": 83085
    },
    {
      "epoch": 9.140814081408141,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01978264262732068,
      "loss": 0.2309,
      "num_input_tokens_seen": 17536320,
      "step": 83090
    },
    {
      "epoch": 9.14136413641364,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.01978127772382644,
      "loss": 0.2288,
      "num_input_tokens_seen": 17537408,
      "step": 83095
    },
    {
      "epoch": 9.141914191419142,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019779912776265455,
      "loss": 0.233,
      "num_input_tokens_seen": 17538496,
      "step": 83100
    },
    {
      "epoch": 9.142464246424643,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.01977854778465031,
      "loss": 0.2278,
      "num_input_tokens_seen": 17539520,
      "step": 83105
    },
    {
      "epoch": 9.143014301430142,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.019777182748993576,
      "loss": 0.2325,
      "num_input_tokens_seen": 17540640,
      "step": 83110
    },
    {
      "epoch": 9.143564356435643,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.01977581766930785,
      "loss": 0.2315,
      "num_input_tokens_seen": 17541664,
      "step": 83115
    },
    {
      "epoch": 9.144114411441144,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0197744525456057,
      "loss": 0.2314,
      "num_input_tokens_seen": 17542720,
      "step": 83120
    },
    {
      "epoch": 9.144664466446645,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.019773087377899713,
      "loss": 0.2299,
      "num_input_tokens_seen": 17543872,
      "step": 83125
    },
    {
      "epoch": 9.145214521452145,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.019771722166202467,
      "loss": 0.2346,
      "num_input_tokens_seen": 17544960,
      "step": 83130
    },
    {
      "epoch": 9.145764576457646,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.019770356910526553,
      "loss": 0.2314,
      "num_input_tokens_seen": 17546016,
      "step": 83135
    },
    {
      "epoch": 9.146314631463147,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.01976899161088454,
      "loss": 0.232,
      "num_input_tokens_seen": 17547104,
      "step": 83140
    },
    {
      "epoch": 9.146864686468646,
      "grad_norm": 0.00457763671875,
      "learning_rate": 0.019767626267289028,
      "loss": 0.2294,
      "num_input_tokens_seen": 17548128,
      "step": 83145
    },
    {
      "epoch": 9.147414741474147,
      "grad_norm": 0.00933837890625,
      "learning_rate": 0.01976626087975259,
      "loss": 0.2304,
      "num_input_tokens_seen": 17549216,
      "step": 83150
    },
    {
      "epoch": 9.147964796479648,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.019764895448287814,
      "loss": 0.2341,
      "num_input_tokens_seen": 17550272,
      "step": 83155
    },
    {
      "epoch": 9.148514851485148,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.01976352997290728,
      "loss": 0.2325,
      "num_input_tokens_seen": 17551328,
      "step": 83160
    },
    {
      "epoch": 9.149064906490649,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01976216445362358,
      "loss": 0.2314,
      "num_input_tokens_seen": 17552384,
      "step": 83165
    },
    {
      "epoch": 9.14961496149615,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.01976079889044929,
      "loss": 0.2273,
      "num_input_tokens_seen": 17553472,
      "step": 83170
    },
    {
      "epoch": 9.150165016501651,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.019759433283397004,
      "loss": 0.2361,
      "num_input_tokens_seen": 17554528,
      "step": 83175
    },
    {
      "epoch": 9.15071507150715,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.019758067632479306,
      "loss": 0.2299,
      "num_input_tokens_seen": 17555584,
      "step": 83180
    },
    {
      "epoch": 9.151265126512651,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.019756701937708777,
      "loss": 0.2304,
      "num_input_tokens_seen": 17556576,
      "step": 83185
    },
    {
      "epoch": 9.151815181518153,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019755336199098018,
      "loss": 0.234,
      "num_input_tokens_seen": 17557600,
      "step": 83190
    },
    {
      "epoch": 9.152365236523652,
      "grad_norm": 0.00457763671875,
      "learning_rate": 0.019753970416659595,
      "loss": 0.233,
      "num_input_tokens_seen": 17558656,
      "step": 83195
    },
    {
      "epoch": 9.152915291529153,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.019752604590406116,
      "loss": 0.2293,
      "num_input_tokens_seen": 17559712,
      "step": 83200
    },
    {
      "epoch": 9.153465346534654,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.019751238720350155,
      "loss": 0.2309,
      "num_input_tokens_seen": 17560736,
      "step": 83205
    },
    {
      "epoch": 9.154015401540153,
      "grad_norm": 0.000698089599609375,
      "learning_rate": 0.019749872806504313,
      "loss": 0.2298,
      "num_input_tokens_seen": 17561792,
      "step": 83210
    },
    {
      "epoch": 9.154565456545654,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.019748506848881173,
      "loss": 0.233,
      "num_input_tokens_seen": 17562752,
      "step": 83215
    },
    {
      "epoch": 9.155115511551156,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.019747140847493322,
      "loss": 0.2351,
      "num_input_tokens_seen": 17563840,
      "step": 83220
    },
    {
      "epoch": 9.155665566556655,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.019745774802353344,
      "loss": 0.2319,
      "num_input_tokens_seen": 17564832,
      "step": 83225
    },
    {
      "epoch": 9.156215621562156,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.019744408713473847,
      "loss": 0.2309,
      "num_input_tokens_seen": 17565888,
      "step": 83230
    },
    {
      "epoch": 9.156765676567657,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.01974304258086741,
      "loss": 0.2314,
      "num_input_tokens_seen": 17566944,
      "step": 83235
    },
    {
      "epoch": 9.157315731573158,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.019741676404546625,
      "loss": 0.2304,
      "num_input_tokens_seen": 17568032,
      "step": 83240
    },
    {
      "epoch": 9.157865786578657,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.019740310184524084,
      "loss": 0.2319,
      "num_input_tokens_seen": 17568992,
      "step": 83245
    },
    {
      "epoch": 9.158415841584159,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.019738943920812373,
      "loss": 0.2329,
      "num_input_tokens_seen": 17569984,
      "step": 83250
    },
    {
      "epoch": 9.15896589658966,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.019737577613424095,
      "loss": 0.2308,
      "num_input_tokens_seen": 17571104,
      "step": 83255
    },
    {
      "epoch": 9.159515951595159,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.019736211262371833,
      "loss": 0.2319,
      "num_input_tokens_seen": 17572128,
      "step": 83260
    },
    {
      "epoch": 9.16006600660066,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.019734844867668192,
      "loss": 0.2298,
      "num_input_tokens_seen": 17573184,
      "step": 83265
    },
    {
      "epoch": 9.160616061606161,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.019733478429325754,
      "loss": 0.2319,
      "num_input_tokens_seen": 17574240,
      "step": 83270
    },
    {
      "epoch": 9.16116611661166,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019732111947357116,
      "loss": 0.233,
      "num_input_tokens_seen": 17575328,
      "step": 83275
    },
    {
      "epoch": 9.161716171617162,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.019730745421774873,
      "loss": 0.2329,
      "num_input_tokens_seen": 17576320,
      "step": 83280
    },
    {
      "epoch": 9.162266226622663,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01972937885259162,
      "loss": 0.2335,
      "num_input_tokens_seen": 17577312,
      "step": 83285
    },
    {
      "epoch": 9.162816281628162,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01972801223981995,
      "loss": 0.2309,
      "num_input_tokens_seen": 17578400,
      "step": 83290
    },
    {
      "epoch": 9.163366336633663,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.019726645583472466,
      "loss": 0.2319,
      "num_input_tokens_seen": 17579392,
      "step": 83295
    },
    {
      "epoch": 9.163916391639164,
      "grad_norm": 0.009765625,
      "learning_rate": 0.01972527888356175,
      "loss": 0.2329,
      "num_input_tokens_seen": 17580480,
      "step": 83300
    },
    {
      "epoch": 9.164466446644665,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.01972391214010041,
      "loss": 0.2319,
      "num_input_tokens_seen": 17581600,
      "step": 83305
    },
    {
      "epoch": 9.165016501650165,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019722545353101036,
      "loss": 0.2293,
      "num_input_tokens_seen": 17582688,
      "step": 83310
    },
    {
      "epoch": 9.165566556655666,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.019721178522576228,
      "loss": 0.2335,
      "num_input_tokens_seen": 17583744,
      "step": 83315
    },
    {
      "epoch": 9.166116611661167,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.019719811648538584,
      "loss": 0.2293,
      "num_input_tokens_seen": 17584832,
      "step": 83320
    },
    {
      "epoch": 9.166666666666666,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.019718444731000703,
      "loss": 0.233,
      "num_input_tokens_seen": 17585888,
      "step": 83325
    },
    {
      "epoch": 9.167216721672167,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019717077769975177,
      "loss": 0.2303,
      "num_input_tokens_seen": 17586880,
      "step": 83330
    },
    {
      "epoch": 9.167766776677668,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.01971571076547461,
      "loss": 0.2314,
      "num_input_tokens_seen": 17587968,
      "step": 83335
    },
    {
      "epoch": 9.168316831683168,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.019714343717511598,
      "loss": 0.2319,
      "num_input_tokens_seen": 17589120,
      "step": 83340
    },
    {
      "epoch": 9.168866886688669,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.019712976626098746,
      "loss": 0.2319,
      "num_input_tokens_seen": 17590208,
      "step": 83345
    },
    {
      "epoch": 9.16941694169417,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01971160949124865,
      "loss": 0.2319,
      "num_input_tokens_seen": 17591296,
      "step": 83350
    },
    {
      "epoch": 9.16996699669967,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01971024231297391,
      "loss": 0.2293,
      "num_input_tokens_seen": 17592352,
      "step": 83355
    },
    {
      "epoch": 9.17051705170517,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.019708875091287124,
      "loss": 0.2319,
      "num_input_tokens_seen": 17593376,
      "step": 83360
    },
    {
      "epoch": 9.171067106710671,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.0197075078262009,
      "loss": 0.2314,
      "num_input_tokens_seen": 17594496,
      "step": 83365
    },
    {
      "epoch": 9.171617161716172,
      "grad_norm": 0.0028533935546875,
      "learning_rate": 0.019706140517727826,
      "loss": 0.2314,
      "num_input_tokens_seen": 17595520,
      "step": 83370
    },
    {
      "epoch": 9.172167216721672,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.019704773165880523,
      "loss": 0.2309,
      "num_input_tokens_seen": 17596640,
      "step": 83375
    },
    {
      "epoch": 9.172717271727173,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.019703405770671586,
      "loss": 0.2303,
      "num_input_tokens_seen": 17597696,
      "step": 83380
    },
    {
      "epoch": 9.173267326732674,
      "grad_norm": 0.000560760498046875,
      "learning_rate": 0.019702038332113602,
      "loss": 0.2314,
      "num_input_tokens_seen": 17598784,
      "step": 83385
    },
    {
      "epoch": 9.173817381738173,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.019700670850219194,
      "loss": 0.2314,
      "num_input_tokens_seen": 17599808,
      "step": 83390
    },
    {
      "epoch": 9.174367436743674,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019699303325000958,
      "loss": 0.2314,
      "num_input_tokens_seen": 17600896,
      "step": 83395
    },
    {
      "epoch": 9.174917491749175,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.019697935756471496,
      "loss": 0.2303,
      "num_input_tokens_seen": 17601920,
      "step": 83400
    },
    {
      "epoch": 9.175467546754675,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01969656814464342,
      "loss": 0.2303,
      "num_input_tokens_seen": 17602944,
      "step": 83405
    },
    {
      "epoch": 9.176017601760176,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.019695200489529326,
      "loss": 0.2345,
      "num_input_tokens_seen": 17604032,
      "step": 83410
    },
    {
      "epoch": 9.176567656765677,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.01969383279114182,
      "loss": 0.2308,
      "num_input_tokens_seen": 17605120,
      "step": 83415
    },
    {
      "epoch": 9.177117711771178,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01969246504949351,
      "loss": 0.2298,
      "num_input_tokens_seen": 17606176,
      "step": 83420
    },
    {
      "epoch": 9.177667766776677,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.019691097264597005,
      "loss": 0.2319,
      "num_input_tokens_seen": 17607232,
      "step": 83425
    },
    {
      "epoch": 9.178217821782178,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.019689729436464905,
      "loss": 0.2319,
      "num_input_tokens_seen": 17608352,
      "step": 83430
    },
    {
      "epoch": 9.17876787678768,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019688361565109818,
      "loss": 0.2314,
      "num_input_tokens_seen": 17609312,
      "step": 83435
    },
    {
      "epoch": 9.179317931793179,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.019686993650544354,
      "loss": 0.2298,
      "num_input_tokens_seen": 17610368,
      "step": 83440
    },
    {
      "epoch": 9.17986798679868,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019685625692781115,
      "loss": 0.2309,
      "num_input_tokens_seen": 17611456,
      "step": 83445
    },
    {
      "epoch": 9.180418041804181,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019684257691832716,
      "loss": 0.2298,
      "num_input_tokens_seen": 17612512,
      "step": 83450
    },
    {
      "epoch": 9.18096809680968,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019682889647711765,
      "loss": 0.2314,
      "num_input_tokens_seen": 17613504,
      "step": 83455
    },
    {
      "epoch": 9.181518151815181,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019681521560430862,
      "loss": 0.2288,
      "num_input_tokens_seen": 17614560,
      "step": 83460
    },
    {
      "epoch": 9.182068206820682,
      "grad_norm": 0.0008392333984375,
      "learning_rate": 0.01968015343000262,
      "loss": 0.2298,
      "num_input_tokens_seen": 17615616,
      "step": 83465
    },
    {
      "epoch": 9.182618261826182,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01967878525643965,
      "loss": 0.2293,
      "num_input_tokens_seen": 17616704,
      "step": 83470
    },
    {
      "epoch": 9.183168316831683,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.019677417039754567,
      "loss": 0.2314,
      "num_input_tokens_seen": 17617824,
      "step": 83475
    },
    {
      "epoch": 9.183718371837184,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.019676048779959973,
      "loss": 0.2319,
      "num_input_tokens_seen": 17618880,
      "step": 83480
    },
    {
      "epoch": 9.184268426842685,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01967468047706848,
      "loss": 0.2319,
      "num_input_tokens_seen": 17619872,
      "step": 83485
    },
    {
      "epoch": 9.184818481848184,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.019673312131092698,
      "loss": 0.2351,
      "num_input_tokens_seen": 17620960,
      "step": 83490
    },
    {
      "epoch": 9.185368536853685,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.019671943742045242,
      "loss": 0.2319,
      "num_input_tokens_seen": 17621952,
      "step": 83495
    },
    {
      "epoch": 9.185918591859187,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019670575309938723,
      "loss": 0.2329,
      "num_input_tokens_seen": 17623072,
      "step": 83500
    },
    {
      "epoch": 9.186468646864686,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019669206834785753,
      "loss": 0.2293,
      "num_input_tokens_seen": 17624096,
      "step": 83505
    },
    {
      "epoch": 9.187018701870187,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019667838316598945,
      "loss": 0.2319,
      "num_input_tokens_seen": 17625088,
      "step": 83510
    },
    {
      "epoch": 9.187568756875688,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.019666469755390905,
      "loss": 0.2314,
      "num_input_tokens_seen": 17626112,
      "step": 83515
    },
    {
      "epoch": 9.188118811881187,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.019665101151174258,
      "loss": 0.2335,
      "num_input_tokens_seen": 17627168,
      "step": 83520
    },
    {
      "epoch": 9.188668866886688,
      "grad_norm": 0.009765625,
      "learning_rate": 0.019663732503961615,
      "loss": 0.2308,
      "num_input_tokens_seen": 17628160,
      "step": 83525
    },
    {
      "epoch": 9.18921892189219,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.01966236381376558,
      "loss": 0.2314,
      "num_input_tokens_seen": 17629184,
      "step": 83530
    },
    {
      "epoch": 9.189768976897689,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.019660995080598778,
      "loss": 0.2314,
      "num_input_tokens_seen": 17630240,
      "step": 83535
    },
    {
      "epoch": 9.19031903190319,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.019659626304473824,
      "loss": 0.2314,
      "num_input_tokens_seen": 17631328,
      "step": 83540
    },
    {
      "epoch": 9.190869086908691,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019658257485403324,
      "loss": 0.2324,
      "num_input_tokens_seen": 17632384,
      "step": 83545
    },
    {
      "epoch": 9.191419141914192,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.01965688862339991,
      "loss": 0.2324,
      "num_input_tokens_seen": 17633408,
      "step": 83550
    },
    {
      "epoch": 9.191969196919691,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01965551971847618,
      "loss": 0.2325,
      "num_input_tokens_seen": 17634496,
      "step": 83555
    },
    {
      "epoch": 9.192519251925193,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019654150770644762,
      "loss": 0.2314,
      "num_input_tokens_seen": 17635616,
      "step": 83560
    },
    {
      "epoch": 9.193069306930694,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.01965278177991827,
      "loss": 0.2298,
      "num_input_tokens_seen": 17636608,
      "step": 83565
    },
    {
      "epoch": 9.193619361936193,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01965141274630932,
      "loss": 0.2314,
      "num_input_tokens_seen": 17637696,
      "step": 83570
    },
    {
      "epoch": 9.194169416941694,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019650043669830527,
      "loss": 0.2314,
      "num_input_tokens_seen": 17638720,
      "step": 83575
    },
    {
      "epoch": 9.194719471947195,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.019648674550494516,
      "loss": 0.2324,
      "num_input_tokens_seen": 17639840,
      "step": 83580
    },
    {
      "epoch": 9.195269526952695,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.019647305388313906,
      "loss": 0.2309,
      "num_input_tokens_seen": 17640928,
      "step": 83585
    },
    {
      "epoch": 9.195819581958196,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.0196459361833013,
      "loss": 0.2288,
      "num_input_tokens_seen": 17642016,
      "step": 83590
    },
    {
      "epoch": 9.196369636963697,
      "grad_norm": 0.009765625,
      "learning_rate": 0.019644566935469343,
      "loss": 0.2309,
      "num_input_tokens_seen": 17643072,
      "step": 83595
    },
    {
      "epoch": 9.196919691969198,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019643197644830634,
      "loss": 0.2324,
      "num_input_tokens_seen": 17644160,
      "step": 83600
    },
    {
      "epoch": 9.197469746974697,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.019641828311397807,
      "loss": 0.2293,
      "num_input_tokens_seen": 17645152,
      "step": 83605
    },
    {
      "epoch": 9.198019801980198,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.01964045893518347,
      "loss": 0.2314,
      "num_input_tokens_seen": 17646176,
      "step": 83610
    },
    {
      "epoch": 9.1985698569857,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01963908951620025,
      "loss": 0.2319,
      "num_input_tokens_seen": 17647264,
      "step": 83615
    },
    {
      "epoch": 9.199119911991199,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.019637720054460767,
      "loss": 0.233,
      "num_input_tokens_seen": 17648384,
      "step": 83620
    },
    {
      "epoch": 9.1996699669967,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.019636350549977647,
      "loss": 0.2309,
      "num_input_tokens_seen": 17649472,
      "step": 83625
    },
    {
      "epoch": 9.2002200220022,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.01963498100276351,
      "loss": 0.2303,
      "num_input_tokens_seen": 17650560,
      "step": 83630
    },
    {
      "epoch": 9.2007700770077,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.019633611412830973,
      "loss": 0.2293,
      "num_input_tokens_seen": 17651616,
      "step": 83635
    },
    {
      "epoch": 9.201320132013201,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.019632241780192662,
      "loss": 0.2324,
      "num_input_tokens_seen": 17652672,
      "step": 83640
    },
    {
      "epoch": 9.201870187018702,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019630872104861207,
      "loss": 0.2298,
      "num_input_tokens_seen": 17653696,
      "step": 83645
    },
    {
      "epoch": 9.202420242024202,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.01962950238684922,
      "loss": 0.2324,
      "num_input_tokens_seen": 17654688,
      "step": 83650
    },
    {
      "epoch": 9.202970297029703,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.019628132626169335,
      "loss": 0.233,
      "num_input_tokens_seen": 17655744,
      "step": 83655
    },
    {
      "epoch": 9.203520352035204,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.01962676282283417,
      "loss": 0.2314,
      "num_input_tokens_seen": 17656832,
      "step": 83660
    },
    {
      "epoch": 9.204070407040705,
      "grad_norm": 0.009765625,
      "learning_rate": 0.019625392976856355,
      "loss": 0.2314,
      "num_input_tokens_seen": 17657856,
      "step": 83665
    },
    {
      "epoch": 9.204620462046204,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019624023088248514,
      "loss": 0.2324,
      "num_input_tokens_seen": 17658880,
      "step": 83670
    },
    {
      "epoch": 9.205170517051705,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019622653157023263,
      "loss": 0.2293,
      "num_input_tokens_seen": 17659936,
      "step": 83675
    },
    {
      "epoch": 9.205720572057206,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01962128318319324,
      "loss": 0.2314,
      "num_input_tokens_seen": 17660992,
      "step": 83680
    },
    {
      "epoch": 9.206270627062706,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.01961991316677107,
      "loss": 0.2335,
      "num_input_tokens_seen": 17662080,
      "step": 83685
    },
    {
      "epoch": 9.206820682068207,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.019618543107769375,
      "loss": 0.2329,
      "num_input_tokens_seen": 17663168,
      "step": 83690
    },
    {
      "epoch": 9.207370737073708,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.019617173006200785,
      "loss": 0.2314,
      "num_input_tokens_seen": 17664256,
      "step": 83695
    },
    {
      "epoch": 9.207920792079207,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.019615802862077928,
      "loss": 0.2308,
      "num_input_tokens_seen": 17665280,
      "step": 83700
    },
    {
      "epoch": 9.208470847084708,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019614432675413428,
      "loss": 0.2324,
      "num_input_tokens_seen": 17666304,
      "step": 83705
    },
    {
      "epoch": 9.20902090209021,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.019613062446219916,
      "loss": 0.2298,
      "num_input_tokens_seen": 17667424,
      "step": 83710
    },
    {
      "epoch": 9.209570957095709,
      "grad_norm": 0.000701904296875,
      "learning_rate": 0.019611692174510028,
      "loss": 0.2314,
      "num_input_tokens_seen": 17668448,
      "step": 83715
    },
    {
      "epoch": 9.21012101210121,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019610321860296376,
      "loss": 0.2314,
      "num_input_tokens_seen": 17669504,
      "step": 83720
    },
    {
      "epoch": 9.210671067106711,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019608951503591605,
      "loss": 0.2308,
      "num_input_tokens_seen": 17670496,
      "step": 83725
    },
    {
      "epoch": 9.211221122112212,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019607581104408342,
      "loss": 0.2314,
      "num_input_tokens_seen": 17671520,
      "step": 83730
    },
    {
      "epoch": 9.211771177117711,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.019606210662759208,
      "loss": 0.2314,
      "num_input_tokens_seen": 17672512,
      "step": 83735
    },
    {
      "epoch": 9.212321232123212,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.019604840178656848,
      "loss": 0.2324,
      "num_input_tokens_seen": 17673504,
      "step": 83740
    },
    {
      "epoch": 9.212871287128714,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.019603469652113886,
      "loss": 0.2314,
      "num_input_tokens_seen": 17674528,
      "step": 83745
    },
    {
      "epoch": 9.213421342134213,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.019602099083142942,
      "loss": 0.2319,
      "num_input_tokens_seen": 17675552,
      "step": 83750
    },
    {
      "epoch": 9.213971397139714,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.01960072847175667,
      "loss": 0.2303,
      "num_input_tokens_seen": 17676640,
      "step": 83755
    },
    {
      "epoch": 9.214521452145215,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019599357817967688,
      "loss": 0.2324,
      "num_input_tokens_seen": 17677728,
      "step": 83760
    },
    {
      "epoch": 9.215071507150714,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.019597987121788632,
      "loss": 0.2314,
      "num_input_tokens_seen": 17678720,
      "step": 83765
    },
    {
      "epoch": 9.215621562156215,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.019596616383232134,
      "loss": 0.2335,
      "num_input_tokens_seen": 17679776,
      "step": 83770
    },
    {
      "epoch": 9.216171617161717,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.01959524560231083,
      "loss": 0.2314,
      "num_input_tokens_seen": 17680832,
      "step": 83775
    },
    {
      "epoch": 9.216721672167218,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.01959387477903735,
      "loss": 0.2303,
      "num_input_tokens_seen": 17681888,
      "step": 83780
    },
    {
      "epoch": 9.217271727172717,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.019592503913424333,
      "loss": 0.2319,
      "num_input_tokens_seen": 17683008,
      "step": 83785
    },
    {
      "epoch": 9.217821782178218,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.019591133005484408,
      "loss": 0.2304,
      "num_input_tokens_seen": 17684064,
      "step": 83790
    },
    {
      "epoch": 9.218371837183719,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.01958976205523022,
      "loss": 0.2303,
      "num_input_tokens_seen": 17685056,
      "step": 83795
    },
    {
      "epoch": 9.218921892189218,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.01958839106267439,
      "loss": 0.2303,
      "num_input_tokens_seen": 17686144,
      "step": 83800
    },
    {
      "epoch": 9.21947194719472,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.019587020027829565,
      "loss": 0.2298,
      "num_input_tokens_seen": 17687168,
      "step": 83805
    },
    {
      "epoch": 9.22002200220022,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019585648950708377,
      "loss": 0.2303,
      "num_input_tokens_seen": 17688160,
      "step": 83810
    },
    {
      "epoch": 9.22057205720572,
      "grad_norm": 0.000766754150390625,
      "learning_rate": 0.019584277831323466,
      "loss": 0.2309,
      "num_input_tokens_seen": 17689248,
      "step": 83815
    },
    {
      "epoch": 9.221122112211221,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.01958290666968746,
      "loss": 0.2308,
      "num_input_tokens_seen": 17690272,
      "step": 83820
    },
    {
      "epoch": 9.221672167216722,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.019581535465813003,
      "loss": 0.2303,
      "num_input_tokens_seen": 17691328,
      "step": 83825
    },
    {
      "epoch": 9.222222222222221,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.019580164219712735,
      "loss": 0.2298,
      "num_input_tokens_seen": 17692416,
      "step": 83830
    },
    {
      "epoch": 9.222772277227723,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019578792931399292,
      "loss": 0.2303,
      "num_input_tokens_seen": 17693472,
      "step": 83835
    },
    {
      "epoch": 9.223322332233224,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.01957742160088531,
      "loss": 0.2319,
      "num_input_tokens_seen": 17694432,
      "step": 83840
    },
    {
      "epoch": 9.223872387238725,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.019576050228183428,
      "loss": 0.2314,
      "num_input_tokens_seen": 17695424,
      "step": 83845
    },
    {
      "epoch": 9.224422442244224,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.01957467881330629,
      "loss": 0.233,
      "num_input_tokens_seen": 17696512,
      "step": 83850
    },
    {
      "epoch": 9.224972497249725,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019573307356266532,
      "loss": 0.2314,
      "num_input_tokens_seen": 17697536,
      "step": 83855
    },
    {
      "epoch": 9.225522552255226,
      "grad_norm": 0.009765625,
      "learning_rate": 0.019571935857076794,
      "loss": 0.2314,
      "num_input_tokens_seen": 17698592,
      "step": 83860
    },
    {
      "epoch": 9.226072607260726,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.019570564315749713,
      "loss": 0.2308,
      "num_input_tokens_seen": 17699680,
      "step": 83865
    },
    {
      "epoch": 9.226622662266227,
      "grad_norm": 0.000659942626953125,
      "learning_rate": 0.019569192732297937,
      "loss": 0.234,
      "num_input_tokens_seen": 17700704,
      "step": 83870
    },
    {
      "epoch": 9.227172717271728,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.019567821106734103,
      "loss": 0.2314,
      "num_input_tokens_seen": 17701728,
      "step": 83875
    },
    {
      "epoch": 9.227722772277227,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01956644943907086,
      "loss": 0.2319,
      "num_input_tokens_seen": 17702784,
      "step": 83880
    },
    {
      "epoch": 9.228272827282728,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019565077729320837,
      "loss": 0.2313,
      "num_input_tokens_seen": 17703840,
      "step": 83885
    },
    {
      "epoch": 9.22882288228823,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.019563705977496685,
      "loss": 0.2319,
      "num_input_tokens_seen": 17704928,
      "step": 83890
    },
    {
      "epoch": 9.229372937293729,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.019562334183611044,
      "loss": 0.2298,
      "num_input_tokens_seen": 17705952,
      "step": 83895
    },
    {
      "epoch": 9.22992299229923,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.01956096234767656,
      "loss": 0.2293,
      "num_input_tokens_seen": 17707104,
      "step": 83900
    },
    {
      "epoch": 9.23047304730473,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019559590469705873,
      "loss": 0.2309,
      "num_input_tokens_seen": 17708160,
      "step": 83905
    },
    {
      "epoch": 9.231023102310232,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.01955821854971163,
      "loss": 0.2314,
      "num_input_tokens_seen": 17709184,
      "step": 83910
    },
    {
      "epoch": 9.231573157315731,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.019556846587706473,
      "loss": 0.2313,
      "num_input_tokens_seen": 17710208,
      "step": 83915
    },
    {
      "epoch": 9.232123212321232,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.019555474583703047,
      "loss": 0.2298,
      "num_input_tokens_seen": 17711296,
      "step": 83920
    },
    {
      "epoch": 9.232673267326733,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019554102537714,
      "loss": 0.2325,
      "num_input_tokens_seen": 17712416,
      "step": 83925
    },
    {
      "epoch": 9.233223322332233,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.019552730449751973,
      "loss": 0.2319,
      "num_input_tokens_seen": 17713440,
      "step": 83930
    },
    {
      "epoch": 9.233773377337734,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019551358319829616,
      "loss": 0.2309,
      "num_input_tokens_seen": 17714528,
      "step": 83935
    },
    {
      "epoch": 9.234323432343235,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.019549986147959567,
      "loss": 0.2298,
      "num_input_tokens_seen": 17715616,
      "step": 83940
    },
    {
      "epoch": 9.234873487348734,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019548613934154486,
      "loss": 0.2309,
      "num_input_tokens_seen": 17716704,
      "step": 83945
    },
    {
      "epoch": 9.235423542354235,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.01954724167842701,
      "loss": 0.2319,
      "num_input_tokens_seen": 17717728,
      "step": 83950
    },
    {
      "epoch": 9.235973597359736,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01954586938078979,
      "loss": 0.2314,
      "num_input_tokens_seen": 17718784,
      "step": 83955
    },
    {
      "epoch": 9.236523652365236,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.019544497041255476,
      "loss": 0.2324,
      "num_input_tokens_seen": 17719840,
      "step": 83960
    },
    {
      "epoch": 9.237073707370737,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.019543124659836708,
      "loss": 0.2309,
      "num_input_tokens_seen": 17720896,
      "step": 83965
    },
    {
      "epoch": 9.237623762376238,
      "grad_norm": 0.000457763671875,
      "learning_rate": 0.019541752236546143,
      "loss": 0.2314,
      "num_input_tokens_seen": 17721952,
      "step": 83970
    },
    {
      "epoch": 9.238173817381739,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.01954037977139643,
      "loss": 0.2319,
      "num_input_tokens_seen": 17723040,
      "step": 83975
    },
    {
      "epoch": 9.238723872387238,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.019539007264400207,
      "loss": 0.2319,
      "num_input_tokens_seen": 17724096,
      "step": 83980
    },
    {
      "epoch": 9.23927392739274,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01953763471557014,
      "loss": 0.2325,
      "num_input_tokens_seen": 17725184,
      "step": 83985
    },
    {
      "epoch": 9.23982398239824,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.01953626212491887,
      "loss": 0.2308,
      "num_input_tokens_seen": 17726272,
      "step": 83990
    },
    {
      "epoch": 9.24037403740374,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019534889492459045,
      "loss": 0.2304,
      "num_input_tokens_seen": 17727328,
      "step": 83995
    },
    {
      "epoch": 9.24092409240924,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.019533516818203322,
      "loss": 0.2304,
      "num_input_tokens_seen": 17728352,
      "step": 84000
    },
    {
      "epoch": 9.241474147414742,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01953214410216435,
      "loss": 0.2314,
      "num_input_tokens_seen": 17729408,
      "step": 84005
    },
    {
      "epoch": 9.242024202420241,
      "grad_norm": 0.000782012939453125,
      "learning_rate": 0.01953077134435478,
      "loss": 0.2324,
      "num_input_tokens_seen": 17730400,
      "step": 84010
    },
    {
      "epoch": 9.242574257425742,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.019529398544787267,
      "loss": 0.2298,
      "num_input_tokens_seen": 17731488,
      "step": 84015
    },
    {
      "epoch": 9.243124312431243,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.019528025703474457,
      "loss": 0.2309,
      "num_input_tokens_seen": 17732480,
      "step": 84020
    },
    {
      "epoch": 9.243674367436745,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.019526652820429008,
      "loss": 0.2309,
      "num_input_tokens_seen": 17733504,
      "step": 84025
    },
    {
      "epoch": 9.244224422442244,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.019525279895663574,
      "loss": 0.2319,
      "num_input_tokens_seen": 17734560,
      "step": 84030
    },
    {
      "epoch": 9.244774477447745,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.019523906929190805,
      "loss": 0.2293,
      "num_input_tokens_seen": 17735648,
      "step": 84035
    },
    {
      "epoch": 9.245324532453246,
      "grad_norm": 0.000545501708984375,
      "learning_rate": 0.019522533921023353,
      "loss": 0.2298,
      "num_input_tokens_seen": 17736672,
      "step": 84040
    },
    {
      "epoch": 9.245874587458745,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.019521160871173882,
      "loss": 0.2319,
      "num_input_tokens_seen": 17737792,
      "step": 84045
    },
    {
      "epoch": 9.246424642464246,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.019519787779655042,
      "loss": 0.2314,
      "num_input_tokens_seen": 17738848,
      "step": 84050
    },
    {
      "epoch": 9.246974697469748,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.01951841464647948,
      "loss": 0.2335,
      "num_input_tokens_seen": 17739904,
      "step": 84055
    },
    {
      "epoch": 9.247524752475247,
      "grad_norm": 0.00063323974609375,
      "learning_rate": 0.01951704147165986,
      "loss": 0.2319,
      "num_input_tokens_seen": 17740896,
      "step": 84060
    },
    {
      "epoch": 9.248074807480748,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.01951566825520884,
      "loss": 0.2298,
      "num_input_tokens_seen": 17741920,
      "step": 84065
    },
    {
      "epoch": 9.248624862486249,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01951429499713907,
      "loss": 0.2314,
      "num_input_tokens_seen": 17742944,
      "step": 84070
    },
    {
      "epoch": 9.249174917491748,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019512921697463215,
      "loss": 0.234,
      "num_input_tokens_seen": 17744000,
      "step": 84075
    },
    {
      "epoch": 9.24972497249725,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01951154835619392,
      "loss": 0.2319,
      "num_input_tokens_seen": 17745088,
      "step": 84080
    },
    {
      "epoch": 9.25027502750275,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019510174973343845,
      "loss": 0.2319,
      "num_input_tokens_seen": 17746112,
      "step": 84085
    },
    {
      "epoch": 9.250825082508252,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.019508801548925658,
      "loss": 0.2314,
      "num_input_tokens_seen": 17747168,
      "step": 84090
    },
    {
      "epoch": 9.251375137513751,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01950742808295201,
      "loss": 0.2309,
      "num_input_tokens_seen": 17748256,
      "step": 84095
    },
    {
      "epoch": 9.251925192519252,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.01950605457543556,
      "loss": 0.2324,
      "num_input_tokens_seen": 17749376,
      "step": 84100
    },
    {
      "epoch": 9.252475247524753,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019504681026388967,
      "loss": 0.2293,
      "num_input_tokens_seen": 17750528,
      "step": 84105
    },
    {
      "epoch": 9.253025302530252,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.019503307435824888,
      "loss": 0.2298,
      "num_input_tokens_seen": 17751552,
      "step": 84110
    },
    {
      "epoch": 9.253575357535754,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.019501933803755984,
      "loss": 0.2298,
      "num_input_tokens_seen": 17752576,
      "step": 84115
    },
    {
      "epoch": 9.254125412541255,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.019500560130194924,
      "loss": 0.2308,
      "num_input_tokens_seen": 17753632,
      "step": 84120
    },
    {
      "epoch": 9.254675467546754,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019499186415154354,
      "loss": 0.2319,
      "num_input_tokens_seen": 17754688,
      "step": 84125
    },
    {
      "epoch": 9.255225522552255,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.019497812658646942,
      "loss": 0.2314,
      "num_input_tokens_seen": 17755744,
      "step": 84130
    },
    {
      "epoch": 9.255775577557756,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.019496438860685354,
      "loss": 0.2303,
      "num_input_tokens_seen": 17756832,
      "step": 84135
    },
    {
      "epoch": 9.256325632563255,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019495065021282242,
      "loss": 0.2314,
      "num_input_tokens_seen": 17757792,
      "step": 84140
    },
    {
      "epoch": 9.256875687568757,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.019493691140450273,
      "loss": 0.2324,
      "num_input_tokens_seen": 17758944,
      "step": 84145
    },
    {
      "epoch": 9.257425742574258,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019492317218202115,
      "loss": 0.2309,
      "num_input_tokens_seen": 17760000,
      "step": 84150
    },
    {
      "epoch": 9.257975797579759,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.019490943254550416,
      "loss": 0.2308,
      "num_input_tokens_seen": 17761056,
      "step": 84155
    },
    {
      "epoch": 9.258525852585258,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019489569249507854,
      "loss": 0.2314,
      "num_input_tokens_seen": 17762048,
      "step": 84160
    },
    {
      "epoch": 9.25907590759076,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019488195203087085,
      "loss": 0.2324,
      "num_input_tokens_seen": 17763072,
      "step": 84165
    },
    {
      "epoch": 9.25962596259626,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01948682111530077,
      "loss": 0.2319,
      "num_input_tokens_seen": 17764160,
      "step": 84170
    },
    {
      "epoch": 9.26017601760176,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.019485446986161583,
      "loss": 0.2283,
      "num_input_tokens_seen": 17765216,
      "step": 84175
    },
    {
      "epoch": 9.26072607260726,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.019484072815682183,
      "loss": 0.2319,
      "num_input_tokens_seen": 17766304,
      "step": 84180
    },
    {
      "epoch": 9.261276127612762,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019482698603875234,
      "loss": 0.2308,
      "num_input_tokens_seen": 17767360,
      "step": 84185
    },
    {
      "epoch": 9.261826182618261,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.019481324350753407,
      "loss": 0.233,
      "num_input_tokens_seen": 17768448,
      "step": 84190
    },
    {
      "epoch": 9.262376237623762,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01947995005632936,
      "loss": 0.2335,
      "num_input_tokens_seen": 17769504,
      "step": 84195
    },
    {
      "epoch": 9.262926292629263,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.01947857572061576,
      "loss": 0.2303,
      "num_input_tokens_seen": 17770560,
      "step": 84200
    },
    {
      "epoch": 9.263476347634764,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.01947720134362528,
      "loss": 0.233,
      "num_input_tokens_seen": 17771680,
      "step": 84205
    },
    {
      "epoch": 9.264026402640264,
      "grad_norm": 0.009765625,
      "learning_rate": 0.019475826925370584,
      "loss": 0.234,
      "num_input_tokens_seen": 17772768,
      "step": 84210
    },
    {
      "epoch": 9.264576457645765,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.019474452465864343,
      "loss": 0.233,
      "num_input_tokens_seen": 17773824,
      "step": 84215
    },
    {
      "epoch": 9.265126512651266,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019473077965119212,
      "loss": 0.2308,
      "num_input_tokens_seen": 17774880,
      "step": 84220
    },
    {
      "epoch": 9.265676567656765,
      "grad_norm": 0.009765625,
      "learning_rate": 0.019471703423147878,
      "loss": 0.2319,
      "num_input_tokens_seen": 17776000,
      "step": 84225
    },
    {
      "epoch": 9.266226622662266,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.01947032883996299,
      "loss": 0.2309,
      "num_input_tokens_seen": 17777024,
      "step": 84230
    },
    {
      "epoch": 9.266776677667767,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019468954215577226,
      "loss": 0.2293,
      "num_input_tokens_seen": 17778048,
      "step": 84235
    },
    {
      "epoch": 9.267326732673267,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.019467579550003264,
      "loss": 0.233,
      "num_input_tokens_seen": 17779008,
      "step": 84240
    },
    {
      "epoch": 9.267876787678768,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01946620484325376,
      "loss": 0.232,
      "num_input_tokens_seen": 17780032,
      "step": 84245
    },
    {
      "epoch": 9.268426842684269,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.019464830095341387,
      "loss": 0.2298,
      "num_input_tokens_seen": 17781088,
      "step": 84250
    },
    {
      "epoch": 9.268976897689768,
      "grad_norm": 0.00946044921875,
      "learning_rate": 0.01946345530627882,
      "loss": 0.2293,
      "num_input_tokens_seen": 17782112,
      "step": 84255
    },
    {
      "epoch": 9.26952695269527,
      "grad_norm": 0.009765625,
      "learning_rate": 0.019462080476078728,
      "loss": 0.2314,
      "num_input_tokens_seen": 17783136,
      "step": 84260
    },
    {
      "epoch": 9.27007700770077,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01946070560475378,
      "loss": 0.2325,
      "num_input_tokens_seen": 17784192,
      "step": 84265
    },
    {
      "epoch": 9.270627062706271,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01945933069231665,
      "loss": 0.2314,
      "num_input_tokens_seen": 17785248,
      "step": 84270
    },
    {
      "epoch": 9.27117711771177,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.019457955738780005,
      "loss": 0.2314,
      "num_input_tokens_seen": 17786304,
      "step": 84275
    },
    {
      "epoch": 9.271727172717272,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.019456580744156526,
      "loss": 0.2324,
      "num_input_tokens_seen": 17787392,
      "step": 84280
    },
    {
      "epoch": 9.272277227722773,
      "grad_norm": 0.009765625,
      "learning_rate": 0.019455205708458878,
      "loss": 0.2314,
      "num_input_tokens_seen": 17788416,
      "step": 84285
    },
    {
      "epoch": 9.272827282728272,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.019453830631699736,
      "loss": 0.2324,
      "num_input_tokens_seen": 17789568,
      "step": 84290
    },
    {
      "epoch": 9.273377337733773,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.019452455513891777,
      "loss": 0.2325,
      "num_input_tokens_seen": 17790656,
      "step": 84295
    },
    {
      "epoch": 9.273927392739274,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01945108035504767,
      "loss": 0.2309,
      "num_input_tokens_seen": 17791776,
      "step": 84300
    },
    {
      "epoch": 9.274477447744774,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019449705155180092,
      "loss": 0.2298,
      "num_input_tokens_seen": 17792800,
      "step": 84305
    },
    {
      "epoch": 9.275027502750275,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01944832991430172,
      "loss": 0.2324,
      "num_input_tokens_seen": 17793792,
      "step": 84310
    },
    {
      "epoch": 9.275577557755776,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.01944695463242522,
      "loss": 0.2298,
      "num_input_tokens_seen": 17794848,
      "step": 84315
    },
    {
      "epoch": 9.276127612761275,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019445579309563275,
      "loss": 0.2314,
      "num_input_tokens_seen": 17795936,
      "step": 84320
    },
    {
      "epoch": 9.276677667766776,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01944420394572856,
      "loss": 0.2324,
      "num_input_tokens_seen": 17796928,
      "step": 84325
    },
    {
      "epoch": 9.277227722772277,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.019442828540933745,
      "loss": 0.2283,
      "num_input_tokens_seen": 17798080,
      "step": 84330
    },
    {
      "epoch": 9.277777777777779,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01944145309519152,
      "loss": 0.2308,
      "num_input_tokens_seen": 17799168,
      "step": 84335
    },
    {
      "epoch": 9.278327832783278,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.019440077608514548,
      "loss": 0.2314,
      "num_input_tokens_seen": 17800160,
      "step": 84340
    },
    {
      "epoch": 9.278877887788779,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.01943870208091551,
      "loss": 0.2329,
      "num_input_tokens_seen": 17801216,
      "step": 84345
    },
    {
      "epoch": 9.27942794279428,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.019437326512407087,
      "loss": 0.2298,
      "num_input_tokens_seen": 17802336,
      "step": 84350
    },
    {
      "epoch": 9.27997799779978,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.019435950903001955,
      "loss": 0.2324,
      "num_input_tokens_seen": 17803360,
      "step": 84355
    },
    {
      "epoch": 9.28052805280528,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019434575252712787,
      "loss": 0.2309,
      "num_input_tokens_seen": 17804416,
      "step": 84360
    },
    {
      "epoch": 9.281078107810782,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.019433199561552275,
      "loss": 0.2319,
      "num_input_tokens_seen": 17805472,
      "step": 84365
    },
    {
      "epoch": 9.281628162816281,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.019431823829533085,
      "loss": 0.2319,
      "num_input_tokens_seen": 17806528,
      "step": 84370
    },
    {
      "epoch": 9.282178217821782,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.0194304480566679,
      "loss": 0.2314,
      "num_input_tokens_seen": 17807584,
      "step": 84375
    },
    {
      "epoch": 9.282728272827283,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019429072242969407,
      "loss": 0.2329,
      "num_input_tokens_seen": 17808576,
      "step": 84380
    },
    {
      "epoch": 9.283278327832782,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019427696388450277,
      "loss": 0.2303,
      "num_input_tokens_seen": 17809664,
      "step": 84385
    },
    {
      "epoch": 9.283828382838283,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.019426320493123192,
      "loss": 0.2303,
      "num_input_tokens_seen": 17810752,
      "step": 84390
    },
    {
      "epoch": 9.284378437843785,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.019424944557000837,
      "loss": 0.2303,
      "num_input_tokens_seen": 17811776,
      "step": 84395
    },
    {
      "epoch": 9.284928492849286,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01942356858009589,
      "loss": 0.2314,
      "num_input_tokens_seen": 17812800,
      "step": 84400
    },
    {
      "epoch": 9.285478547854785,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019422192562421038,
      "loss": 0.2314,
      "num_input_tokens_seen": 17813824,
      "step": 84405
    },
    {
      "epoch": 9.286028602860286,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.019420816503988955,
      "loss": 0.2324,
      "num_input_tokens_seen": 17814944,
      "step": 84410
    },
    {
      "epoch": 9.286578657865787,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01941944040481233,
      "loss": 0.2309,
      "num_input_tokens_seen": 17816032,
      "step": 84415
    },
    {
      "epoch": 9.287128712871286,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.019418064264903837,
      "loss": 0.2314,
      "num_input_tokens_seen": 17817056,
      "step": 84420
    },
    {
      "epoch": 9.287678767876788,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.019416688084276174,
      "loss": 0.2324,
      "num_input_tokens_seen": 17818048,
      "step": 84425
    },
    {
      "epoch": 9.288228822882289,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.019415311862942014,
      "loss": 0.2303,
      "num_input_tokens_seen": 17819072,
      "step": 84430
    },
    {
      "epoch": 9.288778877887788,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.01941393560091404,
      "loss": 0.2303,
      "num_input_tokens_seen": 17820096,
      "step": 84435
    },
    {
      "epoch": 9.289328932893289,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.019412559298204943,
      "loss": 0.2319,
      "num_input_tokens_seen": 17821120,
      "step": 84440
    },
    {
      "epoch": 9.28987898789879,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.019411182954827404,
      "loss": 0.2319,
      "num_input_tokens_seen": 17822144,
      "step": 84445
    },
    {
      "epoch": 9.290429042904291,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.019409806570794108,
      "loss": 0.2319,
      "num_input_tokens_seen": 17823232,
      "step": 84450
    },
    {
      "epoch": 9.29097909790979,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.01940843014611774,
      "loss": 0.2304,
      "num_input_tokens_seen": 17824256,
      "step": 84455
    },
    {
      "epoch": 9.291529152915292,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019407053680810988,
      "loss": 0.2314,
      "num_input_tokens_seen": 17825344,
      "step": 84460
    },
    {
      "epoch": 9.292079207920793,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019405677174886535,
      "loss": 0.2324,
      "num_input_tokens_seen": 17826336,
      "step": 84465
    },
    {
      "epoch": 9.292629262926292,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.01940430062835707,
      "loss": 0.2293,
      "num_input_tokens_seen": 17827360,
      "step": 84470
    },
    {
      "epoch": 9.293179317931793,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.01940292404123528,
      "loss": 0.2303,
      "num_input_tokens_seen": 17828384,
      "step": 84475
    },
    {
      "epoch": 9.293729372937294,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019401547413533855,
      "loss": 0.2314,
      "num_input_tokens_seen": 17829440,
      "step": 84480
    },
    {
      "epoch": 9.294279427942794,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01940017074526548,
      "loss": 0.2314,
      "num_input_tokens_seen": 17830528,
      "step": 84485
    },
    {
      "epoch": 9.294829482948295,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.019398794036442832,
      "loss": 0.2324,
      "num_input_tokens_seen": 17831552,
      "step": 84490
    },
    {
      "epoch": 9.295379537953796,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019397417287078618,
      "loss": 0.2298,
      "num_input_tokens_seen": 17832640,
      "step": 84495
    },
    {
      "epoch": 9.295929592959295,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.01939604049718552,
      "loss": 0.2314,
      "num_input_tokens_seen": 17833664,
      "step": 84500
    },
    {
      "epoch": 9.296479647964796,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019394663666776224,
      "loss": 0.2304,
      "num_input_tokens_seen": 17834752,
      "step": 84505
    },
    {
      "epoch": 9.297029702970297,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.019393286795863423,
      "loss": 0.2345,
      "num_input_tokens_seen": 17835840,
      "step": 84510
    },
    {
      "epoch": 9.297579757975798,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019391909884459806,
      "loss": 0.2319,
      "num_input_tokens_seen": 17836832,
      "step": 84515
    },
    {
      "epoch": 9.298129812981298,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.019390532932578055,
      "loss": 0.2319,
      "num_input_tokens_seen": 17837824,
      "step": 84520
    },
    {
      "epoch": 9.298679867986799,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019389155940230884,
      "loss": 0.2319,
      "num_input_tokens_seen": 17838784,
      "step": 84525
    },
    {
      "epoch": 9.2992299229923,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019387778907430958,
      "loss": 0.2324,
      "num_input_tokens_seen": 17839840,
      "step": 84530
    },
    {
      "epoch": 9.2997799779978,
      "grad_norm": 0.000579833984375,
      "learning_rate": 0.01938640183419098,
      "loss": 0.2293,
      "num_input_tokens_seen": 17840896,
      "step": 84535
    },
    {
      "epoch": 9.3003300330033,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019385024720523645,
      "loss": 0.2329,
      "num_input_tokens_seen": 17842016,
      "step": 84540
    },
    {
      "epoch": 9.300880088008801,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.01938364756644164,
      "loss": 0.2335,
      "num_input_tokens_seen": 17843104,
      "step": 84545
    },
    {
      "epoch": 9.3014301430143,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01938227037195766,
      "loss": 0.2319,
      "num_input_tokens_seen": 17844224,
      "step": 84550
    },
    {
      "epoch": 9.301980198019802,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.019380893137084392,
      "loss": 0.2303,
      "num_input_tokens_seen": 17845248,
      "step": 84555
    },
    {
      "epoch": 9.302530253025303,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.019379515861834538,
      "loss": 0.2314,
      "num_input_tokens_seen": 17846336,
      "step": 84560
    },
    {
      "epoch": 9.303080308030804,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.019378138546220787,
      "loss": 0.2303,
      "num_input_tokens_seen": 17847456,
      "step": 84565
    },
    {
      "epoch": 9.303630363036303,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.019376761190255835,
      "loss": 0.2314,
      "num_input_tokens_seen": 17848512,
      "step": 84570
    },
    {
      "epoch": 9.304180418041804,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.01937538379395237,
      "loss": 0.2319,
      "num_input_tokens_seen": 17849568,
      "step": 84575
    },
    {
      "epoch": 9.304730473047305,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.019374006357323097,
      "loss": 0.2314,
      "num_input_tokens_seen": 17850656,
      "step": 84580
    },
    {
      "epoch": 9.305280528052805,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.019372628880380705,
      "loss": 0.2314,
      "num_input_tokens_seen": 17851712,
      "step": 84585
    },
    {
      "epoch": 9.305830583058306,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.01937125136313789,
      "loss": 0.2309,
      "num_input_tokens_seen": 17852800,
      "step": 84590
    },
    {
      "epoch": 9.306380638063807,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01936987380560735,
      "loss": 0.2319,
      "num_input_tokens_seen": 17853856,
      "step": 84595
    },
    {
      "epoch": 9.306930693069306,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01936849620780178,
      "loss": 0.2304,
      "num_input_tokens_seen": 17854944,
      "step": 84600
    },
    {
      "epoch": 9.307480748074807,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.019367118569733877,
      "loss": 0.2304,
      "num_input_tokens_seen": 17856032,
      "step": 84605
    },
    {
      "epoch": 9.308030803080309,
      "grad_norm": 0.00054168701171875,
      "learning_rate": 0.019365740891416336,
      "loss": 0.2315,
      "num_input_tokens_seen": 17857024,
      "step": 84610
    },
    {
      "epoch": 9.308580858085808,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01936436317286186,
      "loss": 0.2309,
      "num_input_tokens_seen": 17858144,
      "step": 84615
    },
    {
      "epoch": 9.309130913091309,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.01936298541408314,
      "loss": 0.2298,
      "num_input_tokens_seen": 17859200,
      "step": 84620
    },
    {
      "epoch": 9.30968096809681,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.019361607615092875,
      "loss": 0.2309,
      "num_input_tokens_seen": 17860288,
      "step": 84625
    },
    {
      "epoch": 9.310231023102311,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.019360229775903768,
      "loss": 0.2314,
      "num_input_tokens_seen": 17861312,
      "step": 84630
    },
    {
      "epoch": 9.31078107810781,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.01935885189652851,
      "loss": 0.2309,
      "num_input_tokens_seen": 17862368,
      "step": 84635
    },
    {
      "epoch": 9.311331133113312,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.019357473976979817,
      "loss": 0.2319,
      "num_input_tokens_seen": 17863392,
      "step": 84640
    },
    {
      "epoch": 9.311881188118813,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.019356096017270372,
      "loss": 0.2325,
      "num_input_tokens_seen": 17864512,
      "step": 84645
    },
    {
      "epoch": 9.312431243124312,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019354718017412877,
      "loss": 0.2309,
      "num_input_tokens_seen": 17865600,
      "step": 84650
    },
    {
      "epoch": 9.312981298129813,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.019353339977420038,
      "loss": 0.2314,
      "num_input_tokens_seen": 17866656,
      "step": 84655
    },
    {
      "epoch": 9.313531353135314,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.019351961897304554,
      "loss": 0.233,
      "num_input_tokens_seen": 17867680,
      "step": 84660
    },
    {
      "epoch": 9.314081408140813,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.019350583777079124,
      "loss": 0.2324,
      "num_input_tokens_seen": 17868768,
      "step": 84665
    },
    {
      "epoch": 9.314631463146315,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019349205616756456,
      "loss": 0.2304,
      "num_input_tokens_seen": 17869760,
      "step": 84670
    },
    {
      "epoch": 9.315181518151816,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01934782741634924,
      "loss": 0.2303,
      "num_input_tokens_seen": 17870784,
      "step": 84675
    },
    {
      "epoch": 9.315731573157315,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019346449175870185,
      "loss": 0.2319,
      "num_input_tokens_seen": 17871808,
      "step": 84680
    },
    {
      "epoch": 9.316281628162816,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019345070895331995,
      "loss": 0.233,
      "num_input_tokens_seen": 17872896,
      "step": 84685
    },
    {
      "epoch": 9.316831683168317,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.019343692574747377,
      "loss": 0.2319,
      "num_input_tokens_seen": 17873952,
      "step": 84690
    },
    {
      "epoch": 9.317381738173818,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01934231421412902,
      "loss": 0.2319,
      "num_input_tokens_seen": 17875072,
      "step": 84695
    },
    {
      "epoch": 9.317931793179318,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.01934093581348964,
      "loss": 0.2329,
      "num_input_tokens_seen": 17876160,
      "step": 84700
    },
    {
      "epoch": 9.318481848184819,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.01933955737284194,
      "loss": 0.2298,
      "num_input_tokens_seen": 17877248,
      "step": 84705
    },
    {
      "epoch": 9.31903190319032,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.019338178892198618,
      "loss": 0.2319,
      "num_input_tokens_seen": 17878272,
      "step": 84710
    },
    {
      "epoch": 9.319581958195819,
      "grad_norm": 0.0093994140625,
      "learning_rate": 0.01933680037157239,
      "loss": 0.2308,
      "num_input_tokens_seen": 17879296,
      "step": 84715
    },
    {
      "epoch": 9.32013201320132,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.019335421810975948,
      "loss": 0.2324,
      "num_input_tokens_seen": 17880320,
      "step": 84720
    },
    {
      "epoch": 9.320682068206821,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.019334043210422,
      "loss": 0.2324,
      "num_input_tokens_seen": 17881440,
      "step": 84725
    },
    {
      "epoch": 9.32123212321232,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.019332664569923264,
      "loss": 0.2313,
      "num_input_tokens_seen": 17882528,
      "step": 84730
    },
    {
      "epoch": 9.321782178217822,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.019331285889492435,
      "loss": 0.2319,
      "num_input_tokens_seen": 17883520,
      "step": 84735
    },
    {
      "epoch": 9.322332233223323,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.01932990716914222,
      "loss": 0.2319,
      "num_input_tokens_seen": 17884544,
      "step": 84740
    },
    {
      "epoch": 9.322882288228822,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.01932852840888533,
      "loss": 0.2324,
      "num_input_tokens_seen": 17885568,
      "step": 84745
    },
    {
      "epoch": 9.323432343234323,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.01932714960873447,
      "loss": 0.2303,
      "num_input_tokens_seen": 17886624,
      "step": 84750
    },
    {
      "epoch": 9.323982398239824,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.019325770768702344,
      "loss": 0.2319,
      "num_input_tokens_seen": 17887680,
      "step": 84755
    },
    {
      "epoch": 9.324532453245325,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.019324391888801674,
      "loss": 0.2303,
      "num_input_tokens_seen": 17888736,
      "step": 84760
    },
    {
      "epoch": 9.325082508250825,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.019323012969045154,
      "loss": 0.2319,
      "num_input_tokens_seen": 17889792,
      "step": 84765
    },
    {
      "epoch": 9.325632563256326,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019321634009445497,
      "loss": 0.2319,
      "num_input_tokens_seen": 17890848,
      "step": 84770
    },
    {
      "epoch": 9.326182618261827,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019320255010015414,
      "loss": 0.2319,
      "num_input_tokens_seen": 17891936,
      "step": 84775
    },
    {
      "epoch": 9.326732673267326,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.019318875970767615,
      "loss": 0.2303,
      "num_input_tokens_seen": 17893024,
      "step": 84780
    },
    {
      "epoch": 9.327282728272827,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019317496891714806,
      "loss": 0.2329,
      "num_input_tokens_seen": 17894080,
      "step": 84785
    },
    {
      "epoch": 9.327832783278328,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.019316117772869703,
      "loss": 0.2303,
      "num_input_tokens_seen": 17895136,
      "step": 84790
    },
    {
      "epoch": 9.328382838283828,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.01931473861424501,
      "loss": 0.2308,
      "num_input_tokens_seen": 17896288,
      "step": 84795
    },
    {
      "epoch": 9.328932893289329,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019313359415853444,
      "loss": 0.2319,
      "num_input_tokens_seen": 17897376,
      "step": 84800
    },
    {
      "epoch": 9.32948294829483,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019311980177707713,
      "loss": 0.2309,
      "num_input_tokens_seen": 17898464,
      "step": 84805
    },
    {
      "epoch": 9.33003300330033,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01931060089982053,
      "loss": 0.2309,
      "num_input_tokens_seen": 17899552,
      "step": 84810
    },
    {
      "epoch": 9.33058305830583,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01930922158220461,
      "loss": 0.2325,
      "num_input_tokens_seen": 17900576,
      "step": 84815
    },
    {
      "epoch": 9.331133113311331,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.019307842224872657,
      "loss": 0.2334,
      "num_input_tokens_seen": 17901632,
      "step": 84820
    },
    {
      "epoch": 9.331683168316832,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01930646282783739,
      "loss": 0.2308,
      "num_input_tokens_seen": 17902656,
      "step": 84825
    },
    {
      "epoch": 9.332233223322332,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.01930508339111153,
      "loss": 0.2309,
      "num_input_tokens_seen": 17903712,
      "step": 84830
    },
    {
      "epoch": 9.332783278327833,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.01930370391470777,
      "loss": 0.2319,
      "num_input_tokens_seen": 17904704,
      "step": 84835
    },
    {
      "epoch": 9.333333333333334,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.01930232439863884,
      "loss": 0.2308,
      "num_input_tokens_seen": 17905728,
      "step": 84840
    },
    {
      "epoch": 9.333883388338833,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.019300944842917455,
      "loss": 0.2303,
      "num_input_tokens_seen": 17906752,
      "step": 84845
    },
    {
      "epoch": 9.334433443344334,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.01929956524755632,
      "loss": 0.2324,
      "num_input_tokens_seen": 17907840,
      "step": 84850
    },
    {
      "epoch": 9.334983498349835,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.01929818561256816,
      "loss": 0.2314,
      "num_input_tokens_seen": 17908928,
      "step": 84855
    },
    {
      "epoch": 9.335533553355335,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.019296805937965684,
      "loss": 0.2314,
      "num_input_tokens_seen": 17909952,
      "step": 84860
    },
    {
      "epoch": 9.336083608360836,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019295426223761607,
      "loss": 0.2309,
      "num_input_tokens_seen": 17910944,
      "step": 84865
    },
    {
      "epoch": 9.336633663366337,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.019294046469968646,
      "loss": 0.2324,
      "num_input_tokens_seen": 17912000,
      "step": 84870
    },
    {
      "epoch": 9.337183718371838,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.019292666676599526,
      "loss": 0.2308,
      "num_input_tokens_seen": 17913120,
      "step": 84875
    },
    {
      "epoch": 9.337733773377337,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.019291286843666954,
      "loss": 0.233,
      "num_input_tokens_seen": 17914240,
      "step": 84880
    },
    {
      "epoch": 9.338283828382838,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.01928990697118365,
      "loss": 0.233,
      "num_input_tokens_seen": 17915264,
      "step": 84885
    },
    {
      "epoch": 9.33883388338834,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01928852705916233,
      "loss": 0.2283,
      "num_input_tokens_seen": 17916352,
      "step": 84890
    },
    {
      "epoch": 9.339383938393839,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.019287147107615716,
      "loss": 0.2309,
      "num_input_tokens_seen": 17917440,
      "step": 84895
    },
    {
      "epoch": 9.33993399339934,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01928576711655652,
      "loss": 0.233,
      "num_input_tokens_seen": 17918496,
      "step": 84900
    },
    {
      "epoch": 9.340484048404841,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.019284387085997477,
      "loss": 0.2309,
      "num_input_tokens_seen": 17919616,
      "step": 84905
    },
    {
      "epoch": 9.34103410341034,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.01928300701595128,
      "loss": 0.2293,
      "num_input_tokens_seen": 17920672,
      "step": 84910
    },
    {
      "epoch": 9.341584158415841,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.01928162690643067,
      "loss": 0.2324,
      "num_input_tokens_seen": 17921792,
      "step": 84915
    },
    {
      "epoch": 9.342134213421343,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01928024675744836,
      "loss": 0.2319,
      "num_input_tokens_seen": 17922848,
      "step": 84920
    },
    {
      "epoch": 9.342684268426842,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.019278866569017065,
      "loss": 0.2329,
      "num_input_tokens_seen": 17923840,
      "step": 84925
    },
    {
      "epoch": 9.343234323432343,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019277486341149515,
      "loss": 0.233,
      "num_input_tokens_seen": 17924896,
      "step": 84930
    },
    {
      "epoch": 9.343784378437844,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.019276106073858424,
      "loss": 0.2314,
      "num_input_tokens_seen": 17925920,
      "step": 84935
    },
    {
      "epoch": 9.344334433443345,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.01927472576715651,
      "loss": 0.2314,
      "num_input_tokens_seen": 17926944,
      "step": 84940
    },
    {
      "epoch": 9.344884488448844,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.019273345421056506,
      "loss": 0.2314,
      "num_input_tokens_seen": 17927936,
      "step": 84945
    },
    {
      "epoch": 9.345434543454346,
      "grad_norm": 0.009765625,
      "learning_rate": 0.019271965035571128,
      "loss": 0.2309,
      "num_input_tokens_seen": 17928960,
      "step": 84950
    },
    {
      "epoch": 9.345984598459847,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.019270584610713095,
      "loss": 0.2319,
      "num_input_tokens_seen": 17930016,
      "step": 84955
    },
    {
      "epoch": 9.346534653465346,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.019269204146495134,
      "loss": 0.2324,
      "num_input_tokens_seen": 17931040,
      "step": 84960
    },
    {
      "epoch": 9.347084708470847,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01926782364292997,
      "loss": 0.2308,
      "num_input_tokens_seen": 17932160,
      "step": 84965
    },
    {
      "epoch": 9.347634763476348,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.019266443100030315,
      "loss": 0.2309,
      "num_input_tokens_seen": 17933184,
      "step": 84970
    },
    {
      "epoch": 9.348184818481847,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.019265062517808913,
      "loss": 0.2283,
      "num_input_tokens_seen": 17934176,
      "step": 84975
    },
    {
      "epoch": 9.348734873487349,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01926368189627847,
      "loss": 0.2309,
      "num_input_tokens_seen": 17935200,
      "step": 84980
    },
    {
      "epoch": 9.34928492849285,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.019262301235451713,
      "loss": 0.2314,
      "num_input_tokens_seen": 17936320,
      "step": 84985
    },
    {
      "epoch": 9.34983498349835,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.019260920535341376,
      "loss": 0.2324,
      "num_input_tokens_seen": 17937440,
      "step": 84990
    },
    {
      "epoch": 9.35038503850385,
      "grad_norm": 0.000888824462890625,
      "learning_rate": 0.019259539795960175,
      "loss": 0.2319,
      "num_input_tokens_seen": 17938432,
      "step": 84995
    },
    {
      "epoch": 9.350935093509351,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.019258159017320843,
      "loss": 0.2293,
      "num_input_tokens_seen": 17939488,
      "step": 85000
    },
    {
      "epoch": 9.351485148514852,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019256778199436105,
      "loss": 0.2324,
      "num_input_tokens_seen": 17940512,
      "step": 85005
    },
    {
      "epoch": 9.352035203520352,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.01925539734231868,
      "loss": 0.2324,
      "num_input_tokens_seen": 17941568,
      "step": 85010
    },
    {
      "epoch": 9.352585258525853,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.0192540164459813,
      "loss": 0.2303,
      "num_input_tokens_seen": 17942592,
      "step": 85015
    },
    {
      "epoch": 9.353135313531354,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.0192526355104367,
      "loss": 0.2319,
      "num_input_tokens_seen": 17943648,
      "step": 85020
    },
    {
      "epoch": 9.353685368536853,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.01925125453569759,
      "loss": 0.2329,
      "num_input_tokens_seen": 17944640,
      "step": 85025
    },
    {
      "epoch": 9.354235423542354,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.019249873521776706,
      "loss": 0.2324,
      "num_input_tokens_seen": 17945664,
      "step": 85030
    },
    {
      "epoch": 9.354785478547855,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.019248492468686783,
      "loss": 0.2329,
      "num_input_tokens_seen": 17946720,
      "step": 85035
    },
    {
      "epoch": 9.355335533553355,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.019247111376440543,
      "loss": 0.2324,
      "num_input_tokens_seen": 17947808,
      "step": 85040
    },
    {
      "epoch": 9.355885588558856,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01924573024505071,
      "loss": 0.2324,
      "num_input_tokens_seen": 17948864,
      "step": 85045
    },
    {
      "epoch": 9.356435643564357,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.019244349074530022,
      "loss": 0.2324,
      "num_input_tokens_seen": 17949920,
      "step": 85050
    },
    {
      "epoch": 9.356985698569858,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019242967864891204,
      "loss": 0.2335,
      "num_input_tokens_seen": 17950944,
      "step": 85055
    },
    {
      "epoch": 9.357535753575357,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01924158661614699,
      "loss": 0.2319,
      "num_input_tokens_seen": 17952000,
      "step": 85060
    },
    {
      "epoch": 9.358085808580858,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.019240205328310108,
      "loss": 0.2319,
      "num_input_tokens_seen": 17953024,
      "step": 85065
    },
    {
      "epoch": 9.35863586358636,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.019238824001393283,
      "loss": 0.2308,
      "num_input_tokens_seen": 17954080,
      "step": 85070
    },
    {
      "epoch": 9.359185918591859,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.019237442635409253,
      "loss": 0.2298,
      "num_input_tokens_seen": 17955136,
      "step": 85075
    },
    {
      "epoch": 9.35973597359736,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.01923606123037075,
      "loss": 0.2319,
      "num_input_tokens_seen": 17956160,
      "step": 85080
    },
    {
      "epoch": 9.36028602860286,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.0192346797862905,
      "loss": 0.2345,
      "num_input_tokens_seen": 17957184,
      "step": 85085
    },
    {
      "epoch": 9.36083608360836,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.01923329830318124,
      "loss": 0.2319,
      "num_input_tokens_seen": 17958208,
      "step": 85090
    },
    {
      "epoch": 9.361386138613861,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.019231916781055704,
      "loss": 0.2314,
      "num_input_tokens_seen": 17959328,
      "step": 85095
    },
    {
      "epoch": 9.361936193619362,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.01923053521992662,
      "loss": 0.2314,
      "num_input_tokens_seen": 17960448,
      "step": 85100
    },
    {
      "epoch": 9.362486248624862,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.01922915361980672,
      "loss": 0.2324,
      "num_input_tokens_seen": 17961536,
      "step": 85105
    },
    {
      "epoch": 9.363036303630363,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.01922777198070874,
      "loss": 0.2345,
      "num_input_tokens_seen": 17962624,
      "step": 85110
    },
    {
      "epoch": 9.363586358635864,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.019226390302645415,
      "loss": 0.2308,
      "num_input_tokens_seen": 17963744,
      "step": 85115
    },
    {
      "epoch": 9.364136413641365,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.019225008585629484,
      "loss": 0.2329,
      "num_input_tokens_seen": 17964896,
      "step": 85120
    },
    {
      "epoch": 9.364686468646864,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019223626829673674,
      "loss": 0.2314,
      "num_input_tokens_seen": 17965984,
      "step": 85125
    },
    {
      "epoch": 9.365236523652365,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.019222245034790718,
      "loss": 0.233,
      "num_input_tokens_seen": 17966976,
      "step": 85130
    },
    {
      "epoch": 9.365786578657866,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.01922086320099336,
      "loss": 0.2314,
      "num_input_tokens_seen": 17968032,
      "step": 85135
    },
    {
      "epoch": 9.366336633663366,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.01921948132829433,
      "loss": 0.2329,
      "num_input_tokens_seen": 17969056,
      "step": 85140
    },
    {
      "epoch": 9.366886688668867,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.019218099416706363,
      "loss": 0.2309,
      "num_input_tokens_seen": 17970112,
      "step": 85145
    },
    {
      "epoch": 9.367436743674368,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.0192167174662422,
      "loss": 0.2329,
      "num_input_tokens_seen": 17971200,
      "step": 85150
    },
    {
      "epoch": 9.367986798679867,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.019215335476914577,
      "loss": 0.2319,
      "num_input_tokens_seen": 17972288,
      "step": 85155
    },
    {
      "epoch": 9.368536853685368,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01921395344873622,
      "loss": 0.2314,
      "num_input_tokens_seen": 17973376,
      "step": 85160
    },
    {
      "epoch": 9.36908690869087,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.01921257138171989,
      "loss": 0.2314,
      "num_input_tokens_seen": 17974464,
      "step": 85165
    },
    {
      "epoch": 9.369636963696369,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.019211189275878306,
      "loss": 0.2309,
      "num_input_tokens_seen": 17975488,
      "step": 85170
    },
    {
      "epoch": 9.37018701870187,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.019209807131224204,
      "loss": 0.2314,
      "num_input_tokens_seen": 17976480,
      "step": 85175
    },
    {
      "epoch": 9.370737073707371,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019208424947770336,
      "loss": 0.2319,
      "num_input_tokens_seen": 17977536,
      "step": 85180
    },
    {
      "epoch": 9.371287128712872,
      "grad_norm": 0.000762939453125,
      "learning_rate": 0.019207042725529436,
      "loss": 0.2325,
      "num_input_tokens_seen": 17978528,
      "step": 85185
    },
    {
      "epoch": 9.371837183718371,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.01920566046451424,
      "loss": 0.2314,
      "num_input_tokens_seen": 17979648,
      "step": 85190
    },
    {
      "epoch": 9.372387238723872,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.01920427816473749,
      "loss": 0.2283,
      "num_input_tokens_seen": 17980768,
      "step": 85195
    },
    {
      "epoch": 9.372937293729374,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019202895826211926,
      "loss": 0.2345,
      "num_input_tokens_seen": 17981760,
      "step": 85200
    },
    {
      "epoch": 9.373487348734873,
      "grad_norm": 0.009765625,
      "learning_rate": 0.019201513448950283,
      "loss": 0.2319,
      "num_input_tokens_seen": 17982816,
      "step": 85205
    },
    {
      "epoch": 9.374037403740374,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019200131032965313,
      "loss": 0.2319,
      "num_input_tokens_seen": 17983904,
      "step": 85210
    },
    {
      "epoch": 9.374587458745875,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.01919874857826975,
      "loss": 0.2303,
      "num_input_tokens_seen": 17984992,
      "step": 85215
    },
    {
      "epoch": 9.375137513751374,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.019197366084876332,
      "loss": 0.2314,
      "num_input_tokens_seen": 17985984,
      "step": 85220
    },
    {
      "epoch": 9.375687568756875,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.01919598355279781,
      "loss": 0.2314,
      "num_input_tokens_seen": 17987072,
      "step": 85225
    },
    {
      "epoch": 9.376237623762377,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01919460098204692,
      "loss": 0.2304,
      "num_input_tokens_seen": 17988224,
      "step": 85230
    },
    {
      "epoch": 9.376787678767876,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019193218372636402,
      "loss": 0.2335,
      "num_input_tokens_seen": 17989280,
      "step": 85235
    },
    {
      "epoch": 9.377337733773377,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.019191835724579005,
      "loss": 0.2314,
      "num_input_tokens_seen": 17990304,
      "step": 85240
    },
    {
      "epoch": 9.377887788778878,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.019190453037887464,
      "loss": 0.2288,
      "num_input_tokens_seen": 17991392,
      "step": 85245
    },
    {
      "epoch": 9.37843784378438,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.019189070312574535,
      "loss": 0.2319,
      "num_input_tokens_seen": 17992448,
      "step": 85250
    },
    {
      "epoch": 9.378987898789878,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01918768754865295,
      "loss": 0.2325,
      "num_input_tokens_seen": 17993504,
      "step": 85255
    },
    {
      "epoch": 9.37953795379538,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.019186304746135464,
      "loss": 0.2319,
      "num_input_tokens_seen": 17994624,
      "step": 85260
    },
    {
      "epoch": 9.38008800880088,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019184921905034815,
      "loss": 0.233,
      "num_input_tokens_seen": 17995744,
      "step": 85265
    },
    {
      "epoch": 9.38063806380638,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01918353902536375,
      "loss": 0.2319,
      "num_input_tokens_seen": 17996768,
      "step": 85270
    },
    {
      "epoch": 9.381188118811881,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.019182156107135005,
      "loss": 0.2309,
      "num_input_tokens_seen": 17997824,
      "step": 85275
    },
    {
      "epoch": 9.381738173817382,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01918077315036134,
      "loss": 0.2314,
      "num_input_tokens_seen": 17998880,
      "step": 85280
    },
    {
      "epoch": 9.382288228822881,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.019179390155055497,
      "loss": 0.2324,
      "num_input_tokens_seen": 18000032,
      "step": 85285
    },
    {
      "epoch": 9.382838283828383,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.019178007121230217,
      "loss": 0.2304,
      "num_input_tokens_seen": 18001152,
      "step": 85290
    },
    {
      "epoch": 9.383388338833884,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01917662404889825,
      "loss": 0.2314,
      "num_input_tokens_seen": 18002208,
      "step": 85295
    },
    {
      "epoch": 9.383938393839385,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.019175240938072345,
      "loss": 0.2319,
      "num_input_tokens_seen": 18003328,
      "step": 85300
    },
    {
      "epoch": 9.384488448844884,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.019173857788765247,
      "loss": 0.2314,
      "num_input_tokens_seen": 18004384,
      "step": 85305
    },
    {
      "epoch": 9.385038503850385,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.019172474600989704,
      "loss": 0.2346,
      "num_input_tokens_seen": 18005440,
      "step": 85310
    },
    {
      "epoch": 9.385588558855886,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.019171091374758465,
      "loss": 0.2319,
      "num_input_tokens_seen": 18006496,
      "step": 85315
    },
    {
      "epoch": 9.386138613861386,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01916970811008428,
      "loss": 0.2313,
      "num_input_tokens_seen": 18007488,
      "step": 85320
    },
    {
      "epoch": 9.386688668866887,
      "grad_norm": 0.004608154296875,
      "learning_rate": 0.019168324806979893,
      "loss": 0.2303,
      "num_input_tokens_seen": 18008544,
      "step": 85325
    },
    {
      "epoch": 9.387238723872388,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.01916694146545806,
      "loss": 0.2309,
      "num_input_tokens_seen": 18009536,
      "step": 85330
    },
    {
      "epoch": 9.387788778877887,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.019165558085531526,
      "loss": 0.2293,
      "num_input_tokens_seen": 18010624,
      "step": 85335
    },
    {
      "epoch": 9.388338833883388,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.01916417466721304,
      "loss": 0.2309,
      "num_input_tokens_seen": 18011712,
      "step": 85340
    },
    {
      "epoch": 9.38888888888889,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.019162791210515357,
      "loss": 0.2288,
      "num_input_tokens_seen": 18012768,
      "step": 85345
    },
    {
      "epoch": 9.389438943894389,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.01916140771545122,
      "loss": 0.2308,
      "num_input_tokens_seen": 18013824,
      "step": 85350
    },
    {
      "epoch": 9.38998899889989,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01916002418203339,
      "loss": 0.2329,
      "num_input_tokens_seen": 18014848,
      "step": 85355
    },
    {
      "epoch": 9.39053905390539,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.01915864061027461,
      "loss": 0.2329,
      "num_input_tokens_seen": 18015904,
      "step": 85360
    },
    {
      "epoch": 9.391089108910892,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01915725700018764,
      "loss": 0.2329,
      "num_input_tokens_seen": 18017024,
      "step": 85365
    },
    {
      "epoch": 9.391639163916391,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.019155873351785223,
      "loss": 0.2309,
      "num_input_tokens_seen": 18018080,
      "step": 85370
    },
    {
      "epoch": 9.392189218921892,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.019154489665080116,
      "loss": 0.2319,
      "num_input_tokens_seen": 18019104,
      "step": 85375
    },
    {
      "epoch": 9.392739273927393,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.019153105940085072,
      "loss": 0.233,
      "num_input_tokens_seen": 18020160,
      "step": 85380
    },
    {
      "epoch": 9.393289328932893,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01915172217681284,
      "loss": 0.2309,
      "num_input_tokens_seen": 18021184,
      "step": 85385
    },
    {
      "epoch": 9.393839383938394,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.01915033837527618,
      "loss": 0.2314,
      "num_input_tokens_seen": 18022208,
      "step": 85390
    },
    {
      "epoch": 9.394389438943895,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01914895453548784,
      "loss": 0.2314,
      "num_input_tokens_seen": 18023296,
      "step": 85395
    },
    {
      "epoch": 9.394939493949394,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.019147570657460585,
      "loss": 0.2329,
      "num_input_tokens_seen": 18024352,
      "step": 85400
    },
    {
      "epoch": 9.395489548954895,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.019146186741207154,
      "loss": 0.2319,
      "num_input_tokens_seen": 18025312,
      "step": 85405
    },
    {
      "epoch": 9.396039603960396,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01914480278674031,
      "loss": 0.2303,
      "num_input_tokens_seen": 18026336,
      "step": 85410
    },
    {
      "epoch": 9.396589658965897,
      "grad_norm": 0.00077056884765625,
      "learning_rate": 0.01914341879407281,
      "loss": 0.2329,
      "num_input_tokens_seen": 18027392,
      "step": 85415
    },
    {
      "epoch": 9.397139713971397,
      "grad_norm": 0.009765625,
      "learning_rate": 0.0191420347632174,
      "loss": 0.2335,
      "num_input_tokens_seen": 18028448,
      "step": 85420
    },
    {
      "epoch": 9.397689768976898,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.019140650694186856,
      "loss": 0.2293,
      "num_input_tokens_seen": 18029536,
      "step": 85425
    },
    {
      "epoch": 9.398239823982399,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019139266586993914,
      "loss": 0.2319,
      "num_input_tokens_seen": 18030624,
      "step": 85430
    },
    {
      "epoch": 9.398789878987898,
      "grad_norm": 0.000576019287109375,
      "learning_rate": 0.019137882441651335,
      "loss": 0.2314,
      "num_input_tokens_seen": 18031616,
      "step": 85435
    },
    {
      "epoch": 9.3993399339934,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.019136498258171885,
      "loss": 0.2298,
      "num_input_tokens_seen": 18032672,
      "step": 85440
    },
    {
      "epoch": 9.3998899889989,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.019135114036568313,
      "loss": 0.233,
      "num_input_tokens_seen": 18033728,
      "step": 85445
    },
    {
      "epoch": 9.4004400440044,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01913372977685338,
      "loss": 0.2293,
      "num_input_tokens_seen": 18034784,
      "step": 85450
    },
    {
      "epoch": 9.400990099009901,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.01913234547903984,
      "loss": 0.2309,
      "num_input_tokens_seen": 18035872,
      "step": 85455
    },
    {
      "epoch": 9.401540154015402,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.01913096114314046,
      "loss": 0.233,
      "num_input_tokens_seen": 18036864,
      "step": 85460
    },
    {
      "epoch": 9.402090209020901,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019129576769167985,
      "loss": 0.2319,
      "num_input_tokens_seen": 18037952,
      "step": 85465
    },
    {
      "epoch": 9.402640264026402,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01912819235713519,
      "loss": 0.233,
      "num_input_tokens_seen": 18039008,
      "step": 85470
    },
    {
      "epoch": 9.403190319031903,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.019126807907054826,
      "loss": 0.2329,
      "num_input_tokens_seen": 18040096,
      "step": 85475
    },
    {
      "epoch": 9.403740374037405,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019125423418939653,
      "loss": 0.2335,
      "num_input_tokens_seen": 18041248,
      "step": 85480
    },
    {
      "epoch": 9.404290429042904,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.01912403889280243,
      "loss": 0.2314,
      "num_input_tokens_seen": 18042304,
      "step": 85485
    },
    {
      "epoch": 9.404840484048405,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01912265432865592,
      "loss": 0.2319,
      "num_input_tokens_seen": 18043328,
      "step": 85490
    },
    {
      "epoch": 9.405390539053906,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.019121269726512888,
      "loss": 0.2314,
      "num_input_tokens_seen": 18044384,
      "step": 85495
    },
    {
      "epoch": 9.405940594059405,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01911988508638609,
      "loss": 0.2293,
      "num_input_tokens_seen": 18045408,
      "step": 85500
    },
    {
      "epoch": 9.406490649064907,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.01911850040828828,
      "loss": 0.2324,
      "num_input_tokens_seen": 18046464,
      "step": 85505
    },
    {
      "epoch": 9.407040704070408,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019117115692232234,
      "loss": 0.2324,
      "num_input_tokens_seen": 18047520,
      "step": 85510
    },
    {
      "epoch": 9.407590759075907,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.019115730938230706,
      "loss": 0.2303,
      "num_input_tokens_seen": 18048512,
      "step": 85515
    },
    {
      "epoch": 9.408140814081408,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.019114346146296462,
      "loss": 0.2308,
      "num_input_tokens_seen": 18049536,
      "step": 85520
    },
    {
      "epoch": 9.408690869086909,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019112961316442263,
      "loss": 0.2314,
      "num_input_tokens_seen": 18050592,
      "step": 85525
    },
    {
      "epoch": 9.409240924092408,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019111576448680875,
      "loss": 0.2319,
      "num_input_tokens_seen": 18051648,
      "step": 85530
    },
    {
      "epoch": 9.40979097909791,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019110191543025058,
      "loss": 0.2309,
      "num_input_tokens_seen": 18052768,
      "step": 85535
    },
    {
      "epoch": 9.41034103410341,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.019108806599487573,
      "loss": 0.2319,
      "num_input_tokens_seen": 18053856,
      "step": 85540
    },
    {
      "epoch": 9.410891089108912,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.0191074216180812,
      "loss": 0.2314,
      "num_input_tokens_seen": 18054880,
      "step": 85545
    },
    {
      "epoch": 9.411441144114411,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.019106036598818685,
      "loss": 0.2309,
      "num_input_tokens_seen": 18056000,
      "step": 85550
    },
    {
      "epoch": 9.411991199119912,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.019104651541712807,
      "loss": 0.2314,
      "num_input_tokens_seen": 18056992,
      "step": 85555
    },
    {
      "epoch": 9.412541254125413,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.019103266446776318,
      "loss": 0.2308,
      "num_input_tokens_seen": 18058048,
      "step": 85560
    },
    {
      "epoch": 9.413091309130913,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.019101881314021997,
      "loss": 0.2329,
      "num_input_tokens_seen": 18059136,
      "step": 85565
    },
    {
      "epoch": 9.413641364136414,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.0191004961434626,
      "loss": 0.2304,
      "num_input_tokens_seen": 18060160,
      "step": 85570
    },
    {
      "epoch": 9.414191419141915,
      "grad_norm": 0.009765625,
      "learning_rate": 0.019099110935110904,
      "loss": 0.2319,
      "num_input_tokens_seen": 18061184,
      "step": 85575
    },
    {
      "epoch": 9.414741474147414,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.01909772568897966,
      "loss": 0.2319,
      "num_input_tokens_seen": 18062272,
      "step": 85580
    },
    {
      "epoch": 9.415291529152915,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01909634040508165,
      "loss": 0.2329,
      "num_input_tokens_seen": 18063264,
      "step": 85585
    },
    {
      "epoch": 9.415841584158416,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.019094955083429642,
      "loss": 0.2298,
      "num_input_tokens_seen": 18064288,
      "step": 85590
    },
    {
      "epoch": 9.416391639163916,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.019093569724036388,
      "loss": 0.2319,
      "num_input_tokens_seen": 18065344,
      "step": 85595
    },
    {
      "epoch": 9.416941694169417,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01909218432691467,
      "loss": 0.2314,
      "num_input_tokens_seen": 18066400,
      "step": 85600
    },
    {
      "epoch": 9.417491749174918,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.019090798892077255,
      "loss": 0.2329,
      "num_input_tokens_seen": 18067424,
      "step": 85605
    },
    {
      "epoch": 9.418041804180419,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.0190894134195369,
      "loss": 0.2329,
      "num_input_tokens_seen": 18068416,
      "step": 85610
    },
    {
      "epoch": 9.418591859185918,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.019088027909306398,
      "loss": 0.2314,
      "num_input_tokens_seen": 18069472,
      "step": 85615
    },
    {
      "epoch": 9.41914191419142,
      "grad_norm": 0.000728607177734375,
      "learning_rate": 0.0190866423613985,
      "loss": 0.2324,
      "num_input_tokens_seen": 18070528,
      "step": 85620
    },
    {
      "epoch": 9.41969196919692,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.01908525677582597,
      "loss": 0.2319,
      "num_input_tokens_seen": 18071552,
      "step": 85625
    },
    {
      "epoch": 9.42024202420242,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.019083871152601597,
      "loss": 0.2309,
      "num_input_tokens_seen": 18072576,
      "step": 85630
    },
    {
      "epoch": 9.42079207920792,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.019082485491738146,
      "loss": 0.2329,
      "num_input_tokens_seen": 18073600,
      "step": 85635
    },
    {
      "epoch": 9.421342134213422,
      "grad_norm": 0.009765625,
      "learning_rate": 0.01908109979324838,
      "loss": 0.2308,
      "num_input_tokens_seen": 18074656,
      "step": 85640
    },
    {
      "epoch": 9.421892189218921,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.019079714057145078,
      "loss": 0.2324,
      "num_input_tokens_seen": 18075744,
      "step": 85645
    },
    {
      "epoch": 9.422442244224422,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019078328283441012,
      "loss": 0.2319,
      "num_input_tokens_seen": 18076800,
      "step": 85650
    },
    {
      "epoch": 9.422992299229923,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.019076942472148944,
      "loss": 0.2324,
      "num_input_tokens_seen": 18077856,
      "step": 85655
    },
    {
      "epoch": 9.423542354235423,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.019075556623281656,
      "loss": 0.2309,
      "num_input_tokens_seen": 18078912,
      "step": 85660
    },
    {
      "epoch": 9.424092409240924,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.01907417073685192,
      "loss": 0.2324,
      "num_input_tokens_seen": 18079936,
      "step": 85665
    },
    {
      "epoch": 9.424642464246425,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.019072784812872508,
      "loss": 0.2319,
      "num_input_tokens_seen": 18081024,
      "step": 85670
    },
    {
      "epoch": 9.425192519251926,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.019071398851356194,
      "loss": 0.2314,
      "num_input_tokens_seen": 18082048,
      "step": 85675
    },
    {
      "epoch": 9.425742574257425,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.019070012852315747,
      "loss": 0.2309,
      "num_input_tokens_seen": 18083104,
      "step": 85680
    },
    {
      "epoch": 9.426292629262926,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.019068626815763944,
      "loss": 0.2335,
      "num_input_tokens_seen": 18084160,
      "step": 85685
    },
    {
      "epoch": 9.426842684268427,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.01906724074171356,
      "loss": 0.2303,
      "num_input_tokens_seen": 18085248,
      "step": 85690
    },
    {
      "epoch": 9.427392739273927,
      "grad_norm": 0.000797271728515625,
      "learning_rate": 0.019065854630177372,
      "loss": 0.2308,
      "num_input_tokens_seen": 18086304,
      "step": 85695
    },
    {
      "epoch": 9.427942794279428,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.01906446848116815,
      "loss": 0.2303,
      "num_input_tokens_seen": 18087328,
      "step": 85700
    },
    {
      "epoch": 9.428492849284929,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.019063082294698674,
      "loss": 0.2309,
      "num_input_tokens_seen": 18088416,
      "step": 85705
    },
    {
      "epoch": 9.429042904290428,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.01906169607078172,
      "loss": 0.2319,
      "num_input_tokens_seen": 18089536,
      "step": 85710
    },
    {
      "epoch": 9.42959295929593,
      "grad_norm": 0.0006866455078125,
      "learning_rate": 0.019060309809430062,
      "loss": 0.2314,
      "num_input_tokens_seen": 18090592,
      "step": 85715
    },
    {
      "epoch": 9.43014301430143,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.019058923510656474,
      "loss": 0.2304,
      "num_input_tokens_seen": 18091616,
      "step": 85720
    },
    {
      "epoch": 9.430693069306932,
      "grad_norm": 0.000408172607421875,
      "learning_rate": 0.019057537174473733,
      "loss": 0.2308,
      "num_input_tokens_seen": 18092704,
      "step": 85725
    },
    {
      "epoch": 9.43124312431243,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.019056150800894622,
      "loss": 0.2309,
      "num_input_tokens_seen": 18093760,
      "step": 85730
    },
    {
      "epoch": 9.431793179317932,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.019054764389931922,
      "loss": 0.2314,
      "num_input_tokens_seen": 18094784,
      "step": 85735
    },
    {
      "epoch": 9.432343234323433,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.019053377941598396,
      "loss": 0.235,
      "num_input_tokens_seen": 18095904,
      "step": 85740
    },
    {
      "epoch": 9.432893289328932,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.019051991455906837,
      "loss": 0.2303,
      "num_input_tokens_seen": 18096960,
      "step": 85745
    },
    {
      "epoch": 9.433443344334433,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.019050604932870013,
      "loss": 0.2308,
      "num_input_tokens_seen": 18098048,
      "step": 85750
    },
    {
      "epoch": 9.433993399339935,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.0190492183725007,
      "loss": 0.2351,
      "num_input_tokens_seen": 18099136,
      "step": 85755
    },
    {
      "epoch": 9.434543454345434,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.0190478317748117,
      "loss": 0.2303,
      "num_input_tokens_seen": 18100192,
      "step": 85760
    },
    {
      "epoch": 9.435093509350935,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.01904644513981577,
      "loss": 0.2329,
      "num_input_tokens_seen": 18101248,
      "step": 85765
    },
    {
      "epoch": 9.435643564356436,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.019045058467525692,
      "loss": 0.2335,
      "num_input_tokens_seen": 18102336,
      "step": 85770
    },
    {
      "epoch": 9.436193619361935,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.019043671757954253,
      "loss": 0.2309,
      "num_input_tokens_seen": 18103392,
      "step": 85775
    },
    {
      "epoch": 9.436743674367436,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.019042285011114234,
      "loss": 0.2309,
      "num_input_tokens_seen": 18104480,
      "step": 85780
    },
    {
      "epoch": 9.437293729372938,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01904089822701841,
      "loss": 0.2324,
      "num_input_tokens_seen": 18105568,
      "step": 85785
    },
    {
      "epoch": 9.437843784378439,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.019039511405679572,
      "loss": 0.2308,
      "num_input_tokens_seen": 18106624,
      "step": 85790
    },
    {
      "epoch": 9.438393839383938,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.019038124547110492,
      "loss": 0.2314,
      "num_input_tokens_seen": 18107616,
      "step": 85795
    },
    {
      "epoch": 9.438943894389439,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.019036737651323955,
      "loss": 0.2329,
      "num_input_tokens_seen": 18108640,
      "step": 85800
    },
    {
      "epoch": 9.43949394939494,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01903535071833275,
      "loss": 0.2324,
      "num_input_tokens_seen": 18109792,
      "step": 85805
    },
    {
      "epoch": 9.44004400440044,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.019033963748149647,
      "loss": 0.2288,
      "num_input_tokens_seen": 18110784,
      "step": 85810
    },
    {
      "epoch": 9.44059405940594,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.019032576740787437,
      "loss": 0.2304,
      "num_input_tokens_seen": 18111872,
      "step": 85815
    },
    {
      "epoch": 9.441144114411442,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.0190311896962589,
      "loss": 0.2283,
      "num_input_tokens_seen": 18112896,
      "step": 85820
    },
    {
      "epoch": 9.441694169416941,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01902980261457683,
      "loss": 0.2325,
      "num_input_tokens_seen": 18113952,
      "step": 85825
    },
    {
      "epoch": 9.442244224422442,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.019028415495753997,
      "loss": 0.233,
      "num_input_tokens_seen": 18115040,
      "step": 85830
    },
    {
      "epoch": 9.442794279427943,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.019027028339803194,
      "loss": 0.2247,
      "num_input_tokens_seen": 18116096,
      "step": 85835
    },
    {
      "epoch": 9.443344334433444,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.0190256411467372,
      "loss": 0.2294,
      "num_input_tokens_seen": 18117152,
      "step": 85840
    },
    {
      "epoch": 9.443894389438944,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.019024253916568805,
      "loss": 0.2294,
      "num_input_tokens_seen": 18118144,
      "step": 85845
    },
    {
      "epoch": 9.444444444444445,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.019022866649310792,
      "loss": 0.2309,
      "num_input_tokens_seen": 18119168,
      "step": 85850
    },
    {
      "epoch": 9.444994499449946,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.019021479344975948,
      "loss": 0.2315,
      "num_input_tokens_seen": 18120256,
      "step": 85855
    },
    {
      "epoch": 9.445544554455445,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.01902009200357706,
      "loss": 0.2393,
      "num_input_tokens_seen": 18121280,
      "step": 85860
    },
    {
      "epoch": 9.446094609460946,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.01901870462512691,
      "loss": 0.2346,
      "num_input_tokens_seen": 18122400,
      "step": 85865
    },
    {
      "epoch": 9.446644664466447,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.019017317209638285,
      "loss": 0.233,
      "num_input_tokens_seen": 18123424,
      "step": 85870
    },
    {
      "epoch": 9.447194719471947,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.019015929757123982,
      "loss": 0.2283,
      "num_input_tokens_seen": 18124512,
      "step": 85875
    },
    {
      "epoch": 9.447744774477448,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.019014542267596784,
      "loss": 0.2293,
      "num_input_tokens_seen": 18125568,
      "step": 85880
    },
    {
      "epoch": 9.448294829482949,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.019013154741069467,
      "loss": 0.2319,
      "num_input_tokens_seen": 18126656,
      "step": 85885
    },
    {
      "epoch": 9.448844884488448,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.019011767177554834,
      "loss": 0.2304,
      "num_input_tokens_seen": 18127744,
      "step": 85890
    },
    {
      "epoch": 9.44939493949395,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.019010379577065667,
      "loss": 0.2304,
      "num_input_tokens_seen": 18128800,
      "step": 85895
    },
    {
      "epoch": 9.44994499449945,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.019008991939614753,
      "loss": 0.2351,
      "num_input_tokens_seen": 18129888,
      "step": 85900
    },
    {
      "epoch": 9.450495049504951,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.01900760426521489,
      "loss": 0.233,
      "num_input_tokens_seen": 18130944,
      "step": 85905
    },
    {
      "epoch": 9.45104510451045,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.019006216553878855,
      "loss": 0.2303,
      "num_input_tokens_seen": 18132000,
      "step": 85910
    },
    {
      "epoch": 9.451595159515952,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.019004828805619448,
      "loss": 0.233,
      "num_input_tokens_seen": 18133088,
      "step": 85915
    },
    {
      "epoch": 9.452145214521453,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.019003441020449454,
      "loss": 0.2314,
      "num_input_tokens_seen": 18134144,
      "step": 85920
    },
    {
      "epoch": 9.452695269526952,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.01900205319838167,
      "loss": 0.2309,
      "num_input_tokens_seen": 18135232,
      "step": 85925
    },
    {
      "epoch": 9.453245324532453,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.019000665339428877,
      "loss": 0.2309,
      "num_input_tokens_seen": 18136256,
      "step": 85930
    },
    {
      "epoch": 9.453795379537954,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.01899927744360387,
      "loss": 0.2309,
      "num_input_tokens_seen": 18137312,
      "step": 85935
    },
    {
      "epoch": 9.454345434543454,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.018997889510919446,
      "loss": 0.2324,
      "num_input_tokens_seen": 18138400,
      "step": 85940
    },
    {
      "epoch": 9.454895489548955,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.018996501541388384,
      "loss": 0.2309,
      "num_input_tokens_seen": 18139456,
      "step": 85945
    },
    {
      "epoch": 9.455445544554456,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.0189951135350235,
      "loss": 0.2314,
      "num_input_tokens_seen": 18140544,
      "step": 85950
    },
    {
      "epoch": 9.455995599559955,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.018993725491837563,
      "loss": 0.2324,
      "num_input_tokens_seen": 18141632,
      "step": 85955
    },
    {
      "epoch": 9.456545654565456,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01899233741184337,
      "loss": 0.2293,
      "num_input_tokens_seen": 18142656,
      "step": 85960
    },
    {
      "epoch": 9.457095709570957,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.018990949295053728,
      "loss": 0.2314,
      "num_input_tokens_seen": 18143776,
      "step": 85965
    },
    {
      "epoch": 9.457645764576458,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.01898956114148142,
      "loss": 0.2319,
      "num_input_tokens_seen": 18144864,
      "step": 85970
    },
    {
      "epoch": 9.458195819581958,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.018988172951139238,
      "loss": 0.2304,
      "num_input_tokens_seen": 18145856,
      "step": 85975
    },
    {
      "epoch": 9.458745874587459,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.018986784724039977,
      "loss": 0.2335,
      "num_input_tokens_seen": 18146848,
      "step": 85980
    },
    {
      "epoch": 9.45929592959296,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.018985396460196438,
      "loss": 0.2303,
      "num_input_tokens_seen": 18147904,
      "step": 85985
    },
    {
      "epoch": 9.45984598459846,
      "grad_norm": 0.009765625,
      "learning_rate": 0.018984008159621413,
      "loss": 0.2319,
      "num_input_tokens_seen": 18148928,
      "step": 85990
    },
    {
      "epoch": 9.46039603960396,
      "grad_norm": 0.0008392333984375,
      "learning_rate": 0.018982619822327698,
      "loss": 0.2319,
      "num_input_tokens_seen": 18149952,
      "step": 85995
    },
    {
      "epoch": 9.460946094609461,
      "grad_norm": 0.000888824462890625,
      "learning_rate": 0.01898123144832808,
      "loss": 0.2319,
      "num_input_tokens_seen": 18151008,
      "step": 86000
    },
    {
      "epoch": 9.46149614961496,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.018979843037635367,
      "loss": 0.2319,
      "num_input_tokens_seen": 18152096,
      "step": 86005
    },
    {
      "epoch": 9.462046204620462,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.01897845459026235,
      "loss": 0.2309,
      "num_input_tokens_seen": 18153184,
      "step": 86010
    },
    {
      "epoch": 9.462596259625963,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.018977066106221824,
      "loss": 0.2319,
      "num_input_tokens_seen": 18154208,
      "step": 86015
    },
    {
      "epoch": 9.463146314631462,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.018975677585526592,
      "loss": 0.2303,
      "num_input_tokens_seen": 18155264,
      "step": 86020
    },
    {
      "epoch": 9.463696369636963,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.018974289028189446,
      "loss": 0.233,
      "num_input_tokens_seen": 18156288,
      "step": 86025
    },
    {
      "epoch": 9.464246424642464,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.018972900434223183,
      "loss": 0.2309,
      "num_input_tokens_seen": 18157344,
      "step": 86030
    },
    {
      "epoch": 9.464796479647966,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.018971511803640602,
      "loss": 0.2309,
      "num_input_tokens_seen": 18158368,
      "step": 86035
    },
    {
      "epoch": 9.465346534653465,
      "grad_norm": 0.000637054443359375,
      "learning_rate": 0.01897012313645451,
      "loss": 0.2303,
      "num_input_tokens_seen": 18159456,
      "step": 86040
    },
    {
      "epoch": 9.465896589658966,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.018968734432677693,
      "loss": 0.2319,
      "num_input_tokens_seen": 18160480,
      "step": 86045
    },
    {
      "epoch": 9.466446644664467,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.018967345692322952,
      "loss": 0.2298,
      "num_input_tokens_seen": 18161568,
      "step": 86050
    },
    {
      "epoch": 9.466996699669966,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.018965956915403096,
      "loss": 0.2319,
      "num_input_tokens_seen": 18162560,
      "step": 86055
    },
    {
      "epoch": 9.467546754675467,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.01896456810193091,
      "loss": 0.2283,
      "num_input_tokens_seen": 18163680,
      "step": 86060
    },
    {
      "epoch": 9.468096809680969,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.01896317925191921,
      "loss": 0.2303,
      "num_input_tokens_seen": 18164704,
      "step": 86065
    },
    {
      "epoch": 9.468646864686468,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.018961790365380787,
      "loss": 0.2309,
      "num_input_tokens_seen": 18165760,
      "step": 86070
    },
    {
      "epoch": 9.469196919691969,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.01896040144232844,
      "loss": 0.2309,
      "num_input_tokens_seen": 18166880,
      "step": 86075
    },
    {
      "epoch": 9.46974697469747,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.018959012482774977,
      "loss": 0.2319,
      "num_input_tokens_seen": 18167936,
      "step": 86080
    },
    {
      "epoch": 9.47029702970297,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.018957623486733192,
      "loss": 0.2319,
      "num_input_tokens_seen": 18168896,
      "step": 86085
    },
    {
      "epoch": 9.47084708470847,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.018956234454215893,
      "loss": 0.2304,
      "num_input_tokens_seen": 18169984,
      "step": 86090
    },
    {
      "epoch": 9.471397139713972,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01895484538523588,
      "loss": 0.233,
      "num_input_tokens_seen": 18171072,
      "step": 86095
    },
    {
      "epoch": 9.471947194719473,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.018953456279805957,
      "loss": 0.2308,
      "num_input_tokens_seen": 18172064,
      "step": 86100
    },
    {
      "epoch": 9.472497249724972,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.018952067137938917,
      "loss": 0.2303,
      "num_input_tokens_seen": 18173152,
      "step": 86105
    },
    {
      "epoch": 9.473047304730473,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.018950677959647572,
      "loss": 0.2298,
      "num_input_tokens_seen": 18174208,
      "step": 86110
    },
    {
      "epoch": 9.473597359735974,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.01894928874494473,
      "loss": 0.2293,
      "num_input_tokens_seen": 18175264,
      "step": 86115
    },
    {
      "epoch": 9.474147414741473,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.018947899493843182,
      "loss": 0.2314,
      "num_input_tokens_seen": 18176256,
      "step": 86120
    },
    {
      "epoch": 9.474697469746975,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.01894651020635574,
      "loss": 0.2298,
      "num_input_tokens_seen": 18177344,
      "step": 86125
    },
    {
      "epoch": 9.475247524752476,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.01894512088249521,
      "loss": 0.2288,
      "num_input_tokens_seen": 18178400,
      "step": 86130
    },
    {
      "epoch": 9.475797579757975,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.018943731522274392,
      "loss": 0.2299,
      "num_input_tokens_seen": 18179424,
      "step": 86135
    },
    {
      "epoch": 9.476347634763476,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.018942342125706096,
      "loss": 0.2325,
      "num_input_tokens_seen": 18180512,
      "step": 86140
    },
    {
      "epoch": 9.476897689768977,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.01894095269280312,
      "loss": 0.233,
      "num_input_tokens_seen": 18181600,
      "step": 86145
    },
    {
      "epoch": 9.477447744774478,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.018939563223578277,
      "loss": 0.2314,
      "num_input_tokens_seen": 18182688,
      "step": 86150
    },
    {
      "epoch": 9.477997799779978,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.018938173718044373,
      "loss": 0.2309,
      "num_input_tokens_seen": 18183680,
      "step": 86155
    },
    {
      "epoch": 9.478547854785479,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.018936784176214208,
      "loss": 0.2278,
      "num_input_tokens_seen": 18184832,
      "step": 86160
    },
    {
      "epoch": 9.47909790979098,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.018935394598100595,
      "loss": 0.2303,
      "num_input_tokens_seen": 18185856,
      "step": 86165
    },
    {
      "epoch": 9.479647964796479,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.01893400498371633,
      "loss": 0.2346,
      "num_input_tokens_seen": 18186944,
      "step": 86170
    },
    {
      "epoch": 9.48019801980198,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.018932615333074235,
      "loss": 0.2299,
      "num_input_tokens_seen": 18188032,
      "step": 86175
    },
    {
      "epoch": 9.480748074807481,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01893122564618711,
      "loss": 0.234,
      "num_input_tokens_seen": 18189056,
      "step": 86180
    },
    {
      "epoch": 9.48129812981298,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01892983592306777,
      "loss": 0.233,
      "num_input_tokens_seen": 18190176,
      "step": 86185
    },
    {
      "epoch": 9.481848184818482,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01892844616372901,
      "loss": 0.2314,
      "num_input_tokens_seen": 18191168,
      "step": 86190
    },
    {
      "epoch": 9.482398239823983,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.018927056368183655,
      "loss": 0.2324,
      "num_input_tokens_seen": 18192192,
      "step": 86195
    },
    {
      "epoch": 9.482948294829482,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.018925666536444497,
      "loss": 0.2303,
      "num_input_tokens_seen": 18193280,
      "step": 86200
    },
    {
      "epoch": 9.483498349834983,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.018924276668524358,
      "loss": 0.2293,
      "num_input_tokens_seen": 18194400,
      "step": 86205
    },
    {
      "epoch": 9.484048404840484,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01892288676443605,
      "loss": 0.2303,
      "num_input_tokens_seen": 18195488,
      "step": 86210
    },
    {
      "epoch": 9.484598459845985,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.01892149682419237,
      "loss": 0.2319,
      "num_input_tokens_seen": 18196576,
      "step": 86215
    },
    {
      "epoch": 9.485148514851485,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.018920106847806133,
      "loss": 0.2325,
      "num_input_tokens_seen": 18197600,
      "step": 86220
    },
    {
      "epoch": 9.485698569856986,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.01891871683529016,
      "loss": 0.2288,
      "num_input_tokens_seen": 18198720,
      "step": 86225
    },
    {
      "epoch": 9.486248624862487,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.018917326786657248,
      "loss": 0.2341,
      "num_input_tokens_seen": 18199840,
      "step": 86230
    },
    {
      "epoch": 9.486798679867986,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.01891593670192022,
      "loss": 0.2319,
      "num_input_tokens_seen": 18200864,
      "step": 86235
    },
    {
      "epoch": 9.487348734873487,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.018914546581091877,
      "loss": 0.2314,
      "num_input_tokens_seen": 18201984,
      "step": 86240
    },
    {
      "epoch": 9.487898789878988,
      "grad_norm": 0.009521484375,
      "learning_rate": 0.01891315642418504,
      "loss": 0.2309,
      "num_input_tokens_seen": 18203008,
      "step": 86245
    },
    {
      "epoch": 9.488448844884488,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.018911766231212518,
      "loss": 0.2299,
      "num_input_tokens_seen": 18204064,
      "step": 86250
    },
    {
      "epoch": 9.488998899889989,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01891037600218712,
      "loss": 0.2324,
      "num_input_tokens_seen": 18205152,
      "step": 86255
    },
    {
      "epoch": 9.48954895489549,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01890898573712167,
      "loss": 0.2351,
      "num_input_tokens_seen": 18206208,
      "step": 86260
    },
    {
      "epoch": 9.490099009900991,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.01890759543602897,
      "loss": 0.2324,
      "num_input_tokens_seen": 18207296,
      "step": 86265
    },
    {
      "epoch": 9.49064906490649,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.01890620509892184,
      "loss": 0.2325,
      "num_input_tokens_seen": 18208352,
      "step": 86270
    },
    {
      "epoch": 9.491199119911991,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.018904814725813086,
      "loss": 0.2298,
      "num_input_tokens_seen": 18209376,
      "step": 86275
    },
    {
      "epoch": 9.491749174917492,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.018903424316715534,
      "loss": 0.2319,
      "num_input_tokens_seen": 18210560,
      "step": 86280
    },
    {
      "epoch": 9.492299229922992,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.018902033871641993,
      "loss": 0.2325,
      "num_input_tokens_seen": 18211616,
      "step": 86285
    },
    {
      "epoch": 9.492849284928493,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.018900643390605276,
      "loss": 0.2319,
      "num_input_tokens_seen": 18212608,
      "step": 86290
    },
    {
      "epoch": 9.493399339933994,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.018899252873618197,
      "loss": 0.2303,
      "num_input_tokens_seen": 18213664,
      "step": 86295
    },
    {
      "epoch": 9.493949394939493,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01889786232069358,
      "loss": 0.2304,
      "num_input_tokens_seen": 18214752,
      "step": 86300
    },
    {
      "epoch": 9.494499449944994,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.018896471731844236,
      "loss": 0.2325,
      "num_input_tokens_seen": 18215872,
      "step": 86305
    },
    {
      "epoch": 9.495049504950495,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.018895081107082982,
      "loss": 0.2299,
      "num_input_tokens_seen": 18216896,
      "step": 86310
    },
    {
      "epoch": 9.495599559955995,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.018893690446422636,
      "loss": 0.2304,
      "num_input_tokens_seen": 18217920,
      "step": 86315
    },
    {
      "epoch": 9.496149614961496,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01889229974987601,
      "loss": 0.2309,
      "num_input_tokens_seen": 18219040,
      "step": 86320
    },
    {
      "epoch": 9.496699669966997,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.018890909017455926,
      "loss": 0.2325,
      "num_input_tokens_seen": 18220096,
      "step": 86325
    },
    {
      "epoch": 9.497249724972498,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.018889518249175206,
      "loss": 0.2356,
      "num_input_tokens_seen": 18221120,
      "step": 86330
    },
    {
      "epoch": 9.497799779977997,
      "grad_norm": 0.000583648681640625,
      "learning_rate": 0.018888127445046653,
      "loss": 0.2335,
      "num_input_tokens_seen": 18222144,
      "step": 86335
    },
    {
      "epoch": 9.498349834983498,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.0188867366050831,
      "loss": 0.2335,
      "num_input_tokens_seen": 18223168,
      "step": 86340
    },
    {
      "epoch": 9.498899889989,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.018885345729297363,
      "loss": 0.233,
      "num_input_tokens_seen": 18224224,
      "step": 86345
    },
    {
      "epoch": 9.499449944994499,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.018883954817702255,
      "loss": 0.2319,
      "num_input_tokens_seen": 18225280,
      "step": 86350
    },
    {
      "epoch": 9.5,
      "grad_norm": 0.009765625,
      "learning_rate": 0.018882563870310603,
      "loss": 0.2303,
      "num_input_tokens_seen": 18226336,
      "step": 86355
    },
    {
      "epoch": 9.500550055005501,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01888117288713522,
      "loss": 0.233,
      "num_input_tokens_seen": 18227456,
      "step": 86360
    },
    {
      "epoch": 9.501100110011,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.018879781868188925,
      "loss": 0.2309,
      "num_input_tokens_seen": 18228576,
      "step": 86365
    },
    {
      "epoch": 9.501650165016502,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.018878390813484545,
      "loss": 0.2314,
      "num_input_tokens_seen": 18229600,
      "step": 86370
    },
    {
      "epoch": 9.502200220022003,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.018876999723034905,
      "loss": 0.2308,
      "num_input_tokens_seen": 18230688,
      "step": 86375
    },
    {
      "epoch": 9.502750275027502,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.01887560859685281,
      "loss": 0.2325,
      "num_input_tokens_seen": 18231744,
      "step": 86380
    },
    {
      "epoch": 9.503300330033003,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.018874217434951095,
      "loss": 0.2335,
      "num_input_tokens_seen": 18232832,
      "step": 86385
    },
    {
      "epoch": 9.503850385038504,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.018872826237342575,
      "loss": 0.2309,
      "num_input_tokens_seen": 18233888,
      "step": 86390
    },
    {
      "epoch": 9.504400440044005,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.018871435004040073,
      "loss": 0.2309,
      "num_input_tokens_seen": 18234912,
      "step": 86395
    },
    {
      "epoch": 9.504950495049505,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.018870043735056415,
      "loss": 0.2293,
      "num_input_tokens_seen": 18235968,
      "step": 86400
    },
    {
      "epoch": 9.505500550055006,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.01886865243040442,
      "loss": 0.2309,
      "num_input_tokens_seen": 18236992,
      "step": 86405
    },
    {
      "epoch": 9.506050605060507,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.01886726109009691,
      "loss": 0.2314,
      "num_input_tokens_seen": 18238048,
      "step": 86410
    },
    {
      "epoch": 9.506600660066006,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01886586971414671,
      "loss": 0.2314,
      "num_input_tokens_seen": 18239136,
      "step": 86415
    },
    {
      "epoch": 9.507150715071507,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.018864478302566646,
      "loss": 0.2319,
      "num_input_tokens_seen": 18240224,
      "step": 86420
    },
    {
      "epoch": 9.507700770077008,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.018863086855369544,
      "loss": 0.2303,
      "num_input_tokens_seen": 18241248,
      "step": 86425
    },
    {
      "epoch": 9.508250825082508,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.018861695372568217,
      "loss": 0.2319,
      "num_input_tokens_seen": 18242272,
      "step": 86430
    },
    {
      "epoch": 9.508800880088009,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.0188603038541755,
      "loss": 0.2314,
      "num_input_tokens_seen": 18243360,
      "step": 86435
    },
    {
      "epoch": 9.50935093509351,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01885891230020421,
      "loss": 0.2309,
      "num_input_tokens_seen": 18244448,
      "step": 86440
    },
    {
      "epoch": 9.509900990099009,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01885752071066718,
      "loss": 0.2309,
      "num_input_tokens_seen": 18245568,
      "step": 86445
    },
    {
      "epoch": 9.51045104510451,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.018856129085577236,
      "loss": 0.2335,
      "num_input_tokens_seen": 18246688,
      "step": 86450
    },
    {
      "epoch": 9.511001100110011,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.0188547374249472,
      "loss": 0.233,
      "num_input_tokens_seen": 18247744,
      "step": 86455
    },
    {
      "epoch": 9.511551155115512,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.018853345728789897,
      "loss": 0.2304,
      "num_input_tokens_seen": 18248800,
      "step": 86460
    },
    {
      "epoch": 9.512101210121012,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.018851953997118157,
      "loss": 0.2303,
      "num_input_tokens_seen": 18249792,
      "step": 86465
    },
    {
      "epoch": 9.512651265126513,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0188505622299448,
      "loss": 0.2314,
      "num_input_tokens_seen": 18250816,
      "step": 86470
    },
    {
      "epoch": 9.513201320132014,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.01884917042728267,
      "loss": 0.2314,
      "num_input_tokens_seen": 18251904,
      "step": 86475
    },
    {
      "epoch": 9.513751375137513,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.01884777858914457,
      "loss": 0.2324,
      "num_input_tokens_seen": 18252896,
      "step": 86480
    },
    {
      "epoch": 9.514301430143014,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.018846386715543346,
      "loss": 0.2314,
      "num_input_tokens_seen": 18254016,
      "step": 86485
    },
    {
      "epoch": 9.514851485148515,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01884499480649182,
      "loss": 0.2319,
      "num_input_tokens_seen": 18255072,
      "step": 86490
    },
    {
      "epoch": 9.515401540154015,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.018843602862002828,
      "loss": 0.2314,
      "num_input_tokens_seen": 18256128,
      "step": 86495
    },
    {
      "epoch": 9.515951595159516,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.018842210882089188,
      "loss": 0.2298,
      "num_input_tokens_seen": 18257184,
      "step": 86500
    },
    {
      "epoch": 9.516501650165017,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.018840818866763736,
      "loss": 0.2325,
      "num_input_tokens_seen": 18258144,
      "step": 86505
    },
    {
      "epoch": 9.517051705170516,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.018839426816039295,
      "loss": 0.2324,
      "num_input_tokens_seen": 18259232,
      "step": 86510
    },
    {
      "epoch": 9.517601760176017,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.0188380347299287,
      "loss": 0.2298,
      "num_input_tokens_seen": 18260256,
      "step": 86515
    },
    {
      "epoch": 9.518151815181518,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01883664260844479,
      "loss": 0.233,
      "num_input_tokens_seen": 18261312,
      "step": 86520
    },
    {
      "epoch": 9.51870187018702,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.018835250451600376,
      "loss": 0.2309,
      "num_input_tokens_seen": 18262400,
      "step": 86525
    },
    {
      "epoch": 9.519251925192519,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.018833858259408302,
      "loss": 0.2309,
      "num_input_tokens_seen": 18263424,
      "step": 86530
    },
    {
      "epoch": 9.51980198019802,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.018832466031881396,
      "loss": 0.2309,
      "num_input_tokens_seen": 18264512,
      "step": 86535
    },
    {
      "epoch": 9.520352035203521,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01883107376903249,
      "loss": 0.2324,
      "num_input_tokens_seen": 18265568,
      "step": 86540
    },
    {
      "epoch": 9.52090209020902,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.018829681470874417,
      "loss": 0.2293,
      "num_input_tokens_seen": 18266688,
      "step": 86545
    },
    {
      "epoch": 9.521452145214521,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.018828289137420007,
      "loss": 0.2293,
      "num_input_tokens_seen": 18267808,
      "step": 86550
    },
    {
      "epoch": 9.522002200220022,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01882689676868209,
      "loss": 0.2303,
      "num_input_tokens_seen": 18268832,
      "step": 86555
    },
    {
      "epoch": 9.522552255225522,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.018825504364673503,
      "loss": 0.2309,
      "num_input_tokens_seen": 18269888,
      "step": 86560
    },
    {
      "epoch": 9.523102310231023,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.018824111925407082,
      "loss": 0.2314,
      "num_input_tokens_seen": 18271008,
      "step": 86565
    },
    {
      "epoch": 9.523652365236524,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.01882271945089565,
      "loss": 0.2298,
      "num_input_tokens_seen": 18272064,
      "step": 86570
    },
    {
      "epoch": 9.524202420242025,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.018821326941152048,
      "loss": 0.2309,
      "num_input_tokens_seen": 18273056,
      "step": 86575
    },
    {
      "epoch": 9.524752475247524,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.018819934396189116,
      "loss": 0.2324,
      "num_input_tokens_seen": 18274048,
      "step": 86580
    },
    {
      "epoch": 9.525302530253025,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.018818541816019673,
      "loss": 0.2319,
      "num_input_tokens_seen": 18275104,
      "step": 86585
    },
    {
      "epoch": 9.525852585258527,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01881714920065657,
      "loss": 0.2309,
      "num_input_tokens_seen": 18276224,
      "step": 86590
    },
    {
      "epoch": 9.526402640264026,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01881575655011263,
      "loss": 0.2319,
      "num_input_tokens_seen": 18277248,
      "step": 86595
    },
    {
      "epoch": 9.526952695269527,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.018814363864400687,
      "loss": 0.2329,
      "num_input_tokens_seen": 18278240,
      "step": 86600
    },
    {
      "epoch": 9.527502750275028,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.018812971143533593,
      "loss": 0.2319,
      "num_input_tokens_seen": 18279264,
      "step": 86605
    },
    {
      "epoch": 9.528052805280527,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.01881157838752417,
      "loss": 0.2314,
      "num_input_tokens_seen": 18280320,
      "step": 86610
    },
    {
      "epoch": 9.528602860286028,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01881018559638526,
      "loss": 0.2313,
      "num_input_tokens_seen": 18281408,
      "step": 86615
    },
    {
      "epoch": 9.52915291529153,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.018808792770129697,
      "loss": 0.2329,
      "num_input_tokens_seen": 18282432,
      "step": 86620
    },
    {
      "epoch": 9.52970297029703,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.018807399908770315,
      "loss": 0.2308,
      "num_input_tokens_seen": 18283520,
      "step": 86625
    },
    {
      "epoch": 9.53025302530253,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.018806007012319954,
      "loss": 0.2314,
      "num_input_tokens_seen": 18284512,
      "step": 86630
    },
    {
      "epoch": 9.530803080308031,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.018804614080791458,
      "loss": 0.2298,
      "num_input_tokens_seen": 18285600,
      "step": 86635
    },
    {
      "epoch": 9.531353135313532,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.018803221114197657,
      "loss": 0.2304,
      "num_input_tokens_seen": 18286688,
      "step": 86640
    },
    {
      "epoch": 9.531903190319031,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.01880182811255139,
      "loss": 0.2303,
      "num_input_tokens_seen": 18287712,
      "step": 86645
    },
    {
      "epoch": 9.532453245324533,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.018800435075865502,
      "loss": 0.2314,
      "num_input_tokens_seen": 18288800,
      "step": 86650
    },
    {
      "epoch": 9.533003300330034,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.018799042004152822,
      "loss": 0.2308,
      "num_input_tokens_seen": 18289888,
      "step": 86655
    },
    {
      "epoch": 9.533553355335533,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.018797648897426203,
      "loss": 0.2309,
      "num_input_tokens_seen": 18290976,
      "step": 86660
    },
    {
      "epoch": 9.534103410341034,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.018796255755698474,
      "loss": 0.2314,
      "num_input_tokens_seen": 18292064,
      "step": 86665
    },
    {
      "epoch": 9.534653465346535,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.01879486257898247,
      "loss": 0.2298,
      "num_input_tokens_seen": 18293120,
      "step": 86670
    },
    {
      "epoch": 9.535203520352034,
      "grad_norm": 0.00069427490234375,
      "learning_rate": 0.018793469367291045,
      "loss": 0.2319,
      "num_input_tokens_seen": 18294144,
      "step": 86675
    },
    {
      "epoch": 9.535753575357536,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.018792076120637033,
      "loss": 0.2324,
      "num_input_tokens_seen": 18295200,
      "step": 86680
    },
    {
      "epoch": 9.536303630363037,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.018790682839033273,
      "loss": 0.2303,
      "num_input_tokens_seen": 18296352,
      "step": 86685
    },
    {
      "epoch": 9.536853685368538,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.01878928952249261,
      "loss": 0.2314,
      "num_input_tokens_seen": 18297344,
      "step": 86690
    },
    {
      "epoch": 9.537403740374037,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01878789617102788,
      "loss": 0.2314,
      "num_input_tokens_seen": 18298432,
      "step": 86695
    },
    {
      "epoch": 9.537953795379538,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01878650278465193,
      "loss": 0.2314,
      "num_input_tokens_seen": 18299520,
      "step": 86700
    },
    {
      "epoch": 9.53850385038504,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.018785109363377604,
      "loss": 0.2319,
      "num_input_tokens_seen": 18300672,
      "step": 86705
    },
    {
      "epoch": 9.539053905390539,
      "grad_norm": 0.000537872314453125,
      "learning_rate": 0.018783715907217745,
      "loss": 0.234,
      "num_input_tokens_seen": 18301728,
      "step": 86710
    },
    {
      "epoch": 9.53960396039604,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01878232241618518,
      "loss": 0.2329,
      "num_input_tokens_seen": 18302816,
      "step": 86715
    },
    {
      "epoch": 9.54015401540154,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.018780928890292772,
      "loss": 0.2319,
      "num_input_tokens_seen": 18303872,
      "step": 86720
    },
    {
      "epoch": 9.54070407040704,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.018779535329553357,
      "loss": 0.2319,
      "num_input_tokens_seen": 18304896,
      "step": 86725
    },
    {
      "epoch": 9.541254125412541,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.01877814173397977,
      "loss": 0.2283,
      "num_input_tokens_seen": 18305984,
      "step": 86730
    },
    {
      "epoch": 9.541804180418042,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.018776748103584876,
      "loss": 0.2329,
      "num_input_tokens_seen": 18307008,
      "step": 86735
    },
    {
      "epoch": 9.542354235423542,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.018775354438381502,
      "loss": 0.2304,
      "num_input_tokens_seen": 18308064,
      "step": 86740
    },
    {
      "epoch": 9.542904290429043,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.018773960738382493,
      "loss": 0.2309,
      "num_input_tokens_seen": 18309120,
      "step": 86745
    },
    {
      "epoch": 9.543454345434544,
      "grad_norm": 0.009765625,
      "learning_rate": 0.018772567003600703,
      "loss": 0.2319,
      "num_input_tokens_seen": 18310112,
      "step": 86750
    },
    {
      "epoch": 9.544004400440045,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.018771173234048977,
      "loss": 0.2309,
      "num_input_tokens_seen": 18311264,
      "step": 86755
    },
    {
      "epoch": 9.544554455445544,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.018769779429740154,
      "loss": 0.2304,
      "num_input_tokens_seen": 18312384,
      "step": 86760
    },
    {
      "epoch": 9.545104510451045,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01876838559068708,
      "loss": 0.2335,
      "num_input_tokens_seen": 18313440,
      "step": 86765
    },
    {
      "epoch": 9.545654565456546,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.018766991716902607,
      "loss": 0.2314,
      "num_input_tokens_seen": 18314464,
      "step": 86770
    },
    {
      "epoch": 9.546204620462046,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.018765597808399575,
      "loss": 0.2309,
      "num_input_tokens_seen": 18315488,
      "step": 86775
    },
    {
      "epoch": 9.546754675467547,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01876420386519084,
      "loss": 0.2319,
      "num_input_tokens_seen": 18316512,
      "step": 86780
    },
    {
      "epoch": 9.547304730473048,
      "grad_norm": 0.00060272216796875,
      "learning_rate": 0.01876280988728924,
      "loss": 0.2329,
      "num_input_tokens_seen": 18317536,
      "step": 86785
    },
    {
      "epoch": 9.547854785478547,
      "grad_norm": 0.0026397705078125,
      "learning_rate": 0.01876141587470763,
      "loss": 0.2335,
      "num_input_tokens_seen": 18318560,
      "step": 86790
    },
    {
      "epoch": 9.548404840484048,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.018760021827458857,
      "loss": 0.2324,
      "num_input_tokens_seen": 18319616,
      "step": 86795
    },
    {
      "epoch": 9.54895489548955,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.018758627745555763,
      "loss": 0.2304,
      "num_input_tokens_seen": 18320704,
      "step": 86800
    },
    {
      "epoch": 9.549504950495049,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.018757233629011202,
      "loss": 0.2324,
      "num_input_tokens_seen": 18321728,
      "step": 86805
    },
    {
      "epoch": 9.55005500550055,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01875583947783802,
      "loss": 0.2314,
      "num_input_tokens_seen": 18322720,
      "step": 86810
    },
    {
      "epoch": 9.55060506050605,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.018754445292049067,
      "loss": 0.2314,
      "num_input_tokens_seen": 18323808,
      "step": 86815
    },
    {
      "epoch": 9.551155115511552,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.018753051071657197,
      "loss": 0.2309,
      "num_input_tokens_seen": 18324832,
      "step": 86820
    },
    {
      "epoch": 9.551705170517051,
      "grad_norm": 0.0005340576171875,
      "learning_rate": 0.01875165681667525,
      "loss": 0.2308,
      "num_input_tokens_seen": 18325888,
      "step": 86825
    },
    {
      "epoch": 9.552255225522552,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.018750262527116087,
      "loss": 0.2309,
      "num_input_tokens_seen": 18326912,
      "step": 86830
    },
    {
      "epoch": 9.552805280528053,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.018748868202992555,
      "loss": 0.2329,
      "num_input_tokens_seen": 18328064,
      "step": 86835
    },
    {
      "epoch": 9.553355335533553,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.018747473844317502,
      "loss": 0.2335,
      "num_input_tokens_seen": 18329056,
      "step": 86840
    },
    {
      "epoch": 9.553905390539054,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.018746079451103778,
      "loss": 0.2314,
      "num_input_tokens_seen": 18330080,
      "step": 86845
    },
    {
      "epoch": 9.554455445544555,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.018744685023364245,
      "loss": 0.2324,
      "num_input_tokens_seen": 18331168,
      "step": 86850
    },
    {
      "epoch": 9.555005500550054,
      "grad_norm": 0.009765625,
      "learning_rate": 0.018743290561111743,
      "loss": 0.2303,
      "num_input_tokens_seen": 18332160,
      "step": 86855
    },
    {
      "epoch": 9.555555555555555,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.018741896064359123,
      "loss": 0.2319,
      "num_input_tokens_seen": 18333184,
      "step": 86860
    },
    {
      "epoch": 9.556105610561056,
      "grad_norm": 0.0005035400390625,
      "learning_rate": 0.01874050153311925,
      "loss": 0.2303,
      "num_input_tokens_seen": 18334240,
      "step": 86865
    },
    {
      "epoch": 9.556655665566556,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.018739106967404964,
      "loss": 0.2314,
      "num_input_tokens_seen": 18335328,
      "step": 86870
    },
    {
      "epoch": 9.557205720572057,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.018737712367229126,
      "loss": 0.2314,
      "num_input_tokens_seen": 18336416,
      "step": 86875
    },
    {
      "epoch": 9.557755775577558,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.018736317732604587,
      "loss": 0.2303,
      "num_input_tokens_seen": 18337472,
      "step": 86880
    },
    {
      "epoch": 9.558305830583059,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.018734923063544198,
      "loss": 0.2309,
      "num_input_tokens_seen": 18338528,
      "step": 86885
    },
    {
      "epoch": 9.558855885588558,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.018733528360060814,
      "loss": 0.2314,
      "num_input_tokens_seen": 18339648,
      "step": 86890
    },
    {
      "epoch": 9.55940594059406,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.018732133622167295,
      "loss": 0.2309,
      "num_input_tokens_seen": 18340672,
      "step": 86895
    },
    {
      "epoch": 9.55995599559956,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01873073884987649,
      "loss": 0.2308,
      "num_input_tokens_seen": 18341664,
      "step": 86900
    },
    {
      "epoch": 9.56050605060506,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.018729344043201257,
      "loss": 0.2329,
      "num_input_tokens_seen": 18342720,
      "step": 86905
    },
    {
      "epoch": 9.561056105610561,
      "grad_norm": 0.009765625,
      "learning_rate": 0.018727949202154447,
      "loss": 0.2324,
      "num_input_tokens_seen": 18343808,
      "step": 86910
    },
    {
      "epoch": 9.561606160616062,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.01872655432674892,
      "loss": 0.2293,
      "num_input_tokens_seen": 18344832,
      "step": 86915
    },
    {
      "epoch": 9.562156215621561,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.01872515941699752,
      "loss": 0.2314,
      "num_input_tokens_seen": 18345952,
      "step": 86920
    },
    {
      "epoch": 9.562706270627062,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01872376447291313,
      "loss": 0.2303,
      "num_input_tokens_seen": 18347008,
      "step": 86925
    },
    {
      "epoch": 9.563256325632564,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.018722369494508576,
      "loss": 0.2319,
      "num_input_tokens_seen": 18348096,
      "step": 86930
    },
    {
      "epoch": 9.563806380638063,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.018720974481796732,
      "loss": 0.2324,
      "num_input_tokens_seen": 18349152,
      "step": 86935
    },
    {
      "epoch": 9.564356435643564,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.018719579434790454,
      "loss": 0.2319,
      "num_input_tokens_seen": 18350208,
      "step": 86940
    },
    {
      "epoch": 9.564906490649065,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.018718184353502598,
      "loss": 0.2319,
      "num_input_tokens_seen": 18351200,
      "step": 86945
    },
    {
      "epoch": 9.565456545654566,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.01871678923794602,
      "loss": 0.2324,
      "num_input_tokens_seen": 18352224,
      "step": 86950
    },
    {
      "epoch": 9.566006600660065,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.018715394088133577,
      "loss": 0.2324,
      "num_input_tokens_seen": 18353248,
      "step": 86955
    },
    {
      "epoch": 9.566556655665567,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.01871399890407813,
      "loss": 0.2335,
      "num_input_tokens_seen": 18354304,
      "step": 86960
    },
    {
      "epoch": 9.567106710671068,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.01871260368579254,
      "loss": 0.2324,
      "num_input_tokens_seen": 18355360,
      "step": 86965
    },
    {
      "epoch": 9.567656765676567,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01871120843328966,
      "loss": 0.2319,
      "num_input_tokens_seen": 18356448,
      "step": 86970
    },
    {
      "epoch": 9.568206820682068,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.018709813146582353,
      "loss": 0.2314,
      "num_input_tokens_seen": 18357504,
      "step": 86975
    },
    {
      "epoch": 9.56875687568757,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.018708417825683478,
      "loss": 0.2319,
      "num_input_tokens_seen": 18358592,
      "step": 86980
    },
    {
      "epoch": 9.569306930693068,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.0187070224706059,
      "loss": 0.2314,
      "num_input_tokens_seen": 18359616,
      "step": 86985
    },
    {
      "epoch": 9.56985698569857,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.018705627081362468,
      "loss": 0.2303,
      "num_input_tokens_seen": 18360640,
      "step": 86990
    },
    {
      "epoch": 9.57040704070407,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.01870423165796606,
      "loss": 0.2319,
      "num_input_tokens_seen": 18361728,
      "step": 86995
    },
    {
      "epoch": 9.570957095709572,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.018702836200429517,
      "loss": 0.2309,
      "num_input_tokens_seen": 18362720,
      "step": 87000
    },
    {
      "epoch": 9.571507150715071,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01870144070876571,
      "loss": 0.2309,
      "num_input_tokens_seen": 18363840,
      "step": 87005
    },
    {
      "epoch": 9.572057205720572,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.0187000451829875,
      "loss": 0.2304,
      "num_input_tokens_seen": 18364896,
      "step": 87010
    },
    {
      "epoch": 9.572607260726073,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01869864962310775,
      "loss": 0.2324,
      "num_input_tokens_seen": 18366016,
      "step": 87015
    },
    {
      "epoch": 9.573157315731573,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.018697254029139326,
      "loss": 0.2324,
      "num_input_tokens_seen": 18367008,
      "step": 87020
    },
    {
      "epoch": 9.573707370737074,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.01869585840109508,
      "loss": 0.2303,
      "num_input_tokens_seen": 18368064,
      "step": 87025
    },
    {
      "epoch": 9.574257425742575,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.018694462738987883,
      "loss": 0.2335,
      "num_input_tokens_seen": 18369152,
      "step": 87030
    },
    {
      "epoch": 9.574807480748074,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.018693067042830594,
      "loss": 0.2319,
      "num_input_tokens_seen": 18370144,
      "step": 87035
    },
    {
      "epoch": 9.575357535753575,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.018691671312636084,
      "loss": 0.2298,
      "num_input_tokens_seen": 18371232,
      "step": 87040
    },
    {
      "epoch": 9.575907590759076,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.018690275548417203,
      "loss": 0.2303,
      "num_input_tokens_seen": 18372320,
      "step": 87045
    },
    {
      "epoch": 9.576457645764577,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.018688879750186824,
      "loss": 0.2335,
      "num_input_tokens_seen": 18373376,
      "step": 87050
    },
    {
      "epoch": 9.577007700770077,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01868748391795781,
      "loss": 0.2319,
      "num_input_tokens_seen": 18374368,
      "step": 87055
    },
    {
      "epoch": 9.577557755775578,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.018686088051743033,
      "loss": 0.2309,
      "num_input_tokens_seen": 18375392,
      "step": 87060
    },
    {
      "epoch": 9.578107810781079,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.018684692151555347,
      "loss": 0.2314,
      "num_input_tokens_seen": 18376448,
      "step": 87065
    },
    {
      "epoch": 9.578657865786578,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.01868329621740762,
      "loss": 0.234,
      "num_input_tokens_seen": 18377408,
      "step": 87070
    },
    {
      "epoch": 9.57920792079208,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.018681900249312725,
      "loss": 0.2319,
      "num_input_tokens_seen": 18378432,
      "step": 87075
    },
    {
      "epoch": 9.57975797579758,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.01868050424728351,
      "loss": 0.2304,
      "num_input_tokens_seen": 18379488,
      "step": 87080
    },
    {
      "epoch": 9.58030803080308,
      "grad_norm": 0.000774383544921875,
      "learning_rate": 0.018679108211332864,
      "loss": 0.2329,
      "num_input_tokens_seen": 18380512,
      "step": 87085
    },
    {
      "epoch": 9.58085808580858,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01867771214147364,
      "loss": 0.2319,
      "num_input_tokens_seen": 18381568,
      "step": 87090
    },
    {
      "epoch": 9.581408140814082,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01867631603771871,
      "loss": 0.2288,
      "num_input_tokens_seen": 18382592,
      "step": 87095
    },
    {
      "epoch": 9.581958195819581,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.018674919900080935,
      "loss": 0.2314,
      "num_input_tokens_seen": 18383584,
      "step": 87100
    },
    {
      "epoch": 9.582508250825082,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.018673523728573192,
      "loss": 0.2346,
      "num_input_tokens_seen": 18384640,
      "step": 87105
    },
    {
      "epoch": 9.583058305830583,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.018672127523208338,
      "loss": 0.2303,
      "num_input_tokens_seen": 18385728,
      "step": 87110
    },
    {
      "epoch": 9.583608360836084,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.018670731283999254,
      "loss": 0.2335,
      "num_input_tokens_seen": 18386752,
      "step": 87115
    },
    {
      "epoch": 9.584158415841584,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.01866933501095879,
      "loss": 0.2314,
      "num_input_tokens_seen": 18387744,
      "step": 87120
    },
    {
      "epoch": 9.584708470847085,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.018667938704099836,
      "loss": 0.2309,
      "num_input_tokens_seen": 18388864,
      "step": 87125
    },
    {
      "epoch": 9.585258525852586,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.018666542363435245,
      "loss": 0.2329,
      "num_input_tokens_seen": 18389920,
      "step": 87130
    },
    {
      "epoch": 9.585808580858085,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.018665145988977894,
      "loss": 0.2314,
      "num_input_tokens_seen": 18390976,
      "step": 87135
    },
    {
      "epoch": 9.586358635863586,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.018663749580740654,
      "loss": 0.2304,
      "num_input_tokens_seen": 18392032,
      "step": 87140
    },
    {
      "epoch": 9.586908690869087,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.018662353138736393,
      "loss": 0.2308,
      "num_input_tokens_seen": 18393088,
      "step": 87145
    },
    {
      "epoch": 9.587458745874587,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01866095666297797,
      "loss": 0.2304,
      "num_input_tokens_seen": 18394144,
      "step": 87150
    },
    {
      "epoch": 9.588008800880088,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.018659560153478277,
      "loss": 0.2324,
      "num_input_tokens_seen": 18395200,
      "step": 87155
    },
    {
      "epoch": 9.588558855885589,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.018658163610250175,
      "loss": 0.2324,
      "num_input_tokens_seen": 18396256,
      "step": 87160
    },
    {
      "epoch": 9.589108910891088,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.018656767033306525,
      "loss": 0.2324,
      "num_input_tokens_seen": 18397312,
      "step": 87165
    },
    {
      "epoch": 9.58965896589659,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.018655370422660213,
      "loss": 0.2304,
      "num_input_tokens_seen": 18398368,
      "step": 87170
    },
    {
      "epoch": 9.59020902090209,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01865397377832411,
      "loss": 0.2298,
      "num_input_tokens_seen": 18399456,
      "step": 87175
    },
    {
      "epoch": 9.590759075907592,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.018652577100311075,
      "loss": 0.2314,
      "num_input_tokens_seen": 18400512,
      "step": 87180
    },
    {
      "epoch": 9.591309130913091,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.018651180388634,
      "loss": 0.2314,
      "num_input_tokens_seen": 18401568,
      "step": 87185
    },
    {
      "epoch": 9.591859185918592,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01864978364330574,
      "loss": 0.2324,
      "num_input_tokens_seen": 18402624,
      "step": 87190
    },
    {
      "epoch": 9.592409240924093,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.018648386864339173,
      "loss": 0.2319,
      "num_input_tokens_seen": 18403712,
      "step": 87195
    },
    {
      "epoch": 9.592959295929592,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01864699005174718,
      "loss": 0.2314,
      "num_input_tokens_seen": 18404736,
      "step": 87200
    },
    {
      "epoch": 9.593509350935093,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.01864559320554263,
      "loss": 0.2314,
      "num_input_tokens_seen": 18405760,
      "step": 87205
    },
    {
      "epoch": 9.594059405940595,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.018644196325738396,
      "loss": 0.2319,
      "num_input_tokens_seen": 18406848,
      "step": 87210
    },
    {
      "epoch": 9.594609460946094,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.018642799412347352,
      "loss": 0.233,
      "num_input_tokens_seen": 18407936,
      "step": 87215
    },
    {
      "epoch": 9.595159515951595,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01864140246538238,
      "loss": 0.2335,
      "num_input_tokens_seen": 18408960,
      "step": 87220
    },
    {
      "epoch": 9.595709570957096,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.018640005484856337,
      "loss": 0.2309,
      "num_input_tokens_seen": 18410112,
      "step": 87225
    },
    {
      "epoch": 9.596259625962595,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.018638608470782123,
      "loss": 0.2304,
      "num_input_tokens_seen": 18411104,
      "step": 87230
    },
    {
      "epoch": 9.596809680968097,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01863721142317259,
      "loss": 0.2319,
      "num_input_tokens_seen": 18412192,
      "step": 87235
    },
    {
      "epoch": 9.597359735973598,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.01863581434204063,
      "loss": 0.2319,
      "num_input_tokens_seen": 18413216,
      "step": 87240
    },
    {
      "epoch": 9.597909790979099,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.01863441722739911,
      "loss": 0.2319,
      "num_input_tokens_seen": 18414304,
      "step": 87245
    },
    {
      "epoch": 9.598459845984598,
      "grad_norm": 0.009765625,
      "learning_rate": 0.018633020079260917,
      "loss": 0.2314,
      "num_input_tokens_seen": 18415360,
      "step": 87250
    },
    {
      "epoch": 9.599009900990099,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.018631622897638915,
      "loss": 0.233,
      "num_input_tokens_seen": 18416384,
      "step": 87255
    },
    {
      "epoch": 9.5995599559956,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01863022568254599,
      "loss": 0.232,
      "num_input_tokens_seen": 18417472,
      "step": 87260
    },
    {
      "epoch": 9.6001100110011,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.018628828433995014,
      "loss": 0.2298,
      "num_input_tokens_seen": 18418496,
      "step": 87265
    },
    {
      "epoch": 9.6006600660066,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.018627431151998868,
      "loss": 0.2314,
      "num_input_tokens_seen": 18419520,
      "step": 87270
    },
    {
      "epoch": 9.601210121012102,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.018626033836570435,
      "loss": 0.2309,
      "num_input_tokens_seen": 18420544,
      "step": 87275
    },
    {
      "epoch": 9.601760176017601,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.018624636487722585,
      "loss": 0.2329,
      "num_input_tokens_seen": 18421568,
      "step": 87280
    },
    {
      "epoch": 9.602310231023102,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.0186232391054682,
      "loss": 0.2319,
      "num_input_tokens_seen": 18422656,
      "step": 87285
    },
    {
      "epoch": 9.602860286028603,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.01862184168982016,
      "loss": 0.2324,
      "num_input_tokens_seen": 18423680,
      "step": 87290
    },
    {
      "epoch": 9.603410341034103,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.018620444240791338,
      "loss": 0.2319,
      "num_input_tokens_seen": 18424736,
      "step": 87295
    },
    {
      "epoch": 9.603960396039604,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.01861904675839462,
      "loss": 0.2314,
      "num_input_tokens_seen": 18425792,
      "step": 87300
    },
    {
      "epoch": 9.604510451045105,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.018617649242642892,
      "loss": 0.2319,
      "num_input_tokens_seen": 18426816,
      "step": 87305
    },
    {
      "epoch": 9.605060506050606,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01861625169354902,
      "loss": 0.2324,
      "num_input_tokens_seen": 18427808,
      "step": 87310
    },
    {
      "epoch": 9.605610561056105,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.018614854111125893,
      "loss": 0.2309,
      "num_input_tokens_seen": 18428864,
      "step": 87315
    },
    {
      "epoch": 9.606160616061606,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01861345649538639,
      "loss": 0.2309,
      "num_input_tokens_seen": 18429888,
      "step": 87320
    },
    {
      "epoch": 9.606710671067107,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.018612058846343392,
      "loss": 0.233,
      "num_input_tokens_seen": 18430912,
      "step": 87325
    },
    {
      "epoch": 9.607260726072607,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.018610661164009782,
      "loss": 0.2309,
      "num_input_tokens_seen": 18431936,
      "step": 87330
    },
    {
      "epoch": 9.607810781078108,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01860926344839844,
      "loss": 0.2314,
      "num_input_tokens_seen": 18432928,
      "step": 87335
    },
    {
      "epoch": 9.608360836083609,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.018607865699522247,
      "loss": 0.2309,
      "num_input_tokens_seen": 18433952,
      "step": 87340
    },
    {
      "epoch": 9.608910891089108,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01860646791739409,
      "loss": 0.2309,
      "num_input_tokens_seen": 18434976,
      "step": 87345
    },
    {
      "epoch": 9.60946094609461,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.018605070102026854,
      "loss": 0.2324,
      "num_input_tokens_seen": 18436032,
      "step": 87350
    },
    {
      "epoch": 9.61001100110011,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.0186036722534334,
      "loss": 0.2303,
      "num_input_tokens_seen": 18437088,
      "step": 87355
    },
    {
      "epoch": 9.61056105610561,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.018602274371626645,
      "loss": 0.2304,
      "num_input_tokens_seen": 18438176,
      "step": 87360
    },
    {
      "epoch": 9.61111111111111,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.018600876456619446,
      "loss": 0.2319,
      "num_input_tokens_seen": 18439232,
      "step": 87365
    },
    {
      "epoch": 9.611661166116612,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.018599478508424697,
      "loss": 0.2304,
      "num_input_tokens_seen": 18440256,
      "step": 87370
    },
    {
      "epoch": 9.612211221122113,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.01859808052705529,
      "loss": 0.2308,
      "num_input_tokens_seen": 18441344,
      "step": 87375
    },
    {
      "epoch": 9.612761276127612,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.018596682512524092,
      "loss": 0.2303,
      "num_input_tokens_seen": 18442304,
      "step": 87380
    },
    {
      "epoch": 9.613311331133113,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.018595284464844,
      "loss": 0.2314,
      "num_input_tokens_seen": 18443296,
      "step": 87385
    },
    {
      "epoch": 9.613861386138614,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.0185938863840279,
      "loss": 0.2309,
      "num_input_tokens_seen": 18444288,
      "step": 87390
    },
    {
      "epoch": 9.614411441144114,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01859248827008867,
      "loss": 0.2304,
      "num_input_tokens_seen": 18445408,
      "step": 87395
    },
    {
      "epoch": 9.614961496149615,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0185910901230392,
      "loss": 0.2335,
      "num_input_tokens_seen": 18446464,
      "step": 87400
    },
    {
      "epoch": 9.615511551155116,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.018589691942892377,
      "loss": 0.2329,
      "num_input_tokens_seen": 18447456,
      "step": 87405
    },
    {
      "epoch": 9.616061606160617,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.018588293729661084,
      "loss": 0.2314,
      "num_input_tokens_seen": 18448480,
      "step": 87410
    },
    {
      "epoch": 9.616611661166116,
      "grad_norm": 0.009765625,
      "learning_rate": 0.01858689548335821,
      "loss": 0.2298,
      "num_input_tokens_seen": 18449568,
      "step": 87415
    },
    {
      "epoch": 9.617161716171617,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.01858549720399664,
      "loss": 0.2319,
      "num_input_tokens_seen": 18450624,
      "step": 87420
    },
    {
      "epoch": 9.617711771177119,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.018584098891589265,
      "loss": 0.2319,
      "num_input_tokens_seen": 18451712,
      "step": 87425
    },
    {
      "epoch": 9.618261826182618,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.018582700546148967,
      "loss": 0.2299,
      "num_input_tokens_seen": 18452800,
      "step": 87430
    },
    {
      "epoch": 9.618811881188119,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01858130216768864,
      "loss": 0.2298,
      "num_input_tokens_seen": 18453920,
      "step": 87435
    },
    {
      "epoch": 9.61936193619362,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.01857990375622117,
      "loss": 0.233,
      "num_input_tokens_seen": 18455040,
      "step": 87440
    },
    {
      "epoch": 9.61991199119912,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.018578505311759447,
      "loss": 0.2304,
      "num_input_tokens_seen": 18456128,
      "step": 87445
    },
    {
      "epoch": 9.62046204620462,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.018577106834316355,
      "loss": 0.2298,
      "num_input_tokens_seen": 18457152,
      "step": 87450
    },
    {
      "epoch": 9.621012101210122,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01857570832390478,
      "loss": 0.2345,
      "num_input_tokens_seen": 18458176,
      "step": 87455
    },
    {
      "epoch": 9.62156215621562,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.018574309780537628,
      "loss": 0.2298,
      "num_input_tokens_seen": 18459264,
      "step": 87460
    },
    {
      "epoch": 9.622112211221122,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.01857291120422777,
      "loss": 0.2324,
      "num_input_tokens_seen": 18460288,
      "step": 87465
    },
    {
      "epoch": 9.622662266226623,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01857151259498811,
      "loss": 0.234,
      "num_input_tokens_seen": 18461344,
      "step": 87470
    },
    {
      "epoch": 9.623212321232124,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.018570113952831528,
      "loss": 0.2324,
      "num_input_tokens_seen": 18462368,
      "step": 87475
    },
    {
      "epoch": 9.623762376237623,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.01856871527777092,
      "loss": 0.2298,
      "num_input_tokens_seen": 18463456,
      "step": 87480
    },
    {
      "epoch": 9.624312431243125,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.01856731656981917,
      "loss": 0.2325,
      "num_input_tokens_seen": 18464544,
      "step": 87485
    },
    {
      "epoch": 9.624862486248626,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.018565917828989184,
      "loss": 0.2303,
      "num_input_tokens_seen": 18465664,
      "step": 87490
    },
    {
      "epoch": 9.625412541254125,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.018564519055293844,
      "loss": 0.2288,
      "num_input_tokens_seen": 18466752,
      "step": 87495
    },
    {
      "epoch": 9.625962596259626,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.018563120248746033,
      "loss": 0.2308,
      "num_input_tokens_seen": 18467808,
      "step": 87500
    },
    {
      "epoch": 9.626512651265127,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.01856172140935866,
      "loss": 0.2314,
      "num_input_tokens_seen": 18468864,
      "step": 87505
    },
    {
      "epoch": 9.627062706270626,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01856032253714461,
      "loss": 0.234,
      "num_input_tokens_seen": 18469856,
      "step": 87510
    },
    {
      "epoch": 9.627612761276128,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.01855892363211677,
      "loss": 0.2319,
      "num_input_tokens_seen": 18470976,
      "step": 87515
    },
    {
      "epoch": 9.628162816281629,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.01855752469428805,
      "loss": 0.2314,
      "num_input_tokens_seen": 18472032,
      "step": 87520
    },
    {
      "epoch": 9.628712871287128,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01855612572367132,
      "loss": 0.2319,
      "num_input_tokens_seen": 18473024,
      "step": 87525
    },
    {
      "epoch": 9.629262926292629,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01855472672027949,
      "loss": 0.2313,
      "num_input_tokens_seen": 18474048,
      "step": 87530
    },
    {
      "epoch": 9.62981298129813,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.01855332768412545,
      "loss": 0.2319,
      "num_input_tokens_seen": 18475104,
      "step": 87535
    },
    {
      "epoch": 9.630363036303631,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.01855192861522209,
      "loss": 0.234,
      "num_input_tokens_seen": 18476192,
      "step": 87540
    },
    {
      "epoch": 9.63091309130913,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.018550529513582315,
      "loss": 0.2329,
      "num_input_tokens_seen": 18477312,
      "step": 87545
    },
    {
      "epoch": 9.631463146314632,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.01854913037921901,
      "loss": 0.2361,
      "num_input_tokens_seen": 18478400,
      "step": 87550
    },
    {
      "epoch": 9.632013201320133,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.018547731212145074,
      "loss": 0.2303,
      "num_input_tokens_seen": 18479456,
      "step": 87555
    },
    {
      "epoch": 9.632563256325632,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0185463320123734,
      "loss": 0.2335,
      "num_input_tokens_seen": 18480512,
      "step": 87560
    },
    {
      "epoch": 9.633113311331133,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.018544932779916888,
      "loss": 0.2288,
      "num_input_tokens_seen": 18481632,
      "step": 87565
    },
    {
      "epoch": 9.633663366336634,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.018543533514788428,
      "loss": 0.2309,
      "num_input_tokens_seen": 18482688,
      "step": 87570
    },
    {
      "epoch": 9.634213421342134,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.018542134217000922,
      "loss": 0.2335,
      "num_input_tokens_seen": 18483744,
      "step": 87575
    },
    {
      "epoch": 9.634763476347635,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.018540734886567264,
      "loss": 0.2319,
      "num_input_tokens_seen": 18484768,
      "step": 87580
    },
    {
      "epoch": 9.635313531353136,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.01853933552350035,
      "loss": 0.2314,
      "num_input_tokens_seen": 18485792,
      "step": 87585
    },
    {
      "epoch": 9.635863586358635,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.018537936127813088,
      "loss": 0.2309,
      "num_input_tokens_seen": 18486784,
      "step": 87590
    },
    {
      "epoch": 9.636413641364136,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.018536536699518356,
      "loss": 0.2314,
      "num_input_tokens_seen": 18487840,
      "step": 87595
    },
    {
      "epoch": 9.636963696369637,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.018535137238629065,
      "loss": 0.2324,
      "num_input_tokens_seen": 18488928,
      "step": 87600
    },
    {
      "epoch": 9.637513751375138,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.01853373774515811,
      "loss": 0.2303,
      "num_input_tokens_seen": 18489952,
      "step": 87605
    },
    {
      "epoch": 9.638063806380638,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.018532338219118392,
      "loss": 0.2335,
      "num_input_tokens_seen": 18491008,
      "step": 87610
    },
    {
      "epoch": 9.638613861386139,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.018530938660522807,
      "loss": 0.2304,
      "num_input_tokens_seen": 18492160,
      "step": 87615
    },
    {
      "epoch": 9.63916391639164,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.018529539069384256,
      "loss": 0.2314,
      "num_input_tokens_seen": 18493216,
      "step": 87620
    },
    {
      "epoch": 9.63971397139714,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.018528139445715634,
      "loss": 0.2313,
      "num_input_tokens_seen": 18494208,
      "step": 87625
    },
    {
      "epoch": 9.64026402640264,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.01852673978952984,
      "loss": 0.2314,
      "num_input_tokens_seen": 18495232,
      "step": 87630
    },
    {
      "epoch": 9.640814081408141,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01852534010083979,
      "loss": 0.2288,
      "num_input_tokens_seen": 18496256,
      "step": 87635
    },
    {
      "epoch": 9.64136413641364,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.018523940379658363,
      "loss": 0.2319,
      "num_input_tokens_seen": 18497376,
      "step": 87640
    },
    {
      "epoch": 9.641914191419142,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.018522540625998465,
      "loss": 0.2319,
      "num_input_tokens_seen": 18498432,
      "step": 87645
    },
    {
      "epoch": 9.642464246424643,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01852114083987301,
      "loss": 0.2319,
      "num_input_tokens_seen": 18499456,
      "step": 87650
    },
    {
      "epoch": 9.643014301430142,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.018519741021294885,
      "loss": 0.2314,
      "num_input_tokens_seen": 18500416,
      "step": 87655
    },
    {
      "epoch": 9.643564356435643,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.018518341170276995,
      "loss": 0.2309,
      "num_input_tokens_seen": 18501504,
      "step": 87660
    },
    {
      "epoch": 9.644114411441144,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.018516941286832244,
      "loss": 0.233,
      "num_input_tokens_seen": 18502528,
      "step": 87665
    },
    {
      "epoch": 9.644664466446645,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.018515541370973533,
      "loss": 0.2319,
      "num_input_tokens_seen": 18503520,
      "step": 87670
    },
    {
      "epoch": 9.645214521452145,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.018514141422713758,
      "loss": 0.2293,
      "num_input_tokens_seen": 18504608,
      "step": 87675
    },
    {
      "epoch": 9.645764576457646,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.018512741442065835,
      "loss": 0.2304,
      "num_input_tokens_seen": 18505600,
      "step": 87680
    },
    {
      "epoch": 9.646314631463147,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.018511341429042656,
      "loss": 0.2324,
      "num_input_tokens_seen": 18506592,
      "step": 87685
    },
    {
      "epoch": 9.646864686468646,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.018509941383657132,
      "loss": 0.2314,
      "num_input_tokens_seen": 18507648,
      "step": 87690
    },
    {
      "epoch": 9.647414741474147,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.018508541305922162,
      "loss": 0.2335,
      "num_input_tokens_seen": 18508736,
      "step": 87695
    },
    {
      "epoch": 9.647964796479648,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.018507141195850646,
      "loss": 0.2319,
      "num_input_tokens_seen": 18509760,
      "step": 87700
    },
    {
      "epoch": 9.648514851485148,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.018505741053455493,
      "loss": 0.2319,
      "num_input_tokens_seen": 18510848,
      "step": 87705
    },
    {
      "epoch": 9.649064906490649,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.018504340878749612,
      "loss": 0.2319,
      "num_input_tokens_seen": 18511968,
      "step": 87710
    },
    {
      "epoch": 9.64961496149615,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.018502940671745897,
      "loss": 0.2293,
      "num_input_tokens_seen": 18513024,
      "step": 87715
    },
    {
      "epoch": 9.65016501650165,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01850154043245726,
      "loss": 0.2314,
      "num_input_tokens_seen": 18514080,
      "step": 87720
    },
    {
      "epoch": 9.65071507150715,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.01850014016089661,
      "loss": 0.2298,
      "num_input_tokens_seen": 18515104,
      "step": 87725
    },
    {
      "epoch": 9.651265126512651,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.018498739857076844,
      "loss": 0.2293,
      "num_input_tokens_seen": 18516160,
      "step": 87730
    },
    {
      "epoch": 9.651815181518153,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.01849733952101087,
      "loss": 0.2303,
      "num_input_tokens_seen": 18517184,
      "step": 87735
    },
    {
      "epoch": 9.652365236523652,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.0184959391527116,
      "loss": 0.2303,
      "num_input_tokens_seen": 18518272,
      "step": 87740
    },
    {
      "epoch": 9.652915291529153,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01849453875219193,
      "loss": 0.2314,
      "num_input_tokens_seen": 18519328,
      "step": 87745
    },
    {
      "epoch": 9.653465346534654,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01849313831946478,
      "loss": 0.2329,
      "num_input_tokens_seen": 18520384,
      "step": 87750
    },
    {
      "epoch": 9.654015401540153,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.01849173785454305,
      "loss": 0.2319,
      "num_input_tokens_seen": 18521376,
      "step": 87755
    },
    {
      "epoch": 9.654565456545654,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.018490337357439643,
      "loss": 0.2309,
      "num_input_tokens_seen": 18522368,
      "step": 87760
    },
    {
      "epoch": 9.655115511551156,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.018488936828167473,
      "loss": 0.2303,
      "num_input_tokens_seen": 18523456,
      "step": 87765
    },
    {
      "epoch": 9.655665566556655,
      "grad_norm": 0.000644683837890625,
      "learning_rate": 0.018487536266739445,
      "loss": 0.2319,
      "num_input_tokens_seen": 18524512,
      "step": 87770
    },
    {
      "epoch": 9.656215621562156,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01848613567316847,
      "loss": 0.2288,
      "num_input_tokens_seen": 18525568,
      "step": 87775
    },
    {
      "epoch": 9.656765676567657,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01848473504746746,
      "loss": 0.2303,
      "num_input_tokens_seen": 18526592,
      "step": 87780
    },
    {
      "epoch": 9.657315731573158,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.018483334389649313,
      "loss": 0.2304,
      "num_input_tokens_seen": 18527648,
      "step": 87785
    },
    {
      "epoch": 9.657865786578657,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.018481933699726946,
      "loss": 0.2304,
      "num_input_tokens_seen": 18528704,
      "step": 87790
    },
    {
      "epoch": 9.658415841584159,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.018480532977713265,
      "loss": 0.2304,
      "num_input_tokens_seen": 18529760,
      "step": 87795
    },
    {
      "epoch": 9.65896589658966,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.018479132223621183,
      "loss": 0.2335,
      "num_input_tokens_seen": 18530720,
      "step": 87800
    },
    {
      "epoch": 9.659515951595159,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.018477731437463615,
      "loss": 0.2289,
      "num_input_tokens_seen": 18531744,
      "step": 87805
    },
    {
      "epoch": 9.66006600660066,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.018476330619253457,
      "loss": 0.2319,
      "num_input_tokens_seen": 18532800,
      "step": 87810
    },
    {
      "epoch": 9.660616061606161,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.018474929769003633,
      "loss": 0.2309,
      "num_input_tokens_seen": 18533856,
      "step": 87815
    },
    {
      "epoch": 9.66116611661166,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.018473528886727042,
      "loss": 0.2325,
      "num_input_tokens_seen": 18534944,
      "step": 87820
    },
    {
      "epoch": 9.661716171617162,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.018472127972436612,
      "loss": 0.2314,
      "num_input_tokens_seen": 18536032,
      "step": 87825
    },
    {
      "epoch": 9.662266226622663,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.018470727026145238,
      "loss": 0.2325,
      "num_input_tokens_seen": 18537056,
      "step": 87830
    },
    {
      "epoch": 9.662816281628164,
      "grad_norm": 0.001953125,
      "learning_rate": 0.018469326047865837,
      "loss": 0.233,
      "num_input_tokens_seen": 18538080,
      "step": 87835
    },
    {
      "epoch": 9.663366336633663,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.018467925037611323,
      "loss": 0.2314,
      "num_input_tokens_seen": 18539104,
      "step": 87840
    },
    {
      "epoch": 9.663916391639164,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.018466523995394613,
      "loss": 0.2267,
      "num_input_tokens_seen": 18540128,
      "step": 87845
    },
    {
      "epoch": 9.664466446644665,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01846512292122861,
      "loss": 0.2284,
      "num_input_tokens_seen": 18541184,
      "step": 87850
    },
    {
      "epoch": 9.665016501650165,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.018463721815126233,
      "loss": 0.2304,
      "num_input_tokens_seen": 18542272,
      "step": 87855
    },
    {
      "epoch": 9.665566556655666,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.018462320677100394,
      "loss": 0.2342,
      "num_input_tokens_seen": 18543296,
      "step": 87860
    },
    {
      "epoch": 9.666116611661167,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.018460919507164008,
      "loss": 0.2305,
      "num_input_tokens_seen": 18544384,
      "step": 87865
    },
    {
      "epoch": 9.666666666666666,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.018459518305329988,
      "loss": 0.2305,
      "num_input_tokens_seen": 18545408,
      "step": 87870
    },
    {
      "epoch": 9.667216721672167,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.018458117071611244,
      "loss": 0.2336,
      "num_input_tokens_seen": 18546432,
      "step": 87875
    },
    {
      "epoch": 9.667766776677668,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0184567158060207,
      "loss": 0.232,
      "num_input_tokens_seen": 18547456,
      "step": 87880
    },
    {
      "epoch": 9.668316831683168,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.018455314508571262,
      "loss": 0.2294,
      "num_input_tokens_seen": 18548640,
      "step": 87885
    },
    {
      "epoch": 9.668866886688669,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01845391317927585,
      "loss": 0.2309,
      "num_input_tokens_seen": 18549632,
      "step": 87890
    },
    {
      "epoch": 9.66941694169417,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.01845251181814737,
      "loss": 0.2278,
      "num_input_tokens_seen": 18550720,
      "step": 87895
    },
    {
      "epoch": 9.66996699669967,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.018451110425198754,
      "loss": 0.2314,
      "num_input_tokens_seen": 18551808,
      "step": 87900
    },
    {
      "epoch": 9.67051705170517,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.018449709000442904,
      "loss": 0.2294,
      "num_input_tokens_seen": 18552928,
      "step": 87905
    },
    {
      "epoch": 9.671067106710671,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.018448307543892744,
      "loss": 0.2284,
      "num_input_tokens_seen": 18554016,
      "step": 87910
    },
    {
      "epoch": 9.671617161716172,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.018446906055561187,
      "loss": 0.2341,
      "num_input_tokens_seen": 18555072,
      "step": 87915
    },
    {
      "epoch": 9.672167216721672,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.018445504535461153,
      "loss": 0.2357,
      "num_input_tokens_seen": 18556128,
      "step": 87920
    },
    {
      "epoch": 9.672717271727173,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.018444102983605555,
      "loss": 0.2284,
      "num_input_tokens_seen": 18557184,
      "step": 87925
    },
    {
      "epoch": 9.673267326732674,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.018442701400007314,
      "loss": 0.232,
      "num_input_tokens_seen": 18558240,
      "step": 87930
    },
    {
      "epoch": 9.673817381738173,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.018441299784679344,
      "loss": 0.2315,
      "num_input_tokens_seen": 18559232,
      "step": 87935
    },
    {
      "epoch": 9.674367436743674,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.018439898137634567,
      "loss": 0.2315,
      "num_input_tokens_seen": 18560288,
      "step": 87940
    },
    {
      "epoch": 9.674917491749175,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0184384964588859,
      "loss": 0.2325,
      "num_input_tokens_seen": 18561408,
      "step": 87945
    },
    {
      "epoch": 9.675467546754675,
      "grad_norm": 0.0004520416259765625,
      "learning_rate": 0.01843709474844626,
      "loss": 0.2367,
      "num_input_tokens_seen": 18562496,
      "step": 87950
    },
    {
      "epoch": 9.676017601760176,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.018435693006328566,
      "loss": 0.232,
      "num_input_tokens_seen": 18563488,
      "step": 87955
    },
    {
      "epoch": 9.676567656765677,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01843429123254574,
      "loss": 0.232,
      "num_input_tokens_seen": 18564544,
      "step": 87960
    },
    {
      "epoch": 9.677117711771178,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.018432889427110695,
      "loss": 0.234,
      "num_input_tokens_seen": 18565600,
      "step": 87965
    },
    {
      "epoch": 9.677667766776677,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.018431487590036363,
      "loss": 0.2314,
      "num_input_tokens_seen": 18566656,
      "step": 87970
    },
    {
      "epoch": 9.678217821782178,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.018430085721335653,
      "loss": 0.2293,
      "num_input_tokens_seen": 18567744,
      "step": 87975
    },
    {
      "epoch": 9.67876787678768,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.018428683821021485,
      "loss": 0.233,
      "num_input_tokens_seen": 18568768,
      "step": 87980
    },
    {
      "epoch": 9.679317931793179,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.01842728188910679,
      "loss": 0.2293,
      "num_input_tokens_seen": 18569824,
      "step": 87985
    },
    {
      "epoch": 9.67986798679868,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.018425879925604478,
      "loss": 0.2325,
      "num_input_tokens_seen": 18570944,
      "step": 87990
    },
    {
      "epoch": 9.680418041804181,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.018424477930527476,
      "loss": 0.233,
      "num_input_tokens_seen": 18571968,
      "step": 87995
    },
    {
      "epoch": 9.68096809680968,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.018423075903888707,
      "loss": 0.2309,
      "num_input_tokens_seen": 18573088,
      "step": 88000
    },
    {
      "epoch": 9.681518151815181,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.01842167384570109,
      "loss": 0.233,
      "num_input_tokens_seen": 18574208,
      "step": 88005
    },
    {
      "epoch": 9.682068206820682,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.018420271755977536,
      "loss": 0.2325,
      "num_input_tokens_seen": 18575296,
      "step": 88010
    },
    {
      "epoch": 9.682618261826182,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01841886963473099,
      "loss": 0.2304,
      "num_input_tokens_seen": 18576352,
      "step": 88015
    },
    {
      "epoch": 9.683168316831683,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.01841746748197436,
      "loss": 0.2304,
      "num_input_tokens_seen": 18577440,
      "step": 88020
    },
    {
      "epoch": 9.683718371837184,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.01841606529772057,
      "loss": 0.2325,
      "num_input_tokens_seen": 18578528,
      "step": 88025
    },
    {
      "epoch": 9.684268426842685,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.018414663081982547,
      "loss": 0.2335,
      "num_input_tokens_seen": 18579616,
      "step": 88030
    },
    {
      "epoch": 9.684818481848184,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.018413260834773213,
      "loss": 0.2319,
      "num_input_tokens_seen": 18580640,
      "step": 88035
    },
    {
      "epoch": 9.685368536853685,
      "grad_norm": 0.0027313232421875,
      "learning_rate": 0.018411858556105495,
      "loss": 0.2309,
      "num_input_tokens_seen": 18581696,
      "step": 88040
    },
    {
      "epoch": 9.685918591859187,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.01841045624599231,
      "loss": 0.2299,
      "num_input_tokens_seen": 18582752,
      "step": 88045
    },
    {
      "epoch": 9.686468646864686,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.018409053904446582,
      "loss": 0.2341,
      "num_input_tokens_seen": 18583776,
      "step": 88050
    },
    {
      "epoch": 9.687018701870187,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.018407651531481245,
      "loss": 0.2314,
      "num_input_tokens_seen": 18584800,
      "step": 88055
    },
    {
      "epoch": 9.687568756875688,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.018406249127109222,
      "loss": 0.2319,
      "num_input_tokens_seen": 18585888,
      "step": 88060
    },
    {
      "epoch": 9.688118811881187,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.01840484669134343,
      "loss": 0.2299,
      "num_input_tokens_seen": 18586912,
      "step": 88065
    },
    {
      "epoch": 9.688668866886688,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.018403444224196805,
      "loss": 0.232,
      "num_input_tokens_seen": 18587968,
      "step": 88070
    },
    {
      "epoch": 9.68921892189219,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.018402041725682265,
      "loss": 0.2299,
      "num_input_tokens_seen": 18588928,
      "step": 88075
    },
    {
      "epoch": 9.689768976897689,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.018400639195812734,
      "loss": 0.2325,
      "num_input_tokens_seen": 18589952,
      "step": 88080
    },
    {
      "epoch": 9.69031903190319,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.01839923663460115,
      "loss": 0.2283,
      "num_input_tokens_seen": 18590976,
      "step": 88085
    },
    {
      "epoch": 9.690869086908691,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.018397834042060433,
      "loss": 0.234,
      "num_input_tokens_seen": 18592096,
      "step": 88090
    },
    {
      "epoch": 9.691419141914192,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.018396431418203508,
      "loss": 0.2309,
      "num_input_tokens_seen": 18593216,
      "step": 88095
    },
    {
      "epoch": 9.691969196919691,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.018395028763043303,
      "loss": 0.2319,
      "num_input_tokens_seen": 18594272,
      "step": 88100
    },
    {
      "epoch": 9.692519251925193,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01839362607659275,
      "loss": 0.2324,
      "num_input_tokens_seen": 18595360,
      "step": 88105
    },
    {
      "epoch": 9.693069306930694,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01839222335886477,
      "loss": 0.2309,
      "num_input_tokens_seen": 18596384,
      "step": 88110
    },
    {
      "epoch": 9.693619361936193,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0183908206098723,
      "loss": 0.2293,
      "num_input_tokens_seen": 18597440,
      "step": 88115
    },
    {
      "epoch": 9.694169416941694,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.01838941782962826,
      "loss": 0.2335,
      "num_input_tokens_seen": 18598592,
      "step": 88120
    },
    {
      "epoch": 9.694719471947195,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01838801501814558,
      "loss": 0.2309,
      "num_input_tokens_seen": 18599648,
      "step": 88125
    },
    {
      "epoch": 9.695269526952695,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.018386612175437196,
      "loss": 0.2304,
      "num_input_tokens_seen": 18600672,
      "step": 88130
    },
    {
      "epoch": 9.695819581958196,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.01838520930151603,
      "loss": 0.2303,
      "num_input_tokens_seen": 18601760,
      "step": 88135
    },
    {
      "epoch": 9.696369636963697,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.018383806396395017,
      "loss": 0.2314,
      "num_input_tokens_seen": 18602880,
      "step": 88140
    },
    {
      "epoch": 9.696919691969196,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01838240346008708,
      "loss": 0.2309,
      "num_input_tokens_seen": 18603904,
      "step": 88145
    },
    {
      "epoch": 9.697469746974697,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.018381000492605156,
      "loss": 0.2314,
      "num_input_tokens_seen": 18604960,
      "step": 88150
    },
    {
      "epoch": 9.698019801980198,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01837959749396217,
      "loss": 0.2309,
      "num_input_tokens_seen": 18605984,
      "step": 88155
    },
    {
      "epoch": 9.6985698569857,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.018378194464171063,
      "loss": 0.2304,
      "num_input_tokens_seen": 18607104,
      "step": 88160
    },
    {
      "epoch": 9.699119911991199,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.018376791403244754,
      "loss": 0.2314,
      "num_input_tokens_seen": 18608192,
      "step": 88165
    },
    {
      "epoch": 9.6996699669967,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.018375388311196176,
      "loss": 0.2283,
      "num_input_tokens_seen": 18609248,
      "step": 88170
    },
    {
      "epoch": 9.7002200220022,
      "grad_norm": 0.00070953369140625,
      "learning_rate": 0.018373985188038268,
      "loss": 0.2304,
      "num_input_tokens_seen": 18610208,
      "step": 88175
    },
    {
      "epoch": 9.7007700770077,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.018372582033783957,
      "loss": 0.2319,
      "num_input_tokens_seen": 18611264,
      "step": 88180
    },
    {
      "epoch": 9.701320132013201,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.018371178848446173,
      "loss": 0.2298,
      "num_input_tokens_seen": 18612320,
      "step": 88185
    },
    {
      "epoch": 9.701870187018702,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01836977563203785,
      "loss": 0.2324,
      "num_input_tokens_seen": 18613344,
      "step": 88190
    },
    {
      "epoch": 9.702420242024202,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.018368372384571927,
      "loss": 0.2309,
      "num_input_tokens_seen": 18614432,
      "step": 88195
    },
    {
      "epoch": 9.702970297029703,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.018366969106061327,
      "loss": 0.2329,
      "num_input_tokens_seen": 18615424,
      "step": 88200
    },
    {
      "epoch": 9.703520352035204,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.018365565796518993,
      "loss": 0.2303,
      "num_input_tokens_seen": 18616480,
      "step": 88205
    },
    {
      "epoch": 9.704070407040705,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01836416245595785,
      "loss": 0.233,
      "num_input_tokens_seen": 18617568,
      "step": 88210
    },
    {
      "epoch": 9.704620462046204,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01836275908439084,
      "loss": 0.2324,
      "num_input_tokens_seen": 18618592,
      "step": 88215
    },
    {
      "epoch": 9.705170517051705,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.018361355681830893,
      "loss": 0.2319,
      "num_input_tokens_seen": 18619680,
      "step": 88220
    },
    {
      "epoch": 9.705720572057206,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.018359952248290942,
      "loss": 0.2319,
      "num_input_tokens_seen": 18620736,
      "step": 88225
    },
    {
      "epoch": 9.706270627062706,
      "grad_norm": 0.00060272216796875,
      "learning_rate": 0.018358548783783925,
      "loss": 0.2308,
      "num_input_tokens_seen": 18621760,
      "step": 88230
    },
    {
      "epoch": 9.706820682068207,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.01835714528832277,
      "loss": 0.2319,
      "num_input_tokens_seen": 18622848,
      "step": 88235
    },
    {
      "epoch": 9.707370737073708,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01835574176192042,
      "loss": 0.2314,
      "num_input_tokens_seen": 18623872,
      "step": 88240
    },
    {
      "epoch": 9.707920792079207,
      "grad_norm": 0.00041961669921875,
      "learning_rate": 0.01835433820458981,
      "loss": 0.2329,
      "num_input_tokens_seen": 18624928,
      "step": 88245
    },
    {
      "epoch": 9.708470847084708,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.018352934616343877,
      "loss": 0.2319,
      "num_input_tokens_seen": 18625984,
      "step": 88250
    },
    {
      "epoch": 9.70902090209021,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01835153099719555,
      "loss": 0.2308,
      "num_input_tokens_seen": 18627040,
      "step": 88255
    },
    {
      "epoch": 9.70957095709571,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.018350127347157772,
      "loss": 0.2303,
      "num_input_tokens_seen": 18628096,
      "step": 88260
    },
    {
      "epoch": 9.71012101210121,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01834872366624348,
      "loss": 0.2324,
      "num_input_tokens_seen": 18629152,
      "step": 88265
    },
    {
      "epoch": 9.710671067106711,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.0183473199544656,
      "loss": 0.2293,
      "num_input_tokens_seen": 18630176,
      "step": 88270
    },
    {
      "epoch": 9.711221122112212,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.01834591621183709,
      "loss": 0.2309,
      "num_input_tokens_seen": 18631296,
      "step": 88275
    },
    {
      "epoch": 9.711771177117711,
      "grad_norm": 0.005859375,
      "learning_rate": 0.01834451243837087,
      "loss": 0.2324,
      "num_input_tokens_seen": 18632352,
      "step": 88280
    },
    {
      "epoch": 9.712321232123212,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.018343108634079883,
      "loss": 0.2303,
      "num_input_tokens_seen": 18633440,
      "step": 88285
    },
    {
      "epoch": 9.712871287128714,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.018341704798977072,
      "loss": 0.2314,
      "num_input_tokens_seen": 18634464,
      "step": 88290
    },
    {
      "epoch": 9.713421342134213,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.018340300933075367,
      "loss": 0.2288,
      "num_input_tokens_seen": 18635424,
      "step": 88295
    },
    {
      "epoch": 9.713971397139714,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.018338897036387712,
      "loss": 0.2303,
      "num_input_tokens_seen": 18636512,
      "step": 88300
    },
    {
      "epoch": 9.714521452145215,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.018337493108927044,
      "loss": 0.2329,
      "num_input_tokens_seen": 18637632,
      "step": 88305
    },
    {
      "epoch": 9.715071507150714,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01833608915070631,
      "loss": 0.2324,
      "num_input_tokens_seen": 18638752,
      "step": 88310
    },
    {
      "epoch": 9.715621562156215,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.018334685161738435,
      "loss": 0.2309,
      "num_input_tokens_seen": 18639776,
      "step": 88315
    },
    {
      "epoch": 9.716171617161717,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.01833328114203637,
      "loss": 0.2309,
      "num_input_tokens_seen": 18640800,
      "step": 88320
    },
    {
      "epoch": 9.716721672167218,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.01833187709161305,
      "loss": 0.2308,
      "num_input_tokens_seen": 18641888,
      "step": 88325
    },
    {
      "epoch": 9.717271727172717,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.01833047301048142,
      "loss": 0.2303,
      "num_input_tokens_seen": 18642976,
      "step": 88330
    },
    {
      "epoch": 9.717821782178218,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.018329068898654422,
      "loss": 0.2319,
      "num_input_tokens_seen": 18643968,
      "step": 88335
    },
    {
      "epoch": 9.718371837183719,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.01832766475614499,
      "loss": 0.2329,
      "num_input_tokens_seen": 18645024,
      "step": 88340
    },
    {
      "epoch": 9.718921892189218,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.018326260582966065,
      "loss": 0.2319,
      "num_input_tokens_seen": 18646048,
      "step": 88345
    },
    {
      "epoch": 9.71947194719472,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0183248563791306,
      "loss": 0.2319,
      "num_input_tokens_seen": 18647136,
      "step": 88350
    },
    {
      "epoch": 9.72002200220022,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.018323452144651524,
      "loss": 0.2309,
      "num_input_tokens_seen": 18648192,
      "step": 88355
    },
    {
      "epoch": 9.72057205720572,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.018322047879541784,
      "loss": 0.2298,
      "num_input_tokens_seen": 18649248,
      "step": 88360
    },
    {
      "epoch": 9.721122112211221,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.018320643583814324,
      "loss": 0.2308,
      "num_input_tokens_seen": 18650336,
      "step": 88365
    },
    {
      "epoch": 9.721672167216722,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.018319239257482085,
      "loss": 0.2319,
      "num_input_tokens_seen": 18651392,
      "step": 88370
    },
    {
      "epoch": 9.722222222222221,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01831783490055801,
      "loss": 0.2314,
      "num_input_tokens_seen": 18652448,
      "step": 88375
    },
    {
      "epoch": 9.722772277227723,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.018316430513055048,
      "loss": 0.2319,
      "num_input_tokens_seen": 18653600,
      "step": 88380
    },
    {
      "epoch": 9.723322332233224,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.018315026094986128,
      "loss": 0.2309,
      "num_input_tokens_seen": 18654720,
      "step": 88385
    },
    {
      "epoch": 9.723872387238725,
      "grad_norm": 0.000705718994140625,
      "learning_rate": 0.01831362164636421,
      "loss": 0.2303,
      "num_input_tokens_seen": 18655744,
      "step": 88390
    },
    {
      "epoch": 9.724422442244224,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.018312217167202237,
      "loss": 0.2309,
      "num_input_tokens_seen": 18656864,
      "step": 88395
    },
    {
      "epoch": 9.724972497249725,
      "grad_norm": 0.00067901611328125,
      "learning_rate": 0.018310812657513133,
      "loss": 0.2319,
      "num_input_tokens_seen": 18657920,
      "step": 88400
    },
    {
      "epoch": 9.725522552255226,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01830940811730987,
      "loss": 0.2309,
      "num_input_tokens_seen": 18658944,
      "step": 88405
    },
    {
      "epoch": 9.726072607260726,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.018308003546605375,
      "loss": 0.233,
      "num_input_tokens_seen": 18660000,
      "step": 88410
    },
    {
      "epoch": 9.726622662266227,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.018306598945412596,
      "loss": 0.2329,
      "num_input_tokens_seen": 18661120,
      "step": 88415
    },
    {
      "epoch": 9.727172717271728,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.018305194313744486,
      "loss": 0.2329,
      "num_input_tokens_seen": 18662144,
      "step": 88420
    },
    {
      "epoch": 9.727722772277227,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.018303789651613986,
      "loss": 0.2324,
      "num_input_tokens_seen": 18663168,
      "step": 88425
    },
    {
      "epoch": 9.728272827282728,
      "grad_norm": 0.000591278076171875,
      "learning_rate": 0.018302384959034034,
      "loss": 0.2324,
      "num_input_tokens_seen": 18664160,
      "step": 88430
    },
    {
      "epoch": 9.72882288228823,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.018300980236017592,
      "loss": 0.2319,
      "num_input_tokens_seen": 18665184,
      "step": 88435
    },
    {
      "epoch": 9.729372937293729,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0182995754825776,
      "loss": 0.2319,
      "num_input_tokens_seen": 18666240,
      "step": 88440
    },
    {
      "epoch": 9.72992299229923,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.018298170698727002,
      "loss": 0.2288,
      "num_input_tokens_seen": 18667296,
      "step": 88445
    },
    {
      "epoch": 9.73047304730473,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.018296765884478747,
      "loss": 0.2298,
      "num_input_tokens_seen": 18668384,
      "step": 88450
    },
    {
      "epoch": 9.731023102310232,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.018295361039845782,
      "loss": 0.2351,
      "num_input_tokens_seen": 18669504,
      "step": 88455
    },
    {
      "epoch": 9.731573157315731,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.018293956164841054,
      "loss": 0.2319,
      "num_input_tokens_seen": 18670560,
      "step": 88460
    },
    {
      "epoch": 9.732123212321232,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.018292551259477517,
      "loss": 0.2329,
      "num_input_tokens_seen": 18671616,
      "step": 88465
    },
    {
      "epoch": 9.732673267326733,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.018291146323768114,
      "loss": 0.235,
      "num_input_tokens_seen": 18672640,
      "step": 88470
    },
    {
      "epoch": 9.733223322332233,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.01828974135772579,
      "loss": 0.2298,
      "num_input_tokens_seen": 18673728,
      "step": 88475
    },
    {
      "epoch": 9.733773377337734,
      "grad_norm": 0.00048065185546875,
      "learning_rate": 0.018288336361363507,
      "loss": 0.2324,
      "num_input_tokens_seen": 18674752,
      "step": 88480
    },
    {
      "epoch": 9.734323432343235,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0182869313346942,
      "loss": 0.2319,
      "num_input_tokens_seen": 18675776,
      "step": 88485
    },
    {
      "epoch": 9.734873487348734,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.018285526277730824,
      "loss": 0.2298,
      "num_input_tokens_seen": 18676800,
      "step": 88490
    },
    {
      "epoch": 9.735423542354235,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.018284121190486332,
      "loss": 0.2308,
      "num_input_tokens_seen": 18677920,
      "step": 88495
    },
    {
      "epoch": 9.735973597359736,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.018282716072973663,
      "loss": 0.2303,
      "num_input_tokens_seen": 18679040,
      "step": 88500
    },
    {
      "epoch": 9.736523652365236,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.018281310925205782,
      "loss": 0.2308,
      "num_input_tokens_seen": 18680064,
      "step": 88505
    },
    {
      "epoch": 9.737073707370737,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.018279905747195634,
      "loss": 0.2303,
      "num_input_tokens_seen": 18681120,
      "step": 88510
    },
    {
      "epoch": 9.737623762376238,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.018278500538956167,
      "loss": 0.2308,
      "num_input_tokens_seen": 18682208,
      "step": 88515
    },
    {
      "epoch": 9.738173817381739,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.018277095300500333,
      "loss": 0.2329,
      "num_input_tokens_seen": 18683232,
      "step": 88520
    },
    {
      "epoch": 9.738723872387238,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.018275690031841086,
      "loss": 0.2314,
      "num_input_tokens_seen": 18684224,
      "step": 88525
    },
    {
      "epoch": 9.73927392739274,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.01827428473299137,
      "loss": 0.2314,
      "num_input_tokens_seen": 18685248,
      "step": 88530
    },
    {
      "epoch": 9.73982398239824,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.018272879403964146,
      "loss": 0.2309,
      "num_input_tokens_seen": 18686240,
      "step": 88535
    },
    {
      "epoch": 9.74037403740374,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01827147404477237,
      "loss": 0.2298,
      "num_input_tokens_seen": 18687232,
      "step": 88540
    },
    {
      "epoch": 9.74092409240924,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.018270068655428975,
      "loss": 0.2319,
      "num_input_tokens_seen": 18688288,
      "step": 88545
    },
    {
      "epoch": 9.741474147414742,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.018268663235946933,
      "loss": 0.2314,
      "num_input_tokens_seen": 18689408,
      "step": 88550
    },
    {
      "epoch": 9.742024202420241,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01826725778633919,
      "loss": 0.2319,
      "num_input_tokens_seen": 18690400,
      "step": 88555
    },
    {
      "epoch": 9.742574257425742,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0182658523066187,
      "loss": 0.2314,
      "num_input_tokens_seen": 18691488,
      "step": 88560
    },
    {
      "epoch": 9.743124312431243,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.018264446796798416,
      "loss": 0.2303,
      "num_input_tokens_seen": 18692576,
      "step": 88565
    },
    {
      "epoch": 9.743674367436743,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.018263041256891294,
      "loss": 0.2293,
      "num_input_tokens_seen": 18693664,
      "step": 88570
    },
    {
      "epoch": 9.744224422442244,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01826163568691028,
      "loss": 0.2309,
      "num_input_tokens_seen": 18694688,
      "step": 88575
    },
    {
      "epoch": 9.744774477447745,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01826023008686834,
      "loss": 0.2314,
      "num_input_tokens_seen": 18695712,
      "step": 88580
    },
    {
      "epoch": 9.745324532453246,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.018258824456778427,
      "loss": 0.2324,
      "num_input_tokens_seen": 18696832,
      "step": 88585
    },
    {
      "epoch": 9.745874587458745,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.018257418796653487,
      "loss": 0.2314,
      "num_input_tokens_seen": 18697920,
      "step": 88590
    },
    {
      "epoch": 9.746424642464246,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.018256013106506482,
      "loss": 0.2303,
      "num_input_tokens_seen": 18698944,
      "step": 88595
    },
    {
      "epoch": 9.746974697469748,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.01825460738635037,
      "loss": 0.2324,
      "num_input_tokens_seen": 18699968,
      "step": 88600
    },
    {
      "epoch": 9.747524752475247,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.018253201636198098,
      "loss": 0.2314,
      "num_input_tokens_seen": 18700928,
      "step": 88605
    },
    {
      "epoch": 9.748074807480748,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.018251795856062635,
      "loss": 0.2298,
      "num_input_tokens_seen": 18702016,
      "step": 88610
    },
    {
      "epoch": 9.748624862486249,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.018250390045956928,
      "loss": 0.2314,
      "num_input_tokens_seen": 18703040,
      "step": 88615
    },
    {
      "epoch": 9.749174917491748,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01824898420589393,
      "loss": 0.2319,
      "num_input_tokens_seen": 18704032,
      "step": 88620
    },
    {
      "epoch": 9.74972497249725,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.018247578335886607,
      "loss": 0.2314,
      "num_input_tokens_seen": 18705088,
      "step": 88625
    },
    {
      "epoch": 9.75027502750275,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01824617243594791,
      "loss": 0.233,
      "num_input_tokens_seen": 18706176,
      "step": 88630
    },
    {
      "epoch": 9.750825082508252,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.018244766506090804,
      "loss": 0.2314,
      "num_input_tokens_seen": 18707200,
      "step": 88635
    },
    {
      "epoch": 9.751375137513751,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.018243360546328243,
      "loss": 0.2319,
      "num_input_tokens_seen": 18708192,
      "step": 88640
    },
    {
      "epoch": 9.751925192519252,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01824195455667318,
      "loss": 0.2309,
      "num_input_tokens_seen": 18709248,
      "step": 88645
    },
    {
      "epoch": 9.752475247524753,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.018240548537138575,
      "loss": 0.2319,
      "num_input_tokens_seen": 18710304,
      "step": 88650
    },
    {
      "epoch": 9.753025302530252,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.018239142487737395,
      "loss": 0.2314,
      "num_input_tokens_seen": 18711392,
      "step": 88655
    },
    {
      "epoch": 9.753575357535754,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.01823773640848259,
      "loss": 0.2319,
      "num_input_tokens_seen": 18712480,
      "step": 88660
    },
    {
      "epoch": 9.754125412541255,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.018236330299387125,
      "loss": 0.2309,
      "num_input_tokens_seen": 18713536,
      "step": 88665
    },
    {
      "epoch": 9.754675467546754,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.018234924160463954,
      "loss": 0.2304,
      "num_input_tokens_seen": 18714560,
      "step": 88670
    },
    {
      "epoch": 9.755225522552255,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.018233517991726043,
      "loss": 0.2346,
      "num_input_tokens_seen": 18715680,
      "step": 88675
    },
    {
      "epoch": 9.755775577557756,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.018232111793186345,
      "loss": 0.2309,
      "num_input_tokens_seen": 18716736,
      "step": 88680
    },
    {
      "epoch": 9.756325632563257,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.018230705564857824,
      "loss": 0.2314,
      "num_input_tokens_seen": 18717824,
      "step": 88685
    },
    {
      "epoch": 9.756875687568757,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.018229299306753436,
      "loss": 0.232,
      "num_input_tokens_seen": 18718912,
      "step": 88690
    },
    {
      "epoch": 9.757425742574258,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.018227893018886148,
      "loss": 0.2298,
      "num_input_tokens_seen": 18719968,
      "step": 88695
    },
    {
      "epoch": 9.757975797579759,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.01822648670126892,
      "loss": 0.233,
      "num_input_tokens_seen": 18721024,
      "step": 88700
    },
    {
      "epoch": 9.758525852585258,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.018225080353914715,
      "loss": 0.232,
      "num_input_tokens_seen": 18722112,
      "step": 88705
    },
    {
      "epoch": 9.75907590759076,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.018223673976836493,
      "loss": 0.2314,
      "num_input_tokens_seen": 18723168,
      "step": 88710
    },
    {
      "epoch": 9.75962596259626,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.01822226757004721,
      "loss": 0.2309,
      "num_input_tokens_seen": 18724224,
      "step": 88715
    },
    {
      "epoch": 9.76017601760176,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01822086113355983,
      "loss": 0.2303,
      "num_input_tokens_seen": 18725248,
      "step": 88720
    },
    {
      "epoch": 9.76072607260726,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.018219454667387325,
      "loss": 0.2298,
      "num_input_tokens_seen": 18726272,
      "step": 88725
    },
    {
      "epoch": 9.761276127612762,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.018218048171542652,
      "loss": 0.2319,
      "num_input_tokens_seen": 18727232,
      "step": 88730
    },
    {
      "epoch": 9.761826182618261,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.018216641646038766,
      "loss": 0.2293,
      "num_input_tokens_seen": 18728256,
      "step": 88735
    },
    {
      "epoch": 9.762376237623762,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.018215235090888644,
      "loss": 0.2309,
      "num_input_tokens_seen": 18729376,
      "step": 88740
    },
    {
      "epoch": 9.762926292629263,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.01821382850610524,
      "loss": 0.2314,
      "num_input_tokens_seen": 18730528,
      "step": 88745
    },
    {
      "epoch": 9.763476347634764,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01821242189170152,
      "loss": 0.2288,
      "num_input_tokens_seen": 18731584,
      "step": 88750
    },
    {
      "epoch": 9.764026402640264,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.018211015247690455,
      "loss": 0.2356,
      "num_input_tokens_seen": 18732608,
      "step": 88755
    },
    {
      "epoch": 9.764576457645765,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.018209608574084997,
      "loss": 0.2319,
      "num_input_tokens_seen": 18733664,
      "step": 88760
    },
    {
      "epoch": 9.765126512651266,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.018208201870898118,
      "loss": 0.2319,
      "num_input_tokens_seen": 18734656,
      "step": 88765
    },
    {
      "epoch": 9.765676567656765,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01820679513814278,
      "loss": 0.2319,
      "num_input_tokens_seen": 18735680,
      "step": 88770
    },
    {
      "epoch": 9.766226622662266,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.018205388375831955,
      "loss": 0.2308,
      "num_input_tokens_seen": 18736704,
      "step": 88775
    },
    {
      "epoch": 9.766776677667767,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.018203981583978603,
      "loss": 0.2293,
      "num_input_tokens_seen": 18737792,
      "step": 88780
    },
    {
      "epoch": 9.767326732673267,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.018202574762595687,
      "loss": 0.2319,
      "num_input_tokens_seen": 18738912,
      "step": 88785
    },
    {
      "epoch": 9.767876787678768,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.018201167911696178,
      "loss": 0.2308,
      "num_input_tokens_seen": 18740000,
      "step": 88790
    },
    {
      "epoch": 9.768426842684269,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.018199761031293036,
      "loss": 0.2303,
      "num_input_tokens_seen": 18741024,
      "step": 88795
    },
    {
      "epoch": 9.768976897689768,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.018198354121399238,
      "loss": 0.2319,
      "num_input_tokens_seen": 18742080,
      "step": 88800
    },
    {
      "epoch": 9.76952695269527,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.01819694718202774,
      "loss": 0.2325,
      "num_input_tokens_seen": 18743072,
      "step": 88805
    },
    {
      "epoch": 9.77007700770077,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.018195540213191516,
      "loss": 0.2304,
      "num_input_tokens_seen": 18744192,
      "step": 88810
    },
    {
      "epoch": 9.770627062706271,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.018194133214903534,
      "loss": 0.2319,
      "num_input_tokens_seen": 18745216,
      "step": 88815
    },
    {
      "epoch": 9.77117711771177,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.018192726187176756,
      "loss": 0.2298,
      "num_input_tokens_seen": 18746208,
      "step": 88820
    },
    {
      "epoch": 9.771727172717272,
      "grad_norm": 0.009765625,
      "learning_rate": 0.018191319130024156,
      "loss": 0.2303,
      "num_input_tokens_seen": 18747264,
      "step": 88825
    },
    {
      "epoch": 9.772277227722773,
      "grad_norm": 0.000579833984375,
      "learning_rate": 0.018189912043458694,
      "loss": 0.2288,
      "num_input_tokens_seen": 18748352,
      "step": 88830
    },
    {
      "epoch": 9.772827282728272,
      "grad_norm": 0.00286865234375,
      "learning_rate": 0.01818850492749334,
      "loss": 0.232,
      "num_input_tokens_seen": 18749344,
      "step": 88835
    },
    {
      "epoch": 9.773377337733773,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.018187097782141073,
      "loss": 0.2288,
      "num_input_tokens_seen": 18750368,
      "step": 88840
    },
    {
      "epoch": 9.773927392739274,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.018185690607414852,
      "loss": 0.2299,
      "num_input_tokens_seen": 18751392,
      "step": 88845
    },
    {
      "epoch": 9.774477447744774,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.018184283403327655,
      "loss": 0.233,
      "num_input_tokens_seen": 18752480,
      "step": 88850
    },
    {
      "epoch": 9.775027502750275,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.01818287616989244,
      "loss": 0.2309,
      "num_input_tokens_seen": 18753568,
      "step": 88855
    },
    {
      "epoch": 9.775577557755776,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.018181468907122186,
      "loss": 0.2298,
      "num_input_tokens_seen": 18754560,
      "step": 88860
    },
    {
      "epoch": 9.776127612761275,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.018180061615029854,
      "loss": 0.233,
      "num_input_tokens_seen": 18755712,
      "step": 88865
    },
    {
      "epoch": 9.776677667766776,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.01817865429362843,
      "loss": 0.2304,
      "num_input_tokens_seen": 18756736,
      "step": 88870
    },
    {
      "epoch": 9.777227722772277,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.018177246942930866,
      "loss": 0.2304,
      "num_input_tokens_seen": 18757856,
      "step": 88875
    },
    {
      "epoch": 9.777777777777779,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.018175839562950144,
      "loss": 0.2319,
      "num_input_tokens_seen": 18758976,
      "step": 88880
    },
    {
      "epoch": 9.778327832783278,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01817443215369923,
      "loss": 0.2319,
      "num_input_tokens_seen": 18760032,
      "step": 88885
    },
    {
      "epoch": 9.778877887788779,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.018173024715191104,
      "loss": 0.2303,
      "num_input_tokens_seen": 18761152,
      "step": 88890
    },
    {
      "epoch": 9.77942794279428,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.018171617247438727,
      "loss": 0.2324,
      "num_input_tokens_seen": 18762272,
      "step": 88895
    },
    {
      "epoch": 9.77997799779978,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.018170209750455076,
      "loss": 0.2309,
      "num_input_tokens_seen": 18763328,
      "step": 88900
    },
    {
      "epoch": 9.78052805280528,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.018168802224253128,
      "loss": 0.2324,
      "num_input_tokens_seen": 18764384,
      "step": 88905
    },
    {
      "epoch": 9.781078107810782,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.018167394668845843,
      "loss": 0.2309,
      "num_input_tokens_seen": 18765440,
      "step": 88910
    },
    {
      "epoch": 9.781628162816281,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.01816598708424621,
      "loss": 0.2309,
      "num_input_tokens_seen": 18766528,
      "step": 88915
    },
    {
      "epoch": 9.782178217821782,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.018164579470467186,
      "loss": 0.2298,
      "num_input_tokens_seen": 18767552,
      "step": 88920
    },
    {
      "epoch": 9.782728272827283,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.01816317182752175,
      "loss": 0.2324,
      "num_input_tokens_seen": 18768576,
      "step": 88925
    },
    {
      "epoch": 9.783278327832782,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.018161764155422883,
      "loss": 0.2303,
      "num_input_tokens_seen": 18769600,
      "step": 88930
    },
    {
      "epoch": 9.783828382838283,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.01816035645418355,
      "loss": 0.2314,
      "num_input_tokens_seen": 18770656,
      "step": 88935
    },
    {
      "epoch": 9.784378437843785,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01815894872381673,
      "loss": 0.2298,
      "num_input_tokens_seen": 18771712,
      "step": 88940
    },
    {
      "epoch": 9.784928492849286,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.018157540964335396,
      "loss": 0.2293,
      "num_input_tokens_seen": 18772832,
      "step": 88945
    },
    {
      "epoch": 9.785478547854785,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01815613317575252,
      "loss": 0.2267,
      "num_input_tokens_seen": 18773920,
      "step": 88950
    },
    {
      "epoch": 9.786028602860286,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01815472535808108,
      "loss": 0.2335,
      "num_input_tokens_seen": 18774912,
      "step": 88955
    },
    {
      "epoch": 9.786578657865787,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.018153317511334047,
      "loss": 0.2314,
      "num_input_tokens_seen": 18775968,
      "step": 88960
    },
    {
      "epoch": 9.787128712871286,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.0181519096355244,
      "loss": 0.2335,
      "num_input_tokens_seen": 18777024,
      "step": 88965
    },
    {
      "epoch": 9.787678767876788,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.01815050173066512,
      "loss": 0.234,
      "num_input_tokens_seen": 18778112,
      "step": 88970
    },
    {
      "epoch": 9.788228822882289,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.018149093796769173,
      "loss": 0.2308,
      "num_input_tokens_seen": 18779200,
      "step": 88975
    },
    {
      "epoch": 9.788778877887788,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01814768583384954,
      "loss": 0.2283,
      "num_input_tokens_seen": 18780352,
      "step": 88980
    },
    {
      "epoch": 9.789328932893289,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.018146277841919193,
      "loss": 0.2324,
      "num_input_tokens_seen": 18781472,
      "step": 88985
    },
    {
      "epoch": 9.78987898789879,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.018144869820991123,
      "loss": 0.2314,
      "num_input_tokens_seen": 18782464,
      "step": 88990
    },
    {
      "epoch": 9.79042904290429,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.018143461771078288,
      "loss": 0.235,
      "num_input_tokens_seen": 18783552,
      "step": 88995
    },
    {
      "epoch": 9.79097909790979,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.01814205369219368,
      "loss": 0.2293,
      "num_input_tokens_seen": 18784640,
      "step": 89000
    },
    {
      "epoch": 9.791529152915292,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01814064558435027,
      "loss": 0.2309,
      "num_input_tokens_seen": 18785696,
      "step": 89005
    },
    {
      "epoch": 9.792079207920793,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.018139237447561032,
      "loss": 0.2303,
      "num_input_tokens_seen": 18786784,
      "step": 89010
    },
    {
      "epoch": 9.792629262926292,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.018137829281838953,
      "loss": 0.2319,
      "num_input_tokens_seen": 18787872,
      "step": 89015
    },
    {
      "epoch": 9.793179317931793,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.018136421087197006,
      "loss": 0.2319,
      "num_input_tokens_seen": 18788960,
      "step": 89020
    },
    {
      "epoch": 9.793729372937294,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.018135012863648166,
      "loss": 0.2309,
      "num_input_tokens_seen": 18790016,
      "step": 89025
    },
    {
      "epoch": 9.794279427942794,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01813360461120542,
      "loss": 0.2329,
      "num_input_tokens_seen": 18791040,
      "step": 89030
    },
    {
      "epoch": 9.794829482948295,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.01813219632988175,
      "loss": 0.2304,
      "num_input_tokens_seen": 18792128,
      "step": 89035
    },
    {
      "epoch": 9.795379537953796,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.018130788019690126,
      "loss": 0.2304,
      "num_input_tokens_seen": 18793088,
      "step": 89040
    },
    {
      "epoch": 9.795929592959295,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.01812937968064353,
      "loss": 0.2329,
      "num_input_tokens_seen": 18794208,
      "step": 89045
    },
    {
      "epoch": 9.796479647964796,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.018127971312754943,
      "loss": 0.2319,
      "num_input_tokens_seen": 18795296,
      "step": 89050
    },
    {
      "epoch": 9.797029702970297,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.018126562916037342,
      "loss": 0.2308,
      "num_input_tokens_seen": 18796320,
      "step": 89055
    },
    {
      "epoch": 9.797579757975798,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.018125154490503715,
      "loss": 0.2303,
      "num_input_tokens_seen": 18797440,
      "step": 89060
    },
    {
      "epoch": 9.798129812981298,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01812374603616704,
      "loss": 0.2324,
      "num_input_tokens_seen": 18798464,
      "step": 89065
    },
    {
      "epoch": 9.798679867986799,
      "grad_norm": 0.0029144287109375,
      "learning_rate": 0.018122337553040293,
      "loss": 0.2298,
      "num_input_tokens_seen": 18799488,
      "step": 89070
    },
    {
      "epoch": 9.7992299229923,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.01812092904113646,
      "loss": 0.2309,
      "num_input_tokens_seen": 18800544,
      "step": 89075
    },
    {
      "epoch": 9.7997799779978,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.01811952050046852,
      "loss": 0.2324,
      "num_input_tokens_seen": 18801568,
      "step": 89080
    },
    {
      "epoch": 9.8003300330033,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.018118111931049462,
      "loss": 0.2319,
      "num_input_tokens_seen": 18802688,
      "step": 89085
    },
    {
      "epoch": 9.800880088008801,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.01811670333289226,
      "loss": 0.2314,
      "num_input_tokens_seen": 18803776,
      "step": 89090
    },
    {
      "epoch": 9.8014301430143,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.018115294706009897,
      "loss": 0.2319,
      "num_input_tokens_seen": 18804768,
      "step": 89095
    },
    {
      "epoch": 9.801980198019802,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01811388605041536,
      "loss": 0.2314,
      "num_input_tokens_seen": 18805856,
      "step": 89100
    },
    {
      "epoch": 9.802530253025303,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.018112477366121624,
      "loss": 0.2319,
      "num_input_tokens_seen": 18806880,
      "step": 89105
    },
    {
      "epoch": 9.803080308030804,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.018111068653141684,
      "loss": 0.2293,
      "num_input_tokens_seen": 18807936,
      "step": 89110
    },
    {
      "epoch": 9.803630363036303,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.018109659911488513,
      "loss": 0.2303,
      "num_input_tokens_seen": 18808896,
      "step": 89115
    },
    {
      "epoch": 9.804180418041804,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.0181082511411751,
      "loss": 0.2324,
      "num_input_tokens_seen": 18809952,
      "step": 89120
    },
    {
      "epoch": 9.804730473047305,
      "grad_norm": 0.009765625,
      "learning_rate": 0.018106842342214432,
      "loss": 0.2319,
      "num_input_tokens_seen": 18811008,
      "step": 89125
    },
    {
      "epoch": 9.805280528052805,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.018105433514619483,
      "loss": 0.2329,
      "num_input_tokens_seen": 18812064,
      "step": 89130
    },
    {
      "epoch": 9.805830583058306,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01810402465840325,
      "loss": 0.2303,
      "num_input_tokens_seen": 18813088,
      "step": 89135
    },
    {
      "epoch": 9.806380638063807,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.0181026157735787,
      "loss": 0.2335,
      "num_input_tokens_seen": 18814080,
      "step": 89140
    },
    {
      "epoch": 9.806930693069306,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.018101206860158838,
      "loss": 0.2313,
      "num_input_tokens_seen": 18815104,
      "step": 89145
    },
    {
      "epoch": 9.807480748074807,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01809979791815664,
      "loss": 0.2319,
      "num_input_tokens_seen": 18816160,
      "step": 89150
    },
    {
      "epoch": 9.808030803080309,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01809838894758509,
      "loss": 0.233,
      "num_input_tokens_seen": 18817216,
      "step": 89155
    },
    {
      "epoch": 9.808580858085808,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01809697994845718,
      "loss": 0.2324,
      "num_input_tokens_seen": 18818304,
      "step": 89160
    },
    {
      "epoch": 9.809130913091309,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.018095570920785887,
      "loss": 0.2319,
      "num_input_tokens_seen": 18819360,
      "step": 89165
    },
    {
      "epoch": 9.80968096809681,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.0180941618645842,
      "loss": 0.2324,
      "num_input_tokens_seen": 18820448,
      "step": 89170
    },
    {
      "epoch": 9.810231023102311,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.01809275277986511,
      "loss": 0.2314,
      "num_input_tokens_seen": 18821472,
      "step": 89175
    },
    {
      "epoch": 9.81078107810781,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01809134366664161,
      "loss": 0.234,
      "num_input_tokens_seen": 18822496,
      "step": 89180
    },
    {
      "epoch": 9.811331133113312,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.018089934524926667,
      "loss": 0.2314,
      "num_input_tokens_seen": 18823552,
      "step": 89185
    },
    {
      "epoch": 9.811881188118813,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.018088525354733286,
      "loss": 0.2319,
      "num_input_tokens_seen": 18824576,
      "step": 89190
    },
    {
      "epoch": 9.812431243124312,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.01808711615607445,
      "loss": 0.2303,
      "num_input_tokens_seen": 18825664,
      "step": 89195
    },
    {
      "epoch": 9.812981298129813,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.018085706928963136,
      "loss": 0.2309,
      "num_input_tokens_seen": 18826688,
      "step": 89200
    },
    {
      "epoch": 9.813531353135314,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.018084297673412356,
      "loss": 0.2308,
      "num_input_tokens_seen": 18827680,
      "step": 89205
    },
    {
      "epoch": 9.814081408140813,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.018082888389435076,
      "loss": 0.233,
      "num_input_tokens_seen": 18828736,
      "step": 89210
    },
    {
      "epoch": 9.814631463146315,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.018081479077044296,
      "loss": 0.2324,
      "num_input_tokens_seen": 18829760,
      "step": 89215
    },
    {
      "epoch": 9.815181518151816,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.018080069736253,
      "loss": 0.2319,
      "num_input_tokens_seen": 18830784,
      "step": 89220
    },
    {
      "epoch": 9.815731573157315,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.01807866036707418,
      "loss": 0.2304,
      "num_input_tokens_seen": 18831872,
      "step": 89225
    },
    {
      "epoch": 9.816281628162816,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.018077250969520824,
      "loss": 0.2309,
      "num_input_tokens_seen": 18832864,
      "step": 89230
    },
    {
      "epoch": 9.816831683168317,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.018075841543605923,
      "loss": 0.2309,
      "num_input_tokens_seen": 18833888,
      "step": 89235
    },
    {
      "epoch": 9.817381738173818,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.018074432089342465,
      "loss": 0.2325,
      "num_input_tokens_seen": 18834912,
      "step": 89240
    },
    {
      "epoch": 9.817931793179318,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.01807302260674344,
      "loss": 0.2319,
      "num_input_tokens_seen": 18836000,
      "step": 89245
    },
    {
      "epoch": 9.818481848184819,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.018071613095821844,
      "loss": 0.2309,
      "num_input_tokens_seen": 18837120,
      "step": 89250
    },
    {
      "epoch": 9.81903190319032,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01807020355659067,
      "loss": 0.2314,
      "num_input_tokens_seen": 18838176,
      "step": 89255
    },
    {
      "epoch": 9.819581958195819,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01806879398906289,
      "loss": 0.2319,
      "num_input_tokens_seen": 18839264,
      "step": 89260
    },
    {
      "epoch": 9.82013201320132,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.018067384393251517,
      "loss": 0.2278,
      "num_input_tokens_seen": 18840288,
      "step": 89265
    },
    {
      "epoch": 9.820682068206821,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.018065974769169532,
      "loss": 0.2314,
      "num_input_tokens_seen": 18841280,
      "step": 89270
    },
    {
      "epoch": 9.82123212321232,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.01806456511682993,
      "loss": 0.2351,
      "num_input_tokens_seen": 18842336,
      "step": 89275
    },
    {
      "epoch": 9.821782178217822,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.0180631554362457,
      "loss": 0.2319,
      "num_input_tokens_seen": 18843392,
      "step": 89280
    },
    {
      "epoch": 9.822332233223323,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.018061745727429836,
      "loss": 0.2329,
      "num_input_tokens_seen": 18844416,
      "step": 89285
    },
    {
      "epoch": 9.822882288228822,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.018060335990395332,
      "loss": 0.2314,
      "num_input_tokens_seen": 18845472,
      "step": 89290
    },
    {
      "epoch": 9.823432343234323,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.018058926225155175,
      "loss": 0.2298,
      "num_input_tokens_seen": 18846496,
      "step": 89295
    },
    {
      "epoch": 9.823982398239824,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.01805751643172237,
      "loss": 0.2319,
      "num_input_tokens_seen": 18847552,
      "step": 89300
    },
    {
      "epoch": 9.824532453245325,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.0180561066101099,
      "loss": 0.2329,
      "num_input_tokens_seen": 18848608,
      "step": 89305
    },
    {
      "epoch": 9.825082508250825,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.018054696760330762,
      "loss": 0.2314,
      "num_input_tokens_seen": 18849664,
      "step": 89310
    },
    {
      "epoch": 9.825632563256326,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.018053286882397948,
      "loss": 0.2319,
      "num_input_tokens_seen": 18850656,
      "step": 89315
    },
    {
      "epoch": 9.826182618261827,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.018051876976324457,
      "loss": 0.2298,
      "num_input_tokens_seen": 18851712,
      "step": 89320
    },
    {
      "epoch": 9.826732673267326,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.018050467042123278,
      "loss": 0.2303,
      "num_input_tokens_seen": 18852736,
      "step": 89325
    },
    {
      "epoch": 9.827282728272827,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.018049057079807408,
      "loss": 0.2309,
      "num_input_tokens_seen": 18853856,
      "step": 89330
    },
    {
      "epoch": 9.827832783278328,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.018047647089389843,
      "loss": 0.2314,
      "num_input_tokens_seen": 18854912,
      "step": 89335
    },
    {
      "epoch": 9.828382838283828,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.018046237070883576,
      "loss": 0.2314,
      "num_input_tokens_seen": 18855968,
      "step": 89340
    },
    {
      "epoch": 9.828932893289329,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.018044827024301602,
      "loss": 0.2314,
      "num_input_tokens_seen": 18857024,
      "step": 89345
    },
    {
      "epoch": 9.82948294829483,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.018043416949656926,
      "loss": 0.2314,
      "num_input_tokens_seen": 18858080,
      "step": 89350
    },
    {
      "epoch": 9.83003300330033,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.01804200684696253,
      "loss": 0.2303,
      "num_input_tokens_seen": 18859168,
      "step": 89355
    },
    {
      "epoch": 9.83058305830583,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.018040596716231417,
      "loss": 0.2304,
      "num_input_tokens_seen": 18860192,
      "step": 89360
    },
    {
      "epoch": 9.831133113311331,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.018039186557476584,
      "loss": 0.2298,
      "num_input_tokens_seen": 18861216,
      "step": 89365
    },
    {
      "epoch": 9.831683168316832,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01803777637071103,
      "loss": 0.2303,
      "num_input_tokens_seen": 18862304,
      "step": 89370
    },
    {
      "epoch": 9.832233223322332,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.018036366155947742,
      "loss": 0.2324,
      "num_input_tokens_seen": 18863296,
      "step": 89375
    },
    {
      "epoch": 9.832783278327833,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.018034955913199726,
      "loss": 0.2335,
      "num_input_tokens_seen": 18864352,
      "step": 89380
    },
    {
      "epoch": 9.833333333333334,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.018033545642479984,
      "loss": 0.2309,
      "num_input_tokens_seen": 18865376,
      "step": 89385
    },
    {
      "epoch": 9.833883388338833,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.0180321353438015,
      "loss": 0.2303,
      "num_input_tokens_seen": 18866368,
      "step": 89390
    },
    {
      "epoch": 9.834433443344334,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.018030725017177285,
      "loss": 0.2319,
      "num_input_tokens_seen": 18867392,
      "step": 89395
    },
    {
      "epoch": 9.834983498349835,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.018029314662620328,
      "loss": 0.2335,
      "num_input_tokens_seen": 18868416,
      "step": 89400
    },
    {
      "epoch": 9.835533553355335,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.01802790428014363,
      "loss": 0.2319,
      "num_input_tokens_seen": 18869472,
      "step": 89405
    },
    {
      "epoch": 9.836083608360836,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.018026493869760195,
      "loss": 0.2298,
      "num_input_tokens_seen": 18870528,
      "step": 89410
    },
    {
      "epoch": 9.836633663366337,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.018025083431483017,
      "loss": 0.2309,
      "num_input_tokens_seen": 18871584,
      "step": 89415
    },
    {
      "epoch": 9.837183718371836,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.0180236729653251,
      "loss": 0.2325,
      "num_input_tokens_seen": 18872608,
      "step": 89420
    },
    {
      "epoch": 9.837733773377337,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.018022262471299434,
      "loss": 0.2324,
      "num_input_tokens_seen": 18873664,
      "step": 89425
    },
    {
      "epoch": 9.838283828382838,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.018020851949419027,
      "loss": 0.2314,
      "num_input_tokens_seen": 18874720,
      "step": 89430
    },
    {
      "epoch": 9.83883388338834,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.018019441399696877,
      "loss": 0.234,
      "num_input_tokens_seen": 18875744,
      "step": 89435
    },
    {
      "epoch": 9.839383938393839,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.018018030822145984,
      "loss": 0.2309,
      "num_input_tokens_seen": 18876768,
      "step": 89440
    },
    {
      "epoch": 9.83993399339934,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.018016620216779355,
      "loss": 0.2288,
      "num_input_tokens_seen": 18877856,
      "step": 89445
    },
    {
      "epoch": 9.840484048404841,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.01801520958360998,
      "loss": 0.2303,
      "num_input_tokens_seen": 18878912,
      "step": 89450
    },
    {
      "epoch": 9.84103410341034,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.018013798922650866,
      "loss": 0.2304,
      "num_input_tokens_seen": 18879968,
      "step": 89455
    },
    {
      "epoch": 9.841584158415841,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.018012388233915013,
      "loss": 0.2309,
      "num_input_tokens_seen": 18881056,
      "step": 89460
    },
    {
      "epoch": 9.842134213421343,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.018010977517415424,
      "loss": 0.2329,
      "num_input_tokens_seen": 18882048,
      "step": 89465
    },
    {
      "epoch": 9.842684268426842,
      "grad_norm": 0.009765625,
      "learning_rate": 0.0180095667731651,
      "loss": 0.2309,
      "num_input_tokens_seen": 18883136,
      "step": 89470
    },
    {
      "epoch": 9.843234323432343,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01800815600117704,
      "loss": 0.2319,
      "num_input_tokens_seen": 18884128,
      "step": 89475
    },
    {
      "epoch": 9.843784378437844,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01800674520146425,
      "loss": 0.2309,
      "num_input_tokens_seen": 18885216,
      "step": 89480
    },
    {
      "epoch": 9.844334433443345,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.018005334374039735,
      "loss": 0.2309,
      "num_input_tokens_seen": 18886208,
      "step": 89485
    },
    {
      "epoch": 9.844884488448844,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.018003923518916495,
      "loss": 0.2309,
      "num_input_tokens_seen": 18887232,
      "step": 89490
    },
    {
      "epoch": 9.845434543454346,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.018002512636107536,
      "loss": 0.2324,
      "num_input_tokens_seen": 18888256,
      "step": 89495
    },
    {
      "epoch": 9.845984598459847,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.018001101725625853,
      "loss": 0.2324,
      "num_input_tokens_seen": 18889248,
      "step": 89500
    },
    {
      "epoch": 9.846534653465346,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.017999690787484455,
      "loss": 0.2308,
      "num_input_tokens_seen": 18890240,
      "step": 89505
    },
    {
      "epoch": 9.847084708470847,
      "grad_norm": 0.00049591064453125,
      "learning_rate": 0.01799827982169635,
      "loss": 0.2329,
      "num_input_tokens_seen": 18891264,
      "step": 89510
    },
    {
      "epoch": 9.847634763476348,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.017996868828274545,
      "loss": 0.2298,
      "num_input_tokens_seen": 18892288,
      "step": 89515
    },
    {
      "epoch": 9.848184818481847,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.017995457807232027,
      "loss": 0.2309,
      "num_input_tokens_seen": 18893408,
      "step": 89520
    },
    {
      "epoch": 9.848734873487349,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.017994046758581815,
      "loss": 0.2324,
      "num_input_tokens_seen": 18894464,
      "step": 89525
    },
    {
      "epoch": 9.84928492849285,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.017992635682336913,
      "loss": 0.2304,
      "num_input_tokens_seen": 18895552,
      "step": 89530
    },
    {
      "epoch": 9.84983498349835,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01799122457851032,
      "loss": 0.2309,
      "num_input_tokens_seen": 18896640,
      "step": 89535
    },
    {
      "epoch": 9.85038503850385,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01798981344711505,
      "loss": 0.2283,
      "num_input_tokens_seen": 18897696,
      "step": 89540
    },
    {
      "epoch": 9.850935093509351,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.017988402288164104,
      "loss": 0.2314,
      "num_input_tokens_seen": 18898720,
      "step": 89545
    },
    {
      "epoch": 9.851485148514852,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.017986991101670482,
      "loss": 0.2314,
      "num_input_tokens_seen": 18899712,
      "step": 89550
    },
    {
      "epoch": 9.852035203520352,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0179855798876472,
      "loss": 0.2324,
      "num_input_tokens_seen": 18900736,
      "step": 89555
    },
    {
      "epoch": 9.852585258525853,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.017984168646107263,
      "loss": 0.2303,
      "num_input_tokens_seen": 18901824,
      "step": 89560
    },
    {
      "epoch": 9.853135313531354,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.017982757377063674,
      "loss": 0.2303,
      "num_input_tokens_seen": 18902816,
      "step": 89565
    },
    {
      "epoch": 9.853685368536853,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.017981346080529443,
      "loss": 0.2329,
      "num_input_tokens_seen": 18903808,
      "step": 89570
    },
    {
      "epoch": 9.854235423542354,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.017979934756517574,
      "loss": 0.2324,
      "num_input_tokens_seen": 18904832,
      "step": 89575
    },
    {
      "epoch": 9.854785478547855,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.017978523405041073,
      "loss": 0.2319,
      "num_input_tokens_seen": 18905888,
      "step": 89580
    },
    {
      "epoch": 9.855335533553355,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.01797711202611296,
      "loss": 0.2309,
      "num_input_tokens_seen": 18906880,
      "step": 89585
    },
    {
      "epoch": 9.855885588558856,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.017975700619746224,
      "loss": 0.2298,
      "num_input_tokens_seen": 18907904,
      "step": 89590
    },
    {
      "epoch": 9.856435643564357,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.01797428918595389,
      "loss": 0.2288,
      "num_input_tokens_seen": 18909024,
      "step": 89595
    },
    {
      "epoch": 9.856985698569858,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.017972877724748954,
      "loss": 0.2304,
      "num_input_tokens_seen": 18910080,
      "step": 89600
    },
    {
      "epoch": 9.857535753575357,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.017971466236144432,
      "loss": 0.2324,
      "num_input_tokens_seen": 18911104,
      "step": 89605
    },
    {
      "epoch": 9.858085808580858,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.017970054720153335,
      "loss": 0.2335,
      "num_input_tokens_seen": 18912096,
      "step": 89610
    },
    {
      "epoch": 9.85863586358636,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.017968643176788666,
      "loss": 0.2314,
      "num_input_tokens_seen": 18913056,
      "step": 89615
    },
    {
      "epoch": 9.859185918591859,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.01796723160606344,
      "loss": 0.2304,
      "num_input_tokens_seen": 18914080,
      "step": 89620
    },
    {
      "epoch": 9.85973597359736,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01796582000799066,
      "loss": 0.233,
      "num_input_tokens_seen": 18915136,
      "step": 89625
    },
    {
      "epoch": 9.86028602860286,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.017964408382583345,
      "loss": 0.2309,
      "num_input_tokens_seen": 18916192,
      "step": 89630
    },
    {
      "epoch": 9.86083608360836,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.017962996729854498,
      "loss": 0.2314,
      "num_input_tokens_seen": 18917280,
      "step": 89635
    },
    {
      "epoch": 9.861386138613861,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.01796158504981713,
      "loss": 0.2293,
      "num_input_tokens_seen": 18918336,
      "step": 89640
    },
    {
      "epoch": 9.861936193619362,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01796017334248426,
      "loss": 0.2314,
      "num_input_tokens_seen": 18919360,
      "step": 89645
    },
    {
      "epoch": 9.862486248624862,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.017958761607868884,
      "loss": 0.2309,
      "num_input_tokens_seen": 18920416,
      "step": 89650
    },
    {
      "epoch": 9.863036303630363,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.017957349845984034,
      "loss": 0.2335,
      "num_input_tokens_seen": 18921472,
      "step": 89655
    },
    {
      "epoch": 9.863586358635864,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0179559380568427,
      "loss": 0.2324,
      "num_input_tokens_seen": 18922528,
      "step": 89660
    },
    {
      "epoch": 9.864136413641365,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.017954526240457902,
      "loss": 0.2329,
      "num_input_tokens_seen": 18923616,
      "step": 89665
    },
    {
      "epoch": 9.864686468646864,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01795311439684266,
      "loss": 0.2298,
      "num_input_tokens_seen": 18924640,
      "step": 89670
    },
    {
      "epoch": 9.865236523652365,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01795170252600997,
      "loss": 0.2319,
      "num_input_tokens_seen": 18925632,
      "step": 89675
    },
    {
      "epoch": 9.865786578657866,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.017950290627972865,
      "loss": 0.233,
      "num_input_tokens_seen": 18926720,
      "step": 89680
    },
    {
      "epoch": 9.866336633663366,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.017948878702744343,
      "loss": 0.2335,
      "num_input_tokens_seen": 18927808,
      "step": 89685
    },
    {
      "epoch": 9.866886688668867,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.017947466750337423,
      "loss": 0.2309,
      "num_input_tokens_seen": 18928832,
      "step": 89690
    },
    {
      "epoch": 9.867436743674368,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.017946054770765107,
      "loss": 0.2309,
      "num_input_tokens_seen": 18929920,
      "step": 89695
    },
    {
      "epoch": 9.867986798679867,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01794464276404043,
      "loss": 0.2314,
      "num_input_tokens_seen": 18930944,
      "step": 89700
    },
    {
      "epoch": 9.868536853685368,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01794323073017639,
      "loss": 0.2314,
      "num_input_tokens_seen": 18932000,
      "step": 89705
    },
    {
      "epoch": 9.86908690869087,
      "grad_norm": 0.000576019287109375,
      "learning_rate": 0.017941818669186,
      "loss": 0.2314,
      "num_input_tokens_seen": 18932992,
      "step": 89710
    },
    {
      "epoch": 9.869636963696369,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.017940406581082282,
      "loss": 0.2314,
      "num_input_tokens_seen": 18934112,
      "step": 89715
    },
    {
      "epoch": 9.87018701870187,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01793899446587825,
      "loss": 0.2309,
      "num_input_tokens_seen": 18935200,
      "step": 89720
    },
    {
      "epoch": 9.870737073707371,
      "grad_norm": 0.009765625,
      "learning_rate": 0.01793758232358691,
      "loss": 0.2335,
      "num_input_tokens_seen": 18936224,
      "step": 89725
    },
    {
      "epoch": 9.871287128712872,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.01793617015422129,
      "loss": 0.2309,
      "num_input_tokens_seen": 18937248,
      "step": 89730
    },
    {
      "epoch": 9.871837183718371,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.017934757957794395,
      "loss": 0.2309,
      "num_input_tokens_seen": 18938304,
      "step": 89735
    },
    {
      "epoch": 9.872387238723872,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.017933345734319244,
      "loss": 0.2319,
      "num_input_tokens_seen": 18939360,
      "step": 89740
    },
    {
      "epoch": 9.872937293729374,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.01793193348380885,
      "loss": 0.2319,
      "num_input_tokens_seen": 18940416,
      "step": 89745
    },
    {
      "epoch": 9.873487348734873,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.017930521206276237,
      "loss": 0.2309,
      "num_input_tokens_seen": 18941632,
      "step": 89750
    },
    {
      "epoch": 9.874037403740374,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.017929108901734415,
      "loss": 0.2309,
      "num_input_tokens_seen": 18942720,
      "step": 89755
    },
    {
      "epoch": 9.874587458745875,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.017927696570196404,
      "loss": 0.2314,
      "num_input_tokens_seen": 18943776,
      "step": 89760
    },
    {
      "epoch": 9.875137513751374,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.017926284211675216,
      "loss": 0.2324,
      "num_input_tokens_seen": 18944896,
      "step": 89765
    },
    {
      "epoch": 9.875687568756875,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01792487182618387,
      "loss": 0.2314,
      "num_input_tokens_seen": 18945920,
      "step": 89770
    },
    {
      "epoch": 9.876237623762377,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01792345941373539,
      "loss": 0.2293,
      "num_input_tokens_seen": 18946976,
      "step": 89775
    },
    {
      "epoch": 9.876787678767876,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.017922046974342777,
      "loss": 0.2298,
      "num_input_tokens_seen": 18948032,
      "step": 89780
    },
    {
      "epoch": 9.877337733773377,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.017920634508019062,
      "loss": 0.2304,
      "num_input_tokens_seen": 18949056,
      "step": 89785
    },
    {
      "epoch": 9.877887788778878,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.017919222014777265,
      "loss": 0.2298,
      "num_input_tokens_seen": 18950048,
      "step": 89790
    },
    {
      "epoch": 9.87843784378438,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0179178094946304,
      "loss": 0.2298,
      "num_input_tokens_seen": 18951072,
      "step": 89795
    },
    {
      "epoch": 9.878987898789878,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01791639694759148,
      "loss": 0.2314,
      "num_input_tokens_seen": 18952192,
      "step": 89800
    },
    {
      "epoch": 9.87953795379538,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01791498437367353,
      "loss": 0.2278,
      "num_input_tokens_seen": 18953280,
      "step": 89805
    },
    {
      "epoch": 9.88008800880088,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.017913571772889568,
      "loss": 0.2356,
      "num_input_tokens_seen": 18954336,
      "step": 89810
    },
    {
      "epoch": 9.88063806380638,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.017912159145252612,
      "loss": 0.2325,
      "num_input_tokens_seen": 18955488,
      "step": 89815
    },
    {
      "epoch": 9.881188118811881,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.017910746490775686,
      "loss": 0.2283,
      "num_input_tokens_seen": 18956576,
      "step": 89820
    },
    {
      "epoch": 9.881738173817382,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.0179093338094718,
      "loss": 0.2314,
      "num_input_tokens_seen": 18957664,
      "step": 89825
    },
    {
      "epoch": 9.882288228822881,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.017907921101353988,
      "loss": 0.2304,
      "num_input_tokens_seen": 18958656,
      "step": 89830
    },
    {
      "epoch": 9.882838283828383,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01790650836643526,
      "loss": 0.2324,
      "num_input_tokens_seen": 18959648,
      "step": 89835
    },
    {
      "epoch": 9.883388338833884,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.017905095604728635,
      "loss": 0.2325,
      "num_input_tokens_seen": 18960640,
      "step": 89840
    },
    {
      "epoch": 9.883938393839383,
      "grad_norm": 0.009765625,
      "learning_rate": 0.017903682816247143,
      "loss": 0.2309,
      "num_input_tokens_seen": 18961696,
      "step": 89845
    },
    {
      "epoch": 9.884488448844884,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.017902270001003803,
      "loss": 0.233,
      "num_input_tokens_seen": 18962720,
      "step": 89850
    },
    {
      "epoch": 9.885038503850385,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.017900857159011622,
      "loss": 0.2299,
      "num_input_tokens_seen": 18963808,
      "step": 89855
    },
    {
      "epoch": 9.885588558855886,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01789944429028364,
      "loss": 0.2283,
      "num_input_tokens_seen": 18964896,
      "step": 89860
    },
    {
      "epoch": 9.886138613861386,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.01789803139483287,
      "loss": 0.2314,
      "num_input_tokens_seen": 18965920,
      "step": 89865
    },
    {
      "epoch": 9.886688668866887,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.017896618472672335,
      "loss": 0.2325,
      "num_input_tokens_seen": 18967008,
      "step": 89870
    },
    {
      "epoch": 9.887238723872388,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.017895205523815058,
      "loss": 0.2361,
      "num_input_tokens_seen": 18968128,
      "step": 89875
    },
    {
      "epoch": 9.887788778877887,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.01789379254827406,
      "loss": 0.2309,
      "num_input_tokens_seen": 18969152,
      "step": 89880
    },
    {
      "epoch": 9.888338833883388,
      "grad_norm": 0.004669189453125,
      "learning_rate": 0.017892379546062362,
      "loss": 0.2314,
      "num_input_tokens_seen": 18970208,
      "step": 89885
    },
    {
      "epoch": 9.88888888888889,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.017890966517192994,
      "loss": 0.2299,
      "num_input_tokens_seen": 18971232,
      "step": 89890
    },
    {
      "epoch": 9.88943894389439,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.017889553461678975,
      "loss": 0.235,
      "num_input_tokens_seen": 18972384,
      "step": 89895
    },
    {
      "epoch": 9.88998899889989,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.017888140379533325,
      "loss": 0.2319,
      "num_input_tokens_seen": 18973440,
      "step": 89900
    },
    {
      "epoch": 9.89053905390539,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.017886727270769072,
      "loss": 0.2314,
      "num_input_tokens_seen": 18974496,
      "step": 89905
    },
    {
      "epoch": 9.891089108910892,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01788531413539924,
      "loss": 0.2324,
      "num_input_tokens_seen": 18975552,
      "step": 89910
    },
    {
      "epoch": 9.891639163916391,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.017883900973436853,
      "loss": 0.2283,
      "num_input_tokens_seen": 18976544,
      "step": 89915
    },
    {
      "epoch": 9.892189218921892,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.017882487784894933,
      "loss": 0.2325,
      "num_input_tokens_seen": 18977664,
      "step": 89920
    },
    {
      "epoch": 9.892739273927393,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.017881074569786503,
      "loss": 0.2293,
      "num_input_tokens_seen": 18978688,
      "step": 89925
    },
    {
      "epoch": 9.893289328932893,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.0178796613281246,
      "loss": 0.2319,
      "num_input_tokens_seen": 18979744,
      "step": 89930
    },
    {
      "epoch": 9.893839383938394,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.017878248059922233,
      "loss": 0.2314,
      "num_input_tokens_seen": 18980832,
      "step": 89935
    },
    {
      "epoch": 9.894389438943895,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.01787683476519244,
      "loss": 0.2298,
      "num_input_tokens_seen": 18981984,
      "step": 89940
    },
    {
      "epoch": 9.894939493949394,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.017875421443948235,
      "loss": 0.2309,
      "num_input_tokens_seen": 18983072,
      "step": 89945
    },
    {
      "epoch": 9.895489548954895,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.017874008096202654,
      "loss": 0.233,
      "num_input_tokens_seen": 18984096,
      "step": 89950
    },
    {
      "epoch": 9.896039603960396,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01787259472196872,
      "loss": 0.2304,
      "num_input_tokens_seen": 18985152,
      "step": 89955
    },
    {
      "epoch": 9.896589658965897,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.017871181321259456,
      "loss": 0.2351,
      "num_input_tokens_seen": 18986240,
      "step": 89960
    },
    {
      "epoch": 9.897139713971397,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.017869767894087896,
      "loss": 0.232,
      "num_input_tokens_seen": 18987296,
      "step": 89965
    },
    {
      "epoch": 9.897689768976898,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.017868354440467055,
      "loss": 0.2346,
      "num_input_tokens_seen": 18988384,
      "step": 89970
    },
    {
      "epoch": 9.898239823982399,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.017866940960409975,
      "loss": 0.2325,
      "num_input_tokens_seen": 18989344,
      "step": 89975
    },
    {
      "epoch": 9.898789878987898,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.017865527453929674,
      "loss": 0.2288,
      "num_input_tokens_seen": 18990400,
      "step": 89980
    },
    {
      "epoch": 9.8993399339934,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.01786411392103918,
      "loss": 0.232,
      "num_input_tokens_seen": 18991424,
      "step": 89985
    },
    {
      "epoch": 9.8998899889989,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.017862700361751523,
      "loss": 0.2335,
      "num_input_tokens_seen": 18992448,
      "step": 89990
    },
    {
      "epoch": 9.9004400440044,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.01786128677607973,
      "loss": 0.2314,
      "num_input_tokens_seen": 18993536,
      "step": 89995
    },
    {
      "epoch": 9.900990099009901,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.017859873164036828,
      "loss": 0.2314,
      "num_input_tokens_seen": 18994624,
      "step": 90000
    },
    {
      "epoch": 9.901540154015402,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01785845952563585,
      "loss": 0.2288,
      "num_input_tokens_seen": 18995680,
      "step": 90005
    },
    {
      "epoch": 9.902090209020901,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.017857045860889823,
      "loss": 0.2319,
      "num_input_tokens_seen": 18996704,
      "step": 90010
    },
    {
      "epoch": 9.902640264026402,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.017855632169811772,
      "loss": 0.2324,
      "num_input_tokens_seen": 18997696,
      "step": 90015
    },
    {
      "epoch": 9.903190319031903,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.017854218452414736,
      "loss": 0.2314,
      "num_input_tokens_seen": 18998752,
      "step": 90020
    },
    {
      "epoch": 9.903740374037405,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01785280470871173,
      "loss": 0.2298,
      "num_input_tokens_seen": 18999808,
      "step": 90025
    },
    {
      "epoch": 9.904290429042904,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.017851390938715794,
      "loss": 0.2309,
      "num_input_tokens_seen": 19000800,
      "step": 90030
    },
    {
      "epoch": 9.904840484048405,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.017849977142439962,
      "loss": 0.2335,
      "num_input_tokens_seen": 19001920,
      "step": 90035
    },
    {
      "epoch": 9.905390539053906,
      "grad_norm": 0.0029296875,
      "learning_rate": 0.017848563319897256,
      "loss": 0.2319,
      "num_input_tokens_seen": 19002912,
      "step": 90040
    },
    {
      "epoch": 9.905940594059405,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.017847149471100703,
      "loss": 0.2351,
      "num_input_tokens_seen": 19003968,
      "step": 90045
    },
    {
      "epoch": 9.906490649064907,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.01784573559606334,
      "loss": 0.2304,
      "num_input_tokens_seen": 19005024,
      "step": 90050
    },
    {
      "epoch": 9.907040704070408,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.0178443216947982,
      "loss": 0.2335,
      "num_input_tokens_seen": 19006048,
      "step": 90055
    },
    {
      "epoch": 9.907590759075907,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.017842907767318315,
      "loss": 0.2319,
      "num_input_tokens_seen": 19007072,
      "step": 90060
    },
    {
      "epoch": 9.908140814081408,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.01784149381363671,
      "loss": 0.2335,
      "num_input_tokens_seen": 19008096,
      "step": 90065
    },
    {
      "epoch": 9.908690869086909,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.01784007983376642,
      "loss": 0.2319,
      "num_input_tokens_seen": 19009120,
      "step": 90070
    },
    {
      "epoch": 9.909240924092408,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.017838665827720476,
      "loss": 0.2319,
      "num_input_tokens_seen": 19010112,
      "step": 90075
    },
    {
      "epoch": 9.90979097909791,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.017837251795511915,
      "loss": 0.2319,
      "num_input_tokens_seen": 19011200,
      "step": 90080
    },
    {
      "epoch": 9.91034103410341,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.017835837737153762,
      "loss": 0.2288,
      "num_input_tokens_seen": 19012256,
      "step": 90085
    },
    {
      "epoch": 9.910891089108912,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.017834423652659054,
      "loss": 0.233,
      "num_input_tokens_seen": 19013344,
      "step": 90090
    },
    {
      "epoch": 9.911441144114411,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.017833009542040826,
      "loss": 0.2303,
      "num_input_tokens_seen": 19014336,
      "step": 90095
    },
    {
      "epoch": 9.911991199119912,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.017831595405312102,
      "loss": 0.2309,
      "num_input_tokens_seen": 19015360,
      "step": 90100
    },
    {
      "epoch": 9.912541254125413,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.017830181242485928,
      "loss": 0.2314,
      "num_input_tokens_seen": 19016384,
      "step": 90105
    },
    {
      "epoch": 9.913091309130913,
      "grad_norm": 0.000762939453125,
      "learning_rate": 0.01782876705357533,
      "loss": 0.2335,
      "num_input_tokens_seen": 19017440,
      "step": 90110
    },
    {
      "epoch": 9.913641364136414,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.017827352838593345,
      "loss": 0.2293,
      "num_input_tokens_seen": 19018464,
      "step": 90115
    },
    {
      "epoch": 9.914191419141915,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.017825938597553,
      "loss": 0.2304,
      "num_input_tokens_seen": 19019488,
      "step": 90120
    },
    {
      "epoch": 9.914741474147414,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.017824524330467342,
      "loss": 0.2283,
      "num_input_tokens_seen": 19020576,
      "step": 90125
    },
    {
      "epoch": 9.915291529152915,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.017823110037349394,
      "loss": 0.233,
      "num_input_tokens_seen": 19021568,
      "step": 90130
    },
    {
      "epoch": 9.915841584158416,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.0178216957182122,
      "loss": 0.2324,
      "num_input_tokens_seen": 19022560,
      "step": 90135
    },
    {
      "epoch": 9.916391639163916,
      "grad_norm": 0.009765625,
      "learning_rate": 0.017820281373068788,
      "loss": 0.2329,
      "num_input_tokens_seen": 19023584,
      "step": 90140
    },
    {
      "epoch": 9.916941694169417,
      "grad_norm": 0.00069427490234375,
      "learning_rate": 0.01781886700193219,
      "loss": 0.2309,
      "num_input_tokens_seen": 19024608,
      "step": 90145
    },
    {
      "epoch": 9.917491749174918,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.017817452604815456,
      "loss": 0.2324,
      "num_input_tokens_seen": 19025696,
      "step": 90150
    },
    {
      "epoch": 9.918041804180419,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.01781603818173161,
      "loss": 0.2288,
      "num_input_tokens_seen": 19026752,
      "step": 90155
    },
    {
      "epoch": 9.918591859185918,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.01781462373269369,
      "loss": 0.2309,
      "num_input_tokens_seen": 19027744,
      "step": 90160
    },
    {
      "epoch": 9.91914191419142,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.017813209257714735,
      "loss": 0.2309,
      "num_input_tokens_seen": 19028832,
      "step": 90165
    },
    {
      "epoch": 9.91969196919692,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.017811794756807782,
      "loss": 0.2314,
      "num_input_tokens_seen": 19029888,
      "step": 90170
    },
    {
      "epoch": 9.92024202420242,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01781038022998586,
      "loss": 0.233,
      "num_input_tokens_seen": 19030912,
      "step": 90175
    },
    {
      "epoch": 9.92079207920792,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.017808965677262017,
      "loss": 0.2319,
      "num_input_tokens_seen": 19032000,
      "step": 90180
    },
    {
      "epoch": 9.921342134213422,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.017807551098649288,
      "loss": 0.2288,
      "num_input_tokens_seen": 19033024,
      "step": 90185
    },
    {
      "epoch": 9.921892189218921,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.017806136494160702,
      "loss": 0.2288,
      "num_input_tokens_seen": 19034048,
      "step": 90190
    },
    {
      "epoch": 9.922442244224422,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.017804721863809304,
      "loss": 0.2299,
      "num_input_tokens_seen": 19035072,
      "step": 90195
    },
    {
      "epoch": 9.922992299229923,
      "grad_norm": 0.00970458984375,
      "learning_rate": 0.01780330720760813,
      "loss": 0.2309,
      "num_input_tokens_seen": 19036128,
      "step": 90200
    },
    {
      "epoch": 9.923542354235423,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.017801892525570222,
      "loss": 0.2309,
      "num_input_tokens_seen": 19037248,
      "step": 90205
    },
    {
      "epoch": 9.924092409240924,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.01780047781770861,
      "loss": 0.2329,
      "num_input_tokens_seen": 19038272,
      "step": 90210
    },
    {
      "epoch": 9.924642464246425,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.01779906308403634,
      "loss": 0.2303,
      "num_input_tokens_seen": 19039296,
      "step": 90215
    },
    {
      "epoch": 9.925192519251926,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01779764832456645,
      "loss": 0.233,
      "num_input_tokens_seen": 19040416,
      "step": 90220
    },
    {
      "epoch": 9.925742574257425,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01779623353931198,
      "loss": 0.2288,
      "num_input_tokens_seen": 19041504,
      "step": 90225
    },
    {
      "epoch": 9.926292629262926,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.01779481872828596,
      "loss": 0.2314,
      "num_input_tokens_seen": 19042560,
      "step": 90230
    },
    {
      "epoch": 9.926842684268427,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.01779340389150144,
      "loss": 0.233,
      "num_input_tokens_seen": 19043616,
      "step": 90235
    },
    {
      "epoch": 9.927392739273927,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.01779198902897146,
      "loss": 0.2319,
      "num_input_tokens_seen": 19044672,
      "step": 90240
    },
    {
      "epoch": 9.927942794279428,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.017790574140709055,
      "loss": 0.2319,
      "num_input_tokens_seen": 19045728,
      "step": 90245
    },
    {
      "epoch": 9.928492849284929,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.017789159226727268,
      "loss": 0.2304,
      "num_input_tokens_seen": 19046752,
      "step": 90250
    },
    {
      "epoch": 9.929042904290428,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.017787744287039137,
      "loss": 0.2309,
      "num_input_tokens_seen": 19047872,
      "step": 90255
    },
    {
      "epoch": 9.92959295929593,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.017786329321657704,
      "loss": 0.2335,
      "num_input_tokens_seen": 19048960,
      "step": 90260
    },
    {
      "epoch": 9.93014301430143,
      "grad_norm": 0.0028076171875,
      "learning_rate": 0.01778491433059601,
      "loss": 0.2314,
      "num_input_tokens_seen": 19049984,
      "step": 90265
    },
    {
      "epoch": 9.930693069306932,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.0177834993138671,
      "loss": 0.2319,
      "num_input_tokens_seen": 19051040,
      "step": 90270
    },
    {
      "epoch": 9.93124312431243,
      "grad_norm": 0.0096435546875,
      "learning_rate": 0.01778208427148401,
      "loss": 0.2309,
      "num_input_tokens_seen": 19052064,
      "step": 90275
    },
    {
      "epoch": 9.931793179317932,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.017780669203459786,
      "loss": 0.2304,
      "num_input_tokens_seen": 19053088,
      "step": 90280
    },
    {
      "epoch": 9.932343234323433,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01777925410980747,
      "loss": 0.234,
      "num_input_tokens_seen": 19054176,
      "step": 90285
    },
    {
      "epoch": 9.932893289328932,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.017777838990540095,
      "loss": 0.2283,
      "num_input_tokens_seen": 19055200,
      "step": 90290
    },
    {
      "epoch": 9.933443344334433,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.017776423845670717,
      "loss": 0.2314,
      "num_input_tokens_seen": 19056256,
      "step": 90295
    },
    {
      "epoch": 9.933993399339935,
      "grad_norm": 0.00958251953125,
      "learning_rate": 0.017775008675212374,
      "loss": 0.2288,
      "num_input_tokens_seen": 19057344,
      "step": 90300
    },
    {
      "epoch": 9.934543454345434,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.017773593479178104,
      "loss": 0.233,
      "num_input_tokens_seen": 19058496,
      "step": 90305
    },
    {
      "epoch": 9.935093509350935,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.017772178257580956,
      "loss": 0.2314,
      "num_input_tokens_seen": 19059552,
      "step": 90310
    },
    {
      "epoch": 9.935643564356436,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.01777076301043397,
      "loss": 0.2309,
      "num_input_tokens_seen": 19060640,
      "step": 90315
    },
    {
      "epoch": 9.936193619361937,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.017769347737750192,
      "loss": 0.2319,
      "num_input_tokens_seen": 19061696,
      "step": 90320
    },
    {
      "epoch": 9.936743674367436,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.017767932439542666,
      "loss": 0.2309,
      "num_input_tokens_seen": 19062752,
      "step": 90325
    },
    {
      "epoch": 9.937293729372938,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.017766517115824432,
      "loss": 0.2335,
      "num_input_tokens_seen": 19063744,
      "step": 90330
    },
    {
      "epoch": 9.937843784378439,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.017765101766608535,
      "loss": 0.2304,
      "num_input_tokens_seen": 19064832,
      "step": 90335
    },
    {
      "epoch": 9.938393839383938,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.017763686391908028,
      "loss": 0.2335,
      "num_input_tokens_seen": 19065888,
      "step": 90340
    },
    {
      "epoch": 9.938943894389439,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.017762270991735947,
      "loss": 0.2293,
      "num_input_tokens_seen": 19066944,
      "step": 90345
    },
    {
      "epoch": 9.93949394939494,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.01776085556610534,
      "loss": 0.2324,
      "num_input_tokens_seen": 19068000,
      "step": 90350
    },
    {
      "epoch": 9.94004400440044,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.017759440115029248,
      "loss": 0.2304,
      "num_input_tokens_seen": 19069056,
      "step": 90355
    },
    {
      "epoch": 9.94059405940594,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.017758024638520727,
      "loss": 0.2309,
      "num_input_tokens_seen": 19070144,
      "step": 90360
    },
    {
      "epoch": 9.941144114411442,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01775660913659281,
      "loss": 0.2298,
      "num_input_tokens_seen": 19071232,
      "step": 90365
    },
    {
      "epoch": 9.941694169416941,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01775519360925856,
      "loss": 0.2335,
      "num_input_tokens_seen": 19072256,
      "step": 90370
    },
    {
      "epoch": 9.942244224422442,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.017753778056531,
      "loss": 0.2335,
      "num_input_tokens_seen": 19073280,
      "step": 90375
    },
    {
      "epoch": 9.942794279427943,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.017752362478423194,
      "loss": 0.2298,
      "num_input_tokens_seen": 19074272,
      "step": 90380
    },
    {
      "epoch": 9.943344334433444,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.017750946874948183,
      "loss": 0.2308,
      "num_input_tokens_seen": 19075328,
      "step": 90385
    },
    {
      "epoch": 9.943894389438944,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.01774953124611902,
      "loss": 0.2309,
      "num_input_tokens_seen": 19076352,
      "step": 90390
    },
    {
      "epoch": 9.944444444444445,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.017748115591948742,
      "loss": 0.2314,
      "num_input_tokens_seen": 19077344,
      "step": 90395
    },
    {
      "epoch": 9.944994499449946,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.0177466999124504,
      "loss": 0.2314,
      "num_input_tokens_seen": 19078400,
      "step": 90400
    },
    {
      "epoch": 9.945544554455445,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.017745284207637043,
      "loss": 0.2314,
      "num_input_tokens_seen": 19079456,
      "step": 90405
    },
    {
      "epoch": 9.946094609460946,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.017743868477521723,
      "loss": 0.2293,
      "num_input_tokens_seen": 19080480,
      "step": 90410
    },
    {
      "epoch": 9.946644664466447,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.017742452722117486,
      "loss": 0.2335,
      "num_input_tokens_seen": 19081504,
      "step": 90415
    },
    {
      "epoch": 9.947194719471947,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.017741036941437367,
      "loss": 0.233,
      "num_input_tokens_seen": 19082528,
      "step": 90420
    },
    {
      "epoch": 9.947744774477448,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.017739621135494432,
      "loss": 0.2303,
      "num_input_tokens_seen": 19083584,
      "step": 90425
    },
    {
      "epoch": 9.948294829482949,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.017738205304301726,
      "loss": 0.2288,
      "num_input_tokens_seen": 19084640,
      "step": 90430
    },
    {
      "epoch": 9.948844884488448,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.01773678944787229,
      "loss": 0.2299,
      "num_input_tokens_seen": 19085728,
      "step": 90435
    },
    {
      "epoch": 9.94939493949395,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.017735373566219185,
      "loss": 0.2319,
      "num_input_tokens_seen": 19086784,
      "step": 90440
    },
    {
      "epoch": 9.94994499449945,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.017733957659355454,
      "loss": 0.2304,
      "num_input_tokens_seen": 19087840,
      "step": 90445
    },
    {
      "epoch": 9.950495049504951,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.017732541727294143,
      "loss": 0.2345,
      "num_input_tokens_seen": 19088960,
      "step": 90450
    },
    {
      "epoch": 9.95104510451045,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.017731125770048307,
      "loss": 0.2356,
      "num_input_tokens_seen": 19090080,
      "step": 90455
    },
    {
      "epoch": 9.951595159515952,
      "grad_norm": 0.003875732421875,
      "learning_rate": 0.017729709787631,
      "loss": 0.2324,
      "num_input_tokens_seen": 19091232,
      "step": 90460
    },
    {
      "epoch": 9.952145214521453,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.017728293780055263,
      "loss": 0.233,
      "num_input_tokens_seen": 19092288,
      "step": 90465
    },
    {
      "epoch": 9.952695269526952,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.017726877747334155,
      "loss": 0.2283,
      "num_input_tokens_seen": 19093376,
      "step": 90470
    },
    {
      "epoch": 9.953245324532453,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.017725461689480717,
      "loss": 0.2283,
      "num_input_tokens_seen": 19094432,
      "step": 90475
    },
    {
      "epoch": 9.953795379537954,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.017724045606508008,
      "loss": 0.2319,
      "num_input_tokens_seen": 19095488,
      "step": 90480
    },
    {
      "epoch": 9.954345434543454,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.017722629498429083,
      "loss": 0.2324,
      "num_input_tokens_seen": 19096512,
      "step": 90485
    },
    {
      "epoch": 9.954895489548955,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.017721213365256987,
      "loss": 0.2314,
      "num_input_tokens_seen": 19097632,
      "step": 90490
    },
    {
      "epoch": 9.955445544554456,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01771979720700477,
      "loss": 0.2319,
      "num_input_tokens_seen": 19098688,
      "step": 90495
    },
    {
      "epoch": 9.955995599559955,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.01771838102368549,
      "loss": 0.2314,
      "num_input_tokens_seen": 19099776,
      "step": 90500
    },
    {
      "epoch": 9.956545654565456,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.017716964815312198,
      "loss": 0.2293,
      "num_input_tokens_seen": 19100768,
      "step": 90505
    },
    {
      "epoch": 9.957095709570957,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.01771554858189794,
      "loss": 0.2303,
      "num_input_tokens_seen": 19101824,
      "step": 90510
    },
    {
      "epoch": 9.957645764576458,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.017714132323455777,
      "loss": 0.235,
      "num_input_tokens_seen": 19102912,
      "step": 90515
    },
    {
      "epoch": 9.958195819581958,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.017712716039998758,
      "loss": 0.233,
      "num_input_tokens_seen": 19103968,
      "step": 90520
    },
    {
      "epoch": 9.958745874587459,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.017711299731539936,
      "loss": 0.2329,
      "num_input_tokens_seen": 19105056,
      "step": 90525
    },
    {
      "epoch": 9.95929592959296,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.017709883398092364,
      "loss": 0.233,
      "num_input_tokens_seen": 19106144,
      "step": 90530
    },
    {
      "epoch": 9.95984598459846,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.0177084670396691,
      "loss": 0.234,
      "num_input_tokens_seen": 19107200,
      "step": 90535
    },
    {
      "epoch": 9.96039603960396,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.017707050656283196,
      "loss": 0.2309,
      "num_input_tokens_seen": 19108288,
      "step": 90540
    },
    {
      "epoch": 9.960946094609461,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.017705634247947702,
      "loss": 0.2324,
      "num_input_tokens_seen": 19109376,
      "step": 90545
    },
    {
      "epoch": 9.96149614961496,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01770421781467568,
      "loss": 0.2335,
      "num_input_tokens_seen": 19110400,
      "step": 90550
    },
    {
      "epoch": 9.962046204620462,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.017702801356480172,
      "loss": 0.2319,
      "num_input_tokens_seen": 19111424,
      "step": 90555
    },
    {
      "epoch": 9.962596259625963,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.01770138487337425,
      "loss": 0.2319,
      "num_input_tokens_seen": 19112448,
      "step": 90560
    },
    {
      "epoch": 9.963146314631462,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.017699968365370952,
      "loss": 0.2303,
      "num_input_tokens_seen": 19113472,
      "step": 90565
    },
    {
      "epoch": 9.963696369636963,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.017698551832483343,
      "loss": 0.2314,
      "num_input_tokens_seen": 19114560,
      "step": 90570
    },
    {
      "epoch": 9.964246424642464,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.017697135274724482,
      "loss": 0.2308,
      "num_input_tokens_seen": 19115680,
      "step": 90575
    },
    {
      "epoch": 9.964796479647966,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.017695718692107416,
      "loss": 0.2319,
      "num_input_tokens_seen": 19116736,
      "step": 90580
    },
    {
      "epoch": 9.965346534653465,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.017694302084645203,
      "loss": 0.2314,
      "num_input_tokens_seen": 19117792,
      "step": 90585
    },
    {
      "epoch": 9.965896589658966,
      "grad_norm": 0.004302978515625,
      "learning_rate": 0.017692885452350902,
      "loss": 0.2309,
      "num_input_tokens_seen": 19118848,
      "step": 90590
    },
    {
      "epoch": 9.966446644664467,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.017691468795237565,
      "loss": 0.2278,
      "num_input_tokens_seen": 19119904,
      "step": 90595
    },
    {
      "epoch": 9.966996699669966,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.017690052113318252,
      "loss": 0.2324,
      "num_input_tokens_seen": 19120928,
      "step": 90600
    },
    {
      "epoch": 9.967546754675467,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.017688635406606028,
      "loss": 0.2329,
      "num_input_tokens_seen": 19121952,
      "step": 90605
    },
    {
      "epoch": 9.968096809680969,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.017687218675113935,
      "loss": 0.2313,
      "num_input_tokens_seen": 19122912,
      "step": 90610
    },
    {
      "epoch": 9.968646864686468,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.017685801918855035,
      "loss": 0.2298,
      "num_input_tokens_seen": 19123968,
      "step": 90615
    },
    {
      "epoch": 9.969196919691969,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.017684385137842387,
      "loss": 0.2303,
      "num_input_tokens_seen": 19124960,
      "step": 90620
    },
    {
      "epoch": 9.96974697469747,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.01768296833208905,
      "loss": 0.2304,
      "num_input_tokens_seen": 19126016,
      "step": 90625
    },
    {
      "epoch": 9.97029702970297,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01768155150160809,
      "loss": 0.2304,
      "num_input_tokens_seen": 19127104,
      "step": 90630
    },
    {
      "epoch": 9.97084708470847,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.017680134646412545,
      "loss": 0.2298,
      "num_input_tokens_seen": 19128192,
      "step": 90635
    },
    {
      "epoch": 9.971397139713972,
      "grad_norm": 0.006744384765625,
      "learning_rate": 0.01767871776651549,
      "loss": 0.2314,
      "num_input_tokens_seen": 19129216,
      "step": 90640
    },
    {
      "epoch": 9.971947194719473,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.017677300861929975,
      "loss": 0.2319,
      "num_input_tokens_seen": 19130272,
      "step": 90645
    },
    {
      "epoch": 9.972497249724972,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.01767588393266907,
      "loss": 0.2314,
      "num_input_tokens_seen": 19131392,
      "step": 90650
    },
    {
      "epoch": 9.973047304730473,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.017674466978745824,
      "loss": 0.2309,
      "num_input_tokens_seen": 19132448,
      "step": 90655
    },
    {
      "epoch": 9.973597359735974,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.017673050000173298,
      "loss": 0.2309,
      "num_input_tokens_seen": 19133504,
      "step": 90660
    },
    {
      "epoch": 9.974147414741473,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.017671632996964552,
      "loss": 0.2314,
      "num_input_tokens_seen": 19134560,
      "step": 90665
    },
    {
      "epoch": 9.974697469746975,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.017670215969132645,
      "loss": 0.2298,
      "num_input_tokens_seen": 19135616,
      "step": 90670
    },
    {
      "epoch": 9.975247524752476,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.017668798916690643,
      "loss": 0.2314,
      "num_input_tokens_seen": 19136704,
      "step": 90675
    },
    {
      "epoch": 9.975797579757975,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.0176673818396516,
      "loss": 0.2298,
      "num_input_tokens_seen": 19137728,
      "step": 90680
    },
    {
      "epoch": 9.976347634763476,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.01766596473802858,
      "loss": 0.2309,
      "num_input_tokens_seen": 19138752,
      "step": 90685
    },
    {
      "epoch": 9.976897689768977,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.01766454761183464,
      "loss": 0.2304,
      "num_input_tokens_seen": 19139776,
      "step": 90690
    },
    {
      "epoch": 9.977447744774478,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.017663130461082845,
      "loss": 0.2298,
      "num_input_tokens_seen": 19140864,
      "step": 90695
    },
    {
      "epoch": 9.977997799779978,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.01766171328578625,
      "loss": 0.2319,
      "num_input_tokens_seen": 19142048,
      "step": 90700
    },
    {
      "epoch": 9.978547854785479,
      "grad_norm": 0.00421142578125,
      "learning_rate": 0.017660296085957927,
      "loss": 0.2345,
      "num_input_tokens_seen": 19143104,
      "step": 90705
    },
    {
      "epoch": 9.97909790979098,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.017658878861610928,
      "loss": 0.2319,
      "num_input_tokens_seen": 19144096,
      "step": 90710
    },
    {
      "epoch": 9.979647964796479,
      "grad_norm": 0.003082275390625,
      "learning_rate": 0.017657461612758323,
      "loss": 0.2309,
      "num_input_tokens_seen": 19145120,
      "step": 90715
    },
    {
      "epoch": 9.98019801980198,
      "grad_norm": 0.0030670166015625,
      "learning_rate": 0.017656044339413163,
      "loss": 0.2319,
      "num_input_tokens_seen": 19146304,
      "step": 90720
    },
    {
      "epoch": 9.980748074807481,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01765462704158852,
      "loss": 0.2299,
      "num_input_tokens_seen": 19147360,
      "step": 90725
    },
    {
      "epoch": 9.98129812981298,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.017653209719297454,
      "loss": 0.2293,
      "num_input_tokens_seen": 19148448,
      "step": 90730
    },
    {
      "epoch": 9.981848184818482,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.017651792372553026,
      "loss": 0.2299,
      "num_input_tokens_seen": 19149504,
      "step": 90735
    },
    {
      "epoch": 9.982398239823983,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.017650375001368297,
      "loss": 0.2314,
      "num_input_tokens_seen": 19150560,
      "step": 90740
    },
    {
      "epoch": 9.982948294829484,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.017648957605756337,
      "loss": 0.2325,
      "num_input_tokens_seen": 19151680,
      "step": 90745
    },
    {
      "epoch": 9.983498349834983,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.017647540185730212,
      "loss": 0.2304,
      "num_input_tokens_seen": 19152800,
      "step": 90750
    },
    {
      "epoch": 9.984048404840484,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01764612274130297,
      "loss": 0.2325,
      "num_input_tokens_seen": 19153888,
      "step": 90755
    },
    {
      "epoch": 9.984598459845985,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.017644705272487685,
      "loss": 0.2315,
      "num_input_tokens_seen": 19154944,
      "step": 90760
    },
    {
      "epoch": 9.985148514851485,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.017643287779297426,
      "loss": 0.2314,
      "num_input_tokens_seen": 19156000,
      "step": 90765
    },
    {
      "epoch": 9.985698569856986,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.017641870261745244,
      "loss": 0.2319,
      "num_input_tokens_seen": 19157120,
      "step": 90770
    },
    {
      "epoch": 9.986248624862487,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01764045271984422,
      "loss": 0.2278,
      "num_input_tokens_seen": 19158176,
      "step": 90775
    },
    {
      "epoch": 9.986798679867986,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.01763903515360741,
      "loss": 0.2319,
      "num_input_tokens_seen": 19159200,
      "step": 90780
    },
    {
      "epoch": 9.987348734873487,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.017637617563047873,
      "loss": 0.2335,
      "num_input_tokens_seen": 19160224,
      "step": 90785
    },
    {
      "epoch": 9.987898789878988,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.017636199948178683,
      "loss": 0.2315,
      "num_input_tokens_seen": 19161248,
      "step": 90790
    },
    {
      "epoch": 9.988448844884488,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.017634782309012907,
      "loss": 0.2335,
      "num_input_tokens_seen": 19162240,
      "step": 90795
    },
    {
      "epoch": 9.988998899889989,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.0176333646455636,
      "loss": 0.2299,
      "num_input_tokens_seen": 19163296,
      "step": 90800
    },
    {
      "epoch": 9.98954895489549,
      "grad_norm": 0.0004215240478515625,
      "learning_rate": 0.017631946957843838,
      "loss": 0.2319,
      "num_input_tokens_seen": 19164320,
      "step": 90805
    },
    {
      "epoch": 9.990099009900991,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.017630529245866687,
      "loss": 0.2324,
      "num_input_tokens_seen": 19165376,
      "step": 90810
    },
    {
      "epoch": 9.99064906490649,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.017629111509645204,
      "loss": 0.2303,
      "num_input_tokens_seen": 19166432,
      "step": 90815
    },
    {
      "epoch": 9.991199119911991,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.01762769374919247,
      "loss": 0.2329,
      "num_input_tokens_seen": 19167520,
      "step": 90820
    },
    {
      "epoch": 9.991749174917492,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.01762627596452154,
      "loss": 0.2314,
      "num_input_tokens_seen": 19168672,
      "step": 90825
    },
    {
      "epoch": 9.992299229922992,
      "grad_norm": 0.00323486328125,
      "learning_rate": 0.01762485815564548,
      "loss": 0.2309,
      "num_input_tokens_seen": 19169792,
      "step": 90830
    },
    {
      "epoch": 9.992849284928493,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.017623440322577365,
      "loss": 0.2303,
      "num_input_tokens_seen": 19170816,
      "step": 90835
    },
    {
      "epoch": 9.993399339933994,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.017622022465330258,
      "loss": 0.2319,
      "num_input_tokens_seen": 19171872,
      "step": 90840
    },
    {
      "epoch": 9.993949394939493,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.01762060458391723,
      "loss": 0.2293,
      "num_input_tokens_seen": 19172928,
      "step": 90845
    },
    {
      "epoch": 9.994499449944994,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.01761918667835135,
      "loss": 0.2293,
      "num_input_tokens_seen": 19173984,
      "step": 90850
    },
    {
      "epoch": 9.995049504950495,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01761776874864568,
      "loss": 0.233,
      "num_input_tokens_seen": 19175040,
      "step": 90855
    },
    {
      "epoch": 9.995599559955995,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.017616350794813284,
      "loss": 0.2309,
      "num_input_tokens_seen": 19176032,
      "step": 90860
    },
    {
      "epoch": 9.996149614961496,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.017614932816867247,
      "loss": 0.2309,
      "num_input_tokens_seen": 19177056,
      "step": 90865
    },
    {
      "epoch": 9.996699669966997,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.017613514814820628,
      "loss": 0.2314,
      "num_input_tokens_seen": 19178112,
      "step": 90870
    },
    {
      "epoch": 9.997249724972498,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.017612096788686495,
      "loss": 0.2325,
      "num_input_tokens_seen": 19179136,
      "step": 90875
    },
    {
      "epoch": 9.997799779977997,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.017610678738477926,
      "loss": 0.2335,
      "num_input_tokens_seen": 19180192,
      "step": 90880
    },
    {
      "epoch": 9.998349834983498,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.017609260664207975,
      "loss": 0.2324,
      "num_input_tokens_seen": 19181280,
      "step": 90885
    },
    {
      "epoch": 9.998899889989,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.017607842565889723,
      "loss": 0.2324,
      "num_input_tokens_seen": 19182272,
      "step": 90890
    },
    {
      "epoch": 9.999449944994499,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01760642444353624,
      "loss": 0.2314,
      "num_input_tokens_seen": 19183328,
      "step": 90895
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.01760500629716059,
      "loss": 0.2319,
      "num_input_tokens_seen": 19184224,
      "step": 90900
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.23146456480026245,
      "eval_runtime": 60.5533,
      "eval_samples_per_second": 66.718,
      "eval_steps_per_second": 16.68,
      "num_input_tokens_seen": 19184224,
      "step": 90900
    },
    {
      "epoch": 10.000550055005501,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.017603588126775854,
      "loss": 0.2319,
      "num_input_tokens_seen": 19185312,
      "step": 90905
    },
    {
      "epoch": 10.001100110011,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.017602169932395093,
      "loss": 0.2293,
      "num_input_tokens_seen": 19186304,
      "step": 90910
    },
    {
      "epoch": 10.001650165016502,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.017600751714031374,
      "loss": 0.2288,
      "num_input_tokens_seen": 19187360,
      "step": 90915
    },
    {
      "epoch": 10.002200220022003,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.017599333471697783,
      "loss": 0.2303,
      "num_input_tokens_seen": 19188448,
      "step": 90920
    },
    {
      "epoch": 10.002750275027502,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.017597915205407374,
      "loss": 0.2309,
      "num_input_tokens_seen": 19189504,
      "step": 90925
    },
    {
      "epoch": 10.003300330033003,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.017596496915173232,
      "loss": 0.2319,
      "num_input_tokens_seen": 19190560,
      "step": 90930
    },
    {
      "epoch": 10.003850385038504,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.01759507860100842,
      "loss": 0.2314,
      "num_input_tokens_seen": 19191680,
      "step": 90935
    },
    {
      "epoch": 10.004400440044005,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.017593660262926022,
      "loss": 0.233,
      "num_input_tokens_seen": 19192768,
      "step": 90940
    },
    {
      "epoch": 10.004950495049505,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.017592241900939096,
      "loss": 0.2308,
      "num_input_tokens_seen": 19193888,
      "step": 90945
    },
    {
      "epoch": 10.005500550055006,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01759082351506072,
      "loss": 0.2288,
      "num_input_tokens_seen": 19194944,
      "step": 90950
    },
    {
      "epoch": 10.006050605060507,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01758940510530397,
      "loss": 0.2309,
      "num_input_tokens_seen": 19196000,
      "step": 90955
    },
    {
      "epoch": 10.006600660066006,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.01758798667168191,
      "loss": 0.2304,
      "num_input_tokens_seen": 19197024,
      "step": 90960
    },
    {
      "epoch": 10.007150715071507,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.017586568214207628,
      "loss": 0.2329,
      "num_input_tokens_seen": 19198080,
      "step": 90965
    },
    {
      "epoch": 10.007700770077008,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01758514973289418,
      "loss": 0.2319,
      "num_input_tokens_seen": 19199200,
      "step": 90970
    },
    {
      "epoch": 10.008250825082508,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.017583731227754647,
      "loss": 0.2303,
      "num_input_tokens_seen": 19200320,
      "step": 90975
    },
    {
      "epoch": 10.008800880088009,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.017582312698802106,
      "loss": 0.2298,
      "num_input_tokens_seen": 19201408,
      "step": 90980
    },
    {
      "epoch": 10.00935093509351,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.01758089414604963,
      "loss": 0.2309,
      "num_input_tokens_seen": 19202560,
      "step": 90985
    },
    {
      "epoch": 10.009900990099009,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01757947556951029,
      "loss": 0.2309,
      "num_input_tokens_seen": 19203648,
      "step": 90990
    },
    {
      "epoch": 10.01045104510451,
      "grad_norm": 0.000675201416015625,
      "learning_rate": 0.01757805696919716,
      "loss": 0.2325,
      "num_input_tokens_seen": 19204704,
      "step": 90995
    },
    {
      "epoch": 10.011001100110011,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.017576638345123317,
      "loss": 0.2325,
      "num_input_tokens_seen": 19205728,
      "step": 91000
    },
    {
      "epoch": 10.011551155115512,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.01757521969730183,
      "loss": 0.2319,
      "num_input_tokens_seen": 19206784,
      "step": 91005
    },
    {
      "epoch": 10.012101210121012,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.017573801025745787,
      "loss": 0.2324,
      "num_input_tokens_seen": 19207808,
      "step": 91010
    },
    {
      "epoch": 10.012651265126513,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.017572382330468247,
      "loss": 0.2324,
      "num_input_tokens_seen": 19208864,
      "step": 91015
    },
    {
      "epoch": 10.013201320132014,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.0175709636114823,
      "loss": 0.2304,
      "num_input_tokens_seen": 19209920,
      "step": 91020
    },
    {
      "epoch": 10.013751375137513,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.01756954486880101,
      "loss": 0.2319,
      "num_input_tokens_seen": 19210912,
      "step": 91025
    },
    {
      "epoch": 10.014301430143014,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.01756812610243746,
      "loss": 0.2319,
      "num_input_tokens_seen": 19211936,
      "step": 91030
    },
    {
      "epoch": 10.014851485148515,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.017566707312404724,
      "loss": 0.2309,
      "num_input_tokens_seen": 19213056,
      "step": 91035
    },
    {
      "epoch": 10.015401540154015,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.01756528849871588,
      "loss": 0.2319,
      "num_input_tokens_seen": 19214144,
      "step": 91040
    },
    {
      "epoch": 10.015951595159516,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.017563869661384,
      "loss": 0.2314,
      "num_input_tokens_seen": 19215232,
      "step": 91045
    },
    {
      "epoch": 10.016501650165017,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.017562450800422166,
      "loss": 0.2288,
      "num_input_tokens_seen": 19216288,
      "step": 91050
    },
    {
      "epoch": 10.017051705170518,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.01756103191584345,
      "loss": 0.2314,
      "num_input_tokens_seen": 19217440,
      "step": 91055
    },
    {
      "epoch": 10.017601760176017,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.017559613007660936,
      "loss": 0.2335,
      "num_input_tokens_seen": 19218464,
      "step": 91060
    },
    {
      "epoch": 10.018151815181518,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.017558194075887695,
      "loss": 0.2304,
      "num_input_tokens_seen": 19219552,
      "step": 91065
    },
    {
      "epoch": 10.01870187018702,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.017556775120536806,
      "loss": 0.2324,
      "num_input_tokens_seen": 19220608,
      "step": 91070
    },
    {
      "epoch": 10.019251925192519,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.017555356141621344,
      "loss": 0.2293,
      "num_input_tokens_seen": 19221696,
      "step": 91075
    },
    {
      "epoch": 10.01980198019802,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.017553937139154395,
      "loss": 0.2335,
      "num_input_tokens_seen": 19222816,
      "step": 91080
    },
    {
      "epoch": 10.020352035203521,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.017552518113149035,
      "loss": 0.2308,
      "num_input_tokens_seen": 19223936,
      "step": 91085
    },
    {
      "epoch": 10.02090209020902,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.017551099063618337,
      "loss": 0.2304,
      "num_input_tokens_seen": 19225056,
      "step": 91090
    },
    {
      "epoch": 10.021452145214521,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01754967999057538,
      "loss": 0.2298,
      "num_input_tokens_seen": 19226048,
      "step": 91095
    },
    {
      "epoch": 10.022002200220022,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.017548260894033253,
      "loss": 0.2309,
      "num_input_tokens_seen": 19227072,
      "step": 91100
    },
    {
      "epoch": 10.022552255225522,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.017546841774005027,
      "loss": 0.2346,
      "num_input_tokens_seen": 19228128,
      "step": 91105
    },
    {
      "epoch": 10.023102310231023,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.01754542263050378,
      "loss": 0.2319,
      "num_input_tokens_seen": 19229216,
      "step": 91110
    },
    {
      "epoch": 10.023652365236524,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.017544003463542596,
      "loss": 0.2335,
      "num_input_tokens_seen": 19230304,
      "step": 91115
    },
    {
      "epoch": 10.024202420242025,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.017542584273134546,
      "loss": 0.2325,
      "num_input_tokens_seen": 19231424,
      "step": 91120
    },
    {
      "epoch": 10.024752475247524,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.017541165059292724,
      "loss": 0.2314,
      "num_input_tokens_seen": 19232544,
      "step": 91125
    },
    {
      "epoch": 10.025302530253025,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.0175397458220302,
      "loss": 0.2288,
      "num_input_tokens_seen": 19233536,
      "step": 91130
    },
    {
      "epoch": 10.025852585258527,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.01753832656136006,
      "loss": 0.2335,
      "num_input_tokens_seen": 19234592,
      "step": 91135
    },
    {
      "epoch": 10.026402640264026,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.01753690727729538,
      "loss": 0.2314,
      "num_input_tokens_seen": 19235680,
      "step": 91140
    },
    {
      "epoch": 10.026952695269527,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.017535487969849245,
      "loss": 0.2304,
      "num_input_tokens_seen": 19236704,
      "step": 91145
    },
    {
      "epoch": 10.027502750275028,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.017534068639034728,
      "loss": 0.2309,
      "num_input_tokens_seen": 19237728,
      "step": 91150
    },
    {
      "epoch": 10.028052805280527,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.017532649284864924,
      "loss": 0.2325,
      "num_input_tokens_seen": 19238816,
      "step": 91155
    },
    {
      "epoch": 10.028602860286028,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0175312299073529,
      "loss": 0.2309,
      "num_input_tokens_seen": 19239968,
      "step": 91160
    },
    {
      "epoch": 10.02915291529153,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01752981050651175,
      "loss": 0.2304,
      "num_input_tokens_seen": 19241024,
      "step": 91165
    },
    {
      "epoch": 10.029702970297029,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.017528391082354548,
      "loss": 0.2299,
      "num_input_tokens_seen": 19242080,
      "step": 91170
    },
    {
      "epoch": 10.03025302530253,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.01752697163489438,
      "loss": 0.2283,
      "num_input_tokens_seen": 19243072,
      "step": 91175
    },
    {
      "epoch": 10.030803080308031,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.017525552164144326,
      "loss": 0.2314,
      "num_input_tokens_seen": 19244128,
      "step": 91180
    },
    {
      "epoch": 10.031353135313532,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.01752413267011747,
      "loss": 0.2319,
      "num_input_tokens_seen": 19245152,
      "step": 91185
    },
    {
      "epoch": 10.031903190319031,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.017522713152826892,
      "loss": 0.2309,
      "num_input_tokens_seen": 19246240,
      "step": 91190
    },
    {
      "epoch": 10.032453245324533,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.01752129361228568,
      "loss": 0.2325,
      "num_input_tokens_seen": 19247360,
      "step": 91195
    },
    {
      "epoch": 10.033003300330034,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.017519874048506915,
      "loss": 0.2309,
      "num_input_tokens_seen": 19248416,
      "step": 91200
    },
    {
      "epoch": 10.033553355335533,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.01751845446150368,
      "loss": 0.2346,
      "num_input_tokens_seen": 19249504,
      "step": 91205
    },
    {
      "epoch": 10.034103410341034,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.017517034851289057,
      "loss": 0.2294,
      "num_input_tokens_seen": 19250592,
      "step": 91210
    },
    {
      "epoch": 10.034653465346535,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.017515615217876132,
      "loss": 0.232,
      "num_input_tokens_seen": 19251616,
      "step": 91215
    },
    {
      "epoch": 10.035203520352034,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.017514195561277986,
      "loss": 0.2341,
      "num_input_tokens_seen": 19252736,
      "step": 91220
    },
    {
      "epoch": 10.035753575357536,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.017512775881507715,
      "loss": 0.2304,
      "num_input_tokens_seen": 19253792,
      "step": 91225
    },
    {
      "epoch": 10.036303630363037,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.017511356178578387,
      "loss": 0.2314,
      "num_input_tokens_seen": 19254880,
      "step": 91230
    },
    {
      "epoch": 10.036853685368538,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.017509936452503094,
      "loss": 0.2346,
      "num_input_tokens_seen": 19256000,
      "step": 91235
    },
    {
      "epoch": 10.037403740374037,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.017508516703294924,
      "loss": 0.2304,
      "num_input_tokens_seen": 19257088,
      "step": 91240
    },
    {
      "epoch": 10.037953795379538,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.017507096930966956,
      "loss": 0.2309,
      "num_input_tokens_seen": 19258112,
      "step": 91245
    },
    {
      "epoch": 10.03850385038504,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.01750567713553228,
      "loss": 0.2283,
      "num_input_tokens_seen": 19259232,
      "step": 91250
    },
    {
      "epoch": 10.039053905390539,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.017504257317003984,
      "loss": 0.2304,
      "num_input_tokens_seen": 19260320,
      "step": 91255
    },
    {
      "epoch": 10.03960396039604,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.017502837475395147,
      "loss": 0.2298,
      "num_input_tokens_seen": 19261376,
      "step": 91260
    },
    {
      "epoch": 10.04015401540154,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.017501417610718856,
      "loss": 0.232,
      "num_input_tokens_seen": 19262432,
      "step": 91265
    },
    {
      "epoch": 10.04070407040704,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0174999977229882,
      "loss": 0.2319,
      "num_input_tokens_seen": 19263488,
      "step": 91270
    },
    {
      "epoch": 10.041254125412541,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01749857781221627,
      "loss": 0.2293,
      "num_input_tokens_seen": 19264576,
      "step": 91275
    },
    {
      "epoch": 10.041804180418042,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.017497157878416138,
      "loss": 0.2309,
      "num_input_tokens_seen": 19265600,
      "step": 91280
    },
    {
      "epoch": 10.042354235423542,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.017495737921600907,
      "loss": 0.2288,
      "num_input_tokens_seen": 19266592,
      "step": 91285
    },
    {
      "epoch": 10.042904290429043,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.017494317941783657,
      "loss": 0.2319,
      "num_input_tokens_seen": 19267680,
      "step": 91290
    },
    {
      "epoch": 10.043454345434544,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.017492897938977472,
      "loss": 0.2315,
      "num_input_tokens_seen": 19268736,
      "step": 91295
    },
    {
      "epoch": 10.044004400440045,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.017491477913195445,
      "loss": 0.234,
      "num_input_tokens_seen": 19269760,
      "step": 91300
    },
    {
      "epoch": 10.044554455445544,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.017490057864450664,
      "loss": 0.2299,
      "num_input_tokens_seen": 19270880,
      "step": 91305
    },
    {
      "epoch": 10.045104510451045,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.017488637792756206,
      "loss": 0.2289,
      "num_input_tokens_seen": 19272000,
      "step": 91310
    },
    {
      "epoch": 10.045654565456546,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01748721769812517,
      "loss": 0.2304,
      "num_input_tokens_seen": 19273056,
      "step": 91315
    },
    {
      "epoch": 10.046204620462046,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.017485797580570648,
      "loss": 0.233,
      "num_input_tokens_seen": 19274080,
      "step": 91320
    },
    {
      "epoch": 10.046754675467547,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.01748437744010572,
      "loss": 0.2315,
      "num_input_tokens_seen": 19275104,
      "step": 91325
    },
    {
      "epoch": 10.047304730473048,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.01748295727674348,
      "loss": 0.2319,
      "num_input_tokens_seen": 19276224,
      "step": 91330
    },
    {
      "epoch": 10.047854785478547,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01748153709049701,
      "loss": 0.2299,
      "num_input_tokens_seen": 19277280,
      "step": 91335
    },
    {
      "epoch": 10.048404840484048,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0174801168813794,
      "loss": 0.2361,
      "num_input_tokens_seen": 19278272,
      "step": 91340
    },
    {
      "epoch": 10.04895489548955,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.01747869664940375,
      "loss": 0.2314,
      "num_input_tokens_seen": 19279392,
      "step": 91345
    },
    {
      "epoch": 10.049504950495049,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.01747727639458314,
      "loss": 0.2314,
      "num_input_tokens_seen": 19280512,
      "step": 91350
    },
    {
      "epoch": 10.05005500550055,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01747585611693066,
      "loss": 0.2319,
      "num_input_tokens_seen": 19281568,
      "step": 91355
    },
    {
      "epoch": 10.05060506050605,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.017474435816459404,
      "loss": 0.2319,
      "num_input_tokens_seen": 19282688,
      "step": 91360
    },
    {
      "epoch": 10.051155115511552,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.017473015493182465,
      "loss": 0.2325,
      "num_input_tokens_seen": 19283712,
      "step": 91365
    },
    {
      "epoch": 10.051705170517051,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.017471595147112922,
      "loss": 0.233,
      "num_input_tokens_seen": 19284768,
      "step": 91370
    },
    {
      "epoch": 10.052255225522552,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.017470174778263875,
      "loss": 0.2298,
      "num_input_tokens_seen": 19285888,
      "step": 91375
    },
    {
      "epoch": 10.052805280528053,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.01746875438664841,
      "loss": 0.2335,
      "num_input_tokens_seen": 19286944,
      "step": 91380
    },
    {
      "epoch": 10.053355335533553,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.017467333972279623,
      "loss": 0.233,
      "num_input_tokens_seen": 19288032,
      "step": 91385
    },
    {
      "epoch": 10.053905390539054,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.017465913535170605,
      "loss": 0.2314,
      "num_input_tokens_seen": 19289088,
      "step": 91390
    },
    {
      "epoch": 10.054455445544555,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01746449307533444,
      "loss": 0.233,
      "num_input_tokens_seen": 19290176,
      "step": 91395
    },
    {
      "epoch": 10.055005500550054,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.017463072592784228,
      "loss": 0.2335,
      "num_input_tokens_seen": 19291296,
      "step": 91400
    },
    {
      "epoch": 10.055555555555555,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01746165208753306,
      "loss": 0.234,
      "num_input_tokens_seen": 19292256,
      "step": 91405
    },
    {
      "epoch": 10.056105610561056,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.017460231559594017,
      "loss": 0.2304,
      "num_input_tokens_seen": 19293280,
      "step": 91410
    },
    {
      "epoch": 10.056655665566556,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.01745881100898021,
      "loss": 0.2324,
      "num_input_tokens_seen": 19294368,
      "step": 91415
    },
    {
      "epoch": 10.057205720572057,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.01745739043570472,
      "loss": 0.2324,
      "num_input_tokens_seen": 19295392,
      "step": 91420
    },
    {
      "epoch": 10.057755775577558,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.017455969839780636,
      "loss": 0.2288,
      "num_input_tokens_seen": 19296384,
      "step": 91425
    },
    {
      "epoch": 10.058305830583059,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.017454549221221062,
      "loss": 0.2304,
      "num_input_tokens_seen": 19297376,
      "step": 91430
    },
    {
      "epoch": 10.058855885588558,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.01745312858003908,
      "loss": 0.2283,
      "num_input_tokens_seen": 19298400,
      "step": 91435
    },
    {
      "epoch": 10.05940594059406,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.017451707916247797,
      "loss": 0.2335,
      "num_input_tokens_seen": 19299488,
      "step": 91440
    },
    {
      "epoch": 10.05995599559956,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.017450287229860292,
      "loss": 0.2309,
      "num_input_tokens_seen": 19300512,
      "step": 91445
    },
    {
      "epoch": 10.06050605060506,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.017448866520889672,
      "loss": 0.2314,
      "num_input_tokens_seen": 19301568,
      "step": 91450
    },
    {
      "epoch": 10.061056105610561,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.017447445789349016,
      "loss": 0.2283,
      "num_input_tokens_seen": 19302624,
      "step": 91455
    },
    {
      "epoch": 10.061606160616062,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.01744602503525143,
      "loss": 0.2319,
      "num_input_tokens_seen": 19303680,
      "step": 91460
    },
    {
      "epoch": 10.062156215621561,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.017444604258610004,
      "loss": 0.2314,
      "num_input_tokens_seen": 19304736,
      "step": 91465
    },
    {
      "epoch": 10.062706270627062,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01744318345943784,
      "loss": 0.2314,
      "num_input_tokens_seen": 19305824,
      "step": 91470
    },
    {
      "epoch": 10.063256325632564,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.01744176263774802,
      "loss": 0.2304,
      "num_input_tokens_seen": 19306912,
      "step": 91475
    },
    {
      "epoch": 10.063806380638065,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.017440341793553648,
      "loss": 0.2309,
      "num_input_tokens_seen": 19308032,
      "step": 91480
    },
    {
      "epoch": 10.064356435643564,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.017438920926867815,
      "loss": 0.2288,
      "num_input_tokens_seen": 19309088,
      "step": 91485
    },
    {
      "epoch": 10.064906490649065,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01743750003770362,
      "loss": 0.233,
      "num_input_tokens_seen": 19310112,
      "step": 91490
    },
    {
      "epoch": 10.065456545654566,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.017436079126074155,
      "loss": 0.2335,
      "num_input_tokens_seen": 19311136,
      "step": 91495
    },
    {
      "epoch": 10.066006600660065,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.017434658191992518,
      "loss": 0.2325,
      "num_input_tokens_seen": 19312192,
      "step": 91500
    },
    {
      "epoch": 10.066556655665567,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.017433237235471806,
      "loss": 0.2314,
      "num_input_tokens_seen": 19313184,
      "step": 91505
    },
    {
      "epoch": 10.067106710671068,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.017431816256525113,
      "loss": 0.2304,
      "num_input_tokens_seen": 19314144,
      "step": 91510
    },
    {
      "epoch": 10.067656765676567,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.017430395255165536,
      "loss": 0.2314,
      "num_input_tokens_seen": 19315168,
      "step": 91515
    },
    {
      "epoch": 10.068206820682068,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.017428974231406172,
      "loss": 0.2319,
      "num_input_tokens_seen": 19316256,
      "step": 91520
    },
    {
      "epoch": 10.06875687568757,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.017427553185260118,
      "loss": 0.2309,
      "num_input_tokens_seen": 19317312,
      "step": 91525
    },
    {
      "epoch": 10.069306930693068,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01742613211674047,
      "loss": 0.2329,
      "num_input_tokens_seen": 19318336,
      "step": 91530
    },
    {
      "epoch": 10.06985698569857,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.01742471102586033,
      "loss": 0.2329,
      "num_input_tokens_seen": 19319392,
      "step": 91535
    },
    {
      "epoch": 10.07040704070407,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01742328991263279,
      "loss": 0.2298,
      "num_input_tokens_seen": 19320480,
      "step": 91540
    },
    {
      "epoch": 10.070957095709572,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.01742186877707095,
      "loss": 0.2319,
      "num_input_tokens_seen": 19321568,
      "step": 91545
    },
    {
      "epoch": 10.071507150715071,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.017420447619187908,
      "loss": 0.2298,
      "num_input_tokens_seen": 19322688,
      "step": 91550
    },
    {
      "epoch": 10.072057205720572,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.017419026438996756,
      "loss": 0.2303,
      "num_input_tokens_seen": 19323840,
      "step": 91555
    },
    {
      "epoch": 10.072607260726073,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01741760523651061,
      "loss": 0.2319,
      "num_input_tokens_seen": 19324992,
      "step": 91560
    },
    {
      "epoch": 10.073157315731573,
      "grad_norm": 0.000713348388671875,
      "learning_rate": 0.01741618401174255,
      "loss": 0.2314,
      "num_input_tokens_seen": 19325952,
      "step": 91565
    },
    {
      "epoch": 10.073707370737074,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.017414762764705678,
      "loss": 0.2309,
      "num_input_tokens_seen": 19327008,
      "step": 91570
    },
    {
      "epoch": 10.074257425742575,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.017413341495413102,
      "loss": 0.2319,
      "num_input_tokens_seen": 19328096,
      "step": 91575
    },
    {
      "epoch": 10.074807480748074,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.017411920203877915,
      "loss": 0.2324,
      "num_input_tokens_seen": 19329184,
      "step": 91580
    },
    {
      "epoch": 10.075357535753575,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.017410498890113212,
      "loss": 0.2298,
      "num_input_tokens_seen": 19330336,
      "step": 91585
    },
    {
      "epoch": 10.075907590759076,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.0174090775541321,
      "loss": 0.2303,
      "num_input_tokens_seen": 19331456,
      "step": 91590
    },
    {
      "epoch": 10.076457645764576,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.017407656195947675,
      "loss": 0.2308,
      "num_input_tokens_seen": 19332512,
      "step": 91595
    },
    {
      "epoch": 10.077007700770077,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.01740623481557304,
      "loss": 0.2304,
      "num_input_tokens_seen": 19333536,
      "step": 91600
    },
    {
      "epoch": 10.077557755775578,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.017404813413021296,
      "loss": 0.2293,
      "num_input_tokens_seen": 19334560,
      "step": 91605
    },
    {
      "epoch": 10.078107810781079,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.01740339198830554,
      "loss": 0.2303,
      "num_input_tokens_seen": 19335648,
      "step": 91610
    },
    {
      "epoch": 10.078657865786578,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01740197054143887,
      "loss": 0.2309,
      "num_input_tokens_seen": 19336736,
      "step": 91615
    },
    {
      "epoch": 10.07920792079208,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.017400549072434386,
      "loss": 0.2304,
      "num_input_tokens_seen": 19337792,
      "step": 91620
    },
    {
      "epoch": 10.07975797579758,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.017399127581305202,
      "loss": 0.2335,
      "num_input_tokens_seen": 19338816,
      "step": 91625
    },
    {
      "epoch": 10.08030803080308,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.017397706068064404,
      "loss": 0.2334,
      "num_input_tokens_seen": 19339872,
      "step": 91630
    },
    {
      "epoch": 10.08085808580858,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.0173962845327251,
      "loss": 0.2319,
      "num_input_tokens_seen": 19340960,
      "step": 91635
    },
    {
      "epoch": 10.081408140814082,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0173948629753004,
      "loss": 0.2314,
      "num_input_tokens_seen": 19342080,
      "step": 91640
    },
    {
      "epoch": 10.081958195819581,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.017393441395803386,
      "loss": 0.2298,
      "num_input_tokens_seen": 19343168,
      "step": 91645
    },
    {
      "epoch": 10.082508250825082,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.017392019794247177,
      "loss": 0.2298,
      "num_input_tokens_seen": 19344224,
      "step": 91650
    },
    {
      "epoch": 10.083058305830583,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.017390598170644866,
      "loss": 0.2324,
      "num_input_tokens_seen": 19345248,
      "step": 91655
    },
    {
      "epoch": 10.083608360836084,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.017389176525009563,
      "loss": 0.2319,
      "num_input_tokens_seen": 19346240,
      "step": 91660
    },
    {
      "epoch": 10.084158415841584,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.017387754857354363,
      "loss": 0.2314,
      "num_input_tokens_seen": 19347264,
      "step": 91665
    },
    {
      "epoch": 10.084708470847085,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.01738633316769237,
      "loss": 0.2298,
      "num_input_tokens_seen": 19348288,
      "step": 91670
    },
    {
      "epoch": 10.085258525852586,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.017384911456036697,
      "loss": 0.2319,
      "num_input_tokens_seen": 19349376,
      "step": 91675
    },
    {
      "epoch": 10.085808580858085,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.017383489722400437,
      "loss": 0.2309,
      "num_input_tokens_seen": 19350368,
      "step": 91680
    },
    {
      "epoch": 10.086358635863586,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.017382067966796692,
      "loss": 0.2324,
      "num_input_tokens_seen": 19351424,
      "step": 91685
    },
    {
      "epoch": 10.086908690869087,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.017380646189238572,
      "loss": 0.2309,
      "num_input_tokens_seen": 19352480,
      "step": 91690
    },
    {
      "epoch": 10.087458745874587,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.017379224389739177,
      "loss": 0.2329,
      "num_input_tokens_seen": 19353472,
      "step": 91695
    },
    {
      "epoch": 10.088008800880088,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.017377802568311614,
      "loss": 0.2309,
      "num_input_tokens_seen": 19354528,
      "step": 91700
    },
    {
      "epoch": 10.088558855885589,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.017376380724968988,
      "loss": 0.2335,
      "num_input_tokens_seen": 19355584,
      "step": 91705
    },
    {
      "epoch": 10.089108910891088,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0173749588597244,
      "loss": 0.2329,
      "num_input_tokens_seen": 19356608,
      "step": 91710
    },
    {
      "epoch": 10.08965896589659,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.01737353697259095,
      "loss": 0.2309,
      "num_input_tokens_seen": 19357664,
      "step": 91715
    },
    {
      "epoch": 10.09020902090209,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.017372115063581755,
      "loss": 0.2308,
      "num_input_tokens_seen": 19358752,
      "step": 91720
    },
    {
      "epoch": 10.090759075907592,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.017370693132709912,
      "loss": 0.2314,
      "num_input_tokens_seen": 19359776,
      "step": 91725
    },
    {
      "epoch": 10.091309130913091,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.017369271179988526,
      "loss": 0.233,
      "num_input_tokens_seen": 19360800,
      "step": 91730
    },
    {
      "epoch": 10.091859185918592,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.01736784920543071,
      "loss": 0.2309,
      "num_input_tokens_seen": 19361856,
      "step": 91735
    },
    {
      "epoch": 10.092409240924093,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.01736642720904956,
      "loss": 0.2298,
      "num_input_tokens_seen": 19362912,
      "step": 91740
    },
    {
      "epoch": 10.092959295929592,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.017365005190858183,
      "loss": 0.2309,
      "num_input_tokens_seen": 19364000,
      "step": 91745
    },
    {
      "epoch": 10.093509350935093,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.01736358315086969,
      "loss": 0.2319,
      "num_input_tokens_seen": 19365152,
      "step": 91750
    },
    {
      "epoch": 10.094059405940595,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.01736216108909719,
      "loss": 0.2298,
      "num_input_tokens_seen": 19366208,
      "step": 91755
    },
    {
      "epoch": 10.094609460946094,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.017360739005553776,
      "loss": 0.2309,
      "num_input_tokens_seen": 19367264,
      "step": 91760
    },
    {
      "epoch": 10.095159515951595,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.01735931690025257,
      "loss": 0.2314,
      "num_input_tokens_seen": 19368256,
      "step": 91765
    },
    {
      "epoch": 10.095709570957096,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01735789477320667,
      "loss": 0.2325,
      "num_input_tokens_seen": 19369344,
      "step": 91770
    },
    {
      "epoch": 10.096259625962595,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.017356472624429187,
      "loss": 0.2319,
      "num_input_tokens_seen": 19370400,
      "step": 91775
    },
    {
      "epoch": 10.096809680968097,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.017355050453933228,
      "loss": 0.2304,
      "num_input_tokens_seen": 19371456,
      "step": 91780
    },
    {
      "epoch": 10.097359735973598,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.017353628261731896,
      "loss": 0.2319,
      "num_input_tokens_seen": 19372544,
      "step": 91785
    },
    {
      "epoch": 10.097909790979099,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.017352206047838298,
      "loss": 0.2325,
      "num_input_tokens_seen": 19373600,
      "step": 91790
    },
    {
      "epoch": 10.098459845984598,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.017350783812265555,
      "loss": 0.2304,
      "num_input_tokens_seen": 19374688,
      "step": 91795
    },
    {
      "epoch": 10.099009900990099,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.017349361555026758,
      "loss": 0.233,
      "num_input_tokens_seen": 19375712,
      "step": 91800
    },
    {
      "epoch": 10.0995599559956,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.017347939276135024,
      "loss": 0.2314,
      "num_input_tokens_seen": 19376768,
      "step": 91805
    },
    {
      "epoch": 10.1001100110011,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.017346516975603462,
      "loss": 0.2293,
      "num_input_tokens_seen": 19377856,
      "step": 91810
    },
    {
      "epoch": 10.1006600660066,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01734509465344518,
      "loss": 0.2304,
      "num_input_tokens_seen": 19378944,
      "step": 91815
    },
    {
      "epoch": 10.101210121012102,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.017343672309673284,
      "loss": 0.2309,
      "num_input_tokens_seen": 19379968,
      "step": 91820
    },
    {
      "epoch": 10.101760176017601,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.017342249944300888,
      "loss": 0.2319,
      "num_input_tokens_seen": 19381024,
      "step": 91825
    },
    {
      "epoch": 10.102310231023102,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.017340827557341093,
      "loss": 0.2298,
      "num_input_tokens_seen": 19382112,
      "step": 91830
    },
    {
      "epoch": 10.102860286028603,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.017339405148807014,
      "loss": 0.2319,
      "num_input_tokens_seen": 19383104,
      "step": 91835
    },
    {
      "epoch": 10.103410341034103,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.017337982718711765,
      "loss": 0.2319,
      "num_input_tokens_seen": 19384160,
      "step": 91840
    },
    {
      "epoch": 10.103960396039604,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.017336560267068448,
      "loss": 0.2309,
      "num_input_tokens_seen": 19385184,
      "step": 91845
    },
    {
      "epoch": 10.104510451045105,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01733513779389018,
      "loss": 0.2288,
      "num_input_tokens_seen": 19386272,
      "step": 91850
    },
    {
      "epoch": 10.105060506050606,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.01733371529919006,
      "loss": 0.2325,
      "num_input_tokens_seen": 19387360,
      "step": 91855
    },
    {
      "epoch": 10.105610561056105,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.01733229278298121,
      "loss": 0.2314,
      "num_input_tokens_seen": 19388384,
      "step": 91860
    },
    {
      "epoch": 10.106160616061606,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01733087024527674,
      "loss": 0.2319,
      "num_input_tokens_seen": 19389440,
      "step": 91865
    },
    {
      "epoch": 10.106710671067107,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.017329447686089756,
      "loss": 0.2299,
      "num_input_tokens_seen": 19390432,
      "step": 91870
    },
    {
      "epoch": 10.107260726072607,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.017328025105433364,
      "loss": 0.2366,
      "num_input_tokens_seen": 19391424,
      "step": 91875
    },
    {
      "epoch": 10.107810781078108,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.017326602503320687,
      "loss": 0.2319,
      "num_input_tokens_seen": 19392480,
      "step": 91880
    },
    {
      "epoch": 10.108360836083609,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.017325179879764833,
      "loss": 0.2314,
      "num_input_tokens_seen": 19393600,
      "step": 91885
    },
    {
      "epoch": 10.108910891089108,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.01732375723477891,
      "loss": 0.2324,
      "num_input_tokens_seen": 19394624,
      "step": 91890
    },
    {
      "epoch": 10.10946094609461,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.017322334568376032,
      "loss": 0.2324,
      "num_input_tokens_seen": 19395680,
      "step": 91895
    },
    {
      "epoch": 10.11001100110011,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.017320911880569306,
      "loss": 0.2314,
      "num_input_tokens_seen": 19396768,
      "step": 91900
    },
    {
      "epoch": 10.110561056105611,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.01731948917137185,
      "loss": 0.2329,
      "num_input_tokens_seen": 19397824,
      "step": 91905
    },
    {
      "epoch": 10.11111111111111,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01731806644079678,
      "loss": 0.2304,
      "num_input_tokens_seen": 19398944,
      "step": 91910
    },
    {
      "epoch": 10.111661166116612,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.017316643688857204,
      "loss": 0.2324,
      "num_input_tokens_seen": 19400128,
      "step": 91915
    },
    {
      "epoch": 10.112211221122113,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01731522091556623,
      "loss": 0.2335,
      "num_input_tokens_seen": 19401184,
      "step": 91920
    },
    {
      "epoch": 10.112761276127612,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.01731379812093698,
      "loss": 0.2319,
      "num_input_tokens_seen": 19402240,
      "step": 91925
    },
    {
      "epoch": 10.113311331133113,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.01731237530498256,
      "loss": 0.2309,
      "num_input_tokens_seen": 19403328,
      "step": 91930
    },
    {
      "epoch": 10.113861386138614,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.01731095246771609,
      "loss": 0.2308,
      "num_input_tokens_seen": 19404320,
      "step": 91935
    },
    {
      "epoch": 10.114411441144114,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01730952960915068,
      "loss": 0.2293,
      "num_input_tokens_seen": 19405376,
      "step": 91940
    },
    {
      "epoch": 10.114961496149615,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.017308106729299437,
      "loss": 0.2304,
      "num_input_tokens_seen": 19406464,
      "step": 91945
    },
    {
      "epoch": 10.115511551155116,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.017306683828175486,
      "loss": 0.2324,
      "num_input_tokens_seen": 19407552,
      "step": 91950
    },
    {
      "epoch": 10.116061606160615,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.01730526090579194,
      "loss": 0.2298,
      "num_input_tokens_seen": 19408640,
      "step": 91955
    },
    {
      "epoch": 10.116611661166116,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.017303837962161908,
      "loss": 0.2319,
      "num_input_tokens_seen": 19409728,
      "step": 91960
    },
    {
      "epoch": 10.117161716171617,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.017302414997298506,
      "loss": 0.2303,
      "num_input_tokens_seen": 19410752,
      "step": 91965
    },
    {
      "epoch": 10.117711771177119,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.017300992011214853,
      "loss": 0.2324,
      "num_input_tokens_seen": 19411712,
      "step": 91970
    },
    {
      "epoch": 10.118261826182618,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.017299569003924053,
      "loss": 0.2303,
      "num_input_tokens_seen": 19412704,
      "step": 91975
    },
    {
      "epoch": 10.118811881188119,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.017298145975439237,
      "loss": 0.2309,
      "num_input_tokens_seen": 19413760,
      "step": 91980
    },
    {
      "epoch": 10.11936193619362,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.01729672292577351,
      "loss": 0.2335,
      "num_input_tokens_seen": 19414784,
      "step": 91985
    },
    {
      "epoch": 10.11991199119912,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.017295299854939987,
      "loss": 0.2288,
      "num_input_tokens_seen": 19415872,
      "step": 91990
    },
    {
      "epoch": 10.12046204620462,
      "grad_norm": 0.00064849853515625,
      "learning_rate": 0.01729387676295179,
      "loss": 0.233,
      "num_input_tokens_seen": 19416896,
      "step": 91995
    },
    {
      "epoch": 10.121012101210122,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.017292453649822032,
      "loss": 0.2293,
      "num_input_tokens_seen": 19417984,
      "step": 92000
    },
    {
      "epoch": 10.12156215621562,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.017291030515563823,
      "loss": 0.2304,
      "num_input_tokens_seen": 19419072,
      "step": 92005
    },
    {
      "epoch": 10.122112211221122,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.017289607360190296,
      "loss": 0.2335,
      "num_input_tokens_seen": 19420160,
      "step": 92010
    },
    {
      "epoch": 10.122662266226623,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.01728818418371455,
      "loss": 0.2288,
      "num_input_tokens_seen": 19421216,
      "step": 92015
    },
    {
      "epoch": 10.123212321232122,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.017286760986149702,
      "loss": 0.234,
      "num_input_tokens_seen": 19422304,
      "step": 92020
    },
    {
      "epoch": 10.123762376237623,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.017285337767508884,
      "loss": 0.232,
      "num_input_tokens_seen": 19423328,
      "step": 92025
    },
    {
      "epoch": 10.124312431243125,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.017283914527805203,
      "loss": 0.2314,
      "num_input_tokens_seen": 19424384,
      "step": 92030
    },
    {
      "epoch": 10.124862486248626,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.017282491267051777,
      "loss": 0.2294,
      "num_input_tokens_seen": 19425472,
      "step": 92035
    },
    {
      "epoch": 10.125412541254125,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.017281067985261724,
      "loss": 0.2335,
      "num_input_tokens_seen": 19426464,
      "step": 92040
    },
    {
      "epoch": 10.125962596259626,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.017279644682448164,
      "loss": 0.2304,
      "num_input_tokens_seen": 19427488,
      "step": 92045
    },
    {
      "epoch": 10.126512651265127,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.017278221358624207,
      "loss": 0.2309,
      "num_input_tokens_seen": 19428480,
      "step": 92050
    },
    {
      "epoch": 10.127062706270626,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.017276798013802983,
      "loss": 0.2304,
      "num_input_tokens_seen": 19429472,
      "step": 92055
    },
    {
      "epoch": 10.127612761276128,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.017275374647997603,
      "loss": 0.2314,
      "num_input_tokens_seen": 19430528,
      "step": 92060
    },
    {
      "epoch": 10.128162816281629,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.017273951261221183,
      "loss": 0.2325,
      "num_input_tokens_seen": 19431552,
      "step": 92065
    },
    {
      "epoch": 10.128712871287128,
      "grad_norm": 0.0028839111328125,
      "learning_rate": 0.01727252785348685,
      "loss": 0.2324,
      "num_input_tokens_seen": 19432576,
      "step": 92070
    },
    {
      "epoch": 10.129262926292629,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.017271104424807716,
      "loss": 0.2314,
      "num_input_tokens_seen": 19433632,
      "step": 92075
    },
    {
      "epoch": 10.12981298129813,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0172696809751969,
      "loss": 0.2319,
      "num_input_tokens_seen": 19434688,
      "step": 92080
    },
    {
      "epoch": 10.130363036303631,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.01726825750466753,
      "loss": 0.2325,
      "num_input_tokens_seen": 19435680,
      "step": 92085
    },
    {
      "epoch": 10.13091309130913,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.017266834013232715,
      "loss": 0.2314,
      "num_input_tokens_seen": 19436672,
      "step": 92090
    },
    {
      "epoch": 10.131463146314632,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.017265410500905572,
      "loss": 0.2304,
      "num_input_tokens_seen": 19437728,
      "step": 92095
    },
    {
      "epoch": 10.132013201320133,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.01726398696769924,
      "loss": 0.2283,
      "num_input_tokens_seen": 19438848,
      "step": 92100
    },
    {
      "epoch": 10.132563256325632,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.01726256341362682,
      "loss": 0.2329,
      "num_input_tokens_seen": 19439840,
      "step": 92105
    },
    {
      "epoch": 10.133113311331133,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01726113983870144,
      "loss": 0.2319,
      "num_input_tokens_seen": 19440960,
      "step": 92110
    },
    {
      "epoch": 10.133663366336634,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.017259716242936218,
      "loss": 0.2293,
      "num_input_tokens_seen": 19442016,
      "step": 92115
    },
    {
      "epoch": 10.134213421342134,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.017258292626344277,
      "loss": 0.2314,
      "num_input_tokens_seen": 19443104,
      "step": 92120
    },
    {
      "epoch": 10.134763476347635,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01725686898893873,
      "loss": 0.2314,
      "num_input_tokens_seen": 19444160,
      "step": 92125
    },
    {
      "epoch": 10.135313531353136,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.017255445330732713,
      "loss": 0.2319,
      "num_input_tokens_seen": 19445248,
      "step": 92130
    },
    {
      "epoch": 10.135863586358635,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.017254021651739334,
      "loss": 0.2299,
      "num_input_tokens_seen": 19446368,
      "step": 92135
    },
    {
      "epoch": 10.136413641364136,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.017252597951971725,
      "loss": 0.2314,
      "num_input_tokens_seen": 19447360,
      "step": 92140
    },
    {
      "epoch": 10.136963696369637,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.017251174231442998,
      "loss": 0.2319,
      "num_input_tokens_seen": 19448416,
      "step": 92145
    },
    {
      "epoch": 10.137513751375138,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.017249750490166277,
      "loss": 0.2314,
      "num_input_tokens_seen": 19449504,
      "step": 92150
    },
    {
      "epoch": 10.138063806380638,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.017248326728154687,
      "loss": 0.2325,
      "num_input_tokens_seen": 19450528,
      "step": 92155
    },
    {
      "epoch": 10.138613861386139,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.017246902945421347,
      "loss": 0.234,
      "num_input_tokens_seen": 19451584,
      "step": 92160
    },
    {
      "epoch": 10.13916391639164,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.017245479141979377,
      "loss": 0.2314,
      "num_input_tokens_seen": 19452640,
      "step": 92165
    },
    {
      "epoch": 10.13971397139714,
      "grad_norm": 0.0027923583984375,
      "learning_rate": 0.01724405531784191,
      "loss": 0.2309,
      "num_input_tokens_seen": 19453760,
      "step": 92170
    },
    {
      "epoch": 10.14026402640264,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.01724263147302206,
      "loss": 0.2314,
      "num_input_tokens_seen": 19454880,
      "step": 92175
    },
    {
      "epoch": 10.140814081408141,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.01724120760753295,
      "loss": 0.233,
      "num_input_tokens_seen": 19455968,
      "step": 92180
    },
    {
      "epoch": 10.14136413641364,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.017239783721387705,
      "loss": 0.2314,
      "num_input_tokens_seen": 19456992,
      "step": 92185
    },
    {
      "epoch": 10.141914191419142,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.017238359814599452,
      "loss": 0.2324,
      "num_input_tokens_seen": 19457984,
      "step": 92190
    },
    {
      "epoch": 10.142464246424643,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.017236935887181306,
      "loss": 0.234,
      "num_input_tokens_seen": 19459040,
      "step": 92195
    },
    {
      "epoch": 10.143014301430142,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.0172355119391464,
      "loss": 0.2309,
      "num_input_tokens_seen": 19460032,
      "step": 92200
    },
    {
      "epoch": 10.143564356435643,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01723408797050785,
      "loss": 0.2303,
      "num_input_tokens_seen": 19461056,
      "step": 92205
    },
    {
      "epoch": 10.144114411441144,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.017232663981278783,
      "loss": 0.2293,
      "num_input_tokens_seen": 19462112,
      "step": 92210
    },
    {
      "epoch": 10.144664466446645,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.017231239971472325,
      "loss": 0.2309,
      "num_input_tokens_seen": 19463200,
      "step": 92215
    },
    {
      "epoch": 10.145214521452145,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.017229815941101595,
      "loss": 0.2298,
      "num_input_tokens_seen": 19464224,
      "step": 92220
    },
    {
      "epoch": 10.145764576457646,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.017228391890179724,
      "loss": 0.2309,
      "num_input_tokens_seen": 19465280,
      "step": 92225
    },
    {
      "epoch": 10.146314631463147,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.017226967818719838,
      "loss": 0.2314,
      "num_input_tokens_seen": 19466368,
      "step": 92230
    },
    {
      "epoch": 10.146864686468646,
      "grad_norm": 0.005859375,
      "learning_rate": 0.017225543726735053,
      "loss": 0.2298,
      "num_input_tokens_seen": 19467456,
      "step": 92235
    },
    {
      "epoch": 10.147414741474147,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.0172241196142385,
      "loss": 0.2325,
      "num_input_tokens_seen": 19468512,
      "step": 92240
    },
    {
      "epoch": 10.147964796479648,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.017222695481243305,
      "loss": 0.2303,
      "num_input_tokens_seen": 19469504,
      "step": 92245
    },
    {
      "epoch": 10.148514851485148,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.01722127132776259,
      "loss": 0.2303,
      "num_input_tokens_seen": 19470560,
      "step": 92250
    },
    {
      "epoch": 10.149064906490649,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.017219847153809486,
      "loss": 0.2319,
      "num_input_tokens_seen": 19471552,
      "step": 92255
    },
    {
      "epoch": 10.14961496149615,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.017218422959397118,
      "loss": 0.2298,
      "num_input_tokens_seen": 19472608,
      "step": 92260
    },
    {
      "epoch": 10.150165016501651,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.017216998744538605,
      "loss": 0.2298,
      "num_input_tokens_seen": 19473632,
      "step": 92265
    },
    {
      "epoch": 10.15071507150715,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.017215574509247083,
      "loss": 0.2319,
      "num_input_tokens_seen": 19474592,
      "step": 92270
    },
    {
      "epoch": 10.151265126512651,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.017214150253535673,
      "loss": 0.2314,
      "num_input_tokens_seen": 19475616,
      "step": 92275
    },
    {
      "epoch": 10.151815181518153,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.017212725977417496,
      "loss": 0.2335,
      "num_input_tokens_seen": 19476672,
      "step": 92280
    },
    {
      "epoch": 10.152365236523652,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.01721130168090569,
      "loss": 0.2314,
      "num_input_tokens_seen": 19477760,
      "step": 92285
    },
    {
      "epoch": 10.152915291529153,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.01720987736401338,
      "loss": 0.2309,
      "num_input_tokens_seen": 19478816,
      "step": 92290
    },
    {
      "epoch": 10.153465346534654,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.01720845302675369,
      "loss": 0.2278,
      "num_input_tokens_seen": 19479840,
      "step": 92295
    },
    {
      "epoch": 10.154015401540153,
      "grad_norm": 0.0030364990234375,
      "learning_rate": 0.01720702866913975,
      "loss": 0.2309,
      "num_input_tokens_seen": 19480928,
      "step": 92300
    },
    {
      "epoch": 10.154565456545654,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01720560429118468,
      "loss": 0.2324,
      "num_input_tokens_seen": 19481984,
      "step": 92305
    },
    {
      "epoch": 10.155115511551156,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.017204179892901614,
      "loss": 0.2309,
      "num_input_tokens_seen": 19483072,
      "step": 92310
    },
    {
      "epoch": 10.155665566556655,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.017202755474303683,
      "loss": 0.233,
      "num_input_tokens_seen": 19484096,
      "step": 92315
    },
    {
      "epoch": 10.156215621562156,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.017201331035404013,
      "loss": 0.2288,
      "num_input_tokens_seen": 19485216,
      "step": 92320
    },
    {
      "epoch": 10.156765676567657,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01719990657621573,
      "loss": 0.2325,
      "num_input_tokens_seen": 19486208,
      "step": 92325
    },
    {
      "epoch": 10.157315731573158,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.017198482096751958,
      "loss": 0.233,
      "num_input_tokens_seen": 19487328,
      "step": 92330
    },
    {
      "epoch": 10.157865786578657,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.01719705759702584,
      "loss": 0.2314,
      "num_input_tokens_seen": 19488416,
      "step": 92335
    },
    {
      "epoch": 10.158415841584159,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.017195633077050486,
      "loss": 0.2314,
      "num_input_tokens_seen": 19489408,
      "step": 92340
    },
    {
      "epoch": 10.15896589658966,
      "grad_norm": 0.00347900390625,
      "learning_rate": 0.017194208536839048,
      "loss": 0.2303,
      "num_input_tokens_seen": 19490496,
      "step": 92345
    },
    {
      "epoch": 10.159515951595159,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.017192783976404637,
      "loss": 0.2315,
      "num_input_tokens_seen": 19491648,
      "step": 92350
    },
    {
      "epoch": 10.16006600660066,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.017191359395760383,
      "loss": 0.2283,
      "num_input_tokens_seen": 19492672,
      "step": 92355
    },
    {
      "epoch": 10.160616061606161,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01718993479491943,
      "loss": 0.233,
      "num_input_tokens_seen": 19493760,
      "step": 92360
    },
    {
      "epoch": 10.16116611661166,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.017188510173894896,
      "loss": 0.2288,
      "num_input_tokens_seen": 19494784,
      "step": 92365
    },
    {
      "epoch": 10.161716171617162,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.01718708553269991,
      "loss": 0.2335,
      "num_input_tokens_seen": 19495776,
      "step": 92370
    },
    {
      "epoch": 10.162266226622663,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.017185660871347607,
      "loss": 0.2304,
      "num_input_tokens_seen": 19496800,
      "step": 92375
    },
    {
      "epoch": 10.162816281628162,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.017184236189851123,
      "loss": 0.2309,
      "num_input_tokens_seen": 19497824,
      "step": 92380
    },
    {
      "epoch": 10.163366336633663,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.017182811488223572,
      "loss": 0.233,
      "num_input_tokens_seen": 19498912,
      "step": 92385
    },
    {
      "epoch": 10.163916391639164,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.017181386766478103,
      "loss": 0.2298,
      "num_input_tokens_seen": 19499968,
      "step": 92390
    },
    {
      "epoch": 10.164466446644665,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.017179962024627838,
      "loss": 0.2303,
      "num_input_tokens_seen": 19500992,
      "step": 92395
    },
    {
      "epoch": 10.165016501650165,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.017178537262685905,
      "loss": 0.234,
      "num_input_tokens_seen": 19502048,
      "step": 92400
    },
    {
      "epoch": 10.165566556655666,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.017177112480665443,
      "loss": 0.2309,
      "num_input_tokens_seen": 19503040,
      "step": 92405
    },
    {
      "epoch": 10.166116611661167,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.01717568767857958,
      "loss": 0.2309,
      "num_input_tokens_seen": 19504032,
      "step": 92410
    },
    {
      "epoch": 10.166666666666666,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.017174262856441445,
      "loss": 0.2314,
      "num_input_tokens_seen": 19505088,
      "step": 92415
    },
    {
      "epoch": 10.167216721672167,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.017172838014264176,
      "loss": 0.2304,
      "num_input_tokens_seen": 19506112,
      "step": 92420
    },
    {
      "epoch": 10.167766776677668,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0171714131520609,
      "loss": 0.2324,
      "num_input_tokens_seen": 19507104,
      "step": 92425
    },
    {
      "epoch": 10.168316831683168,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.017169988269844748,
      "loss": 0.2319,
      "num_input_tokens_seen": 19508192,
      "step": 92430
    },
    {
      "epoch": 10.168866886688669,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.01716856336762886,
      "loss": 0.2325,
      "num_input_tokens_seen": 19509248,
      "step": 92435
    },
    {
      "epoch": 10.16941694169417,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01716713844542636,
      "loss": 0.233,
      "num_input_tokens_seen": 19510304,
      "step": 92440
    },
    {
      "epoch": 10.16996699669967,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.017165713503250385,
      "loss": 0.2298,
      "num_input_tokens_seen": 19511360,
      "step": 92445
    },
    {
      "epoch": 10.17051705170517,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.01716428854111407,
      "loss": 0.2309,
      "num_input_tokens_seen": 19512320,
      "step": 92450
    },
    {
      "epoch": 10.171067106710671,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.017162863559030548,
      "loss": 0.2298,
      "num_input_tokens_seen": 19513312,
      "step": 92455
    },
    {
      "epoch": 10.171617161716172,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.017161438557012947,
      "loss": 0.2319,
      "num_input_tokens_seen": 19514304,
      "step": 92460
    },
    {
      "epoch": 10.172167216721672,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.017160013535074407,
      "loss": 0.2314,
      "num_input_tokens_seen": 19515392,
      "step": 92465
    },
    {
      "epoch": 10.172717271727173,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.017158588493228055,
      "loss": 0.2309,
      "num_input_tokens_seen": 19516448,
      "step": 92470
    },
    {
      "epoch": 10.173267326732674,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01715716343148703,
      "loss": 0.2314,
      "num_input_tokens_seen": 19517504,
      "step": 92475
    },
    {
      "epoch": 10.173817381738173,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.017155738349864464,
      "loss": 0.233,
      "num_input_tokens_seen": 19518592,
      "step": 92480
    },
    {
      "epoch": 10.174367436743674,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.017154313248373496,
      "loss": 0.2335,
      "num_input_tokens_seen": 19519648,
      "step": 92485
    },
    {
      "epoch": 10.174917491749175,
      "grad_norm": 0.0029449462890625,
      "learning_rate": 0.01715288812702725,
      "loss": 0.233,
      "num_input_tokens_seen": 19520704,
      "step": 92490
    },
    {
      "epoch": 10.175467546754675,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.01715146298583887,
      "loss": 0.2314,
      "num_input_tokens_seen": 19521728,
      "step": 92495
    },
    {
      "epoch": 10.176017601760176,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.017150037824821492,
      "loss": 0.2298,
      "num_input_tokens_seen": 19522784,
      "step": 92500
    },
    {
      "epoch": 10.176567656765677,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.017148612643988242,
      "loss": 0.2298,
      "num_input_tokens_seen": 19523872,
      "step": 92505
    },
    {
      "epoch": 10.177117711771178,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.017147187443352266,
      "loss": 0.2293,
      "num_input_tokens_seen": 19524960,
      "step": 92510
    },
    {
      "epoch": 10.177667766776677,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.01714576222292669,
      "loss": 0.2303,
      "num_input_tokens_seen": 19526048,
      "step": 92515
    },
    {
      "epoch": 10.178217821782178,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.01714433698272465,
      "loss": 0.234,
      "num_input_tokens_seen": 19527104,
      "step": 92520
    },
    {
      "epoch": 10.17876787678768,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.01714291172275929,
      "loss": 0.2335,
      "num_input_tokens_seen": 19528160,
      "step": 92525
    },
    {
      "epoch": 10.179317931793179,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.017141486443043735,
      "loss": 0.2335,
      "num_input_tokens_seen": 19529152,
      "step": 92530
    },
    {
      "epoch": 10.17986798679868,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.017140061143591137,
      "loss": 0.2314,
      "num_input_tokens_seen": 19530176,
      "step": 92535
    },
    {
      "epoch": 10.180418041804181,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.017138635824414616,
      "loss": 0.2298,
      "num_input_tokens_seen": 19531232,
      "step": 92540
    },
    {
      "epoch": 10.18096809680968,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.017137210485527313,
      "loss": 0.2319,
      "num_input_tokens_seen": 19532288,
      "step": 92545
    },
    {
      "epoch": 10.181518151815181,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.017135785126942367,
      "loss": 0.2324,
      "num_input_tokens_seen": 19533280,
      "step": 92550
    },
    {
      "epoch": 10.182068206820682,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.017134359748672917,
      "loss": 0.2309,
      "num_input_tokens_seen": 19534400,
      "step": 92555
    },
    {
      "epoch": 10.182618261826182,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.017132934350732095,
      "loss": 0.2298,
      "num_input_tokens_seen": 19535424,
      "step": 92560
    },
    {
      "epoch": 10.183168316831683,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.017131508933133043,
      "loss": 0.2335,
      "num_input_tokens_seen": 19536448,
      "step": 92565
    },
    {
      "epoch": 10.183718371837184,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.017130083495888897,
      "loss": 0.234,
      "num_input_tokens_seen": 19537536,
      "step": 92570
    },
    {
      "epoch": 10.184268426842685,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.017128658039012788,
      "loss": 0.2314,
      "num_input_tokens_seen": 19538624,
      "step": 92575
    },
    {
      "epoch": 10.184818481848184,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.01712723256251787,
      "loss": 0.2309,
      "num_input_tokens_seen": 19539680,
      "step": 92580
    },
    {
      "epoch": 10.185368536853685,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.017125807066417257,
      "loss": 0.2329,
      "num_input_tokens_seen": 19540704,
      "step": 92585
    },
    {
      "epoch": 10.185918591859187,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.017124381550724106,
      "loss": 0.2303,
      "num_input_tokens_seen": 19541760,
      "step": 92590
    },
    {
      "epoch": 10.186468646864686,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01712295601545155,
      "loss": 0.2324,
      "num_input_tokens_seen": 19542880,
      "step": 92595
    },
    {
      "epoch": 10.187018701870187,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.01712153046061273,
      "loss": 0.2324,
      "num_input_tokens_seen": 19543872,
      "step": 92600
    },
    {
      "epoch": 10.187568756875688,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.017120104886220778,
      "loss": 0.2288,
      "num_input_tokens_seen": 19544928,
      "step": 92605
    },
    {
      "epoch": 10.188118811881187,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.017118679292288837,
      "loss": 0.2304,
      "num_input_tokens_seen": 19545920,
      "step": 92610
    },
    {
      "epoch": 10.188668866886688,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.017117253678830043,
      "loss": 0.2314,
      "num_input_tokens_seen": 19546944,
      "step": 92615
    },
    {
      "epoch": 10.18921892189219,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.017115828045857543,
      "loss": 0.2309,
      "num_input_tokens_seen": 19548032,
      "step": 92620
    },
    {
      "epoch": 10.189768976897689,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.017114402393384468,
      "loss": 0.2335,
      "num_input_tokens_seen": 19549120,
      "step": 92625
    },
    {
      "epoch": 10.19031903190319,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.017112976721423964,
      "loss": 0.2314,
      "num_input_tokens_seen": 19550144,
      "step": 92630
    },
    {
      "epoch": 10.190869086908691,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.017111551029989163,
      "loss": 0.232,
      "num_input_tokens_seen": 19551232,
      "step": 92635
    },
    {
      "epoch": 10.191419141914192,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.017110125319093217,
      "loss": 0.2314,
      "num_input_tokens_seen": 19552224,
      "step": 92640
    },
    {
      "epoch": 10.191969196919691,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.01710869958874925,
      "loss": 0.233,
      "num_input_tokens_seen": 19553248,
      "step": 92645
    },
    {
      "epoch": 10.192519251925193,
      "grad_norm": 0.005859375,
      "learning_rate": 0.017107273838970417,
      "loss": 0.2335,
      "num_input_tokens_seen": 19554272,
      "step": 92650
    },
    {
      "epoch": 10.193069306930694,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.017105848069769846,
      "loss": 0.233,
      "num_input_tokens_seen": 19555328,
      "step": 92655
    },
    {
      "epoch": 10.193619361936193,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.017104422281160686,
      "loss": 0.2324,
      "num_input_tokens_seen": 19556384,
      "step": 92660
    },
    {
      "epoch": 10.194169416941694,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.017102996473156076,
      "loss": 0.2308,
      "num_input_tokens_seen": 19557408,
      "step": 92665
    },
    {
      "epoch": 10.194719471947195,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01710157064576916,
      "loss": 0.2293,
      "num_input_tokens_seen": 19558432,
      "step": 92670
    },
    {
      "epoch": 10.195269526952695,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.017100144799013072,
      "loss": 0.2294,
      "num_input_tokens_seen": 19559552,
      "step": 92675
    },
    {
      "epoch": 10.195819581958196,
      "grad_norm": 0.0030517578125,
      "learning_rate": 0.017098718932900957,
      "loss": 0.2325,
      "num_input_tokens_seen": 19560672,
      "step": 92680
    },
    {
      "epoch": 10.196369636963697,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.017097293047445956,
      "loss": 0.2268,
      "num_input_tokens_seen": 19561792,
      "step": 92685
    },
    {
      "epoch": 10.196919691969198,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01709586714266121,
      "loss": 0.2283,
      "num_input_tokens_seen": 19562848,
      "step": 92690
    },
    {
      "epoch": 10.197469746974697,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01709444121855987,
      "loss": 0.2309,
      "num_input_tokens_seen": 19563936,
      "step": 92695
    },
    {
      "epoch": 10.198019801980198,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.01709301527515506,
      "loss": 0.2294,
      "num_input_tokens_seen": 19565024,
      "step": 92700
    },
    {
      "epoch": 10.1985698569857,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.017091589312459944,
      "loss": 0.2321,
      "num_input_tokens_seen": 19566080,
      "step": 92705
    },
    {
      "epoch": 10.199119911991199,
      "grad_norm": 0.0026397705078125,
      "learning_rate": 0.017090163330487646,
      "loss": 0.2294,
      "num_input_tokens_seen": 19567136,
      "step": 92710
    },
    {
      "epoch": 10.1996699669967,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.017088737329251317,
      "loss": 0.2279,
      "num_input_tokens_seen": 19568192,
      "step": 92715
    },
    {
      "epoch": 10.2002200220022,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.017087311308764095,
      "loss": 0.229,
      "num_input_tokens_seen": 19569280,
      "step": 92720
    },
    {
      "epoch": 10.2007700770077,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.017085885269039133,
      "loss": 0.2321,
      "num_input_tokens_seen": 19570336,
      "step": 92725
    },
    {
      "epoch": 10.201320132013201,
      "grad_norm": 0.006988525390625,
      "learning_rate": 0.01708445921008956,
      "loss": 0.2337,
      "num_input_tokens_seen": 19571328,
      "step": 92730
    },
    {
      "epoch": 10.201870187018702,
      "grad_norm": 0.004425048828125,
      "learning_rate": 0.01708303313192853,
      "loss": 0.2311,
      "num_input_tokens_seen": 19572352,
      "step": 92735
    },
    {
      "epoch": 10.202420242024202,
      "grad_norm": 0.007110595703125,
      "learning_rate": 0.017081607034569183,
      "loss": 0.2342,
      "num_input_tokens_seen": 19573440,
      "step": 92740
    },
    {
      "epoch": 10.202970297029703,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.017080180918024664,
      "loss": 0.2342,
      "num_input_tokens_seen": 19574432,
      "step": 92745
    },
    {
      "epoch": 10.203520352035204,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.01707875478230812,
      "loss": 0.2326,
      "num_input_tokens_seen": 19575488,
      "step": 92750
    },
    {
      "epoch": 10.204070407040705,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.017077328627432685,
      "loss": 0.231,
      "num_input_tokens_seen": 19576512,
      "step": 92755
    },
    {
      "epoch": 10.204620462046204,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.017075902453411503,
      "loss": 0.2336,
      "num_input_tokens_seen": 19577600,
      "step": 92760
    },
    {
      "epoch": 10.205170517051705,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.017074476260257734,
      "loss": 0.23,
      "num_input_tokens_seen": 19578592,
      "step": 92765
    },
    {
      "epoch": 10.205720572057206,
      "grad_norm": 0.001953125,
      "learning_rate": 0.01707305004798451,
      "loss": 0.2321,
      "num_input_tokens_seen": 19579680,
      "step": 92770
    },
    {
      "epoch": 10.206270627062706,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.017071623816604976,
      "loss": 0.2341,
      "num_input_tokens_seen": 19580704,
      "step": 92775
    },
    {
      "epoch": 10.206820682068207,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.017070197566132282,
      "loss": 0.2315,
      "num_input_tokens_seen": 19581696,
      "step": 92780
    },
    {
      "epoch": 10.207370737073708,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.01706877129657957,
      "loss": 0.231,
      "num_input_tokens_seen": 19582752,
      "step": 92785
    },
    {
      "epoch": 10.207920792079207,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.01706734500795998,
      "loss": 0.2352,
      "num_input_tokens_seen": 19583840,
      "step": 92790
    },
    {
      "epoch": 10.208470847084708,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.017065918700286672,
      "loss": 0.2351,
      "num_input_tokens_seen": 19584832,
      "step": 92795
    },
    {
      "epoch": 10.20902090209021,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.01706449237357278,
      "loss": 0.233,
      "num_input_tokens_seen": 19585920,
      "step": 92800
    },
    {
      "epoch": 10.209570957095709,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01706306602783145,
      "loss": 0.2319,
      "num_input_tokens_seen": 19586976,
      "step": 92805
    },
    {
      "epoch": 10.21012101210121,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.017061639663075834,
      "loss": 0.2293,
      "num_input_tokens_seen": 19588096,
      "step": 92810
    },
    {
      "epoch": 10.210671067106711,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.017060213279319072,
      "loss": 0.2324,
      "num_input_tokens_seen": 19589088,
      "step": 92815
    },
    {
      "epoch": 10.211221122112212,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.017058786876574313,
      "loss": 0.2314,
      "num_input_tokens_seen": 19590112,
      "step": 92820
    },
    {
      "epoch": 10.211771177117711,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.017057360454854703,
      "loss": 0.2324,
      "num_input_tokens_seen": 19591168,
      "step": 92825
    },
    {
      "epoch": 10.212321232123212,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.01705593401417339,
      "loss": 0.2309,
      "num_input_tokens_seen": 19592192,
      "step": 92830
    },
    {
      "epoch": 10.212871287128714,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.017054507554543515,
      "loss": 0.2298,
      "num_input_tokens_seen": 19593184,
      "step": 92835
    },
    {
      "epoch": 10.213421342134213,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.01705308107597824,
      "loss": 0.2277,
      "num_input_tokens_seen": 19594272,
      "step": 92840
    },
    {
      "epoch": 10.213971397139714,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.017051654578490696,
      "loss": 0.2293,
      "num_input_tokens_seen": 19595232,
      "step": 92845
    },
    {
      "epoch": 10.214521452145215,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.01705022806209403,
      "loss": 0.2314,
      "num_input_tokens_seen": 19596256,
      "step": 92850
    },
    {
      "epoch": 10.215071507150714,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0170488015268014,
      "loss": 0.2314,
      "num_input_tokens_seen": 19597280,
      "step": 92855
    },
    {
      "epoch": 10.215621562156215,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.017047374972625953,
      "loss": 0.2314,
      "num_input_tokens_seen": 19598336,
      "step": 92860
    },
    {
      "epoch": 10.216171617161717,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.017045948399580832,
      "loss": 0.2303,
      "num_input_tokens_seen": 19599360,
      "step": 92865
    },
    {
      "epoch": 10.216721672167218,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.017044521807679187,
      "loss": 0.2303,
      "num_input_tokens_seen": 19600384,
      "step": 92870
    },
    {
      "epoch": 10.217271727172717,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.017043095196934165,
      "loss": 0.2324,
      "num_input_tokens_seen": 19601408,
      "step": 92875
    },
    {
      "epoch": 10.217821782178218,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01704166856735891,
      "loss": 0.2298,
      "num_input_tokens_seen": 19602464,
      "step": 92880
    },
    {
      "epoch": 10.218371837183719,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.01704024191896658,
      "loss": 0.2314,
      "num_input_tokens_seen": 19603456,
      "step": 92885
    },
    {
      "epoch": 10.218921892189218,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01703881525177032,
      "loss": 0.2314,
      "num_input_tokens_seen": 19604576,
      "step": 92890
    },
    {
      "epoch": 10.21947194719472,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.01703738856578328,
      "loss": 0.2304,
      "num_input_tokens_seen": 19605696,
      "step": 92895
    },
    {
      "epoch": 10.22002200220022,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.017035961861018602,
      "loss": 0.2298,
      "num_input_tokens_seen": 19606816,
      "step": 92900
    },
    {
      "epoch": 10.22057205720572,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.017034535137489443,
      "loss": 0.2314,
      "num_input_tokens_seen": 19607904,
      "step": 92905
    },
    {
      "epoch": 10.221122112211221,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.017033108395208944,
      "loss": 0.2309,
      "num_input_tokens_seen": 19608928,
      "step": 92910
    },
    {
      "epoch": 10.221672167216722,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.01703168163419027,
      "loss": 0.2304,
      "num_input_tokens_seen": 19609920,
      "step": 92915
    },
    {
      "epoch": 10.222222222222221,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.017030254854446557,
      "loss": 0.2324,
      "num_input_tokens_seen": 19611008,
      "step": 92920
    },
    {
      "epoch": 10.222772277227723,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.017028828055990958,
      "loss": 0.2298,
      "num_input_tokens_seen": 19612064,
      "step": 92925
    },
    {
      "epoch": 10.223322332233224,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.017027401238836625,
      "loss": 0.2309,
      "num_input_tokens_seen": 19613120,
      "step": 92930
    },
    {
      "epoch": 10.223872387238725,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.017025974402996708,
      "loss": 0.2309,
      "num_input_tokens_seen": 19614208,
      "step": 92935
    },
    {
      "epoch": 10.224422442244224,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01702454754848436,
      "loss": 0.2273,
      "num_input_tokens_seen": 19615264,
      "step": 92940
    },
    {
      "epoch": 10.224972497249725,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.017023120675312725,
      "loss": 0.2304,
      "num_input_tokens_seen": 19616320,
      "step": 92945
    },
    {
      "epoch": 10.225522552255226,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.017021693783494957,
      "loss": 0.2303,
      "num_input_tokens_seen": 19617408,
      "step": 92950
    },
    {
      "epoch": 10.226072607260726,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.01702026687304421,
      "loss": 0.2314,
      "num_input_tokens_seen": 19618528,
      "step": 92955
    },
    {
      "epoch": 10.226622662266227,
      "grad_norm": 0.005859375,
      "learning_rate": 0.017018839943973636,
      "loss": 0.2309,
      "num_input_tokens_seen": 19619552,
      "step": 92960
    },
    {
      "epoch": 10.227172717271728,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.017017412996296374,
      "loss": 0.2314,
      "num_input_tokens_seen": 19620608,
      "step": 92965
    },
    {
      "epoch": 10.227722772277227,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.017015986030025593,
      "loss": 0.2314,
      "num_input_tokens_seen": 19621728,
      "step": 92970
    },
    {
      "epoch": 10.228272827282728,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.01701455904517443,
      "loss": 0.2351,
      "num_input_tokens_seen": 19622816,
      "step": 92975
    },
    {
      "epoch": 10.22882288228823,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.017013132041756043,
      "loss": 0.2304,
      "num_input_tokens_seen": 19623872,
      "step": 92980
    },
    {
      "epoch": 10.229372937293729,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.01701170501978359,
      "loss": 0.2299,
      "num_input_tokens_seen": 19624960,
      "step": 92985
    },
    {
      "epoch": 10.22992299229923,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.017010277979270218,
      "loss": 0.2309,
      "num_input_tokens_seen": 19626016,
      "step": 92990
    },
    {
      "epoch": 10.23047304730473,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.01700885092022907,
      "loss": 0.2309,
      "num_input_tokens_seen": 19627072,
      "step": 92995
    },
    {
      "epoch": 10.231023102310232,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.01700742384267331,
      "loss": 0.2331,
      "num_input_tokens_seen": 19628096,
      "step": 93000
    },
    {
      "epoch": 10.231573157315731,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.01700599674661609,
      "loss": 0.2294,
      "num_input_tokens_seen": 19629152,
      "step": 93005
    },
    {
      "epoch": 10.232123212321232,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.017004569632070556,
      "loss": 0.231,
      "num_input_tokens_seen": 19630176,
      "step": 93010
    },
    {
      "epoch": 10.232673267326733,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.017003142499049873,
      "loss": 0.2294,
      "num_input_tokens_seen": 19631200,
      "step": 93015
    },
    {
      "epoch": 10.233223322332233,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.01700171534756718,
      "loss": 0.2352,
      "num_input_tokens_seen": 19632192,
      "step": 93020
    },
    {
      "epoch": 10.233773377337734,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.01700028817763564,
      "loss": 0.2274,
      "num_input_tokens_seen": 19633216,
      "step": 93025
    },
    {
      "epoch": 10.234323432343235,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.016998860989268408,
      "loss": 0.2299,
      "num_input_tokens_seen": 19634272,
      "step": 93030
    },
    {
      "epoch": 10.234873487348734,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.016997433782478623,
      "loss": 0.2289,
      "num_input_tokens_seen": 19635296,
      "step": 93035
    },
    {
      "epoch": 10.235423542354235,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.01699600655727946,
      "loss": 0.23,
      "num_input_tokens_seen": 19636320,
      "step": 93040
    },
    {
      "epoch": 10.235973597359736,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.016994579313684057,
      "loss": 0.23,
      "num_input_tokens_seen": 19637408,
      "step": 93045
    },
    {
      "epoch": 10.236523652365236,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.01699315205170557,
      "loss": 0.2289,
      "num_input_tokens_seen": 19638464,
      "step": 93050
    },
    {
      "epoch": 10.237073707370737,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.016991724771357162,
      "loss": 0.2316,
      "num_input_tokens_seen": 19639456,
      "step": 93055
    },
    {
      "epoch": 10.237623762376238,
      "grad_norm": 0.01239013671875,
      "learning_rate": 0.016990297472651985,
      "loss": 0.2353,
      "num_input_tokens_seen": 19640512,
      "step": 93060
    },
    {
      "epoch": 10.238173817381739,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.016988870155603186,
      "loss": 0.2315,
      "num_input_tokens_seen": 19641504,
      "step": 93065
    },
    {
      "epoch": 10.238723872387238,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.016987442820223925,
      "loss": 0.2331,
      "num_input_tokens_seen": 19642528,
      "step": 93070
    },
    {
      "epoch": 10.23927392739274,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.01698601546652736,
      "loss": 0.2305,
      "num_input_tokens_seen": 19643584,
      "step": 93075
    },
    {
      "epoch": 10.23982398239824,
      "grad_norm": 0.01251220703125,
      "learning_rate": 0.016984588094526645,
      "loss": 0.2367,
      "num_input_tokens_seen": 19644672,
      "step": 93080
    },
    {
      "epoch": 10.24037403740374,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.01698316070423493,
      "loss": 0.23,
      "num_input_tokens_seen": 19645728,
      "step": 93085
    },
    {
      "epoch": 10.24092409240924,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.016981733295665376,
      "loss": 0.232,
      "num_input_tokens_seen": 19646848,
      "step": 93090
    },
    {
      "epoch": 10.241474147414742,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.016980305868831133,
      "loss": 0.2315,
      "num_input_tokens_seen": 19647904,
      "step": 93095
    },
    {
      "epoch": 10.242024202420241,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.016978878423745368,
      "loss": 0.2305,
      "num_input_tokens_seen": 19648960,
      "step": 93100
    },
    {
      "epoch": 10.242574257425742,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01697745096042123,
      "loss": 0.232,
      "num_input_tokens_seen": 19649984,
      "step": 93105
    },
    {
      "epoch": 10.243124312431243,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.01697602347887187,
      "loss": 0.2299,
      "num_input_tokens_seen": 19651072,
      "step": 93110
    },
    {
      "epoch": 10.243674367436745,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.01697459597911045,
      "loss": 0.2268,
      "num_input_tokens_seen": 19652064,
      "step": 93115
    },
    {
      "epoch": 10.244224422442244,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.016973168461150132,
      "loss": 0.2315,
      "num_input_tokens_seen": 19653088,
      "step": 93120
    },
    {
      "epoch": 10.244774477447745,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.016971740925004062,
      "loss": 0.2316,
      "num_input_tokens_seen": 19654112,
      "step": 93125
    },
    {
      "epoch": 10.245324532453246,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.016970313370685407,
      "loss": 0.2315,
      "num_input_tokens_seen": 19655200,
      "step": 93130
    },
    {
      "epoch": 10.245874587458745,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.016968885798207312,
      "loss": 0.2336,
      "num_input_tokens_seen": 19656256,
      "step": 93135
    },
    {
      "epoch": 10.246424642464246,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.016967458207582945,
      "loss": 0.232,
      "num_input_tokens_seen": 19657344,
      "step": 93140
    },
    {
      "epoch": 10.246974697469748,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.016966030598825458,
      "loss": 0.2325,
      "num_input_tokens_seen": 19658336,
      "step": 93145
    },
    {
      "epoch": 10.247524752475247,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.016964602971948017,
      "loss": 0.2315,
      "num_input_tokens_seen": 19659456,
      "step": 93150
    },
    {
      "epoch": 10.248074807480748,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.016963175326963766,
      "loss": 0.2335,
      "num_input_tokens_seen": 19660576,
      "step": 93155
    },
    {
      "epoch": 10.248624862486249,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.016961747663885873,
      "loss": 0.233,
      "num_input_tokens_seen": 19661600,
      "step": 93160
    },
    {
      "epoch": 10.249174917491748,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.016960319982727495,
      "loss": 0.2309,
      "num_input_tokens_seen": 19662656,
      "step": 93165
    },
    {
      "epoch": 10.24972497249725,
      "grad_norm": 0.0005645751953125,
      "learning_rate": 0.01695889228350178,
      "loss": 0.2304,
      "num_input_tokens_seen": 19663776,
      "step": 93170
    },
    {
      "epoch": 10.25027502750275,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.016957464566221908,
      "loss": 0.2324,
      "num_input_tokens_seen": 19664896,
      "step": 93175
    },
    {
      "epoch": 10.250825082508252,
      "grad_norm": 0.00665283203125,
      "learning_rate": 0.016956036830901015,
      "loss": 0.2314,
      "num_input_tokens_seen": 19665952,
      "step": 93180
    },
    {
      "epoch": 10.251375137513751,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.016954609077552273,
      "loss": 0.2293,
      "num_input_tokens_seen": 19667008,
      "step": 93185
    },
    {
      "epoch": 10.251925192519252,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.016953181306188832,
      "loss": 0.2309,
      "num_input_tokens_seen": 19668096,
      "step": 93190
    },
    {
      "epoch": 10.252475247524753,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.01695175351682386,
      "loss": 0.2304,
      "num_input_tokens_seen": 19669184,
      "step": 93195
    },
    {
      "epoch": 10.253025302530252,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.016950325709470512,
      "loss": 0.2335,
      "num_input_tokens_seen": 19670208,
      "step": 93200
    },
    {
      "epoch": 10.253575357535754,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.016948897884141948,
      "loss": 0.2298,
      "num_input_tokens_seen": 19671264,
      "step": 93205
    },
    {
      "epoch": 10.254125412541255,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.01694747004085133,
      "loss": 0.2319,
      "num_input_tokens_seen": 19672224,
      "step": 93210
    },
    {
      "epoch": 10.254675467546754,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.01694604217961181,
      "loss": 0.2298,
      "num_input_tokens_seen": 19673248,
      "step": 93215
    },
    {
      "epoch": 10.255225522552255,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.016944614300436558,
      "loss": 0.2325,
      "num_input_tokens_seen": 19674304,
      "step": 93220
    },
    {
      "epoch": 10.255775577557756,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.016943186403338723,
      "loss": 0.233,
      "num_input_tokens_seen": 19675360,
      "step": 93225
    },
    {
      "epoch": 10.256325632563255,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.016941758488331474,
      "loss": 0.2324,
      "num_input_tokens_seen": 19676352,
      "step": 93230
    },
    {
      "epoch": 10.256875687568757,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.016940330555427967,
      "loss": 0.2335,
      "num_input_tokens_seen": 19677408,
      "step": 93235
    },
    {
      "epoch": 10.257425742574258,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01693890260464137,
      "loss": 0.2298,
      "num_input_tokens_seen": 19678528,
      "step": 93240
    },
    {
      "epoch": 10.257975797579759,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.016937474635984835,
      "loss": 0.2309,
      "num_input_tokens_seen": 19679520,
      "step": 93245
    },
    {
      "epoch": 10.258525852585258,
      "grad_norm": 0.00058746337890625,
      "learning_rate": 0.016936046649471525,
      "loss": 0.2309,
      "num_input_tokens_seen": 19680480,
      "step": 93250
    },
    {
      "epoch": 10.25907590759076,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.016934618645114603,
      "loss": 0.2319,
      "num_input_tokens_seen": 19681600,
      "step": 93255
    },
    {
      "epoch": 10.25962596259626,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.016933190622927228,
      "loss": 0.2293,
      "num_input_tokens_seen": 19682624,
      "step": 93260
    },
    {
      "epoch": 10.26017601760176,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.016931762582922564,
      "loss": 0.2314,
      "num_input_tokens_seen": 19683616,
      "step": 93265
    },
    {
      "epoch": 10.26072607260726,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.016930334525113774,
      "loss": 0.2309,
      "num_input_tokens_seen": 19684672,
      "step": 93270
    },
    {
      "epoch": 10.261276127612762,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.016928906449514013,
      "loss": 0.2335,
      "num_input_tokens_seen": 19685664,
      "step": 93275
    },
    {
      "epoch": 10.261826182618261,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01692747835613645,
      "loss": 0.2335,
      "num_input_tokens_seen": 19686720,
      "step": 93280
    },
    {
      "epoch": 10.262376237623762,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.016926050244994235,
      "loss": 0.2309,
      "num_input_tokens_seen": 19687776,
      "step": 93285
    },
    {
      "epoch": 10.262926292629263,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.01692462211610055,
      "loss": 0.2314,
      "num_input_tokens_seen": 19688896,
      "step": 93290
    },
    {
      "epoch": 10.263476347634764,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.016923193969468543,
      "loss": 0.2288,
      "num_input_tokens_seen": 19689952,
      "step": 93295
    },
    {
      "epoch": 10.264026402640264,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.016921765805111375,
      "loss": 0.2324,
      "num_input_tokens_seen": 19691040,
      "step": 93300
    },
    {
      "epoch": 10.264576457645765,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.016920337623042223,
      "loss": 0.2319,
      "num_input_tokens_seen": 19692064,
      "step": 93305
    },
    {
      "epoch": 10.265126512651266,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.016918909423274234,
      "loss": 0.2319,
      "num_input_tokens_seen": 19693152,
      "step": 93310
    },
    {
      "epoch": 10.265676567656765,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.016917481205820577,
      "loss": 0.2324,
      "num_input_tokens_seen": 19694208,
      "step": 93315
    },
    {
      "epoch": 10.266226622662266,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.016916052970694418,
      "loss": 0.2309,
      "num_input_tokens_seen": 19695328,
      "step": 93320
    },
    {
      "epoch": 10.266776677667767,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.016914624717908923,
      "loss": 0.2324,
      "num_input_tokens_seen": 19696448,
      "step": 93325
    },
    {
      "epoch": 10.267326732673267,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.01691319644747724,
      "loss": 0.2314,
      "num_input_tokens_seen": 19697472,
      "step": 93330
    },
    {
      "epoch": 10.267876787678768,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.01691176815941255,
      "loss": 0.2324,
      "num_input_tokens_seen": 19698496,
      "step": 93335
    },
    {
      "epoch": 10.268426842684269,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.016910339853728012,
      "loss": 0.2324,
      "num_input_tokens_seen": 19699520,
      "step": 93340
    },
    {
      "epoch": 10.268976897689768,
      "grad_norm": 0.0007476806640625,
      "learning_rate": 0.016908911530436783,
      "loss": 0.2298,
      "num_input_tokens_seen": 19700544,
      "step": 93345
    },
    {
      "epoch": 10.26952695269527,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.016907483189552038,
      "loss": 0.233,
      "num_input_tokens_seen": 19701568,
      "step": 93350
    },
    {
      "epoch": 10.27007700770077,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.016906054831086932,
      "loss": 0.2319,
      "num_input_tokens_seen": 19702688,
      "step": 93355
    },
    {
      "epoch": 10.270627062706271,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.016904626455054632,
      "loss": 0.2299,
      "num_input_tokens_seen": 19703776,
      "step": 93360
    },
    {
      "epoch": 10.27117711771177,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.016903198061468307,
      "loss": 0.233,
      "num_input_tokens_seen": 19704800,
      "step": 93365
    },
    {
      "epoch": 10.271727172717272,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.016901769650341115,
      "loss": 0.2288,
      "num_input_tokens_seen": 19705888,
      "step": 93370
    },
    {
      "epoch": 10.272277227722773,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.01690034122168623,
      "loss": 0.2325,
      "num_input_tokens_seen": 19707008,
      "step": 93375
    },
    {
      "epoch": 10.272827282728272,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.01689891277551681,
      "loss": 0.2325,
      "num_input_tokens_seen": 19708064,
      "step": 93380
    },
    {
      "epoch": 10.273377337733773,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.016897484311846022,
      "loss": 0.2288,
      "num_input_tokens_seen": 19709120,
      "step": 93385
    },
    {
      "epoch": 10.273927392739274,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.01689605583068703,
      "loss": 0.2351,
      "num_input_tokens_seen": 19710208,
      "step": 93390
    },
    {
      "epoch": 10.274477447744774,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.016894627332053002,
      "loss": 0.2288,
      "num_input_tokens_seen": 19711264,
      "step": 93395
    },
    {
      "epoch": 10.275027502750275,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.016893198815957102,
      "loss": 0.234,
      "num_input_tokens_seen": 19712352,
      "step": 93400
    },
    {
      "epoch": 10.275577557755776,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.0168917702824125,
      "loss": 0.2335,
      "num_input_tokens_seen": 19713408,
      "step": 93405
    },
    {
      "epoch": 10.276127612761275,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.01689034173143236,
      "loss": 0.2309,
      "num_input_tokens_seen": 19714496,
      "step": 93410
    },
    {
      "epoch": 10.276677667766776,
      "grad_norm": 0.000499725341796875,
      "learning_rate": 0.01688891316302984,
      "loss": 0.2293,
      "num_input_tokens_seen": 19715520,
      "step": 93415
    },
    {
      "epoch": 10.277227722772277,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01688748457721812,
      "loss": 0.2298,
      "num_input_tokens_seen": 19716544,
      "step": 93420
    },
    {
      "epoch": 10.277777777777779,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.01688605597401036,
      "loss": 0.2324,
      "num_input_tokens_seen": 19717600,
      "step": 93425
    },
    {
      "epoch": 10.278327832783278,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.01688462735341972,
      "loss": 0.2298,
      "num_input_tokens_seen": 19718624,
      "step": 93430
    },
    {
      "epoch": 10.278877887788779,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.016883198715459385,
      "loss": 0.2314,
      "num_input_tokens_seen": 19719648,
      "step": 93435
    },
    {
      "epoch": 10.27942794279428,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.016881770060142506,
      "loss": 0.2319,
      "num_input_tokens_seen": 19720768,
      "step": 93440
    },
    {
      "epoch": 10.27997799779978,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.016880341387482255,
      "loss": 0.2314,
      "num_input_tokens_seen": 19721856,
      "step": 93445
    },
    {
      "epoch": 10.28052805280528,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0168789126974918,
      "loss": 0.233,
      "num_input_tokens_seen": 19722848,
      "step": 93450
    },
    {
      "epoch": 10.281078107810782,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.016877483990184308,
      "loss": 0.2325,
      "num_input_tokens_seen": 19723936,
      "step": 93455
    },
    {
      "epoch": 10.281628162816281,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.016876055265572946,
      "loss": 0.2346,
      "num_input_tokens_seen": 19725024,
      "step": 93460
    },
    {
      "epoch": 10.282178217821782,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.016874626523670888,
      "loss": 0.2298,
      "num_input_tokens_seen": 19726016,
      "step": 93465
    },
    {
      "epoch": 10.282728272827283,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.01687319776449129,
      "loss": 0.2324,
      "num_input_tokens_seen": 19727104,
      "step": 93470
    },
    {
      "epoch": 10.283278327832782,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.01687176898804733,
      "loss": 0.2309,
      "num_input_tokens_seen": 19728192,
      "step": 93475
    },
    {
      "epoch": 10.283828382838283,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.01687034019435217,
      "loss": 0.2324,
      "num_input_tokens_seen": 19729312,
      "step": 93480
    },
    {
      "epoch": 10.284378437843785,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.01686891138341899,
      "loss": 0.2324,
      "num_input_tokens_seen": 19730336,
      "step": 93485
    },
    {
      "epoch": 10.284928492849286,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.016867482555260944,
      "loss": 0.2303,
      "num_input_tokens_seen": 19731360,
      "step": 93490
    },
    {
      "epoch": 10.285478547854785,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.016866053709891207,
      "loss": 0.234,
      "num_input_tokens_seen": 19732384,
      "step": 93495
    },
    {
      "epoch": 10.286028602860286,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.016864624847322953,
      "loss": 0.2319,
      "num_input_tokens_seen": 19733376,
      "step": 93500
    },
    {
      "epoch": 10.286578657865787,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.016863195967569345,
      "loss": 0.2298,
      "num_input_tokens_seen": 19734400,
      "step": 93505
    },
    {
      "epoch": 10.287128712871286,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.016861767070643553,
      "loss": 0.2298,
      "num_input_tokens_seen": 19735424,
      "step": 93510
    },
    {
      "epoch": 10.287678767876788,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.016860338156558744,
      "loss": 0.2319,
      "num_input_tokens_seen": 19736416,
      "step": 93515
    },
    {
      "epoch": 10.288228822882289,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.016858909225328095,
      "loss": 0.2319,
      "num_input_tokens_seen": 19737472,
      "step": 93520
    },
    {
      "epoch": 10.288778877887788,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.01685748027696477,
      "loss": 0.2319,
      "num_input_tokens_seen": 19738528,
      "step": 93525
    },
    {
      "epoch": 10.289328932893289,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.01685605131148194,
      "loss": 0.2303,
      "num_input_tokens_seen": 19739584,
      "step": 93530
    },
    {
      "epoch": 10.28987898789879,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01685462232889278,
      "loss": 0.2319,
      "num_input_tokens_seen": 19740608,
      "step": 93535
    },
    {
      "epoch": 10.290429042904291,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.016853193329210453,
      "loss": 0.2293,
      "num_input_tokens_seen": 19741696,
      "step": 93540
    },
    {
      "epoch": 10.29097909790979,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.01685176431244813,
      "loss": 0.2324,
      "num_input_tokens_seen": 19742816,
      "step": 93545
    },
    {
      "epoch": 10.291529152915292,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.016850335278618987,
      "loss": 0.2309,
      "num_input_tokens_seen": 19743840,
      "step": 93550
    },
    {
      "epoch": 10.292079207920793,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.016848906227736195,
      "loss": 0.2298,
      "num_input_tokens_seen": 19744960,
      "step": 93555
    },
    {
      "epoch": 10.292629262926292,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.016847477159812912,
      "loss": 0.2309,
      "num_input_tokens_seen": 19746048,
      "step": 93560
    },
    {
      "epoch": 10.293179317931793,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.016846048074862327,
      "loss": 0.2303,
      "num_input_tokens_seen": 19747136,
      "step": 93565
    },
    {
      "epoch": 10.293729372937294,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0168446189728976,
      "loss": 0.2324,
      "num_input_tokens_seen": 19748192,
      "step": 93570
    },
    {
      "epoch": 10.294279427942794,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.016843189853931902,
      "loss": 0.2324,
      "num_input_tokens_seen": 19749184,
      "step": 93575
    },
    {
      "epoch": 10.294829482948295,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.016841760717978414,
      "loss": 0.2324,
      "num_input_tokens_seen": 19750208,
      "step": 93580
    },
    {
      "epoch": 10.295379537953796,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.016840331565050295,
      "loss": 0.2335,
      "num_input_tokens_seen": 19751296,
      "step": 93585
    },
    {
      "epoch": 10.295929592959295,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016838902395160723,
      "loss": 0.2308,
      "num_input_tokens_seen": 19752288,
      "step": 93590
    },
    {
      "epoch": 10.296479647964796,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.016837473208322875,
      "loss": 0.2335,
      "num_input_tokens_seen": 19753344,
      "step": 93595
    },
    {
      "epoch": 10.297029702970297,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.016836044004549917,
      "loss": 0.2308,
      "num_input_tokens_seen": 19754400,
      "step": 93600
    },
    {
      "epoch": 10.297579757975798,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.016834614783855017,
      "loss": 0.2324,
      "num_input_tokens_seen": 19755456,
      "step": 93605
    },
    {
      "epoch": 10.298129812981298,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.01683318554625136,
      "loss": 0.2319,
      "num_input_tokens_seen": 19756576,
      "step": 93610
    },
    {
      "epoch": 10.298679867986799,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.01683175629175211,
      "loss": 0.2314,
      "num_input_tokens_seen": 19757600,
      "step": 93615
    },
    {
      "epoch": 10.2992299229923,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.016830327020370434,
      "loss": 0.2335,
      "num_input_tokens_seen": 19758656,
      "step": 93620
    },
    {
      "epoch": 10.2997799779978,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.01682889773211952,
      "loss": 0.2314,
      "num_input_tokens_seen": 19759680,
      "step": 93625
    },
    {
      "epoch": 10.3003300330033,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.016827468427012535,
      "loss": 0.2314,
      "num_input_tokens_seen": 19760672,
      "step": 93630
    },
    {
      "epoch": 10.300880088008801,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.016826039105062644,
      "loss": 0.2324,
      "num_input_tokens_seen": 19761728,
      "step": 93635
    },
    {
      "epoch": 10.3014301430143,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.016824609766283027,
      "loss": 0.2298,
      "num_input_tokens_seen": 19762752,
      "step": 93640
    },
    {
      "epoch": 10.301980198019802,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.016823180410686863,
      "loss": 0.2324,
      "num_input_tokens_seen": 19763840,
      "step": 93645
    },
    {
      "epoch": 10.302530253025303,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.016821751038287314,
      "loss": 0.2293,
      "num_input_tokens_seen": 19764960,
      "step": 93650
    },
    {
      "epoch": 10.303080308030804,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.016820321649097563,
      "loss": 0.2304,
      "num_input_tokens_seen": 19766016,
      "step": 93655
    },
    {
      "epoch": 10.303630363036303,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.01681889224313078,
      "loss": 0.2299,
      "num_input_tokens_seen": 19767072,
      "step": 93660
    },
    {
      "epoch": 10.304180418041804,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01681746282040014,
      "loss": 0.2298,
      "num_input_tokens_seen": 19768064,
      "step": 93665
    },
    {
      "epoch": 10.304730473047305,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.016816033380918815,
      "loss": 0.2283,
      "num_input_tokens_seen": 19769216,
      "step": 93670
    },
    {
      "epoch": 10.305280528052805,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.016814603924699984,
      "loss": 0.2299,
      "num_input_tokens_seen": 19770336,
      "step": 93675
    },
    {
      "epoch": 10.305830583058306,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.01681317445175682,
      "loss": 0.2278,
      "num_input_tokens_seen": 19771392,
      "step": 93680
    },
    {
      "epoch": 10.306380638063807,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.0168117449621025,
      "loss": 0.231,
      "num_input_tokens_seen": 19772480,
      "step": 93685
    },
    {
      "epoch": 10.306930693069306,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01681031545575019,
      "loss": 0.2288,
      "num_input_tokens_seen": 19773504,
      "step": 93690
    },
    {
      "epoch": 10.307480748074807,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.016808885932713073,
      "loss": 0.2314,
      "num_input_tokens_seen": 19774528,
      "step": 93695
    },
    {
      "epoch": 10.308030803080309,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.016807456393004325,
      "loss": 0.2341,
      "num_input_tokens_seen": 19775616,
      "step": 93700
    },
    {
      "epoch": 10.308580858085808,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.016806026836637112,
      "loss": 0.232,
      "num_input_tokens_seen": 19776640,
      "step": 93705
    },
    {
      "epoch": 10.309130913091309,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.016804597263624627,
      "loss": 0.2325,
      "num_input_tokens_seen": 19777696,
      "step": 93710
    },
    {
      "epoch": 10.30968096809681,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.016803167673980025,
      "loss": 0.2336,
      "num_input_tokens_seen": 19778688,
      "step": 93715
    },
    {
      "epoch": 10.310231023102311,
      "grad_norm": 0.01165771484375,
      "learning_rate": 0.0168017380677165,
      "loss": 0.233,
      "num_input_tokens_seen": 19779744,
      "step": 93720
    },
    {
      "epoch": 10.31078107810781,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.01680030844484721,
      "loss": 0.2304,
      "num_input_tokens_seen": 19780736,
      "step": 93725
    },
    {
      "epoch": 10.311331133113312,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.016798878805385352,
      "loss": 0.2315,
      "num_input_tokens_seen": 19781856,
      "step": 93730
    },
    {
      "epoch": 10.311881188118813,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01679744914934408,
      "loss": 0.2335,
      "num_input_tokens_seen": 19782880,
      "step": 93735
    },
    {
      "epoch": 10.312431243124312,
      "grad_norm": 0.000774383544921875,
      "learning_rate": 0.01679601947673659,
      "loss": 0.2335,
      "num_input_tokens_seen": 19783904,
      "step": 93740
    },
    {
      "epoch": 10.312981298129813,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.016794589787576048,
      "loss": 0.232,
      "num_input_tokens_seen": 19784960,
      "step": 93745
    },
    {
      "epoch": 10.313531353135314,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.01679316008187563,
      "loss": 0.2335,
      "num_input_tokens_seen": 19786016,
      "step": 93750
    },
    {
      "epoch": 10.314081408140813,
      "grad_norm": 0.005859375,
      "learning_rate": 0.01679173035964852,
      "loss": 0.233,
      "num_input_tokens_seen": 19787040,
      "step": 93755
    },
    {
      "epoch": 10.314631463146315,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.016790300620907892,
      "loss": 0.2309,
      "num_input_tokens_seen": 19788032,
      "step": 93760
    },
    {
      "epoch": 10.315181518151816,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.016788870865666914,
      "loss": 0.2319,
      "num_input_tokens_seen": 19789120,
      "step": 93765
    },
    {
      "epoch": 10.315731573157315,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.016787441093938784,
      "loss": 0.2298,
      "num_input_tokens_seen": 19790208,
      "step": 93770
    },
    {
      "epoch": 10.316281628162816,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.01678601130573666,
      "loss": 0.2335,
      "num_input_tokens_seen": 19791264,
      "step": 93775
    },
    {
      "epoch": 10.316831683168317,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.016784581501073727,
      "loss": 0.2351,
      "num_input_tokens_seen": 19792288,
      "step": 93780
    },
    {
      "epoch": 10.317381738173818,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.016783151679963162,
      "loss": 0.2304,
      "num_input_tokens_seen": 19793312,
      "step": 93785
    },
    {
      "epoch": 10.317931793179318,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.016781721842418145,
      "loss": 0.2304,
      "num_input_tokens_seen": 19794400,
      "step": 93790
    },
    {
      "epoch": 10.318481848184819,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.016780291988451854,
      "loss": 0.2324,
      "num_input_tokens_seen": 19795424,
      "step": 93795
    },
    {
      "epoch": 10.31903190319032,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.016778862118077463,
      "loss": 0.2299,
      "num_input_tokens_seen": 19796480,
      "step": 93800
    },
    {
      "epoch": 10.319581958195819,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.01677743223130816,
      "loss": 0.2314,
      "num_input_tokens_seen": 19797504,
      "step": 93805
    },
    {
      "epoch": 10.32013201320132,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.016776002328157107,
      "loss": 0.2303,
      "num_input_tokens_seen": 19798528,
      "step": 93810
    },
    {
      "epoch": 10.320682068206821,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0167745724086375,
      "loss": 0.233,
      "num_input_tokens_seen": 19799616,
      "step": 93815
    },
    {
      "epoch": 10.32123212321232,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.016773142472762504,
      "loss": 0.2324,
      "num_input_tokens_seen": 19800704,
      "step": 93820
    },
    {
      "epoch": 10.321782178217822,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.016771712520545313,
      "loss": 0.2314,
      "num_input_tokens_seen": 19801728,
      "step": 93825
    },
    {
      "epoch": 10.322332233223323,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.016770282551999093,
      "loss": 0.2304,
      "num_input_tokens_seen": 19802848,
      "step": 93830
    },
    {
      "epoch": 10.322882288228822,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.01676885256713703,
      "loss": 0.2298,
      "num_input_tokens_seen": 19803968,
      "step": 93835
    },
    {
      "epoch": 10.323432343234323,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.0167674225659723,
      "loss": 0.2309,
      "num_input_tokens_seen": 19804992,
      "step": 93840
    },
    {
      "epoch": 10.323982398239824,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.016765992548518086,
      "loss": 0.2304,
      "num_input_tokens_seen": 19806016,
      "step": 93845
    },
    {
      "epoch": 10.324532453245325,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.01676456251478756,
      "loss": 0.2314,
      "num_input_tokens_seen": 19807136,
      "step": 93850
    },
    {
      "epoch": 10.325082508250825,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.016763132464793915,
      "loss": 0.2324,
      "num_input_tokens_seen": 19808224,
      "step": 93855
    },
    {
      "epoch": 10.325632563256326,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.016761702398550325,
      "loss": 0.2309,
      "num_input_tokens_seen": 19809248,
      "step": 93860
    },
    {
      "epoch": 10.326182618261827,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.016760272316069965,
      "loss": 0.2324,
      "num_input_tokens_seen": 19810336,
      "step": 93865
    },
    {
      "epoch": 10.326732673267326,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.016758842217366024,
      "loss": 0.2309,
      "num_input_tokens_seen": 19811424,
      "step": 93870
    },
    {
      "epoch": 10.327282728272827,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.016757412102451675,
      "loss": 0.2319,
      "num_input_tokens_seen": 19812480,
      "step": 93875
    },
    {
      "epoch": 10.327832783278328,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.016755981971340102,
      "loss": 0.2329,
      "num_input_tokens_seen": 19813504,
      "step": 93880
    },
    {
      "epoch": 10.328382838283828,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01675455182404449,
      "loss": 0.2329,
      "num_input_tokens_seen": 19814560,
      "step": 93885
    },
    {
      "epoch": 10.328932893289329,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01675312166057801,
      "loss": 0.2319,
      "num_input_tokens_seen": 19815584,
      "step": 93890
    },
    {
      "epoch": 10.32948294829483,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01675169148095385,
      "loss": 0.2309,
      "num_input_tokens_seen": 19816640,
      "step": 93895
    },
    {
      "epoch": 10.33003300330033,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01675026128518519,
      "loss": 0.2319,
      "num_input_tokens_seen": 19817760,
      "step": 93900
    },
    {
      "epoch": 10.33058305830583,
      "grad_norm": 0.005859375,
      "learning_rate": 0.016748831073285213,
      "loss": 0.2319,
      "num_input_tokens_seen": 19818720,
      "step": 93905
    },
    {
      "epoch": 10.331133113311331,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.0167474008452671,
      "loss": 0.2303,
      "num_input_tokens_seen": 19819744,
      "step": 93910
    },
    {
      "epoch": 10.331683168316832,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.01674597060114403,
      "loss": 0.2309,
      "num_input_tokens_seen": 19820768,
      "step": 93915
    },
    {
      "epoch": 10.332233223322332,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.016744540340929186,
      "loss": 0.2309,
      "num_input_tokens_seen": 19821824,
      "step": 93920
    },
    {
      "epoch": 10.332783278327833,
      "grad_norm": 0.0005645751953125,
      "learning_rate": 0.016743110064635754,
      "loss": 0.2335,
      "num_input_tokens_seen": 19822848,
      "step": 93925
    },
    {
      "epoch": 10.333333333333334,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.016741679772276908,
      "loss": 0.2298,
      "num_input_tokens_seen": 19823872,
      "step": 93930
    },
    {
      "epoch": 10.333883388338833,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.016740249463865847,
      "loss": 0.2293,
      "num_input_tokens_seen": 19824896,
      "step": 93935
    },
    {
      "epoch": 10.334433443344334,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.016738819139415727,
      "loss": 0.2324,
      "num_input_tokens_seen": 19825984,
      "step": 93940
    },
    {
      "epoch": 10.334983498349835,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.01673738879893975,
      "loss": 0.2314,
      "num_input_tokens_seen": 19827104,
      "step": 93945
    },
    {
      "epoch": 10.335533553355335,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0167359584424511,
      "loss": 0.2319,
      "num_input_tokens_seen": 19828160,
      "step": 93950
    },
    {
      "epoch": 10.336083608360836,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.016734528069962947,
      "loss": 0.2335,
      "num_input_tokens_seen": 19829152,
      "step": 93955
    },
    {
      "epoch": 10.336633663366337,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.016733097681488486,
      "loss": 0.2319,
      "num_input_tokens_seen": 19830176,
      "step": 93960
    },
    {
      "epoch": 10.337183718371838,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.016731667277040896,
      "loss": 0.2319,
      "num_input_tokens_seen": 19831264,
      "step": 93965
    },
    {
      "epoch": 10.337733773377337,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.016730236856633354,
      "loss": 0.2314,
      "num_input_tokens_seen": 19832288,
      "step": 93970
    },
    {
      "epoch": 10.338283828382838,
      "grad_norm": 0.000530242919921875,
      "learning_rate": 0.01672880642027905,
      "loss": 0.2309,
      "num_input_tokens_seen": 19833344,
      "step": 93975
    },
    {
      "epoch": 10.33883388338834,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.01672737596799117,
      "loss": 0.233,
      "num_input_tokens_seen": 19834368,
      "step": 93980
    },
    {
      "epoch": 10.339383938393839,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.016725945499782895,
      "loss": 0.2319,
      "num_input_tokens_seen": 19835392,
      "step": 93985
    },
    {
      "epoch": 10.33993399339934,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016724515015667407,
      "loss": 0.2298,
      "num_input_tokens_seen": 19836352,
      "step": 93990
    },
    {
      "epoch": 10.340484048404841,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.01672308451565789,
      "loss": 0.2314,
      "num_input_tokens_seen": 19837408,
      "step": 93995
    },
    {
      "epoch": 10.34103410341034,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.016721653999767532,
      "loss": 0.2329,
      "num_input_tokens_seen": 19838528,
      "step": 94000
    },
    {
      "epoch": 10.341584158415841,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.016720223468009516,
      "loss": 0.2319,
      "num_input_tokens_seen": 19839552,
      "step": 94005
    },
    {
      "epoch": 10.342134213421343,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.016718792920397022,
      "loss": 0.2314,
      "num_input_tokens_seen": 19840576,
      "step": 94010
    },
    {
      "epoch": 10.342684268426842,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.016717362356943242,
      "loss": 0.2304,
      "num_input_tokens_seen": 19841600,
      "step": 94015
    },
    {
      "epoch": 10.343234323432343,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.016715931777661355,
      "loss": 0.2324,
      "num_input_tokens_seen": 19842720,
      "step": 94020
    },
    {
      "epoch": 10.343784378437844,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01671450118256455,
      "loss": 0.2309,
      "num_input_tokens_seen": 19843744,
      "step": 94025
    },
    {
      "epoch": 10.344334433443345,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01671307057166601,
      "loss": 0.2303,
      "num_input_tokens_seen": 19844736,
      "step": 94030
    },
    {
      "epoch": 10.344884488448844,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.01671163994497892,
      "loss": 0.2324,
      "num_input_tokens_seen": 19845856,
      "step": 94035
    },
    {
      "epoch": 10.345434543454346,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.016710209302516463,
      "loss": 0.2335,
      "num_input_tokens_seen": 19846912,
      "step": 94040
    },
    {
      "epoch": 10.345984598459847,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.016708778644291832,
      "loss": 0.2309,
      "num_input_tokens_seen": 19847872,
      "step": 94045
    },
    {
      "epoch": 10.346534653465346,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.016707347970318205,
      "loss": 0.2314,
      "num_input_tokens_seen": 19848864,
      "step": 94050
    },
    {
      "epoch": 10.347084708470847,
      "grad_norm": 0.000728607177734375,
      "learning_rate": 0.016705917280608773,
      "loss": 0.2304,
      "num_input_tokens_seen": 19849888,
      "step": 94055
    },
    {
      "epoch": 10.347634763476348,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.016704486575176723,
      "loss": 0.2319,
      "num_input_tokens_seen": 19850880,
      "step": 94060
    },
    {
      "epoch": 10.348184818481847,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.016703055854035233,
      "loss": 0.2293,
      "num_input_tokens_seen": 19851936,
      "step": 94065
    },
    {
      "epoch": 10.348734873487349,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.016701625117197492,
      "loss": 0.2309,
      "num_input_tokens_seen": 19852928,
      "step": 94070
    },
    {
      "epoch": 10.34928492849285,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.016700194364676697,
      "loss": 0.2309,
      "num_input_tokens_seen": 19853952,
      "step": 94075
    },
    {
      "epoch": 10.34983498349835,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.016698763596486026,
      "loss": 0.2309,
      "num_input_tokens_seen": 19855008,
      "step": 94080
    },
    {
      "epoch": 10.35038503850385,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.016697332812638657,
      "loss": 0.2319,
      "num_input_tokens_seen": 19856128,
      "step": 94085
    },
    {
      "epoch": 10.350935093509351,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.01669590201314779,
      "loss": 0.2303,
      "num_input_tokens_seen": 19857184,
      "step": 94090
    },
    {
      "epoch": 10.351485148514852,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01669447119802661,
      "loss": 0.2324,
      "num_input_tokens_seen": 19858336,
      "step": 94095
    },
    {
      "epoch": 10.352035203520352,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.016693040367288303,
      "loss": 0.2288,
      "num_input_tokens_seen": 19859360,
      "step": 94100
    },
    {
      "epoch": 10.352585258525853,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.016691609520946057,
      "loss": 0.2314,
      "num_input_tokens_seen": 19860448,
      "step": 94105
    },
    {
      "epoch": 10.353135313531354,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.01669017865901305,
      "loss": 0.2314,
      "num_input_tokens_seen": 19861440,
      "step": 94110
    },
    {
      "epoch": 10.353685368536853,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.016688747781502478,
      "loss": 0.2325,
      "num_input_tokens_seen": 19862400,
      "step": 94115
    },
    {
      "epoch": 10.354235423542354,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.016687316888427534,
      "loss": 0.2319,
      "num_input_tokens_seen": 19863456,
      "step": 94120
    },
    {
      "epoch": 10.354785478547855,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.016685885979801397,
      "loss": 0.2319,
      "num_input_tokens_seen": 19864480,
      "step": 94125
    },
    {
      "epoch": 10.355335533553355,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.016684455055637263,
      "loss": 0.2319,
      "num_input_tokens_seen": 19865600,
      "step": 94130
    },
    {
      "epoch": 10.355885588558856,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.016683024115948308,
      "loss": 0.2335,
      "num_input_tokens_seen": 19866624,
      "step": 94135
    },
    {
      "epoch": 10.356435643564357,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.01668159316074773,
      "loss": 0.2309,
      "num_input_tokens_seen": 19867648,
      "step": 94140
    },
    {
      "epoch": 10.356985698569858,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.01668016219004871,
      "loss": 0.2319,
      "num_input_tokens_seen": 19868672,
      "step": 94145
    },
    {
      "epoch": 10.357535753575357,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.016678731203864452,
      "loss": 0.2304,
      "num_input_tokens_seen": 19869728,
      "step": 94150
    },
    {
      "epoch": 10.358085808580858,
      "grad_norm": 0.000682830810546875,
      "learning_rate": 0.016677300202208128,
      "loss": 0.2314,
      "num_input_tokens_seen": 19870752,
      "step": 94155
    },
    {
      "epoch": 10.35863586358636,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01667586918509293,
      "loss": 0.2298,
      "num_input_tokens_seen": 19871808,
      "step": 94160
    },
    {
      "epoch": 10.359185918591859,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.016674438152532056,
      "loss": 0.2319,
      "num_input_tokens_seen": 19872864,
      "step": 94165
    },
    {
      "epoch": 10.35973597359736,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.016673007104538687,
      "loss": 0.2314,
      "num_input_tokens_seen": 19873984,
      "step": 94170
    },
    {
      "epoch": 10.36028602860286,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.016671576041126013,
      "loss": 0.2324,
      "num_input_tokens_seen": 19875008,
      "step": 94175
    },
    {
      "epoch": 10.36083608360836,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.016670144962307225,
      "loss": 0.2324,
      "num_input_tokens_seen": 19876064,
      "step": 94180
    },
    {
      "epoch": 10.361386138613861,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.016668713868095507,
      "loss": 0.2324,
      "num_input_tokens_seen": 19877120,
      "step": 94185
    },
    {
      "epoch": 10.361936193619362,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.01666728275850406,
      "loss": 0.233,
      "num_input_tokens_seen": 19878208,
      "step": 94190
    },
    {
      "epoch": 10.362486248624862,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01666585163354607,
      "loss": 0.2324,
      "num_input_tokens_seen": 19879232,
      "step": 94195
    },
    {
      "epoch": 10.363036303630363,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.016664420493234716,
      "loss": 0.2319,
      "num_input_tokens_seen": 19880288,
      "step": 94200
    },
    {
      "epoch": 10.363586358635864,
      "grad_norm": 0.0026397705078125,
      "learning_rate": 0.016662989337583207,
      "loss": 0.2324,
      "num_input_tokens_seen": 19881344,
      "step": 94205
    },
    {
      "epoch": 10.364136413641365,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.01666155816660472,
      "loss": 0.2298,
      "num_input_tokens_seen": 19882400,
      "step": 94210
    },
    {
      "epoch": 10.364686468646864,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.016660126980312443,
      "loss": 0.2313,
      "num_input_tokens_seen": 19883488,
      "step": 94215
    },
    {
      "epoch": 10.365236523652365,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.016658695778719583,
      "loss": 0.2329,
      "num_input_tokens_seen": 19884544,
      "step": 94220
    },
    {
      "epoch": 10.365786578657866,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.016657264561839313,
      "loss": 0.2303,
      "num_input_tokens_seen": 19885536,
      "step": 94225
    },
    {
      "epoch": 10.366336633663366,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.016655833329684826,
      "loss": 0.2319,
      "num_input_tokens_seen": 19886592,
      "step": 94230
    },
    {
      "epoch": 10.366886688668867,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.016654402082269325,
      "loss": 0.2324,
      "num_input_tokens_seen": 19887712,
      "step": 94235
    },
    {
      "epoch": 10.367436743674368,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.016652970819605994,
      "loss": 0.2324,
      "num_input_tokens_seen": 19888832,
      "step": 94240
    },
    {
      "epoch": 10.367986798679867,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.016651539541708022,
      "loss": 0.2314,
      "num_input_tokens_seen": 19889920,
      "step": 94245
    },
    {
      "epoch": 10.368536853685368,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.016650108248588605,
      "loss": 0.2324,
      "num_input_tokens_seen": 19890976,
      "step": 94250
    },
    {
      "epoch": 10.36908690869087,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.01664867694026093,
      "loss": 0.2334,
      "num_input_tokens_seen": 19892064,
      "step": 94255
    },
    {
      "epoch": 10.369636963696369,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.016647245616738186,
      "loss": 0.2303,
      "num_input_tokens_seen": 19893056,
      "step": 94260
    },
    {
      "epoch": 10.37018701870187,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.016645814278033577,
      "loss": 0.2298,
      "num_input_tokens_seen": 19894240,
      "step": 94265
    },
    {
      "epoch": 10.370737073707371,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01664438292416029,
      "loss": 0.2314,
      "num_input_tokens_seen": 19895264,
      "step": 94270
    },
    {
      "epoch": 10.371287128712872,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016642951555131506,
      "loss": 0.2329,
      "num_input_tokens_seen": 19896288,
      "step": 94275
    },
    {
      "epoch": 10.371837183718371,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.016641520170960426,
      "loss": 0.2303,
      "num_input_tokens_seen": 19897344,
      "step": 94280
    },
    {
      "epoch": 10.372387238723872,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.016640088771660247,
      "loss": 0.2313,
      "num_input_tokens_seen": 19898432,
      "step": 94285
    },
    {
      "epoch": 10.372937293729374,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.016638657357244156,
      "loss": 0.2309,
      "num_input_tokens_seen": 19899488,
      "step": 94290
    },
    {
      "epoch": 10.373487348734873,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.016637225927725346,
      "loss": 0.2335,
      "num_input_tokens_seen": 19900480,
      "step": 94295
    },
    {
      "epoch": 10.374037403740374,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.016635794483117007,
      "loss": 0.2298,
      "num_input_tokens_seen": 19901536,
      "step": 94300
    },
    {
      "epoch": 10.374587458745875,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.016634363023432342,
      "loss": 0.2298,
      "num_input_tokens_seen": 19902560,
      "step": 94305
    },
    {
      "epoch": 10.375137513751374,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.016632931548684532,
      "loss": 0.2319,
      "num_input_tokens_seen": 19903616,
      "step": 94310
    },
    {
      "epoch": 10.375687568756875,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.016631500058886778,
      "loss": 0.2304,
      "num_input_tokens_seen": 19904672,
      "step": 94315
    },
    {
      "epoch": 10.376237623762377,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.016630068554052267,
      "loss": 0.2329,
      "num_input_tokens_seen": 19905664,
      "step": 94320
    },
    {
      "epoch": 10.376787678767876,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0166286370341942,
      "loss": 0.2329,
      "num_input_tokens_seen": 19906688,
      "step": 94325
    },
    {
      "epoch": 10.377337733773377,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.016627205499325767,
      "loss": 0.2304,
      "num_input_tokens_seen": 19907776,
      "step": 94330
    },
    {
      "epoch": 10.377887788778878,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01662577394946016,
      "loss": 0.2309,
      "num_input_tokens_seen": 19908768,
      "step": 94335
    },
    {
      "epoch": 10.37843784378438,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.01662434238461058,
      "loss": 0.2309,
      "num_input_tokens_seen": 19909856,
      "step": 94340
    },
    {
      "epoch": 10.378987898789878,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.016622910804790204,
      "loss": 0.2324,
      "num_input_tokens_seen": 19910944,
      "step": 94345
    },
    {
      "epoch": 10.37953795379538,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01662147921001225,
      "loss": 0.2304,
      "num_input_tokens_seen": 19912096,
      "step": 94350
    },
    {
      "epoch": 10.38008800880088,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.016620047600289893,
      "loss": 0.2303,
      "num_input_tokens_seen": 19913120,
      "step": 94355
    },
    {
      "epoch": 10.38063806380638,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.016618615975636335,
      "loss": 0.2319,
      "num_input_tokens_seen": 19914272,
      "step": 94360
    },
    {
      "epoch": 10.381188118811881,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.016617184336064777,
      "loss": 0.2309,
      "num_input_tokens_seen": 19915296,
      "step": 94365
    },
    {
      "epoch": 10.381738173817382,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0166157526815884,
      "loss": 0.2283,
      "num_input_tokens_seen": 19916320,
      "step": 94370
    },
    {
      "epoch": 10.382288228822881,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.016614321012220407,
      "loss": 0.2308,
      "num_input_tokens_seen": 19917376,
      "step": 94375
    },
    {
      "epoch": 10.382838283828383,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.01661288932797399,
      "loss": 0.233,
      "num_input_tokens_seen": 19918432,
      "step": 94380
    },
    {
      "epoch": 10.383388338833884,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.016611457628862354,
      "loss": 0.2304,
      "num_input_tokens_seen": 19919488,
      "step": 94385
    },
    {
      "epoch": 10.383938393839385,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.016610025914898678,
      "loss": 0.2329,
      "num_input_tokens_seen": 19920480,
      "step": 94390
    },
    {
      "epoch": 10.384488448844884,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01660859418609617,
      "loss": 0.2324,
      "num_input_tokens_seen": 19921568,
      "step": 94395
    },
    {
      "epoch": 10.385038503850385,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.01660716244246802,
      "loss": 0.2319,
      "num_input_tokens_seen": 19922560,
      "step": 94400
    },
    {
      "epoch": 10.385588558855886,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.016605730684027424,
      "loss": 0.233,
      "num_input_tokens_seen": 19923616,
      "step": 94405
    },
    {
      "epoch": 10.386138613861386,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01660429891078758,
      "loss": 0.2298,
      "num_input_tokens_seen": 19924672,
      "step": 94410
    },
    {
      "epoch": 10.386688668866887,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.016602867122761683,
      "loss": 0.2309,
      "num_input_tokens_seen": 19925760,
      "step": 94415
    },
    {
      "epoch": 10.387238723872388,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016601435319962925,
      "loss": 0.2293,
      "num_input_tokens_seen": 19926880,
      "step": 94420
    },
    {
      "epoch": 10.387788778877887,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.016600003502404512,
      "loss": 0.2314,
      "num_input_tokens_seen": 19927968,
      "step": 94425
    },
    {
      "epoch": 10.388338833883388,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01659857167009963,
      "loss": 0.2324,
      "num_input_tokens_seen": 19929024,
      "step": 94430
    },
    {
      "epoch": 10.38888888888889,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.016597139823061487,
      "loss": 0.2319,
      "num_input_tokens_seen": 19930112,
      "step": 94435
    },
    {
      "epoch": 10.389438943894389,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.016595707961303264,
      "loss": 0.2303,
      "num_input_tokens_seen": 19931168,
      "step": 94440
    },
    {
      "epoch": 10.38998899889989,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.016594276084838173,
      "loss": 0.2314,
      "num_input_tokens_seen": 19932160,
      "step": 94445
    },
    {
      "epoch": 10.39053905390539,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.016592844193679398,
      "loss": 0.2298,
      "num_input_tokens_seen": 19933248,
      "step": 94450
    },
    {
      "epoch": 10.391089108910892,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.016591412287840144,
      "loss": 0.2309,
      "num_input_tokens_seen": 19934336,
      "step": 94455
    },
    {
      "epoch": 10.391639163916391,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.016589980367333612,
      "loss": 0.2303,
      "num_input_tokens_seen": 19935392,
      "step": 94460
    },
    {
      "epoch": 10.392189218921892,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01658854843217299,
      "loss": 0.2314,
      "num_input_tokens_seen": 19936480,
      "step": 94465
    },
    {
      "epoch": 10.392739273927393,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.01658711648237148,
      "loss": 0.2309,
      "num_input_tokens_seen": 19937536,
      "step": 94470
    },
    {
      "epoch": 10.393289328932893,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.016585684517942276,
      "loss": 0.2314,
      "num_input_tokens_seen": 19938624,
      "step": 94475
    },
    {
      "epoch": 10.393839383938394,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.016584252538898586,
      "loss": 0.2298,
      "num_input_tokens_seen": 19939712,
      "step": 94480
    },
    {
      "epoch": 10.394389438943895,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.016582820545253594,
      "loss": 0.2319,
      "num_input_tokens_seen": 19940736,
      "step": 94485
    },
    {
      "epoch": 10.394939493949394,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.016581388537020503,
      "loss": 0.233,
      "num_input_tokens_seen": 19941856,
      "step": 94490
    },
    {
      "epoch": 10.395489548954895,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.016579956514212516,
      "loss": 0.2298,
      "num_input_tokens_seen": 19942848,
      "step": 94495
    },
    {
      "epoch": 10.396039603960396,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.016578524476842826,
      "loss": 0.2319,
      "num_input_tokens_seen": 19943872,
      "step": 94500
    },
    {
      "epoch": 10.396589658965897,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.01657709242492464,
      "loss": 0.2283,
      "num_input_tokens_seen": 19944928,
      "step": 94505
    },
    {
      "epoch": 10.397139713971397,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.016575660358471144,
      "loss": 0.2314,
      "num_input_tokens_seen": 19946016,
      "step": 94510
    },
    {
      "epoch": 10.397689768976898,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.016574228277495547,
      "loss": 0.2335,
      "num_input_tokens_seen": 19947072,
      "step": 94515
    },
    {
      "epoch": 10.398239823982399,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.016572796182011032,
      "loss": 0.2303,
      "num_input_tokens_seen": 19948128,
      "step": 94520
    },
    {
      "epoch": 10.398789878987898,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.01657136407203082,
      "loss": 0.2324,
      "num_input_tokens_seen": 19949184,
      "step": 94525
    },
    {
      "epoch": 10.3993399339934,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.0165699319475681,
      "loss": 0.2319,
      "num_input_tokens_seen": 19950240,
      "step": 94530
    },
    {
      "epoch": 10.3998899889989,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.016568499808636065,
      "loss": 0.2314,
      "num_input_tokens_seen": 19951296,
      "step": 94535
    },
    {
      "epoch": 10.4004400440044,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.016567067655247923,
      "loss": 0.2309,
      "num_input_tokens_seen": 19952320,
      "step": 94540
    },
    {
      "epoch": 10.400990099009901,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01656563548741687,
      "loss": 0.2319,
      "num_input_tokens_seen": 19953376,
      "step": 94545
    },
    {
      "epoch": 10.401540154015402,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.0165642033051561,
      "loss": 0.2308,
      "num_input_tokens_seen": 19954400,
      "step": 94550
    },
    {
      "epoch": 10.402090209020901,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01656277110847883,
      "loss": 0.2324,
      "num_input_tokens_seen": 19955360,
      "step": 94555
    },
    {
      "epoch": 10.402640264026402,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.01656133889739824,
      "loss": 0.2314,
      "num_input_tokens_seen": 19956416,
      "step": 94560
    },
    {
      "epoch": 10.403190319031903,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.01655990667192754,
      "loss": 0.2303,
      "num_input_tokens_seen": 19957536,
      "step": 94565
    },
    {
      "epoch": 10.403740374037405,
      "grad_norm": 0.001953125,
      "learning_rate": 0.016558474432079934,
      "loss": 0.2303,
      "num_input_tokens_seen": 19958560,
      "step": 94570
    },
    {
      "epoch": 10.404290429042904,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.016557042177868613,
      "loss": 0.2288,
      "num_input_tokens_seen": 19959616,
      "step": 94575
    },
    {
      "epoch": 10.404840484048405,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01655560990930678,
      "loss": 0.2319,
      "num_input_tokens_seen": 19960736,
      "step": 94580
    },
    {
      "epoch": 10.405390539053906,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.01655417762640764,
      "loss": 0.2314,
      "num_input_tokens_seen": 19961728,
      "step": 94585
    },
    {
      "epoch": 10.405940594059405,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.01655274532918439,
      "loss": 0.2303,
      "num_input_tokens_seen": 19962848,
      "step": 94590
    },
    {
      "epoch": 10.406490649064907,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.016551313017650227,
      "loss": 0.2308,
      "num_input_tokens_seen": 19963872,
      "step": 94595
    },
    {
      "epoch": 10.407040704070408,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.016549880691818362,
      "loss": 0.2314,
      "num_input_tokens_seen": 19964896,
      "step": 94600
    },
    {
      "epoch": 10.407590759075907,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.016548448351701987,
      "loss": 0.2309,
      "num_input_tokens_seen": 19965920,
      "step": 94605
    },
    {
      "epoch": 10.408140814081408,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.016547015997314306,
      "loss": 0.2298,
      "num_input_tokens_seen": 19966912,
      "step": 94610
    },
    {
      "epoch": 10.408690869086909,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.016545583628668523,
      "loss": 0.2298,
      "num_input_tokens_seen": 19968000,
      "step": 94615
    },
    {
      "epoch": 10.409240924092408,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.01654415124577784,
      "loss": 0.233,
      "num_input_tokens_seen": 19969120,
      "step": 94620
    },
    {
      "epoch": 10.40979097909791,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01654271884865545,
      "loss": 0.2314,
      "num_input_tokens_seen": 19970176,
      "step": 94625
    },
    {
      "epoch": 10.41034103410341,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.016541286437314562,
      "loss": 0.2309,
      "num_input_tokens_seen": 19971200,
      "step": 94630
    },
    {
      "epoch": 10.410891089108912,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01653985401176838,
      "loss": 0.2314,
      "num_input_tokens_seen": 19972224,
      "step": 94635
    },
    {
      "epoch": 10.411441144114411,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.016538421572030098,
      "loss": 0.2309,
      "num_input_tokens_seen": 19973280,
      "step": 94640
    },
    {
      "epoch": 10.411991199119912,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.016536989118112924,
      "loss": 0.2335,
      "num_input_tokens_seen": 19974336,
      "step": 94645
    },
    {
      "epoch": 10.412541254125413,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.01653555665003006,
      "loss": 0.2314,
      "num_input_tokens_seen": 19975488,
      "step": 94650
    },
    {
      "epoch": 10.413091309130913,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.01653412416779471,
      "loss": 0.2335,
      "num_input_tokens_seen": 19976448,
      "step": 94655
    },
    {
      "epoch": 10.413641364136414,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01653269167142007,
      "loss": 0.2319,
      "num_input_tokens_seen": 19977504,
      "step": 94660
    },
    {
      "epoch": 10.414191419141915,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.016531259160919347,
      "loss": 0.2319,
      "num_input_tokens_seen": 19978560,
      "step": 94665
    },
    {
      "epoch": 10.414741474147414,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.016529826636305745,
      "loss": 0.2303,
      "num_input_tokens_seen": 19979584,
      "step": 94670
    },
    {
      "epoch": 10.415291529152915,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.016528394097592464,
      "loss": 0.2335,
      "num_input_tokens_seen": 19980640,
      "step": 94675
    },
    {
      "epoch": 10.415841584158416,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.016526961544792704,
      "loss": 0.2309,
      "num_input_tokens_seen": 19981664,
      "step": 94680
    },
    {
      "epoch": 10.416391639163916,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.016525528977919673,
      "loss": 0.2314,
      "num_input_tokens_seen": 19982688,
      "step": 94685
    },
    {
      "epoch": 10.416941694169417,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.016524096396986578,
      "loss": 0.234,
      "num_input_tokens_seen": 19983776,
      "step": 94690
    },
    {
      "epoch": 10.417491749174918,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.016522663802006614,
      "loss": 0.2303,
      "num_input_tokens_seen": 19984832,
      "step": 94695
    },
    {
      "epoch": 10.418041804180419,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.01652123119299299,
      "loss": 0.2309,
      "num_input_tokens_seen": 19985984,
      "step": 94700
    },
    {
      "epoch": 10.418591859185918,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.01651979856995891,
      "loss": 0.2314,
      "num_input_tokens_seen": 19987040,
      "step": 94705
    },
    {
      "epoch": 10.41914191419142,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.01651836593291757,
      "loss": 0.2303,
      "num_input_tokens_seen": 19988096,
      "step": 94710
    },
    {
      "epoch": 10.41969196919692,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.016516933281882185,
      "loss": 0.2314,
      "num_input_tokens_seen": 19989088,
      "step": 94715
    },
    {
      "epoch": 10.42024202420242,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.016515500616865953,
      "loss": 0.2303,
      "num_input_tokens_seen": 19990176,
      "step": 94720
    },
    {
      "epoch": 10.42079207920792,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.016514067937882078,
      "loss": 0.2308,
      "num_input_tokens_seen": 19991200,
      "step": 94725
    },
    {
      "epoch": 10.421342134213422,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.016512635244943768,
      "loss": 0.2314,
      "num_input_tokens_seen": 19992256,
      "step": 94730
    },
    {
      "epoch": 10.421892189218921,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01651120253806422,
      "loss": 0.2303,
      "num_input_tokens_seen": 19993344,
      "step": 94735
    },
    {
      "epoch": 10.422442244224422,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.016509769817256645,
      "loss": 0.2319,
      "num_input_tokens_seen": 19994432,
      "step": 94740
    },
    {
      "epoch": 10.422992299229923,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.016508337082534248,
      "loss": 0.2298,
      "num_input_tokens_seen": 19995488,
      "step": 94745
    },
    {
      "epoch": 10.423542354235423,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.016506904333910232,
      "loss": 0.2308,
      "num_input_tokens_seen": 19996576,
      "step": 94750
    },
    {
      "epoch": 10.424092409240924,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.016505471571397798,
      "loss": 0.2324,
      "num_input_tokens_seen": 19997696,
      "step": 94755
    },
    {
      "epoch": 10.424642464246425,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.016504038795010157,
      "loss": 0.2329,
      "num_input_tokens_seen": 19998784,
      "step": 94760
    },
    {
      "epoch": 10.425192519251926,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.016502606004760514,
      "loss": 0.2329,
      "num_input_tokens_seen": 19999872,
      "step": 94765
    },
    {
      "epoch": 10.425742574257425,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.016501173200662075,
      "loss": 0.2303,
      "num_input_tokens_seen": 20000896,
      "step": 94770
    },
    {
      "epoch": 10.426292629262926,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.016499740382728036,
      "loss": 0.233,
      "num_input_tokens_seen": 20001952,
      "step": 94775
    },
    {
      "epoch": 10.426842684268427,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.016498307550971616,
      "loss": 0.2314,
      "num_input_tokens_seen": 20002976,
      "step": 94780
    },
    {
      "epoch": 10.427392739273927,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.016496874705406006,
      "loss": 0.2298,
      "num_input_tokens_seen": 20004000,
      "step": 94785
    },
    {
      "epoch": 10.427942794279428,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.01649544184604443,
      "loss": 0.2298,
      "num_input_tokens_seen": 20005024,
      "step": 94790
    },
    {
      "epoch": 10.428492849284929,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.016494008972900074,
      "loss": 0.2303,
      "num_input_tokens_seen": 20006144,
      "step": 94795
    },
    {
      "epoch": 10.429042904290428,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.01649257608598616,
      "loss": 0.2319,
      "num_input_tokens_seen": 20007168,
      "step": 94800
    },
    {
      "epoch": 10.42959295929593,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.016491143185315887,
      "loss": 0.2314,
      "num_input_tokens_seen": 20008288,
      "step": 94805
    },
    {
      "epoch": 10.43014301430143,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.016489710270902465,
      "loss": 0.2314,
      "num_input_tokens_seen": 20009344,
      "step": 94810
    },
    {
      "epoch": 10.430693069306932,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.016488277342759098,
      "loss": 0.2293,
      "num_input_tokens_seen": 20010464,
      "step": 94815
    },
    {
      "epoch": 10.43124312431243,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01648684440089899,
      "loss": 0.2303,
      "num_input_tokens_seen": 20011488,
      "step": 94820
    },
    {
      "epoch": 10.431793179317932,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01648541144533535,
      "loss": 0.2303,
      "num_input_tokens_seen": 20012480,
      "step": 94825
    },
    {
      "epoch": 10.432343234323433,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.01648397847608139,
      "loss": 0.2308,
      "num_input_tokens_seen": 20013536,
      "step": 94830
    },
    {
      "epoch": 10.432893289328932,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.01648254549315031,
      "loss": 0.2298,
      "num_input_tokens_seen": 20014560,
      "step": 94835
    },
    {
      "epoch": 10.433443344334433,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.016481112496555317,
      "loss": 0.2319,
      "num_input_tokens_seen": 20015616,
      "step": 94840
    },
    {
      "epoch": 10.433993399339935,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016479679486309625,
      "loss": 0.2309,
      "num_input_tokens_seen": 20016704,
      "step": 94845
    },
    {
      "epoch": 10.434543454345434,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.016478246462426436,
      "loss": 0.2293,
      "num_input_tokens_seen": 20017728,
      "step": 94850
    },
    {
      "epoch": 10.435093509350935,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016476813424918954,
      "loss": 0.2298,
      "num_input_tokens_seen": 20018752,
      "step": 94855
    },
    {
      "epoch": 10.435643564356436,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.0164753803738004,
      "loss": 0.2298,
      "num_input_tokens_seen": 20019808,
      "step": 94860
    },
    {
      "epoch": 10.436193619361935,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01647394730908397,
      "loss": 0.2324,
      "num_input_tokens_seen": 20020832,
      "step": 94865
    },
    {
      "epoch": 10.436743674367436,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.016472514230782873,
      "loss": 0.2314,
      "num_input_tokens_seen": 20021920,
      "step": 94870
    },
    {
      "epoch": 10.437293729372938,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01647108113891032,
      "loss": 0.2324,
      "num_input_tokens_seen": 20022944,
      "step": 94875
    },
    {
      "epoch": 10.437843784378439,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.016469648033479518,
      "loss": 0.2351,
      "num_input_tokens_seen": 20024032,
      "step": 94880
    },
    {
      "epoch": 10.438393839383938,
      "grad_norm": 0.00063323974609375,
      "learning_rate": 0.016468214914503676,
      "loss": 0.2303,
      "num_input_tokens_seen": 20025120,
      "step": 94885
    },
    {
      "epoch": 10.438943894389439,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.016466781781996002,
      "loss": 0.2303,
      "num_input_tokens_seen": 20026144,
      "step": 94890
    },
    {
      "epoch": 10.43949394939494,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0164653486359697,
      "loss": 0.2303,
      "num_input_tokens_seen": 20027168,
      "step": 94895
    },
    {
      "epoch": 10.44004400440044,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.016463915476437982,
      "loss": 0.2304,
      "num_input_tokens_seen": 20028192,
      "step": 94900
    },
    {
      "epoch": 10.44059405940594,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.016462482303414062,
      "loss": 0.2309,
      "num_input_tokens_seen": 20029216,
      "step": 94905
    },
    {
      "epoch": 10.441144114411442,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.016461049116911142,
      "loss": 0.2319,
      "num_input_tokens_seen": 20030272,
      "step": 94910
    },
    {
      "epoch": 10.441694169416941,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.016459615916942436,
      "loss": 0.2314,
      "num_input_tokens_seen": 20031360,
      "step": 94915
    },
    {
      "epoch": 10.442244224422442,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.01645818270352115,
      "loss": 0.2351,
      "num_input_tokens_seen": 20032416,
      "step": 94920
    },
    {
      "epoch": 10.442794279427943,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.016456749476660495,
      "loss": 0.2309,
      "num_input_tokens_seen": 20033472,
      "step": 94925
    },
    {
      "epoch": 10.443344334433444,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01645531623637367,
      "loss": 0.2324,
      "num_input_tokens_seen": 20034528,
      "step": 94930
    },
    {
      "epoch": 10.443894389438944,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.016453882982673907,
      "loss": 0.2314,
      "num_input_tokens_seen": 20035616,
      "step": 94935
    },
    {
      "epoch": 10.444444444444445,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.016452449715574392,
      "loss": 0.2314,
      "num_input_tokens_seen": 20036672,
      "step": 94940
    },
    {
      "epoch": 10.444994499449946,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01645101643508835,
      "loss": 0.2314,
      "num_input_tokens_seen": 20037728,
      "step": 94945
    },
    {
      "epoch": 10.445544554455445,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.016449583141228987,
      "loss": 0.2319,
      "num_input_tokens_seen": 20038784,
      "step": 94950
    },
    {
      "epoch": 10.446094609460946,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01644814983400951,
      "loss": 0.2303,
      "num_input_tokens_seen": 20039904,
      "step": 94955
    },
    {
      "epoch": 10.446644664466447,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.016446716513443132,
      "loss": 0.2298,
      "num_input_tokens_seen": 20040992,
      "step": 94960
    },
    {
      "epoch": 10.447194719471947,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.01644528317954306,
      "loss": 0.2324,
      "num_input_tokens_seen": 20041984,
      "step": 94965
    },
    {
      "epoch": 10.447744774477448,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0164438498323225,
      "loss": 0.2303,
      "num_input_tokens_seen": 20043008,
      "step": 94970
    },
    {
      "epoch": 10.448294829482949,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.016442416471794678,
      "loss": 0.2335,
      "num_input_tokens_seen": 20044096,
      "step": 94975
    },
    {
      "epoch": 10.448844884488448,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.016440983097972796,
      "loss": 0.2309,
      "num_input_tokens_seen": 20045120,
      "step": 94980
    },
    {
      "epoch": 10.44939493949395,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016439549710870062,
      "loss": 0.2329,
      "num_input_tokens_seen": 20046208,
      "step": 94985
    },
    {
      "epoch": 10.44994499449945,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.016438116310499684,
      "loss": 0.2298,
      "num_input_tokens_seen": 20047296,
      "step": 94990
    },
    {
      "epoch": 10.450495049504951,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.016436682896874888,
      "loss": 0.2319,
      "num_input_tokens_seen": 20048416,
      "step": 94995
    },
    {
      "epoch": 10.45104510451045,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.016435249470008864,
      "loss": 0.2324,
      "num_input_tokens_seen": 20049408,
      "step": 95000
    },
    {
      "epoch": 10.451595159515952,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016433816029914845,
      "loss": 0.2314,
      "num_input_tokens_seen": 20050496,
      "step": 95005
    },
    {
      "epoch": 10.452145214521453,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.016432382576606026,
      "loss": 0.2303,
      "num_input_tokens_seen": 20051584,
      "step": 95010
    },
    {
      "epoch": 10.452695269526952,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01643094911009562,
      "loss": 0.2304,
      "num_input_tokens_seen": 20052608,
      "step": 95015
    },
    {
      "epoch": 10.453245324532453,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.016429515630396848,
      "loss": 0.2298,
      "num_input_tokens_seen": 20053632,
      "step": 95020
    },
    {
      "epoch": 10.453795379537954,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01642808213752292,
      "loss": 0.2329,
      "num_input_tokens_seen": 20054688,
      "step": 95025
    },
    {
      "epoch": 10.454345434543454,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01642664863148704,
      "loss": 0.2329,
      "num_input_tokens_seen": 20055712,
      "step": 95030
    },
    {
      "epoch": 10.454895489548955,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.016425215112302424,
      "loss": 0.2298,
      "num_input_tokens_seen": 20056736,
      "step": 95035
    },
    {
      "epoch": 10.455445544554456,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.016423781579982283,
      "loss": 0.2329,
      "num_input_tokens_seen": 20057760,
      "step": 95040
    },
    {
      "epoch": 10.455995599559955,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01642234803453983,
      "loss": 0.2319,
      "num_input_tokens_seen": 20058816,
      "step": 95045
    },
    {
      "epoch": 10.456545654565456,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.016420914475988285,
      "loss": 0.2329,
      "num_input_tokens_seen": 20059776,
      "step": 95050
    },
    {
      "epoch": 10.457095709570957,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.016419480904340845,
      "loss": 0.2319,
      "num_input_tokens_seen": 20060832,
      "step": 95055
    },
    {
      "epoch": 10.457645764576458,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01641804731961073,
      "loss": 0.2298,
      "num_input_tokens_seen": 20061920,
      "step": 95060
    },
    {
      "epoch": 10.458195819581958,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01641661372181116,
      "loss": 0.2309,
      "num_input_tokens_seen": 20062976,
      "step": 95065
    },
    {
      "epoch": 10.458745874587459,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.016415180110955337,
      "loss": 0.2319,
      "num_input_tokens_seen": 20064096,
      "step": 95070
    },
    {
      "epoch": 10.45929592959296,
      "grad_norm": 0.002716064453125,
      "learning_rate": 0.01641374648705648,
      "loss": 0.2314,
      "num_input_tokens_seen": 20065216,
      "step": 95075
    },
    {
      "epoch": 10.45984598459846,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.016412312850127796,
      "loss": 0.2309,
      "num_input_tokens_seen": 20066272,
      "step": 95080
    },
    {
      "epoch": 10.46039603960396,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.016410879200182502,
      "loss": 0.2308,
      "num_input_tokens_seen": 20067296,
      "step": 95085
    },
    {
      "epoch": 10.460946094609461,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.016409445537233815,
      "loss": 0.2303,
      "num_input_tokens_seen": 20068352,
      "step": 95090
    },
    {
      "epoch": 10.46149614961496,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.016408011861294945,
      "loss": 0.2329,
      "num_input_tokens_seen": 20069440,
      "step": 95095
    },
    {
      "epoch": 10.462046204620462,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.016406578172379106,
      "loss": 0.2314,
      "num_input_tokens_seen": 20070528,
      "step": 95100
    },
    {
      "epoch": 10.462596259625963,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.016405144470499507,
      "loss": 0.2309,
      "num_input_tokens_seen": 20071648,
      "step": 95105
    },
    {
      "epoch": 10.463146314631462,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.01640371075566937,
      "loss": 0.2288,
      "num_input_tokens_seen": 20072672,
      "step": 95110
    },
    {
      "epoch": 10.463696369636963,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.016402277027901896,
      "loss": 0.2329,
      "num_input_tokens_seen": 20073760,
      "step": 95115
    },
    {
      "epoch": 10.464246424642464,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.016400843287210316,
      "loss": 0.2303,
      "num_input_tokens_seen": 20074720,
      "step": 95120
    },
    {
      "epoch": 10.464796479647966,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.01639940953360784,
      "loss": 0.2303,
      "num_input_tokens_seen": 20075744,
      "step": 95125
    },
    {
      "epoch": 10.465346534653465,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016397975767107664,
      "loss": 0.2319,
      "num_input_tokens_seen": 20076800,
      "step": 95130
    },
    {
      "epoch": 10.465896589658966,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.01639654198772303,
      "loss": 0.2298,
      "num_input_tokens_seen": 20077920,
      "step": 95135
    },
    {
      "epoch": 10.466446644664467,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016395108195467127,
      "loss": 0.2304,
      "num_input_tokens_seen": 20078944,
      "step": 95140
    },
    {
      "epoch": 10.466996699669966,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.016393674390353188,
      "loss": 0.2308,
      "num_input_tokens_seen": 20079904,
      "step": 95145
    },
    {
      "epoch": 10.467546754675467,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01639224057239442,
      "loss": 0.2329,
      "num_input_tokens_seen": 20081024,
      "step": 95150
    },
    {
      "epoch": 10.468096809680969,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.016390806741604036,
      "loss": 0.2293,
      "num_input_tokens_seen": 20082080,
      "step": 95155
    },
    {
      "epoch": 10.468646864686468,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.016389372897995254,
      "loss": 0.2324,
      "num_input_tokens_seen": 20083104,
      "step": 95160
    },
    {
      "epoch": 10.469196919691969,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01638793904158129,
      "loss": 0.2319,
      "num_input_tokens_seen": 20084128,
      "step": 95165
    },
    {
      "epoch": 10.46974697469747,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.016386505172375363,
      "loss": 0.2319,
      "num_input_tokens_seen": 20085120,
      "step": 95170
    },
    {
      "epoch": 10.47029702970297,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016385071290390674,
      "loss": 0.2308,
      "num_input_tokens_seen": 20086176,
      "step": 95175
    },
    {
      "epoch": 10.47084708470847,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.016383637395640454,
      "loss": 0.2309,
      "num_input_tokens_seen": 20087264,
      "step": 95180
    },
    {
      "epoch": 10.471397139713972,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.01638220348813791,
      "loss": 0.233,
      "num_input_tokens_seen": 20088352,
      "step": 95185
    },
    {
      "epoch": 10.471947194719473,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.016380769567896254,
      "loss": 0.2314,
      "num_input_tokens_seen": 20089408,
      "step": 95190
    },
    {
      "epoch": 10.472497249724972,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.016379335634928716,
      "loss": 0.2335,
      "num_input_tokens_seen": 20090464,
      "step": 95195
    },
    {
      "epoch": 10.473047304730473,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.016377901689248506,
      "loss": 0.2308,
      "num_input_tokens_seen": 20091488,
      "step": 95200
    },
    {
      "epoch": 10.473597359735974,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.016376467730868826,
      "loss": 0.2314,
      "num_input_tokens_seen": 20092576,
      "step": 95205
    },
    {
      "epoch": 10.474147414741473,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.016375033759802912,
      "loss": 0.2356,
      "num_input_tokens_seen": 20093696,
      "step": 95210
    },
    {
      "epoch": 10.474697469746975,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.016373599776063966,
      "loss": 0.2309,
      "num_input_tokens_seen": 20094784,
      "step": 95215
    },
    {
      "epoch": 10.475247524752476,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.016372165779665215,
      "loss": 0.2314,
      "num_input_tokens_seen": 20095808,
      "step": 95220
    },
    {
      "epoch": 10.475797579757975,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01637073177061987,
      "loss": 0.2324,
      "num_input_tokens_seen": 20096864,
      "step": 95225
    },
    {
      "epoch": 10.476347634763476,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.016369297748941143,
      "loss": 0.2314,
      "num_input_tokens_seen": 20097952,
      "step": 95230
    },
    {
      "epoch": 10.476897689768977,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.016367863714642255,
      "loss": 0.2309,
      "num_input_tokens_seen": 20099008,
      "step": 95235
    },
    {
      "epoch": 10.477447744774478,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.016366429667736434,
      "loss": 0.2345,
      "num_input_tokens_seen": 20100160,
      "step": 95240
    },
    {
      "epoch": 10.477997799779978,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.016364995608236874,
      "loss": 0.2319,
      "num_input_tokens_seen": 20101152,
      "step": 95245
    },
    {
      "epoch": 10.478547854785479,
      "grad_norm": 0.00078582763671875,
      "learning_rate": 0.016363561536156812,
      "loss": 0.2298,
      "num_input_tokens_seen": 20102272,
      "step": 95250
    },
    {
      "epoch": 10.47909790979098,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.016362127451509455,
      "loss": 0.234,
      "num_input_tokens_seen": 20103328,
      "step": 95255
    },
    {
      "epoch": 10.479647964796479,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.016360693354308023,
      "loss": 0.2324,
      "num_input_tokens_seen": 20104352,
      "step": 95260
    },
    {
      "epoch": 10.48019801980198,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.016359259244565737,
      "loss": 0.2324,
      "num_input_tokens_seen": 20105440,
      "step": 95265
    },
    {
      "epoch": 10.480748074807481,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.01635782512229581,
      "loss": 0.2303,
      "num_input_tokens_seen": 20106528,
      "step": 95270
    },
    {
      "epoch": 10.48129812981298,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.016356390987511453,
      "loss": 0.2319,
      "num_input_tokens_seen": 20107552,
      "step": 95275
    },
    {
      "epoch": 10.481848184818482,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.016354956840225896,
      "loss": 0.2309,
      "num_input_tokens_seen": 20108544,
      "step": 95280
    },
    {
      "epoch": 10.482398239823983,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.016353522680452353,
      "loss": 0.2288,
      "num_input_tokens_seen": 20109632,
      "step": 95285
    },
    {
      "epoch": 10.482948294829482,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.01635208850820404,
      "loss": 0.2314,
      "num_input_tokens_seen": 20110688,
      "step": 95290
    },
    {
      "epoch": 10.483498349834983,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.01635065432349418,
      "loss": 0.2303,
      "num_input_tokens_seen": 20111680,
      "step": 95295
    },
    {
      "epoch": 10.484048404840484,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.01634922012633598,
      "loss": 0.233,
      "num_input_tokens_seen": 20112704,
      "step": 95300
    },
    {
      "epoch": 10.484598459845985,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.016347785916742665,
      "loss": 0.2335,
      "num_input_tokens_seen": 20113760,
      "step": 95305
    },
    {
      "epoch": 10.485148514851485,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.01634635169472746,
      "loss": 0.2309,
      "num_input_tokens_seen": 20114848,
      "step": 95310
    },
    {
      "epoch": 10.485698569856986,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01634491746030358,
      "loss": 0.233,
      "num_input_tokens_seen": 20115968,
      "step": 95315
    },
    {
      "epoch": 10.486248624862487,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.016343483213484233,
      "loss": 0.233,
      "num_input_tokens_seen": 20116992,
      "step": 95320
    },
    {
      "epoch": 10.486798679867986,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01634204895428265,
      "loss": 0.2314,
      "num_input_tokens_seen": 20117984,
      "step": 95325
    },
    {
      "epoch": 10.487348734873487,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.016340614682712046,
      "loss": 0.234,
      "num_input_tokens_seen": 20119008,
      "step": 95330
    },
    {
      "epoch": 10.487898789878988,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.016339180398785638,
      "loss": 0.2314,
      "num_input_tokens_seen": 20120128,
      "step": 95335
    },
    {
      "epoch": 10.488448844884488,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.016337746102516648,
      "loss": 0.2324,
      "num_input_tokens_seen": 20121152,
      "step": 95340
    },
    {
      "epoch": 10.488998899889989,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.016336311793918295,
      "loss": 0.2298,
      "num_input_tokens_seen": 20122144,
      "step": 95345
    },
    {
      "epoch": 10.48954895489549,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.016334877473003794,
      "loss": 0.234,
      "num_input_tokens_seen": 20123264,
      "step": 95350
    },
    {
      "epoch": 10.490099009900991,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01633344313978637,
      "loss": 0.2319,
      "num_input_tokens_seen": 20124352,
      "step": 95355
    },
    {
      "epoch": 10.49064906490649,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01633200879427924,
      "loss": 0.2293,
      "num_input_tokens_seen": 20125472,
      "step": 95360
    },
    {
      "epoch": 10.491199119911991,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.01633057443649563,
      "loss": 0.2293,
      "num_input_tokens_seen": 20126528,
      "step": 95365
    },
    {
      "epoch": 10.491749174917492,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.016329140066448745,
      "loss": 0.2335,
      "num_input_tokens_seen": 20127552,
      "step": 95370
    },
    {
      "epoch": 10.492299229922992,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.016327705684151816,
      "loss": 0.2324,
      "num_input_tokens_seen": 20128608,
      "step": 95375
    },
    {
      "epoch": 10.492849284928493,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01632627128961806,
      "loss": 0.2325,
      "num_input_tokens_seen": 20129664,
      "step": 95380
    },
    {
      "epoch": 10.493399339933994,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.016324836882860704,
      "loss": 0.2324,
      "num_input_tokens_seen": 20130688,
      "step": 95385
    },
    {
      "epoch": 10.493949394939493,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.016323402463892956,
      "loss": 0.2298,
      "num_input_tokens_seen": 20131744,
      "step": 95390
    },
    {
      "epoch": 10.494499449944994,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.016321968032728045,
      "loss": 0.2319,
      "num_input_tokens_seen": 20132800,
      "step": 95395
    },
    {
      "epoch": 10.495049504950495,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.01632053358937919,
      "loss": 0.2314,
      "num_input_tokens_seen": 20133760,
      "step": 95400
    },
    {
      "epoch": 10.495599559955995,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.016319099133859608,
      "loss": 0.234,
      "num_input_tokens_seen": 20134816,
      "step": 95405
    },
    {
      "epoch": 10.496149614961496,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016317664666182525,
      "loss": 0.2329,
      "num_input_tokens_seen": 20135776,
      "step": 95410
    },
    {
      "epoch": 10.496699669966997,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.016316230186361157,
      "loss": 0.2314,
      "num_input_tokens_seen": 20136768,
      "step": 95415
    },
    {
      "epoch": 10.497249724972498,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.016314795694408725,
      "loss": 0.2324,
      "num_input_tokens_seen": 20137792,
      "step": 95420
    },
    {
      "epoch": 10.497799779977997,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.016313361190338455,
      "loss": 0.2314,
      "num_input_tokens_seen": 20138848,
      "step": 95425
    },
    {
      "epoch": 10.498349834983498,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01631192667416357,
      "loss": 0.2313,
      "num_input_tokens_seen": 20139904,
      "step": 95430
    },
    {
      "epoch": 10.498899889989,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.016310492145897276,
      "loss": 0.2308,
      "num_input_tokens_seen": 20140896,
      "step": 95435
    },
    {
      "epoch": 10.499449944994499,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01630905760555281,
      "loss": 0.234,
      "num_input_tokens_seen": 20141920,
      "step": 95440
    },
    {
      "epoch": 10.5,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01630762305314339,
      "loss": 0.2319,
      "num_input_tokens_seen": 20143040,
      "step": 95445
    },
    {
      "epoch": 10.500550055005501,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.016306188488682233,
      "loss": 0.2319,
      "num_input_tokens_seen": 20144096,
      "step": 95450
    },
    {
      "epoch": 10.501100110011,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.016304753912182564,
      "loss": 0.2309,
      "num_input_tokens_seen": 20145152,
      "step": 95455
    },
    {
      "epoch": 10.501650165016502,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.016303319323657604,
      "loss": 0.2309,
      "num_input_tokens_seen": 20146176,
      "step": 95460
    },
    {
      "epoch": 10.502200220022003,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.016301884723120573,
      "loss": 0.2319,
      "num_input_tokens_seen": 20147232,
      "step": 95465
    },
    {
      "epoch": 10.502750275027502,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.016300450110584702,
      "loss": 0.2309,
      "num_input_tokens_seen": 20148320,
      "step": 95470
    },
    {
      "epoch": 10.503300330033003,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.0162990154860632,
      "loss": 0.233,
      "num_input_tokens_seen": 20149376,
      "step": 95475
    },
    {
      "epoch": 10.503850385038504,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.0162975808495693,
      "loss": 0.2304,
      "num_input_tokens_seen": 20150496,
      "step": 95480
    },
    {
      "epoch": 10.504400440044005,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.01629614620111622,
      "loss": 0.2298,
      "num_input_tokens_seen": 20151552,
      "step": 95485
    },
    {
      "epoch": 10.504950495049505,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.016294711540717176,
      "loss": 0.2314,
      "num_input_tokens_seen": 20152608,
      "step": 95490
    },
    {
      "epoch": 10.505500550055006,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.016293276868385397,
      "loss": 0.2303,
      "num_input_tokens_seen": 20153664,
      "step": 95495
    },
    {
      "epoch": 10.506050605060507,
      "grad_norm": 0.0004329681396484375,
      "learning_rate": 0.01629184218413411,
      "loss": 0.2319,
      "num_input_tokens_seen": 20154656,
      "step": 95500
    },
    {
      "epoch": 10.506600660066006,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.016290407487976537,
      "loss": 0.2314,
      "num_input_tokens_seen": 20155744,
      "step": 95505
    },
    {
      "epoch": 10.507150715071507,
      "grad_norm": 0.000667572021484375,
      "learning_rate": 0.016288972779925893,
      "loss": 0.2329,
      "num_input_tokens_seen": 20156800,
      "step": 95510
    },
    {
      "epoch": 10.507700770077008,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.016287538059995405,
      "loss": 0.2319,
      "num_input_tokens_seen": 20157888,
      "step": 95515
    },
    {
      "epoch": 10.508250825082508,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.016286103328198298,
      "loss": 0.2298,
      "num_input_tokens_seen": 20158880,
      "step": 95520
    },
    {
      "epoch": 10.508800880088009,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.016284668584547798,
      "loss": 0.2298,
      "num_input_tokens_seen": 20160000,
      "step": 95525
    },
    {
      "epoch": 10.50935093509351,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01628323382905712,
      "loss": 0.2319,
      "num_input_tokens_seen": 20161088,
      "step": 95530
    },
    {
      "epoch": 10.509900990099009,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.016281799061739488,
      "loss": 0.2298,
      "num_input_tokens_seen": 20162144,
      "step": 95535
    },
    {
      "epoch": 10.51045104510451,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.01628036428260813,
      "loss": 0.2308,
      "num_input_tokens_seen": 20163232,
      "step": 95540
    },
    {
      "epoch": 10.511001100110011,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.016278929491676274,
      "loss": 0.2345,
      "num_input_tokens_seen": 20164320,
      "step": 95545
    },
    {
      "epoch": 10.511551155115512,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.016277494688957134,
      "loss": 0.2324,
      "num_input_tokens_seen": 20165408,
      "step": 95550
    },
    {
      "epoch": 10.512101210121012,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.016276059874463938,
      "loss": 0.2288,
      "num_input_tokens_seen": 20166432,
      "step": 95555
    },
    {
      "epoch": 10.512651265126513,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.016274625048209916,
      "loss": 0.2309,
      "num_input_tokens_seen": 20167552,
      "step": 95560
    },
    {
      "epoch": 10.513201320132014,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.016273190210208285,
      "loss": 0.2308,
      "num_input_tokens_seen": 20168576,
      "step": 95565
    },
    {
      "epoch": 10.513751375137513,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.016271755360472262,
      "loss": 0.2324,
      "num_input_tokens_seen": 20169696,
      "step": 95570
    },
    {
      "epoch": 10.514301430143014,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.01627032049901509,
      "loss": 0.2293,
      "num_input_tokens_seen": 20170752,
      "step": 95575
    },
    {
      "epoch": 10.514851485148515,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01626888562584998,
      "loss": 0.2314,
      "num_input_tokens_seen": 20171872,
      "step": 95580
    },
    {
      "epoch": 10.515401540154015,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.016267450740990157,
      "loss": 0.2309,
      "num_input_tokens_seen": 20172960,
      "step": 95585
    },
    {
      "epoch": 10.515951595159516,
      "grad_norm": 0.00058746337890625,
      "learning_rate": 0.016266015844448856,
      "loss": 0.2324,
      "num_input_tokens_seen": 20174016,
      "step": 95590
    },
    {
      "epoch": 10.516501650165017,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.01626458093623929,
      "loss": 0.2329,
      "num_input_tokens_seen": 20175072,
      "step": 95595
    },
    {
      "epoch": 10.517051705170516,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.01626314601637469,
      "loss": 0.2335,
      "num_input_tokens_seen": 20176192,
      "step": 95600
    },
    {
      "epoch": 10.517601760176017,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.016261711084868277,
      "loss": 0.2319,
      "num_input_tokens_seen": 20177184,
      "step": 95605
    },
    {
      "epoch": 10.518151815181518,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01626027614173328,
      "loss": 0.2293,
      "num_input_tokens_seen": 20178272,
      "step": 95610
    },
    {
      "epoch": 10.51870187018702,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.016258841186982922,
      "loss": 0.2303,
      "num_input_tokens_seen": 20179264,
      "step": 95615
    },
    {
      "epoch": 10.519251925192519,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.01625740622063043,
      "loss": 0.2324,
      "num_input_tokens_seen": 20180352,
      "step": 95620
    },
    {
      "epoch": 10.51980198019802,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.016255971242689026,
      "loss": 0.2303,
      "num_input_tokens_seen": 20181344,
      "step": 95625
    },
    {
      "epoch": 10.520352035203521,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.016254536253171938,
      "loss": 0.2293,
      "num_input_tokens_seen": 20182400,
      "step": 95630
    },
    {
      "epoch": 10.52090209020902,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01625310125209239,
      "loss": 0.2314,
      "num_input_tokens_seen": 20183360,
      "step": 95635
    },
    {
      "epoch": 10.521452145214521,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.01625166623946361,
      "loss": 0.2319,
      "num_input_tokens_seen": 20184480,
      "step": 95640
    },
    {
      "epoch": 10.522002200220022,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.016250231215298828,
      "loss": 0.2314,
      "num_input_tokens_seen": 20185536,
      "step": 95645
    },
    {
      "epoch": 10.522552255225522,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.016248796179611263,
      "loss": 0.2298,
      "num_input_tokens_seen": 20186656,
      "step": 95650
    },
    {
      "epoch": 10.523102310231023,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.016247361132414134,
      "loss": 0.2309,
      "num_input_tokens_seen": 20187680,
      "step": 95655
    },
    {
      "epoch": 10.523652365236524,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.016245926073720683,
      "loss": 0.2309,
      "num_input_tokens_seen": 20188672,
      "step": 95660
    },
    {
      "epoch": 10.524202420242025,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.016244491003544134,
      "loss": 0.2335,
      "num_input_tokens_seen": 20189728,
      "step": 95665
    },
    {
      "epoch": 10.524752475247524,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.016243055921897703,
      "loss": 0.2309,
      "num_input_tokens_seen": 20190816,
      "step": 95670
    },
    {
      "epoch": 10.525302530253025,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.01624162082879462,
      "loss": 0.2324,
      "num_input_tokens_seen": 20191936,
      "step": 95675
    },
    {
      "epoch": 10.525852585258527,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.016240185724248117,
      "loss": 0.2319,
      "num_input_tokens_seen": 20192992,
      "step": 95680
    },
    {
      "epoch": 10.526402640264026,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.016238750608271416,
      "loss": 0.2303,
      "num_input_tokens_seen": 20194016,
      "step": 95685
    },
    {
      "epoch": 10.526952695269527,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.016237315480877746,
      "loss": 0.2319,
      "num_input_tokens_seen": 20195072,
      "step": 95690
    },
    {
      "epoch": 10.527502750275028,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.01623588034208033,
      "loss": 0.2324,
      "num_input_tokens_seen": 20196064,
      "step": 95695
    },
    {
      "epoch": 10.528052805280527,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.0162344451918924,
      "loss": 0.2324,
      "num_input_tokens_seen": 20197120,
      "step": 95700
    },
    {
      "epoch": 10.528602860286028,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.016233010030327184,
      "loss": 0.2308,
      "num_input_tokens_seen": 20198176,
      "step": 95705
    },
    {
      "epoch": 10.52915291529153,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.016231574857397905,
      "loss": 0.2319,
      "num_input_tokens_seen": 20199232,
      "step": 95710
    },
    {
      "epoch": 10.52970297029703,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.016230139673117785,
      "loss": 0.2319,
      "num_input_tokens_seen": 20200320,
      "step": 95715
    },
    {
      "epoch": 10.53025302530253,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01622870447750007,
      "loss": 0.2298,
      "num_input_tokens_seen": 20201344,
      "step": 95720
    },
    {
      "epoch": 10.530803080308031,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.016227269270557964,
      "loss": 0.2314,
      "num_input_tokens_seen": 20202432,
      "step": 95725
    },
    {
      "epoch": 10.531353135313532,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01622583405230471,
      "loss": 0.2319,
      "num_input_tokens_seen": 20203520,
      "step": 95730
    },
    {
      "epoch": 10.531903190319031,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.016224398822753533,
      "loss": 0.2314,
      "num_input_tokens_seen": 20204608,
      "step": 95735
    },
    {
      "epoch": 10.532453245324533,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.01622296358191766,
      "loss": 0.2304,
      "num_input_tokens_seen": 20205632,
      "step": 95740
    },
    {
      "epoch": 10.533003300330034,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.016221528329810317,
      "loss": 0.2319,
      "num_input_tokens_seen": 20206656,
      "step": 95745
    },
    {
      "epoch": 10.533553355335533,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.016220093066444737,
      "loss": 0.2319,
      "num_input_tokens_seen": 20207712,
      "step": 95750
    },
    {
      "epoch": 10.534103410341034,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.016218657791834137,
      "loss": 0.2319,
      "num_input_tokens_seen": 20208704,
      "step": 95755
    },
    {
      "epoch": 10.534653465346535,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.016217222505991755,
      "loss": 0.2314,
      "num_input_tokens_seen": 20209792,
      "step": 95760
    },
    {
      "epoch": 10.535203520352034,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.016215787208930826,
      "loss": 0.2319,
      "num_input_tokens_seen": 20210816,
      "step": 95765
    },
    {
      "epoch": 10.535753575357536,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01621435190066456,
      "loss": 0.233,
      "num_input_tokens_seen": 20211936,
      "step": 95770
    },
    {
      "epoch": 10.536303630363037,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.0162129165812062,
      "loss": 0.2288,
      "num_input_tokens_seen": 20213024,
      "step": 95775
    },
    {
      "epoch": 10.536853685368538,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.016211481250568968,
      "loss": 0.2329,
      "num_input_tokens_seen": 20214112,
      "step": 95780
    },
    {
      "epoch": 10.537403740374037,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.016210045908766093,
      "loss": 0.2308,
      "num_input_tokens_seen": 20215168,
      "step": 95785
    },
    {
      "epoch": 10.537953795379538,
      "grad_norm": 0.00299072265625,
      "learning_rate": 0.01620861055581081,
      "loss": 0.2304,
      "num_input_tokens_seen": 20216224,
      "step": 95790
    },
    {
      "epoch": 10.53850385038504,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.01620717519171634,
      "loss": 0.2303,
      "num_input_tokens_seen": 20217312,
      "step": 95795
    },
    {
      "epoch": 10.539053905390539,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.016205739816495916,
      "loss": 0.2319,
      "num_input_tokens_seen": 20218336,
      "step": 95800
    },
    {
      "epoch": 10.53960396039604,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.016204304430162766,
      "loss": 0.2319,
      "num_input_tokens_seen": 20219392,
      "step": 95805
    },
    {
      "epoch": 10.54015401540154,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.016202869032730122,
      "loss": 0.2314,
      "num_input_tokens_seen": 20220448,
      "step": 95810
    },
    {
      "epoch": 10.54070407040704,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01620143362421121,
      "loss": 0.2329,
      "num_input_tokens_seen": 20221472,
      "step": 95815
    },
    {
      "epoch": 10.541254125412541,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.016199998204619263,
      "loss": 0.2304,
      "num_input_tokens_seen": 20222528,
      "step": 95820
    },
    {
      "epoch": 10.541804180418042,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.016198562773967508,
      "loss": 0.2314,
      "num_input_tokens_seen": 20223552,
      "step": 95825
    },
    {
      "epoch": 10.542354235423542,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.01619712733226917,
      "loss": 0.234,
      "num_input_tokens_seen": 20224544,
      "step": 95830
    },
    {
      "epoch": 10.542904290429043,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01619569187953749,
      "loss": 0.2314,
      "num_input_tokens_seen": 20225600,
      "step": 95835
    },
    {
      "epoch": 10.543454345434544,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01619425641578569,
      "loss": 0.2314,
      "num_input_tokens_seen": 20226656,
      "step": 95840
    },
    {
      "epoch": 10.544004400440045,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.016192820941027,
      "loss": 0.2288,
      "num_input_tokens_seen": 20227744,
      "step": 95845
    },
    {
      "epoch": 10.544554455445544,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.016191385455274654,
      "loss": 0.2299,
      "num_input_tokens_seen": 20228800,
      "step": 95850
    },
    {
      "epoch": 10.545104510451045,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.016189949958541876,
      "loss": 0.2314,
      "num_input_tokens_seen": 20229856,
      "step": 95855
    },
    {
      "epoch": 10.545654565456546,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.016188514450841907,
      "loss": 0.2309,
      "num_input_tokens_seen": 20230880,
      "step": 95860
    },
    {
      "epoch": 10.546204620462046,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.016187078932187963,
      "loss": 0.2293,
      "num_input_tokens_seen": 20231936,
      "step": 95865
    },
    {
      "epoch": 10.546754675467547,
      "grad_norm": 0.009765625,
      "learning_rate": 0.016185643402593285,
      "loss": 0.2309,
      "num_input_tokens_seen": 20233056,
      "step": 95870
    },
    {
      "epoch": 10.547304730473048,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0161842078620711,
      "loss": 0.2314,
      "num_input_tokens_seen": 20234112,
      "step": 95875
    },
    {
      "epoch": 10.547854785478547,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01618277231063464,
      "loss": 0.2304,
      "num_input_tokens_seen": 20235232,
      "step": 95880
    },
    {
      "epoch": 10.548404840484048,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.016181336748297137,
      "loss": 0.2325,
      "num_input_tokens_seen": 20236224,
      "step": 95885
    },
    {
      "epoch": 10.54895489548955,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.016179901175071818,
      "loss": 0.2298,
      "num_input_tokens_seen": 20237312,
      "step": 95890
    },
    {
      "epoch": 10.549504950495049,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.01617846559097192,
      "loss": 0.2325,
      "num_input_tokens_seen": 20238432,
      "step": 95895
    },
    {
      "epoch": 10.55005500550055,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.016177029996010662,
      "loss": 0.2304,
      "num_input_tokens_seen": 20239488,
      "step": 95900
    },
    {
      "epoch": 10.55060506050605,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01617559439020129,
      "loss": 0.2314,
      "num_input_tokens_seen": 20240576,
      "step": 95905
    },
    {
      "epoch": 10.551155115511552,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01617415877355703,
      "loss": 0.2308,
      "num_input_tokens_seen": 20241696,
      "step": 95910
    },
    {
      "epoch": 10.551705170517051,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.016172723146091104,
      "loss": 0.2293,
      "num_input_tokens_seen": 20242752,
      "step": 95915
    },
    {
      "epoch": 10.552255225522552,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.016171287507816757,
      "loss": 0.2309,
      "num_input_tokens_seen": 20243776,
      "step": 95920
    },
    {
      "epoch": 10.552805280528053,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.016169851858747215,
      "loss": 0.2319,
      "num_input_tokens_seen": 20244800,
      "step": 95925
    },
    {
      "epoch": 10.553355335533553,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01616841619889571,
      "loss": 0.2308,
      "num_input_tokens_seen": 20245824,
      "step": 95930
    },
    {
      "epoch": 10.553905390539054,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016166980528275476,
      "loss": 0.2319,
      "num_input_tokens_seen": 20246816,
      "step": 95935
    },
    {
      "epoch": 10.554455445544555,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01616554484689974,
      "loss": 0.2314,
      "num_input_tokens_seen": 20247904,
      "step": 95940
    },
    {
      "epoch": 10.555005500550054,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.016164109154781732,
      "loss": 0.2293,
      "num_input_tokens_seen": 20248928,
      "step": 95945
    },
    {
      "epoch": 10.555555555555555,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.01616267345193469,
      "loss": 0.2319,
      "num_input_tokens_seen": 20250016,
      "step": 95950
    },
    {
      "epoch": 10.556105610561056,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01616123773837185,
      "loss": 0.2314,
      "num_input_tokens_seen": 20251040,
      "step": 95955
    },
    {
      "epoch": 10.556655665566556,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.016159802014106436,
      "loss": 0.2314,
      "num_input_tokens_seen": 20252128,
      "step": 95960
    },
    {
      "epoch": 10.557205720572057,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01615836627915168,
      "loss": 0.2293,
      "num_input_tokens_seen": 20253120,
      "step": 95965
    },
    {
      "epoch": 10.557755775577558,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.016156930533520825,
      "loss": 0.2309,
      "num_input_tokens_seen": 20254144,
      "step": 95970
    },
    {
      "epoch": 10.558305830583059,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.016155494777227088,
      "loss": 0.2303,
      "num_input_tokens_seen": 20255168,
      "step": 95975
    },
    {
      "epoch": 10.558855885588558,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.016154059010283717,
      "loss": 0.2324,
      "num_input_tokens_seen": 20256192,
      "step": 95980
    },
    {
      "epoch": 10.55940594059406,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.016152623232703937,
      "loss": 0.2309,
      "num_input_tokens_seen": 20257216,
      "step": 95985
    },
    {
      "epoch": 10.55995599559956,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.016151187444500975,
      "loss": 0.2304,
      "num_input_tokens_seen": 20258304,
      "step": 95990
    },
    {
      "epoch": 10.56050605060506,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01614975164568808,
      "loss": 0.2293,
      "num_input_tokens_seen": 20259296,
      "step": 95995
    },
    {
      "epoch": 10.561056105610561,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.01614831583627847,
      "loss": 0.2319,
      "num_input_tokens_seen": 20260320,
      "step": 96000
    },
    {
      "epoch": 10.561606160616062,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.016146880016285385,
      "loss": 0.2303,
      "num_input_tokens_seen": 20261440,
      "step": 96005
    },
    {
      "epoch": 10.562156215621561,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.01614544418572206,
      "loss": 0.2319,
      "num_input_tokens_seen": 20262464,
      "step": 96010
    },
    {
      "epoch": 10.562706270627062,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.016144008344601724,
      "loss": 0.2319,
      "num_input_tokens_seen": 20263552,
      "step": 96015
    },
    {
      "epoch": 10.563256325632564,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.016142572492937608,
      "loss": 0.2309,
      "num_input_tokens_seen": 20264608,
      "step": 96020
    },
    {
      "epoch": 10.563806380638063,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.016141136630742958,
      "loss": 0.2324,
      "num_input_tokens_seen": 20265632,
      "step": 96025
    },
    {
      "epoch": 10.564356435643564,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.01613970075803099,
      "loss": 0.2288,
      "num_input_tokens_seen": 20266688,
      "step": 96030
    },
    {
      "epoch": 10.564906490649065,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.016138264874814953,
      "loss": 0.234,
      "num_input_tokens_seen": 20267808,
      "step": 96035
    },
    {
      "epoch": 10.565456545654566,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.016136828981108072,
      "loss": 0.2319,
      "num_input_tokens_seen": 20268928,
      "step": 96040
    },
    {
      "epoch": 10.566006600660065,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.016135393076923587,
      "loss": 0.2319,
      "num_input_tokens_seen": 20270016,
      "step": 96045
    },
    {
      "epoch": 10.566556655665567,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.016133957162274726,
      "loss": 0.2314,
      "num_input_tokens_seen": 20271040,
      "step": 96050
    },
    {
      "epoch": 10.567106710671068,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.016132521237174728,
      "loss": 0.2304,
      "num_input_tokens_seen": 20272032,
      "step": 96055
    },
    {
      "epoch": 10.567656765676567,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.016131085301636823,
      "loss": 0.2283,
      "num_input_tokens_seen": 20273056,
      "step": 96060
    },
    {
      "epoch": 10.568206820682068,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.016129649355674246,
      "loss": 0.2309,
      "num_input_tokens_seen": 20274048,
      "step": 96065
    },
    {
      "epoch": 10.56875687568757,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.016128213399300234,
      "loss": 0.2298,
      "num_input_tokens_seen": 20275168,
      "step": 96070
    },
    {
      "epoch": 10.569306930693068,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.016126777432528026,
      "loss": 0.233,
      "num_input_tokens_seen": 20276224,
      "step": 96075
    },
    {
      "epoch": 10.56985698569857,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.016125341455370847,
      "loss": 0.2309,
      "num_input_tokens_seen": 20277312,
      "step": 96080
    },
    {
      "epoch": 10.57040704070407,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.016123905467841933,
      "loss": 0.2303,
      "num_input_tokens_seen": 20278400,
      "step": 96085
    },
    {
      "epoch": 10.570957095709572,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.016122469469954524,
      "loss": 0.2293,
      "num_input_tokens_seen": 20279392,
      "step": 96090
    },
    {
      "epoch": 10.571507150715071,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01612103346172185,
      "loss": 0.2303,
      "num_input_tokens_seen": 20280416,
      "step": 96095
    },
    {
      "epoch": 10.572057205720572,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.01611959744315716,
      "loss": 0.2304,
      "num_input_tokens_seen": 20281504,
      "step": 96100
    },
    {
      "epoch": 10.572607260726073,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.016118161414273663,
      "loss": 0.2319,
      "num_input_tokens_seen": 20282560,
      "step": 96105
    },
    {
      "epoch": 10.573157315731573,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.016116725375084615,
      "loss": 0.2288,
      "num_input_tokens_seen": 20283648,
      "step": 96110
    },
    {
      "epoch": 10.573707370737074,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.016115289325603244,
      "loss": 0.2314,
      "num_input_tokens_seen": 20284704,
      "step": 96115
    },
    {
      "epoch": 10.574257425742575,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.016113853265842793,
      "loss": 0.2319,
      "num_input_tokens_seen": 20285792,
      "step": 96120
    },
    {
      "epoch": 10.574807480748074,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.016112417195816485,
      "loss": 0.2314,
      "num_input_tokens_seen": 20286880,
      "step": 96125
    },
    {
      "epoch": 10.575357535753575,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01611098111553756,
      "loss": 0.2319,
      "num_input_tokens_seen": 20287936,
      "step": 96130
    },
    {
      "epoch": 10.575907590759076,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.016109545025019253,
      "loss": 0.2324,
      "num_input_tokens_seen": 20289024,
      "step": 96135
    },
    {
      "epoch": 10.576457645764577,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.016108108924274805,
      "loss": 0.2324,
      "num_input_tokens_seen": 20290016,
      "step": 96140
    },
    {
      "epoch": 10.577007700770077,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.01610667281331745,
      "loss": 0.2304,
      "num_input_tokens_seen": 20291104,
      "step": 96145
    },
    {
      "epoch": 10.577557755775578,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.016105236692160425,
      "loss": 0.2304,
      "num_input_tokens_seen": 20292160,
      "step": 96150
    },
    {
      "epoch": 10.578107810781079,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.016103800560816964,
      "loss": 0.2314,
      "num_input_tokens_seen": 20293216,
      "step": 96155
    },
    {
      "epoch": 10.578657865786578,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.016102364419300297,
      "loss": 0.2324,
      "num_input_tokens_seen": 20294240,
      "step": 96160
    },
    {
      "epoch": 10.57920792079208,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.016100928267623667,
      "loss": 0.2298,
      "num_input_tokens_seen": 20295296,
      "step": 96165
    },
    {
      "epoch": 10.57975797579758,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.016099492105800313,
      "loss": 0.2314,
      "num_input_tokens_seen": 20296320,
      "step": 96170
    },
    {
      "epoch": 10.58030803080308,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.016098055933843464,
      "loss": 0.2314,
      "num_input_tokens_seen": 20297376,
      "step": 96175
    },
    {
      "epoch": 10.58085808580858,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.016096619751766364,
      "loss": 0.2324,
      "num_input_tokens_seen": 20298432,
      "step": 96180
    },
    {
      "epoch": 10.581408140814082,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.016095183559582248,
      "loss": 0.2324,
      "num_input_tokens_seen": 20299488,
      "step": 96185
    },
    {
      "epoch": 10.581958195819581,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.016093747357304344,
      "loss": 0.233,
      "num_input_tokens_seen": 20300576,
      "step": 96190
    },
    {
      "epoch": 10.582508250825082,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0160923111449459,
      "loss": 0.2309,
      "num_input_tokens_seen": 20301632,
      "step": 96195
    },
    {
      "epoch": 10.583058305830583,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.016090874922520145,
      "loss": 0.2309,
      "num_input_tokens_seen": 20302656,
      "step": 96200
    },
    {
      "epoch": 10.583608360836084,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.01608943869004032,
      "loss": 0.2309,
      "num_input_tokens_seen": 20303712,
      "step": 96205
    },
    {
      "epoch": 10.584158415841584,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.016088002447519664,
      "loss": 0.2335,
      "num_input_tokens_seen": 20304704,
      "step": 96210
    },
    {
      "epoch": 10.584708470847085,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.016086566194971414,
      "loss": 0.2303,
      "num_input_tokens_seen": 20305728,
      "step": 96215
    },
    {
      "epoch": 10.585258525852586,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0160851299324088,
      "loss": 0.2319,
      "num_input_tokens_seen": 20306720,
      "step": 96220
    },
    {
      "epoch": 10.585808580858085,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.016083693659845064,
      "loss": 0.2309,
      "num_input_tokens_seen": 20307808,
      "step": 96225
    },
    {
      "epoch": 10.586358635863586,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.016082257377293446,
      "loss": 0.2314,
      "num_input_tokens_seen": 20308896,
      "step": 96230
    },
    {
      "epoch": 10.586908690869087,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.016080821084767175,
      "loss": 0.2294,
      "num_input_tokens_seen": 20309952,
      "step": 96235
    },
    {
      "epoch": 10.587458745874587,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.016079384782279504,
      "loss": 0.2309,
      "num_input_tokens_seen": 20310944,
      "step": 96240
    },
    {
      "epoch": 10.588008800880088,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.016077948469843655,
      "loss": 0.2314,
      "num_input_tokens_seen": 20312000,
      "step": 96245
    },
    {
      "epoch": 10.588558855885589,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016076512147472872,
      "loss": 0.233,
      "num_input_tokens_seen": 20313056,
      "step": 96250
    },
    {
      "epoch": 10.589108910891088,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.016075075815180393,
      "loss": 0.2298,
      "num_input_tokens_seen": 20314208,
      "step": 96255
    },
    {
      "epoch": 10.58965896589659,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.016073639472979462,
      "loss": 0.2308,
      "num_input_tokens_seen": 20315296,
      "step": 96260
    },
    {
      "epoch": 10.59020902090209,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.016072203120883307,
      "loss": 0.2319,
      "num_input_tokens_seen": 20316320,
      "step": 96265
    },
    {
      "epoch": 10.590759075907592,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01607076675890517,
      "loss": 0.2335,
      "num_input_tokens_seen": 20317344,
      "step": 96270
    },
    {
      "epoch": 10.591309130913091,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.016069330387058295,
      "loss": 0.2303,
      "num_input_tokens_seen": 20318336,
      "step": 96275
    },
    {
      "epoch": 10.591859185918592,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.016067894005355902,
      "loss": 0.2288,
      "num_input_tokens_seen": 20319328,
      "step": 96280
    },
    {
      "epoch": 10.592409240924093,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.016066457613811255,
      "loss": 0.2309,
      "num_input_tokens_seen": 20320384,
      "step": 96285
    },
    {
      "epoch": 10.592959295929592,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.016065021212437577,
      "loss": 0.2319,
      "num_input_tokens_seen": 20321504,
      "step": 96290
    },
    {
      "epoch": 10.593509350935093,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.016063584801248105,
      "loss": 0.233,
      "num_input_tokens_seen": 20322592,
      "step": 96295
    },
    {
      "epoch": 10.594059405940595,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.016062148380256085,
      "loss": 0.2319,
      "num_input_tokens_seen": 20323648,
      "step": 96300
    },
    {
      "epoch": 10.594609460946094,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.016060711949474754,
      "loss": 0.2309,
      "num_input_tokens_seen": 20324704,
      "step": 96305
    },
    {
      "epoch": 10.595159515951595,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.01605927550891735,
      "loss": 0.2314,
      "num_input_tokens_seen": 20325760,
      "step": 96310
    },
    {
      "epoch": 10.595709570957096,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.01605783905859711,
      "loss": 0.2309,
      "num_input_tokens_seen": 20326752,
      "step": 96315
    },
    {
      "epoch": 10.596259625962595,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01605640259852728,
      "loss": 0.2319,
      "num_input_tokens_seen": 20327840,
      "step": 96320
    },
    {
      "epoch": 10.596809680968097,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.016054966128721086,
      "loss": 0.2304,
      "num_input_tokens_seen": 20328864,
      "step": 96325
    },
    {
      "epoch": 10.597359735973598,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.016053529649191777,
      "loss": 0.2309,
      "num_input_tokens_seen": 20329920,
      "step": 96330
    },
    {
      "epoch": 10.597909790979099,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.016052093159952596,
      "loss": 0.2309,
      "num_input_tokens_seen": 20330976,
      "step": 96335
    },
    {
      "epoch": 10.598459845984598,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.016050656661016776,
      "loss": 0.2303,
      "num_input_tokens_seen": 20332064,
      "step": 96340
    },
    {
      "epoch": 10.599009900990099,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.016049220152397556,
      "loss": 0.2314,
      "num_input_tokens_seen": 20333120,
      "step": 96345
    },
    {
      "epoch": 10.5995599559956,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.01604778363410818,
      "loss": 0.2309,
      "num_input_tokens_seen": 20334176,
      "step": 96350
    },
    {
      "epoch": 10.6001100110011,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.016046347106161877,
      "loss": 0.2314,
      "num_input_tokens_seen": 20335200,
      "step": 96355
    },
    {
      "epoch": 10.6006600660066,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.016044910568571904,
      "loss": 0.233,
      "num_input_tokens_seen": 20336256,
      "step": 96360
    },
    {
      "epoch": 10.601210121012102,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.016043474021351485,
      "loss": 0.2325,
      "num_input_tokens_seen": 20337312,
      "step": 96365
    },
    {
      "epoch": 10.601760176017601,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.016042037464513874,
      "loss": 0.2324,
      "num_input_tokens_seen": 20338400,
      "step": 96370
    },
    {
      "epoch": 10.602310231023102,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.0160406008980723,
      "loss": 0.2319,
      "num_input_tokens_seen": 20339552,
      "step": 96375
    },
    {
      "epoch": 10.602860286028603,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.016039164322040004,
      "loss": 0.2319,
      "num_input_tokens_seen": 20340608,
      "step": 96380
    },
    {
      "epoch": 10.603410341034103,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01603772773643023,
      "loss": 0.2324,
      "num_input_tokens_seen": 20341664,
      "step": 96385
    },
    {
      "epoch": 10.603960396039604,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.016036291141256223,
      "loss": 0.2309,
      "num_input_tokens_seen": 20342720,
      "step": 96390
    },
    {
      "epoch": 10.604510451045105,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01603485453653121,
      "loss": 0.2325,
      "num_input_tokens_seen": 20343712,
      "step": 96395
    },
    {
      "epoch": 10.605060506050606,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.016033417922268443,
      "loss": 0.2324,
      "num_input_tokens_seen": 20344768,
      "step": 96400
    },
    {
      "epoch": 10.605610561056105,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01603198129848116,
      "loss": 0.2319,
      "num_input_tokens_seen": 20345824,
      "step": 96405
    },
    {
      "epoch": 10.606160616061606,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.016030544665182597,
      "loss": 0.2293,
      "num_input_tokens_seen": 20346912,
      "step": 96410
    },
    {
      "epoch": 10.606710671067107,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.016029108022386003,
      "loss": 0.2325,
      "num_input_tokens_seen": 20348000,
      "step": 96415
    },
    {
      "epoch": 10.607260726072607,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.016027671370104614,
      "loss": 0.2278,
      "num_input_tokens_seen": 20349120,
      "step": 96420
    },
    {
      "epoch": 10.607810781078108,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.016026234708351662,
      "loss": 0.2304,
      "num_input_tokens_seen": 20350272,
      "step": 96425
    },
    {
      "epoch": 10.608360836083609,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.01602479803714041,
      "loss": 0.2329,
      "num_input_tokens_seen": 20351296,
      "step": 96430
    },
    {
      "epoch": 10.608910891089108,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.01602336135648408,
      "loss": 0.2309,
      "num_input_tokens_seen": 20352416,
      "step": 96435
    },
    {
      "epoch": 10.60946094609461,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.016021924666395918,
      "loss": 0.2319,
      "num_input_tokens_seen": 20353536,
      "step": 96440
    },
    {
      "epoch": 10.61001100110011,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01602048796688917,
      "loss": 0.2309,
      "num_input_tokens_seen": 20354656,
      "step": 96445
    },
    {
      "epoch": 10.61056105610561,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016019051257977073,
      "loss": 0.2314,
      "num_input_tokens_seen": 20355744,
      "step": 96450
    },
    {
      "epoch": 10.61111111111111,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.016017614539672872,
      "loss": 0.2319,
      "num_input_tokens_seen": 20356800,
      "step": 96455
    },
    {
      "epoch": 10.611661166116612,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.0160161778119898,
      "loss": 0.2298,
      "num_input_tokens_seen": 20357824,
      "step": 96460
    },
    {
      "epoch": 10.612211221122113,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.01601474107494111,
      "loss": 0.2304,
      "num_input_tokens_seen": 20358816,
      "step": 96465
    },
    {
      "epoch": 10.612761276127612,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.016013304328540034,
      "loss": 0.2319,
      "num_input_tokens_seen": 20359872,
      "step": 96470
    },
    {
      "epoch": 10.613311331133113,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.01601186757279982,
      "loss": 0.2309,
      "num_input_tokens_seen": 20360928,
      "step": 96475
    },
    {
      "epoch": 10.613861386138614,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.01601043080773371,
      "loss": 0.234,
      "num_input_tokens_seen": 20362080,
      "step": 96480
    },
    {
      "epoch": 10.614411441144114,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.016008994033354942,
      "loss": 0.2308,
      "num_input_tokens_seen": 20363104,
      "step": 96485
    },
    {
      "epoch": 10.614961496149615,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01600755724967676,
      "loss": 0.2324,
      "num_input_tokens_seen": 20364128,
      "step": 96490
    },
    {
      "epoch": 10.615511551155116,
      "grad_norm": 0.00074005126953125,
      "learning_rate": 0.01600612045671241,
      "loss": 0.2314,
      "num_input_tokens_seen": 20365152,
      "step": 96495
    },
    {
      "epoch": 10.616061606160617,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.016004683654475126,
      "loss": 0.2293,
      "num_input_tokens_seen": 20366208,
      "step": 96500
    },
    {
      "epoch": 10.616611661166116,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.016003246842978158,
      "loss": 0.2314,
      "num_input_tokens_seen": 20367264,
      "step": 96505
    },
    {
      "epoch": 10.617161716171617,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.016001810022234744,
      "loss": 0.2324,
      "num_input_tokens_seen": 20368320,
      "step": 96510
    },
    {
      "epoch": 10.617711771177119,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.016000373192258124,
      "loss": 0.2304,
      "num_input_tokens_seen": 20369408,
      "step": 96515
    },
    {
      "epoch": 10.618261826182618,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01599893635306155,
      "loss": 0.2329,
      "num_input_tokens_seen": 20370496,
      "step": 96520
    },
    {
      "epoch": 10.618811881188119,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.015997499504658258,
      "loss": 0.2314,
      "num_input_tokens_seen": 20371584,
      "step": 96525
    },
    {
      "epoch": 10.61936193619362,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.015996062647061493,
      "loss": 0.2319,
      "num_input_tokens_seen": 20372640,
      "step": 96530
    },
    {
      "epoch": 10.61991199119912,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.015994625780284495,
      "loss": 0.2314,
      "num_input_tokens_seen": 20373696,
      "step": 96535
    },
    {
      "epoch": 10.62046204620462,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.015993188904340504,
      "loss": 0.2298,
      "num_input_tokens_seen": 20374656,
      "step": 96540
    },
    {
      "epoch": 10.621012101210122,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.015991752019242772,
      "loss": 0.2298,
      "num_input_tokens_seen": 20375712,
      "step": 96545
    },
    {
      "epoch": 10.62156215621562,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.01599031512500454,
      "loss": 0.2288,
      "num_input_tokens_seen": 20376832,
      "step": 96550
    },
    {
      "epoch": 10.622112211221122,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.015988878221639042,
      "loss": 0.2314,
      "num_input_tokens_seen": 20377920,
      "step": 96555
    },
    {
      "epoch": 10.622662266226623,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.015987441309159533,
      "loss": 0.2309,
      "num_input_tokens_seen": 20378976,
      "step": 96560
    },
    {
      "epoch": 10.623212321232124,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.015986004387579248,
      "loss": 0.2303,
      "num_input_tokens_seen": 20380064,
      "step": 96565
    },
    {
      "epoch": 10.623762376237623,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.01598456745691143,
      "loss": 0.2308,
      "num_input_tokens_seen": 20381120,
      "step": 96570
    },
    {
      "epoch": 10.624312431243125,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.015983130517169337,
      "loss": 0.2298,
      "num_input_tokens_seen": 20382208,
      "step": 96575
    },
    {
      "epoch": 10.624862486248626,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.015981693568366196,
      "loss": 0.2308,
      "num_input_tokens_seen": 20383296,
      "step": 96580
    },
    {
      "epoch": 10.625412541254125,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.015980256610515255,
      "loss": 0.234,
      "num_input_tokens_seen": 20384352,
      "step": 96585
    },
    {
      "epoch": 10.625962596259626,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.015978819643629762,
      "loss": 0.2319,
      "num_input_tokens_seen": 20385472,
      "step": 96590
    },
    {
      "epoch": 10.626512651265127,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.015977382667722962,
      "loss": 0.2324,
      "num_input_tokens_seen": 20386496,
      "step": 96595
    },
    {
      "epoch": 10.627062706270626,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.015975945682808083,
      "loss": 0.2319,
      "num_input_tokens_seen": 20387552,
      "step": 96600
    },
    {
      "epoch": 10.627612761276128,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.015974508688898388,
      "loss": 0.2298,
      "num_input_tokens_seen": 20388608,
      "step": 96605
    },
    {
      "epoch": 10.628162816281629,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015973071686007118,
      "loss": 0.2319,
      "num_input_tokens_seen": 20389664,
      "step": 96610
    },
    {
      "epoch": 10.628712871287128,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015971634674147507,
      "loss": 0.2335,
      "num_input_tokens_seen": 20390688,
      "step": 96615
    },
    {
      "epoch": 10.629262926292629,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.01597019765333281,
      "loss": 0.2314,
      "num_input_tokens_seen": 20391776,
      "step": 96620
    },
    {
      "epoch": 10.62981298129813,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.015968760623576262,
      "loss": 0.2314,
      "num_input_tokens_seen": 20392832,
      "step": 96625
    },
    {
      "epoch": 10.630363036303631,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.015967323584891113,
      "loss": 0.2324,
      "num_input_tokens_seen": 20393824,
      "step": 96630
    },
    {
      "epoch": 10.63091309130913,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01596588653729061,
      "loss": 0.2314,
      "num_input_tokens_seen": 20394944,
      "step": 96635
    },
    {
      "epoch": 10.631463146314632,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01596444948078799,
      "loss": 0.233,
      "num_input_tokens_seen": 20395936,
      "step": 96640
    },
    {
      "epoch": 10.632013201320133,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.015963012415396508,
      "loss": 0.2329,
      "num_input_tokens_seen": 20396928,
      "step": 96645
    },
    {
      "epoch": 10.632563256325632,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015961575341129398,
      "loss": 0.2324,
      "num_input_tokens_seen": 20398016,
      "step": 96650
    },
    {
      "epoch": 10.633113311331133,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.01596013825799991,
      "loss": 0.2314,
      "num_input_tokens_seen": 20399072,
      "step": 96655
    },
    {
      "epoch": 10.633663366336634,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01595870116602129,
      "loss": 0.2303,
      "num_input_tokens_seen": 20400192,
      "step": 96660
    },
    {
      "epoch": 10.634213421342134,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.01595726406520678,
      "loss": 0.2324,
      "num_input_tokens_seen": 20401216,
      "step": 96665
    },
    {
      "epoch": 10.634763476347635,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.015955826955569626,
      "loss": 0.234,
      "num_input_tokens_seen": 20402304,
      "step": 96670
    },
    {
      "epoch": 10.635313531353136,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.015954389837123075,
      "loss": 0.2298,
      "num_input_tokens_seen": 20403360,
      "step": 96675
    },
    {
      "epoch": 10.635863586358635,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.01595295270988037,
      "loss": 0.2298,
      "num_input_tokens_seen": 20404384,
      "step": 96680
    },
    {
      "epoch": 10.636413641364136,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.015951515573854755,
      "loss": 0.2324,
      "num_input_tokens_seen": 20405440,
      "step": 96685
    },
    {
      "epoch": 10.636963696369637,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.015950078429059482,
      "loss": 0.2314,
      "num_input_tokens_seen": 20406496,
      "step": 96690
    },
    {
      "epoch": 10.637513751375138,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01594864127550779,
      "loss": 0.2308,
      "num_input_tokens_seen": 20407520,
      "step": 96695
    },
    {
      "epoch": 10.638063806380638,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.015947204113212925,
      "loss": 0.2314,
      "num_input_tokens_seen": 20408576,
      "step": 96700
    },
    {
      "epoch": 10.638613861386139,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.015945766942188137,
      "loss": 0.2298,
      "num_input_tokens_seen": 20409632,
      "step": 96705
    },
    {
      "epoch": 10.63916391639164,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.015944329762446665,
      "loss": 0.2288,
      "num_input_tokens_seen": 20410688,
      "step": 96710
    },
    {
      "epoch": 10.63971397139714,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.015942892574001763,
      "loss": 0.2308,
      "num_input_tokens_seen": 20411712,
      "step": 96715
    },
    {
      "epoch": 10.64026402640264,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.01594145537686667,
      "loss": 0.2319,
      "num_input_tokens_seen": 20412704,
      "step": 96720
    },
    {
      "epoch": 10.640814081408141,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.015940018171054636,
      "loss": 0.2309,
      "num_input_tokens_seen": 20413760,
      "step": 96725
    },
    {
      "epoch": 10.64136413641364,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015938580956578904,
      "loss": 0.2298,
      "num_input_tokens_seen": 20414816,
      "step": 96730
    },
    {
      "epoch": 10.641914191419142,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.015937143733452723,
      "loss": 0.2304,
      "num_input_tokens_seen": 20415808,
      "step": 96735
    },
    {
      "epoch": 10.642464246424643,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.015935706501689337,
      "loss": 0.2319,
      "num_input_tokens_seen": 20416832,
      "step": 96740
    },
    {
      "epoch": 10.643014301430142,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.015934269261301995,
      "loss": 0.2309,
      "num_input_tokens_seen": 20417888,
      "step": 96745
    },
    {
      "epoch": 10.643564356435643,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01593283201230394,
      "loss": 0.2288,
      "num_input_tokens_seen": 20418976,
      "step": 96750
    },
    {
      "epoch": 10.644114411441144,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.01593139475470842,
      "loss": 0.2314,
      "num_input_tokens_seen": 20419968,
      "step": 96755
    },
    {
      "epoch": 10.644664466446645,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.015929957488528677,
      "loss": 0.2329,
      "num_input_tokens_seen": 20421024,
      "step": 96760
    },
    {
      "epoch": 10.645214521452145,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.015928520213777973,
      "loss": 0.2308,
      "num_input_tokens_seen": 20422144,
      "step": 96765
    },
    {
      "epoch": 10.645764576457646,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.015927082930469537,
      "loss": 0.2288,
      "num_input_tokens_seen": 20423264,
      "step": 96770
    },
    {
      "epoch": 10.646314631463147,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01592564563861662,
      "loss": 0.2293,
      "num_input_tokens_seen": 20424288,
      "step": 96775
    },
    {
      "epoch": 10.646864686468646,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.015924208338232472,
      "loss": 0.2314,
      "num_input_tokens_seen": 20425280,
      "step": 96780
    },
    {
      "epoch": 10.647414741474147,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015922771029330342,
      "loss": 0.234,
      "num_input_tokens_seen": 20426368,
      "step": 96785
    },
    {
      "epoch": 10.647964796479648,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.015921333711923474,
      "loss": 0.2319,
      "num_input_tokens_seen": 20427360,
      "step": 96790
    },
    {
      "epoch": 10.648514851485148,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.015919896386025117,
      "loss": 0.233,
      "num_input_tokens_seen": 20428416,
      "step": 96795
    },
    {
      "epoch": 10.649064906490649,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.015918459051648515,
      "loss": 0.2324,
      "num_input_tokens_seen": 20429440,
      "step": 96800
    },
    {
      "epoch": 10.64961496149615,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.015917021708806912,
      "loss": 0.2329,
      "num_input_tokens_seen": 20430496,
      "step": 96805
    },
    {
      "epoch": 10.65016501650165,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01591558435751357,
      "loss": 0.2304,
      "num_input_tokens_seen": 20431552,
      "step": 96810
    },
    {
      "epoch": 10.65071507150715,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.015914146997781715,
      "loss": 0.2303,
      "num_input_tokens_seen": 20432640,
      "step": 96815
    },
    {
      "epoch": 10.651265126512651,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.015912709629624613,
      "loss": 0.2309,
      "num_input_tokens_seen": 20433728,
      "step": 96820
    },
    {
      "epoch": 10.651815181518153,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.0159112722530555,
      "loss": 0.2303,
      "num_input_tokens_seen": 20434784,
      "step": 96825
    },
    {
      "epoch": 10.652365236523652,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.015909834868087634,
      "loss": 0.2293,
      "num_input_tokens_seen": 20435776,
      "step": 96830
    },
    {
      "epoch": 10.652915291529153,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.015908397474734253,
      "loss": 0.2324,
      "num_input_tokens_seen": 20436800,
      "step": 96835
    },
    {
      "epoch": 10.653465346534654,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01590696007300861,
      "loss": 0.2324,
      "num_input_tokens_seen": 20437952,
      "step": 96840
    },
    {
      "epoch": 10.654015401540153,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.015905522662923943,
      "loss": 0.2319,
      "num_input_tokens_seen": 20438976,
      "step": 96845
    },
    {
      "epoch": 10.654565456545654,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.015904085244493518,
      "loss": 0.2329,
      "num_input_tokens_seen": 20440032,
      "step": 96850
    },
    {
      "epoch": 10.655115511551156,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01590264781773057,
      "loss": 0.2293,
      "num_input_tokens_seen": 20441056,
      "step": 96855
    },
    {
      "epoch": 10.655665566556655,
      "grad_norm": 0.0030364990234375,
      "learning_rate": 0.01590121038264835,
      "loss": 0.2293,
      "num_input_tokens_seen": 20442144,
      "step": 96860
    },
    {
      "epoch": 10.656215621562156,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015899772939260107,
      "loss": 0.2293,
      "num_input_tokens_seen": 20443168,
      "step": 96865
    },
    {
      "epoch": 10.656765676567657,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.015898335487579088,
      "loss": 0.2314,
      "num_input_tokens_seen": 20444288,
      "step": 96870
    },
    {
      "epoch": 10.657315731573158,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.01589689802761854,
      "loss": 0.2308,
      "num_input_tokens_seen": 20445344,
      "step": 96875
    },
    {
      "epoch": 10.657865786578657,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.015895460559391717,
      "loss": 0.2308,
      "num_input_tokens_seen": 20446464,
      "step": 96880
    },
    {
      "epoch": 10.658415841584159,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.01589402308291186,
      "loss": 0.2298,
      "num_input_tokens_seen": 20447488,
      "step": 96885
    },
    {
      "epoch": 10.65896589658966,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015892585598192222,
      "loss": 0.234,
      "num_input_tokens_seen": 20448544,
      "step": 96890
    },
    {
      "epoch": 10.659515951595159,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015891148105246052,
      "loss": 0.2324,
      "num_input_tokens_seen": 20449568,
      "step": 96895
    },
    {
      "epoch": 10.66006600660066,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.015889710604086598,
      "loss": 0.2298,
      "num_input_tokens_seen": 20450688,
      "step": 96900
    },
    {
      "epoch": 10.660616061606161,
      "grad_norm": 0.000843048095703125,
      "learning_rate": 0.015888273094727106,
      "loss": 0.233,
      "num_input_tokens_seen": 20451776,
      "step": 96905
    },
    {
      "epoch": 10.66116611661166,
      "grad_norm": 0.000621795654296875,
      "learning_rate": 0.015886835577180833,
      "loss": 0.2298,
      "num_input_tokens_seen": 20452800,
      "step": 96910
    },
    {
      "epoch": 10.661716171617162,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.015885398051461018,
      "loss": 0.2309,
      "num_input_tokens_seen": 20453824,
      "step": 96915
    },
    {
      "epoch": 10.662266226622663,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.015883960517580908,
      "loss": 0.2314,
      "num_input_tokens_seen": 20454912,
      "step": 96920
    },
    {
      "epoch": 10.662816281628164,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015882522975553766,
      "loss": 0.2325,
      "num_input_tokens_seen": 20455968,
      "step": 96925
    },
    {
      "epoch": 10.663366336633663,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.01588108542539283,
      "loss": 0.2303,
      "num_input_tokens_seen": 20456992,
      "step": 96930
    },
    {
      "epoch": 10.663916391639164,
      "grad_norm": 0.000713348388671875,
      "learning_rate": 0.015879647867111355,
      "loss": 0.2288,
      "num_input_tokens_seen": 20458048,
      "step": 96935
    },
    {
      "epoch": 10.664466446644665,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.015878210300722586,
      "loss": 0.2329,
      "num_input_tokens_seen": 20459040,
      "step": 96940
    },
    {
      "epoch": 10.665016501650165,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.015876772726239774,
      "loss": 0.2309,
      "num_input_tokens_seen": 20460160,
      "step": 96945
    },
    {
      "epoch": 10.665566556655666,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.01587533514367617,
      "loss": 0.2319,
      "num_input_tokens_seen": 20461184,
      "step": 96950
    },
    {
      "epoch": 10.666116611661167,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.01587389755304502,
      "loss": 0.2319,
      "num_input_tokens_seen": 20462272,
      "step": 96955
    },
    {
      "epoch": 10.666666666666666,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.015872459954359576,
      "loss": 0.2288,
      "num_input_tokens_seen": 20463328,
      "step": 96960
    },
    {
      "epoch": 10.667216721672167,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.015871022347633088,
      "loss": 0.2319,
      "num_input_tokens_seen": 20464352,
      "step": 96965
    },
    {
      "epoch": 10.667766776677668,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.015869584732878805,
      "loss": 0.2319,
      "num_input_tokens_seen": 20465440,
      "step": 96970
    },
    {
      "epoch": 10.668316831683168,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.015868147110109976,
      "loss": 0.2319,
      "num_input_tokens_seen": 20466432,
      "step": 96975
    },
    {
      "epoch": 10.668866886688669,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.015866709479339852,
      "loss": 0.2314,
      "num_input_tokens_seen": 20467520,
      "step": 96980
    },
    {
      "epoch": 10.66941694169417,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01586527184058168,
      "loss": 0.2304,
      "num_input_tokens_seen": 20468608,
      "step": 96985
    },
    {
      "epoch": 10.66996699669967,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015863834193848715,
      "loss": 0.2303,
      "num_input_tokens_seen": 20469632,
      "step": 96990
    },
    {
      "epoch": 10.67051705170517,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.015862396539154205,
      "loss": 0.2308,
      "num_input_tokens_seen": 20470656,
      "step": 96995
    },
    {
      "epoch": 10.671067106710671,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.015860958876511403,
      "loss": 0.2308,
      "num_input_tokens_seen": 20471680,
      "step": 97000
    },
    {
      "epoch": 10.671617161716172,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.01585952120593355,
      "loss": 0.2288,
      "num_input_tokens_seen": 20472736,
      "step": 97005
    },
    {
      "epoch": 10.672167216721672,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015858083527433908,
      "loss": 0.2329,
      "num_input_tokens_seen": 20473824,
      "step": 97010
    },
    {
      "epoch": 10.672717271727173,
      "grad_norm": 0.002838134765625,
      "learning_rate": 0.015856645841025717,
      "loss": 0.2309,
      "num_input_tokens_seen": 20474880,
      "step": 97015
    },
    {
      "epoch": 10.673267326732674,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.015855208146722234,
      "loss": 0.2304,
      "num_input_tokens_seen": 20475936,
      "step": 97020
    },
    {
      "epoch": 10.673817381738173,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01585377044453671,
      "loss": 0.2324,
      "num_input_tokens_seen": 20476992,
      "step": 97025
    },
    {
      "epoch": 10.674367436743674,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.015852332734482395,
      "loss": 0.2303,
      "num_input_tokens_seen": 20478048,
      "step": 97030
    },
    {
      "epoch": 10.674917491749175,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015850895016572532,
      "loss": 0.2319,
      "num_input_tokens_seen": 20479104,
      "step": 97035
    },
    {
      "epoch": 10.675467546754675,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.015849457290820382,
      "loss": 0.2298,
      "num_input_tokens_seen": 20480192,
      "step": 97040
    },
    {
      "epoch": 10.676017601760176,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01584801955723919,
      "loss": 0.2303,
      "num_input_tokens_seen": 20481280,
      "step": 97045
    },
    {
      "epoch": 10.676567656765677,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.01584658181584221,
      "loss": 0.2293,
      "num_input_tokens_seen": 20482336,
      "step": 97050
    },
    {
      "epoch": 10.677117711771178,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.015845144066642693,
      "loss": 0.2308,
      "num_input_tokens_seen": 20483424,
      "step": 97055
    },
    {
      "epoch": 10.677667766776677,
      "grad_norm": 0.0028076171875,
      "learning_rate": 0.01584370630965389,
      "loss": 0.2329,
      "num_input_tokens_seen": 20484480,
      "step": 97060
    },
    {
      "epoch": 10.678217821782178,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.015842268544889043,
      "loss": 0.2324,
      "num_input_tokens_seen": 20485536,
      "step": 97065
    },
    {
      "epoch": 10.67876787678768,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.015840830772361418,
      "loss": 0.2304,
      "num_input_tokens_seen": 20486624,
      "step": 97070
    },
    {
      "epoch": 10.679317931793179,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.01583939299208426,
      "loss": 0.2319,
      "num_input_tokens_seen": 20487680,
      "step": 97075
    },
    {
      "epoch": 10.67986798679868,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01583795520407081,
      "loss": 0.2308,
      "num_input_tokens_seen": 20488704,
      "step": 97080
    },
    {
      "epoch": 10.680418041804181,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.015836517408334336,
      "loss": 0.2314,
      "num_input_tokens_seen": 20489792,
      "step": 97085
    },
    {
      "epoch": 10.68096809680968,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.015835079604888086,
      "loss": 0.2324,
      "num_input_tokens_seen": 20490848,
      "step": 97090
    },
    {
      "epoch": 10.681518151815181,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.0158336417937453,
      "loss": 0.2303,
      "num_input_tokens_seen": 20491936,
      "step": 97095
    },
    {
      "epoch": 10.682068206820682,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.015832203974919242,
      "loss": 0.2309,
      "num_input_tokens_seen": 20493056,
      "step": 97100
    },
    {
      "epoch": 10.682618261826182,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.015830766148423162,
      "loss": 0.2309,
      "num_input_tokens_seen": 20494112,
      "step": 97105
    },
    {
      "epoch": 10.683168316831683,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.0158293283142703,
      "loss": 0.233,
      "num_input_tokens_seen": 20495136,
      "step": 97110
    },
    {
      "epoch": 10.683718371837184,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.015827890472473925,
      "loss": 0.2304,
      "num_input_tokens_seen": 20496224,
      "step": 97115
    },
    {
      "epoch": 10.684268426842685,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.01582645262304728,
      "loss": 0.2324,
      "num_input_tokens_seen": 20497376,
      "step": 97120
    },
    {
      "epoch": 10.684818481848184,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.01582501476600362,
      "loss": 0.2314,
      "num_input_tokens_seen": 20498400,
      "step": 97125
    },
    {
      "epoch": 10.685368536853685,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.01582357690135619,
      "loss": 0.2293,
      "num_input_tokens_seen": 20499488,
      "step": 97130
    },
    {
      "epoch": 10.685918591859187,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01582213902911825,
      "loss": 0.2293,
      "num_input_tokens_seen": 20500608,
      "step": 97135
    },
    {
      "epoch": 10.686468646864686,
      "grad_norm": 0.000782012939453125,
      "learning_rate": 0.015820701149303038,
      "loss": 0.2304,
      "num_input_tokens_seen": 20501600,
      "step": 97140
    },
    {
      "epoch": 10.687018701870187,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.015819263261923833,
      "loss": 0.2309,
      "num_input_tokens_seen": 20502656,
      "step": 97145
    },
    {
      "epoch": 10.687568756875688,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.015817825366993863,
      "loss": 0.2298,
      "num_input_tokens_seen": 20503680,
      "step": 97150
    },
    {
      "epoch": 10.688118811881187,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.01581638746452639,
      "loss": 0.2309,
      "num_input_tokens_seen": 20504768,
      "step": 97155
    },
    {
      "epoch": 10.688668866886688,
      "grad_norm": 0.000690460205078125,
      "learning_rate": 0.015814949554534666,
      "loss": 0.2314,
      "num_input_tokens_seen": 20505824,
      "step": 97160
    },
    {
      "epoch": 10.68921892189219,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015813511637031943,
      "loss": 0.2319,
      "num_input_tokens_seen": 20506880,
      "step": 97165
    },
    {
      "epoch": 10.689768976897689,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.015812073712031475,
      "loss": 0.2288,
      "num_input_tokens_seen": 20508000,
      "step": 97170
    },
    {
      "epoch": 10.69031903190319,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.015810635779546518,
      "loss": 0.2319,
      "num_input_tokens_seen": 20508992,
      "step": 97175
    },
    {
      "epoch": 10.690869086908691,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.015809197839590308,
      "loss": 0.2351,
      "num_input_tokens_seen": 20510144,
      "step": 97180
    },
    {
      "epoch": 10.691419141914192,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.015807759892176115,
      "loss": 0.2304,
      "num_input_tokens_seen": 20511168,
      "step": 97185
    },
    {
      "epoch": 10.691969196919691,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01580632193731719,
      "loss": 0.2299,
      "num_input_tokens_seen": 20512192,
      "step": 97190
    },
    {
      "epoch": 10.692519251925193,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.01580488397502678,
      "loss": 0.2288,
      "num_input_tokens_seen": 20513184,
      "step": 97195
    },
    {
      "epoch": 10.693069306930694,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01580344600531814,
      "loss": 0.2299,
      "num_input_tokens_seen": 20514240,
      "step": 97200
    },
    {
      "epoch": 10.693619361936193,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.01580200802820453,
      "loss": 0.2314,
      "num_input_tokens_seen": 20515360,
      "step": 97205
    },
    {
      "epoch": 10.694169416941694,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.015800570043699187,
      "loss": 0.2314,
      "num_input_tokens_seen": 20516512,
      "step": 97210
    },
    {
      "epoch": 10.694719471947195,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01579913205181538,
      "loss": 0.2351,
      "num_input_tokens_seen": 20517568,
      "step": 97215
    },
    {
      "epoch": 10.695269526952695,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01579769405256636,
      "loss": 0.2325,
      "num_input_tokens_seen": 20518560,
      "step": 97220
    },
    {
      "epoch": 10.695819581958196,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.015796256045965365,
      "loss": 0.2319,
      "num_input_tokens_seen": 20519616,
      "step": 97225
    },
    {
      "epoch": 10.696369636963697,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01579481803202567,
      "loss": 0.2329,
      "num_input_tokens_seen": 20520640,
      "step": 97230
    },
    {
      "epoch": 10.696919691969196,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.015793380010760513,
      "loss": 0.2329,
      "num_input_tokens_seen": 20521696,
      "step": 97235
    },
    {
      "epoch": 10.697469746974697,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.015791941982183158,
      "loss": 0.2288,
      "num_input_tokens_seen": 20522688,
      "step": 97240
    },
    {
      "epoch": 10.698019801980198,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01579050394630685,
      "loss": 0.2293,
      "num_input_tokens_seen": 20523744,
      "step": 97245
    },
    {
      "epoch": 10.6985698569857,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.015789065903144852,
      "loss": 0.2324,
      "num_input_tokens_seen": 20524832,
      "step": 97250
    },
    {
      "epoch": 10.699119911991199,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.0157876278527104,
      "loss": 0.2319,
      "num_input_tokens_seen": 20525824,
      "step": 97255
    },
    {
      "epoch": 10.6996699669967,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.015786189795016774,
      "loss": 0.2329,
      "num_input_tokens_seen": 20526912,
      "step": 97260
    },
    {
      "epoch": 10.7002200220022,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015784751730077207,
      "loss": 0.2329,
      "num_input_tokens_seen": 20528000,
      "step": 97265
    },
    {
      "epoch": 10.7007700770077,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.01578331365790496,
      "loss": 0.2298,
      "num_input_tokens_seen": 20529088,
      "step": 97270
    },
    {
      "epoch": 10.701320132013201,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.015781875578513287,
      "loss": 0.2298,
      "num_input_tokens_seen": 20530144,
      "step": 97275
    },
    {
      "epoch": 10.701870187018702,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.015780437491915444,
      "loss": 0.2319,
      "num_input_tokens_seen": 20531104,
      "step": 97280
    },
    {
      "epoch": 10.702420242024202,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.01577899939812468,
      "loss": 0.2303,
      "num_input_tokens_seen": 20532256,
      "step": 97285
    },
    {
      "epoch": 10.702970297029703,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.015777561297154255,
      "loss": 0.2303,
      "num_input_tokens_seen": 20533280,
      "step": 97290
    },
    {
      "epoch": 10.703520352035204,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.015776123189017416,
      "loss": 0.2324,
      "num_input_tokens_seen": 20534240,
      "step": 97295
    },
    {
      "epoch": 10.704070407040705,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.015774685073727427,
      "loss": 0.2319,
      "num_input_tokens_seen": 20535328,
      "step": 97300
    },
    {
      "epoch": 10.704620462046204,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.01577324695129754,
      "loss": 0.2314,
      "num_input_tokens_seen": 20536416,
      "step": 97305
    },
    {
      "epoch": 10.705170517051705,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.015771808821741,
      "loss": 0.2319,
      "num_input_tokens_seen": 20537408,
      "step": 97310
    },
    {
      "epoch": 10.705720572057206,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01577037068507107,
      "loss": 0.2298,
      "num_input_tokens_seen": 20538432,
      "step": 97315
    },
    {
      "epoch": 10.706270627062706,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015768932541301007,
      "loss": 0.2303,
      "num_input_tokens_seen": 20539456,
      "step": 97320
    },
    {
      "epoch": 10.706820682068207,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.01576749439044406,
      "loss": 0.2309,
      "num_input_tokens_seen": 20540480,
      "step": 97325
    },
    {
      "epoch": 10.707370737073708,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.015766056232513485,
      "loss": 0.2314,
      "num_input_tokens_seen": 20541600,
      "step": 97330
    },
    {
      "epoch": 10.707920792079207,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.015764618067522537,
      "loss": 0.2304,
      "num_input_tokens_seen": 20542656,
      "step": 97335
    },
    {
      "epoch": 10.708470847084708,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01576317989548447,
      "loss": 0.2314,
      "num_input_tokens_seen": 20543680,
      "step": 97340
    },
    {
      "epoch": 10.70902090209021,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.01576174171641254,
      "loss": 0.234,
      "num_input_tokens_seen": 20544800,
      "step": 97345
    },
    {
      "epoch": 10.70957095709571,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.015760303530320003,
      "loss": 0.2314,
      "num_input_tokens_seen": 20545792,
      "step": 97350
    },
    {
      "epoch": 10.71012101210121,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.015758865337220113,
      "loss": 0.2303,
      "num_input_tokens_seen": 20546848,
      "step": 97355
    },
    {
      "epoch": 10.710671067106711,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.01575742713712613,
      "loss": 0.2309,
      "num_input_tokens_seen": 20548000,
      "step": 97360
    },
    {
      "epoch": 10.711221122112212,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.015755988930051302,
      "loss": 0.2304,
      "num_input_tokens_seen": 20549088,
      "step": 97365
    },
    {
      "epoch": 10.711771177117711,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.015754550716008883,
      "loss": 0.2294,
      "num_input_tokens_seen": 20550112,
      "step": 97370
    },
    {
      "epoch": 10.712321232123212,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.01575311249501213,
      "loss": 0.2325,
      "num_input_tokens_seen": 20551232,
      "step": 97375
    },
    {
      "epoch": 10.712871287128714,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01575167426707431,
      "loss": 0.2314,
      "num_input_tokens_seen": 20552288,
      "step": 97380
    },
    {
      "epoch": 10.713421342134213,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.01575023603220866,
      "loss": 0.2304,
      "num_input_tokens_seen": 20553376,
      "step": 97385
    },
    {
      "epoch": 10.713971397139714,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01574879779042845,
      "loss": 0.2335,
      "num_input_tokens_seen": 20554464,
      "step": 97390
    },
    {
      "epoch": 10.714521452145215,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.015747359541746928,
      "loss": 0.2325,
      "num_input_tokens_seen": 20555456,
      "step": 97395
    },
    {
      "epoch": 10.715071507150714,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.015745921286177352,
      "loss": 0.2346,
      "num_input_tokens_seen": 20556448,
      "step": 97400
    },
    {
      "epoch": 10.715621562156215,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01574448302373298,
      "loss": 0.2335,
      "num_input_tokens_seen": 20557536,
      "step": 97405
    },
    {
      "epoch": 10.716171617161717,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.015743044754427064,
      "loss": 0.234,
      "num_input_tokens_seen": 20558528,
      "step": 97410
    },
    {
      "epoch": 10.716721672167218,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015741606478272857,
      "loss": 0.2303,
      "num_input_tokens_seen": 20559520,
      "step": 97415
    },
    {
      "epoch": 10.717271727172717,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.01574016819528362,
      "loss": 0.2345,
      "num_input_tokens_seen": 20560576,
      "step": 97420
    },
    {
      "epoch": 10.717821782178218,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01573872990547261,
      "loss": 0.233,
      "num_input_tokens_seen": 20561664,
      "step": 97425
    },
    {
      "epoch": 10.718371837183719,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.015737291608853083,
      "loss": 0.2283,
      "num_input_tokens_seen": 20562720,
      "step": 97430
    },
    {
      "epoch": 10.718921892189218,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01573585330543829,
      "loss": 0.2293,
      "num_input_tokens_seen": 20563776,
      "step": 97435
    },
    {
      "epoch": 10.71947194719472,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015734414995241492,
      "loss": 0.2324,
      "num_input_tokens_seen": 20564832,
      "step": 97440
    },
    {
      "epoch": 10.72002200220022,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01573297667827594,
      "loss": 0.2293,
      "num_input_tokens_seen": 20565888,
      "step": 97445
    },
    {
      "epoch": 10.72057205720572,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.0157315383545549,
      "loss": 0.2324,
      "num_input_tokens_seen": 20567008,
      "step": 97450
    },
    {
      "epoch": 10.721122112211221,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.015730100024091614,
      "loss": 0.2303,
      "num_input_tokens_seen": 20568096,
      "step": 97455
    },
    {
      "epoch": 10.721672167216722,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01572866168689935,
      "loss": 0.2319,
      "num_input_tokens_seen": 20569120,
      "step": 97460
    },
    {
      "epoch": 10.722222222222221,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.015727223342991363,
      "loss": 0.234,
      "num_input_tokens_seen": 20570112,
      "step": 97465
    },
    {
      "epoch": 10.722772277227723,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.015725784992380906,
      "loss": 0.2319,
      "num_input_tokens_seen": 20571200,
      "step": 97470
    },
    {
      "epoch": 10.723322332233224,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.015724346635081238,
      "loss": 0.2304,
      "num_input_tokens_seen": 20572288,
      "step": 97475
    },
    {
      "epoch": 10.723872387238725,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.015722908271105614,
      "loss": 0.2308,
      "num_input_tokens_seen": 20573248,
      "step": 97480
    },
    {
      "epoch": 10.724422442244224,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.015721469900467286,
      "loss": 0.2314,
      "num_input_tokens_seen": 20574368,
      "step": 97485
    },
    {
      "epoch": 10.724972497249725,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.015720031523179523,
      "loss": 0.2319,
      "num_input_tokens_seen": 20575392,
      "step": 97490
    },
    {
      "epoch": 10.725522552255226,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.015718593139255572,
      "loss": 0.2303,
      "num_input_tokens_seen": 20576384,
      "step": 97495
    },
    {
      "epoch": 10.726072607260726,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.015717154748708693,
      "loss": 0.233,
      "num_input_tokens_seen": 20577440,
      "step": 97500
    },
    {
      "epoch": 10.726622662266227,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.015715716351552143,
      "loss": 0.2319,
      "num_input_tokens_seen": 20578496,
      "step": 97505
    },
    {
      "epoch": 10.727172717271728,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01571427794779918,
      "loss": 0.2288,
      "num_input_tokens_seen": 20579584,
      "step": 97510
    },
    {
      "epoch": 10.727722772277227,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.015712839537463056,
      "loss": 0.2314,
      "num_input_tokens_seen": 20580672,
      "step": 97515
    },
    {
      "epoch": 10.728272827282728,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.015711401120557036,
      "loss": 0.2329,
      "num_input_tokens_seen": 20581728,
      "step": 97520
    },
    {
      "epoch": 10.72882288228823,
      "grad_norm": 0.0028076171875,
      "learning_rate": 0.015709962697094376,
      "loss": 0.2303,
      "num_input_tokens_seen": 20582784,
      "step": 97525
    },
    {
      "epoch": 10.729372937293729,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.015708524267088325,
      "loss": 0.2319,
      "num_input_tokens_seen": 20583872,
      "step": 97530
    },
    {
      "epoch": 10.72992299229923,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.015707085830552147,
      "loss": 0.2335,
      "num_input_tokens_seen": 20584992,
      "step": 97535
    },
    {
      "epoch": 10.73047304730473,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.015705647387499102,
      "loss": 0.2308,
      "num_input_tokens_seen": 20586048,
      "step": 97540
    },
    {
      "epoch": 10.731023102310232,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.015704208937942436,
      "loss": 0.2298,
      "num_input_tokens_seen": 20587072,
      "step": 97545
    },
    {
      "epoch": 10.731573157315731,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.015702770481895424,
      "loss": 0.2319,
      "num_input_tokens_seen": 20588192,
      "step": 97550
    },
    {
      "epoch": 10.732123212321232,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.01570133201937131,
      "loss": 0.2303,
      "num_input_tokens_seen": 20589216,
      "step": 97555
    },
    {
      "epoch": 10.732673267326733,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.015699893550383346,
      "loss": 0.2309,
      "num_input_tokens_seen": 20590336,
      "step": 97560
    },
    {
      "epoch": 10.733223322332233,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.01569845507494481,
      "loss": 0.2303,
      "num_input_tokens_seen": 20591360,
      "step": 97565
    },
    {
      "epoch": 10.733773377337734,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.015697016593068946,
      "loss": 0.2303,
      "num_input_tokens_seen": 20592416,
      "step": 97570
    },
    {
      "epoch": 10.734323432343235,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.01569557810476901,
      "loss": 0.2329,
      "num_input_tokens_seen": 20593408,
      "step": 97575
    },
    {
      "epoch": 10.734873487348734,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.01569413961005827,
      "loss": 0.2298,
      "num_input_tokens_seen": 20594464,
      "step": 97580
    },
    {
      "epoch": 10.735423542354235,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01569270110894998,
      "loss": 0.2308,
      "num_input_tokens_seen": 20595488,
      "step": 97585
    },
    {
      "epoch": 10.735973597359736,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.01569126260145739,
      "loss": 0.2314,
      "num_input_tokens_seen": 20596576,
      "step": 97590
    },
    {
      "epoch": 10.736523652365236,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.015689824087593767,
      "loss": 0.2324,
      "num_input_tokens_seen": 20597664,
      "step": 97595
    },
    {
      "epoch": 10.737073707370737,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.015688385567372367,
      "loss": 0.234,
      "num_input_tokens_seen": 20598720,
      "step": 97600
    },
    {
      "epoch": 10.737623762376238,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015686947040806448,
      "loss": 0.2314,
      "num_input_tokens_seen": 20599744,
      "step": 97605
    },
    {
      "epoch": 10.738173817381739,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.015685508507909264,
      "loss": 0.2324,
      "num_input_tokens_seen": 20600768,
      "step": 97610
    },
    {
      "epoch": 10.738723872387238,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01568406996869408,
      "loss": 0.2329,
      "num_input_tokens_seen": 20601856,
      "step": 97615
    },
    {
      "epoch": 10.73927392739274,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.015682631423174154,
      "loss": 0.2324,
      "num_input_tokens_seen": 20602912,
      "step": 97620
    },
    {
      "epoch": 10.73982398239824,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01568119287136274,
      "loss": 0.2319,
      "num_input_tokens_seen": 20603904,
      "step": 97625
    },
    {
      "epoch": 10.74037403740374,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.015679754313273092,
      "loss": 0.2319,
      "num_input_tokens_seen": 20604928,
      "step": 97630
    },
    {
      "epoch": 10.74092409240924,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01567831574891848,
      "loss": 0.2313,
      "num_input_tokens_seen": 20605984,
      "step": 97635
    },
    {
      "epoch": 10.741474147414742,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01567687717831216,
      "loss": 0.2329,
      "num_input_tokens_seen": 20607040,
      "step": 97640
    },
    {
      "epoch": 10.742024202420241,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.015675438601467383,
      "loss": 0.2329,
      "num_input_tokens_seen": 20608064,
      "step": 97645
    },
    {
      "epoch": 10.742574257425742,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.015674000018397416,
      "loss": 0.2314,
      "num_input_tokens_seen": 20609088,
      "step": 97650
    },
    {
      "epoch": 10.743124312431243,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.015672561429115512,
      "loss": 0.234,
      "num_input_tokens_seen": 20610144,
      "step": 97655
    },
    {
      "epoch": 10.743674367436743,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.015671122833634928,
      "loss": 0.2293,
      "num_input_tokens_seen": 20611168,
      "step": 97660
    },
    {
      "epoch": 10.744224422442244,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.015669684231968937,
      "loss": 0.2303,
      "num_input_tokens_seen": 20612288,
      "step": 97665
    },
    {
      "epoch": 10.744774477447745,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.015668245624130783,
      "loss": 0.2329,
      "num_input_tokens_seen": 20613312,
      "step": 97670
    },
    {
      "epoch": 10.745324532453246,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.015666807010133725,
      "loss": 0.2298,
      "num_input_tokens_seen": 20614368,
      "step": 97675
    },
    {
      "epoch": 10.745874587458745,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01566536838999103,
      "loss": 0.2319,
      "num_input_tokens_seen": 20615424,
      "step": 97680
    },
    {
      "epoch": 10.746424642464246,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01566392976371596,
      "loss": 0.2324,
      "num_input_tokens_seen": 20616480,
      "step": 97685
    },
    {
      "epoch": 10.746974697469748,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.01566249113132176,
      "loss": 0.2313,
      "num_input_tokens_seen": 20617568,
      "step": 97690
    },
    {
      "epoch": 10.747524752475247,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0156610524928217,
      "loss": 0.2314,
      "num_input_tokens_seen": 20618656,
      "step": 97695
    },
    {
      "epoch": 10.748074807480748,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01565961384822904,
      "loss": 0.2304,
      "num_input_tokens_seen": 20619648,
      "step": 97700
    },
    {
      "epoch": 10.748624862486249,
      "grad_norm": 0.00066375732421875,
      "learning_rate": 0.015658175197557026,
      "loss": 0.2298,
      "num_input_tokens_seen": 20620704,
      "step": 97705
    },
    {
      "epoch": 10.749174917491748,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.015656736540818932,
      "loss": 0.2314,
      "num_input_tokens_seen": 20621792,
      "step": 97710
    },
    {
      "epoch": 10.74972497249725,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.015655297878028013,
      "loss": 0.2298,
      "num_input_tokens_seen": 20622816,
      "step": 97715
    },
    {
      "epoch": 10.75027502750275,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.015653859209197526,
      "loss": 0.2303,
      "num_input_tokens_seen": 20623936,
      "step": 97720
    },
    {
      "epoch": 10.750825082508252,
      "grad_norm": 0.000774383544921875,
      "learning_rate": 0.01565242053434074,
      "loss": 0.2335,
      "num_input_tokens_seen": 20624896,
      "step": 97725
    },
    {
      "epoch": 10.751375137513751,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0156509818534709,
      "loss": 0.2314,
      "num_input_tokens_seen": 20625888,
      "step": 97730
    },
    {
      "epoch": 10.751925192519252,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.015649543166601273,
      "loss": 0.2298,
      "num_input_tokens_seen": 20626880,
      "step": 97735
    },
    {
      "epoch": 10.752475247524753,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01564810447374512,
      "loss": 0.2314,
      "num_input_tokens_seen": 20628000,
      "step": 97740
    },
    {
      "epoch": 10.753025302530252,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.015646665774915695,
      "loss": 0.2309,
      "num_input_tokens_seen": 20629056,
      "step": 97745
    },
    {
      "epoch": 10.753575357535754,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.015645227070126267,
      "loss": 0.2319,
      "num_input_tokens_seen": 20630048,
      "step": 97750
    },
    {
      "epoch": 10.754125412541255,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.015643788359390093,
      "loss": 0.2298,
      "num_input_tokens_seen": 20631168,
      "step": 97755
    },
    {
      "epoch": 10.754675467546754,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.015642349642720425,
      "loss": 0.2319,
      "num_input_tokens_seen": 20632224,
      "step": 97760
    },
    {
      "epoch": 10.755225522552255,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01564091092013053,
      "loss": 0.2319,
      "num_input_tokens_seen": 20633280,
      "step": 97765
    },
    {
      "epoch": 10.755775577557756,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.015639472191633665,
      "loss": 0.2309,
      "num_input_tokens_seen": 20634368,
      "step": 97770
    },
    {
      "epoch": 10.756325632563257,
      "grad_norm": 0.0006103515625,
      "learning_rate": 0.015638033457243093,
      "loss": 0.2319,
      "num_input_tokens_seen": 20635488,
      "step": 97775
    },
    {
      "epoch": 10.756875687568757,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.015636594716972076,
      "loss": 0.2314,
      "num_input_tokens_seen": 20636608,
      "step": 97780
    },
    {
      "epoch": 10.757425742574258,
      "grad_norm": 0.000591278076171875,
      "learning_rate": 0.01563515597083387,
      "loss": 0.2319,
      "num_input_tokens_seen": 20637632,
      "step": 97785
    },
    {
      "epoch": 10.757975797579759,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01563371721884173,
      "loss": 0.2314,
      "num_input_tokens_seen": 20638656,
      "step": 97790
    },
    {
      "epoch": 10.758525852585258,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.01563227846100893,
      "loss": 0.233,
      "num_input_tokens_seen": 20639680,
      "step": 97795
    },
    {
      "epoch": 10.75907590759076,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.015630839697348718,
      "loss": 0.2324,
      "num_input_tokens_seen": 20640736,
      "step": 97800
    },
    {
      "epoch": 10.75962596259626,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.015629400927874355,
      "loss": 0.2298,
      "num_input_tokens_seen": 20641856,
      "step": 97805
    },
    {
      "epoch": 10.76017601760176,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.015627962152599116,
      "loss": 0.2314,
      "num_input_tokens_seen": 20642944,
      "step": 97810
    },
    {
      "epoch": 10.76072607260726,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01562652337153625,
      "loss": 0.2304,
      "num_input_tokens_seen": 20644000,
      "step": 97815
    },
    {
      "epoch": 10.761276127612762,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.015625084584699012,
      "loss": 0.2298,
      "num_input_tokens_seen": 20645024,
      "step": 97820
    },
    {
      "epoch": 10.761826182618261,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.015623645792100673,
      "loss": 0.234,
      "num_input_tokens_seen": 20646016,
      "step": 97825
    },
    {
      "epoch": 10.762376237623762,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.015622206993754491,
      "loss": 0.2304,
      "num_input_tokens_seen": 20647008,
      "step": 97830
    },
    {
      "epoch": 10.762926292629263,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.015620768189673723,
      "loss": 0.2314,
      "num_input_tokens_seen": 20648064,
      "step": 97835
    },
    {
      "epoch": 10.763476347634764,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.015619329379871635,
      "loss": 0.2293,
      "num_input_tokens_seen": 20649152,
      "step": 97840
    },
    {
      "epoch": 10.764026402640264,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.015617890564361487,
      "loss": 0.2303,
      "num_input_tokens_seen": 20650240,
      "step": 97845
    },
    {
      "epoch": 10.764576457645765,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.015616451743156531,
      "loss": 0.2314,
      "num_input_tokens_seen": 20651296,
      "step": 97850
    },
    {
      "epoch": 10.765126512651266,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.015615012916270045,
      "loss": 0.2309,
      "num_input_tokens_seen": 20652448,
      "step": 97855
    },
    {
      "epoch": 10.765676567656765,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.015613574083715274,
      "loss": 0.2303,
      "num_input_tokens_seen": 20653472,
      "step": 97860
    },
    {
      "epoch": 10.766226622662266,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.015612135245505483,
      "loss": 0.2324,
      "num_input_tokens_seen": 20654560,
      "step": 97865
    },
    {
      "epoch": 10.766776677667767,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01561069640165394,
      "loss": 0.2314,
      "num_input_tokens_seen": 20655648,
      "step": 97870
    },
    {
      "epoch": 10.767326732673267,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.015609257552173897,
      "loss": 0.233,
      "num_input_tokens_seen": 20656768,
      "step": 97875
    },
    {
      "epoch": 10.767876787678768,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.015607818697078623,
      "loss": 0.2314,
      "num_input_tokens_seen": 20657792,
      "step": 97880
    },
    {
      "epoch": 10.768426842684269,
      "grad_norm": 0.002655029296875,
      "learning_rate": 0.015606379836381374,
      "loss": 0.2309,
      "num_input_tokens_seen": 20658816,
      "step": 97885
    },
    {
      "epoch": 10.768976897689768,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.015604940970095415,
      "loss": 0.2319,
      "num_input_tokens_seen": 20659904,
      "step": 97890
    },
    {
      "epoch": 10.76952695269527,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.015603502098233999,
      "loss": 0.2298,
      "num_input_tokens_seen": 20660960,
      "step": 97895
    },
    {
      "epoch": 10.77007700770077,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.0156020632208104,
      "loss": 0.2324,
      "num_input_tokens_seen": 20662080,
      "step": 97900
    },
    {
      "epoch": 10.770627062706271,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.015600624337837873,
      "loss": 0.2298,
      "num_input_tokens_seen": 20663232,
      "step": 97905
    },
    {
      "epoch": 10.77117711771177,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.015599185449329676,
      "loss": 0.2299,
      "num_input_tokens_seen": 20664256,
      "step": 97910
    },
    {
      "epoch": 10.771727172717272,
      "grad_norm": 0.000812530517578125,
      "learning_rate": 0.015597746555299077,
      "loss": 0.233,
      "num_input_tokens_seen": 20665280,
      "step": 97915
    },
    {
      "epoch": 10.772277227722773,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.015596307655759336,
      "loss": 0.2319,
      "num_input_tokens_seen": 20666304,
      "step": 97920
    },
    {
      "epoch": 10.772827282728272,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01559486875072371,
      "loss": 0.2324,
      "num_input_tokens_seen": 20667328,
      "step": 97925
    },
    {
      "epoch": 10.773377337733773,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01559342984020547,
      "loss": 0.2314,
      "num_input_tokens_seen": 20668416,
      "step": 97930
    },
    {
      "epoch": 10.773927392739274,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.015591990924217865,
      "loss": 0.2298,
      "num_input_tokens_seen": 20669472,
      "step": 97935
    },
    {
      "epoch": 10.774477447744774,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.015590552002774165,
      "loss": 0.2309,
      "num_input_tokens_seen": 20670560,
      "step": 97940
    },
    {
      "epoch": 10.775027502750275,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.015589113075887635,
      "loss": 0.2324,
      "num_input_tokens_seen": 20671680,
      "step": 97945
    },
    {
      "epoch": 10.775577557755776,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.015587674143571527,
      "loss": 0.2314,
      "num_input_tokens_seen": 20672704,
      "step": 97950
    },
    {
      "epoch": 10.776127612761275,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015586235205839113,
      "loss": 0.2303,
      "num_input_tokens_seen": 20673728,
      "step": 97955
    },
    {
      "epoch": 10.776677667766776,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.015584796262703648,
      "loss": 0.2324,
      "num_input_tokens_seen": 20674880,
      "step": 97960
    },
    {
      "epoch": 10.777227722772277,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.015583357314178398,
      "loss": 0.2313,
      "num_input_tokens_seen": 20675936,
      "step": 97965
    },
    {
      "epoch": 10.777777777777779,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015581918360276621,
      "loss": 0.2309,
      "num_input_tokens_seen": 20677024,
      "step": 97970
    },
    {
      "epoch": 10.778327832783278,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.015580479401011586,
      "loss": 0.2319,
      "num_input_tokens_seen": 20678048,
      "step": 97975
    },
    {
      "epoch": 10.778877887788779,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.015579040436396546,
      "loss": 0.2314,
      "num_input_tokens_seen": 20679136,
      "step": 97980
    },
    {
      "epoch": 10.77942794279428,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.015577601466444772,
      "loss": 0.2308,
      "num_input_tokens_seen": 20680160,
      "step": 97985
    },
    {
      "epoch": 10.77997799779978,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01557616249116952,
      "loss": 0.2314,
      "num_input_tokens_seen": 20681184,
      "step": 97990
    },
    {
      "epoch": 10.78052805280528,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015574723510584053,
      "loss": 0.2324,
      "num_input_tokens_seen": 20682304,
      "step": 97995
    },
    {
      "epoch": 10.781078107810782,
      "grad_norm": 0.005859375,
      "learning_rate": 0.015573284524701643,
      "loss": 0.2303,
      "num_input_tokens_seen": 20683360,
      "step": 98000
    },
    {
      "epoch": 10.781628162816281,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.015571845533535538,
      "loss": 0.2309,
      "num_input_tokens_seen": 20684416,
      "step": 98005
    },
    {
      "epoch": 10.782178217821782,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.015570406537099004,
      "loss": 0.2319,
      "num_input_tokens_seen": 20685472,
      "step": 98010
    },
    {
      "epoch": 10.782728272827283,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.015568967535405313,
      "loss": 0.2329,
      "num_input_tokens_seen": 20686528,
      "step": 98015
    },
    {
      "epoch": 10.783278327832782,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.015567528528467718,
      "loss": 0.2314,
      "num_input_tokens_seen": 20687584,
      "step": 98020
    },
    {
      "epoch": 10.783828382838283,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.01556608951629949,
      "loss": 0.2308,
      "num_input_tokens_seen": 20688640,
      "step": 98025
    },
    {
      "epoch": 10.784378437843785,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.015564650498913882,
      "loss": 0.2324,
      "num_input_tokens_seen": 20689632,
      "step": 98030
    },
    {
      "epoch": 10.784928492849286,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.015563211476324163,
      "loss": 0.2329,
      "num_input_tokens_seen": 20690688,
      "step": 98035
    },
    {
      "epoch": 10.785478547854785,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.01556177244854359,
      "loss": 0.2303,
      "num_input_tokens_seen": 20691712,
      "step": 98040
    },
    {
      "epoch": 10.786028602860286,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015560333415585435,
      "loss": 0.2324,
      "num_input_tokens_seen": 20692768,
      "step": 98045
    },
    {
      "epoch": 10.786578657865787,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.015558894377462952,
      "loss": 0.2335,
      "num_input_tokens_seen": 20693824,
      "step": 98050
    },
    {
      "epoch": 10.787128712871286,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.015557455334189412,
      "loss": 0.2345,
      "num_input_tokens_seen": 20694880,
      "step": 98055
    },
    {
      "epoch": 10.787678767876788,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01555601628577807,
      "loss": 0.2308,
      "num_input_tokens_seen": 20695936,
      "step": 98060
    },
    {
      "epoch": 10.788228822882289,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.015554577232242192,
      "loss": 0.2324,
      "num_input_tokens_seen": 20696960,
      "step": 98065
    },
    {
      "epoch": 10.788778877887788,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.015553138173595047,
      "loss": 0.2308,
      "num_input_tokens_seen": 20697952,
      "step": 98070
    },
    {
      "epoch": 10.789328932893289,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01555169910984989,
      "loss": 0.2298,
      "num_input_tokens_seen": 20699008,
      "step": 98075
    },
    {
      "epoch": 10.78987898789879,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.015550260041019982,
      "loss": 0.2329,
      "num_input_tokens_seen": 20700032,
      "step": 98080
    },
    {
      "epoch": 10.79042904290429,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.015548820967118598,
      "loss": 0.2319,
      "num_input_tokens_seen": 20701056,
      "step": 98085
    },
    {
      "epoch": 10.79097909790979,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.015547381888158992,
      "loss": 0.2304,
      "num_input_tokens_seen": 20702144,
      "step": 98090
    },
    {
      "epoch": 10.791529152915292,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015545942804154429,
      "loss": 0.2303,
      "num_input_tokens_seen": 20703200,
      "step": 98095
    },
    {
      "epoch": 10.792079207920793,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.015544503715118176,
      "loss": 0.2298,
      "num_input_tokens_seen": 20704224,
      "step": 98100
    },
    {
      "epoch": 10.792629262926292,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.015543064621063493,
      "loss": 0.2319,
      "num_input_tokens_seen": 20705280,
      "step": 98105
    },
    {
      "epoch": 10.793179317931793,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.015541625522003642,
      "loss": 0.2308,
      "num_input_tokens_seen": 20706368,
      "step": 98110
    },
    {
      "epoch": 10.793729372937294,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015540186417951888,
      "loss": 0.2303,
      "num_input_tokens_seen": 20707392,
      "step": 98115
    },
    {
      "epoch": 10.794279427942794,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.015538747308921498,
      "loss": 0.2303,
      "num_input_tokens_seen": 20708480,
      "step": 98120
    },
    {
      "epoch": 10.794829482948295,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.01553730819492573,
      "loss": 0.2314,
      "num_input_tokens_seen": 20709472,
      "step": 98125
    },
    {
      "epoch": 10.795379537953796,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.01553586907597785,
      "loss": 0.2314,
      "num_input_tokens_seen": 20710496,
      "step": 98130
    },
    {
      "epoch": 10.795929592959295,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.015534429952091124,
      "loss": 0.2314,
      "num_input_tokens_seen": 20711520,
      "step": 98135
    },
    {
      "epoch": 10.796479647964796,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.015532990823278815,
      "loss": 0.2325,
      "num_input_tokens_seen": 20712576,
      "step": 98140
    },
    {
      "epoch": 10.797029702970297,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01553155168955418,
      "loss": 0.2324,
      "num_input_tokens_seen": 20713600,
      "step": 98145
    },
    {
      "epoch": 10.797579757975798,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.015530112550930493,
      "loss": 0.2335,
      "num_input_tokens_seen": 20714688,
      "step": 98150
    },
    {
      "epoch": 10.798129812981298,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.015528673407421006,
      "loss": 0.2329,
      "num_input_tokens_seen": 20715680,
      "step": 98155
    },
    {
      "epoch": 10.798679867986799,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.015527234259038998,
      "loss": 0.2325,
      "num_input_tokens_seen": 20716736,
      "step": 98160
    },
    {
      "epoch": 10.7992299229923,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.015525795105797723,
      "loss": 0.2319,
      "num_input_tokens_seen": 20717824,
      "step": 98165
    },
    {
      "epoch": 10.7997799779978,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.015524355947710445,
      "loss": 0.2299,
      "num_input_tokens_seen": 20718848,
      "step": 98170
    },
    {
      "epoch": 10.8003300330033,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.015522916784790428,
      "loss": 0.2304,
      "num_input_tokens_seen": 20719936,
      "step": 98175
    },
    {
      "epoch": 10.800880088008801,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01552147761705094,
      "loss": 0.2335,
      "num_input_tokens_seen": 20721056,
      "step": 98180
    },
    {
      "epoch": 10.8014301430143,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.015520038444505242,
      "loss": 0.2336,
      "num_input_tokens_seen": 20722080,
      "step": 98185
    },
    {
      "epoch": 10.801980198019802,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.015518599267166606,
      "loss": 0.2304,
      "num_input_tokens_seen": 20723104,
      "step": 98190
    },
    {
      "epoch": 10.802530253025303,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.015517160085048283,
      "loss": 0.2304,
      "num_input_tokens_seen": 20724192,
      "step": 98195
    },
    {
      "epoch": 10.803080308030804,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.01551572089816354,
      "loss": 0.2325,
      "num_input_tokens_seen": 20725280,
      "step": 98200
    },
    {
      "epoch": 10.803630363036303,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.015514281706525654,
      "loss": 0.2309,
      "num_input_tokens_seen": 20726400,
      "step": 98205
    },
    {
      "epoch": 10.804180418041804,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.015512842510147874,
      "loss": 0.2304,
      "num_input_tokens_seen": 20727456,
      "step": 98210
    },
    {
      "epoch": 10.804730473047305,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.015511403309043471,
      "loss": 0.2335,
      "num_input_tokens_seen": 20728576,
      "step": 98215
    },
    {
      "epoch": 10.805280528052805,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.015509964103225714,
      "loss": 0.232,
      "num_input_tokens_seen": 20729664,
      "step": 98220
    },
    {
      "epoch": 10.805830583058306,
      "grad_norm": 0.0030670166015625,
      "learning_rate": 0.015508524892707858,
      "loss": 0.2298,
      "num_input_tokens_seen": 20730688,
      "step": 98225
    },
    {
      "epoch": 10.806380638063807,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.015507085677503168,
      "loss": 0.2314,
      "num_input_tokens_seen": 20731776,
      "step": 98230
    },
    {
      "epoch": 10.806930693069306,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.015505646457624925,
      "loss": 0.2309,
      "num_input_tokens_seen": 20732832,
      "step": 98235
    },
    {
      "epoch": 10.807480748074807,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.015504207233086369,
      "loss": 0.2309,
      "num_input_tokens_seen": 20733856,
      "step": 98240
    },
    {
      "epoch": 10.808030803080309,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.015502768003900783,
      "loss": 0.233,
      "num_input_tokens_seen": 20734848,
      "step": 98245
    },
    {
      "epoch": 10.808580858085808,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.015501328770081427,
      "loss": 0.2304,
      "num_input_tokens_seen": 20735936,
      "step": 98250
    },
    {
      "epoch": 10.809130913091309,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.01549988953164156,
      "loss": 0.2314,
      "num_input_tokens_seen": 20736992,
      "step": 98255
    },
    {
      "epoch": 10.80968096809681,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.015498450288594453,
      "loss": 0.2335,
      "num_input_tokens_seen": 20738080,
      "step": 98260
    },
    {
      "epoch": 10.810231023102311,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.015497011040953369,
      "loss": 0.2304,
      "num_input_tokens_seen": 20739168,
      "step": 98265
    },
    {
      "epoch": 10.81078107810781,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.015495571788731571,
      "loss": 0.2314,
      "num_input_tokens_seen": 20740224,
      "step": 98270
    },
    {
      "epoch": 10.811331133113312,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.015494132531942326,
      "loss": 0.232,
      "num_input_tokens_seen": 20741312,
      "step": 98275
    },
    {
      "epoch": 10.811881188118813,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.0154926932705989,
      "loss": 0.2309,
      "num_input_tokens_seen": 20742368,
      "step": 98280
    },
    {
      "epoch": 10.812431243124312,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.015491254004714556,
      "loss": 0.2304,
      "num_input_tokens_seen": 20743392,
      "step": 98285
    },
    {
      "epoch": 10.812981298129813,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.015489814734302563,
      "loss": 0.2293,
      "num_input_tokens_seen": 20744512,
      "step": 98290
    },
    {
      "epoch": 10.813531353135314,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.015488375459376178,
      "loss": 0.2335,
      "num_input_tokens_seen": 20745600,
      "step": 98295
    },
    {
      "epoch": 10.814081408140813,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.015486936179948669,
      "loss": 0.233,
      "num_input_tokens_seen": 20746688,
      "step": 98300
    },
    {
      "epoch": 10.814631463146315,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.015485496896033306,
      "loss": 0.2335,
      "num_input_tokens_seen": 20747776,
      "step": 98305
    },
    {
      "epoch": 10.815181518151816,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015484057607643355,
      "loss": 0.2309,
      "num_input_tokens_seen": 20748864,
      "step": 98310
    },
    {
      "epoch": 10.815731573157315,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01548261831479207,
      "loss": 0.2298,
      "num_input_tokens_seen": 20749888,
      "step": 98315
    },
    {
      "epoch": 10.816281628162816,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.015481179017492729,
      "loss": 0.2356,
      "num_input_tokens_seen": 20750976,
      "step": 98320
    },
    {
      "epoch": 10.816831683168317,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01547973971575859,
      "loss": 0.2304,
      "num_input_tokens_seen": 20752064,
      "step": 98325
    },
    {
      "epoch": 10.817381738173818,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01547830040960292,
      "loss": 0.2314,
      "num_input_tokens_seen": 20753152,
      "step": 98330
    },
    {
      "epoch": 10.817931793179318,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.015476861099038983,
      "loss": 0.2314,
      "num_input_tokens_seen": 20754176,
      "step": 98335
    },
    {
      "epoch": 10.818481848184819,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.015475421784080049,
      "loss": 0.2303,
      "num_input_tokens_seen": 20755200,
      "step": 98340
    },
    {
      "epoch": 10.81903190319032,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.015473982464739376,
      "loss": 0.2303,
      "num_input_tokens_seen": 20756288,
      "step": 98345
    },
    {
      "epoch": 10.819581958195819,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.01547254314103024,
      "loss": 0.2304,
      "num_input_tokens_seen": 20757344,
      "step": 98350
    },
    {
      "epoch": 10.82013201320132,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.015471103812965898,
      "loss": 0.2283,
      "num_input_tokens_seen": 20758432,
      "step": 98355
    },
    {
      "epoch": 10.820682068206821,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015469664480559617,
      "loss": 0.2319,
      "num_input_tokens_seen": 20759488,
      "step": 98360
    },
    {
      "epoch": 10.82123212321232,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.015468225143824664,
      "loss": 0.2335,
      "num_input_tokens_seen": 20760576,
      "step": 98365
    },
    {
      "epoch": 10.821782178217822,
      "grad_norm": 0.0007171630859375,
      "learning_rate": 0.015466785802774309,
      "loss": 0.2293,
      "num_input_tokens_seen": 20761632,
      "step": 98370
    },
    {
      "epoch": 10.822332233223323,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.015465346457421807,
      "loss": 0.2319,
      "num_input_tokens_seen": 20762688,
      "step": 98375
    },
    {
      "epoch": 10.822882288228822,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.015463907107780435,
      "loss": 0.2314,
      "num_input_tokens_seen": 20763648,
      "step": 98380
    },
    {
      "epoch": 10.823432343234323,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.015462467753863451,
      "loss": 0.2304,
      "num_input_tokens_seen": 20764736,
      "step": 98385
    },
    {
      "epoch": 10.823982398239824,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.015461028395684123,
      "loss": 0.2304,
      "num_input_tokens_seen": 20765728,
      "step": 98390
    },
    {
      "epoch": 10.824532453245325,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01545958903325572,
      "loss": 0.2314,
      "num_input_tokens_seen": 20766688,
      "step": 98395
    },
    {
      "epoch": 10.825082508250825,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.015458149666591502,
      "loss": 0.233,
      "num_input_tokens_seen": 20767776,
      "step": 98400
    },
    {
      "epoch": 10.825632563256326,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.015456710295704742,
      "loss": 0.2298,
      "num_input_tokens_seen": 20768832,
      "step": 98405
    },
    {
      "epoch": 10.826182618261827,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.015455270920608702,
      "loss": 0.2278,
      "num_input_tokens_seen": 20769888,
      "step": 98410
    },
    {
      "epoch": 10.826732673267326,
      "grad_norm": 0.00041961669921875,
      "learning_rate": 0.015453831541316644,
      "loss": 0.234,
      "num_input_tokens_seen": 20770944,
      "step": 98415
    },
    {
      "epoch": 10.827282728272827,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.015452392157841843,
      "loss": 0.2309,
      "num_input_tokens_seen": 20772032,
      "step": 98420
    },
    {
      "epoch": 10.827832783278328,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.01545095277019756,
      "loss": 0.2341,
      "num_input_tokens_seen": 20773088,
      "step": 98425
    },
    {
      "epoch": 10.828382838283828,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.01544951337839706,
      "loss": 0.2309,
      "num_input_tokens_seen": 20774176,
      "step": 98430
    },
    {
      "epoch": 10.828932893289329,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.015448073982453613,
      "loss": 0.2314,
      "num_input_tokens_seen": 20775232,
      "step": 98435
    },
    {
      "epoch": 10.82948294829483,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01544663458238048,
      "loss": 0.233,
      "num_input_tokens_seen": 20776256,
      "step": 98440
    },
    {
      "epoch": 10.83003300330033,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.015445195178190929,
      "loss": 0.2314,
      "num_input_tokens_seen": 20777344,
      "step": 98445
    },
    {
      "epoch": 10.83058305830583,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.015443755769898239,
      "loss": 0.2314,
      "num_input_tokens_seen": 20778432,
      "step": 98450
    },
    {
      "epoch": 10.831133113311331,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.015442316357515655,
      "loss": 0.233,
      "num_input_tokens_seen": 20779456,
      "step": 98455
    },
    {
      "epoch": 10.831683168316832,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.015440876941056452,
      "loss": 0.2325,
      "num_input_tokens_seen": 20780512,
      "step": 98460
    },
    {
      "epoch": 10.832233223322332,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.015439437520533903,
      "loss": 0.2314,
      "num_input_tokens_seen": 20781568,
      "step": 98465
    },
    {
      "epoch": 10.832783278327833,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015437998095961268,
      "loss": 0.2309,
      "num_input_tokens_seen": 20782560,
      "step": 98470
    },
    {
      "epoch": 10.833333333333334,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.015436558667351814,
      "loss": 0.2309,
      "num_input_tokens_seen": 20783648,
      "step": 98475
    },
    {
      "epoch": 10.833883388338833,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.015435119234718806,
      "loss": 0.2304,
      "num_input_tokens_seen": 20784736,
      "step": 98480
    },
    {
      "epoch": 10.834433443344334,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.015433679798075516,
      "loss": 0.2298,
      "num_input_tokens_seen": 20785728,
      "step": 98485
    },
    {
      "epoch": 10.834983498349835,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.015432240357435206,
      "loss": 0.2314,
      "num_input_tokens_seen": 20786880,
      "step": 98490
    },
    {
      "epoch": 10.835533553355335,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.015430800912811144,
      "loss": 0.2293,
      "num_input_tokens_seen": 20787936,
      "step": 98495
    },
    {
      "epoch": 10.836083608360836,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0154293614642166,
      "loss": 0.2304,
      "num_input_tokens_seen": 20788992,
      "step": 98500
    },
    {
      "epoch": 10.836633663366337,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.015427922011664832,
      "loss": 0.2319,
      "num_input_tokens_seen": 20790048,
      "step": 98505
    },
    {
      "epoch": 10.837183718371836,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.015426482555169116,
      "loss": 0.2319,
      "num_input_tokens_seen": 20791072,
      "step": 98510
    },
    {
      "epoch": 10.837733773377337,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015425043094742713,
      "loss": 0.2293,
      "num_input_tokens_seen": 20792128,
      "step": 98515
    },
    {
      "epoch": 10.838283828382838,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015423603630398896,
      "loss": 0.2319,
      "num_input_tokens_seen": 20793216,
      "step": 98520
    },
    {
      "epoch": 10.83883388338834,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015422164162150923,
      "loss": 0.233,
      "num_input_tokens_seen": 20794272,
      "step": 98525
    },
    {
      "epoch": 10.839383938393839,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.015420724690012063,
      "loss": 0.2324,
      "num_input_tokens_seen": 20795360,
      "step": 98530
    },
    {
      "epoch": 10.83993399339934,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.015419285213995592,
      "loss": 0.2314,
      "num_input_tokens_seen": 20796448,
      "step": 98535
    },
    {
      "epoch": 10.840484048404841,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.015417845734114767,
      "loss": 0.2319,
      "num_input_tokens_seen": 20797536,
      "step": 98540
    },
    {
      "epoch": 10.84103410341034,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.015416406250382857,
      "loss": 0.2304,
      "num_input_tokens_seen": 20798592,
      "step": 98545
    },
    {
      "epoch": 10.841584158415841,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015414966762813134,
      "loss": 0.2319,
      "num_input_tokens_seen": 20799648,
      "step": 98550
    },
    {
      "epoch": 10.842134213421343,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.015413527271418861,
      "loss": 0.2304,
      "num_input_tokens_seen": 20800736,
      "step": 98555
    },
    {
      "epoch": 10.842684268426842,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.015412087776213302,
      "loss": 0.2314,
      "num_input_tokens_seen": 20801792,
      "step": 98560
    },
    {
      "epoch": 10.843234323432343,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.015410648277209728,
      "loss": 0.2329,
      "num_input_tokens_seen": 20802880,
      "step": 98565
    },
    {
      "epoch": 10.843784378437844,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.015409208774421411,
      "loss": 0.2319,
      "num_input_tokens_seen": 20803936,
      "step": 98570
    },
    {
      "epoch": 10.844334433443345,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.01540776926786161,
      "loss": 0.2303,
      "num_input_tokens_seen": 20804928,
      "step": 98575
    },
    {
      "epoch": 10.844884488448844,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.015406329757543592,
      "loss": 0.2314,
      "num_input_tokens_seen": 20806016,
      "step": 98580
    },
    {
      "epoch": 10.845434543454346,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.015404890243480632,
      "loss": 0.2319,
      "num_input_tokens_seen": 20807040,
      "step": 98585
    },
    {
      "epoch": 10.845984598459847,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.01540345072568599,
      "loss": 0.234,
      "num_input_tokens_seen": 20808064,
      "step": 98590
    },
    {
      "epoch": 10.846534653465346,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015402011204172941,
      "loss": 0.2335,
      "num_input_tokens_seen": 20809152,
      "step": 98595
    },
    {
      "epoch": 10.847084708470847,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.015400571678954747,
      "loss": 0.2303,
      "num_input_tokens_seen": 20810208,
      "step": 98600
    },
    {
      "epoch": 10.847634763476348,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.015399132150044669,
      "loss": 0.2303,
      "num_input_tokens_seen": 20811264,
      "step": 98605
    },
    {
      "epoch": 10.848184818481847,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.015397692617455987,
      "loss": 0.2303,
      "num_input_tokens_seen": 20812288,
      "step": 98610
    },
    {
      "epoch": 10.848734873487349,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.015396253081201964,
      "loss": 0.2293,
      "num_input_tokens_seen": 20813344,
      "step": 98615
    },
    {
      "epoch": 10.84928492849285,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.015394813541295863,
      "loss": 0.2313,
      "num_input_tokens_seen": 20814336,
      "step": 98620
    },
    {
      "epoch": 10.84983498349835,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015393373997750958,
      "loss": 0.2314,
      "num_input_tokens_seen": 20815328,
      "step": 98625
    },
    {
      "epoch": 10.85038503850385,
      "grad_norm": 0.00072479248046875,
      "learning_rate": 0.015391934450580513,
      "loss": 0.2335,
      "num_input_tokens_seen": 20816416,
      "step": 98630
    },
    {
      "epoch": 10.850935093509351,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.015390494899797793,
      "loss": 0.2345,
      "num_input_tokens_seen": 20817504,
      "step": 98635
    },
    {
      "epoch": 10.851485148514852,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015389055345416073,
      "loss": 0.2329,
      "num_input_tokens_seen": 20818496,
      "step": 98640
    },
    {
      "epoch": 10.852035203520352,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.015387615787448615,
      "loss": 0.2319,
      "num_input_tokens_seen": 20819488,
      "step": 98645
    },
    {
      "epoch": 10.852585258525853,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.015386176225908686,
      "loss": 0.2314,
      "num_input_tokens_seen": 20820544,
      "step": 98650
    },
    {
      "epoch": 10.853135313531354,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.01538473666080956,
      "loss": 0.2324,
      "num_input_tokens_seen": 20821632,
      "step": 98655
    },
    {
      "epoch": 10.853685368536853,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0153832970921645,
      "loss": 0.2345,
      "num_input_tokens_seen": 20822688,
      "step": 98660
    },
    {
      "epoch": 10.854235423542354,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.015381857519986773,
      "loss": 0.2319,
      "num_input_tokens_seen": 20823776,
      "step": 98665
    },
    {
      "epoch": 10.854785478547855,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.01538041794428965,
      "loss": 0.2319,
      "num_input_tokens_seen": 20824896,
      "step": 98670
    },
    {
      "epoch": 10.855335533553355,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.015378978365086397,
      "loss": 0.2324,
      "num_input_tokens_seen": 20825984,
      "step": 98675
    },
    {
      "epoch": 10.855885588558856,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.01537753878239028,
      "loss": 0.2329,
      "num_input_tokens_seen": 20827072,
      "step": 98680
    },
    {
      "epoch": 10.856435643564357,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.01537609919621457,
      "loss": 0.2308,
      "num_input_tokens_seen": 20828064,
      "step": 98685
    },
    {
      "epoch": 10.856985698569858,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.015374659606572536,
      "loss": 0.2313,
      "num_input_tokens_seen": 20829152,
      "step": 98690
    },
    {
      "epoch": 10.857535753575357,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.015373220013477445,
      "loss": 0.2309,
      "num_input_tokens_seen": 20830304,
      "step": 98695
    },
    {
      "epoch": 10.858085808580858,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015371780416942562,
      "loss": 0.2308,
      "num_input_tokens_seen": 20831296,
      "step": 98700
    },
    {
      "epoch": 10.85863586358636,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01537034081698116,
      "loss": 0.2303,
      "num_input_tokens_seen": 20832320,
      "step": 98705
    },
    {
      "epoch": 10.859185918591859,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.0153689012136065,
      "loss": 0.2298,
      "num_input_tokens_seen": 20833408,
      "step": 98710
    },
    {
      "epoch": 10.85973597359736,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.015367461606831858,
      "loss": 0.2313,
      "num_input_tokens_seen": 20834496,
      "step": 98715
    },
    {
      "epoch": 10.86028602860286,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.015366021996670495,
      "loss": 0.2314,
      "num_input_tokens_seen": 20835584,
      "step": 98720
    },
    {
      "epoch": 10.86083608360836,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.015364582383135686,
      "loss": 0.2288,
      "num_input_tokens_seen": 20836672,
      "step": 98725
    },
    {
      "epoch": 10.861386138613861,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.015363142766240696,
      "loss": 0.2308,
      "num_input_tokens_seen": 20837696,
      "step": 98730
    },
    {
      "epoch": 10.861936193619362,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015361703145998792,
      "loss": 0.2314,
      "num_input_tokens_seen": 20838752,
      "step": 98735
    },
    {
      "epoch": 10.862486248624862,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.015360263522423246,
      "loss": 0.2303,
      "num_input_tokens_seen": 20839744,
      "step": 98740
    },
    {
      "epoch": 10.863036303630363,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.015358823895527321,
      "loss": 0.2309,
      "num_input_tokens_seen": 20840768,
      "step": 98745
    },
    {
      "epoch": 10.863586358635864,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.01535738426532429,
      "loss": 0.2324,
      "num_input_tokens_seen": 20841856,
      "step": 98750
    },
    {
      "epoch": 10.864136413641365,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01535594463182742,
      "loss": 0.2314,
      "num_input_tokens_seen": 20842912,
      "step": 98755
    },
    {
      "epoch": 10.864686468646864,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.015354504995049981,
      "loss": 0.2324,
      "num_input_tokens_seen": 20843968,
      "step": 98760
    },
    {
      "epoch": 10.865236523652365,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.015353065355005234,
      "loss": 0.2324,
      "num_input_tokens_seen": 20845024,
      "step": 98765
    },
    {
      "epoch": 10.865786578657866,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.015351625711706458,
      "loss": 0.2298,
      "num_input_tokens_seen": 20846080,
      "step": 98770
    },
    {
      "epoch": 10.866336633663366,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.015350186065166915,
      "loss": 0.2319,
      "num_input_tokens_seen": 20847168,
      "step": 98775
    },
    {
      "epoch": 10.866886688668867,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.015348746415399872,
      "loss": 0.2329,
      "num_input_tokens_seen": 20848224,
      "step": 98780
    },
    {
      "epoch": 10.867436743674368,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.015347306762418609,
      "loss": 0.2335,
      "num_input_tokens_seen": 20849248,
      "step": 98785
    },
    {
      "epoch": 10.867986798679867,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.015345867106236382,
      "loss": 0.2309,
      "num_input_tokens_seen": 20850368,
      "step": 98790
    },
    {
      "epoch": 10.868536853685368,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015344427446866458,
      "loss": 0.2308,
      "num_input_tokens_seen": 20851424,
      "step": 98795
    },
    {
      "epoch": 10.86908690869087,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.015342987784322117,
      "loss": 0.2314,
      "num_input_tokens_seen": 20852480,
      "step": 98800
    },
    {
      "epoch": 10.869636963696369,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01534154811861662,
      "loss": 0.2298,
      "num_input_tokens_seen": 20853504,
      "step": 98805
    },
    {
      "epoch": 10.87018701870187,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.015340108449763242,
      "loss": 0.2329,
      "num_input_tokens_seen": 20854528,
      "step": 98810
    },
    {
      "epoch": 10.870737073707371,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.015338668777775245,
      "loss": 0.2345,
      "num_input_tokens_seen": 20855552,
      "step": 98815
    },
    {
      "epoch": 10.871287128712872,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.015337229102665897,
      "loss": 0.2335,
      "num_input_tokens_seen": 20856672,
      "step": 98820
    },
    {
      "epoch": 10.871837183718371,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.015335789424448474,
      "loss": 0.232,
      "num_input_tokens_seen": 20857728,
      "step": 98825
    },
    {
      "epoch": 10.872387238723872,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.01533434974313624,
      "loss": 0.2304,
      "num_input_tokens_seen": 20858848,
      "step": 98830
    },
    {
      "epoch": 10.872937293729374,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.015332910058742464,
      "loss": 0.2335,
      "num_input_tokens_seen": 20859808,
      "step": 98835
    },
    {
      "epoch": 10.873487348734873,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.015331470371280418,
      "loss": 0.2309,
      "num_input_tokens_seen": 20860864,
      "step": 98840
    },
    {
      "epoch": 10.874037403740374,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.015330030680763369,
      "loss": 0.2299,
      "num_input_tokens_seen": 20861920,
      "step": 98845
    },
    {
      "epoch": 10.874587458745875,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.015328590987204583,
      "loss": 0.2308,
      "num_input_tokens_seen": 20862976,
      "step": 98850
    },
    {
      "epoch": 10.875137513751374,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.015327151290617335,
      "loss": 0.2329,
      "num_input_tokens_seen": 20864000,
      "step": 98855
    },
    {
      "epoch": 10.875687568756875,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.015325711591014889,
      "loss": 0.2319,
      "num_input_tokens_seen": 20865024,
      "step": 98860
    },
    {
      "epoch": 10.876237623762377,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.015324271888410509,
      "loss": 0.2335,
      "num_input_tokens_seen": 20866016,
      "step": 98865
    },
    {
      "epoch": 10.876787678767876,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01532283218281748,
      "loss": 0.2345,
      "num_input_tokens_seen": 20867008,
      "step": 98870
    },
    {
      "epoch": 10.877337733773377,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.015321392474249058,
      "loss": 0.2314,
      "num_input_tokens_seen": 20868064,
      "step": 98875
    },
    {
      "epoch": 10.877887788778878,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.015319952762718515,
      "loss": 0.2319,
      "num_input_tokens_seen": 20869216,
      "step": 98880
    },
    {
      "epoch": 10.87843784378438,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.015318513048239123,
      "loss": 0.2309,
      "num_input_tokens_seen": 20870336,
      "step": 98885
    },
    {
      "epoch": 10.878987898789878,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.015317073330824148,
      "loss": 0.2288,
      "num_input_tokens_seen": 20871360,
      "step": 98890
    },
    {
      "epoch": 10.87953795379538,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01531563361048686,
      "loss": 0.2308,
      "num_input_tokens_seen": 20872384,
      "step": 98895
    },
    {
      "epoch": 10.88008800880088,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.015314193887240528,
      "loss": 0.2314,
      "num_input_tokens_seen": 20873408,
      "step": 98900
    },
    {
      "epoch": 10.88063806380638,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.015312754161098426,
      "loss": 0.2308,
      "num_input_tokens_seen": 20874464,
      "step": 98905
    },
    {
      "epoch": 10.881188118811881,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.01531131443207381,
      "loss": 0.2303,
      "num_input_tokens_seen": 20875456,
      "step": 98910
    },
    {
      "epoch": 10.881738173817382,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.015309874700179965,
      "loss": 0.2324,
      "num_input_tokens_seen": 20876512,
      "step": 98915
    },
    {
      "epoch": 10.882288228822881,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.015308434965430153,
      "loss": 0.2308,
      "num_input_tokens_seen": 20877568,
      "step": 98920
    },
    {
      "epoch": 10.882838283828383,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.015306995227837643,
      "loss": 0.2298,
      "num_input_tokens_seen": 20878624,
      "step": 98925
    },
    {
      "epoch": 10.883388338833884,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.015305555487415705,
      "loss": 0.2314,
      "num_input_tokens_seen": 20879648,
      "step": 98930
    },
    {
      "epoch": 10.883938393839383,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.015304115744177612,
      "loss": 0.2319,
      "num_input_tokens_seen": 20880704,
      "step": 98935
    },
    {
      "epoch": 10.884488448844884,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.015302675998136624,
      "loss": 0.2303,
      "num_input_tokens_seen": 20881696,
      "step": 98940
    },
    {
      "epoch": 10.885038503850385,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.015301236249306018,
      "loss": 0.2319,
      "num_input_tokens_seen": 20882720,
      "step": 98945
    },
    {
      "epoch": 10.885588558855886,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015299796497699069,
      "loss": 0.2293,
      "num_input_tokens_seen": 20883744,
      "step": 98950
    },
    {
      "epoch": 10.886138613861386,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.015298356743329031,
      "loss": 0.2309,
      "num_input_tokens_seen": 20884800,
      "step": 98955
    },
    {
      "epoch": 10.886688668866887,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.015296916986209184,
      "loss": 0.2319,
      "num_input_tokens_seen": 20885888,
      "step": 98960
    },
    {
      "epoch": 10.887238723872388,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.0152954772263528,
      "loss": 0.2303,
      "num_input_tokens_seen": 20886944,
      "step": 98965
    },
    {
      "epoch": 10.887788778877887,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015294037463773138,
      "loss": 0.2288,
      "num_input_tokens_seen": 20887968,
      "step": 98970
    },
    {
      "epoch": 10.888338833883388,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.015292597698483481,
      "loss": 0.2288,
      "num_input_tokens_seen": 20889056,
      "step": 98975
    },
    {
      "epoch": 10.88888888888889,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.015291157930497087,
      "loss": 0.2303,
      "num_input_tokens_seen": 20890048,
      "step": 98980
    },
    {
      "epoch": 10.88943894389439,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015289718159827227,
      "loss": 0.2319,
      "num_input_tokens_seen": 20891104,
      "step": 98985
    },
    {
      "epoch": 10.88998899889989,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.015288278386487176,
      "loss": 0.2324,
      "num_input_tokens_seen": 20892096,
      "step": 98990
    },
    {
      "epoch": 10.89053905390539,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.015286838610490204,
      "loss": 0.2314,
      "num_input_tokens_seen": 20893184,
      "step": 98995
    },
    {
      "epoch": 10.891089108910892,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.015285398831849577,
      "loss": 0.2319,
      "num_input_tokens_seen": 20894304,
      "step": 99000
    },
    {
      "epoch": 10.891639163916391,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.015283959050578567,
      "loss": 0.2303,
      "num_input_tokens_seen": 20895392,
      "step": 99005
    },
    {
      "epoch": 10.892189218921892,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.015282519266690442,
      "loss": 0.2324,
      "num_input_tokens_seen": 20896448,
      "step": 99010
    },
    {
      "epoch": 10.892739273927393,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.015281079480198467,
      "loss": 0.2308,
      "num_input_tokens_seen": 20897504,
      "step": 99015
    },
    {
      "epoch": 10.893289328932893,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.015279639691115926,
      "loss": 0.2308,
      "num_input_tokens_seen": 20898560,
      "step": 99020
    },
    {
      "epoch": 10.893839383938394,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.015278199899456077,
      "loss": 0.2319,
      "num_input_tokens_seen": 20899648,
      "step": 99025
    },
    {
      "epoch": 10.894389438943895,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.015276760105232191,
      "loss": 0.2335,
      "num_input_tokens_seen": 20900736,
      "step": 99030
    },
    {
      "epoch": 10.894939493949394,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.015275320308457543,
      "loss": 0.2309,
      "num_input_tokens_seen": 20901792,
      "step": 99035
    },
    {
      "epoch": 10.895489548954895,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.015273880509145398,
      "loss": 0.2324,
      "num_input_tokens_seen": 20902816,
      "step": 99040
    },
    {
      "epoch": 10.896039603960396,
      "grad_norm": 0.00058746337890625,
      "learning_rate": 0.015272440707309029,
      "loss": 0.2329,
      "num_input_tokens_seen": 20903904,
      "step": 99045
    },
    {
      "epoch": 10.896589658965897,
      "grad_norm": 0.00081634521484375,
      "learning_rate": 0.015271000902961702,
      "loss": 0.2308,
      "num_input_tokens_seen": 20904896,
      "step": 99050
    },
    {
      "epoch": 10.897139713971397,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.015269561096116688,
      "loss": 0.2319,
      "num_input_tokens_seen": 20905952,
      "step": 99055
    },
    {
      "epoch": 10.897689768976898,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.015268121286787263,
      "loss": 0.2313,
      "num_input_tokens_seen": 20906944,
      "step": 99060
    },
    {
      "epoch": 10.898239823982399,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.015266681474986691,
      "loss": 0.2293,
      "num_input_tokens_seen": 20908000,
      "step": 99065
    },
    {
      "epoch": 10.898789878987898,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.015265241660728245,
      "loss": 0.235,
      "num_input_tokens_seen": 20908992,
      "step": 99070
    },
    {
      "epoch": 10.8993399339934,
      "grad_norm": 0.001953125,
      "learning_rate": 0.015263801844025196,
      "loss": 0.2298,
      "num_input_tokens_seen": 20910080,
      "step": 99075
    },
    {
      "epoch": 10.8998899889989,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.01526236202489081,
      "loss": 0.2319,
      "num_input_tokens_seen": 20911104,
      "step": 99080
    },
    {
      "epoch": 10.9004400440044,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.015260922203338355,
      "loss": 0.2329,
      "num_input_tokens_seen": 20912128,
      "step": 99085
    },
    {
      "epoch": 10.900990099009901,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01525948237938111,
      "loss": 0.2319,
      "num_input_tokens_seen": 20913184,
      "step": 99090
    },
    {
      "epoch": 10.901540154015402,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.015258042553032342,
      "loss": 0.2324,
      "num_input_tokens_seen": 20914272,
      "step": 99095
    },
    {
      "epoch": 10.902090209020901,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.015256602724305313,
      "loss": 0.2313,
      "num_input_tokens_seen": 20915296,
      "step": 99100
    },
    {
      "epoch": 10.902640264026402,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.015255162893213306,
      "loss": 0.2319,
      "num_input_tokens_seen": 20916320,
      "step": 99105
    },
    {
      "epoch": 10.903190319031903,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.015253723059769584,
      "loss": 0.2293,
      "num_input_tokens_seen": 20917376,
      "step": 99110
    },
    {
      "epoch": 10.903740374037405,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.015252283223987417,
      "loss": 0.2319,
      "num_input_tokens_seen": 20918400,
      "step": 99115
    },
    {
      "epoch": 10.904290429042904,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.015250843385880076,
      "loss": 0.2324,
      "num_input_tokens_seen": 20919488,
      "step": 99120
    },
    {
      "epoch": 10.904840484048405,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.015249403545460833,
      "loss": 0.2329,
      "num_input_tokens_seen": 20920544,
      "step": 99125
    },
    {
      "epoch": 10.905390539053906,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.015247963702742956,
      "loss": 0.2319,
      "num_input_tokens_seen": 20921600,
      "step": 99130
    },
    {
      "epoch": 10.905940594059405,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015246523857739716,
      "loss": 0.2319,
      "num_input_tokens_seen": 20922688,
      "step": 99135
    },
    {
      "epoch": 10.906490649064907,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.015245084010464384,
      "loss": 0.2319,
      "num_input_tokens_seen": 20923744,
      "step": 99140
    },
    {
      "epoch": 10.907040704070408,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.015243644160930233,
      "loss": 0.2314,
      "num_input_tokens_seen": 20924768,
      "step": 99145
    },
    {
      "epoch": 10.907590759075907,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01524220430915053,
      "loss": 0.2319,
      "num_input_tokens_seen": 20925824,
      "step": 99150
    },
    {
      "epoch": 10.908140814081408,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.015240764455138544,
      "loss": 0.2298,
      "num_input_tokens_seen": 20926848,
      "step": 99155
    },
    {
      "epoch": 10.908690869086909,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.015239324598907545,
      "loss": 0.2298,
      "num_input_tokens_seen": 20927904,
      "step": 99160
    },
    {
      "epoch": 10.909240924092408,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015237884740470815,
      "loss": 0.2319,
      "num_input_tokens_seen": 20928960,
      "step": 99165
    },
    {
      "epoch": 10.90979097909791,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.01523644487984161,
      "loss": 0.233,
      "num_input_tokens_seen": 20930080,
      "step": 99170
    },
    {
      "epoch": 10.91034103410341,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.015235005017033206,
      "loss": 0.2335,
      "num_input_tokens_seen": 20931136,
      "step": 99175
    },
    {
      "epoch": 10.910891089108912,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.015233565152058872,
      "loss": 0.2298,
      "num_input_tokens_seen": 20932160,
      "step": 99180
    },
    {
      "epoch": 10.911441144114411,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015232125284931883,
      "loss": 0.2314,
      "num_input_tokens_seen": 20933184,
      "step": 99185
    },
    {
      "epoch": 10.911991199119912,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.015230685415665505,
      "loss": 0.2283,
      "num_input_tokens_seen": 20934240,
      "step": 99190
    },
    {
      "epoch": 10.912541254125413,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.015229245544273011,
      "loss": 0.2329,
      "num_input_tokens_seen": 20935328,
      "step": 99195
    },
    {
      "epoch": 10.913091309130913,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.015227805670767668,
      "loss": 0.2319,
      "num_input_tokens_seen": 20936448,
      "step": 99200
    },
    {
      "epoch": 10.913641364136414,
      "grad_norm": 0.000568389892578125,
      "learning_rate": 0.015226365795162754,
      "loss": 0.2319,
      "num_input_tokens_seen": 20937504,
      "step": 99205
    },
    {
      "epoch": 10.914191419141915,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.015224925917471537,
      "loss": 0.2313,
      "num_input_tokens_seen": 20938592,
      "step": 99210
    },
    {
      "epoch": 10.914741474147414,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.01522348603770728,
      "loss": 0.2309,
      "num_input_tokens_seen": 20939680,
      "step": 99215
    },
    {
      "epoch": 10.915291529152915,
      "grad_norm": 0.000591278076171875,
      "learning_rate": 0.01522204615588326,
      "loss": 0.2314,
      "num_input_tokens_seen": 20940704,
      "step": 99220
    },
    {
      "epoch": 10.915841584158416,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.015220606272012749,
      "loss": 0.2303,
      "num_input_tokens_seen": 20941760,
      "step": 99225
    },
    {
      "epoch": 10.916391639163916,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.015219166386109012,
      "loss": 0.2303,
      "num_input_tokens_seen": 20942752,
      "step": 99230
    },
    {
      "epoch": 10.916941694169417,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.015217726498185329,
      "loss": 0.2314,
      "num_input_tokens_seen": 20943776,
      "step": 99235
    },
    {
      "epoch": 10.917491749174918,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015216286608254964,
      "loss": 0.2324,
      "num_input_tokens_seen": 20944832,
      "step": 99240
    },
    {
      "epoch": 10.918041804180419,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.015214846716331188,
      "loss": 0.2319,
      "num_input_tokens_seen": 20945824,
      "step": 99245
    },
    {
      "epoch": 10.918591859185918,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.015213406822427271,
      "loss": 0.2308,
      "num_input_tokens_seen": 20946912,
      "step": 99250
    },
    {
      "epoch": 10.91914191419142,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.015211966926556488,
      "loss": 0.2314,
      "num_input_tokens_seen": 20947936,
      "step": 99255
    },
    {
      "epoch": 10.91969196919692,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.015210527028732108,
      "loss": 0.2309,
      "num_input_tokens_seen": 20948928,
      "step": 99260
    },
    {
      "epoch": 10.92024202420242,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.015209087128967404,
      "loss": 0.2309,
      "num_input_tokens_seen": 20949984,
      "step": 99265
    },
    {
      "epoch": 10.92079207920792,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01520764722727564,
      "loss": 0.2324,
      "num_input_tokens_seen": 20951008,
      "step": 99270
    },
    {
      "epoch": 10.921342134213422,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015206207323670088,
      "loss": 0.233,
      "num_input_tokens_seen": 20952096,
      "step": 99275
    },
    {
      "epoch": 10.921892189218921,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.01520476741816403,
      "loss": 0.2319,
      "num_input_tokens_seen": 20953088,
      "step": 99280
    },
    {
      "epoch": 10.922442244224422,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.015203327510770727,
      "loss": 0.2314,
      "num_input_tokens_seen": 20954144,
      "step": 99285
    },
    {
      "epoch": 10.922992299229923,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.015201887601503447,
      "loss": 0.2319,
      "num_input_tokens_seen": 20955264,
      "step": 99290
    },
    {
      "epoch": 10.923542354235423,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01520044769037547,
      "loss": 0.2298,
      "num_input_tokens_seen": 20956320,
      "step": 99295
    },
    {
      "epoch": 10.924092409240924,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015199007777400061,
      "loss": 0.2314,
      "num_input_tokens_seen": 20957408,
      "step": 99300
    },
    {
      "epoch": 10.924642464246425,
      "grad_norm": 0.000537872314453125,
      "learning_rate": 0.015197567862590494,
      "loss": 0.2298,
      "num_input_tokens_seen": 20958432,
      "step": 99305
    },
    {
      "epoch": 10.925192519251926,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.015196127945960041,
      "loss": 0.2309,
      "num_input_tokens_seen": 20959520,
      "step": 99310
    },
    {
      "epoch": 10.925742574257425,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.015194688027521963,
      "loss": 0.2304,
      "num_input_tokens_seen": 20960608,
      "step": 99315
    },
    {
      "epoch": 10.926292629262926,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.015193248107289545,
      "loss": 0.2319,
      "num_input_tokens_seen": 20961664,
      "step": 99320
    },
    {
      "epoch": 10.926842684268427,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015191808185276054,
      "loss": 0.2314,
      "num_input_tokens_seen": 20962688,
      "step": 99325
    },
    {
      "epoch": 10.927392739273927,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015190368261494754,
      "loss": 0.2303,
      "num_input_tokens_seen": 20963808,
      "step": 99330
    },
    {
      "epoch": 10.927942794279428,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.015188928335958925,
      "loss": 0.2324,
      "num_input_tokens_seen": 20964896,
      "step": 99335
    },
    {
      "epoch": 10.928492849284929,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.015187488408681834,
      "loss": 0.2314,
      "num_input_tokens_seen": 20965952,
      "step": 99340
    },
    {
      "epoch": 10.929042904290428,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015186048479676745,
      "loss": 0.2298,
      "num_input_tokens_seen": 20967008,
      "step": 99345
    },
    {
      "epoch": 10.92959295929593,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.015184608548956943,
      "loss": 0.2329,
      "num_input_tokens_seen": 20968064,
      "step": 99350
    },
    {
      "epoch": 10.93014301430143,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.015183168616535696,
      "loss": 0.2309,
      "num_input_tokens_seen": 20969152,
      "step": 99355
    },
    {
      "epoch": 10.930693069306932,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.015181728682426265,
      "loss": 0.2329,
      "num_input_tokens_seen": 20970176,
      "step": 99360
    },
    {
      "epoch": 10.93124312431243,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.015180288746641929,
      "loss": 0.2345,
      "num_input_tokens_seen": 20971200,
      "step": 99365
    },
    {
      "epoch": 10.931793179317932,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01517884880919596,
      "loss": 0.2309,
      "num_input_tokens_seen": 20972224,
      "step": 99370
    },
    {
      "epoch": 10.932343234323433,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015177408870101626,
      "loss": 0.2309,
      "num_input_tokens_seen": 20973216,
      "step": 99375
    },
    {
      "epoch": 10.932893289328932,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.015175968929372202,
      "loss": 0.2314,
      "num_input_tokens_seen": 20974304,
      "step": 99380
    },
    {
      "epoch": 10.933443344334433,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.015174528987020958,
      "loss": 0.2309,
      "num_input_tokens_seen": 20975328,
      "step": 99385
    },
    {
      "epoch": 10.933993399339935,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015173089043061158,
      "loss": 0.2319,
      "num_input_tokens_seen": 20976352,
      "step": 99390
    },
    {
      "epoch": 10.934543454345434,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015171649097506082,
      "loss": 0.2314,
      "num_input_tokens_seen": 20977472,
      "step": 99395
    },
    {
      "epoch": 10.935093509350935,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.015170209150369004,
      "loss": 0.2319,
      "num_input_tokens_seen": 20978496,
      "step": 99400
    },
    {
      "epoch": 10.935643564356436,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015168769201663181,
      "loss": 0.2298,
      "num_input_tokens_seen": 20979552,
      "step": 99405
    },
    {
      "epoch": 10.936193619361937,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.015167329251401898,
      "loss": 0.2329,
      "num_input_tokens_seen": 20980576,
      "step": 99410
    },
    {
      "epoch": 10.936743674367436,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.015165889299598423,
      "loss": 0.2319,
      "num_input_tokens_seen": 20981664,
      "step": 99415
    },
    {
      "epoch": 10.937293729372938,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.015164449346266021,
      "loss": 0.2308,
      "num_input_tokens_seen": 20982752,
      "step": 99420
    },
    {
      "epoch": 10.937843784378439,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.015163009391417974,
      "loss": 0.2308,
      "num_input_tokens_seen": 20983712,
      "step": 99425
    },
    {
      "epoch": 10.938393839383938,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.015161569435067544,
      "loss": 0.2335,
      "num_input_tokens_seen": 20984768,
      "step": 99430
    },
    {
      "epoch": 10.938943894389439,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.015160129477228005,
      "loss": 0.2293,
      "num_input_tokens_seen": 20985792,
      "step": 99435
    },
    {
      "epoch": 10.93949394939494,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.015158689517912632,
      "loss": 0.2308,
      "num_input_tokens_seen": 20986784,
      "step": 99440
    },
    {
      "epoch": 10.94004400440044,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.015157249557134692,
      "loss": 0.2298,
      "num_input_tokens_seen": 20987872,
      "step": 99445
    },
    {
      "epoch": 10.94059405940594,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.015155809594907463,
      "loss": 0.2303,
      "num_input_tokens_seen": 20988928,
      "step": 99450
    },
    {
      "epoch": 10.941144114411442,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.015154369631244209,
      "loss": 0.2293,
      "num_input_tokens_seen": 20989952,
      "step": 99455
    },
    {
      "epoch": 10.941694169416941,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.015152929666158202,
      "loss": 0.2314,
      "num_input_tokens_seen": 20990944,
      "step": 99460
    },
    {
      "epoch": 10.942244224422442,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.015151489699662716,
      "loss": 0.2314,
      "num_input_tokens_seen": 20992064,
      "step": 99465
    },
    {
      "epoch": 10.942794279427943,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015150049731771024,
      "loss": 0.2293,
      "num_input_tokens_seen": 20993088,
      "step": 99470
    },
    {
      "epoch": 10.943344334433444,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.015148609762496395,
      "loss": 0.2283,
      "num_input_tokens_seen": 20994208,
      "step": 99475
    },
    {
      "epoch": 10.943894389438944,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.015147169791852102,
      "loss": 0.2309,
      "num_input_tokens_seen": 20995296,
      "step": 99480
    },
    {
      "epoch": 10.944444444444445,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.015145729819851413,
      "loss": 0.2319,
      "num_input_tokens_seen": 20996384,
      "step": 99485
    },
    {
      "epoch": 10.944994499449946,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.015144289846507605,
      "loss": 0.2303,
      "num_input_tokens_seen": 20997440,
      "step": 99490
    },
    {
      "epoch": 10.945544554455445,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.015142849871833945,
      "loss": 0.2324,
      "num_input_tokens_seen": 20998592,
      "step": 99495
    },
    {
      "epoch": 10.946094609460946,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.015141409895843708,
      "loss": 0.2324,
      "num_input_tokens_seen": 20999584,
      "step": 99500
    },
    {
      "epoch": 10.946644664466447,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01513996991855016,
      "loss": 0.2329,
      "num_input_tokens_seen": 21000608,
      "step": 99505
    },
    {
      "epoch": 10.947194719471947,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.01513852993996658,
      "loss": 0.2319,
      "num_input_tokens_seen": 21001696,
      "step": 99510
    },
    {
      "epoch": 10.947744774477448,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.015137089960106235,
      "loss": 0.234,
      "num_input_tokens_seen": 21002816,
      "step": 99515
    },
    {
      "epoch": 10.948294829482949,
      "grad_norm": 0.009765625,
      "learning_rate": 0.015135649978982398,
      "loss": 0.2309,
      "num_input_tokens_seen": 21003872,
      "step": 99520
    },
    {
      "epoch": 10.948844884488448,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01513420999660834,
      "loss": 0.2303,
      "num_input_tokens_seen": 21004896,
      "step": 99525
    },
    {
      "epoch": 10.94939493949395,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.015132770012997333,
      "loss": 0.2319,
      "num_input_tokens_seen": 21005952,
      "step": 99530
    },
    {
      "epoch": 10.94994499449945,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.015131330028162646,
      "loss": 0.233,
      "num_input_tokens_seen": 21007008,
      "step": 99535
    },
    {
      "epoch": 10.950495049504951,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.015129890042117556,
      "loss": 0.2314,
      "num_input_tokens_seen": 21008032,
      "step": 99540
    },
    {
      "epoch": 10.95104510451045,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.015128450054875335,
      "loss": 0.2303,
      "num_input_tokens_seen": 21009056,
      "step": 99545
    },
    {
      "epoch": 10.951595159515952,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015127010066449247,
      "loss": 0.2303,
      "num_input_tokens_seen": 21010048,
      "step": 99550
    },
    {
      "epoch": 10.952145214521453,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.015125570076852566,
      "loss": 0.2303,
      "num_input_tokens_seen": 21011072,
      "step": 99555
    },
    {
      "epoch": 10.952695269526952,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.015124130086098569,
      "loss": 0.2319,
      "num_input_tokens_seen": 21012096,
      "step": 99560
    },
    {
      "epoch": 10.953245324532453,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.015122690094200524,
      "loss": 0.2319,
      "num_input_tokens_seen": 21013184,
      "step": 99565
    },
    {
      "epoch": 10.953795379537954,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.015121250101171706,
      "loss": 0.2283,
      "num_input_tokens_seen": 21014208,
      "step": 99570
    },
    {
      "epoch": 10.954345434543454,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01511981010702538,
      "loss": 0.2303,
      "num_input_tokens_seen": 21015264,
      "step": 99575
    },
    {
      "epoch": 10.954895489548955,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.015118370111774821,
      "loss": 0.2319,
      "num_input_tokens_seen": 21016288,
      "step": 99580
    },
    {
      "epoch": 10.955445544554456,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.015116930115433306,
      "loss": 0.2308,
      "num_input_tokens_seen": 21017280,
      "step": 99585
    },
    {
      "epoch": 10.955995599559955,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0151154901180141,
      "loss": 0.2303,
      "num_input_tokens_seen": 21018272,
      "step": 99590
    },
    {
      "epoch": 10.956545654565456,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01511405011953048,
      "loss": 0.2309,
      "num_input_tokens_seen": 21019232,
      "step": 99595
    },
    {
      "epoch": 10.957095709570957,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01511261011999571,
      "loss": 0.2319,
      "num_input_tokens_seen": 21020288,
      "step": 99600
    },
    {
      "epoch": 10.957645764576458,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.015111170119423069,
      "loss": 0.2319,
      "num_input_tokens_seen": 21021344,
      "step": 99605
    },
    {
      "epoch": 10.958195819581958,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.015109730117825822,
      "loss": 0.2324,
      "num_input_tokens_seen": 21022400,
      "step": 99610
    },
    {
      "epoch": 10.958745874587459,
      "grad_norm": 0.00058746337890625,
      "learning_rate": 0.015108290115217254,
      "loss": 0.2319,
      "num_input_tokens_seen": 21023424,
      "step": 99615
    },
    {
      "epoch": 10.95929592959296,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.015106850111610622,
      "loss": 0.2319,
      "num_input_tokens_seen": 21024480,
      "step": 99620
    },
    {
      "epoch": 10.95984598459846,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.015105410107019202,
      "loss": 0.2303,
      "num_input_tokens_seen": 21025568,
      "step": 99625
    },
    {
      "epoch": 10.96039603960396,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01510397010145627,
      "loss": 0.2319,
      "num_input_tokens_seen": 21026656,
      "step": 99630
    },
    {
      "epoch": 10.960946094609461,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.0151025300949351,
      "loss": 0.2319,
      "num_input_tokens_seen": 21027680,
      "step": 99635
    },
    {
      "epoch": 10.96149614961496,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.015101090087468953,
      "loss": 0.2309,
      "num_input_tokens_seen": 21028768,
      "step": 99640
    },
    {
      "epoch": 10.962046204620462,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.01509965007907111,
      "loss": 0.233,
      "num_input_tokens_seen": 21029792,
      "step": 99645
    },
    {
      "epoch": 10.962596259625963,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.015098210069754836,
      "loss": 0.233,
      "num_input_tokens_seen": 21030848,
      "step": 99650
    },
    {
      "epoch": 10.963146314631462,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.015096770059533412,
      "loss": 0.2309,
      "num_input_tokens_seen": 21031904,
      "step": 99655
    },
    {
      "epoch": 10.963696369636963,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.015095330048420104,
      "loss": 0.2309,
      "num_input_tokens_seen": 21032896,
      "step": 99660
    },
    {
      "epoch": 10.964246424642464,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01509389003642818,
      "loss": 0.2309,
      "num_input_tokens_seen": 21033952,
      "step": 99665
    },
    {
      "epoch": 10.964796479647966,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.015092450023570923,
      "loss": 0.2314,
      "num_input_tokens_seen": 21034976,
      "step": 99670
    },
    {
      "epoch": 10.965346534653465,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.015091010009861596,
      "loss": 0.2314,
      "num_input_tokens_seen": 21036000,
      "step": 99675
    },
    {
      "epoch": 10.965896589658966,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.015089569995313471,
      "loss": 0.2299,
      "num_input_tokens_seen": 21037024,
      "step": 99680
    },
    {
      "epoch": 10.966446644664467,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.015088129979939828,
      "loss": 0.2303,
      "num_input_tokens_seen": 21038144,
      "step": 99685
    },
    {
      "epoch": 10.966996699669966,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.015086689963753931,
      "loss": 0.2309,
      "num_input_tokens_seen": 21039168,
      "step": 99690
    },
    {
      "epoch": 10.967546754675467,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.015085249946769049,
      "loss": 0.2303,
      "num_input_tokens_seen": 21040224,
      "step": 99695
    },
    {
      "epoch": 10.968096809680969,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015083809928998462,
      "loss": 0.2329,
      "num_input_tokens_seen": 21041248,
      "step": 99700
    },
    {
      "epoch": 10.968646864686468,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01508236991045544,
      "loss": 0.2324,
      "num_input_tokens_seen": 21042400,
      "step": 99705
    },
    {
      "epoch": 10.969196919691969,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.015080929891153255,
      "loss": 0.2324,
      "num_input_tokens_seen": 21043424,
      "step": 99710
    },
    {
      "epoch": 10.96974697469747,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.015079489871105177,
      "loss": 0.233,
      "num_input_tokens_seen": 21044512,
      "step": 99715
    },
    {
      "epoch": 10.97029702970297,
      "grad_norm": 0.009765625,
      "learning_rate": 0.015078049850324482,
      "loss": 0.2314,
      "num_input_tokens_seen": 21045536,
      "step": 99720
    },
    {
      "epoch": 10.97084708470847,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.015076609828824432,
      "loss": 0.2308,
      "num_input_tokens_seen": 21046656,
      "step": 99725
    },
    {
      "epoch": 10.971397139713972,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.015075169806618311,
      "loss": 0.2299,
      "num_input_tokens_seen": 21047712,
      "step": 99730
    },
    {
      "epoch": 10.971947194719473,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.01507372978371939,
      "loss": 0.233,
      "num_input_tokens_seen": 21048800,
      "step": 99735
    },
    {
      "epoch": 10.972497249724972,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01507228976014093,
      "loss": 0.2299,
      "num_input_tokens_seen": 21049824,
      "step": 99740
    },
    {
      "epoch": 10.973047304730473,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.015070849735896211,
      "loss": 0.2293,
      "num_input_tokens_seen": 21050848,
      "step": 99745
    },
    {
      "epoch": 10.973597359735974,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.015069409710998507,
      "loss": 0.2335,
      "num_input_tokens_seen": 21051872,
      "step": 99750
    },
    {
      "epoch": 10.974147414741473,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.015067969685461082,
      "loss": 0.2324,
      "num_input_tokens_seen": 21052960,
      "step": 99755
    },
    {
      "epoch": 10.974697469746975,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.01506652965929722,
      "loss": 0.2319,
      "num_input_tokens_seen": 21054048,
      "step": 99760
    },
    {
      "epoch": 10.975247524752476,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.015065089632520182,
      "loss": 0.2314,
      "num_input_tokens_seen": 21055136,
      "step": 99765
    },
    {
      "epoch": 10.975797579757975,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.015063649605143243,
      "loss": 0.2293,
      "num_input_tokens_seen": 21056224,
      "step": 99770
    },
    {
      "epoch": 10.976347634763476,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.015062209577179677,
      "loss": 0.2319,
      "num_input_tokens_seen": 21057248,
      "step": 99775
    },
    {
      "epoch": 10.976897689768977,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.015060769548642756,
      "loss": 0.2314,
      "num_input_tokens_seen": 21058368,
      "step": 99780
    },
    {
      "epoch": 10.977447744774478,
      "grad_norm": 0.0005340576171875,
      "learning_rate": 0.01505932951954575,
      "loss": 0.2309,
      "num_input_tokens_seen": 21059360,
      "step": 99785
    },
    {
      "epoch": 10.977997799779978,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.015057889489901935,
      "loss": 0.234,
      "num_input_tokens_seen": 21060448,
      "step": 99790
    },
    {
      "epoch": 10.978547854785479,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.015056449459724579,
      "loss": 0.233,
      "num_input_tokens_seen": 21061504,
      "step": 99795
    },
    {
      "epoch": 10.97909790979098,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.015055009429026952,
      "loss": 0.233,
      "num_input_tokens_seen": 21062560,
      "step": 99800
    },
    {
      "epoch": 10.979647964796479,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.015053569397822335,
      "loss": 0.2309,
      "num_input_tokens_seen": 21063680,
      "step": 99805
    },
    {
      "epoch": 10.98019801980198,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.01505212936612399,
      "loss": 0.2335,
      "num_input_tokens_seen": 21064736,
      "step": 99810
    },
    {
      "epoch": 10.980748074807481,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.015050689333945195,
      "loss": 0.2309,
      "num_input_tokens_seen": 21065824,
      "step": 99815
    },
    {
      "epoch": 10.98129812981298,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.015049249301299222,
      "loss": 0.2294,
      "num_input_tokens_seen": 21066816,
      "step": 99820
    },
    {
      "epoch": 10.981848184818482,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01504780926819934,
      "loss": 0.2293,
      "num_input_tokens_seen": 21067840,
      "step": 99825
    },
    {
      "epoch": 10.982398239823983,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.015046369234658826,
      "loss": 0.2309,
      "num_input_tokens_seen": 21068928,
      "step": 99830
    },
    {
      "epoch": 10.982948294829484,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.015044929200690949,
      "loss": 0.2314,
      "num_input_tokens_seen": 21069952,
      "step": 99835
    },
    {
      "epoch": 10.983498349834983,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.015043489166308974,
      "loss": 0.2314,
      "num_input_tokens_seen": 21070976,
      "step": 99840
    },
    {
      "epoch": 10.984048404840484,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.015042049131526186,
      "loss": 0.2309,
      "num_input_tokens_seen": 21072096,
      "step": 99845
    },
    {
      "epoch": 10.984598459845985,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.01504060909635585,
      "loss": 0.2314,
      "num_input_tokens_seen": 21073184,
      "step": 99850
    },
    {
      "epoch": 10.985148514851485,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.015039169060811243,
      "loss": 0.2309,
      "num_input_tokens_seen": 21074208,
      "step": 99855
    },
    {
      "epoch": 10.985698569856986,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.015037729024905633,
      "loss": 0.234,
      "num_input_tokens_seen": 21075264,
      "step": 99860
    },
    {
      "epoch": 10.986248624862487,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.01503628898865229,
      "loss": 0.2314,
      "num_input_tokens_seen": 21076352,
      "step": 99865
    },
    {
      "epoch": 10.986798679867986,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.015034848952064486,
      "loss": 0.2308,
      "num_input_tokens_seen": 21077408,
      "step": 99870
    },
    {
      "epoch": 10.987348734873487,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.015033408915155502,
      "loss": 0.2324,
      "num_input_tokens_seen": 21078464,
      "step": 99875
    },
    {
      "epoch": 10.987898789878988,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.015031968877938602,
      "loss": 0.2314,
      "num_input_tokens_seen": 21079584,
      "step": 99880
    },
    {
      "epoch": 10.988448844884488,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.015030528840427056,
      "loss": 0.2303,
      "num_input_tokens_seen": 21080672,
      "step": 99885
    },
    {
      "epoch": 10.988998899889989,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.015029088802634146,
      "loss": 0.2314,
      "num_input_tokens_seen": 21081824,
      "step": 99890
    },
    {
      "epoch": 10.98954895489549,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015027648764573137,
      "loss": 0.2309,
      "num_input_tokens_seen": 21082848,
      "step": 99895
    },
    {
      "epoch": 10.990099009900991,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.015026208726257302,
      "loss": 0.2319,
      "num_input_tokens_seen": 21083904,
      "step": 99900
    },
    {
      "epoch": 10.99064906490649,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.015024768687699912,
      "loss": 0.2319,
      "num_input_tokens_seen": 21084928,
      "step": 99905
    },
    {
      "epoch": 10.991199119911991,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.015023328648914243,
      "loss": 0.2308,
      "num_input_tokens_seen": 21086016,
      "step": 99910
    },
    {
      "epoch": 10.991749174917492,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.015021888609913563,
      "loss": 0.2314,
      "num_input_tokens_seen": 21087040,
      "step": 99915
    },
    {
      "epoch": 10.992299229922992,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.015020448570711145,
      "loss": 0.2303,
      "num_input_tokens_seen": 21088064,
      "step": 99920
    },
    {
      "epoch": 10.992849284928493,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.015019008531320268,
      "loss": 0.2329,
      "num_input_tokens_seen": 21089152,
      "step": 99925
    },
    {
      "epoch": 10.993399339933994,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.015017568491754195,
      "loss": 0.2319,
      "num_input_tokens_seen": 21090240,
      "step": 99930
    },
    {
      "epoch": 10.993949394939493,
      "grad_norm": 0.000637054443359375,
      "learning_rate": 0.015016128452026203,
      "loss": 0.2319,
      "num_input_tokens_seen": 21091296,
      "step": 99935
    },
    {
      "epoch": 10.994499449944994,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.015014688412149563,
      "loss": 0.2319,
      "num_input_tokens_seen": 21092416,
      "step": 99940
    },
    {
      "epoch": 10.995049504950495,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.015013248372137541,
      "loss": 0.2309,
      "num_input_tokens_seen": 21093472,
      "step": 99945
    },
    {
      "epoch": 10.995599559955995,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.015011808332003425,
      "loss": 0.2313,
      "num_input_tokens_seen": 21094432,
      "step": 99950
    },
    {
      "epoch": 10.996149614961496,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.01501036829176047,
      "loss": 0.2303,
      "num_input_tokens_seen": 21095552,
      "step": 99955
    },
    {
      "epoch": 10.996699669966997,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.015008928251421956,
      "loss": 0.2319,
      "num_input_tokens_seen": 21096576,
      "step": 99960
    },
    {
      "epoch": 10.997249724972498,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.015007488211001159,
      "loss": 0.2288,
      "num_input_tokens_seen": 21097696,
      "step": 99965
    },
    {
      "epoch": 10.997799779977997,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.015006048170511346,
      "loss": 0.2325,
      "num_input_tokens_seen": 21098784,
      "step": 99970
    },
    {
      "epoch": 10.998349834983498,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.015004608129965787,
      "loss": 0.2303,
      "num_input_tokens_seen": 21099840,
      "step": 99975
    },
    {
      "epoch": 10.998899889989,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.01500316808937776,
      "loss": 0.2298,
      "num_input_tokens_seen": 21100864,
      "step": 99980
    },
    {
      "epoch": 10.999449944994499,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.015001728048760527,
      "loss": 0.2319,
      "num_input_tokens_seen": 21101952,
      "step": 99985
    },
    {
      "epoch": 11.0,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.015000288008127373,
      "loss": 0.2319,
      "num_input_tokens_seen": 21102912,
      "step": 99990
    },
    {
      "epoch": 11.0,
      "eval_loss": 0.23138663172721863,
      "eval_runtime": 60.5774,
      "eval_samples_per_second": 66.692,
      "eval_steps_per_second": 16.673,
      "num_input_tokens_seen": 21102912,
      "step": 99990
    },
    {
      "epoch": 11.000550055005501,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.014998847967491565,
      "loss": 0.2319,
      "num_input_tokens_seen": 21103936,
      "step": 99995
    },
    {
      "epoch": 11.001100110011,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014997407926866374,
      "loss": 0.2283,
      "num_input_tokens_seen": 21105024,
      "step": 100000
    },
    {
      "epoch": 11.001650165016502,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01499596788626507,
      "loss": 0.2324,
      "num_input_tokens_seen": 21106016,
      "step": 100005
    },
    {
      "epoch": 11.002200220022003,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.014994527845700935,
      "loss": 0.2298,
      "num_input_tokens_seen": 21107136,
      "step": 100010
    },
    {
      "epoch": 11.002750275027502,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01499308780518723,
      "loss": 0.2329,
      "num_input_tokens_seen": 21108192,
      "step": 100015
    },
    {
      "epoch": 11.003300330033003,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.01499164776473723,
      "loss": 0.2303,
      "num_input_tokens_seen": 21109248,
      "step": 100020
    },
    {
      "epoch": 11.003850385038504,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.014990207724364213,
      "loss": 0.2303,
      "num_input_tokens_seen": 21110368,
      "step": 100025
    },
    {
      "epoch": 11.004400440044005,
      "grad_norm": 0.002593994140625,
      "learning_rate": 0.014988767684081444,
      "loss": 0.2324,
      "num_input_tokens_seen": 21111392,
      "step": 100030
    },
    {
      "epoch": 11.004950495049505,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.014987327643902195,
      "loss": 0.2314,
      "num_input_tokens_seen": 21112512,
      "step": 100035
    },
    {
      "epoch": 11.005500550055006,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014985887603839747,
      "loss": 0.2324,
      "num_input_tokens_seen": 21113632,
      "step": 100040
    },
    {
      "epoch": 11.006050605060507,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.014984447563907356,
      "loss": 0.2319,
      "num_input_tokens_seen": 21114720,
      "step": 100045
    },
    {
      "epoch": 11.006600660066006,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.014983007524118316,
      "loss": 0.2324,
      "num_input_tokens_seen": 21115776,
      "step": 100050
    },
    {
      "epoch": 11.007150715071507,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.014981567484485883,
      "loss": 0.2303,
      "num_input_tokens_seen": 21116864,
      "step": 100055
    },
    {
      "epoch": 11.007700770077008,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.014980127445023332,
      "loss": 0.2304,
      "num_input_tokens_seen": 21117952,
      "step": 100060
    },
    {
      "epoch": 11.008250825082508,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01497868740574394,
      "loss": 0.2314,
      "num_input_tokens_seen": 21119008,
      "step": 100065
    },
    {
      "epoch": 11.008800880088009,
      "grad_norm": 0.000621795654296875,
      "learning_rate": 0.014977247366660972,
      "loss": 0.2293,
      "num_input_tokens_seen": 21120032,
      "step": 100070
    },
    {
      "epoch": 11.00935093509351,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014975807327787706,
      "loss": 0.2335,
      "num_input_tokens_seen": 21121120,
      "step": 100075
    },
    {
      "epoch": 11.009900990099009,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014974367289137416,
      "loss": 0.2335,
      "num_input_tokens_seen": 21122176,
      "step": 100080
    },
    {
      "epoch": 11.01045104510451,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.014972927250723364,
      "loss": 0.2303,
      "num_input_tokens_seen": 21123168,
      "step": 100085
    },
    {
      "epoch": 11.011001100110011,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014971487212558832,
      "loss": 0.2324,
      "num_input_tokens_seen": 21124192,
      "step": 100090
    },
    {
      "epoch": 11.011551155115512,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.014970047174657086,
      "loss": 0.2324,
      "num_input_tokens_seen": 21125280,
      "step": 100095
    },
    {
      "epoch": 11.012101210121012,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014968607137031402,
      "loss": 0.2314,
      "num_input_tokens_seen": 21126336,
      "step": 100100
    },
    {
      "epoch": 11.012651265126513,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.014967167099695056,
      "loss": 0.2298,
      "num_input_tokens_seen": 21127360,
      "step": 100105
    },
    {
      "epoch": 11.013201320132014,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.014965727062661308,
      "loss": 0.2308,
      "num_input_tokens_seen": 21128480,
      "step": 100110
    },
    {
      "epoch": 11.013751375137513,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.014964287025943442,
      "loss": 0.2308,
      "num_input_tokens_seen": 21129664,
      "step": 100115
    },
    {
      "epoch": 11.014301430143014,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01496284698955472,
      "loss": 0.2314,
      "num_input_tokens_seen": 21130720,
      "step": 100120
    },
    {
      "epoch": 11.014851485148515,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.014961406953508427,
      "loss": 0.2314,
      "num_input_tokens_seen": 21131712,
      "step": 100125
    },
    {
      "epoch": 11.015401540154015,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014959966917817823,
      "loss": 0.2335,
      "num_input_tokens_seen": 21132736,
      "step": 100130
    },
    {
      "epoch": 11.015951595159516,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.014958526882496184,
      "loss": 0.2298,
      "num_input_tokens_seen": 21133824,
      "step": 100135
    },
    {
      "epoch": 11.016501650165017,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.014957086847556787,
      "loss": 0.2314,
      "num_input_tokens_seen": 21134848,
      "step": 100140
    },
    {
      "epoch": 11.017051705170518,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.014955646813012894,
      "loss": 0.2303,
      "num_input_tokens_seen": 21135936,
      "step": 100145
    },
    {
      "epoch": 11.017601760176017,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.014954206778877786,
      "loss": 0.2319,
      "num_input_tokens_seen": 21137024,
      "step": 100150
    },
    {
      "epoch": 11.018151815181518,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014952766745164736,
      "loss": 0.2309,
      "num_input_tokens_seen": 21138080,
      "step": 100155
    },
    {
      "epoch": 11.01870187018702,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014951326711887002,
      "loss": 0.2324,
      "num_input_tokens_seen": 21139136,
      "step": 100160
    },
    {
      "epoch": 11.019251925192519,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.014949886679057877,
      "loss": 0.2309,
      "num_input_tokens_seen": 21140192,
      "step": 100165
    },
    {
      "epoch": 11.01980198019802,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.014948446646690615,
      "loss": 0.2329,
      "num_input_tokens_seen": 21141184,
      "step": 100170
    },
    {
      "epoch": 11.020352035203521,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014947006614798499,
      "loss": 0.2324,
      "num_input_tokens_seen": 21142208,
      "step": 100175
    },
    {
      "epoch": 11.02090209020902,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014945566583394802,
      "loss": 0.2308,
      "num_input_tokens_seen": 21143232,
      "step": 100180
    },
    {
      "epoch": 11.021452145214521,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.014944126552492784,
      "loss": 0.2308,
      "num_input_tokens_seen": 21144288,
      "step": 100185
    },
    {
      "epoch": 11.022002200220022,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.014942686522105729,
      "loss": 0.2303,
      "num_input_tokens_seen": 21145280,
      "step": 100190
    },
    {
      "epoch": 11.022552255225522,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.0149412464922469,
      "loss": 0.2314,
      "num_input_tokens_seen": 21146272,
      "step": 100195
    },
    {
      "epoch": 11.023102310231023,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.014939806462929582,
      "loss": 0.2288,
      "num_input_tokens_seen": 21147360,
      "step": 100200
    },
    {
      "epoch": 11.023652365236524,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.014938366434167034,
      "loss": 0.2304,
      "num_input_tokens_seen": 21148384,
      "step": 100205
    },
    {
      "epoch": 11.024202420242025,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014936926405972531,
      "loss": 0.2329,
      "num_input_tokens_seen": 21149408,
      "step": 100210
    },
    {
      "epoch": 11.024752475247524,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014935486378359352,
      "loss": 0.2303,
      "num_input_tokens_seen": 21150528,
      "step": 100215
    },
    {
      "epoch": 11.025302530253025,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.014934046351340761,
      "loss": 0.2298,
      "num_input_tokens_seen": 21151552,
      "step": 100220
    },
    {
      "epoch": 11.025852585258527,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.014932606324930032,
      "loss": 0.2319,
      "num_input_tokens_seen": 21152576,
      "step": 100225
    },
    {
      "epoch": 11.026402640264026,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.01493116629914044,
      "loss": 0.2303,
      "num_input_tokens_seen": 21153600,
      "step": 100230
    },
    {
      "epoch": 11.026952695269527,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.014929726273985252,
      "loss": 0.2309,
      "num_input_tokens_seen": 21154688,
      "step": 100235
    },
    {
      "epoch": 11.027502750275028,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014928286249477752,
      "loss": 0.2319,
      "num_input_tokens_seen": 21155744,
      "step": 100240
    },
    {
      "epoch": 11.028052805280527,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.014926846225631197,
      "loss": 0.2309,
      "num_input_tokens_seen": 21156800,
      "step": 100245
    },
    {
      "epoch": 11.028602860286028,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014925406202458864,
      "loss": 0.2324,
      "num_input_tokens_seen": 21157856,
      "step": 100250
    },
    {
      "epoch": 11.02915291529153,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.01492396617997403,
      "loss": 0.2314,
      "num_input_tokens_seen": 21158816,
      "step": 100255
    },
    {
      "epoch": 11.029702970297029,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01492252615818996,
      "loss": 0.2298,
      "num_input_tokens_seen": 21159872,
      "step": 100260
    },
    {
      "epoch": 11.03025302530253,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01492108613711993,
      "loss": 0.2314,
      "num_input_tokens_seen": 21160832,
      "step": 100265
    },
    {
      "epoch": 11.030803080308031,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014919646116777216,
      "loss": 0.2304,
      "num_input_tokens_seen": 21161952,
      "step": 100270
    },
    {
      "epoch": 11.031353135313532,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.014918206097175079,
      "loss": 0.2319,
      "num_input_tokens_seen": 21163040,
      "step": 100275
    },
    {
      "epoch": 11.031903190319031,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014916766078326796,
      "loss": 0.2293,
      "num_input_tokens_seen": 21164128,
      "step": 100280
    },
    {
      "epoch": 11.032453245324533,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.014915326060245641,
      "loss": 0.2335,
      "num_input_tokens_seen": 21165184,
      "step": 100285
    },
    {
      "epoch": 11.033003300330034,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.014913886042944888,
      "loss": 0.2298,
      "num_input_tokens_seen": 21166304,
      "step": 100290
    },
    {
      "epoch": 11.033553355335533,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01491244602643781,
      "loss": 0.2304,
      "num_input_tokens_seen": 21167296,
      "step": 100295
    },
    {
      "epoch": 11.034103410341034,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014911006010737667,
      "loss": 0.235,
      "num_input_tokens_seen": 21168416,
      "step": 100300
    },
    {
      "epoch": 11.034653465346535,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.014909565995857741,
      "loss": 0.2304,
      "num_input_tokens_seen": 21169408,
      "step": 100305
    },
    {
      "epoch": 11.035203520352034,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0149081259818113,
      "loss": 0.2314,
      "num_input_tokens_seen": 21170496,
      "step": 100310
    },
    {
      "epoch": 11.035753575357536,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.014906685968611624,
      "loss": 0.2314,
      "num_input_tokens_seen": 21171648,
      "step": 100315
    },
    {
      "epoch": 11.036303630363037,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.014905245956271978,
      "loss": 0.2324,
      "num_input_tokens_seen": 21172800,
      "step": 100320
    },
    {
      "epoch": 11.036853685368538,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.014903805944805629,
      "loss": 0.2319,
      "num_input_tokens_seen": 21173856,
      "step": 100325
    },
    {
      "epoch": 11.037403740374037,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014902365934225859,
      "loss": 0.2314,
      "num_input_tokens_seen": 21174848,
      "step": 100330
    },
    {
      "epoch": 11.037953795379538,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.014900925924545932,
      "loss": 0.2319,
      "num_input_tokens_seen": 21176000,
      "step": 100335
    },
    {
      "epoch": 11.03850385038504,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.014899485915779127,
      "loss": 0.2314,
      "num_input_tokens_seen": 21177120,
      "step": 100340
    },
    {
      "epoch": 11.039053905390539,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014898045907938707,
      "loss": 0.2319,
      "num_input_tokens_seen": 21178208,
      "step": 100345
    },
    {
      "epoch": 11.03960396039604,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01489660590103795,
      "loss": 0.2298,
      "num_input_tokens_seen": 21179232,
      "step": 100350
    },
    {
      "epoch": 11.04015401540154,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.014895165895090132,
      "loss": 0.2303,
      "num_input_tokens_seen": 21180288,
      "step": 100355
    },
    {
      "epoch": 11.04070407040704,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.014893725890108515,
      "loss": 0.2308,
      "num_input_tokens_seen": 21181312,
      "step": 100360
    },
    {
      "epoch": 11.041254125412541,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.014892285886106376,
      "loss": 0.2314,
      "num_input_tokens_seen": 21182432,
      "step": 100365
    },
    {
      "epoch": 11.041804180418042,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014890845883096991,
      "loss": 0.2329,
      "num_input_tokens_seen": 21183456,
      "step": 100370
    },
    {
      "epoch": 11.042354235423542,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.014889405881093621,
      "loss": 0.2298,
      "num_input_tokens_seen": 21184544,
      "step": 100375
    },
    {
      "epoch": 11.042904290429043,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.014887965880109546,
      "loss": 0.2298,
      "num_input_tokens_seen": 21185664,
      "step": 100380
    },
    {
      "epoch": 11.043454345434544,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014886525880158034,
      "loss": 0.2303,
      "num_input_tokens_seen": 21186720,
      "step": 100385
    },
    {
      "epoch": 11.044004400440045,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014885085881252364,
      "loss": 0.2314,
      "num_input_tokens_seen": 21187776,
      "step": 100390
    },
    {
      "epoch": 11.044554455445544,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.014883645883405797,
      "loss": 0.2319,
      "num_input_tokens_seen": 21188832,
      "step": 100395
    },
    {
      "epoch": 11.045104510451045,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.014882205886631611,
      "loss": 0.2324,
      "num_input_tokens_seen": 21189856,
      "step": 100400
    },
    {
      "epoch": 11.045654565456546,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.014880765890943077,
      "loss": 0.2314,
      "num_input_tokens_seen": 21190816,
      "step": 100405
    },
    {
      "epoch": 11.046204620462046,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.014879325896353471,
      "loss": 0.2335,
      "num_input_tokens_seen": 21191840,
      "step": 100410
    },
    {
      "epoch": 11.046754675467547,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014877885902876054,
      "loss": 0.2319,
      "num_input_tokens_seen": 21192832,
      "step": 100415
    },
    {
      "epoch": 11.047304730473048,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.014876445910524106,
      "loss": 0.2345,
      "num_input_tokens_seen": 21193856,
      "step": 100420
    },
    {
      "epoch": 11.047854785478547,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.014875005919310895,
      "loss": 0.2324,
      "num_input_tokens_seen": 21194912,
      "step": 100425
    },
    {
      "epoch": 11.048404840484048,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.0148735659292497,
      "loss": 0.2303,
      "num_input_tokens_seen": 21195936,
      "step": 100430
    },
    {
      "epoch": 11.04895489548955,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014872125940353784,
      "loss": 0.2319,
      "num_input_tokens_seen": 21196960,
      "step": 100435
    },
    {
      "epoch": 11.049504950495049,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01487068595263642,
      "loss": 0.2314,
      "num_input_tokens_seen": 21198048,
      "step": 100440
    },
    {
      "epoch": 11.05005500550055,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014869245966110885,
      "loss": 0.2298,
      "num_input_tokens_seen": 21199104,
      "step": 100445
    },
    {
      "epoch": 11.05060506050605,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.014867805980790443,
      "loss": 0.2324,
      "num_input_tokens_seen": 21200192,
      "step": 100450
    },
    {
      "epoch": 11.051155115511552,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014866365996688374,
      "loss": 0.2335,
      "num_input_tokens_seen": 21201312,
      "step": 100455
    },
    {
      "epoch": 11.051705170517051,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.014864926013817947,
      "loss": 0.2314,
      "num_input_tokens_seen": 21202400,
      "step": 100460
    },
    {
      "epoch": 11.052255225522552,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.014863486032192423,
      "loss": 0.2298,
      "num_input_tokens_seen": 21203456,
      "step": 100465
    },
    {
      "epoch": 11.052805280528053,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.014862046051825091,
      "loss": 0.2294,
      "num_input_tokens_seen": 21204512,
      "step": 100470
    },
    {
      "epoch": 11.053355335533553,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.01486060607272921,
      "loss": 0.2294,
      "num_input_tokens_seen": 21205536,
      "step": 100475
    },
    {
      "epoch": 11.053905390539054,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.014859166094918058,
      "loss": 0.2309,
      "num_input_tokens_seen": 21206656,
      "step": 100480
    },
    {
      "epoch": 11.054455445544555,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.014857726118404909,
      "loss": 0.2309,
      "num_input_tokens_seen": 21207776,
      "step": 100485
    },
    {
      "epoch": 11.055005500550054,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01485628614320302,
      "loss": 0.2314,
      "num_input_tokens_seen": 21208832,
      "step": 100490
    },
    {
      "epoch": 11.055555555555555,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.01485484616932568,
      "loss": 0.232,
      "num_input_tokens_seen": 21209920,
      "step": 100495
    },
    {
      "epoch": 11.056105610561056,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.01485340619678615,
      "loss": 0.2294,
      "num_input_tokens_seen": 21210976,
      "step": 100500
    },
    {
      "epoch": 11.056655665566556,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.014851966225597708,
      "loss": 0.2304,
      "num_input_tokens_seen": 21212064,
      "step": 100505
    },
    {
      "epoch": 11.057205720572057,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.01485052625577362,
      "loss": 0.2309,
      "num_input_tokens_seen": 21213088,
      "step": 100510
    },
    {
      "epoch": 11.057755775577558,
      "grad_norm": 0.00072479248046875,
      "learning_rate": 0.014849086287327158,
      "loss": 0.2335,
      "num_input_tokens_seen": 21214144,
      "step": 100515
    },
    {
      "epoch": 11.058305830583059,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.014847646320271602,
      "loss": 0.2293,
      "num_input_tokens_seen": 21215168,
      "step": 100520
    },
    {
      "epoch": 11.058855885588558,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.014846206354620205,
      "loss": 0.2293,
      "num_input_tokens_seen": 21216256,
      "step": 100525
    },
    {
      "epoch": 11.05940594059406,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014844766390386262,
      "loss": 0.2278,
      "num_input_tokens_seen": 21217280,
      "step": 100530
    },
    {
      "epoch": 11.05995599559956,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.014843326427583027,
      "loss": 0.2299,
      "num_input_tokens_seen": 21218336,
      "step": 100535
    },
    {
      "epoch": 11.06050605060506,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.014841886466223777,
      "loss": 0.2299,
      "num_input_tokens_seen": 21219424,
      "step": 100540
    },
    {
      "epoch": 11.061056105610561,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.014840446506321789,
      "loss": 0.2309,
      "num_input_tokens_seen": 21220448,
      "step": 100545
    },
    {
      "epoch": 11.061606160616062,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.01483900654789032,
      "loss": 0.232,
      "num_input_tokens_seen": 21221472,
      "step": 100550
    },
    {
      "epoch": 11.062156215621561,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.014837566590942656,
      "loss": 0.2335,
      "num_input_tokens_seen": 21222432,
      "step": 100555
    },
    {
      "epoch": 11.062706270627062,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014836126635492067,
      "loss": 0.233,
      "num_input_tokens_seen": 21223488,
      "step": 100560
    },
    {
      "epoch": 11.063256325632564,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.014834686681551811,
      "loss": 0.2309,
      "num_input_tokens_seen": 21224544,
      "step": 100565
    },
    {
      "epoch": 11.063806380638065,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.014833246729135174,
      "loss": 0.2283,
      "num_input_tokens_seen": 21225568,
      "step": 100570
    },
    {
      "epoch": 11.064356435643564,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.014831806778255416,
      "loss": 0.2325,
      "num_input_tokens_seen": 21226624,
      "step": 100575
    },
    {
      "epoch": 11.064906490649065,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.014830366828925823,
      "loss": 0.2335,
      "num_input_tokens_seen": 21227680,
      "step": 100580
    },
    {
      "epoch": 11.065456545654566,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.014828926881159655,
      "loss": 0.2314,
      "num_input_tokens_seen": 21228672,
      "step": 100585
    },
    {
      "epoch": 11.066006600660065,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01482748693497018,
      "loss": 0.2298,
      "num_input_tokens_seen": 21229728,
      "step": 100590
    },
    {
      "epoch": 11.066556655665567,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.01482604699037068,
      "loss": 0.2298,
      "num_input_tokens_seen": 21230784,
      "step": 100595
    },
    {
      "epoch": 11.067106710671068,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.014824607047374426,
      "loss": 0.2294,
      "num_input_tokens_seen": 21231840,
      "step": 100600
    },
    {
      "epoch": 11.067656765676567,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.014823167105994677,
      "loss": 0.233,
      "num_input_tokens_seen": 21232960,
      "step": 100605
    },
    {
      "epoch": 11.068206820682068,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014821727166244712,
      "loss": 0.2351,
      "num_input_tokens_seen": 21234016,
      "step": 100610
    },
    {
      "epoch": 11.06875687568757,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.014820287228137802,
      "loss": 0.2298,
      "num_input_tokens_seen": 21235104,
      "step": 100615
    },
    {
      "epoch": 11.069306930693068,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.014818847291687224,
      "loss": 0.2309,
      "num_input_tokens_seen": 21236128,
      "step": 100620
    },
    {
      "epoch": 11.06985698569857,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.014817407356906242,
      "loss": 0.2304,
      "num_input_tokens_seen": 21237152,
      "step": 100625
    },
    {
      "epoch": 11.07040704070407,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.014815967423808123,
      "loss": 0.2294,
      "num_input_tokens_seen": 21238304,
      "step": 100630
    },
    {
      "epoch": 11.070957095709572,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.014814527492406153,
      "loss": 0.2314,
      "num_input_tokens_seen": 21239328,
      "step": 100635
    },
    {
      "epoch": 11.071507150715071,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.014813087562713587,
      "loss": 0.2324,
      "num_input_tokens_seen": 21240448,
      "step": 100640
    },
    {
      "epoch": 11.072057205720572,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.014811647634743706,
      "loss": 0.2294,
      "num_input_tokens_seen": 21241472,
      "step": 100645
    },
    {
      "epoch": 11.072607260726073,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.014810207708509777,
      "loss": 0.2304,
      "num_input_tokens_seen": 21242592,
      "step": 100650
    },
    {
      "epoch": 11.073157315731573,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.014808767784025071,
      "loss": 0.2299,
      "num_input_tokens_seen": 21243680,
      "step": 100655
    },
    {
      "epoch": 11.073707370737074,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014807327861302868,
      "loss": 0.2289,
      "num_input_tokens_seen": 21244736,
      "step": 100660
    },
    {
      "epoch": 11.074257425742575,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.014805887940356423,
      "loss": 0.2299,
      "num_input_tokens_seen": 21245824,
      "step": 100665
    },
    {
      "epoch": 11.074807480748074,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.01480444802119902,
      "loss": 0.2325,
      "num_input_tokens_seen": 21246880,
      "step": 100670
    },
    {
      "epoch": 11.075357535753575,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.01480300810384393,
      "loss": 0.2299,
      "num_input_tokens_seen": 21247904,
      "step": 100675
    },
    {
      "epoch": 11.075907590759076,
      "grad_norm": 0.00067138671875,
      "learning_rate": 0.014801568188304411,
      "loss": 0.2325,
      "num_input_tokens_seen": 21248896,
      "step": 100680
    },
    {
      "epoch": 11.076457645764576,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.014800128274593746,
      "loss": 0.2325,
      "num_input_tokens_seen": 21250016,
      "step": 100685
    },
    {
      "epoch": 11.077007700770077,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.014798688362725201,
      "loss": 0.2346,
      "num_input_tokens_seen": 21251072,
      "step": 100690
    },
    {
      "epoch": 11.077557755775578,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.014797248452712054,
      "loss": 0.232,
      "num_input_tokens_seen": 21252160,
      "step": 100695
    },
    {
      "epoch": 11.078107810781079,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.014795808544567569,
      "loss": 0.233,
      "num_input_tokens_seen": 21253216,
      "step": 100700
    },
    {
      "epoch": 11.078657865786578,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.014794368638305013,
      "loss": 0.2319,
      "num_input_tokens_seen": 21254304,
      "step": 100705
    },
    {
      "epoch": 11.07920792079208,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01479292873393767,
      "loss": 0.2294,
      "num_input_tokens_seen": 21255328,
      "step": 100710
    },
    {
      "epoch": 11.07975797579758,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.014791488831478796,
      "loss": 0.2325,
      "num_input_tokens_seen": 21256384,
      "step": 100715
    },
    {
      "epoch": 11.08030803080308,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01479004893094168,
      "loss": 0.2319,
      "num_input_tokens_seen": 21257376,
      "step": 100720
    },
    {
      "epoch": 11.08085808580858,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.014788609032339575,
      "loss": 0.2304,
      "num_input_tokens_seen": 21258400,
      "step": 100725
    },
    {
      "epoch": 11.081408140814082,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.014787169135685759,
      "loss": 0.2309,
      "num_input_tokens_seen": 21259424,
      "step": 100730
    },
    {
      "epoch": 11.081958195819581,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.014785729240993507,
      "loss": 0.233,
      "num_input_tokens_seen": 21260448,
      "step": 100735
    },
    {
      "epoch": 11.082508250825082,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01478428934827608,
      "loss": 0.2309,
      "num_input_tokens_seen": 21261536,
      "step": 100740
    },
    {
      "epoch": 11.083058305830583,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.014782849457546758,
      "loss": 0.2324,
      "num_input_tokens_seen": 21262592,
      "step": 100745
    },
    {
      "epoch": 11.083608360836084,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.014781409568818812,
      "loss": 0.2335,
      "num_input_tokens_seen": 21263744,
      "step": 100750
    },
    {
      "epoch": 11.084158415841584,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.014779969682105503,
      "loss": 0.2324,
      "num_input_tokens_seen": 21264768,
      "step": 100755
    },
    {
      "epoch": 11.084708470847085,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01477852979742011,
      "loss": 0.2314,
      "num_input_tokens_seen": 21265792,
      "step": 100760
    },
    {
      "epoch": 11.085258525852586,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0147770899147759,
      "loss": 0.2324,
      "num_input_tokens_seen": 21266816,
      "step": 100765
    },
    {
      "epoch": 11.085808580858085,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.014775650034186147,
      "loss": 0.2324,
      "num_input_tokens_seen": 21267904,
      "step": 100770
    },
    {
      "epoch": 11.086358635863586,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014774210155664122,
      "loss": 0.2319,
      "num_input_tokens_seen": 21268928,
      "step": 100775
    },
    {
      "epoch": 11.086908690869087,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.014772770279223088,
      "loss": 0.2335,
      "num_input_tokens_seen": 21270048,
      "step": 100780
    },
    {
      "epoch": 11.087458745874587,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.014771330404876326,
      "loss": 0.2314,
      "num_input_tokens_seen": 21271072,
      "step": 100785
    },
    {
      "epoch": 11.088008800880088,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.014769890532637105,
      "loss": 0.2319,
      "num_input_tokens_seen": 21272064,
      "step": 100790
    },
    {
      "epoch": 11.088558855885589,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.014768450662518685,
      "loss": 0.2298,
      "num_input_tokens_seen": 21273152,
      "step": 100795
    },
    {
      "epoch": 11.089108910891088,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014767010794534349,
      "loss": 0.2314,
      "num_input_tokens_seen": 21274208,
      "step": 100800
    },
    {
      "epoch": 11.08965896589659,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.014765570928697357,
      "loss": 0.2319,
      "num_input_tokens_seen": 21275328,
      "step": 100805
    },
    {
      "epoch": 11.09020902090209,
      "grad_norm": 0.000614166259765625,
      "learning_rate": 0.014764131065020994,
      "loss": 0.2319,
      "num_input_tokens_seen": 21276320,
      "step": 100810
    },
    {
      "epoch": 11.090759075907592,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014762691203518518,
      "loss": 0.2314,
      "num_input_tokens_seen": 21277376,
      "step": 100815
    },
    {
      "epoch": 11.091309130913091,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0147612513442032,
      "loss": 0.2309,
      "num_input_tokens_seen": 21278464,
      "step": 100820
    },
    {
      "epoch": 11.091859185918592,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014759811487088319,
      "loss": 0.2314,
      "num_input_tokens_seen": 21279552,
      "step": 100825
    },
    {
      "epoch": 11.092409240924093,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014758371632187132,
      "loss": 0.2319,
      "num_input_tokens_seen": 21280608,
      "step": 100830
    },
    {
      "epoch": 11.092959295929592,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014756931779512927,
      "loss": 0.2308,
      "num_input_tokens_seen": 21281632,
      "step": 100835
    },
    {
      "epoch": 11.093509350935093,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.014755491929078962,
      "loss": 0.2308,
      "num_input_tokens_seen": 21282784,
      "step": 100840
    },
    {
      "epoch": 11.094059405940595,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.014754052080898507,
      "loss": 0.2319,
      "num_input_tokens_seen": 21283776,
      "step": 100845
    },
    {
      "epoch": 11.094609460946094,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.014752612234984842,
      "loss": 0.2303,
      "num_input_tokens_seen": 21284832,
      "step": 100850
    },
    {
      "epoch": 11.095159515951595,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014751172391351226,
      "loss": 0.2288,
      "num_input_tokens_seen": 21285888,
      "step": 100855
    },
    {
      "epoch": 11.095709570957096,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.014749732550010936,
      "loss": 0.2324,
      "num_input_tokens_seen": 21287008,
      "step": 100860
    },
    {
      "epoch": 11.096259625962595,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.014748292710977247,
      "loss": 0.233,
      "num_input_tokens_seen": 21288064,
      "step": 100865
    },
    {
      "epoch": 11.096809680968097,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.014746852874263415,
      "loss": 0.2303,
      "num_input_tokens_seen": 21289120,
      "step": 100870
    },
    {
      "epoch": 11.097359735973598,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.014745413039882723,
      "loss": 0.2324,
      "num_input_tokens_seen": 21290208,
      "step": 100875
    },
    {
      "epoch": 11.097909790979099,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.014743973207848432,
      "loss": 0.2329,
      "num_input_tokens_seen": 21291232,
      "step": 100880
    },
    {
      "epoch": 11.098459845984598,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.014742533378173824,
      "loss": 0.2308,
      "num_input_tokens_seen": 21292320,
      "step": 100885
    },
    {
      "epoch": 11.099009900990099,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01474109355087216,
      "loss": 0.2303,
      "num_input_tokens_seen": 21293344,
      "step": 100890
    },
    {
      "epoch": 11.0995599559956,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01473965372595671,
      "loss": 0.2298,
      "num_input_tokens_seen": 21294400,
      "step": 100895
    },
    {
      "epoch": 11.1001100110011,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.014738213903440746,
      "loss": 0.2319,
      "num_input_tokens_seen": 21295520,
      "step": 100900
    },
    {
      "epoch": 11.1006600660066,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.01473677408333754,
      "loss": 0.2288,
      "num_input_tokens_seen": 21296608,
      "step": 100905
    },
    {
      "epoch": 11.101210121012102,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.014735334265660364,
      "loss": 0.2319,
      "num_input_tokens_seen": 21297632,
      "step": 100910
    },
    {
      "epoch": 11.101760176017601,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014733894450422485,
      "loss": 0.2303,
      "num_input_tokens_seen": 21298624,
      "step": 100915
    },
    {
      "epoch": 11.102310231023102,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.014732454637637168,
      "loss": 0.2319,
      "num_input_tokens_seen": 21299744,
      "step": 100920
    },
    {
      "epoch": 11.102860286028603,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.014731014827317696,
      "loss": 0.234,
      "num_input_tokens_seen": 21300800,
      "step": 100925
    },
    {
      "epoch": 11.103410341034103,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014729575019477322,
      "loss": 0.2293,
      "num_input_tokens_seen": 21301824,
      "step": 100930
    },
    {
      "epoch": 11.103960396039604,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.014728135214129332,
      "loss": 0.2288,
      "num_input_tokens_seen": 21302848,
      "step": 100935
    },
    {
      "epoch": 11.104510451045105,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.014726695411286992,
      "loss": 0.2319,
      "num_input_tokens_seen": 21303936,
      "step": 100940
    },
    {
      "epoch": 11.105060506050606,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.014725255610963557,
      "loss": 0.2314,
      "num_input_tokens_seen": 21304960,
      "step": 100945
    },
    {
      "epoch": 11.105610561056105,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.014723815813172322,
      "loss": 0.2314,
      "num_input_tokens_seen": 21306016,
      "step": 100950
    },
    {
      "epoch": 11.106160616061606,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.014722376017926536,
      "loss": 0.2319,
      "num_input_tokens_seen": 21307072,
      "step": 100955
    },
    {
      "epoch": 11.106710671067107,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.014720936225239479,
      "loss": 0.2303,
      "num_input_tokens_seen": 21308160,
      "step": 100960
    },
    {
      "epoch": 11.107260726072607,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.014719496435124425,
      "loss": 0.2304,
      "num_input_tokens_seen": 21309184,
      "step": 100965
    },
    {
      "epoch": 11.107810781078108,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01471805664759463,
      "loss": 0.2309,
      "num_input_tokens_seen": 21310208,
      "step": 100970
    },
    {
      "epoch": 11.108360836083609,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.014716616862663373,
      "loss": 0.2319,
      "num_input_tokens_seen": 21311232,
      "step": 100975
    },
    {
      "epoch": 11.108910891089108,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.014715177080343927,
      "loss": 0.2324,
      "num_input_tokens_seen": 21312256,
      "step": 100980
    },
    {
      "epoch": 11.10946094609461,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.014713737300649551,
      "loss": 0.2319,
      "num_input_tokens_seen": 21313280,
      "step": 100985
    },
    {
      "epoch": 11.11001100110011,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.014712297523593525,
      "loss": 0.2324,
      "num_input_tokens_seen": 21314304,
      "step": 100990
    },
    {
      "epoch": 11.110561056105611,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.01471085774918911,
      "loss": 0.2335,
      "num_input_tokens_seen": 21315360,
      "step": 100995
    },
    {
      "epoch": 11.11111111111111,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.014709417977449589,
      "loss": 0.2303,
      "num_input_tokens_seen": 21316448,
      "step": 101000
    },
    {
      "epoch": 11.111661166116612,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014707978208388216,
      "loss": 0.2335,
      "num_input_tokens_seen": 21317536,
      "step": 101005
    },
    {
      "epoch": 11.112211221122113,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.014706538442018267,
      "loss": 0.2314,
      "num_input_tokens_seen": 21318560,
      "step": 101010
    },
    {
      "epoch": 11.112761276127612,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014705098678353016,
      "loss": 0.2303,
      "num_input_tokens_seen": 21319552,
      "step": 101015
    },
    {
      "epoch": 11.113311331133113,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.014703658917405726,
      "loss": 0.2314,
      "num_input_tokens_seen": 21320640,
      "step": 101020
    },
    {
      "epoch": 11.113861386138614,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.014702219159189674,
      "loss": 0.2319,
      "num_input_tokens_seen": 21321632,
      "step": 101025
    },
    {
      "epoch": 11.114411441144114,
      "grad_norm": 0.00078582763671875,
      "learning_rate": 0.014700779403718122,
      "loss": 0.2324,
      "num_input_tokens_seen": 21322720,
      "step": 101030
    },
    {
      "epoch": 11.114961496149615,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.014699339651004341,
      "loss": 0.2303,
      "num_input_tokens_seen": 21323744,
      "step": 101035
    },
    {
      "epoch": 11.115511551155116,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.01469789990106161,
      "loss": 0.2293,
      "num_input_tokens_seen": 21324768,
      "step": 101040
    },
    {
      "epoch": 11.116061606160615,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014696460153903182,
      "loss": 0.2319,
      "num_input_tokens_seen": 21325856,
      "step": 101045
    },
    {
      "epoch": 11.116611661166116,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.014695020409542337,
      "loss": 0.2304,
      "num_input_tokens_seen": 21326880,
      "step": 101050
    },
    {
      "epoch": 11.117161716171617,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014693580667992347,
      "loss": 0.2308,
      "num_input_tokens_seen": 21327904,
      "step": 101055
    },
    {
      "epoch": 11.117711771177119,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.014692140929266472,
      "loss": 0.2319,
      "num_input_tokens_seen": 21328960,
      "step": 101060
    },
    {
      "epoch": 11.118261826182618,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.014690701193377988,
      "loss": 0.2325,
      "num_input_tokens_seen": 21330016,
      "step": 101065
    },
    {
      "epoch": 11.118811881188119,
      "grad_norm": 0.000751495361328125,
      "learning_rate": 0.01468926146034016,
      "loss": 0.2319,
      "num_input_tokens_seen": 21331008,
      "step": 101070
    },
    {
      "epoch": 11.11936193619362,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.014687821730166264,
      "loss": 0.2303,
      "num_input_tokens_seen": 21332128,
      "step": 101075
    },
    {
      "epoch": 11.11991199119912,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014686382002869567,
      "loss": 0.2313,
      "num_input_tokens_seen": 21333152,
      "step": 101080
    },
    {
      "epoch": 11.12046204620462,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01468494227846333,
      "loss": 0.2319,
      "num_input_tokens_seen": 21334144,
      "step": 101085
    },
    {
      "epoch": 11.121012101210122,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.014683502556960834,
      "loss": 0.2314,
      "num_input_tokens_seen": 21335200,
      "step": 101090
    },
    {
      "epoch": 11.12156215621562,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.014682062838375337,
      "loss": 0.2314,
      "num_input_tokens_seen": 21336256,
      "step": 101095
    },
    {
      "epoch": 11.122112211221122,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.014680623122720124,
      "loss": 0.2319,
      "num_input_tokens_seen": 21337344,
      "step": 101100
    },
    {
      "epoch": 11.122662266226623,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.01467918341000845,
      "loss": 0.2313,
      "num_input_tokens_seen": 21338368,
      "step": 101105
    },
    {
      "epoch": 11.123212321232122,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.014677743700253584,
      "loss": 0.2329,
      "num_input_tokens_seen": 21339424,
      "step": 101110
    },
    {
      "epoch": 11.123762376237623,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014676303993468807,
      "loss": 0.2314,
      "num_input_tokens_seen": 21340512,
      "step": 101115
    },
    {
      "epoch": 11.124312431243125,
      "grad_norm": 0.000705718994140625,
      "learning_rate": 0.014674864289667376,
      "loss": 0.2329,
      "num_input_tokens_seen": 21341568,
      "step": 101120
    },
    {
      "epoch": 11.124862486248626,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.014673424588862565,
      "loss": 0.2329,
      "num_input_tokens_seen": 21342560,
      "step": 101125
    },
    {
      "epoch": 11.125412541254125,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014671984891067644,
      "loss": 0.2319,
      "num_input_tokens_seen": 21343616,
      "step": 101130
    },
    {
      "epoch": 11.125962596259626,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.014670545196295879,
      "loss": 0.2308,
      "num_input_tokens_seen": 21344672,
      "step": 101135
    },
    {
      "epoch": 11.126512651265127,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.014669105504560546,
      "loss": 0.2324,
      "num_input_tokens_seen": 21345792,
      "step": 101140
    },
    {
      "epoch": 11.127062706270626,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014667665815874902,
      "loss": 0.2319,
      "num_input_tokens_seen": 21346912,
      "step": 101145
    },
    {
      "epoch": 11.127612761276128,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014666226130252227,
      "loss": 0.2303,
      "num_input_tokens_seen": 21348064,
      "step": 101150
    },
    {
      "epoch": 11.128162816281629,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.014664786447705788,
      "loss": 0.2324,
      "num_input_tokens_seen": 21349152,
      "step": 101155
    },
    {
      "epoch": 11.128712871287128,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014663346768248845,
      "loss": 0.2329,
      "num_input_tokens_seen": 21350176,
      "step": 101160
    },
    {
      "epoch": 11.129262926292629,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.014661907091894678,
      "loss": 0.2313,
      "num_input_tokens_seen": 21351264,
      "step": 101165
    },
    {
      "epoch": 11.12981298129813,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.014660467418656552,
      "loss": 0.2313,
      "num_input_tokens_seen": 21352352,
      "step": 101170
    },
    {
      "epoch": 11.130363036303631,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01465902774854773,
      "loss": 0.2324,
      "num_input_tokens_seen": 21353472,
      "step": 101175
    },
    {
      "epoch": 11.13091309130913,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.014657588081581489,
      "loss": 0.2308,
      "num_input_tokens_seen": 21354496,
      "step": 101180
    },
    {
      "epoch": 11.131463146314632,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.014656148417771091,
      "loss": 0.2309,
      "num_input_tokens_seen": 21355488,
      "step": 101185
    },
    {
      "epoch": 11.132013201320133,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.014654708757129811,
      "loss": 0.2314,
      "num_input_tokens_seen": 21356512,
      "step": 101190
    },
    {
      "epoch": 11.132563256325632,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.014653269099670916,
      "loss": 0.2298,
      "num_input_tokens_seen": 21357568,
      "step": 101195
    },
    {
      "epoch": 11.133113311331133,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.014651829445407669,
      "loss": 0.2335,
      "num_input_tokens_seen": 21358720,
      "step": 101200
    },
    {
      "epoch": 11.133663366336634,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014650389794353347,
      "loss": 0.2324,
      "num_input_tokens_seen": 21359744,
      "step": 101205
    },
    {
      "epoch": 11.134213421342134,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01464895014652121,
      "loss": 0.2309,
      "num_input_tokens_seen": 21360736,
      "step": 101210
    },
    {
      "epoch": 11.134763476347635,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014647510501924537,
      "loss": 0.233,
      "num_input_tokens_seen": 21361728,
      "step": 101215
    },
    {
      "epoch": 11.135313531353136,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.014646070860576588,
      "loss": 0.2324,
      "num_input_tokens_seen": 21362816,
      "step": 101220
    },
    {
      "epoch": 11.135863586358635,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.01464463122249063,
      "loss": 0.2288,
      "num_input_tokens_seen": 21363904,
      "step": 101225
    },
    {
      "epoch": 11.136413641364136,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.014643191587679943,
      "loss": 0.2314,
      "num_input_tokens_seen": 21364896,
      "step": 101230
    },
    {
      "epoch": 11.136963696369637,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.014641751956157781,
      "loss": 0.2329,
      "num_input_tokens_seen": 21365920,
      "step": 101235
    },
    {
      "epoch": 11.137513751375138,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.014640312327937423,
      "loss": 0.2293,
      "num_input_tokens_seen": 21367008,
      "step": 101240
    },
    {
      "epoch": 11.138063806380638,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.014638872703032137,
      "loss": 0.2293,
      "num_input_tokens_seen": 21368032,
      "step": 101245
    },
    {
      "epoch": 11.138613861386139,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014637433081455178,
      "loss": 0.2299,
      "num_input_tokens_seen": 21369088,
      "step": 101250
    },
    {
      "epoch": 11.13916391639164,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.014635993463219832,
      "loss": 0.2293,
      "num_input_tokens_seen": 21370144,
      "step": 101255
    },
    {
      "epoch": 11.13971397139714,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014634553848339355,
      "loss": 0.2314,
      "num_input_tokens_seen": 21371232,
      "step": 101260
    },
    {
      "epoch": 11.14026402640264,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.014633114236827024,
      "loss": 0.2303,
      "num_input_tokens_seen": 21372320,
      "step": 101265
    },
    {
      "epoch": 11.140814081408141,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014631674628696105,
      "loss": 0.2308,
      "num_input_tokens_seen": 21373376,
      "step": 101270
    },
    {
      "epoch": 11.14136413641364,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.014630235023959859,
      "loss": 0.2319,
      "num_input_tokens_seen": 21374400,
      "step": 101275
    },
    {
      "epoch": 11.141914191419142,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.014628795422631562,
      "loss": 0.2309,
      "num_input_tokens_seen": 21375424,
      "step": 101280
    },
    {
      "epoch": 11.142464246424643,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.014627355824724476,
      "loss": 0.2329,
      "num_input_tokens_seen": 21376512,
      "step": 101285
    },
    {
      "epoch": 11.143014301430142,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.01462591623025188,
      "loss": 0.2329,
      "num_input_tokens_seen": 21377568,
      "step": 101290
    },
    {
      "epoch": 11.143564356435643,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.014624476639227028,
      "loss": 0.2319,
      "num_input_tokens_seen": 21378688,
      "step": 101295
    },
    {
      "epoch": 11.144114411441144,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.014623037051663195,
      "loss": 0.2313,
      "num_input_tokens_seen": 21379744,
      "step": 101300
    },
    {
      "epoch": 11.144664466446645,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.014621597467573653,
      "loss": 0.2309,
      "num_input_tokens_seen": 21380832,
      "step": 101305
    },
    {
      "epoch": 11.145214521452145,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.014620157886971656,
      "loss": 0.2314,
      "num_input_tokens_seen": 21381888,
      "step": 101310
    },
    {
      "epoch": 11.145764576457646,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014618718309870493,
      "loss": 0.2314,
      "num_input_tokens_seen": 21382976,
      "step": 101315
    },
    {
      "epoch": 11.146314631463147,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.014617278736283415,
      "loss": 0.2324,
      "num_input_tokens_seen": 21384032,
      "step": 101320
    },
    {
      "epoch": 11.146864686468646,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.014615839166223693,
      "loss": 0.2314,
      "num_input_tokens_seen": 21385088,
      "step": 101325
    },
    {
      "epoch": 11.147414741474147,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014614399599704604,
      "loss": 0.2309,
      "num_input_tokens_seen": 21386176,
      "step": 101330
    },
    {
      "epoch": 11.147964796479648,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014612960036739401,
      "loss": 0.2324,
      "num_input_tokens_seen": 21387200,
      "step": 101335
    },
    {
      "epoch": 11.148514851485148,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.014611520477341365,
      "loss": 0.2319,
      "num_input_tokens_seen": 21388288,
      "step": 101340
    },
    {
      "epoch": 11.149064906490649,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014610080921523761,
      "loss": 0.2308,
      "num_input_tokens_seen": 21389312,
      "step": 101345
    },
    {
      "epoch": 11.14961496149615,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.014608641369299846,
      "loss": 0.2324,
      "num_input_tokens_seen": 21390400,
      "step": 101350
    },
    {
      "epoch": 11.150165016501651,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.0146072018206829,
      "loss": 0.2324,
      "num_input_tokens_seen": 21391456,
      "step": 101355
    },
    {
      "epoch": 11.15071507150715,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.014605762275686184,
      "loss": 0.2309,
      "num_input_tokens_seen": 21392576,
      "step": 101360
    },
    {
      "epoch": 11.151265126512651,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.014604322734322973,
      "loss": 0.2319,
      "num_input_tokens_seen": 21393696,
      "step": 101365
    },
    {
      "epoch": 11.151815181518153,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.014602883196606527,
      "loss": 0.2314,
      "num_input_tokens_seen": 21394784,
      "step": 101370
    },
    {
      "epoch": 11.152365236523652,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01460144366255011,
      "loss": 0.2309,
      "num_input_tokens_seen": 21395808,
      "step": 101375
    },
    {
      "epoch": 11.152915291529153,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.014600004132167003,
      "loss": 0.2314,
      "num_input_tokens_seen": 21396896,
      "step": 101380
    },
    {
      "epoch": 11.153465346534654,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.014598564605470467,
      "loss": 0.234,
      "num_input_tokens_seen": 21398016,
      "step": 101385
    },
    {
      "epoch": 11.154015401540153,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.014597125082473763,
      "loss": 0.2319,
      "num_input_tokens_seen": 21399040,
      "step": 101390
    },
    {
      "epoch": 11.154565456545654,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.014595685563190165,
      "loss": 0.2303,
      "num_input_tokens_seen": 21400160,
      "step": 101395
    },
    {
      "epoch": 11.155115511551156,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.014594246047632937,
      "loss": 0.2329,
      "num_input_tokens_seen": 21401248,
      "step": 101400
    },
    {
      "epoch": 11.155665566556655,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014592806535815357,
      "loss": 0.2314,
      "num_input_tokens_seen": 21402304,
      "step": 101405
    },
    {
      "epoch": 11.156215621562156,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.014591367027750678,
      "loss": 0.2319,
      "num_input_tokens_seen": 21403360,
      "step": 101410
    },
    {
      "epoch": 11.156765676567657,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.014589927523452171,
      "loss": 0.2314,
      "num_input_tokens_seen": 21404416,
      "step": 101415
    },
    {
      "epoch": 11.157315731573158,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.014588488022933112,
      "loss": 0.2319,
      "num_input_tokens_seen": 21405472,
      "step": 101420
    },
    {
      "epoch": 11.157865786578657,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014587048526206754,
      "loss": 0.2329,
      "num_input_tokens_seen": 21406528,
      "step": 101425
    },
    {
      "epoch": 11.158415841584159,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.014585609033286375,
      "loss": 0.2319,
      "num_input_tokens_seen": 21407648,
      "step": 101430
    },
    {
      "epoch": 11.15896589658966,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014584169544185238,
      "loss": 0.2314,
      "num_input_tokens_seen": 21408736,
      "step": 101435
    },
    {
      "epoch": 11.159515951595159,
      "grad_norm": 0.0047607421875,
      "learning_rate": 0.014582730058916608,
      "loss": 0.2319,
      "num_input_tokens_seen": 21409792,
      "step": 101440
    },
    {
      "epoch": 11.16006600660066,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.014581290577493763,
      "loss": 0.2335,
      "num_input_tokens_seen": 21410816,
      "step": 101445
    },
    {
      "epoch": 11.160616061606161,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.014579851099929951,
      "loss": 0.2313,
      "num_input_tokens_seen": 21411968,
      "step": 101450
    },
    {
      "epoch": 11.16116611661166,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.014578411626238456,
      "loss": 0.2303,
      "num_input_tokens_seen": 21413024,
      "step": 101455
    },
    {
      "epoch": 11.161716171617162,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014576972156432545,
      "loss": 0.2288,
      "num_input_tokens_seen": 21414048,
      "step": 101460
    },
    {
      "epoch": 11.162266226622663,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.014575532690525467,
      "loss": 0.2309,
      "num_input_tokens_seen": 21415072,
      "step": 101465
    },
    {
      "epoch": 11.162816281628162,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.014574093228530507,
      "loss": 0.2324,
      "num_input_tokens_seen": 21416096,
      "step": 101470
    },
    {
      "epoch": 11.163366336633663,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01457265377046092,
      "loss": 0.2314,
      "num_input_tokens_seen": 21417152,
      "step": 101475
    },
    {
      "epoch": 11.163916391639164,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.014571214316329986,
      "loss": 0.2319,
      "num_input_tokens_seen": 21418272,
      "step": 101480
    },
    {
      "epoch": 11.164466446644665,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.01456977486615096,
      "loss": 0.2314,
      "num_input_tokens_seen": 21419328,
      "step": 101485
    },
    {
      "epoch": 11.165016501650165,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014568335419937111,
      "loss": 0.2324,
      "num_input_tokens_seen": 21420384,
      "step": 101490
    },
    {
      "epoch": 11.165566556655666,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01456689597770171,
      "loss": 0.2309,
      "num_input_tokens_seen": 21421376,
      "step": 101495
    },
    {
      "epoch": 11.166116611661167,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014565456539458016,
      "loss": 0.2309,
      "num_input_tokens_seen": 21422432,
      "step": 101500
    },
    {
      "epoch": 11.166666666666666,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014564017105219308,
      "loss": 0.2335,
      "num_input_tokens_seen": 21423424,
      "step": 101505
    },
    {
      "epoch": 11.167216721672167,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01456257767499884,
      "loss": 0.2293,
      "num_input_tokens_seen": 21424416,
      "step": 101510
    },
    {
      "epoch": 11.167766776677668,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.014561138248809884,
      "loss": 0.2319,
      "num_input_tokens_seen": 21425440,
      "step": 101515
    },
    {
      "epoch": 11.168316831683168,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01455969882666571,
      "loss": 0.233,
      "num_input_tokens_seen": 21426592,
      "step": 101520
    },
    {
      "epoch": 11.168866886688669,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.014558259408579575,
      "loss": 0.2303,
      "num_input_tokens_seen": 21427616,
      "step": 101525
    },
    {
      "epoch": 11.16941694169417,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.014556819994564754,
      "loss": 0.2314,
      "num_input_tokens_seen": 21428672,
      "step": 101530
    },
    {
      "epoch": 11.16996699669967,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.014555380584634514,
      "loss": 0.2319,
      "num_input_tokens_seen": 21429728,
      "step": 101535
    },
    {
      "epoch": 11.17051705170517,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.014553941178802111,
      "loss": 0.2303,
      "num_input_tokens_seen": 21430784,
      "step": 101540
    },
    {
      "epoch": 11.171067106710671,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014552501777080822,
      "loss": 0.2308,
      "num_input_tokens_seen": 21431808,
      "step": 101545
    },
    {
      "epoch": 11.171617161716172,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.014551062379483906,
      "loss": 0.2324,
      "num_input_tokens_seen": 21432832,
      "step": 101550
    },
    {
      "epoch": 11.172167216721672,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.014549622986024637,
      "loss": 0.2324,
      "num_input_tokens_seen": 21433888,
      "step": 101555
    },
    {
      "epoch": 11.172717271727173,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014548183596716278,
      "loss": 0.2324,
      "num_input_tokens_seen": 21434944,
      "step": 101560
    },
    {
      "epoch": 11.173267326732674,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.014546744211572087,
      "loss": 0.2298,
      "num_input_tokens_seen": 21435968,
      "step": 101565
    },
    {
      "epoch": 11.173817381738173,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014545304830605342,
      "loss": 0.2319,
      "num_input_tokens_seen": 21436992,
      "step": 101570
    },
    {
      "epoch": 11.174367436743674,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014543865453829309,
      "loss": 0.2303,
      "num_input_tokens_seen": 21438016,
      "step": 101575
    },
    {
      "epoch": 11.174917491749175,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.01454242608125724,
      "loss": 0.2309,
      "num_input_tokens_seen": 21439104,
      "step": 101580
    },
    {
      "epoch": 11.175467546754675,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014540986712902415,
      "loss": 0.2319,
      "num_input_tokens_seen": 21440128,
      "step": 101585
    },
    {
      "epoch": 11.176017601760176,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.01453954734877809,
      "loss": 0.233,
      "num_input_tokens_seen": 21441184,
      "step": 101590
    },
    {
      "epoch": 11.176567656765677,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.014538107988897545,
      "loss": 0.2329,
      "num_input_tokens_seen": 21442240,
      "step": 101595
    },
    {
      "epoch": 11.177117711771178,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014536668633274031,
      "loss": 0.2298,
      "num_input_tokens_seen": 21443264,
      "step": 101600
    },
    {
      "epoch": 11.177667766776677,
      "grad_norm": 0.00469970703125,
      "learning_rate": 0.01453522928192082,
      "loss": 0.2293,
      "num_input_tokens_seen": 21444288,
      "step": 101605
    },
    {
      "epoch": 11.178217821782178,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.014533789934851181,
      "loss": 0.2303,
      "num_input_tokens_seen": 21445280,
      "step": 101610
    },
    {
      "epoch": 11.17876787678768,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.01453235059207837,
      "loss": 0.2356,
      "num_input_tokens_seen": 21446336,
      "step": 101615
    },
    {
      "epoch": 11.179317931793179,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.014530911253615667,
      "loss": 0.2309,
      "num_input_tokens_seen": 21447328,
      "step": 101620
    },
    {
      "epoch": 11.17986798679868,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.014529471919476327,
      "loss": 0.2329,
      "num_input_tokens_seen": 21448416,
      "step": 101625
    },
    {
      "epoch": 11.180418041804181,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.014528032589673615,
      "loss": 0.234,
      "num_input_tokens_seen": 21449472,
      "step": 101630
    },
    {
      "epoch": 11.18096809680968,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.014526593264220806,
      "loss": 0.234,
      "num_input_tokens_seen": 21450496,
      "step": 101635
    },
    {
      "epoch": 11.181518151815181,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014525153943131154,
      "loss": 0.2309,
      "num_input_tokens_seen": 21451584,
      "step": 101640
    },
    {
      "epoch": 11.182068206820682,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.014523714626417933,
      "loss": 0.2314,
      "num_input_tokens_seen": 21452608,
      "step": 101645
    },
    {
      "epoch": 11.182618261826182,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.014522275314094411,
      "loss": 0.2319,
      "num_input_tokens_seen": 21453664,
      "step": 101650
    },
    {
      "epoch": 11.183168316831683,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.01452083600617384,
      "loss": 0.2309,
      "num_input_tokens_seen": 21454688,
      "step": 101655
    },
    {
      "epoch": 11.183718371837184,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014519396702669496,
      "loss": 0.2309,
      "num_input_tokens_seen": 21455808,
      "step": 101660
    },
    {
      "epoch": 11.184268426842685,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01451795740359464,
      "loss": 0.235,
      "num_input_tokens_seen": 21456800,
      "step": 101665
    },
    {
      "epoch": 11.184818481848184,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.014516518108962546,
      "loss": 0.234,
      "num_input_tokens_seen": 21457888,
      "step": 101670
    },
    {
      "epoch": 11.185368536853685,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.014515078818786468,
      "loss": 0.2288,
      "num_input_tokens_seen": 21458912,
      "step": 101675
    },
    {
      "epoch": 11.185918591859187,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014513639533079674,
      "loss": 0.2304,
      "num_input_tokens_seen": 21459936,
      "step": 101680
    },
    {
      "epoch": 11.186468646864686,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.014512200251855435,
      "loss": 0.2298,
      "num_input_tokens_seen": 21460992,
      "step": 101685
    },
    {
      "epoch": 11.187018701870187,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014510760975127008,
      "loss": 0.2314,
      "num_input_tokens_seen": 21462080,
      "step": 101690
    },
    {
      "epoch": 11.187568756875688,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.014509321702907666,
      "loss": 0.2324,
      "num_input_tokens_seen": 21463072,
      "step": 101695
    },
    {
      "epoch": 11.188118811881187,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.014507882435210671,
      "loss": 0.2319,
      "num_input_tokens_seen": 21464160,
      "step": 101700
    },
    {
      "epoch": 11.188668866886688,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.014506443172049282,
      "loss": 0.2308,
      "num_input_tokens_seen": 21465184,
      "step": 101705
    },
    {
      "epoch": 11.18921892189219,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014505003913436778,
      "loss": 0.2319,
      "num_input_tokens_seen": 21466208,
      "step": 101710
    },
    {
      "epoch": 11.189768976897689,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014503564659386408,
      "loss": 0.2324,
      "num_input_tokens_seen": 21467264,
      "step": 101715
    },
    {
      "epoch": 11.19031903190319,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.014502125409911446,
      "loss": 0.2319,
      "num_input_tokens_seen": 21468320,
      "step": 101720
    },
    {
      "epoch": 11.190869086908691,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01450068616502516,
      "loss": 0.2314,
      "num_input_tokens_seen": 21469376,
      "step": 101725
    },
    {
      "epoch": 11.191419141914192,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.0144992469247408,
      "loss": 0.2314,
      "num_input_tokens_seen": 21470336,
      "step": 101730
    },
    {
      "epoch": 11.191969196919691,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014497807689071652,
      "loss": 0.2308,
      "num_input_tokens_seen": 21471392,
      "step": 101735
    },
    {
      "epoch": 11.192519251925193,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014496368458030961,
      "loss": 0.2303,
      "num_input_tokens_seen": 21472416,
      "step": 101740
    },
    {
      "epoch": 11.193069306930694,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.014494929231632004,
      "loss": 0.2298,
      "num_input_tokens_seen": 21473472,
      "step": 101745
    },
    {
      "epoch": 11.193619361936193,
      "grad_norm": 0.000553131103515625,
      "learning_rate": 0.014493490009888046,
      "loss": 0.2324,
      "num_input_tokens_seen": 21474464,
      "step": 101750
    },
    {
      "epoch": 11.194169416941694,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01449205079281234,
      "loss": 0.2303,
      "num_input_tokens_seen": 21475520,
      "step": 101755
    },
    {
      "epoch": 11.194719471947195,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.014490611580418162,
      "loss": 0.2303,
      "num_input_tokens_seen": 21476576,
      "step": 101760
    },
    {
      "epoch": 11.195269526952695,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014489172372718776,
      "loss": 0.2309,
      "num_input_tokens_seen": 21477536,
      "step": 101765
    },
    {
      "epoch": 11.195819581958196,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.014487733169727436,
      "loss": 0.2309,
      "num_input_tokens_seen": 21478528,
      "step": 101770
    },
    {
      "epoch": 11.196369636963697,
      "grad_norm": 0.000415802001953125,
      "learning_rate": 0.014486293971457417,
      "loss": 0.2298,
      "num_input_tokens_seen": 21479584,
      "step": 101775
    },
    {
      "epoch": 11.196919691969198,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.014484854777921977,
      "loss": 0.2293,
      "num_input_tokens_seen": 21480608,
      "step": 101780
    },
    {
      "epoch": 11.197469746974697,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.014483415589134388,
      "loss": 0.2309,
      "num_input_tokens_seen": 21481632,
      "step": 101785
    },
    {
      "epoch": 11.198019801980198,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01448197640510791,
      "loss": 0.2303,
      "num_input_tokens_seen": 21482624,
      "step": 101790
    },
    {
      "epoch": 11.1985698569857,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014480537225855801,
      "loss": 0.2309,
      "num_input_tokens_seen": 21483648,
      "step": 101795
    },
    {
      "epoch": 11.199119911991199,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.014479098051391335,
      "loss": 0.2319,
      "num_input_tokens_seen": 21484704,
      "step": 101800
    },
    {
      "epoch": 11.1996699669967,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01447765888172777,
      "loss": 0.2298,
      "num_input_tokens_seen": 21485728,
      "step": 101805
    },
    {
      "epoch": 11.2002200220022,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.014476219716878376,
      "loss": 0.2324,
      "num_input_tokens_seen": 21486784,
      "step": 101810
    },
    {
      "epoch": 11.2007700770077,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.014474780556856415,
      "loss": 0.2309,
      "num_input_tokens_seen": 21487872,
      "step": 101815
    },
    {
      "epoch": 11.201320132013201,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.014473341401675143,
      "loss": 0.2314,
      "num_input_tokens_seen": 21488832,
      "step": 101820
    },
    {
      "epoch": 11.201870187018702,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.014471902251347837,
      "loss": 0.2314,
      "num_input_tokens_seen": 21489952,
      "step": 101825
    },
    {
      "epoch": 11.202420242024202,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.01447046310588775,
      "loss": 0.2324,
      "num_input_tokens_seen": 21491008,
      "step": 101830
    },
    {
      "epoch": 11.202970297029703,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01446902396530815,
      "loss": 0.2324,
      "num_input_tokens_seen": 21492064,
      "step": 101835
    },
    {
      "epoch": 11.203520352035204,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.014467584829622309,
      "loss": 0.2324,
      "num_input_tokens_seen": 21493152,
      "step": 101840
    },
    {
      "epoch": 11.204070407040705,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.014466145698843476,
      "loss": 0.2324,
      "num_input_tokens_seen": 21494272,
      "step": 101845
    },
    {
      "epoch": 11.204620462046204,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014464706572984922,
      "loss": 0.2308,
      "num_input_tokens_seen": 21495392,
      "step": 101850
    },
    {
      "epoch": 11.205170517051705,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.014463267452059909,
      "loss": 0.2319,
      "num_input_tokens_seen": 21496352,
      "step": 101855
    },
    {
      "epoch": 11.205720572057206,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014461828336081707,
      "loss": 0.2319,
      "num_input_tokens_seen": 21497376,
      "step": 101860
    },
    {
      "epoch": 11.206270627062706,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014460389225063576,
      "loss": 0.2329,
      "num_input_tokens_seen": 21498464,
      "step": 101865
    },
    {
      "epoch": 11.206820682068207,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.014458950119018774,
      "loss": 0.2314,
      "num_input_tokens_seen": 21499520,
      "step": 101870
    },
    {
      "epoch": 11.207370737073708,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.014457511017960572,
      "loss": 0.2314,
      "num_input_tokens_seen": 21500608,
      "step": 101875
    },
    {
      "epoch": 11.207920792079207,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014456071921902227,
      "loss": 0.2309,
      "num_input_tokens_seen": 21501696,
      "step": 101880
    },
    {
      "epoch": 11.208470847084708,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.014454632830857012,
      "loss": 0.2324,
      "num_input_tokens_seen": 21502752,
      "step": 101885
    },
    {
      "epoch": 11.20902090209021,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014453193744838182,
      "loss": 0.2298,
      "num_input_tokens_seen": 21503840,
      "step": 101890
    },
    {
      "epoch": 11.209570957095709,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.014451754663859,
      "loss": 0.2309,
      "num_input_tokens_seen": 21504832,
      "step": 101895
    },
    {
      "epoch": 11.21012101210121,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.014450315587932737,
      "loss": 0.2314,
      "num_input_tokens_seen": 21505888,
      "step": 101900
    },
    {
      "epoch": 11.210671067106711,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.014448876517072646,
      "loss": 0.2309,
      "num_input_tokens_seen": 21506912,
      "step": 101905
    },
    {
      "epoch": 11.211221122112212,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.014447437451291999,
      "loss": 0.2319,
      "num_input_tokens_seen": 21508032,
      "step": 101910
    },
    {
      "epoch": 11.211771177117711,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.01444599839060406,
      "loss": 0.2309,
      "num_input_tokens_seen": 21509088,
      "step": 101915
    },
    {
      "epoch": 11.212321232123212,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014444559335022077,
      "loss": 0.233,
      "num_input_tokens_seen": 21510144,
      "step": 101920
    },
    {
      "epoch": 11.212871287128714,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.014443120284559333,
      "loss": 0.2314,
      "num_input_tokens_seen": 21511168,
      "step": 101925
    },
    {
      "epoch": 11.213421342134213,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.014441681239229079,
      "loss": 0.2308,
      "num_input_tokens_seen": 21512320,
      "step": 101930
    },
    {
      "epoch": 11.213971397139714,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.014440242199044582,
      "loss": 0.2298,
      "num_input_tokens_seen": 21513440,
      "step": 101935
    },
    {
      "epoch": 11.214521452145215,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014438803164019107,
      "loss": 0.233,
      "num_input_tokens_seen": 21514528,
      "step": 101940
    },
    {
      "epoch": 11.215071507150714,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.014437364134165909,
      "loss": 0.234,
      "num_input_tokens_seen": 21515584,
      "step": 101945
    },
    {
      "epoch": 11.215621562156215,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.014435925109498256,
      "loss": 0.2314,
      "num_input_tokens_seen": 21516672,
      "step": 101950
    },
    {
      "epoch": 11.216171617161717,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.014434486090029416,
      "loss": 0.2303,
      "num_input_tokens_seen": 21517728,
      "step": 101955
    },
    {
      "epoch": 11.216721672167218,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.01443304707577264,
      "loss": 0.2303,
      "num_input_tokens_seen": 21518784,
      "step": 101960
    },
    {
      "epoch": 11.217271727172717,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.014431608066741201,
      "loss": 0.233,
      "num_input_tokens_seen": 21519840,
      "step": 101965
    },
    {
      "epoch": 11.217821782178218,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.014430169062948353,
      "loss": 0.2324,
      "num_input_tokens_seen": 21520832,
      "step": 101970
    },
    {
      "epoch": 11.218371837183719,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.014428730064407367,
      "loss": 0.2303,
      "num_input_tokens_seen": 21521952,
      "step": 101975
    },
    {
      "epoch": 11.218921892189218,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014427291071131505,
      "loss": 0.2324,
      "num_input_tokens_seen": 21523008,
      "step": 101980
    },
    {
      "epoch": 11.21947194719472,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01442585208313402,
      "loss": 0.2303,
      "num_input_tokens_seen": 21524064,
      "step": 101985
    },
    {
      "epoch": 11.22002200220022,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.014424413100428184,
      "loss": 0.2308,
      "num_input_tokens_seen": 21525184,
      "step": 101990
    },
    {
      "epoch": 11.22057205720572,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.014422974123027253,
      "loss": 0.2314,
      "num_input_tokens_seen": 21526240,
      "step": 101995
    },
    {
      "epoch": 11.221122112211221,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.0144215351509445,
      "loss": 0.2309,
      "num_input_tokens_seen": 21527328,
      "step": 102000
    },
    {
      "epoch": 11.221672167216722,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.014420096184193175,
      "loss": 0.2309,
      "num_input_tokens_seen": 21528384,
      "step": 102005
    },
    {
      "epoch": 11.222222222222221,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.014418657222786544,
      "loss": 0.2313,
      "num_input_tokens_seen": 21529408,
      "step": 102010
    },
    {
      "epoch": 11.222772277227723,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.014417218266737876,
      "loss": 0.2303,
      "num_input_tokens_seen": 21530496,
      "step": 102015
    },
    {
      "epoch": 11.223322332233224,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.01441577931606042,
      "loss": 0.2319,
      "num_input_tokens_seen": 21531552,
      "step": 102020
    },
    {
      "epoch": 11.223872387238725,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01441434037076745,
      "loss": 0.2309,
      "num_input_tokens_seen": 21532640,
      "step": 102025
    },
    {
      "epoch": 11.224422442244224,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.014412901430872227,
      "loss": 0.2272,
      "num_input_tokens_seen": 21533728,
      "step": 102030
    },
    {
      "epoch": 11.224972497249725,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.014411462496388,
      "loss": 0.2319,
      "num_input_tokens_seen": 21534752,
      "step": 102035
    },
    {
      "epoch": 11.225522552255226,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01441002356732805,
      "loss": 0.2329,
      "num_input_tokens_seen": 21535840,
      "step": 102040
    },
    {
      "epoch": 11.226072607260726,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.014408584643705624,
      "loss": 0.2308,
      "num_input_tokens_seen": 21536832,
      "step": 102045
    },
    {
      "epoch": 11.226622662266227,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.014407145725533994,
      "loss": 0.2324,
      "num_input_tokens_seen": 21537888,
      "step": 102050
    },
    {
      "epoch": 11.227172717271728,
      "grad_norm": 0.000530242919921875,
      "learning_rate": 0.014405706812826418,
      "loss": 0.2335,
      "num_input_tokens_seen": 21538976,
      "step": 102055
    },
    {
      "epoch": 11.227722772277227,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01440426790559615,
      "loss": 0.2319,
      "num_input_tokens_seen": 21540064,
      "step": 102060
    },
    {
      "epoch": 11.228272827282728,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014402829003856466,
      "loss": 0.2335,
      "num_input_tokens_seen": 21541152,
      "step": 102065
    },
    {
      "epoch": 11.22882288228823,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.014401390107620614,
      "loss": 0.2314,
      "num_input_tokens_seen": 21542240,
      "step": 102070
    },
    {
      "epoch": 11.229372937293729,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014399951216901869,
      "loss": 0.2319,
      "num_input_tokens_seen": 21543296,
      "step": 102075
    },
    {
      "epoch": 11.22992299229923,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.014398512331713485,
      "loss": 0.2298,
      "num_input_tokens_seen": 21544288,
      "step": 102080
    },
    {
      "epoch": 11.23047304730473,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.014397073452068719,
      "loss": 0.2324,
      "num_input_tokens_seen": 21545280,
      "step": 102085
    },
    {
      "epoch": 11.231023102310232,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.014395634577980843,
      "loss": 0.2309,
      "num_input_tokens_seen": 21546336,
      "step": 102090
    },
    {
      "epoch": 11.231573157315731,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.014394195709463105,
      "loss": 0.2329,
      "num_input_tokens_seen": 21547360,
      "step": 102095
    },
    {
      "epoch": 11.232123212321232,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.014392756846528784,
      "loss": 0.2298,
      "num_input_tokens_seen": 21548416,
      "step": 102100
    },
    {
      "epoch": 11.232673267326733,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014391317989191129,
      "loss": 0.2319,
      "num_input_tokens_seen": 21549504,
      "step": 102105
    },
    {
      "epoch": 11.233223322332233,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.0143898791374634,
      "loss": 0.2345,
      "num_input_tokens_seen": 21550560,
      "step": 102110
    },
    {
      "epoch": 11.233773377337734,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.014388440291358868,
      "loss": 0.2308,
      "num_input_tokens_seen": 21551648,
      "step": 102115
    },
    {
      "epoch": 11.234323432343235,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014387001450890783,
      "loss": 0.2308,
      "num_input_tokens_seen": 21552768,
      "step": 102120
    },
    {
      "epoch": 11.234873487348734,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.014385562616072415,
      "loss": 0.2335,
      "num_input_tokens_seen": 21553888,
      "step": 102125
    },
    {
      "epoch": 11.235423542354235,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.014384123786917023,
      "loss": 0.233,
      "num_input_tokens_seen": 21554880,
      "step": 102130
    },
    {
      "epoch": 11.235973597359736,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.014382684963437864,
      "loss": 0.2309,
      "num_input_tokens_seen": 21555936,
      "step": 102135
    },
    {
      "epoch": 11.236523652365236,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.0143812461456482,
      "loss": 0.2309,
      "num_input_tokens_seen": 21556992,
      "step": 102140
    },
    {
      "epoch": 11.237073707370737,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.0143798073335613,
      "loss": 0.2293,
      "num_input_tokens_seen": 21558080,
      "step": 102145
    },
    {
      "epoch": 11.237623762376238,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.014378368527190408,
      "loss": 0.2308,
      "num_input_tokens_seen": 21559104,
      "step": 102150
    },
    {
      "epoch": 11.238173817381739,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.014376929726548802,
      "loss": 0.2303,
      "num_input_tokens_seen": 21560096,
      "step": 102155
    },
    {
      "epoch": 11.238723872387238,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.01437549093164973,
      "loss": 0.233,
      "num_input_tokens_seen": 21561152,
      "step": 102160
    },
    {
      "epoch": 11.23927392739274,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.014374052142506465,
      "loss": 0.2303,
      "num_input_tokens_seen": 21562240,
      "step": 102165
    },
    {
      "epoch": 11.23982398239824,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014372613359132261,
      "loss": 0.2314,
      "num_input_tokens_seen": 21563296,
      "step": 102170
    },
    {
      "epoch": 11.24037403740374,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014371174581540374,
      "loss": 0.2324,
      "num_input_tokens_seen": 21564288,
      "step": 102175
    },
    {
      "epoch": 11.24092409240924,
      "grad_norm": 0.000698089599609375,
      "learning_rate": 0.01436973580974407,
      "loss": 0.2319,
      "num_input_tokens_seen": 21565312,
      "step": 102180
    },
    {
      "epoch": 11.241474147414742,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.014368297043756608,
      "loss": 0.2314,
      "num_input_tokens_seen": 21566368,
      "step": 102185
    },
    {
      "epoch": 11.242024202420241,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014366858283591254,
      "loss": 0.2324,
      "num_input_tokens_seen": 21567456,
      "step": 102190
    },
    {
      "epoch": 11.242574257425742,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.01436541952926126,
      "loss": 0.2324,
      "num_input_tokens_seen": 21568480,
      "step": 102195
    },
    {
      "epoch": 11.243124312431243,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.014363980780779887,
      "loss": 0.2324,
      "num_input_tokens_seen": 21569472,
      "step": 102200
    },
    {
      "epoch": 11.243674367436745,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.014362542038160406,
      "loss": 0.2319,
      "num_input_tokens_seen": 21570592,
      "step": 102205
    },
    {
      "epoch": 11.244224422442244,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.01436110330141606,
      "loss": 0.2303,
      "num_input_tokens_seen": 21571616,
      "step": 102210
    },
    {
      "epoch": 11.244774477447745,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.014359664570560123,
      "loss": 0.2303,
      "num_input_tokens_seen": 21572672,
      "step": 102215
    },
    {
      "epoch": 11.245324532453246,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01435822584560585,
      "loss": 0.2308,
      "num_input_tokens_seen": 21573664,
      "step": 102220
    },
    {
      "epoch": 11.245874587458745,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.014356787126566498,
      "loss": 0.2298,
      "num_input_tokens_seen": 21574720,
      "step": 102225
    },
    {
      "epoch": 11.246424642464246,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.014355348413455336,
      "loss": 0.2324,
      "num_input_tokens_seen": 21575808,
      "step": 102230
    },
    {
      "epoch": 11.246974697469748,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014353909706285615,
      "loss": 0.2319,
      "num_input_tokens_seen": 21576896,
      "step": 102235
    },
    {
      "epoch": 11.247524752475247,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.014352471005070599,
      "loss": 0.2303,
      "num_input_tokens_seen": 21577952,
      "step": 102240
    },
    {
      "epoch": 11.248074807480748,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01435103230982355,
      "loss": 0.2314,
      "num_input_tokens_seen": 21579008,
      "step": 102245
    },
    {
      "epoch": 11.248624862486249,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014349593620557719,
      "loss": 0.2308,
      "num_input_tokens_seen": 21580064,
      "step": 102250
    },
    {
      "epoch": 11.249174917491748,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.014348154937286375,
      "loss": 0.2324,
      "num_input_tokens_seen": 21581088,
      "step": 102255
    },
    {
      "epoch": 11.24972497249725,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014346716260022771,
      "loss": 0.2309,
      "num_input_tokens_seen": 21582144,
      "step": 102260
    },
    {
      "epoch": 11.25027502750275,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.014345277588780177,
      "loss": 0.2314,
      "num_input_tokens_seen": 21583200,
      "step": 102265
    },
    {
      "epoch": 11.250825082508252,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01434383892357184,
      "loss": 0.2313,
      "num_input_tokens_seen": 21584256,
      "step": 102270
    },
    {
      "epoch": 11.251375137513751,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.014342400264411023,
      "loss": 0.2314,
      "num_input_tokens_seen": 21585312,
      "step": 102275
    },
    {
      "epoch": 11.251925192519252,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.014340961611310989,
      "loss": 0.2335,
      "num_input_tokens_seen": 21586400,
      "step": 102280
    },
    {
      "epoch": 11.252475247524753,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014339522964284994,
      "loss": 0.233,
      "num_input_tokens_seen": 21587456,
      "step": 102285
    },
    {
      "epoch": 11.253025302530252,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.014338084323346304,
      "loss": 0.2308,
      "num_input_tokens_seen": 21588544,
      "step": 102290
    },
    {
      "epoch": 11.253575357535754,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.014336645688508171,
      "loss": 0.2319,
      "num_input_tokens_seen": 21589600,
      "step": 102295
    },
    {
      "epoch": 11.254125412541255,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.014335207059783854,
      "loss": 0.2329,
      "num_input_tokens_seen": 21590720,
      "step": 102300
    },
    {
      "epoch": 11.254675467546754,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.014333768437186619,
      "loss": 0.2324,
      "num_input_tokens_seen": 21591776,
      "step": 102305
    },
    {
      "epoch": 11.255225522552255,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014332329820729716,
      "loss": 0.2308,
      "num_input_tokens_seen": 21592832,
      "step": 102310
    },
    {
      "epoch": 11.255775577557756,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.01433089121042641,
      "loss": 0.2314,
      "num_input_tokens_seen": 21593952,
      "step": 102315
    },
    {
      "epoch": 11.256325632563255,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014329452606289964,
      "loss": 0.2309,
      "num_input_tokens_seen": 21595072,
      "step": 102320
    },
    {
      "epoch": 11.256875687568757,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.014328014008333624,
      "loss": 0.2329,
      "num_input_tokens_seen": 21596160,
      "step": 102325
    },
    {
      "epoch": 11.257425742574258,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.01432657541657066,
      "loss": 0.2319,
      "num_input_tokens_seen": 21597216,
      "step": 102330
    },
    {
      "epoch": 11.257975797579759,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014325136831014327,
      "loss": 0.2335,
      "num_input_tokens_seen": 21598272,
      "step": 102335
    },
    {
      "epoch": 11.258525852585258,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.014323698251677877,
      "loss": 0.2335,
      "num_input_tokens_seen": 21599360,
      "step": 102340
    },
    {
      "epoch": 11.25907590759076,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.014322259678574588,
      "loss": 0.2298,
      "num_input_tokens_seen": 21600416,
      "step": 102345
    },
    {
      "epoch": 11.25962596259626,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.014320821111717697,
      "loss": 0.2303,
      "num_input_tokens_seen": 21601408,
      "step": 102350
    },
    {
      "epoch": 11.26017601760176,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.014319382551120476,
      "loss": 0.233,
      "num_input_tokens_seen": 21602464,
      "step": 102355
    },
    {
      "epoch": 11.26072607260726,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.014317943996796182,
      "loss": 0.2303,
      "num_input_tokens_seen": 21603488,
      "step": 102360
    },
    {
      "epoch": 11.261276127612762,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014316505448758064,
      "loss": 0.2335,
      "num_input_tokens_seen": 21604544,
      "step": 102365
    },
    {
      "epoch": 11.261826182618261,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.014315066907019391,
      "loss": 0.2319,
      "num_input_tokens_seen": 21605632,
      "step": 102370
    },
    {
      "epoch": 11.262376237623762,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.014313628371593416,
      "loss": 0.2314,
      "num_input_tokens_seen": 21606624,
      "step": 102375
    },
    {
      "epoch": 11.262926292629263,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.014312189842493404,
      "loss": 0.233,
      "num_input_tokens_seen": 21607680,
      "step": 102380
    },
    {
      "epoch": 11.263476347634764,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.014310751319732605,
      "loss": 0.2324,
      "num_input_tokens_seen": 21608704,
      "step": 102385
    },
    {
      "epoch": 11.264026402640264,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.014309312803324276,
      "loss": 0.2314,
      "num_input_tokens_seen": 21609792,
      "step": 102390
    },
    {
      "epoch": 11.264576457645765,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014307874293281687,
      "loss": 0.2319,
      "num_input_tokens_seen": 21610816,
      "step": 102395
    },
    {
      "epoch": 11.265126512651266,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01430643578961808,
      "loss": 0.2329,
      "num_input_tokens_seen": 21611904,
      "step": 102400
    },
    {
      "epoch": 11.265676567656765,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014304997292346732,
      "loss": 0.2303,
      "num_input_tokens_seen": 21612960,
      "step": 102405
    },
    {
      "epoch": 11.266226622662266,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.014303558801480884,
      "loss": 0.2319,
      "num_input_tokens_seen": 21613952,
      "step": 102410
    },
    {
      "epoch": 11.266776677667767,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014302120317033798,
      "loss": 0.233,
      "num_input_tokens_seen": 21614976,
      "step": 102415
    },
    {
      "epoch": 11.267326732673267,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014300681839018742,
      "loss": 0.2324,
      "num_input_tokens_seen": 21615968,
      "step": 102420
    },
    {
      "epoch": 11.267876787678768,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.014299243367448958,
      "loss": 0.2314,
      "num_input_tokens_seen": 21617056,
      "step": 102425
    },
    {
      "epoch": 11.268426842684269,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.014297804902337716,
      "loss": 0.2314,
      "num_input_tokens_seen": 21618144,
      "step": 102430
    },
    {
      "epoch": 11.268976897689768,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.014296366443698272,
      "loss": 0.2303,
      "num_input_tokens_seen": 21619136,
      "step": 102435
    },
    {
      "epoch": 11.26952695269527,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014294927991543874,
      "loss": 0.2314,
      "num_input_tokens_seen": 21620192,
      "step": 102440
    },
    {
      "epoch": 11.27007700770077,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.01429348954588779,
      "loss": 0.2319,
      "num_input_tokens_seen": 21621280,
      "step": 102445
    },
    {
      "epoch": 11.270627062706271,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.014292051106743272,
      "loss": 0.2314,
      "num_input_tokens_seen": 21622304,
      "step": 102450
    },
    {
      "epoch": 11.27117711771177,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.014290612674123584,
      "loss": 0.2324,
      "num_input_tokens_seen": 21623360,
      "step": 102455
    },
    {
      "epoch": 11.271727172717272,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.014289174248041974,
      "loss": 0.2309,
      "num_input_tokens_seen": 21624384,
      "step": 102460
    },
    {
      "epoch": 11.272277227722773,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014287735828511703,
      "loss": 0.233,
      "num_input_tokens_seen": 21625408,
      "step": 102465
    },
    {
      "epoch": 11.272827282728272,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014286297415546033,
      "loss": 0.2319,
      "num_input_tokens_seen": 21626528,
      "step": 102470
    },
    {
      "epoch": 11.273377337733773,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.014284859009158211,
      "loss": 0.2304,
      "num_input_tokens_seen": 21627552,
      "step": 102475
    },
    {
      "epoch": 11.273927392739274,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014283420609361509,
      "loss": 0.2309,
      "num_input_tokens_seen": 21628576,
      "step": 102480
    },
    {
      "epoch": 11.274477447744774,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014281982216169171,
      "loss": 0.2303,
      "num_input_tokens_seen": 21629600,
      "step": 102485
    },
    {
      "epoch": 11.275027502750275,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014280543829594455,
      "loss": 0.2293,
      "num_input_tokens_seen": 21630656,
      "step": 102490
    },
    {
      "epoch": 11.275577557755776,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01427910544965063,
      "loss": 0.234,
      "num_input_tokens_seen": 21631712,
      "step": 102495
    },
    {
      "epoch": 11.276127612761275,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.014277667076350937,
      "loss": 0.2309,
      "num_input_tokens_seen": 21632736,
      "step": 102500
    },
    {
      "epoch": 11.276677667766776,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.014276228709708641,
      "loss": 0.2314,
      "num_input_tokens_seen": 21633824,
      "step": 102505
    },
    {
      "epoch": 11.277227722772277,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.014274790349737002,
      "loss": 0.2293,
      "num_input_tokens_seen": 21634912,
      "step": 102510
    },
    {
      "epoch": 11.277777777777779,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.014273351996449265,
      "loss": 0.2314,
      "num_input_tokens_seen": 21635968,
      "step": 102515
    },
    {
      "epoch": 11.278327832783278,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.0142719136498587,
      "loss": 0.2335,
      "num_input_tokens_seen": 21637024,
      "step": 102520
    },
    {
      "epoch": 11.278877887788779,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.014270475309978556,
      "loss": 0.2303,
      "num_input_tokens_seen": 21638112,
      "step": 102525
    },
    {
      "epoch": 11.27942794279428,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.014269036976822087,
      "loss": 0.2319,
      "num_input_tokens_seen": 21639168,
      "step": 102530
    },
    {
      "epoch": 11.27997799779978,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014267598650402563,
      "loss": 0.2324,
      "num_input_tokens_seen": 21640192,
      "step": 102535
    },
    {
      "epoch": 11.28052805280528,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01426616033073322,
      "loss": 0.2309,
      "num_input_tokens_seen": 21641184,
      "step": 102540
    },
    {
      "epoch": 11.281078107810782,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.014264722017827332,
      "loss": 0.2309,
      "num_input_tokens_seen": 21642240,
      "step": 102545
    },
    {
      "epoch": 11.281628162816281,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.014263283711698148,
      "loss": 0.2329,
      "num_input_tokens_seen": 21643296,
      "step": 102550
    },
    {
      "epoch": 11.282178217821782,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.014261845412358922,
      "loss": 0.233,
      "num_input_tokens_seen": 21644352,
      "step": 102555
    },
    {
      "epoch": 11.282728272827283,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.014260407119822913,
      "loss": 0.2314,
      "num_input_tokens_seen": 21645408,
      "step": 102560
    },
    {
      "epoch": 11.283278327832782,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.014258968834103374,
      "loss": 0.2319,
      "num_input_tokens_seen": 21646528,
      "step": 102565
    },
    {
      "epoch": 11.283828382838283,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01425753055521357,
      "loss": 0.2314,
      "num_input_tokens_seen": 21647584,
      "step": 102570
    },
    {
      "epoch": 11.284378437843785,
      "grad_norm": 0.00064849853515625,
      "learning_rate": 0.014256092283166748,
      "loss": 0.2335,
      "num_input_tokens_seen": 21648640,
      "step": 102575
    },
    {
      "epoch": 11.284928492849286,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.014254654017976163,
      "loss": 0.2304,
      "num_input_tokens_seen": 21649728,
      "step": 102580
    },
    {
      "epoch": 11.285478547854785,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.014253215759655078,
      "loss": 0.2309,
      "num_input_tokens_seen": 21650752,
      "step": 102585
    },
    {
      "epoch": 11.286028602860286,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.01425177750821674,
      "loss": 0.2319,
      "num_input_tokens_seen": 21651776,
      "step": 102590
    },
    {
      "epoch": 11.286578657865787,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.014250339263674418,
      "loss": 0.2319,
      "num_input_tokens_seen": 21652832,
      "step": 102595
    },
    {
      "epoch": 11.287128712871286,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014248901026041354,
      "loss": 0.2324,
      "num_input_tokens_seen": 21653920,
      "step": 102600
    },
    {
      "epoch": 11.287678767876788,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.014247462795330809,
      "loss": 0.2324,
      "num_input_tokens_seen": 21655104,
      "step": 102605
    },
    {
      "epoch": 11.288228822882289,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.014246024571556042,
      "loss": 0.2304,
      "num_input_tokens_seen": 21656128,
      "step": 102610
    },
    {
      "epoch": 11.288778877887788,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014244586354730299,
      "loss": 0.2335,
      "num_input_tokens_seen": 21657184,
      "step": 102615
    },
    {
      "epoch": 11.289328932893289,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014243148144866845,
      "loss": 0.233,
      "num_input_tokens_seen": 21658208,
      "step": 102620
    },
    {
      "epoch": 11.28987898789879,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.014241709941978932,
      "loss": 0.233,
      "num_input_tokens_seen": 21659328,
      "step": 102625
    },
    {
      "epoch": 11.290429042904291,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.014240271746079811,
      "loss": 0.2309,
      "num_input_tokens_seen": 21660416,
      "step": 102630
    },
    {
      "epoch": 11.29097909790979,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.014238833557182744,
      "loss": 0.2298,
      "num_input_tokens_seen": 21661472,
      "step": 102635
    },
    {
      "epoch": 11.291529152915292,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.01423739537530098,
      "loss": 0.2303,
      "num_input_tokens_seen": 21662592,
      "step": 102640
    },
    {
      "epoch": 11.292079207920793,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014235957200447779,
      "loss": 0.234,
      "num_input_tokens_seen": 21663584,
      "step": 102645
    },
    {
      "epoch": 11.292629262926292,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.014234519032636396,
      "loss": 0.2298,
      "num_input_tokens_seen": 21664640,
      "step": 102650
    },
    {
      "epoch": 11.293179317931793,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.014233080871880077,
      "loss": 0.2304,
      "num_input_tokens_seen": 21665664,
      "step": 102655
    },
    {
      "epoch": 11.293729372937294,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014231642718192088,
      "loss": 0.2319,
      "num_input_tokens_seen": 21666784,
      "step": 102660
    },
    {
      "epoch": 11.294279427942794,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.014230204571585678,
      "loss": 0.234,
      "num_input_tokens_seen": 21667808,
      "step": 102665
    },
    {
      "epoch": 11.294829482948295,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.014228766432074106,
      "loss": 0.2304,
      "num_input_tokens_seen": 21668832,
      "step": 102670
    },
    {
      "epoch": 11.295379537953796,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.014227328299670624,
      "loss": 0.232,
      "num_input_tokens_seen": 21669856,
      "step": 102675
    },
    {
      "epoch": 11.295929592959295,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.014225890174388479,
      "loss": 0.2335,
      "num_input_tokens_seen": 21670848,
      "step": 102680
    },
    {
      "epoch": 11.296479647964796,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.014224452056240943,
      "loss": 0.2304,
      "num_input_tokens_seen": 21671872,
      "step": 102685
    },
    {
      "epoch": 11.297029702970297,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.014223013945241251,
      "loss": 0.2293,
      "num_input_tokens_seen": 21672960,
      "step": 102690
    },
    {
      "epoch": 11.297579757975798,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014221575841402672,
      "loss": 0.2293,
      "num_input_tokens_seen": 21674048,
      "step": 102695
    },
    {
      "epoch": 11.298129812981298,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.014220137744738457,
      "loss": 0.2325,
      "num_input_tokens_seen": 21675136,
      "step": 102700
    },
    {
      "epoch": 11.298679867986799,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01421869965526185,
      "loss": 0.2345,
      "num_input_tokens_seen": 21676192,
      "step": 102705
    },
    {
      "epoch": 11.2992299229923,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014217261572986122,
      "loss": 0.2314,
      "num_input_tokens_seen": 21677280,
      "step": 102710
    },
    {
      "epoch": 11.2997799779978,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.014215823497924517,
      "loss": 0.2335,
      "num_input_tokens_seen": 21678368,
      "step": 102715
    },
    {
      "epoch": 11.3003300330033,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.014214385430090284,
      "loss": 0.2309,
      "num_input_tokens_seen": 21679456,
      "step": 102720
    },
    {
      "epoch": 11.300880088008801,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01421294736949669,
      "loss": 0.2304,
      "num_input_tokens_seen": 21680512,
      "step": 102725
    },
    {
      "epoch": 11.3014301430143,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01421150931615698,
      "loss": 0.2309,
      "num_input_tokens_seen": 21681568,
      "step": 102730
    },
    {
      "epoch": 11.301980198019802,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.01421007127008441,
      "loss": 0.2309,
      "num_input_tokens_seen": 21682656,
      "step": 102735
    },
    {
      "epoch": 11.302530253025303,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.014208633231292239,
      "loss": 0.2304,
      "num_input_tokens_seen": 21683712,
      "step": 102740
    },
    {
      "epoch": 11.303080308030804,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014207195199793708,
      "loss": 0.234,
      "num_input_tokens_seen": 21684864,
      "step": 102745
    },
    {
      "epoch": 11.303630363036303,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.014205757175602081,
      "loss": 0.2314,
      "num_input_tokens_seen": 21685952,
      "step": 102750
    },
    {
      "epoch": 11.304180418041804,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.014204319158730608,
      "loss": 0.233,
      "num_input_tokens_seen": 21686976,
      "step": 102755
    },
    {
      "epoch": 11.304730473047305,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014202881149192544,
      "loss": 0.2324,
      "num_input_tokens_seen": 21688064,
      "step": 102760
    },
    {
      "epoch": 11.305280528052805,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.014201443147001148,
      "loss": 0.2335,
      "num_input_tokens_seen": 21689120,
      "step": 102765
    },
    {
      "epoch": 11.305830583058306,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01420000515216966,
      "loss": 0.2298,
      "num_input_tokens_seen": 21690240,
      "step": 102770
    },
    {
      "epoch": 11.306380638063807,
      "grad_norm": 0.0004367828369140625,
      "learning_rate": 0.014198567164711344,
      "loss": 0.2314,
      "num_input_tokens_seen": 21691296,
      "step": 102775
    },
    {
      "epoch": 11.306930693069306,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014197129184639447,
      "loss": 0.2298,
      "num_input_tokens_seen": 21692320,
      "step": 102780
    },
    {
      "epoch": 11.307480748074807,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01419569121196723,
      "loss": 0.2314,
      "num_input_tokens_seen": 21693376,
      "step": 102785
    },
    {
      "epoch": 11.308030803080309,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01419425324670794,
      "loss": 0.2329,
      "num_input_tokens_seen": 21694432,
      "step": 102790
    },
    {
      "epoch": 11.308580858085808,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.014192815288874826,
      "loss": 0.2303,
      "num_input_tokens_seen": 21695488,
      "step": 102795
    },
    {
      "epoch": 11.309130913091309,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.014191377338481153,
      "loss": 0.2303,
      "num_input_tokens_seen": 21696576,
      "step": 102800
    },
    {
      "epoch": 11.30968096809681,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014189939395540162,
      "loss": 0.2308,
      "num_input_tokens_seen": 21697664,
      "step": 102805
    },
    {
      "epoch": 11.310231023102311,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014188501460065112,
      "loss": 0.2309,
      "num_input_tokens_seen": 21698752,
      "step": 102810
    },
    {
      "epoch": 11.31078107810781,
      "grad_norm": 0.0005645751953125,
      "learning_rate": 0.01418706353206926,
      "loss": 0.2303,
      "num_input_tokens_seen": 21699744,
      "step": 102815
    },
    {
      "epoch": 11.311331133113312,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.014185625611565842,
      "loss": 0.234,
      "num_input_tokens_seen": 21700704,
      "step": 102820
    },
    {
      "epoch": 11.311881188118813,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014184187698568133,
      "loss": 0.2314,
      "num_input_tokens_seen": 21701728,
      "step": 102825
    },
    {
      "epoch": 11.312431243124312,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.014182749793089366,
      "loss": 0.2309,
      "num_input_tokens_seen": 21702880,
      "step": 102830
    },
    {
      "epoch": 11.312981298129813,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.014181311895142807,
      "loss": 0.2319,
      "num_input_tokens_seen": 21703968,
      "step": 102835
    },
    {
      "epoch": 11.313531353135314,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.014179874004741707,
      "loss": 0.2309,
      "num_input_tokens_seen": 21705024,
      "step": 102840
    },
    {
      "epoch": 11.314081408140813,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.014178436121899306,
      "loss": 0.2314,
      "num_input_tokens_seen": 21706112,
      "step": 102845
    },
    {
      "epoch": 11.314631463146315,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.01417699824662887,
      "loss": 0.2309,
      "num_input_tokens_seen": 21707168,
      "step": 102850
    },
    {
      "epoch": 11.315181518151816,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014175560378943644,
      "loss": 0.2319,
      "num_input_tokens_seen": 21708256,
      "step": 102855
    },
    {
      "epoch": 11.315731573157315,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.014174122518856887,
      "loss": 0.2324,
      "num_input_tokens_seen": 21709312,
      "step": 102860
    },
    {
      "epoch": 11.316281628162816,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.014172684666381843,
      "loss": 0.2319,
      "num_input_tokens_seen": 21710336,
      "step": 102865
    },
    {
      "epoch": 11.316831683168317,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014171246821531765,
      "loss": 0.2309,
      "num_input_tokens_seen": 21711360,
      "step": 102870
    },
    {
      "epoch": 11.317381738173818,
      "grad_norm": 0.004730224609375,
      "learning_rate": 0.014169808984319915,
      "loss": 0.2324,
      "num_input_tokens_seen": 21712416,
      "step": 102875
    },
    {
      "epoch": 11.317931793179318,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.014168371154759525,
      "loss": 0.2314,
      "num_input_tokens_seen": 21713504,
      "step": 102880
    },
    {
      "epoch": 11.318481848184819,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014166933332863871,
      "loss": 0.2319,
      "num_input_tokens_seen": 21714528,
      "step": 102885
    },
    {
      "epoch": 11.31903190319032,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.014165495518646189,
      "loss": 0.2319,
      "num_input_tokens_seen": 21715584,
      "step": 102890
    },
    {
      "epoch": 11.319581958195819,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01416405771211973,
      "loss": 0.2298,
      "num_input_tokens_seen": 21716672,
      "step": 102895
    },
    {
      "epoch": 11.32013201320132,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.014162619913297758,
      "loss": 0.2309,
      "num_input_tokens_seen": 21717792,
      "step": 102900
    },
    {
      "epoch": 11.320682068206821,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.014161182122193512,
      "loss": 0.2314,
      "num_input_tokens_seen": 21718848,
      "step": 102905
    },
    {
      "epoch": 11.32123212321232,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014159744338820245,
      "loss": 0.2319,
      "num_input_tokens_seen": 21719872,
      "step": 102910
    },
    {
      "epoch": 11.321782178217822,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.014158306563191218,
      "loss": 0.2319,
      "num_input_tokens_seen": 21720928,
      "step": 102915
    },
    {
      "epoch": 11.322332233223323,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.014156868795319669,
      "loss": 0.2304,
      "num_input_tokens_seen": 21722016,
      "step": 102920
    },
    {
      "epoch": 11.322882288228822,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.014155431035218859,
      "loss": 0.2324,
      "num_input_tokens_seen": 21723072,
      "step": 102925
    },
    {
      "epoch": 11.323432343234323,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.014153993282902039,
      "loss": 0.2298,
      "num_input_tokens_seen": 21724160,
      "step": 102930
    },
    {
      "epoch": 11.323982398239824,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014152555538382449,
      "loss": 0.2309,
      "num_input_tokens_seen": 21725184,
      "step": 102935
    },
    {
      "epoch": 11.324532453245325,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014151117801673352,
      "loss": 0.2335,
      "num_input_tokens_seen": 21726240,
      "step": 102940
    },
    {
      "epoch": 11.325082508250825,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.014149680072787994,
      "loss": 0.2314,
      "num_input_tokens_seen": 21727200,
      "step": 102945
    },
    {
      "epoch": 11.325632563256326,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.014148242351739627,
      "loss": 0.2314,
      "num_input_tokens_seen": 21728192,
      "step": 102950
    },
    {
      "epoch": 11.326182618261827,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014146804638541507,
      "loss": 0.2319,
      "num_input_tokens_seen": 21729216,
      "step": 102955
    },
    {
      "epoch": 11.326732673267326,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01414536693320687,
      "loss": 0.2324,
      "num_input_tokens_seen": 21730240,
      "step": 102960
    },
    {
      "epoch": 11.327282728272827,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.014143929235748982,
      "loss": 0.2345,
      "num_input_tokens_seen": 21731296,
      "step": 102965
    },
    {
      "epoch": 11.327832783278328,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.014142491546181083,
      "loss": 0.2319,
      "num_input_tokens_seen": 21732288,
      "step": 102970
    },
    {
      "epoch": 11.328382838283828,
      "grad_norm": 0.000568389892578125,
      "learning_rate": 0.014141053864516435,
      "loss": 0.2314,
      "num_input_tokens_seen": 21733312,
      "step": 102975
    },
    {
      "epoch": 11.328932893289329,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014139616190768277,
      "loss": 0.2303,
      "num_input_tokens_seen": 21734304,
      "step": 102980
    },
    {
      "epoch": 11.32948294829483,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.014138178524949861,
      "loss": 0.2304,
      "num_input_tokens_seen": 21735392,
      "step": 102985
    },
    {
      "epoch": 11.33003300330033,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014136740867074447,
      "loss": 0.2303,
      "num_input_tokens_seen": 21736448,
      "step": 102990
    },
    {
      "epoch": 11.33058305830583,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014135303217155271,
      "loss": 0.2309,
      "num_input_tokens_seen": 21737472,
      "step": 102995
    },
    {
      "epoch": 11.331133113311331,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014133865575205594,
      "loss": 0.2303,
      "num_input_tokens_seen": 21738496,
      "step": 103000
    },
    {
      "epoch": 11.331683168316832,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014132427941238663,
      "loss": 0.2309,
      "num_input_tokens_seen": 21739520,
      "step": 103005
    },
    {
      "epoch": 11.332233223322332,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.014130990315267723,
      "loss": 0.2314,
      "num_input_tokens_seen": 21740544,
      "step": 103010
    },
    {
      "epoch": 11.332783278327833,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014129552697306034,
      "loss": 0.2319,
      "num_input_tokens_seen": 21741600,
      "step": 103015
    },
    {
      "epoch": 11.333333333333334,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.014128115087366836,
      "loss": 0.2324,
      "num_input_tokens_seen": 21742656,
      "step": 103020
    },
    {
      "epoch": 11.333883388338833,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.014126677485463384,
      "loss": 0.2288,
      "num_input_tokens_seen": 21743744,
      "step": 103025
    },
    {
      "epoch": 11.334433443344334,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.014125239891608932,
      "loss": 0.2309,
      "num_input_tokens_seen": 21744736,
      "step": 103030
    },
    {
      "epoch": 11.334983498349835,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.014123802305816715,
      "loss": 0.2314,
      "num_input_tokens_seen": 21745728,
      "step": 103035
    },
    {
      "epoch": 11.335533553355335,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.014122364728099997,
      "loss": 0.2319,
      "num_input_tokens_seen": 21746784,
      "step": 103040
    },
    {
      "epoch": 11.336083608360836,
      "grad_norm": 0.00079345703125,
      "learning_rate": 0.014120927158472017,
      "loss": 0.2303,
      "num_input_tokens_seen": 21747840,
      "step": 103045
    },
    {
      "epoch": 11.336633663366337,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.014119489596946038,
      "loss": 0.2335,
      "num_input_tokens_seen": 21748800,
      "step": 103050
    },
    {
      "epoch": 11.337183718371838,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.014118052043535296,
      "loss": 0.2314,
      "num_input_tokens_seen": 21749888,
      "step": 103055
    },
    {
      "epoch": 11.337733773377337,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.014116614498253044,
      "loss": 0.2309,
      "num_input_tokens_seen": 21751040,
      "step": 103060
    },
    {
      "epoch": 11.338283828382838,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.014115176961112533,
      "loss": 0.2308,
      "num_input_tokens_seen": 21752064,
      "step": 103065
    },
    {
      "epoch": 11.33883388338834,
      "grad_norm": 0.00982666015625,
      "learning_rate": 0.01411373943212701,
      "loss": 0.2319,
      "num_input_tokens_seen": 21753120,
      "step": 103070
    },
    {
      "epoch": 11.339383938393839,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.014112301911309732,
      "loss": 0.2314,
      "num_input_tokens_seen": 21754080,
      "step": 103075
    },
    {
      "epoch": 11.33993399339934,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.014110864398673937,
      "loss": 0.2308,
      "num_input_tokens_seen": 21755104,
      "step": 103080
    },
    {
      "epoch": 11.340484048404841,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.014109426894232875,
      "loss": 0.2314,
      "num_input_tokens_seen": 21756192,
      "step": 103085
    },
    {
      "epoch": 11.34103410341034,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.014107989397999805,
      "loss": 0.2324,
      "num_input_tokens_seen": 21757216,
      "step": 103090
    },
    {
      "epoch": 11.341584158415841,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.014106551909987962,
      "loss": 0.2293,
      "num_input_tokens_seen": 21758272,
      "step": 103095
    },
    {
      "epoch": 11.342134213421343,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.014105114430210605,
      "loss": 0.2308,
      "num_input_tokens_seen": 21759392,
      "step": 103100
    },
    {
      "epoch": 11.342684268426842,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014103676958680983,
      "loss": 0.2303,
      "num_input_tokens_seen": 21760416,
      "step": 103105
    },
    {
      "epoch": 11.343234323432343,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.01410223949541233,
      "loss": 0.2314,
      "num_input_tokens_seen": 21761472,
      "step": 103110
    },
    {
      "epoch": 11.343784378437844,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.014100802040417911,
      "loss": 0.2314,
      "num_input_tokens_seen": 21762528,
      "step": 103115
    },
    {
      "epoch": 11.344334433443345,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01409936459371097,
      "loss": 0.2308,
      "num_input_tokens_seen": 21763648,
      "step": 103120
    },
    {
      "epoch": 11.344884488448844,
      "grad_norm": 0.0008392333984375,
      "learning_rate": 0.014097927155304746,
      "loss": 0.2308,
      "num_input_tokens_seen": 21764768,
      "step": 103125
    },
    {
      "epoch": 11.345434543454346,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.014096489725212503,
      "loss": 0.2298,
      "num_input_tokens_seen": 21765824,
      "step": 103130
    },
    {
      "epoch": 11.345984598459847,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014095052303447474,
      "loss": 0.2314,
      "num_input_tokens_seen": 21766848,
      "step": 103135
    },
    {
      "epoch": 11.346534653465346,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.014093614890022916,
      "loss": 0.2319,
      "num_input_tokens_seen": 21767872,
      "step": 103140
    },
    {
      "epoch": 11.347084708470847,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01409217748495208,
      "loss": 0.2345,
      "num_input_tokens_seen": 21768960,
      "step": 103145
    },
    {
      "epoch": 11.347634763476348,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.0140907400882482,
      "loss": 0.2304,
      "num_input_tokens_seen": 21770080,
      "step": 103150
    },
    {
      "epoch": 11.348184818481847,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.014089302699924538,
      "loss": 0.2309,
      "num_input_tokens_seen": 21771104,
      "step": 103155
    },
    {
      "epoch": 11.348734873487349,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014087865319994331,
      "loss": 0.2335,
      "num_input_tokens_seen": 21772192,
      "step": 103160
    },
    {
      "epoch": 11.34928492849285,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.014086427948470837,
      "loss": 0.233,
      "num_input_tokens_seen": 21773280,
      "step": 103165
    },
    {
      "epoch": 11.34983498349835,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.014084990585367297,
      "loss": 0.2309,
      "num_input_tokens_seen": 21774336,
      "step": 103170
    },
    {
      "epoch": 11.35038503850385,
      "grad_norm": 0.00081634521484375,
      "learning_rate": 0.014083553230696955,
      "loss": 0.233,
      "num_input_tokens_seen": 21775360,
      "step": 103175
    },
    {
      "epoch": 11.350935093509351,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014082115884473072,
      "loss": 0.2324,
      "num_input_tokens_seen": 21776448,
      "step": 103180
    },
    {
      "epoch": 11.351485148514852,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.014080678546708874,
      "loss": 0.2335,
      "num_input_tokens_seen": 21777536,
      "step": 103185
    },
    {
      "epoch": 11.352035203520352,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014079241217417632,
      "loss": 0.2298,
      "num_input_tokens_seen": 21778528,
      "step": 103190
    },
    {
      "epoch": 11.352585258525853,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.014077803896612577,
      "loss": 0.2319,
      "num_input_tokens_seen": 21779552,
      "step": 103195
    },
    {
      "epoch": 11.353135313531354,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01407636658430696,
      "loss": 0.2303,
      "num_input_tokens_seen": 21780544,
      "step": 103200
    },
    {
      "epoch": 11.353685368536853,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.014074929280514034,
      "loss": 0.2298,
      "num_input_tokens_seen": 21781536,
      "step": 103205
    },
    {
      "epoch": 11.354235423542354,
      "grad_norm": 0.000888824462890625,
      "learning_rate": 0.014073491985247035,
      "loss": 0.2308,
      "num_input_tokens_seen": 21782560,
      "step": 103210
    },
    {
      "epoch": 11.354785478547855,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.014072054698519219,
      "loss": 0.2293,
      "num_input_tokens_seen": 21783616,
      "step": 103215
    },
    {
      "epoch": 11.355335533553355,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.014070617420343831,
      "loss": 0.233,
      "num_input_tokens_seen": 21784704,
      "step": 103220
    },
    {
      "epoch": 11.355885588558856,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.014069180150734113,
      "loss": 0.2324,
      "num_input_tokens_seen": 21785824,
      "step": 103225
    },
    {
      "epoch": 11.356435643564357,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.014067742889703316,
      "loss": 0.2309,
      "num_input_tokens_seen": 21786848,
      "step": 103230
    },
    {
      "epoch": 11.356985698569858,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.014066305637264683,
      "loss": 0.2314,
      "num_input_tokens_seen": 21788000,
      "step": 103235
    },
    {
      "epoch": 11.357535753575357,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.01406486839343147,
      "loss": 0.2314,
      "num_input_tokens_seen": 21789024,
      "step": 103240
    },
    {
      "epoch": 11.358085808580858,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.014063431158216911,
      "loss": 0.2319,
      "num_input_tokens_seen": 21790048,
      "step": 103245
    },
    {
      "epoch": 11.35863586358636,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014061993931634256,
      "loss": 0.2324,
      "num_input_tokens_seen": 21791072,
      "step": 103250
    },
    {
      "epoch": 11.359185918591859,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014060556713696755,
      "loss": 0.2303,
      "num_input_tokens_seen": 21792160,
      "step": 103255
    },
    {
      "epoch": 11.35973597359736,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01405911950441765,
      "loss": 0.2319,
      "num_input_tokens_seen": 21793216,
      "step": 103260
    },
    {
      "epoch": 11.36028602860286,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.014057682303810194,
      "loss": 0.2319,
      "num_input_tokens_seen": 21794176,
      "step": 103265
    },
    {
      "epoch": 11.36083608360836,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014056245111887628,
      "loss": 0.2314,
      "num_input_tokens_seen": 21795296,
      "step": 103270
    },
    {
      "epoch": 11.361386138613861,
      "grad_norm": 0.00060272216796875,
      "learning_rate": 0.014054807928663192,
      "loss": 0.2298,
      "num_input_tokens_seen": 21796320,
      "step": 103275
    },
    {
      "epoch": 11.361936193619362,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.014053370754150144,
      "loss": 0.2298,
      "num_input_tokens_seen": 21797376,
      "step": 103280
    },
    {
      "epoch": 11.362486248624862,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.014051933588361718,
      "loss": 0.2304,
      "num_input_tokens_seen": 21798432,
      "step": 103285
    },
    {
      "epoch": 11.363036303630363,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014050496431311168,
      "loss": 0.2329,
      "num_input_tokens_seen": 21799520,
      "step": 103290
    },
    {
      "epoch": 11.363586358635864,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01404905928301174,
      "loss": 0.2324,
      "num_input_tokens_seen": 21800576,
      "step": 103295
    },
    {
      "epoch": 11.364136413641365,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.014047622143476667,
      "loss": 0.2308,
      "num_input_tokens_seen": 21801600,
      "step": 103300
    },
    {
      "epoch": 11.364686468646864,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.014046185012719212,
      "loss": 0.2314,
      "num_input_tokens_seen": 21802624,
      "step": 103305
    },
    {
      "epoch": 11.365236523652365,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.014044747890752612,
      "loss": 0.2319,
      "num_input_tokens_seen": 21803648,
      "step": 103310
    },
    {
      "epoch": 11.365786578657866,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.014043310777590105,
      "loss": 0.2298,
      "num_input_tokens_seen": 21804704,
      "step": 103315
    },
    {
      "epoch": 11.366336633663366,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014041873673244953,
      "loss": 0.234,
      "num_input_tokens_seen": 21805728,
      "step": 103320
    },
    {
      "epoch": 11.366886688668867,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.014040436577730384,
      "loss": 0.2319,
      "num_input_tokens_seen": 21806784,
      "step": 103325
    },
    {
      "epoch": 11.367436743674368,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.014038999491059656,
      "loss": 0.2319,
      "num_input_tokens_seen": 21807904,
      "step": 103330
    },
    {
      "epoch": 11.367986798679867,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.014037562413246009,
      "loss": 0.2314,
      "num_input_tokens_seen": 21808992,
      "step": 103335
    },
    {
      "epoch": 11.368536853685368,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.014036125344302682,
      "loss": 0.2324,
      "num_input_tokens_seen": 21809984,
      "step": 103340
    },
    {
      "epoch": 11.36908690869087,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014034688284242928,
      "loss": 0.2319,
      "num_input_tokens_seen": 21811040,
      "step": 103345
    },
    {
      "epoch": 11.369636963696369,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.014033251233079986,
      "loss": 0.2319,
      "num_input_tokens_seen": 21812128,
      "step": 103350
    },
    {
      "epoch": 11.37018701870187,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014031814190827108,
      "loss": 0.2314,
      "num_input_tokens_seen": 21813216,
      "step": 103355
    },
    {
      "epoch": 11.370737073707371,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014030377157497534,
      "loss": 0.2324,
      "num_input_tokens_seen": 21814240,
      "step": 103360
    },
    {
      "epoch": 11.371287128712872,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.014028940133104503,
      "loss": 0.2319,
      "num_input_tokens_seen": 21815296,
      "step": 103365
    },
    {
      "epoch": 11.371837183718371,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.01402750311766127,
      "loss": 0.2293,
      "num_input_tokens_seen": 21816352,
      "step": 103370
    },
    {
      "epoch": 11.372387238723872,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01402606611118107,
      "loss": 0.2308,
      "num_input_tokens_seen": 21817472,
      "step": 103375
    },
    {
      "epoch": 11.372937293729374,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.014024629113677156,
      "loss": 0.2319,
      "num_input_tokens_seen": 21818432,
      "step": 103380
    },
    {
      "epoch": 11.373487348734873,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.014023192125162766,
      "loss": 0.2345,
      "num_input_tokens_seen": 21819488,
      "step": 103385
    },
    {
      "epoch": 11.374037403740374,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.014021755145651143,
      "loss": 0.2314,
      "num_input_tokens_seen": 21820480,
      "step": 103390
    },
    {
      "epoch": 11.374587458745875,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.014020318175155538,
      "loss": 0.2329,
      "num_input_tokens_seen": 21821536,
      "step": 103395
    },
    {
      "epoch": 11.375137513751374,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.014018881213689185,
      "loss": 0.2303,
      "num_input_tokens_seen": 21822560,
      "step": 103400
    },
    {
      "epoch": 11.375687568756875,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.014017444261265336,
      "loss": 0.2324,
      "num_input_tokens_seen": 21823584,
      "step": 103405
    },
    {
      "epoch": 11.376237623762377,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.014016007317897237,
      "loss": 0.2308,
      "num_input_tokens_seen": 21824640,
      "step": 103410
    },
    {
      "epoch": 11.376787678767876,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.014014570383598116,
      "loss": 0.2314,
      "num_input_tokens_seen": 21825696,
      "step": 103415
    },
    {
      "epoch": 11.377337733773377,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.014013133458381233,
      "loss": 0.2298,
      "num_input_tokens_seen": 21826784,
      "step": 103420
    },
    {
      "epoch": 11.377887788778878,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.014011696542259821,
      "loss": 0.2308,
      "num_input_tokens_seen": 21827744,
      "step": 103425
    },
    {
      "epoch": 11.37843784378438,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01401025963524713,
      "loss": 0.2319,
      "num_input_tokens_seen": 21828800,
      "step": 103430
    },
    {
      "epoch": 11.378987898789878,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.014008822737356407,
      "loss": 0.2314,
      "num_input_tokens_seen": 21829824,
      "step": 103435
    },
    {
      "epoch": 11.37953795379538,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.014007385848600881,
      "loss": 0.2309,
      "num_input_tokens_seen": 21830944,
      "step": 103440
    },
    {
      "epoch": 11.38008800880088,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.014005948968993806,
      "loss": 0.2329,
      "num_input_tokens_seen": 21832064,
      "step": 103445
    },
    {
      "epoch": 11.38063806380638,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01400451209854842,
      "loss": 0.2308,
      "num_input_tokens_seen": 21833120,
      "step": 103450
    },
    {
      "epoch": 11.381188118811881,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.014003075237277974,
      "loss": 0.2298,
      "num_input_tokens_seen": 21834112,
      "step": 103455
    },
    {
      "epoch": 11.381738173817382,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.014001638385195702,
      "loss": 0.2303,
      "num_input_tokens_seen": 21835104,
      "step": 103460
    },
    {
      "epoch": 11.382288228822881,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.014000201542314848,
      "loss": 0.2298,
      "num_input_tokens_seen": 21836192,
      "step": 103465
    },
    {
      "epoch": 11.382838283828383,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.013998764708648663,
      "loss": 0.2324,
      "num_input_tokens_seen": 21837216,
      "step": 103470
    },
    {
      "epoch": 11.383388338833884,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.013997327884210375,
      "loss": 0.2309,
      "num_input_tokens_seen": 21838240,
      "step": 103475
    },
    {
      "epoch": 11.383938393839385,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.01399589106901324,
      "loss": 0.2319,
      "num_input_tokens_seen": 21839264,
      "step": 103480
    },
    {
      "epoch": 11.384488448844884,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.013994454263070498,
      "loss": 0.2288,
      "num_input_tokens_seen": 21840320,
      "step": 103485
    },
    {
      "epoch": 11.385038503850385,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013993017466395377,
      "loss": 0.2319,
      "num_input_tokens_seen": 21841344,
      "step": 103490
    },
    {
      "epoch": 11.385588558855886,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.013991580679001143,
      "loss": 0.2309,
      "num_input_tokens_seen": 21842464,
      "step": 103495
    },
    {
      "epoch": 11.386138613861386,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.013990143900901022,
      "loss": 0.233,
      "num_input_tokens_seen": 21843520,
      "step": 103500
    },
    {
      "epoch": 11.386688668866887,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.013988707132108256,
      "loss": 0.2304,
      "num_input_tokens_seen": 21844544,
      "step": 103505
    },
    {
      "epoch": 11.387238723872388,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.013987270372636097,
      "loss": 0.234,
      "num_input_tokens_seen": 21845568,
      "step": 103510
    },
    {
      "epoch": 11.387788778877887,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.013985833622497777,
      "loss": 0.2345,
      "num_input_tokens_seen": 21846624,
      "step": 103515
    },
    {
      "epoch": 11.388338833883388,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.013984396881706544,
      "loss": 0.233,
      "num_input_tokens_seen": 21847680,
      "step": 103520
    },
    {
      "epoch": 11.38888888888889,
      "grad_norm": 0.00064849853515625,
      "learning_rate": 0.013982960150275642,
      "loss": 0.2319,
      "num_input_tokens_seen": 21848672,
      "step": 103525
    },
    {
      "epoch": 11.389438943894389,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013981523428218299,
      "loss": 0.2298,
      "num_input_tokens_seen": 21849760,
      "step": 103530
    },
    {
      "epoch": 11.38998899889989,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.013980086715547771,
      "loss": 0.2335,
      "num_input_tokens_seen": 21850784,
      "step": 103535
    },
    {
      "epoch": 11.39053905390539,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.01397865001227729,
      "loss": 0.2314,
      "num_input_tokens_seen": 21851840,
      "step": 103540
    },
    {
      "epoch": 11.391089108910892,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013977213318420105,
      "loss": 0.2314,
      "num_input_tokens_seen": 21852864,
      "step": 103545
    },
    {
      "epoch": 11.391639163916391,
      "grad_norm": 0.000514984130859375,
      "learning_rate": 0.013975776633989459,
      "loss": 0.2308,
      "num_input_tokens_seen": 21853888,
      "step": 103550
    },
    {
      "epoch": 11.392189218921892,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.013974339958998579,
      "loss": 0.2308,
      "num_input_tokens_seen": 21854944,
      "step": 103555
    },
    {
      "epoch": 11.392739273927393,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.013972903293460721,
      "loss": 0.2319,
      "num_input_tokens_seen": 21856032,
      "step": 103560
    },
    {
      "epoch": 11.393289328932893,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.013971466637389117,
      "loss": 0.2319,
      "num_input_tokens_seen": 21857024,
      "step": 103565
    },
    {
      "epoch": 11.393839383938394,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.013970029990797017,
      "loss": 0.2329,
      "num_input_tokens_seen": 21858080,
      "step": 103570
    },
    {
      "epoch": 11.394389438943895,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013968593353697656,
      "loss": 0.2309,
      "num_input_tokens_seen": 21859136,
      "step": 103575
    },
    {
      "epoch": 11.394939493949394,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.013967156726104268,
      "loss": 0.2309,
      "num_input_tokens_seen": 21860192,
      "step": 103580
    },
    {
      "epoch": 11.395489548954895,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.013965720108030108,
      "loss": 0.2324,
      "num_input_tokens_seen": 21861312,
      "step": 103585
    },
    {
      "epoch": 11.396039603960396,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.013964283499488405,
      "loss": 0.2303,
      "num_input_tokens_seen": 21862336,
      "step": 103590
    },
    {
      "epoch": 11.396589658965897,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.013962846900492406,
      "loss": 0.2314,
      "num_input_tokens_seen": 21863328,
      "step": 103595
    },
    {
      "epoch": 11.397139713971397,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013961410311055352,
      "loss": 0.2283,
      "num_input_tokens_seen": 21864384,
      "step": 103600
    },
    {
      "epoch": 11.397689768976898,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013959973731190473,
      "loss": 0.2319,
      "num_input_tokens_seen": 21865440,
      "step": 103605
    },
    {
      "epoch": 11.398239823982399,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.013958537160911026,
      "loss": 0.2303,
      "num_input_tokens_seen": 21866496,
      "step": 103610
    },
    {
      "epoch": 11.398789878987898,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013957100600230235,
      "loss": 0.2309,
      "num_input_tokens_seen": 21867552,
      "step": 103615
    },
    {
      "epoch": 11.3993399339934,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013955664049161352,
      "loss": 0.2324,
      "num_input_tokens_seen": 21868544,
      "step": 103620
    },
    {
      "epoch": 11.3998899889989,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.013954227507717612,
      "loss": 0.2324,
      "num_input_tokens_seen": 21869568,
      "step": 103625
    },
    {
      "epoch": 11.4004400440044,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.013952790975912253,
      "loss": 0.2319,
      "num_input_tokens_seen": 21870656,
      "step": 103630
    },
    {
      "epoch": 11.400990099009901,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.01395135445375852,
      "loss": 0.2303,
      "num_input_tokens_seen": 21871680,
      "step": 103635
    },
    {
      "epoch": 11.401540154015402,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013949917941269645,
      "loss": 0.2324,
      "num_input_tokens_seen": 21872736,
      "step": 103640
    },
    {
      "epoch": 11.402090209020901,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.013948481438458877,
      "loss": 0.2293,
      "num_input_tokens_seen": 21873792,
      "step": 103645
    },
    {
      "epoch": 11.402640264026402,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.013947044945339451,
      "loss": 0.233,
      "num_input_tokens_seen": 21874880,
      "step": 103650
    },
    {
      "epoch": 11.403190319031903,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.013945608461924603,
      "loss": 0.2335,
      "num_input_tokens_seen": 21875936,
      "step": 103655
    },
    {
      "epoch": 11.403740374037405,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.013944171988227581,
      "loss": 0.2324,
      "num_input_tokens_seen": 21876960,
      "step": 103660
    },
    {
      "epoch": 11.404290429042904,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.01394273552426161,
      "loss": 0.2303,
      "num_input_tokens_seen": 21877952,
      "step": 103665
    },
    {
      "epoch": 11.404840484048405,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.013941299070039947,
      "loss": 0.2324,
      "num_input_tokens_seen": 21878976,
      "step": 103670
    },
    {
      "epoch": 11.405390539053906,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.013939862625575821,
      "loss": 0.2298,
      "num_input_tokens_seen": 21880064,
      "step": 103675
    },
    {
      "epoch": 11.405940594059405,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013938426190882468,
      "loss": 0.2303,
      "num_input_tokens_seen": 21881120,
      "step": 103680
    },
    {
      "epoch": 11.406490649064907,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01393698976597314,
      "loss": 0.2319,
      "num_input_tokens_seen": 21882176,
      "step": 103685
    },
    {
      "epoch": 11.407040704070408,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013935553350861062,
      "loss": 0.2303,
      "num_input_tokens_seen": 21883200,
      "step": 103690
    },
    {
      "epoch": 11.407590759075907,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.013934116945559474,
      "loss": 0.2303,
      "num_input_tokens_seen": 21884288,
      "step": 103695
    },
    {
      "epoch": 11.408140814081408,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.013932680550081627,
      "loss": 0.2319,
      "num_input_tokens_seen": 21885376,
      "step": 103700
    },
    {
      "epoch": 11.408690869086909,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.013931244164440745,
      "loss": 0.2314,
      "num_input_tokens_seen": 21886432,
      "step": 103705
    },
    {
      "epoch": 11.409240924092408,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.013929807788650075,
      "loss": 0.2324,
      "num_input_tokens_seen": 21887488,
      "step": 103710
    },
    {
      "epoch": 11.40979097909791,
      "grad_norm": 0.000560760498046875,
      "learning_rate": 0.013928371422722855,
      "loss": 0.2324,
      "num_input_tokens_seen": 21888512,
      "step": 103715
    },
    {
      "epoch": 11.41034103410341,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.013926935066672317,
      "loss": 0.2334,
      "num_input_tokens_seen": 21889600,
      "step": 103720
    },
    {
      "epoch": 11.410891089108912,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.013925498720511706,
      "loss": 0.2309,
      "num_input_tokens_seen": 21890720,
      "step": 103725
    },
    {
      "epoch": 11.411441144114411,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.013924062384254254,
      "loss": 0.2335,
      "num_input_tokens_seen": 21891744,
      "step": 103730
    },
    {
      "epoch": 11.411991199119912,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013922626057913206,
      "loss": 0.2319,
      "num_input_tokens_seen": 21892800,
      "step": 103735
    },
    {
      "epoch": 11.412541254125413,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.013921189741501801,
      "loss": 0.2329,
      "num_input_tokens_seen": 21893824,
      "step": 103740
    },
    {
      "epoch": 11.413091309130913,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.013919753435033265,
      "loss": 0.2324,
      "num_input_tokens_seen": 21894848,
      "step": 103745
    },
    {
      "epoch": 11.413641364136414,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.013918317138520847,
      "loss": 0.233,
      "num_input_tokens_seen": 21895904,
      "step": 103750
    },
    {
      "epoch": 11.414191419141915,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.013916880851977778,
      "loss": 0.2319,
      "num_input_tokens_seen": 21896960,
      "step": 103755
    },
    {
      "epoch": 11.414741474147414,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.013915444575417305,
      "loss": 0.2314,
      "num_input_tokens_seen": 21897984,
      "step": 103760
    },
    {
      "epoch": 11.415291529152915,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.013914008308852656,
      "loss": 0.2319,
      "num_input_tokens_seen": 21898976,
      "step": 103765
    },
    {
      "epoch": 11.415841584158416,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013912572052297064,
      "loss": 0.2298,
      "num_input_tokens_seen": 21900000,
      "step": 103770
    },
    {
      "epoch": 11.416391639163916,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.013911135805763784,
      "loss": 0.2293,
      "num_input_tokens_seen": 21901024,
      "step": 103775
    },
    {
      "epoch": 11.416941694169417,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.013909699569266036,
      "loss": 0.2314,
      "num_input_tokens_seen": 21902080,
      "step": 103780
    },
    {
      "epoch": 11.417491749174918,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013908263342817065,
      "loss": 0.2314,
      "num_input_tokens_seen": 21903104,
      "step": 103785
    },
    {
      "epoch": 11.418041804180419,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.013906827126430111,
      "loss": 0.2329,
      "num_input_tokens_seen": 21904128,
      "step": 103790
    },
    {
      "epoch": 11.418591859185918,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013905390920118402,
      "loss": 0.2335,
      "num_input_tokens_seen": 21905216,
      "step": 103795
    },
    {
      "epoch": 11.41914191419142,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013903954723895183,
      "loss": 0.2309,
      "num_input_tokens_seen": 21906304,
      "step": 103800
    },
    {
      "epoch": 11.41969196919692,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.013902518537773684,
      "loss": 0.2309,
      "num_input_tokens_seen": 21907328,
      "step": 103805
    },
    {
      "epoch": 11.42024202420242,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.013901082361767148,
      "loss": 0.2309,
      "num_input_tokens_seen": 21908384,
      "step": 103810
    },
    {
      "epoch": 11.42079207920792,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.01389964619588881,
      "loss": 0.2314,
      "num_input_tokens_seen": 21909376,
      "step": 103815
    },
    {
      "epoch": 11.421342134213422,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.0138982100401519,
      "loss": 0.2314,
      "num_input_tokens_seen": 21910400,
      "step": 103820
    },
    {
      "epoch": 11.421892189218921,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.013896773894569663,
      "loss": 0.2314,
      "num_input_tokens_seen": 21911584,
      "step": 103825
    },
    {
      "epoch": 11.422442244224422,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.013895337759155328,
      "loss": 0.2324,
      "num_input_tokens_seen": 21912640,
      "step": 103830
    },
    {
      "epoch": 11.422992299229923,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013893901633922142,
      "loss": 0.2324,
      "num_input_tokens_seen": 21913696,
      "step": 103835
    },
    {
      "epoch": 11.423542354235423,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.013892465518883328,
      "loss": 0.2329,
      "num_input_tokens_seen": 21914656,
      "step": 103840
    },
    {
      "epoch": 11.424092409240924,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013891029414052127,
      "loss": 0.2314,
      "num_input_tokens_seen": 21915744,
      "step": 103845
    },
    {
      "epoch": 11.424642464246425,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.013889593319441777,
      "loss": 0.2309,
      "num_input_tokens_seen": 21916832,
      "step": 103850
    },
    {
      "epoch": 11.425192519251926,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.013888157235065512,
      "loss": 0.2303,
      "num_input_tokens_seen": 21917888,
      "step": 103855
    },
    {
      "epoch": 11.425742574257425,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.013886721160936573,
      "loss": 0.2324,
      "num_input_tokens_seen": 21918912,
      "step": 103860
    },
    {
      "epoch": 11.426292629262926,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.01388528509706819,
      "loss": 0.2314,
      "num_input_tokens_seen": 21919936,
      "step": 103865
    },
    {
      "epoch": 11.426842684268427,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.013883849043473591,
      "loss": 0.2335,
      "num_input_tokens_seen": 21920960,
      "step": 103870
    },
    {
      "epoch": 11.427392739273927,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.01388241300016603,
      "loss": 0.233,
      "num_input_tokens_seen": 21922048,
      "step": 103875
    },
    {
      "epoch": 11.427942794279428,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.013880976967158728,
      "loss": 0.2309,
      "num_input_tokens_seen": 21923040,
      "step": 103880
    },
    {
      "epoch": 11.428492849284929,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.013879540944464922,
      "loss": 0.2314,
      "num_input_tokens_seen": 21924096,
      "step": 103885
    },
    {
      "epoch": 11.429042904290428,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013878104932097856,
      "loss": 0.2324,
      "num_input_tokens_seen": 21925088,
      "step": 103890
    },
    {
      "epoch": 11.42959295929593,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.01387666893007075,
      "loss": 0.2308,
      "num_input_tokens_seen": 21926208,
      "step": 103895
    },
    {
      "epoch": 11.43014301430143,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.013875232938396853,
      "loss": 0.2293,
      "num_input_tokens_seen": 21927232,
      "step": 103900
    },
    {
      "epoch": 11.430693069306932,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013873796957089396,
      "loss": 0.2324,
      "num_input_tokens_seen": 21928224,
      "step": 103905
    },
    {
      "epoch": 11.43124312431243,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.013872360986161604,
      "loss": 0.2293,
      "num_input_tokens_seen": 21929248,
      "step": 103910
    },
    {
      "epoch": 11.431793179317932,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.013870925025626727,
      "loss": 0.2314,
      "num_input_tokens_seen": 21930304,
      "step": 103915
    },
    {
      "epoch": 11.432343234323433,
      "grad_norm": 0.00057220458984375,
      "learning_rate": 0.013869489075497989,
      "loss": 0.2313,
      "num_input_tokens_seen": 21931360,
      "step": 103920
    },
    {
      "epoch": 11.432893289328932,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.013868053135788627,
      "loss": 0.2324,
      "num_input_tokens_seen": 21932384,
      "step": 103925
    },
    {
      "epoch": 11.433443344334433,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.013866617206511882,
      "loss": 0.2309,
      "num_input_tokens_seen": 21933376,
      "step": 103930
    },
    {
      "epoch": 11.433993399339935,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.013865181287680975,
      "loss": 0.2314,
      "num_input_tokens_seen": 21934400,
      "step": 103935
    },
    {
      "epoch": 11.434543454345434,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013863745379309152,
      "loss": 0.2319,
      "num_input_tokens_seen": 21935424,
      "step": 103940
    },
    {
      "epoch": 11.435093509350935,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013862309481409638,
      "loss": 0.2308,
      "num_input_tokens_seen": 21936480,
      "step": 103945
    },
    {
      "epoch": 11.435643564356436,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.013860873593995678,
      "loss": 0.233,
      "num_input_tokens_seen": 21937600,
      "step": 103950
    },
    {
      "epoch": 11.436193619361935,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.013859437717080496,
      "loss": 0.2309,
      "num_input_tokens_seen": 21938656,
      "step": 103955
    },
    {
      "epoch": 11.436743674367436,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.013858001850677327,
      "loss": 0.2309,
      "num_input_tokens_seen": 21939680,
      "step": 103960
    },
    {
      "epoch": 11.437293729372938,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.013856565994799414,
      "loss": 0.2314,
      "num_input_tokens_seen": 21940768,
      "step": 103965
    },
    {
      "epoch": 11.437843784378439,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.013855130149459972,
      "loss": 0.2309,
      "num_input_tokens_seen": 21941824,
      "step": 103970
    },
    {
      "epoch": 11.438393839383938,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013853694314672257,
      "loss": 0.2314,
      "num_input_tokens_seen": 21942880,
      "step": 103975
    },
    {
      "epoch": 11.438943894389439,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013852258490449488,
      "loss": 0.2309,
      "num_input_tokens_seen": 21943904,
      "step": 103980
    },
    {
      "epoch": 11.43949394939494,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.013850822676804899,
      "loss": 0.2309,
      "num_input_tokens_seen": 21944992,
      "step": 103985
    },
    {
      "epoch": 11.44004400440044,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01384938687375173,
      "loss": 0.2309,
      "num_input_tokens_seen": 21945984,
      "step": 103990
    },
    {
      "epoch": 11.44059405940594,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.013847951081303206,
      "loss": 0.2329,
      "num_input_tokens_seen": 21947104,
      "step": 103995
    },
    {
      "epoch": 11.441144114411442,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.013846515299472567,
      "loss": 0.2303,
      "num_input_tokens_seen": 21948192,
      "step": 104000
    },
    {
      "epoch": 11.441694169416941,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013845079528273047,
      "loss": 0.2319,
      "num_input_tokens_seen": 21949248,
      "step": 104005
    },
    {
      "epoch": 11.442244224422442,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013843643767717868,
      "loss": 0.2314,
      "num_input_tokens_seen": 21950304,
      "step": 104010
    },
    {
      "epoch": 11.442794279427943,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.013842208017820271,
      "loss": 0.2314,
      "num_input_tokens_seen": 21951296,
      "step": 104015
    },
    {
      "epoch": 11.443344334433444,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.013840772278593486,
      "loss": 0.2303,
      "num_input_tokens_seen": 21952288,
      "step": 104020
    },
    {
      "epoch": 11.443894389438944,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.01383933655005075,
      "loss": 0.2309,
      "num_input_tokens_seen": 21953312,
      "step": 104025
    },
    {
      "epoch": 11.444444444444445,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.013837900832205291,
      "loss": 0.2309,
      "num_input_tokens_seen": 21954336,
      "step": 104030
    },
    {
      "epoch": 11.444994499449946,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01383646512507034,
      "loss": 0.2314,
      "num_input_tokens_seen": 21955360,
      "step": 104035
    },
    {
      "epoch": 11.445544554455445,
      "grad_norm": 0.00081634521484375,
      "learning_rate": 0.013835029428659135,
      "loss": 0.2319,
      "num_input_tokens_seen": 21956448,
      "step": 104040
    },
    {
      "epoch": 11.446094609460946,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.0138335937429849,
      "loss": 0.233,
      "num_input_tokens_seen": 21957504,
      "step": 104045
    },
    {
      "epoch": 11.446644664466447,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.013832158068060879,
      "loss": 0.233,
      "num_input_tokens_seen": 21958624,
      "step": 104050
    },
    {
      "epoch": 11.447194719471947,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013830722403900294,
      "loss": 0.2319,
      "num_input_tokens_seen": 21959712,
      "step": 104055
    },
    {
      "epoch": 11.447744774477448,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.013829286750516374,
      "loss": 0.2303,
      "num_input_tokens_seen": 21960768,
      "step": 104060
    },
    {
      "epoch": 11.448294829482949,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013827851107922364,
      "loss": 0.2314,
      "num_input_tokens_seen": 21961824,
      "step": 104065
    },
    {
      "epoch": 11.448844884488448,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013826415476131485,
      "loss": 0.2298,
      "num_input_tokens_seen": 21962880,
      "step": 104070
    },
    {
      "epoch": 11.44939493949395,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013824979855156969,
      "loss": 0.2309,
      "num_input_tokens_seen": 21963936,
      "step": 104075
    },
    {
      "epoch": 11.44994499449945,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.013823544245012054,
      "loss": 0.2303,
      "num_input_tokens_seen": 21965024,
      "step": 104080
    },
    {
      "epoch": 11.450495049504951,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.013822108645709959,
      "loss": 0.2298,
      "num_input_tokens_seen": 21966080,
      "step": 104085
    },
    {
      "epoch": 11.45104510451045,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.013820673057263934,
      "loss": 0.2293,
      "num_input_tokens_seen": 21967136,
      "step": 104090
    },
    {
      "epoch": 11.451595159515952,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.013819237479687194,
      "loss": 0.2309,
      "num_input_tokens_seen": 21968224,
      "step": 104095
    },
    {
      "epoch": 11.452145214521453,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.013817801912992974,
      "loss": 0.2314,
      "num_input_tokens_seen": 21969312,
      "step": 104100
    },
    {
      "epoch": 11.452695269526952,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013816366357194514,
      "loss": 0.2303,
      "num_input_tokens_seen": 21970432,
      "step": 104105
    },
    {
      "epoch": 11.453245324532453,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.013814930812305028,
      "loss": 0.2314,
      "num_input_tokens_seen": 21971520,
      "step": 104110
    },
    {
      "epoch": 11.453795379537954,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.013813495278337761,
      "loss": 0.233,
      "num_input_tokens_seen": 21972608,
      "step": 104115
    },
    {
      "epoch": 11.454345434543454,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013812059755305943,
      "loss": 0.2314,
      "num_input_tokens_seen": 21973600,
      "step": 104120
    },
    {
      "epoch": 11.454895489548955,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013810624243222791,
      "loss": 0.2308,
      "num_input_tokens_seen": 21974720,
      "step": 104125
    },
    {
      "epoch": 11.455445544554456,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.013809188742101549,
      "loss": 0.2324,
      "num_input_tokens_seen": 21975744,
      "step": 104130
    },
    {
      "epoch": 11.455995599559955,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.013807753251955442,
      "loss": 0.2319,
      "num_input_tokens_seen": 21976800,
      "step": 104135
    },
    {
      "epoch": 11.456545654565456,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.013806317772797703,
      "loss": 0.2329,
      "num_input_tokens_seen": 21977824,
      "step": 104140
    },
    {
      "epoch": 11.457095709570957,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.013804882304641561,
      "loss": 0.2303,
      "num_input_tokens_seen": 21978880,
      "step": 104145
    },
    {
      "epoch": 11.457645764576458,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01380344684750024,
      "loss": 0.2319,
      "num_input_tokens_seen": 21979904,
      "step": 104150
    },
    {
      "epoch": 11.458195819581958,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.013802011401386979,
      "loss": 0.2283,
      "num_input_tokens_seen": 21980992,
      "step": 104155
    },
    {
      "epoch": 11.458745874587459,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.013800575966315003,
      "loss": 0.2303,
      "num_input_tokens_seen": 21982016,
      "step": 104160
    },
    {
      "epoch": 11.45929592959296,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.013799140542297547,
      "loss": 0.2298,
      "num_input_tokens_seen": 21983104,
      "step": 104165
    },
    {
      "epoch": 11.45984598459846,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.013797705129347833,
      "loss": 0.2298,
      "num_input_tokens_seen": 21984192,
      "step": 104170
    },
    {
      "epoch": 11.46039603960396,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.01379626972747909,
      "loss": 0.2324,
      "num_input_tokens_seen": 21985248,
      "step": 104175
    },
    {
      "epoch": 11.460946094609461,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013794834336704559,
      "loss": 0.2319,
      "num_input_tokens_seen": 21986272,
      "step": 104180
    },
    {
      "epoch": 11.46149614961496,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013793398957037456,
      "loss": 0.2319,
      "num_input_tokens_seen": 21987328,
      "step": 104185
    },
    {
      "epoch": 11.462046204620462,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.013791963588491019,
      "loss": 0.2319,
      "num_input_tokens_seen": 21988448,
      "step": 104190
    },
    {
      "epoch": 11.462596259625963,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.013790528231078474,
      "loss": 0.2314,
      "num_input_tokens_seen": 21989504,
      "step": 104195
    },
    {
      "epoch": 11.463146314631462,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.013789092884813047,
      "loss": 0.2324,
      "num_input_tokens_seen": 21990560,
      "step": 104200
    },
    {
      "epoch": 11.463696369636963,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.013787657549707972,
      "loss": 0.2324,
      "num_input_tokens_seen": 21991648,
      "step": 104205
    },
    {
      "epoch": 11.464246424642464,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013786222225776473,
      "loss": 0.2324,
      "num_input_tokens_seen": 21992704,
      "step": 104210
    },
    {
      "epoch": 11.464796479647966,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.013784786913031782,
      "loss": 0.2303,
      "num_input_tokens_seen": 21993696,
      "step": 104215
    },
    {
      "epoch": 11.465346534653465,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.013783351611487133,
      "loss": 0.2324,
      "num_input_tokens_seen": 21994752,
      "step": 104220
    },
    {
      "epoch": 11.465896589658966,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01378191632115574,
      "loss": 0.2308,
      "num_input_tokens_seen": 21995744,
      "step": 104225
    },
    {
      "epoch": 11.466446644664467,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013780481042050843,
      "loss": 0.2298,
      "num_input_tokens_seen": 21996704,
      "step": 104230
    },
    {
      "epoch": 11.466996699669966,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.013779045774185665,
      "loss": 0.2293,
      "num_input_tokens_seen": 21997824,
      "step": 104235
    },
    {
      "epoch": 11.467546754675467,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.013777610517573443,
      "loss": 0.2308,
      "num_input_tokens_seen": 21998848,
      "step": 104240
    },
    {
      "epoch": 11.468096809680969,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.013776175272227395,
      "loss": 0.2308,
      "num_input_tokens_seen": 21999936,
      "step": 104245
    },
    {
      "epoch": 11.468646864686468,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.013774740038160748,
      "loss": 0.2314,
      "num_input_tokens_seen": 22000992,
      "step": 104250
    },
    {
      "epoch": 11.469196919691969,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013773304815386741,
      "loss": 0.2329,
      "num_input_tokens_seen": 22002080,
      "step": 104255
    },
    {
      "epoch": 11.46974697469747,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013771869603918593,
      "loss": 0.2324,
      "num_input_tokens_seen": 22003040,
      "step": 104260
    },
    {
      "epoch": 11.47029702970297,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.013770434403769528,
      "loss": 0.2314,
      "num_input_tokens_seen": 22004160,
      "step": 104265
    },
    {
      "epoch": 11.47084708470847,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013768999214952787,
      "loss": 0.2324,
      "num_input_tokens_seen": 22005280,
      "step": 104270
    },
    {
      "epoch": 11.471397139713972,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.01376756403748158,
      "loss": 0.2303,
      "num_input_tokens_seen": 22006336,
      "step": 104275
    },
    {
      "epoch": 11.471947194719473,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013766128871369151,
      "loss": 0.2319,
      "num_input_tokens_seen": 22007392,
      "step": 104280
    },
    {
      "epoch": 11.472497249724972,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01376469371662872,
      "loss": 0.2335,
      "num_input_tokens_seen": 22008448,
      "step": 104285
    },
    {
      "epoch": 11.473047304730473,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01376325857327351,
      "loss": 0.2329,
      "num_input_tokens_seen": 22009472,
      "step": 104290
    },
    {
      "epoch": 11.473597359735974,
      "grad_norm": 0.000659942626953125,
      "learning_rate": 0.013761823441316759,
      "loss": 0.2319,
      "num_input_tokens_seen": 22010592,
      "step": 104295
    },
    {
      "epoch": 11.474147414741473,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.01376038832077168,
      "loss": 0.2319,
      "num_input_tokens_seen": 22011648,
      "step": 104300
    },
    {
      "epoch": 11.474697469746975,
      "grad_norm": 0.000690460205078125,
      "learning_rate": 0.013758953211651511,
      "loss": 0.2324,
      "num_input_tokens_seen": 22012672,
      "step": 104305
    },
    {
      "epoch": 11.475247524752476,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.013757518113969478,
      "loss": 0.2298,
      "num_input_tokens_seen": 22013792,
      "step": 104310
    },
    {
      "epoch": 11.475797579757975,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013756083027738798,
      "loss": 0.2308,
      "num_input_tokens_seen": 22014816,
      "step": 104315
    },
    {
      "epoch": 11.476347634763476,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.013754647952972704,
      "loss": 0.2329,
      "num_input_tokens_seen": 22015872,
      "step": 104320
    },
    {
      "epoch": 11.476897689768977,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.013753212889684422,
      "loss": 0.2314,
      "num_input_tokens_seen": 22017024,
      "step": 104325
    },
    {
      "epoch": 11.477447744774478,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.013751777837887183,
      "loss": 0.2324,
      "num_input_tokens_seen": 22018048,
      "step": 104330
    },
    {
      "epoch": 11.477997799779978,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.01375034279759421,
      "loss": 0.2293,
      "num_input_tokens_seen": 22019104,
      "step": 104335
    },
    {
      "epoch": 11.478547854785479,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.013748907768818722,
      "loss": 0.2298,
      "num_input_tokens_seen": 22020096,
      "step": 104340
    },
    {
      "epoch": 11.47909790979098,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.013747472751573954,
      "loss": 0.2324,
      "num_input_tokens_seen": 22021184,
      "step": 104345
    },
    {
      "epoch": 11.479647964796479,
      "grad_norm": 0.000614166259765625,
      "learning_rate": 0.013746037745873125,
      "loss": 0.2319,
      "num_input_tokens_seen": 22022208,
      "step": 104350
    },
    {
      "epoch": 11.48019801980198,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.01374460275172947,
      "loss": 0.2298,
      "num_input_tokens_seen": 22023328,
      "step": 104355
    },
    {
      "epoch": 11.480748074807481,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.013743167769156206,
      "loss": 0.2304,
      "num_input_tokens_seen": 22024352,
      "step": 104360
    },
    {
      "epoch": 11.48129812981298,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.01374173279816656,
      "loss": 0.2324,
      "num_input_tokens_seen": 22025472,
      "step": 104365
    },
    {
      "epoch": 11.481848184818482,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.013740297838773764,
      "loss": 0.2319,
      "num_input_tokens_seen": 22026496,
      "step": 104370
    },
    {
      "epoch": 11.482398239823983,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.013738862890991034,
      "loss": 0.2319,
      "num_input_tokens_seen": 22027552,
      "step": 104375
    },
    {
      "epoch": 11.482948294829482,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.013737427954831601,
      "loss": 0.2298,
      "num_input_tokens_seen": 22028640,
      "step": 104380
    },
    {
      "epoch": 11.483498349834983,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.013735993030308691,
      "loss": 0.2298,
      "num_input_tokens_seen": 22029696,
      "step": 104385
    },
    {
      "epoch": 11.484048404840484,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013734558117435519,
      "loss": 0.2324,
      "num_input_tokens_seen": 22030720,
      "step": 104390
    },
    {
      "epoch": 11.484598459845985,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.013733123216225327,
      "loss": 0.2304,
      "num_input_tokens_seen": 22031840,
      "step": 104395
    },
    {
      "epoch": 11.485148514851485,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013731688326691322,
      "loss": 0.2314,
      "num_input_tokens_seen": 22032896,
      "step": 104400
    },
    {
      "epoch": 11.485698569856986,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013730253448846742,
      "loss": 0.2303,
      "num_input_tokens_seen": 22033888,
      "step": 104405
    },
    {
      "epoch": 11.486248624862487,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01372881858270481,
      "loss": 0.2314,
      "num_input_tokens_seen": 22034944,
      "step": 104410
    },
    {
      "epoch": 11.486798679867986,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01372738372827874,
      "loss": 0.2309,
      "num_input_tokens_seen": 22035968,
      "step": 104415
    },
    {
      "epoch": 11.487348734873487,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.01372594888558177,
      "loss": 0.2309,
      "num_input_tokens_seen": 22037024,
      "step": 104420
    },
    {
      "epoch": 11.487898789878988,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.013724514054627109,
      "loss": 0.2319,
      "num_input_tokens_seen": 22038112,
      "step": 104425
    },
    {
      "epoch": 11.488448844884488,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.013723079235428,
      "loss": 0.2298,
      "num_input_tokens_seen": 22039168,
      "step": 104430
    },
    {
      "epoch": 11.488998899889989,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013721644427997651,
      "loss": 0.2324,
      "num_input_tokens_seen": 22040256,
      "step": 104435
    },
    {
      "epoch": 11.48954895489549,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013720209632349292,
      "loss": 0.2314,
      "num_input_tokens_seen": 22041344,
      "step": 104440
    },
    {
      "epoch": 11.490099009900991,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013718774848496151,
      "loss": 0.2314,
      "num_input_tokens_seen": 22042464,
      "step": 104445
    },
    {
      "epoch": 11.49064906490649,
      "grad_norm": 0.00482177734375,
      "learning_rate": 0.013717340076451446,
      "loss": 0.2293,
      "num_input_tokens_seen": 22043616,
      "step": 104450
    },
    {
      "epoch": 11.491199119911991,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.013715905316228396,
      "loss": 0.2298,
      "num_input_tokens_seen": 22044704,
      "step": 104455
    },
    {
      "epoch": 11.491749174917492,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.013714470567840236,
      "loss": 0.2303,
      "num_input_tokens_seen": 22045728,
      "step": 104460
    },
    {
      "epoch": 11.492299229922992,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.01371303583130018,
      "loss": 0.2324,
      "num_input_tokens_seen": 22046816,
      "step": 104465
    },
    {
      "epoch": 11.492849284928493,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013711601106621462,
      "loss": 0.2319,
      "num_input_tokens_seen": 22047840,
      "step": 104470
    },
    {
      "epoch": 11.493399339933994,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.013710166393817294,
      "loss": 0.2314,
      "num_input_tokens_seen": 22048864,
      "step": 104475
    },
    {
      "epoch": 11.493949394939493,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013708731692900902,
      "loss": 0.2319,
      "num_input_tokens_seen": 22050016,
      "step": 104480
    },
    {
      "epoch": 11.494499449944994,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.013707297003885515,
      "loss": 0.2314,
      "num_input_tokens_seen": 22051040,
      "step": 104485
    },
    {
      "epoch": 11.495049504950495,
      "grad_norm": 0.000797271728515625,
      "learning_rate": 0.013705862326784345,
      "loss": 0.2319,
      "num_input_tokens_seen": 22052096,
      "step": 104490
    },
    {
      "epoch": 11.495599559955995,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.013704427661610627,
      "loss": 0.2293,
      "num_input_tokens_seen": 22053184,
      "step": 104495
    },
    {
      "epoch": 11.496149614961496,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.013702993008377578,
      "loss": 0.2309,
      "num_input_tokens_seen": 22054208,
      "step": 104500
    },
    {
      "epoch": 11.496699669966997,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013701558367098413,
      "loss": 0.2319,
      "num_input_tokens_seen": 22055264,
      "step": 104505
    },
    {
      "epoch": 11.497249724972498,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.013700123737786367,
      "loss": 0.233,
      "num_input_tokens_seen": 22056352,
      "step": 104510
    },
    {
      "epoch": 11.497799779977997,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.013698689120454653,
      "loss": 0.2319,
      "num_input_tokens_seen": 22057376,
      "step": 104515
    },
    {
      "epoch": 11.498349834983498,
      "grad_norm": 0.0007171630859375,
      "learning_rate": 0.013697254515116499,
      "loss": 0.2324,
      "num_input_tokens_seen": 22058432,
      "step": 104520
    },
    {
      "epoch": 11.498899889989,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.01369581992178513,
      "loss": 0.2319,
      "num_input_tokens_seen": 22059488,
      "step": 104525
    },
    {
      "epoch": 11.499449944994499,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.013694385340473755,
      "loss": 0.2309,
      "num_input_tokens_seen": 22060608,
      "step": 104530
    },
    {
      "epoch": 11.5,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.013692950771195608,
      "loss": 0.2293,
      "num_input_tokens_seen": 22061664,
      "step": 104535
    },
    {
      "epoch": 11.500550055005501,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.013691516213963901,
      "loss": 0.2293,
      "num_input_tokens_seen": 22062784,
      "step": 104540
    },
    {
      "epoch": 11.501100110011,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.01369008166879187,
      "loss": 0.2303,
      "num_input_tokens_seen": 22063840,
      "step": 104545
    },
    {
      "epoch": 11.501650165016502,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.013688647135692724,
      "loss": 0.2335,
      "num_input_tokens_seen": 22064896,
      "step": 104550
    },
    {
      "epoch": 11.502200220022003,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.013687212614679685,
      "loss": 0.233,
      "num_input_tokens_seen": 22066016,
      "step": 104555
    },
    {
      "epoch": 11.502750275027502,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.013685778105765984,
      "loss": 0.2329,
      "num_input_tokens_seen": 22067136,
      "step": 104560
    },
    {
      "epoch": 11.503300330033003,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01368434360896483,
      "loss": 0.2329,
      "num_input_tokens_seen": 22068192,
      "step": 104565
    },
    {
      "epoch": 11.503850385038504,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01368290912428945,
      "loss": 0.2298,
      "num_input_tokens_seen": 22069216,
      "step": 104570
    },
    {
      "epoch": 11.504400440044005,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.013681474651753067,
      "loss": 0.2309,
      "num_input_tokens_seen": 22070272,
      "step": 104575
    },
    {
      "epoch": 11.504950495049505,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.013680040191368894,
      "loss": 0.2324,
      "num_input_tokens_seen": 22071392,
      "step": 104580
    },
    {
      "epoch": 11.505500550055006,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.013678605743150166,
      "loss": 0.2309,
      "num_input_tokens_seen": 22072384,
      "step": 104585
    },
    {
      "epoch": 11.506050605060507,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.013677171307110087,
      "loss": 0.2314,
      "num_input_tokens_seen": 22073472,
      "step": 104590
    },
    {
      "epoch": 11.506600660066006,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.01367573688326189,
      "loss": 0.2308,
      "num_input_tokens_seen": 22074560,
      "step": 104595
    },
    {
      "epoch": 11.507150715071507,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.013674302471618792,
      "loss": 0.2303,
      "num_input_tokens_seen": 22075616,
      "step": 104600
    },
    {
      "epoch": 11.507700770077008,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.013672868072194005,
      "loss": 0.2308,
      "num_input_tokens_seen": 22076768,
      "step": 104605
    },
    {
      "epoch": 11.508250825082508,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.013671433685000763,
      "loss": 0.2309,
      "num_input_tokens_seen": 22077856,
      "step": 104610
    },
    {
      "epoch": 11.508800880088009,
      "grad_norm": 0.00069427490234375,
      "learning_rate": 0.013669999310052274,
      "loss": 0.2308,
      "num_input_tokens_seen": 22078944,
      "step": 104615
    },
    {
      "epoch": 11.50935093509351,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.013668564947361768,
      "loss": 0.2304,
      "num_input_tokens_seen": 22080000,
      "step": 104620
    },
    {
      "epoch": 11.509900990099009,
      "grad_norm": 0.000484466552734375,
      "learning_rate": 0.013667130596942459,
      "loss": 0.2308,
      "num_input_tokens_seen": 22081088,
      "step": 104625
    },
    {
      "epoch": 11.51045104510451,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.013665696258807564,
      "loss": 0.233,
      "num_input_tokens_seen": 22082176,
      "step": 104630
    },
    {
      "epoch": 11.511001100110011,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013664261932970309,
      "loss": 0.2314,
      "num_input_tokens_seen": 22083200,
      "step": 104635
    },
    {
      "epoch": 11.511551155115512,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.013662827619443908,
      "loss": 0.2309,
      "num_input_tokens_seen": 22084256,
      "step": 104640
    },
    {
      "epoch": 11.512101210121012,
      "grad_norm": 0.0006561279296875,
      "learning_rate": 0.01366139331824159,
      "loss": 0.2298,
      "num_input_tokens_seen": 22085248,
      "step": 104645
    },
    {
      "epoch": 11.512651265126513,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.013659959029376563,
      "loss": 0.2324,
      "num_input_tokens_seen": 22086336,
      "step": 104650
    },
    {
      "epoch": 11.513201320132014,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.013658524752862047,
      "loss": 0.2335,
      "num_input_tokens_seen": 22087392,
      "step": 104655
    },
    {
      "epoch": 11.513751375137513,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.013657090488711273,
      "loss": 0.2309,
      "num_input_tokens_seen": 22088416,
      "step": 104660
    },
    {
      "epoch": 11.514301430143014,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.01365565623693745,
      "loss": 0.2314,
      "num_input_tokens_seen": 22089440,
      "step": 104665
    },
    {
      "epoch": 11.514851485148515,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01365422199755379,
      "loss": 0.2324,
      "num_input_tokens_seen": 22090464,
      "step": 104670
    },
    {
      "epoch": 11.515401540154015,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.01365278777057353,
      "loss": 0.2324,
      "num_input_tokens_seen": 22091488,
      "step": 104675
    },
    {
      "epoch": 11.515951595159516,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01365135355600987,
      "loss": 0.2324,
      "num_input_tokens_seen": 22092576,
      "step": 104680
    },
    {
      "epoch": 11.516501650165017,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.013649919353876043,
      "loss": 0.2314,
      "num_input_tokens_seen": 22093600,
      "step": 104685
    },
    {
      "epoch": 11.517051705170516,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.013648485164185263,
      "loss": 0.234,
      "num_input_tokens_seen": 22094688,
      "step": 104690
    },
    {
      "epoch": 11.517601760176017,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.013647050986950739,
      "loss": 0.2319,
      "num_input_tokens_seen": 22095712,
      "step": 104695
    },
    {
      "epoch": 11.518151815181518,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013645616822185704,
      "loss": 0.2309,
      "num_input_tokens_seen": 22096736,
      "step": 104700
    },
    {
      "epoch": 11.51870187018702,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.013644182669903363,
      "loss": 0.2308,
      "num_input_tokens_seen": 22097696,
      "step": 104705
    },
    {
      "epoch": 11.519251925192519,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013642748530116943,
      "loss": 0.2324,
      "num_input_tokens_seen": 22098720,
      "step": 104710
    },
    {
      "epoch": 11.51980198019802,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.013641314402839664,
      "loss": 0.2319,
      "num_input_tokens_seen": 22099872,
      "step": 104715
    },
    {
      "epoch": 11.520352035203521,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.013639880288084728,
      "loss": 0.2314,
      "num_input_tokens_seen": 22100864,
      "step": 104720
    },
    {
      "epoch": 11.52090209020902,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.01363844618586537,
      "loss": 0.2324,
      "num_input_tokens_seen": 22101920,
      "step": 104725
    },
    {
      "epoch": 11.521452145214521,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.013637012096194794,
      "loss": 0.2319,
      "num_input_tokens_seen": 22103008,
      "step": 104730
    },
    {
      "epoch": 11.522002200220022,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.013635578019086231,
      "loss": 0.2319,
      "num_input_tokens_seen": 22104032,
      "step": 104735
    },
    {
      "epoch": 11.522552255225522,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.013634143954552886,
      "loss": 0.2308,
      "num_input_tokens_seen": 22105024,
      "step": 104740
    },
    {
      "epoch": 11.523102310231023,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.01363270990260798,
      "loss": 0.2303,
      "num_input_tokens_seen": 22106112,
      "step": 104745
    },
    {
      "epoch": 11.523652365236524,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013631275863264736,
      "loss": 0.2329,
      "num_input_tokens_seen": 22107168,
      "step": 104750
    },
    {
      "epoch": 11.524202420242025,
      "grad_norm": 0.0004749298095703125,
      "learning_rate": 0.013629841836536354,
      "loss": 0.2303,
      "num_input_tokens_seen": 22108192,
      "step": 104755
    },
    {
      "epoch": 11.524752475247524,
      "grad_norm": 0.000797271728515625,
      "learning_rate": 0.013628407822436076,
      "loss": 0.2303,
      "num_input_tokens_seen": 22109312,
      "step": 104760
    },
    {
      "epoch": 11.525302530253025,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.013626973820977098,
      "loss": 0.2314,
      "num_input_tokens_seen": 22110368,
      "step": 104765
    },
    {
      "epoch": 11.525852585258527,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013625539832172642,
      "loss": 0.2308,
      "num_input_tokens_seen": 22111328,
      "step": 104770
    },
    {
      "epoch": 11.526402640264026,
      "grad_norm": 0.00077056884765625,
      "learning_rate": 0.013624105856035932,
      "loss": 0.2314,
      "num_input_tokens_seen": 22112384,
      "step": 104775
    },
    {
      "epoch": 11.526952695269527,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013622671892580172,
      "loss": 0.2303,
      "num_input_tokens_seen": 22113344,
      "step": 104780
    },
    {
      "epoch": 11.527502750275028,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013621237941818588,
      "loss": 0.2303,
      "num_input_tokens_seen": 22114368,
      "step": 104785
    },
    {
      "epoch": 11.528052805280527,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013619804003764396,
      "loss": 0.2345,
      "num_input_tokens_seen": 22115520,
      "step": 104790
    },
    {
      "epoch": 11.528602860286028,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.0136183700784308,
      "loss": 0.2319,
      "num_input_tokens_seen": 22116576,
      "step": 104795
    },
    {
      "epoch": 11.52915291529153,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013616936165831026,
      "loss": 0.2308,
      "num_input_tokens_seen": 22117632,
      "step": 104800
    },
    {
      "epoch": 11.52970297029703,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.013615502265978288,
      "loss": 0.2308,
      "num_input_tokens_seen": 22118688,
      "step": 104805
    },
    {
      "epoch": 11.53025302530253,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013614068378885806,
      "loss": 0.2319,
      "num_input_tokens_seen": 22119744,
      "step": 104810
    },
    {
      "epoch": 11.530803080308031,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.013612634504566786,
      "loss": 0.2298,
      "num_input_tokens_seen": 22120768,
      "step": 104815
    },
    {
      "epoch": 11.531353135313532,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013611200643034446,
      "loss": 0.2298,
      "num_input_tokens_seen": 22121824,
      "step": 104820
    },
    {
      "epoch": 11.531903190319031,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013609766794302007,
      "loss": 0.2319,
      "num_input_tokens_seen": 22122944,
      "step": 104825
    },
    {
      "epoch": 11.532453245324533,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.013608332958382678,
      "loss": 0.2309,
      "num_input_tokens_seen": 22123968,
      "step": 104830
    },
    {
      "epoch": 11.533003300330034,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.013606899135289681,
      "loss": 0.2304,
      "num_input_tokens_seen": 22125088,
      "step": 104835
    },
    {
      "epoch": 11.533553355335533,
      "grad_norm": 0.0004119873046875,
      "learning_rate": 0.013605465325036224,
      "loss": 0.2293,
      "num_input_tokens_seen": 22126176,
      "step": 104840
    },
    {
      "epoch": 11.534103410341034,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.013604031527635521,
      "loss": 0.2319,
      "num_input_tokens_seen": 22127200,
      "step": 104845
    },
    {
      "epoch": 11.534653465346535,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.013602597743100795,
      "loss": 0.234,
      "num_input_tokens_seen": 22128256,
      "step": 104850
    },
    {
      "epoch": 11.535203520352034,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.013601163971445252,
      "loss": 0.2314,
      "num_input_tokens_seen": 22129344,
      "step": 104855
    },
    {
      "epoch": 11.535753575357536,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.013599730212682107,
      "loss": 0.2298,
      "num_input_tokens_seen": 22130336,
      "step": 104860
    },
    {
      "epoch": 11.536303630363037,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013598296466824582,
      "loss": 0.2319,
      "num_input_tokens_seen": 22131360,
      "step": 104865
    },
    {
      "epoch": 11.536853685368538,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01359686273388588,
      "loss": 0.2319,
      "num_input_tokens_seen": 22132384,
      "step": 104870
    },
    {
      "epoch": 11.537403740374037,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013595429013879225,
      "loss": 0.2319,
      "num_input_tokens_seen": 22133504,
      "step": 104875
    },
    {
      "epoch": 11.537953795379538,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.013593995306817826,
      "loss": 0.2313,
      "num_input_tokens_seen": 22134496,
      "step": 104880
    },
    {
      "epoch": 11.53850385038504,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013592561612714892,
      "loss": 0.2324,
      "num_input_tokens_seen": 22135552,
      "step": 104885
    },
    {
      "epoch": 11.539053905390539,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013591127931583654,
      "loss": 0.2319,
      "num_input_tokens_seen": 22136544,
      "step": 104890
    },
    {
      "epoch": 11.53960396039604,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.013589694263437303,
      "loss": 0.2319,
      "num_input_tokens_seen": 22137568,
      "step": 104895
    },
    {
      "epoch": 11.54015401540154,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.013588260608289068,
      "loss": 0.2313,
      "num_input_tokens_seen": 22138624,
      "step": 104900
    },
    {
      "epoch": 11.54070407040704,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01358682696615216,
      "loss": 0.2298,
      "num_input_tokens_seen": 22139680,
      "step": 104905
    },
    {
      "epoch": 11.541254125412541,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013585393337039784,
      "loss": 0.2309,
      "num_input_tokens_seen": 22140736,
      "step": 104910
    },
    {
      "epoch": 11.541804180418042,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013583959720965164,
      "loss": 0.2308,
      "num_input_tokens_seen": 22141792,
      "step": 104915
    },
    {
      "epoch": 11.542354235423542,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.0135825261179415,
      "loss": 0.2324,
      "num_input_tokens_seen": 22142848,
      "step": 104920
    },
    {
      "epoch": 11.542904290429043,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013581092527982022,
      "loss": 0.2319,
      "num_input_tokens_seen": 22143872,
      "step": 104925
    },
    {
      "epoch": 11.543454345434544,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01357965895109993,
      "loss": 0.2314,
      "num_input_tokens_seen": 22144864,
      "step": 104930
    },
    {
      "epoch": 11.544004400440045,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013578225387308436,
      "loss": 0.2303,
      "num_input_tokens_seen": 22145920,
      "step": 104935
    },
    {
      "epoch": 11.544554455445544,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.01357679183662076,
      "loss": 0.2309,
      "num_input_tokens_seen": 22146976,
      "step": 104940
    },
    {
      "epoch": 11.545104510451045,
      "grad_norm": 0.000682830810546875,
      "learning_rate": 0.013575358299050109,
      "loss": 0.2324,
      "num_input_tokens_seen": 22147968,
      "step": 104945
    },
    {
      "epoch": 11.545654565456546,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.0135739247746097,
      "loss": 0.2324,
      "num_input_tokens_seen": 22149088,
      "step": 104950
    },
    {
      "epoch": 11.546204620462046,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.013572491263312742,
      "loss": 0.2335,
      "num_input_tokens_seen": 22150048,
      "step": 104955
    },
    {
      "epoch": 11.546754675467547,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013571057765172442,
      "loss": 0.2319,
      "num_input_tokens_seen": 22151072,
      "step": 104960
    },
    {
      "epoch": 11.547304730473048,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013569624280202024,
      "loss": 0.2329,
      "num_input_tokens_seen": 22152096,
      "step": 104965
    },
    {
      "epoch": 11.547854785478547,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.013568190808414685,
      "loss": 0.2319,
      "num_input_tokens_seen": 22153152,
      "step": 104970
    },
    {
      "epoch": 11.548404840484048,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01356675734982365,
      "loss": 0.2309,
      "num_input_tokens_seen": 22154208,
      "step": 104975
    },
    {
      "epoch": 11.54895489548955,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.013565323904442126,
      "loss": 0.2329,
      "num_input_tokens_seen": 22155232,
      "step": 104980
    },
    {
      "epoch": 11.549504950495049,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.013563890472283317,
      "loss": 0.2308,
      "num_input_tokens_seen": 22156352,
      "step": 104985
    },
    {
      "epoch": 11.55005500550055,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013562457053360443,
      "loss": 0.2308,
      "num_input_tokens_seen": 22157408,
      "step": 104990
    },
    {
      "epoch": 11.55060506050605,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01356102364768671,
      "loss": 0.2319,
      "num_input_tokens_seen": 22158432,
      "step": 104995
    },
    {
      "epoch": 11.551155115511552,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.013559590255275335,
      "loss": 0.2319,
      "num_input_tokens_seen": 22159552,
      "step": 105000
    },
    {
      "epoch": 11.551705170517051,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.013558156876139529,
      "loss": 0.2314,
      "num_input_tokens_seen": 22160640,
      "step": 105005
    },
    {
      "epoch": 11.552255225522552,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.013556723510292492,
      "loss": 0.2308,
      "num_input_tokens_seen": 22161728,
      "step": 105010
    },
    {
      "epoch": 11.552805280528053,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013555290157747446,
      "loss": 0.2308,
      "num_input_tokens_seen": 22162784,
      "step": 105015
    },
    {
      "epoch": 11.553355335533553,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013553856818517594,
      "loss": 0.2298,
      "num_input_tokens_seen": 22163872,
      "step": 105020
    },
    {
      "epoch": 11.553905390539054,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.013552423492616155,
      "loss": 0.2308,
      "num_input_tokens_seen": 22164864,
      "step": 105025
    },
    {
      "epoch": 11.554455445544555,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.013550990180056333,
      "loss": 0.2324,
      "num_input_tokens_seen": 22165888,
      "step": 105030
    },
    {
      "epoch": 11.555005500550054,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.013549556880851334,
      "loss": 0.2303,
      "num_input_tokens_seen": 22166944,
      "step": 105035
    },
    {
      "epoch": 11.555555555555555,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01354812359501438,
      "loss": 0.2314,
      "num_input_tokens_seen": 22168064,
      "step": 105040
    },
    {
      "epoch": 11.556105610561056,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.01354669032255867,
      "loss": 0.2314,
      "num_input_tokens_seen": 22169120,
      "step": 105045
    },
    {
      "epoch": 11.556655665566556,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.013545257063497417,
      "loss": 0.2324,
      "num_input_tokens_seen": 22170208,
      "step": 105050
    },
    {
      "epoch": 11.557205720572057,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013543823817843838,
      "loss": 0.2324,
      "num_input_tokens_seen": 22171168,
      "step": 105055
    },
    {
      "epoch": 11.557755775577558,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013542390585611125,
      "loss": 0.2319,
      "num_input_tokens_seen": 22172224,
      "step": 105060
    },
    {
      "epoch": 11.558305830583059,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.013540957366812508,
      "loss": 0.2303,
      "num_input_tokens_seen": 22173280,
      "step": 105065
    },
    {
      "epoch": 11.558855885588558,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013539524161461184,
      "loss": 0.2288,
      "num_input_tokens_seen": 22174336,
      "step": 105070
    },
    {
      "epoch": 11.55940594059406,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.013538090969570362,
      "loss": 0.2303,
      "num_input_tokens_seen": 22175456,
      "step": 105075
    },
    {
      "epoch": 11.55995599559956,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.013536657791153262,
      "loss": 0.2319,
      "num_input_tokens_seen": 22176512,
      "step": 105080
    },
    {
      "epoch": 11.56050605060506,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013535224626223077,
      "loss": 0.2293,
      "num_input_tokens_seen": 22177536,
      "step": 105085
    },
    {
      "epoch": 11.561056105610561,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.013533791474793026,
      "loss": 0.2309,
      "num_input_tokens_seen": 22178528,
      "step": 105090
    },
    {
      "epoch": 11.561606160616062,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.01353235833687632,
      "loss": 0.2298,
      "num_input_tokens_seen": 22179584,
      "step": 105095
    },
    {
      "epoch": 11.562156215621561,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013530925212486156,
      "loss": 0.2303,
      "num_input_tokens_seen": 22180608,
      "step": 105100
    },
    {
      "epoch": 11.562706270627062,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013529492101635752,
      "loss": 0.2303,
      "num_input_tokens_seen": 22181632,
      "step": 105105
    },
    {
      "epoch": 11.563256325632564,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.013528059004338311,
      "loss": 0.2319,
      "num_input_tokens_seen": 22182688,
      "step": 105110
    },
    {
      "epoch": 11.563806380638063,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013526625920607048,
      "loss": 0.2319,
      "num_input_tokens_seen": 22183744,
      "step": 105115
    },
    {
      "epoch": 11.564356435643564,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.01352519285045517,
      "loss": 0.2303,
      "num_input_tokens_seen": 22184768,
      "step": 105120
    },
    {
      "epoch": 11.564906490649065,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.013523759793895874,
      "loss": 0.2324,
      "num_input_tokens_seen": 22185888,
      "step": 105125
    },
    {
      "epoch": 11.565456545654566,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01352232675094238,
      "loss": 0.2319,
      "num_input_tokens_seen": 22186944,
      "step": 105130
    },
    {
      "epoch": 11.566006600660065,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013520893721607887,
      "loss": 0.2314,
      "num_input_tokens_seen": 22188000,
      "step": 105135
    },
    {
      "epoch": 11.566556655665567,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.013519460705905614,
      "loss": 0.2319,
      "num_input_tokens_seen": 22189120,
      "step": 105140
    },
    {
      "epoch": 11.567106710671068,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.013518027703848759,
      "loss": 0.2303,
      "num_input_tokens_seen": 22190176,
      "step": 105145
    },
    {
      "epoch": 11.567656765676567,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.013516594715450527,
      "loss": 0.2308,
      "num_input_tokens_seen": 22191200,
      "step": 105150
    },
    {
      "epoch": 11.568206820682068,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.013515161740724138,
      "loss": 0.2309,
      "num_input_tokens_seen": 22192256,
      "step": 105155
    },
    {
      "epoch": 11.56875687568757,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013513728779682781,
      "loss": 0.2314,
      "num_input_tokens_seen": 22193248,
      "step": 105160
    },
    {
      "epoch": 11.569306930693068,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013512295832339679,
      "loss": 0.2319,
      "num_input_tokens_seen": 22194272,
      "step": 105165
    },
    {
      "epoch": 11.56985698569857,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.013510862898708034,
      "loss": 0.2314,
      "num_input_tokens_seen": 22195328,
      "step": 105170
    },
    {
      "epoch": 11.57040704070407,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.013509429978801042,
      "loss": 0.2303,
      "num_input_tokens_seen": 22196352,
      "step": 105175
    },
    {
      "epoch": 11.570957095709572,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.013507997072631927,
      "loss": 0.2293,
      "num_input_tokens_seen": 22197408,
      "step": 105180
    },
    {
      "epoch": 11.571507150715071,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01350656418021388,
      "loss": 0.2303,
      "num_input_tokens_seen": 22198464,
      "step": 105185
    },
    {
      "epoch": 11.572057205720572,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01350513130156012,
      "loss": 0.2324,
      "num_input_tokens_seen": 22199520,
      "step": 105190
    },
    {
      "epoch": 11.572607260726073,
      "grad_norm": 0.000782012939453125,
      "learning_rate": 0.01350369843668385,
      "loss": 0.2319,
      "num_input_tokens_seen": 22200608,
      "step": 105195
    },
    {
      "epoch": 11.573157315731573,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013502265585598269,
      "loss": 0.2314,
      "num_input_tokens_seen": 22201632,
      "step": 105200
    },
    {
      "epoch": 11.573707370737074,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013500832748316587,
      "loss": 0.2329,
      "num_input_tokens_seen": 22202656,
      "step": 105205
    },
    {
      "epoch": 11.574257425742575,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.01349939992485201,
      "loss": 0.2303,
      "num_input_tokens_seen": 22203744,
      "step": 105210
    },
    {
      "epoch": 11.574807480748074,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.013497967115217748,
      "loss": 0.2319,
      "num_input_tokens_seen": 22204864,
      "step": 105215
    },
    {
      "epoch": 11.575357535753575,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.013496534319427,
      "loss": 0.2335,
      "num_input_tokens_seen": 22205888,
      "step": 105220
    },
    {
      "epoch": 11.575907590759076,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013495101537492971,
      "loss": 0.2303,
      "num_input_tokens_seen": 22207008,
      "step": 105225
    },
    {
      "epoch": 11.576457645764577,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.013493668769428875,
      "loss": 0.2314,
      "num_input_tokens_seen": 22208032,
      "step": 105230
    },
    {
      "epoch": 11.577007700770077,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013492236015247908,
      "loss": 0.2319,
      "num_input_tokens_seen": 22209056,
      "step": 105235
    },
    {
      "epoch": 11.577557755775578,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013490803274963274,
      "loss": 0.2304,
      "num_input_tokens_seen": 22210144,
      "step": 105240
    },
    {
      "epoch": 11.578107810781079,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013489370548588186,
      "loss": 0.2303,
      "num_input_tokens_seen": 22211232,
      "step": 105245
    },
    {
      "epoch": 11.578657865786578,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.013487937836135841,
      "loss": 0.2314,
      "num_input_tokens_seen": 22212224,
      "step": 105250
    },
    {
      "epoch": 11.57920792079208,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.013486505137619455,
      "loss": 0.233,
      "num_input_tokens_seen": 22213216,
      "step": 105255
    },
    {
      "epoch": 11.57975797579758,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.013485072453052221,
      "loss": 0.234,
      "num_input_tokens_seen": 22214304,
      "step": 105260
    },
    {
      "epoch": 11.58030803080308,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.013483639782447343,
      "loss": 0.2314,
      "num_input_tokens_seen": 22215360,
      "step": 105265
    },
    {
      "epoch": 11.58085808580858,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01348220712581804,
      "loss": 0.2319,
      "num_input_tokens_seen": 22216512,
      "step": 105270
    },
    {
      "epoch": 11.581408140814082,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.013480774483177495,
      "loss": 0.2314,
      "num_input_tokens_seen": 22217536,
      "step": 105275
    },
    {
      "epoch": 11.581958195819581,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.013479341854538926,
      "loss": 0.2314,
      "num_input_tokens_seen": 22218560,
      "step": 105280
    },
    {
      "epoch": 11.582508250825082,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.013477909239915539,
      "loss": 0.2304,
      "num_input_tokens_seen": 22219648,
      "step": 105285
    },
    {
      "epoch": 11.583058305830583,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.013476476639320525,
      "loss": 0.2335,
      "num_input_tokens_seen": 22220672,
      "step": 105290
    },
    {
      "epoch": 11.583608360836084,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013475044052767097,
      "loss": 0.2308,
      "num_input_tokens_seen": 22221856,
      "step": 105295
    },
    {
      "epoch": 11.584158415841584,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.013473611480268455,
      "loss": 0.2324,
      "num_input_tokens_seen": 22222880,
      "step": 105300
    },
    {
      "epoch": 11.584708470847085,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013472178921837804,
      "loss": 0.2298,
      "num_input_tokens_seen": 22224000,
      "step": 105305
    },
    {
      "epoch": 11.585258525852586,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01347074637748835,
      "loss": 0.2309,
      "num_input_tokens_seen": 22225088,
      "step": 105310
    },
    {
      "epoch": 11.585808580858085,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013469313847233289,
      "loss": 0.2303,
      "num_input_tokens_seen": 22226112,
      "step": 105315
    },
    {
      "epoch": 11.586358635863586,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.01346788133108583,
      "loss": 0.2314,
      "num_input_tokens_seen": 22227168,
      "step": 105320
    },
    {
      "epoch": 11.586908690869087,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.01346644882905917,
      "loss": 0.2345,
      "num_input_tokens_seen": 22228256,
      "step": 105325
    },
    {
      "epoch": 11.587458745874587,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.01346501634116652,
      "loss": 0.2319,
      "num_input_tokens_seen": 22229344,
      "step": 105330
    },
    {
      "epoch": 11.588008800880088,
      "grad_norm": 0.000659942626953125,
      "learning_rate": 0.013463583867421077,
      "loss": 0.2319,
      "num_input_tokens_seen": 22230304,
      "step": 105335
    },
    {
      "epoch": 11.588558855885589,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.013462151407836041,
      "loss": 0.2309,
      "num_input_tokens_seen": 22231360,
      "step": 105340
    },
    {
      "epoch": 11.589108910891088,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.013460718962424623,
      "loss": 0.2298,
      "num_input_tokens_seen": 22232416,
      "step": 105345
    },
    {
      "epoch": 11.58965896589659,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.013459286531200012,
      "loss": 0.2293,
      "num_input_tokens_seen": 22233504,
      "step": 105350
    },
    {
      "epoch": 11.59020902090209,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.013457854114175424,
      "loss": 0.2319,
      "num_input_tokens_seen": 22234592,
      "step": 105355
    },
    {
      "epoch": 11.590759075907592,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.013456421711364054,
      "loss": 0.2314,
      "num_input_tokens_seen": 22235648,
      "step": 105360
    },
    {
      "epoch": 11.591309130913091,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013454989322779099,
      "loss": 0.2308,
      "num_input_tokens_seen": 22236736,
      "step": 105365
    },
    {
      "epoch": 11.591859185918592,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.013453556948433776,
      "loss": 0.2314,
      "num_input_tokens_seen": 22237856,
      "step": 105370
    },
    {
      "epoch": 11.592409240924093,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.013452124588341267,
      "loss": 0.2293,
      "num_input_tokens_seen": 22238976,
      "step": 105375
    },
    {
      "epoch": 11.592959295929592,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.013450692242514788,
      "loss": 0.2324,
      "num_input_tokens_seen": 22240000,
      "step": 105380
    },
    {
      "epoch": 11.593509350935093,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.013449259910967536,
      "loss": 0.2314,
      "num_input_tokens_seen": 22241024,
      "step": 105385
    },
    {
      "epoch": 11.594059405940595,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.013447827593712708,
      "loss": 0.2324,
      "num_input_tokens_seen": 22242112,
      "step": 105390
    },
    {
      "epoch": 11.594609460946094,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013446395290763509,
      "loss": 0.2303,
      "num_input_tokens_seen": 22243136,
      "step": 105395
    },
    {
      "epoch": 11.595159515951595,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013444963002133136,
      "loss": 0.2309,
      "num_input_tokens_seen": 22244192,
      "step": 105400
    },
    {
      "epoch": 11.595709570957096,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013443530727834802,
      "loss": 0.2319,
      "num_input_tokens_seen": 22245248,
      "step": 105405
    },
    {
      "epoch": 11.596259625962595,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.013442098467881691,
      "loss": 0.2319,
      "num_input_tokens_seen": 22246336,
      "step": 105410
    },
    {
      "epoch": 11.596809680968097,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.013440666222287009,
      "loss": 0.2319,
      "num_input_tokens_seen": 22247328,
      "step": 105415
    },
    {
      "epoch": 11.597359735973598,
      "grad_norm": 0.000751495361328125,
      "learning_rate": 0.013439233991063964,
      "loss": 0.2314,
      "num_input_tokens_seen": 22248416,
      "step": 105420
    },
    {
      "epoch": 11.597909790979099,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.013437801774225752,
      "loss": 0.2303,
      "num_input_tokens_seen": 22249600,
      "step": 105425
    },
    {
      "epoch": 11.598459845984598,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.013436369571785565,
      "loss": 0.2298,
      "num_input_tokens_seen": 22250624,
      "step": 105430
    },
    {
      "epoch": 11.599009900990099,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.013434937383756613,
      "loss": 0.2309,
      "num_input_tokens_seen": 22251648,
      "step": 105435
    },
    {
      "epoch": 11.5995599559956,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.013433505210152087,
      "loss": 0.2304,
      "num_input_tokens_seen": 22252800,
      "step": 105440
    },
    {
      "epoch": 11.6001100110011,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0134320730509852,
      "loss": 0.2314,
      "num_input_tokens_seen": 22253856,
      "step": 105445
    },
    {
      "epoch": 11.6006600660066,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013430640906269138,
      "loss": 0.2303,
      "num_input_tokens_seen": 22254944,
      "step": 105450
    },
    {
      "epoch": 11.601210121012102,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013429208776017105,
      "loss": 0.2309,
      "num_input_tokens_seen": 22256032,
      "step": 105455
    },
    {
      "epoch": 11.601760176017601,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.013427776660242306,
      "loss": 0.2319,
      "num_input_tokens_seen": 22257024,
      "step": 105460
    },
    {
      "epoch": 11.602310231023102,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.01342634455895793,
      "loss": 0.2319,
      "num_input_tokens_seen": 22258080,
      "step": 105465
    },
    {
      "epoch": 11.602860286028603,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.013424912472177183,
      "loss": 0.2303,
      "num_input_tokens_seen": 22259136,
      "step": 105470
    },
    {
      "epoch": 11.603410341034103,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.013423480399913265,
      "loss": 0.2324,
      "num_input_tokens_seen": 22260224,
      "step": 105475
    },
    {
      "epoch": 11.603960396039604,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.013422048342179365,
      "loss": 0.2319,
      "num_input_tokens_seen": 22261248,
      "step": 105480
    },
    {
      "epoch": 11.604510451045105,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013420616298988695,
      "loss": 0.2314,
      "num_input_tokens_seen": 22262208,
      "step": 105485
    },
    {
      "epoch": 11.605060506050606,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013419184270354441,
      "loss": 0.2324,
      "num_input_tokens_seen": 22263296,
      "step": 105490
    },
    {
      "epoch": 11.605610561056105,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01341775225628981,
      "loss": 0.2324,
      "num_input_tokens_seen": 22264320,
      "step": 105495
    },
    {
      "epoch": 11.606160616061606,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013416320256808001,
      "loss": 0.2324,
      "num_input_tokens_seen": 22265344,
      "step": 105500
    },
    {
      "epoch": 11.606710671067107,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.013414888271922202,
      "loss": 0.2314,
      "num_input_tokens_seen": 22266400,
      "step": 105505
    },
    {
      "epoch": 11.607260726072607,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.013413456301645619,
      "loss": 0.2324,
      "num_input_tokens_seen": 22267488,
      "step": 105510
    },
    {
      "epoch": 11.607810781078108,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.013412024345991445,
      "loss": 0.2324,
      "num_input_tokens_seen": 22268544,
      "step": 105515
    },
    {
      "epoch": 11.608360836083609,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013410592404972888,
      "loss": 0.2309,
      "num_input_tokens_seen": 22269632,
      "step": 105520
    },
    {
      "epoch": 11.608910891089108,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.013409160478603134,
      "loss": 0.233,
      "num_input_tokens_seen": 22270688,
      "step": 105525
    },
    {
      "epoch": 11.60946094609461,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.013407728566895382,
      "loss": 0.233,
      "num_input_tokens_seen": 22271744,
      "step": 105530
    },
    {
      "epoch": 11.61001100110011,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013406296669862839,
      "loss": 0.2319,
      "num_input_tokens_seen": 22272800,
      "step": 105535
    },
    {
      "epoch": 11.61056105610561,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013404864787518685,
      "loss": 0.2319,
      "num_input_tokens_seen": 22273856,
      "step": 105540
    },
    {
      "epoch": 11.61111111111111,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.013403432919876135,
      "loss": 0.2314,
      "num_input_tokens_seen": 22274944,
      "step": 105545
    },
    {
      "epoch": 11.611661166116612,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.013402001066948376,
      "loss": 0.2303,
      "num_input_tokens_seen": 22276000,
      "step": 105550
    },
    {
      "epoch": 11.612211221122113,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013400569228748601,
      "loss": 0.2293,
      "num_input_tokens_seen": 22277088,
      "step": 105555
    },
    {
      "epoch": 11.612761276127612,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01339913740529002,
      "loss": 0.2329,
      "num_input_tokens_seen": 22278144,
      "step": 105560
    },
    {
      "epoch": 11.613311331133113,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.013397705596585813,
      "loss": 0.2309,
      "num_input_tokens_seen": 22279264,
      "step": 105565
    },
    {
      "epoch": 11.613861386138614,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.013396273802649192,
      "loss": 0.2324,
      "num_input_tokens_seen": 22280384,
      "step": 105570
    },
    {
      "epoch": 11.614411441144114,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013394842023493345,
      "loss": 0.2314,
      "num_input_tokens_seen": 22281472,
      "step": 105575
    },
    {
      "epoch": 11.614961496149615,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.013393410259131464,
      "loss": 0.2319,
      "num_input_tokens_seen": 22282528,
      "step": 105580
    },
    {
      "epoch": 11.615511551155116,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.013391978509576753,
      "loss": 0.2293,
      "num_input_tokens_seen": 22283584,
      "step": 105585
    },
    {
      "epoch": 11.616061606160617,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013390546774842402,
      "loss": 0.2308,
      "num_input_tokens_seen": 22284640,
      "step": 105590
    },
    {
      "epoch": 11.616611661166116,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013389115054941614,
      "loss": 0.2314,
      "num_input_tokens_seen": 22285632,
      "step": 105595
    },
    {
      "epoch": 11.617161716171617,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.013387683349887576,
      "loss": 0.2288,
      "num_input_tokens_seen": 22286720,
      "step": 105600
    },
    {
      "epoch": 11.617711771177119,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013386251659693486,
      "loss": 0.2293,
      "num_input_tokens_seen": 22287808,
      "step": 105605
    },
    {
      "epoch": 11.618261826182618,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013384819984372541,
      "loss": 0.2319,
      "num_input_tokens_seen": 22288832,
      "step": 105610
    },
    {
      "epoch": 11.618811881188119,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.01338338832393794,
      "loss": 0.2324,
      "num_input_tokens_seen": 22289984,
      "step": 105615
    },
    {
      "epoch": 11.61936193619362,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.013381956678402867,
      "loss": 0.2293,
      "num_input_tokens_seen": 22291072,
      "step": 105620
    },
    {
      "epoch": 11.61991199119912,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013380525047780525,
      "loss": 0.2319,
      "num_input_tokens_seen": 22292064,
      "step": 105625
    },
    {
      "epoch": 11.62046204620462,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013379093432084103,
      "loss": 0.2309,
      "num_input_tokens_seen": 22293088,
      "step": 105630
    },
    {
      "epoch": 11.621012101210122,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013377661831326807,
      "loss": 0.2298,
      "num_input_tokens_seen": 22294176,
      "step": 105635
    },
    {
      "epoch": 11.62156215621562,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01337623024552182,
      "loss": 0.234,
      "num_input_tokens_seen": 22295264,
      "step": 105640
    },
    {
      "epoch": 11.622112211221122,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.013374798674682333,
      "loss": 0.2335,
      "num_input_tokens_seen": 22296320,
      "step": 105645
    },
    {
      "epoch": 11.622662266226623,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.013373367118821556,
      "loss": 0.2298,
      "num_input_tokens_seen": 22297376,
      "step": 105650
    },
    {
      "epoch": 11.623212321232124,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013371935577952668,
      "loss": 0.2309,
      "num_input_tokens_seen": 22298432,
      "step": 105655
    },
    {
      "epoch": 11.623762376237623,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.01337050405208887,
      "loss": 0.2309,
      "num_input_tokens_seen": 22299552,
      "step": 105660
    },
    {
      "epoch": 11.624312431243125,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.013369072541243355,
      "loss": 0.2298,
      "num_input_tokens_seen": 22300672,
      "step": 105665
    },
    {
      "epoch": 11.624862486248626,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.013367641045429314,
      "loss": 0.2324,
      "num_input_tokens_seen": 22301792,
      "step": 105670
    },
    {
      "epoch": 11.625412541254125,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.013366209564659946,
      "loss": 0.2304,
      "num_input_tokens_seen": 22302784,
      "step": 105675
    },
    {
      "epoch": 11.625962596259626,
      "grad_norm": 0.001953125,
      "learning_rate": 0.013364778098948435,
      "loss": 0.2283,
      "num_input_tokens_seen": 22303840,
      "step": 105680
    },
    {
      "epoch": 11.626512651265127,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.013363346648307985,
      "loss": 0.2345,
      "num_input_tokens_seen": 22304896,
      "step": 105685
    },
    {
      "epoch": 11.627062706270626,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013361915212751785,
      "loss": 0.2303,
      "num_input_tokens_seen": 22305920,
      "step": 105690
    },
    {
      "epoch": 11.627612761276128,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013360483792293019,
      "loss": 0.2309,
      "num_input_tokens_seen": 22307040,
      "step": 105695
    },
    {
      "epoch": 11.628162816281629,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.013359052386944893,
      "loss": 0.2324,
      "num_input_tokens_seen": 22308096,
      "step": 105700
    },
    {
      "epoch": 11.628712871287128,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.013357620996720589,
      "loss": 0.2324,
      "num_input_tokens_seen": 22309184,
      "step": 105705
    },
    {
      "epoch": 11.629262926292629,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01335618962163331,
      "loss": 0.2314,
      "num_input_tokens_seen": 22310176,
      "step": 105710
    },
    {
      "epoch": 11.62981298129813,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.013354758261696239,
      "loss": 0.2304,
      "num_input_tokens_seen": 22311200,
      "step": 105715
    },
    {
      "epoch": 11.630363036303631,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.01335332691692257,
      "loss": 0.2309,
      "num_input_tokens_seen": 22312256,
      "step": 105720
    },
    {
      "epoch": 11.63091309130913,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013351895587325498,
      "loss": 0.2335,
      "num_input_tokens_seen": 22313312,
      "step": 105725
    },
    {
      "epoch": 11.631463146314632,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.013350464272918212,
      "loss": 0.2319,
      "num_input_tokens_seen": 22314368,
      "step": 105730
    },
    {
      "epoch": 11.632013201320133,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01334903297371391,
      "loss": 0.2319,
      "num_input_tokens_seen": 22315392,
      "step": 105735
    },
    {
      "epoch": 11.632563256325632,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.013347601689725779,
      "loss": 0.2304,
      "num_input_tokens_seen": 22316480,
      "step": 105740
    },
    {
      "epoch": 11.633113311331133,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.013346170420967004,
      "loss": 0.2298,
      "num_input_tokens_seen": 22317536,
      "step": 105745
    },
    {
      "epoch": 11.633663366336634,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013344739167450789,
      "loss": 0.2309,
      "num_input_tokens_seen": 22318624,
      "step": 105750
    },
    {
      "epoch": 11.634213421342134,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013343307929190315,
      "loss": 0.2303,
      "num_input_tokens_seen": 22319744,
      "step": 105755
    },
    {
      "epoch": 11.634763476347635,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013341876706198777,
      "loss": 0.2309,
      "num_input_tokens_seen": 22320800,
      "step": 105760
    },
    {
      "epoch": 11.635313531353136,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.01334044549848937,
      "loss": 0.2309,
      "num_input_tokens_seen": 22321920,
      "step": 105765
    },
    {
      "epoch": 11.635863586358635,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013339014306075276,
      "loss": 0.2319,
      "num_input_tokens_seen": 22323040,
      "step": 105770
    },
    {
      "epoch": 11.636413641364136,
      "grad_norm": 0.000579833984375,
      "learning_rate": 0.013337583128969693,
      "loss": 0.2309,
      "num_input_tokens_seen": 22324096,
      "step": 105775
    },
    {
      "epoch": 11.636963696369637,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.013336151967185803,
      "loss": 0.2304,
      "num_input_tokens_seen": 22325120,
      "step": 105780
    },
    {
      "epoch": 11.637513751375138,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013334720820736806,
      "loss": 0.2324,
      "num_input_tokens_seen": 22326144,
      "step": 105785
    },
    {
      "epoch": 11.638063806380638,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013333289689635893,
      "loss": 0.2314,
      "num_input_tokens_seen": 22327264,
      "step": 105790
    },
    {
      "epoch": 11.638613861386139,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.01333185857389624,
      "loss": 0.2335,
      "num_input_tokens_seen": 22328384,
      "step": 105795
    },
    {
      "epoch": 11.63916391639164,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.013330427473531052,
      "loss": 0.233,
      "num_input_tokens_seen": 22329408,
      "step": 105800
    },
    {
      "epoch": 11.63971397139714,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013328996388553513,
      "loss": 0.2309,
      "num_input_tokens_seen": 22330496,
      "step": 105805
    },
    {
      "epoch": 11.64026402640264,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.013327565318976808,
      "loss": 0.2319,
      "num_input_tokens_seen": 22331520,
      "step": 105810
    },
    {
      "epoch": 11.640814081408141,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.013326134264814134,
      "loss": 0.2298,
      "num_input_tokens_seen": 22332576,
      "step": 105815
    },
    {
      "epoch": 11.64136413641364,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013324703226078675,
      "loss": 0.2309,
      "num_input_tokens_seen": 22333600,
      "step": 105820
    },
    {
      "epoch": 11.641914191419142,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.013323272202783627,
      "loss": 0.2314,
      "num_input_tokens_seen": 22334624,
      "step": 105825
    },
    {
      "epoch": 11.642464246424643,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.013321841194942174,
      "loss": 0.2319,
      "num_input_tokens_seen": 22335648,
      "step": 105830
    },
    {
      "epoch": 11.643014301430142,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0133204102025675,
      "loss": 0.2298,
      "num_input_tokens_seen": 22336704,
      "step": 105835
    },
    {
      "epoch": 11.643564356435643,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.013318979225672807,
      "loss": 0.2298,
      "num_input_tokens_seen": 22337760,
      "step": 105840
    },
    {
      "epoch": 11.644114411441144,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.013317548264271265,
      "loss": 0.2319,
      "num_input_tokens_seen": 22338848,
      "step": 105845
    },
    {
      "epoch": 11.644664466446645,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013316117318376083,
      "loss": 0.2314,
      "num_input_tokens_seen": 22339872,
      "step": 105850
    },
    {
      "epoch": 11.645214521452145,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013314686388000437,
      "loss": 0.2314,
      "num_input_tokens_seen": 22341056,
      "step": 105855
    },
    {
      "epoch": 11.645764576457646,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.013313255473157514,
      "loss": 0.2314,
      "num_input_tokens_seen": 22342080,
      "step": 105860
    },
    {
      "epoch": 11.646314631463147,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013311824573860513,
      "loss": 0.233,
      "num_input_tokens_seen": 22343104,
      "step": 105865
    },
    {
      "epoch": 11.646864686468646,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.013310393690122609,
      "loss": 0.2303,
      "num_input_tokens_seen": 22344192,
      "step": 105870
    },
    {
      "epoch": 11.647414741474147,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013308962821956996,
      "loss": 0.233,
      "num_input_tokens_seen": 22345248,
      "step": 105875
    },
    {
      "epoch": 11.647964796479648,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013307531969376866,
      "loss": 0.2298,
      "num_input_tokens_seen": 22346336,
      "step": 105880
    },
    {
      "epoch": 11.648514851485148,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.013306101132395397,
      "loss": 0.2324,
      "num_input_tokens_seen": 22347360,
      "step": 105885
    },
    {
      "epoch": 11.649064906490649,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013304670311025783,
      "loss": 0.2309,
      "num_input_tokens_seen": 22348480,
      "step": 105890
    },
    {
      "epoch": 11.64961496149615,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013303239505281205,
      "loss": 0.2308,
      "num_input_tokens_seen": 22349472,
      "step": 105895
    },
    {
      "epoch": 11.65016501650165,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.013301808715174862,
      "loss": 0.2319,
      "num_input_tokens_seen": 22350560,
      "step": 105900
    },
    {
      "epoch": 11.65071507150715,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.01330037794071993,
      "loss": 0.2298,
      "num_input_tokens_seen": 22351648,
      "step": 105905
    },
    {
      "epoch": 11.651265126512651,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013298947181929597,
      "loss": 0.2324,
      "num_input_tokens_seen": 22352704,
      "step": 105910
    },
    {
      "epoch": 11.651815181518153,
      "grad_norm": 0.001953125,
      "learning_rate": 0.013297516438817051,
      "loss": 0.2314,
      "num_input_tokens_seen": 22353792,
      "step": 105915
    },
    {
      "epoch": 11.652365236523652,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.013296085711395478,
      "loss": 0.2325,
      "num_input_tokens_seen": 22354880,
      "step": 105920
    },
    {
      "epoch": 11.652915291529153,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.013294654999678074,
      "loss": 0.2314,
      "num_input_tokens_seen": 22356000,
      "step": 105925
    },
    {
      "epoch": 11.653465346534654,
      "grad_norm": 0.000698089599609375,
      "learning_rate": 0.013293224303678011,
      "loss": 0.2324,
      "num_input_tokens_seen": 22356992,
      "step": 105930
    },
    {
      "epoch": 11.654015401540153,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013291793623408477,
      "loss": 0.2303,
      "num_input_tokens_seen": 22358048,
      "step": 105935
    },
    {
      "epoch": 11.654565456545654,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.01329036295888267,
      "loss": 0.2299,
      "num_input_tokens_seen": 22359072,
      "step": 105940
    },
    {
      "epoch": 11.655115511551156,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.01328893231011376,
      "loss": 0.2314,
      "num_input_tokens_seen": 22360128,
      "step": 105945
    },
    {
      "epoch": 11.655665566556655,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.01328750167711494,
      "loss": 0.233,
      "num_input_tokens_seen": 22361120,
      "step": 105950
    },
    {
      "epoch": 11.656215621562156,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.013286071059899401,
      "loss": 0.2309,
      "num_input_tokens_seen": 22362144,
      "step": 105955
    },
    {
      "epoch": 11.656765676567657,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013284640458480313,
      "loss": 0.2309,
      "num_input_tokens_seen": 22363168,
      "step": 105960
    },
    {
      "epoch": 11.657315731573158,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.01328320987287088,
      "loss": 0.233,
      "num_input_tokens_seen": 22364192,
      "step": 105965
    },
    {
      "epoch": 11.657865786578657,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.013281779303084273,
      "loss": 0.2309,
      "num_input_tokens_seen": 22365248,
      "step": 105970
    },
    {
      "epoch": 11.658415841584159,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.013280348749133685,
      "loss": 0.2319,
      "num_input_tokens_seen": 22366272,
      "step": 105975
    },
    {
      "epoch": 11.65896589658966,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013278918211032298,
      "loss": 0.2319,
      "num_input_tokens_seen": 22367296,
      "step": 105980
    },
    {
      "epoch": 11.659515951595159,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.01327748768879329,
      "loss": 0.2319,
      "num_input_tokens_seen": 22368352,
      "step": 105985
    },
    {
      "epoch": 11.66006600660066,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013276057182429855,
      "loss": 0.2324,
      "num_input_tokens_seen": 22369440,
      "step": 105990
    },
    {
      "epoch": 11.660616061606161,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.013274626691955178,
      "loss": 0.2314,
      "num_input_tokens_seen": 22370560,
      "step": 105995
    },
    {
      "epoch": 11.66116611661166,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013273196217382432,
      "loss": 0.2308,
      "num_input_tokens_seen": 22371552,
      "step": 106000
    },
    {
      "epoch": 11.661716171617162,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013271765758724812,
      "loss": 0.2303,
      "num_input_tokens_seen": 22372544,
      "step": 106005
    },
    {
      "epoch": 11.662266226622663,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.013270335315995493,
      "loss": 0.2319,
      "num_input_tokens_seen": 22373568,
      "step": 106010
    },
    {
      "epoch": 11.662816281628164,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.01326890488920767,
      "loss": 0.2324,
      "num_input_tokens_seen": 22374560,
      "step": 106015
    },
    {
      "epoch": 11.663366336633663,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.013267474478374517,
      "loss": 0.2324,
      "num_input_tokens_seen": 22375648,
      "step": 106020
    },
    {
      "epoch": 11.663916391639164,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013266044083509216,
      "loss": 0.2293,
      "num_input_tokens_seen": 22376704,
      "step": 106025
    },
    {
      "epoch": 11.664466446644665,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.01326461370462496,
      "loss": 0.2319,
      "num_input_tokens_seen": 22377760,
      "step": 106030
    },
    {
      "epoch": 11.665016501650165,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.013263183341734923,
      "loss": 0.234,
      "num_input_tokens_seen": 22378816,
      "step": 106035
    },
    {
      "epoch": 11.665566556655666,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.0132617529948523,
      "loss": 0.234,
      "num_input_tokens_seen": 22379872,
      "step": 106040
    },
    {
      "epoch": 11.666116611661167,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.013260322663990258,
      "loss": 0.2319,
      "num_input_tokens_seen": 22380992,
      "step": 106045
    },
    {
      "epoch": 11.666666666666666,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.013258892349161985,
      "loss": 0.2314,
      "num_input_tokens_seen": 22382112,
      "step": 106050
    },
    {
      "epoch": 11.667216721672167,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013257462050380675,
      "loss": 0.2303,
      "num_input_tokens_seen": 22383200,
      "step": 106055
    },
    {
      "epoch": 11.667766776677668,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013256031767659493,
      "loss": 0.2319,
      "num_input_tokens_seen": 22384224,
      "step": 106060
    },
    {
      "epoch": 11.668316831683168,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.013254601501011633,
      "loss": 0.2308,
      "num_input_tokens_seen": 22385216,
      "step": 106065
    },
    {
      "epoch": 11.668866886688669,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.013253171250450276,
      "loss": 0.2324,
      "num_input_tokens_seen": 22386272,
      "step": 106070
    },
    {
      "epoch": 11.66941694169417,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013251741015988595,
      "loss": 0.2303,
      "num_input_tokens_seen": 22387328,
      "step": 106075
    },
    {
      "epoch": 11.66996699669967,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013250310797639782,
      "loss": 0.2303,
      "num_input_tokens_seen": 22388352,
      "step": 106080
    },
    {
      "epoch": 11.67051705170517,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013248880595417012,
      "loss": 0.2298,
      "num_input_tokens_seen": 22389408,
      "step": 106085
    },
    {
      "epoch": 11.671067106710671,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.01324745040933347,
      "loss": 0.2303,
      "num_input_tokens_seen": 22390496,
      "step": 106090
    },
    {
      "epoch": 11.671617161716172,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013246020239402342,
      "loss": 0.2329,
      "num_input_tokens_seen": 22391552,
      "step": 106095
    },
    {
      "epoch": 11.672167216721672,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013244590085636797,
      "loss": 0.2314,
      "num_input_tokens_seen": 22392704,
      "step": 106100
    },
    {
      "epoch": 11.672717271727173,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013243159948050027,
      "loss": 0.2319,
      "num_input_tokens_seen": 22393760,
      "step": 106105
    },
    {
      "epoch": 11.673267326732674,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.013241729826655203,
      "loss": 0.2319,
      "num_input_tokens_seen": 22394816,
      "step": 106110
    },
    {
      "epoch": 11.673817381738173,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01324029972146552,
      "loss": 0.234,
      "num_input_tokens_seen": 22395872,
      "step": 106115
    },
    {
      "epoch": 11.674367436743674,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.013238869632494146,
      "loss": 0.2308,
      "num_input_tokens_seen": 22396896,
      "step": 106120
    },
    {
      "epoch": 11.674917491749175,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.013237439559754262,
      "loss": 0.2309,
      "num_input_tokens_seen": 22397920,
      "step": 106125
    },
    {
      "epoch": 11.675467546754675,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013236009503259061,
      "loss": 0.2308,
      "num_input_tokens_seen": 22398976,
      "step": 106130
    },
    {
      "epoch": 11.676017601760176,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.013234579463021704,
      "loss": 0.2319,
      "num_input_tokens_seen": 22399968,
      "step": 106135
    },
    {
      "epoch": 11.676567656765677,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.013233149439055385,
      "loss": 0.2324,
      "num_input_tokens_seen": 22400992,
      "step": 106140
    },
    {
      "epoch": 11.677117711771178,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013231719431373282,
      "loss": 0.2319,
      "num_input_tokens_seen": 22401952,
      "step": 106145
    },
    {
      "epoch": 11.677667766776677,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.013230289439988566,
      "loss": 0.2309,
      "num_input_tokens_seen": 22403040,
      "step": 106150
    },
    {
      "epoch": 11.678217821782178,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.013228859464914432,
      "loss": 0.2319,
      "num_input_tokens_seen": 22404032,
      "step": 106155
    },
    {
      "epoch": 11.67876787678768,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013227429506164044,
      "loss": 0.2298,
      "num_input_tokens_seen": 22405120,
      "step": 106160
    },
    {
      "epoch": 11.679317931793179,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.013225999563750588,
      "loss": 0.2319,
      "num_input_tokens_seen": 22406112,
      "step": 106165
    },
    {
      "epoch": 11.67986798679868,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.013224569637687249,
      "loss": 0.2319,
      "num_input_tokens_seen": 22407232,
      "step": 106170
    },
    {
      "epoch": 11.680418041804181,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.013223139727987193,
      "loss": 0.2298,
      "num_input_tokens_seen": 22408352,
      "step": 106175
    },
    {
      "epoch": 11.68096809680968,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.013221709834663607,
      "loss": 0.2319,
      "num_input_tokens_seen": 22409504,
      "step": 106180
    },
    {
      "epoch": 11.681518151815181,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.013220279957729673,
      "loss": 0.2314,
      "num_input_tokens_seen": 22410496,
      "step": 106185
    },
    {
      "epoch": 11.682068206820682,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.013218850097198557,
      "loss": 0.2308,
      "num_input_tokens_seen": 22411552,
      "step": 106190
    },
    {
      "epoch": 11.682618261826182,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.013217420253083449,
      "loss": 0.2329,
      "num_input_tokens_seen": 22412544,
      "step": 106195
    },
    {
      "epoch": 11.683168316831683,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.01321599042539752,
      "loss": 0.2329,
      "num_input_tokens_seen": 22413632,
      "step": 106200
    },
    {
      "epoch": 11.683718371837184,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013214560614153954,
      "loss": 0.2324,
      "num_input_tokens_seen": 22414688,
      "step": 106205
    },
    {
      "epoch": 11.684268426842685,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01321313081936593,
      "loss": 0.2309,
      "num_input_tokens_seen": 22415680,
      "step": 106210
    },
    {
      "epoch": 11.684818481848184,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.013211701041046613,
      "loss": 0.2304,
      "num_input_tokens_seen": 22416832,
      "step": 106215
    },
    {
      "epoch": 11.685368536853685,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013210271279209195,
      "loss": 0.2319,
      "num_input_tokens_seen": 22417888,
      "step": 106220
    },
    {
      "epoch": 11.685918591859187,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.013208841533866846,
      "loss": 0.2319,
      "num_input_tokens_seen": 22418912,
      "step": 106225
    },
    {
      "epoch": 11.686468646864686,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.013207411805032748,
      "loss": 0.2308,
      "num_input_tokens_seen": 22420000,
      "step": 106230
    },
    {
      "epoch": 11.687018701870187,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013205982092720073,
      "loss": 0.2329,
      "num_input_tokens_seen": 22420992,
      "step": 106235
    },
    {
      "epoch": 11.687568756875688,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.013204552396941998,
      "loss": 0.2308,
      "num_input_tokens_seen": 22422016,
      "step": 106240
    },
    {
      "epoch": 11.688118811881187,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013203122717711709,
      "loss": 0.2303,
      "num_input_tokens_seen": 22423168,
      "step": 106245
    },
    {
      "epoch": 11.688668866886688,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.013201693055042367,
      "loss": 0.2314,
      "num_input_tokens_seen": 22424224,
      "step": 106250
    },
    {
      "epoch": 11.68921892189219,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.013200263408947163,
      "loss": 0.2329,
      "num_input_tokens_seen": 22425248,
      "step": 106255
    },
    {
      "epoch": 11.689768976897689,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.013198833779439268,
      "loss": 0.2309,
      "num_input_tokens_seen": 22426368,
      "step": 106260
    },
    {
      "epoch": 11.69031903190319,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01319740416653185,
      "loss": 0.234,
      "num_input_tokens_seen": 22427424,
      "step": 106265
    },
    {
      "epoch": 11.690869086908691,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013195974570238102,
      "loss": 0.2303,
      "num_input_tokens_seen": 22428544,
      "step": 106270
    },
    {
      "epoch": 11.691419141914192,
      "grad_norm": 0.000667572021484375,
      "learning_rate": 0.013194544990571183,
      "loss": 0.2293,
      "num_input_tokens_seen": 22429600,
      "step": 106275
    },
    {
      "epoch": 11.691969196919691,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.013193115427544283,
      "loss": 0.2298,
      "num_input_tokens_seen": 22430688,
      "step": 106280
    },
    {
      "epoch": 11.692519251925193,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01319168588117057,
      "loss": 0.2314,
      "num_input_tokens_seen": 22431680,
      "step": 106285
    },
    {
      "epoch": 11.693069306930694,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.013190256351463217,
      "loss": 0.2335,
      "num_input_tokens_seen": 22432800,
      "step": 106290
    },
    {
      "epoch": 11.693619361936193,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.013188826838435407,
      "loss": 0.2283,
      "num_input_tokens_seen": 22433920,
      "step": 106295
    },
    {
      "epoch": 11.694169416941694,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.013187397342100305,
      "loss": 0.2314,
      "num_input_tokens_seen": 22434976,
      "step": 106300
    },
    {
      "epoch": 11.694719471947195,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013185967862471101,
      "loss": 0.2309,
      "num_input_tokens_seen": 22436032,
      "step": 106305
    },
    {
      "epoch": 11.695269526952695,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013184538399560956,
      "loss": 0.2319,
      "num_input_tokens_seen": 22437088,
      "step": 106310
    },
    {
      "epoch": 11.695819581958196,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.013183108953383045,
      "loss": 0.2314,
      "num_input_tokens_seen": 22438080,
      "step": 106315
    },
    {
      "epoch": 11.696369636963697,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.013181679523950555,
      "loss": 0.2361,
      "num_input_tokens_seen": 22439200,
      "step": 106320
    },
    {
      "epoch": 11.696919691969196,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.013180250111276643,
      "loss": 0.2319,
      "num_input_tokens_seen": 22440256,
      "step": 106325
    },
    {
      "epoch": 11.697469746974697,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.013178820715374502,
      "loss": 0.2304,
      "num_input_tokens_seen": 22441312,
      "step": 106330
    },
    {
      "epoch": 11.698019801980198,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.013177391336257291,
      "loss": 0.233,
      "num_input_tokens_seen": 22442368,
      "step": 106335
    },
    {
      "epoch": 11.6985698569857,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.013175961973938189,
      "loss": 0.2314,
      "num_input_tokens_seen": 22443360,
      "step": 106340
    },
    {
      "epoch": 11.699119911991199,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.013174532628430376,
      "loss": 0.2288,
      "num_input_tokens_seen": 22444384,
      "step": 106345
    },
    {
      "epoch": 11.6996699669967,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013173103299747014,
      "loss": 0.2298,
      "num_input_tokens_seen": 22445408,
      "step": 106350
    },
    {
      "epoch": 11.7002200220022,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.013171673987901283,
      "loss": 0.2314,
      "num_input_tokens_seen": 22446432,
      "step": 106355
    },
    {
      "epoch": 11.7007700770077,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.013170244692906362,
      "loss": 0.2309,
      "num_input_tokens_seen": 22447552,
      "step": 106360
    },
    {
      "epoch": 11.701320132013201,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.013168815414775409,
      "loss": 0.2324,
      "num_input_tokens_seen": 22448544,
      "step": 106365
    },
    {
      "epoch": 11.701870187018702,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.01316738615352161,
      "loss": 0.2324,
      "num_input_tokens_seen": 22449600,
      "step": 106370
    },
    {
      "epoch": 11.702420242024202,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.013165956909158128,
      "loss": 0.2314,
      "num_input_tokens_seen": 22450720,
      "step": 106375
    },
    {
      "epoch": 11.702970297029703,
      "grad_norm": 0.000782012939453125,
      "learning_rate": 0.01316452768169815,
      "loss": 0.2298,
      "num_input_tokens_seen": 22451808,
      "step": 106380
    },
    {
      "epoch": 11.703520352035204,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.013163098471154835,
      "loss": 0.2335,
      "num_input_tokens_seen": 22452928,
      "step": 106385
    },
    {
      "epoch": 11.704070407040705,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.013161669277541359,
      "loss": 0.2309,
      "num_input_tokens_seen": 22454016,
      "step": 106390
    },
    {
      "epoch": 11.704620462046204,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.013160240100870895,
      "loss": 0.2314,
      "num_input_tokens_seen": 22455008,
      "step": 106395
    },
    {
      "epoch": 11.705170517051705,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.013158810941156621,
      "loss": 0.2293,
      "num_input_tokens_seen": 22456064,
      "step": 106400
    },
    {
      "epoch": 11.705720572057206,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.013157381798411695,
      "loss": 0.2288,
      "num_input_tokens_seen": 22457088,
      "step": 106405
    },
    {
      "epoch": 11.706270627062706,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.013155952672649303,
      "loss": 0.2303,
      "num_input_tokens_seen": 22458112,
      "step": 106410
    },
    {
      "epoch": 11.706820682068207,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.013154523563882605,
      "loss": 0.2304,
      "num_input_tokens_seen": 22459200,
      "step": 106415
    },
    {
      "epoch": 11.707370737073708,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.013153094472124783,
      "loss": 0.2298,
      "num_input_tokens_seen": 22460192,
      "step": 106420
    },
    {
      "epoch": 11.707920792079207,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.013151665397389002,
      "loss": 0.2319,
      "num_input_tokens_seen": 22461280,
      "step": 106425
    },
    {
      "epoch": 11.708470847084708,
      "grad_norm": 0.0098876953125,
      "learning_rate": 0.01315023633968843,
      "loss": 0.2314,
      "num_input_tokens_seen": 22462304,
      "step": 106430
    },
    {
      "epoch": 11.70902090209021,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.013148807299036247,
      "loss": 0.2314,
      "num_input_tokens_seen": 22463328,
      "step": 106435
    },
    {
      "epoch": 11.70957095709571,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.013147378275445614,
      "loss": 0.2319,
      "num_input_tokens_seen": 22464352,
      "step": 106440
    },
    {
      "epoch": 11.71012101210121,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013145949268929709,
      "loss": 0.2309,
      "num_input_tokens_seen": 22465376,
      "step": 106445
    },
    {
      "epoch": 11.710671067106711,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.013144520279501701,
      "loss": 0.2314,
      "num_input_tokens_seen": 22466400,
      "step": 106450
    },
    {
      "epoch": 11.711221122112212,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.013143091307174755,
      "loss": 0.2329,
      "num_input_tokens_seen": 22467584,
      "step": 106455
    },
    {
      "epoch": 11.711771177117711,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013141662351962052,
      "loss": 0.2345,
      "num_input_tokens_seen": 22468640,
      "step": 106460
    },
    {
      "epoch": 11.712321232123212,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01314023341387675,
      "loss": 0.2299,
      "num_input_tokens_seen": 22469760,
      "step": 106465
    },
    {
      "epoch": 11.712871287128714,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.013138804492932027,
      "loss": 0.2309,
      "num_input_tokens_seen": 22470816,
      "step": 106470
    },
    {
      "epoch": 11.713421342134213,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.013137375589141052,
      "loss": 0.2298,
      "num_input_tokens_seen": 22471872,
      "step": 106475
    },
    {
      "epoch": 11.713971397139714,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.013135946702516986,
      "loss": 0.2304,
      "num_input_tokens_seen": 22472928,
      "step": 106480
    },
    {
      "epoch": 11.714521452145215,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.01313451783307301,
      "loss": 0.2324,
      "num_input_tokens_seen": 22474016,
      "step": 106485
    },
    {
      "epoch": 11.715071507150714,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013133088980822282,
      "loss": 0.2324,
      "num_input_tokens_seen": 22475072,
      "step": 106490
    },
    {
      "epoch": 11.715621562156215,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.013131660145777985,
      "loss": 0.2303,
      "num_input_tokens_seen": 22476064,
      "step": 106495
    },
    {
      "epoch": 11.716171617161717,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.013130231327953277,
      "loss": 0.2298,
      "num_input_tokens_seen": 22477120,
      "step": 106500
    },
    {
      "epoch": 11.716721672167218,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013128802527361327,
      "loss": 0.2324,
      "num_input_tokens_seen": 22478176,
      "step": 106505
    },
    {
      "epoch": 11.717271727172717,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.013127373744015307,
      "loss": 0.2319,
      "num_input_tokens_seen": 22479168,
      "step": 106510
    },
    {
      "epoch": 11.717821782178218,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.013125944977928385,
      "loss": 0.2324,
      "num_input_tokens_seen": 22480224,
      "step": 106515
    },
    {
      "epoch": 11.718371837183719,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.013124516229113732,
      "loss": 0.2335,
      "num_input_tokens_seen": 22481312,
      "step": 106520
    },
    {
      "epoch": 11.718921892189218,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.01312308749758451,
      "loss": 0.2303,
      "num_input_tokens_seen": 22482336,
      "step": 106525
    },
    {
      "epoch": 11.71947194719472,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.013121658783353888,
      "loss": 0.2319,
      "num_input_tokens_seen": 22483296,
      "step": 106530
    },
    {
      "epoch": 11.72002200220022,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.013120230086435042,
      "loss": 0.2319,
      "num_input_tokens_seen": 22484384,
      "step": 106535
    },
    {
      "epoch": 11.72057205720572,
      "grad_norm": 0.00067138671875,
      "learning_rate": 0.013118801406841126,
      "loss": 0.2319,
      "num_input_tokens_seen": 22485376,
      "step": 106540
    },
    {
      "epoch": 11.721122112211221,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.01311737274458532,
      "loss": 0.2319,
      "num_input_tokens_seen": 22486400,
      "step": 106545
    },
    {
      "epoch": 11.721672167216722,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.013115944099680787,
      "loss": 0.2288,
      "num_input_tokens_seen": 22487424,
      "step": 106550
    },
    {
      "epoch": 11.722222222222221,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.013114515472140688,
      "loss": 0.2319,
      "num_input_tokens_seen": 22488448,
      "step": 106555
    },
    {
      "epoch": 11.722772277227723,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013113086861978198,
      "loss": 0.2329,
      "num_input_tokens_seen": 22489472,
      "step": 106560
    },
    {
      "epoch": 11.723322332233224,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013111658269206478,
      "loss": 0.234,
      "num_input_tokens_seen": 22490528,
      "step": 106565
    },
    {
      "epoch": 11.723872387238725,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.013110229693838701,
      "loss": 0.2314,
      "num_input_tokens_seen": 22491616,
      "step": 106570
    },
    {
      "epoch": 11.724422442244224,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013108801135888032,
      "loss": 0.2324,
      "num_input_tokens_seen": 22492672,
      "step": 106575
    },
    {
      "epoch": 11.724972497249725,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.013107372595367631,
      "loss": 0.2319,
      "num_input_tokens_seen": 22493760,
      "step": 106580
    },
    {
      "epoch": 11.725522552255226,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.01310594407229067,
      "loss": 0.2319,
      "num_input_tokens_seen": 22494816,
      "step": 106585
    },
    {
      "epoch": 11.726072607260726,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.013104515566670317,
      "loss": 0.2324,
      "num_input_tokens_seen": 22495872,
      "step": 106590
    },
    {
      "epoch": 11.726622662266227,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.013103087078519728,
      "loss": 0.2314,
      "num_input_tokens_seen": 22496928,
      "step": 106595
    },
    {
      "epoch": 11.727172717271728,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.01310165860785208,
      "loss": 0.2324,
      "num_input_tokens_seen": 22497984,
      "step": 106600
    },
    {
      "epoch": 11.727722772277227,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01310023015468053,
      "loss": 0.2303,
      "num_input_tokens_seen": 22499008,
      "step": 106605
    },
    {
      "epoch": 11.728272827282728,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.013098801719018251,
      "loss": 0.2308,
      "num_input_tokens_seen": 22500032,
      "step": 106610
    },
    {
      "epoch": 11.72882288228823,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.013097373300878404,
      "loss": 0.2319,
      "num_input_tokens_seen": 22501024,
      "step": 106615
    },
    {
      "epoch": 11.729372937293729,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01309594490027415,
      "loss": 0.2329,
      "num_input_tokens_seen": 22502048,
      "step": 106620
    },
    {
      "epoch": 11.72992299229923,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.013094516517218666,
      "loss": 0.2319,
      "num_input_tokens_seen": 22503136,
      "step": 106625
    },
    {
      "epoch": 11.73047304730473,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.013093088151725097,
      "loss": 0.2314,
      "num_input_tokens_seen": 22504128,
      "step": 106630
    },
    {
      "epoch": 11.731023102310232,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.013091659803806631,
      "loss": 0.2283,
      "num_input_tokens_seen": 22505152,
      "step": 106635
    },
    {
      "epoch": 11.731573157315731,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.013090231473476417,
      "loss": 0.2303,
      "num_input_tokens_seen": 22506240,
      "step": 106640
    },
    {
      "epoch": 11.732123212321232,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01308880316074762,
      "loss": 0.2314,
      "num_input_tokens_seen": 22507296,
      "step": 106645
    },
    {
      "epoch": 11.732673267326733,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.013087374865633411,
      "loss": 0.2313,
      "num_input_tokens_seen": 22508352,
      "step": 106650
    },
    {
      "epoch": 11.733223322332233,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.013085946588146947,
      "loss": 0.2298,
      "num_input_tokens_seen": 22509408,
      "step": 106655
    },
    {
      "epoch": 11.733773377337734,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.013084518328301398,
      "loss": 0.2319,
      "num_input_tokens_seen": 22510400,
      "step": 106660
    },
    {
      "epoch": 11.734323432343235,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.013083090086109925,
      "loss": 0.2309,
      "num_input_tokens_seen": 22511392,
      "step": 106665
    },
    {
      "epoch": 11.734873487348734,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.013081661861585688,
      "loss": 0.2308,
      "num_input_tokens_seen": 22512544,
      "step": 106670
    },
    {
      "epoch": 11.735423542354235,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.013080233654741855,
      "loss": 0.2345,
      "num_input_tokens_seen": 22513664,
      "step": 106675
    },
    {
      "epoch": 11.735973597359736,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.013078805465591583,
      "loss": 0.2308,
      "num_input_tokens_seen": 22514720,
      "step": 106680
    },
    {
      "epoch": 11.736523652365236,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.013077377294148046,
      "loss": 0.2335,
      "num_input_tokens_seen": 22515712,
      "step": 106685
    },
    {
      "epoch": 11.737073707370737,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013075949140424396,
      "loss": 0.2319,
      "num_input_tokens_seen": 22516800,
      "step": 106690
    },
    {
      "epoch": 11.737623762376238,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.013074521004433797,
      "loss": 0.2314,
      "num_input_tokens_seen": 22517856,
      "step": 106695
    },
    {
      "epoch": 11.738173817381739,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.013073092886189417,
      "loss": 0.2308,
      "num_input_tokens_seen": 22518848,
      "step": 106700
    },
    {
      "epoch": 11.738723872387238,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013071664785704411,
      "loss": 0.2319,
      "num_input_tokens_seen": 22519968,
      "step": 106705
    },
    {
      "epoch": 11.73927392739274,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.013070236702991954,
      "loss": 0.2319,
      "num_input_tokens_seen": 22521024,
      "step": 106710
    },
    {
      "epoch": 11.73982398239824,
      "grad_norm": 0.000797271728515625,
      "learning_rate": 0.013068808638065196,
      "loss": 0.2314,
      "num_input_tokens_seen": 22522048,
      "step": 106715
    },
    {
      "epoch": 11.74037403740374,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.013067380590937297,
      "loss": 0.2319,
      "num_input_tokens_seen": 22523104,
      "step": 106720
    },
    {
      "epoch": 11.74092409240924,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013065952561621432,
      "loss": 0.2303,
      "num_input_tokens_seen": 22524064,
      "step": 106725
    },
    {
      "epoch": 11.741474147414742,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013064524550130746,
      "loss": 0.2319,
      "num_input_tokens_seen": 22525088,
      "step": 106730
    },
    {
      "epoch": 11.742024202420241,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013063096556478412,
      "loss": 0.2319,
      "num_input_tokens_seen": 22526112,
      "step": 106735
    },
    {
      "epoch": 11.742574257425742,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.013061668580677595,
      "loss": 0.2314,
      "num_input_tokens_seen": 22527168,
      "step": 106740
    },
    {
      "epoch": 11.743124312431243,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.013060240622741434,
      "loss": 0.2308,
      "num_input_tokens_seen": 22528128,
      "step": 106745
    },
    {
      "epoch": 11.743674367436743,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.013058812682683116,
      "loss": 0.2303,
      "num_input_tokens_seen": 22529152,
      "step": 106750
    },
    {
      "epoch": 11.744224422442244,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.013057384760515782,
      "loss": 0.2298,
      "num_input_tokens_seen": 22530240,
      "step": 106755
    },
    {
      "epoch": 11.744774477447745,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.013055956856252606,
      "loss": 0.2303,
      "num_input_tokens_seen": 22531296,
      "step": 106760
    },
    {
      "epoch": 11.745324532453246,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013054528969906744,
      "loss": 0.2324,
      "num_input_tokens_seen": 22532448,
      "step": 106765
    },
    {
      "epoch": 11.745874587458745,
      "grad_norm": 0.00072479248046875,
      "learning_rate": 0.01305310110149135,
      "loss": 0.2335,
      "num_input_tokens_seen": 22533536,
      "step": 106770
    },
    {
      "epoch": 11.746424642464246,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.013051673251019592,
      "loss": 0.2304,
      "num_input_tokens_seen": 22534656,
      "step": 106775
    },
    {
      "epoch": 11.746974697469748,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.013050245418504629,
      "loss": 0.234,
      "num_input_tokens_seen": 22535744,
      "step": 106780
    },
    {
      "epoch": 11.747524752475247,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.013048817603959612,
      "loss": 0.2324,
      "num_input_tokens_seen": 22536832,
      "step": 106785
    },
    {
      "epoch": 11.748074807480748,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.01304738980739771,
      "loss": 0.2324,
      "num_input_tokens_seen": 22537856,
      "step": 106790
    },
    {
      "epoch": 11.748624862486249,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.013045962028832077,
      "loss": 0.2308,
      "num_input_tokens_seen": 22538912,
      "step": 106795
    },
    {
      "epoch": 11.749174917491748,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01304453426827588,
      "loss": 0.2319,
      "num_input_tokens_seen": 22539968,
      "step": 106800
    },
    {
      "epoch": 11.74972497249725,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.01304310652574227,
      "loss": 0.2303,
      "num_input_tokens_seen": 22541024,
      "step": 106805
    },
    {
      "epoch": 11.75027502750275,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013041678801244403,
      "loss": 0.2314,
      "num_input_tokens_seen": 22542048,
      "step": 106810
    },
    {
      "epoch": 11.750825082508252,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013040251094795448,
      "loss": 0.2314,
      "num_input_tokens_seen": 22543104,
      "step": 106815
    },
    {
      "epoch": 11.751375137513751,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.013038823406408554,
      "loss": 0.2329,
      "num_input_tokens_seen": 22544192,
      "step": 106820
    },
    {
      "epoch": 11.751925192519252,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.013037395736096892,
      "loss": 0.2319,
      "num_input_tokens_seen": 22545248,
      "step": 106825
    },
    {
      "epoch": 11.752475247524753,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.013035968083873606,
      "loss": 0.2314,
      "num_input_tokens_seen": 22546272,
      "step": 106830
    },
    {
      "epoch": 11.753025302530252,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.013034540449751858,
      "loss": 0.2314,
      "num_input_tokens_seen": 22547424,
      "step": 106835
    },
    {
      "epoch": 11.753575357535754,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013033112833744814,
      "loss": 0.2335,
      "num_input_tokens_seen": 22548512,
      "step": 106840
    },
    {
      "epoch": 11.754125412541255,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.013031685235865616,
      "loss": 0.2319,
      "num_input_tokens_seen": 22549632,
      "step": 106845
    },
    {
      "epoch": 11.754675467546754,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.013030257656127437,
      "loss": 0.2314,
      "num_input_tokens_seen": 22550720,
      "step": 106850
    },
    {
      "epoch": 11.755225522552255,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01302883009454343,
      "loss": 0.2309,
      "num_input_tokens_seen": 22551872,
      "step": 106855
    },
    {
      "epoch": 11.755775577557756,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.013027402551126744,
      "loss": 0.2314,
      "num_input_tokens_seen": 22552928,
      "step": 106860
    },
    {
      "epoch": 11.756325632563257,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013025975025890547,
      "loss": 0.2314,
      "num_input_tokens_seen": 22553888,
      "step": 106865
    },
    {
      "epoch": 11.756875687568757,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.013024547518847985,
      "loss": 0.2319,
      "num_input_tokens_seen": 22554848,
      "step": 106870
    },
    {
      "epoch": 11.757425742574258,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.013023120030012224,
      "loss": 0.2298,
      "num_input_tokens_seen": 22555872,
      "step": 106875
    },
    {
      "epoch": 11.757975797579759,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.013021692559396423,
      "loss": 0.2324,
      "num_input_tokens_seen": 22556960,
      "step": 106880
    },
    {
      "epoch": 11.758525852585258,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.013020265107013724,
      "loss": 0.2309,
      "num_input_tokens_seen": 22558016,
      "step": 106885
    },
    {
      "epoch": 11.75907590759076,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.013018837672877294,
      "loss": 0.2309,
      "num_input_tokens_seen": 22559104,
      "step": 106890
    },
    {
      "epoch": 11.75962596259626,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.013017410257000284,
      "loss": 0.2314,
      "num_input_tokens_seen": 22560192,
      "step": 106895
    },
    {
      "epoch": 11.76017601760176,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.01301598285939586,
      "loss": 0.2309,
      "num_input_tokens_seen": 22561216,
      "step": 106900
    },
    {
      "epoch": 11.76072607260726,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.013014555480077168,
      "loss": 0.2314,
      "num_input_tokens_seen": 22562272,
      "step": 106905
    },
    {
      "epoch": 11.761276127612762,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.013013128119057358,
      "loss": 0.2298,
      "num_input_tokens_seen": 22563296,
      "step": 106910
    },
    {
      "epoch": 11.761826182618261,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.013011700776349602,
      "loss": 0.2319,
      "num_input_tokens_seen": 22564448,
      "step": 106915
    },
    {
      "epoch": 11.762376237623762,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01301027345196704,
      "loss": 0.2324,
      "num_input_tokens_seen": 22565536,
      "step": 106920
    },
    {
      "epoch": 11.762926292629263,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.013008846145922834,
      "loss": 0.2334,
      "num_input_tokens_seen": 22566528,
      "step": 106925
    },
    {
      "epoch": 11.763476347634764,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.013007418858230141,
      "loss": 0.2313,
      "num_input_tokens_seen": 22567552,
      "step": 106930
    },
    {
      "epoch": 11.764026402640264,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.013005991588902102,
      "loss": 0.2303,
      "num_input_tokens_seen": 22568576,
      "step": 106935
    },
    {
      "epoch": 11.764576457645765,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.013004564337951894,
      "loss": 0.2324,
      "num_input_tokens_seen": 22569632,
      "step": 106940
    },
    {
      "epoch": 11.765126512651266,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.013003137105392651,
      "loss": 0.2356,
      "num_input_tokens_seen": 22570688,
      "step": 106945
    },
    {
      "epoch": 11.765676567656765,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.013001709891237537,
      "loss": 0.2309,
      "num_input_tokens_seen": 22571744,
      "step": 106950
    },
    {
      "epoch": 11.766226622662266,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01300028269549971,
      "loss": 0.2298,
      "num_input_tokens_seen": 22572832,
      "step": 106955
    },
    {
      "epoch": 11.766776677667767,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.012998855518192309,
      "loss": 0.2319,
      "num_input_tokens_seen": 22573952,
      "step": 106960
    },
    {
      "epoch": 11.767326732673267,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.012997428359328501,
      "loss": 0.2324,
      "num_input_tokens_seen": 22574976,
      "step": 106965
    },
    {
      "epoch": 11.767876787678768,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.012996001218921438,
      "loss": 0.2314,
      "num_input_tokens_seen": 22576032,
      "step": 106970
    },
    {
      "epoch": 11.768426842684269,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012994574096984264,
      "loss": 0.2319,
      "num_input_tokens_seen": 22577120,
      "step": 106975
    },
    {
      "epoch": 11.768976897689768,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01299314699353014,
      "loss": 0.234,
      "num_input_tokens_seen": 22578208,
      "step": 106980
    },
    {
      "epoch": 11.76952695269527,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.012991719908572216,
      "loss": 0.2324,
      "num_input_tokens_seen": 22579264,
      "step": 106985
    },
    {
      "epoch": 11.77007700770077,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012990292842123647,
      "loss": 0.2319,
      "num_input_tokens_seen": 22580288,
      "step": 106990
    },
    {
      "epoch": 11.770627062706271,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01298886579419759,
      "loss": 0.2308,
      "num_input_tokens_seen": 22581344,
      "step": 106995
    },
    {
      "epoch": 11.77117711771177,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.012987438764807185,
      "loss": 0.2293,
      "num_input_tokens_seen": 22582368,
      "step": 107000
    },
    {
      "epoch": 11.771727172717272,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.012986011753965592,
      "loss": 0.2314,
      "num_input_tokens_seen": 22583392,
      "step": 107005
    },
    {
      "epoch": 11.772277227722773,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012984584761685961,
      "loss": 0.2309,
      "num_input_tokens_seen": 22584512,
      "step": 107010
    },
    {
      "epoch": 11.772827282728272,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.012983157787981453,
      "loss": 0.2303,
      "num_input_tokens_seen": 22585600,
      "step": 107015
    },
    {
      "epoch": 11.773377337733773,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012981730832865206,
      "loss": 0.2309,
      "num_input_tokens_seen": 22586656,
      "step": 107020
    },
    {
      "epoch": 11.773927392739274,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.012980303896350376,
      "loss": 0.2303,
      "num_input_tokens_seen": 22587744,
      "step": 107025
    },
    {
      "epoch": 11.774477447744774,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012978876978450122,
      "loss": 0.2303,
      "num_input_tokens_seen": 22588736,
      "step": 107030
    },
    {
      "epoch": 11.775027502750275,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.012977450079177581,
      "loss": 0.2319,
      "num_input_tokens_seen": 22589824,
      "step": 107035
    },
    {
      "epoch": 11.775577557755776,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.012976023198545917,
      "loss": 0.2278,
      "num_input_tokens_seen": 22590848,
      "step": 107040
    },
    {
      "epoch": 11.776127612761275,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.012974596336568278,
      "loss": 0.2304,
      "num_input_tokens_seen": 22591904,
      "step": 107045
    },
    {
      "epoch": 11.776677667766776,
      "grad_norm": 0.000652313232421875,
      "learning_rate": 0.012973169493257806,
      "loss": 0.2309,
      "num_input_tokens_seen": 22592928,
      "step": 107050
    },
    {
      "epoch": 11.777227722772277,
      "grad_norm": 0.00069427490234375,
      "learning_rate": 0.012971742668627663,
      "loss": 0.2293,
      "num_input_tokens_seen": 22593952,
      "step": 107055
    },
    {
      "epoch": 11.777777777777779,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.012970315862690993,
      "loss": 0.2303,
      "num_input_tokens_seen": 22594976,
      "step": 107060
    },
    {
      "epoch": 11.778327832783278,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.012968889075460947,
      "loss": 0.2309,
      "num_input_tokens_seen": 22596032,
      "step": 107065
    },
    {
      "epoch": 11.778877887788779,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.012967462306950681,
      "loss": 0.2309,
      "num_input_tokens_seen": 22597120,
      "step": 107070
    },
    {
      "epoch": 11.77942794279428,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.012966035557173333,
      "loss": 0.2309,
      "num_input_tokens_seen": 22598080,
      "step": 107075
    },
    {
      "epoch": 11.77997799779978,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012964608826142063,
      "loss": 0.2319,
      "num_input_tokens_seen": 22599072,
      "step": 107080
    },
    {
      "epoch": 11.78052805280528,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.012963182113870013,
      "loss": 0.2319,
      "num_input_tokens_seen": 22600096,
      "step": 107085
    },
    {
      "epoch": 11.781078107810782,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.012961755420370341,
      "loss": 0.2314,
      "num_input_tokens_seen": 22601088,
      "step": 107090
    },
    {
      "epoch": 11.781628162816281,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.012960328745656191,
      "loss": 0.2314,
      "num_input_tokens_seen": 22602176,
      "step": 107095
    },
    {
      "epoch": 11.782178217821782,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.012958902089740707,
      "loss": 0.2298,
      "num_input_tokens_seen": 22603200,
      "step": 107100
    },
    {
      "epoch": 11.782728272827283,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01295747545263705,
      "loss": 0.2314,
      "num_input_tokens_seen": 22604288,
      "step": 107105
    },
    {
      "epoch": 11.783278327832782,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.012956048834358353,
      "loss": 0.233,
      "num_input_tokens_seen": 22605280,
      "step": 107110
    },
    {
      "epoch": 11.783828382838283,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01295462223491778,
      "loss": 0.2314,
      "num_input_tokens_seen": 22606400,
      "step": 107115
    },
    {
      "epoch": 11.784378437843785,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.012953195654328472,
      "loss": 0.2309,
      "num_input_tokens_seen": 22607424,
      "step": 107120
    },
    {
      "epoch": 11.784928492849286,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012951769092603574,
      "loss": 0.2298,
      "num_input_tokens_seen": 22608480,
      "step": 107125
    },
    {
      "epoch": 11.785478547854785,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.012950342549756242,
      "loss": 0.233,
      "num_input_tokens_seen": 22609504,
      "step": 107130
    },
    {
      "epoch": 11.786028602860286,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.012948916025799614,
      "loss": 0.2314,
      "num_input_tokens_seen": 22610560,
      "step": 107135
    },
    {
      "epoch": 11.786578657865787,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012947489520746848,
      "loss": 0.2314,
      "num_input_tokens_seen": 22611520,
      "step": 107140
    },
    {
      "epoch": 11.787128712871286,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.012946063034611087,
      "loss": 0.2283,
      "num_input_tokens_seen": 22612640,
      "step": 107145
    },
    {
      "epoch": 11.787678767876788,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.01294463656740547,
      "loss": 0.2324,
      "num_input_tokens_seen": 22613728,
      "step": 107150
    },
    {
      "epoch": 11.788228822882289,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.012943210119143156,
      "loss": 0.2298,
      "num_input_tokens_seen": 22614848,
      "step": 107155
    },
    {
      "epoch": 11.788778877887788,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.012941783689837292,
      "loss": 0.2304,
      "num_input_tokens_seen": 22615840,
      "step": 107160
    },
    {
      "epoch": 11.789328932893289,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.01294035727950101,
      "loss": 0.2303,
      "num_input_tokens_seen": 22616832,
      "step": 107165
    },
    {
      "epoch": 11.78987898789879,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012938930888147472,
      "loss": 0.2314,
      "num_input_tokens_seen": 22617920,
      "step": 107170
    },
    {
      "epoch": 11.79042904290429,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012937504515789818,
      "loss": 0.2314,
      "num_input_tokens_seen": 22619008,
      "step": 107175
    },
    {
      "epoch": 11.79097909790979,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.012936078162441194,
      "loss": 0.2298,
      "num_input_tokens_seen": 22620064,
      "step": 107180
    },
    {
      "epoch": 11.791529152915292,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.012934651828114754,
      "loss": 0.233,
      "num_input_tokens_seen": 22621152,
      "step": 107185
    },
    {
      "epoch": 11.792079207920793,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.012933225512823629,
      "loss": 0.2298,
      "num_input_tokens_seen": 22622240,
      "step": 107190
    },
    {
      "epoch": 11.792629262926292,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.012931799216580976,
      "loss": 0.2303,
      "num_input_tokens_seen": 22623328,
      "step": 107195
    },
    {
      "epoch": 11.793179317931793,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.012930372939399935,
      "loss": 0.2319,
      "num_input_tokens_seen": 22624352,
      "step": 107200
    },
    {
      "epoch": 11.793729372937294,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.012928946681293659,
      "loss": 0.2303,
      "num_input_tokens_seen": 22625408,
      "step": 107205
    },
    {
      "epoch": 11.794279427942794,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.012927520442275285,
      "loss": 0.2309,
      "num_input_tokens_seen": 22626496,
      "step": 107210
    },
    {
      "epoch": 11.794829482948295,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.012926094222357958,
      "loss": 0.2303,
      "num_input_tokens_seen": 22627520,
      "step": 107215
    },
    {
      "epoch": 11.795379537953796,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.012924668021554831,
      "loss": 0.2303,
      "num_input_tokens_seen": 22628608,
      "step": 107220
    },
    {
      "epoch": 11.795929592959295,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.012923241839879035,
      "loss": 0.2304,
      "num_input_tokens_seen": 22629568,
      "step": 107225
    },
    {
      "epoch": 11.796479647964796,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012921815677343728,
      "loss": 0.2309,
      "num_input_tokens_seen": 22630592,
      "step": 107230
    },
    {
      "epoch": 11.797029702970297,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.012920389533962047,
      "loss": 0.233,
      "num_input_tokens_seen": 22631616,
      "step": 107235
    },
    {
      "epoch": 11.797579757975798,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.012918963409747136,
      "loss": 0.2303,
      "num_input_tokens_seen": 22632704,
      "step": 107240
    },
    {
      "epoch": 11.798129812981298,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.012917537304712146,
      "loss": 0.2309,
      "num_input_tokens_seen": 22633696,
      "step": 107245
    },
    {
      "epoch": 11.798679867986799,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01291611121887021,
      "loss": 0.2324,
      "num_input_tokens_seen": 22634752,
      "step": 107250
    },
    {
      "epoch": 11.7992299229923,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.012914685152234479,
      "loss": 0.2309,
      "num_input_tokens_seen": 22635808,
      "step": 107255
    },
    {
      "epoch": 11.7997799779978,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.012913259104818098,
      "loss": 0.2314,
      "num_input_tokens_seen": 22636768,
      "step": 107260
    },
    {
      "epoch": 11.8003300330033,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.0129118330766342,
      "loss": 0.2324,
      "num_input_tokens_seen": 22637760,
      "step": 107265
    },
    {
      "epoch": 11.800880088008801,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.012910407067695938,
      "loss": 0.2319,
      "num_input_tokens_seen": 22638816,
      "step": 107270
    },
    {
      "epoch": 11.8014301430143,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012908981078016449,
      "loss": 0.2314,
      "num_input_tokens_seen": 22639904,
      "step": 107275
    },
    {
      "epoch": 11.801980198019802,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.012907555107608883,
      "loss": 0.2298,
      "num_input_tokens_seen": 22640928,
      "step": 107280
    },
    {
      "epoch": 11.802530253025303,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012906129156486376,
      "loss": 0.233,
      "num_input_tokens_seen": 22641952,
      "step": 107285
    },
    {
      "epoch": 11.803080308030804,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012904703224662065,
      "loss": 0.2335,
      "num_input_tokens_seen": 22643008,
      "step": 107290
    },
    {
      "epoch": 11.803630363036303,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012903277312149106,
      "loss": 0.2309,
      "num_input_tokens_seen": 22644032,
      "step": 107295
    },
    {
      "epoch": 11.804180418041804,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.012901851418960629,
      "loss": 0.2319,
      "num_input_tokens_seen": 22645152,
      "step": 107300
    },
    {
      "epoch": 11.804730473047305,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.012900425545109787,
      "loss": 0.2309,
      "num_input_tokens_seen": 22646240,
      "step": 107305
    },
    {
      "epoch": 11.805280528052805,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.012898999690609715,
      "loss": 0.2314,
      "num_input_tokens_seen": 22647328,
      "step": 107310
    },
    {
      "epoch": 11.805830583058306,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.012897573855473549,
      "loss": 0.2345,
      "num_input_tokens_seen": 22648448,
      "step": 107315
    },
    {
      "epoch": 11.806380638063807,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012896148039714441,
      "loss": 0.2303,
      "num_input_tokens_seen": 22649568,
      "step": 107320
    },
    {
      "epoch": 11.806930693069306,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.012894722243345522,
      "loss": 0.2298,
      "num_input_tokens_seen": 22650624,
      "step": 107325
    },
    {
      "epoch": 11.807480748074807,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01289329646637994,
      "loss": 0.2293,
      "num_input_tokens_seen": 22651712,
      "step": 107330
    },
    {
      "epoch": 11.808030803080309,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.01289187070883084,
      "loss": 0.2304,
      "num_input_tokens_seen": 22652800,
      "step": 107335
    },
    {
      "epoch": 11.808580858085808,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.012890444970711346,
      "loss": 0.2303,
      "num_input_tokens_seen": 22653888,
      "step": 107340
    },
    {
      "epoch": 11.809130913091309,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.012889019252034613,
      "loss": 0.2303,
      "num_input_tokens_seen": 22654912,
      "step": 107345
    },
    {
      "epoch": 11.80968096809681,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012887593552813782,
      "loss": 0.2324,
      "num_input_tokens_seen": 22655968,
      "step": 107350
    },
    {
      "epoch": 11.810231023102311,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.012886167873061974,
      "loss": 0.233,
      "num_input_tokens_seen": 22657024,
      "step": 107355
    },
    {
      "epoch": 11.81078107810781,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.012884742212792356,
      "loss": 0.2293,
      "num_input_tokens_seen": 22658048,
      "step": 107360
    },
    {
      "epoch": 11.811331133113312,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.012883316572018045,
      "loss": 0.2303,
      "num_input_tokens_seen": 22659104,
      "step": 107365
    },
    {
      "epoch": 11.811881188118813,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.012881890950752192,
      "loss": 0.2324,
      "num_input_tokens_seen": 22660128,
      "step": 107370
    },
    {
      "epoch": 11.812431243124312,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.01288046534900794,
      "loss": 0.2319,
      "num_input_tokens_seen": 22661216,
      "step": 107375
    },
    {
      "epoch": 11.812981298129813,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.012879039766798413,
      "loss": 0.2303,
      "num_input_tokens_seen": 22662240,
      "step": 107380
    },
    {
      "epoch": 11.813531353135314,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.012877614204136762,
      "loss": 0.2324,
      "num_input_tokens_seen": 22663264,
      "step": 107385
    },
    {
      "epoch": 11.814081408140813,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.01287618866103612,
      "loss": 0.2319,
      "num_input_tokens_seen": 22664352,
      "step": 107390
    },
    {
      "epoch": 11.814631463146315,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.012874763137509635,
      "loss": 0.2335,
      "num_input_tokens_seen": 22665344,
      "step": 107395
    },
    {
      "epoch": 11.815181518151816,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.012873337633570433,
      "loss": 0.2314,
      "num_input_tokens_seen": 22666368,
      "step": 107400
    },
    {
      "epoch": 11.815731573157315,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.012871912149231656,
      "loss": 0.2309,
      "num_input_tokens_seen": 22667520,
      "step": 107405
    },
    {
      "epoch": 11.816281628162816,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01287048668450645,
      "loss": 0.2309,
      "num_input_tokens_seen": 22668640,
      "step": 107410
    },
    {
      "epoch": 11.816831683168317,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.012869061239407936,
      "loss": 0.2309,
      "num_input_tokens_seen": 22669664,
      "step": 107415
    },
    {
      "epoch": 11.817381738173818,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.012867635813949273,
      "loss": 0.2314,
      "num_input_tokens_seen": 22670752,
      "step": 107420
    },
    {
      "epoch": 11.817931793179318,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.012866210408143583,
      "loss": 0.2329,
      "num_input_tokens_seen": 22671840,
      "step": 107425
    },
    {
      "epoch": 11.818481848184819,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012864785022004005,
      "loss": 0.2314,
      "num_input_tokens_seen": 22672896,
      "step": 107430
    },
    {
      "epoch": 11.81903190319032,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012863359655543685,
      "loss": 0.2303,
      "num_input_tokens_seen": 22673984,
      "step": 107435
    },
    {
      "epoch": 11.819581958195819,
      "grad_norm": 0.000812530517578125,
      "learning_rate": 0.012861934308775748,
      "loss": 0.2324,
      "num_input_tokens_seen": 22674944,
      "step": 107440
    },
    {
      "epoch": 11.82013201320132,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012860508981713339,
      "loss": 0.2319,
      "num_input_tokens_seen": 22675968,
      "step": 107445
    },
    {
      "epoch": 11.820682068206821,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.012859083674369595,
      "loss": 0.2308,
      "num_input_tokens_seen": 22677056,
      "step": 107450
    },
    {
      "epoch": 11.82123212321232,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.012857658386757644,
      "loss": 0.2314,
      "num_input_tokens_seen": 22678112,
      "step": 107455
    },
    {
      "epoch": 11.821782178217822,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012856233118890632,
      "loss": 0.2319,
      "num_input_tokens_seen": 22679136,
      "step": 107460
    },
    {
      "epoch": 11.822332233223323,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012854807870781686,
      "loss": 0.2308,
      "num_input_tokens_seen": 22680192,
      "step": 107465
    },
    {
      "epoch": 11.822882288228822,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012853382642443952,
      "loss": 0.2319,
      "num_input_tokens_seen": 22681248,
      "step": 107470
    },
    {
      "epoch": 11.823432343234323,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01285195743389056,
      "loss": 0.2319,
      "num_input_tokens_seen": 22682336,
      "step": 107475
    },
    {
      "epoch": 11.823982398239824,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01285053224513464,
      "loss": 0.233,
      "num_input_tokens_seen": 22683392,
      "step": 107480
    },
    {
      "epoch": 11.824532453245325,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.012849107076189335,
      "loss": 0.2314,
      "num_input_tokens_seen": 22684416,
      "step": 107485
    },
    {
      "epoch": 11.825082508250825,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012847681927067778,
      "loss": 0.2303,
      "num_input_tokens_seen": 22685504,
      "step": 107490
    },
    {
      "epoch": 11.825632563256326,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.012846256797783107,
      "loss": 0.2324,
      "num_input_tokens_seen": 22686592,
      "step": 107495
    },
    {
      "epoch": 11.826182618261827,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.01284483168834845,
      "loss": 0.2335,
      "num_input_tokens_seen": 22687616,
      "step": 107500
    },
    {
      "epoch": 11.826732673267326,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012843406598776945,
      "loss": 0.234,
      "num_input_tokens_seen": 22688640,
      "step": 107505
    },
    {
      "epoch": 11.827282728272827,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.012841981529081732,
      "loss": 0.2293,
      "num_input_tokens_seen": 22689696,
      "step": 107510
    },
    {
      "epoch": 11.827832783278328,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.012840556479275934,
      "loss": 0.2303,
      "num_input_tokens_seen": 22690688,
      "step": 107515
    },
    {
      "epoch": 11.828382838283828,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.012839131449372692,
      "loss": 0.2319,
      "num_input_tokens_seen": 22691744,
      "step": 107520
    },
    {
      "epoch": 11.828932893289329,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.012837706439385144,
      "loss": 0.2298,
      "num_input_tokens_seen": 22692704,
      "step": 107525
    },
    {
      "epoch": 11.82948294829483,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.012836281449326406,
      "loss": 0.2299,
      "num_input_tokens_seen": 22693696,
      "step": 107530
    },
    {
      "epoch": 11.83003300330033,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.012834856479209636,
      "loss": 0.2319,
      "num_input_tokens_seen": 22694720,
      "step": 107535
    },
    {
      "epoch": 11.83058305830583,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.01283343152904795,
      "loss": 0.232,
      "num_input_tokens_seen": 22695808,
      "step": 107540
    },
    {
      "epoch": 11.831133113311331,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.012832006598854483,
      "loss": 0.2299,
      "num_input_tokens_seen": 22696864,
      "step": 107545
    },
    {
      "epoch": 11.831683168316832,
      "grad_norm": 0.000621795654296875,
      "learning_rate": 0.012830581688642378,
      "loss": 0.232,
      "num_input_tokens_seen": 22697856,
      "step": 107550
    },
    {
      "epoch": 11.832233223322332,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.012829156798424753,
      "loss": 0.233,
      "num_input_tokens_seen": 22698880,
      "step": 107555
    },
    {
      "epoch": 11.832783278327833,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.012827731928214752,
      "loss": 0.2303,
      "num_input_tokens_seen": 22699904,
      "step": 107560
    },
    {
      "epoch": 11.833333333333334,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.012826307078025507,
      "loss": 0.2288,
      "num_input_tokens_seen": 22700928,
      "step": 107565
    },
    {
      "epoch": 11.833883388338833,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.012824882247870139,
      "loss": 0.2314,
      "num_input_tokens_seen": 22701984,
      "step": 107570
    },
    {
      "epoch": 11.834433443344334,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.012823457437761792,
      "loss": 0.233,
      "num_input_tokens_seen": 22702976,
      "step": 107575
    },
    {
      "epoch": 11.834983498349835,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012822032647713588,
      "loss": 0.2325,
      "num_input_tokens_seen": 22703968,
      "step": 107580
    },
    {
      "epoch": 11.835533553355335,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012820607877738673,
      "loss": 0.2335,
      "num_input_tokens_seen": 22705024,
      "step": 107585
    },
    {
      "epoch": 11.836083608360836,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012819183127850166,
      "loss": 0.2304,
      "num_input_tokens_seen": 22706144,
      "step": 107590
    },
    {
      "epoch": 11.836633663366337,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.012817758398061199,
      "loss": 0.2309,
      "num_input_tokens_seen": 22707168,
      "step": 107595
    },
    {
      "epoch": 11.837183718371836,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.012816333688384907,
      "loss": 0.2293,
      "num_input_tokens_seen": 22708224,
      "step": 107600
    },
    {
      "epoch": 11.837733773377337,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.012814908998834418,
      "loss": 0.2314,
      "num_input_tokens_seen": 22709280,
      "step": 107605
    },
    {
      "epoch": 11.838283828382838,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.012813484329422867,
      "loss": 0.2314,
      "num_input_tokens_seen": 22710336,
      "step": 107610
    },
    {
      "epoch": 11.83883388338834,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.012812059680163381,
      "loss": 0.2314,
      "num_input_tokens_seen": 22711488,
      "step": 107615
    },
    {
      "epoch": 11.839383938393839,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.01281063505106909,
      "loss": 0.2309,
      "num_input_tokens_seen": 22712640,
      "step": 107620
    },
    {
      "epoch": 11.83993399339934,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01280921044215313,
      "loss": 0.2314,
      "num_input_tokens_seen": 22713696,
      "step": 107625
    },
    {
      "epoch": 11.840484048404841,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012807785853428618,
      "loss": 0.2335,
      "num_input_tokens_seen": 22714720,
      "step": 107630
    },
    {
      "epoch": 11.84103410341034,
      "grad_norm": 0.00077056884765625,
      "learning_rate": 0.012806361284908698,
      "loss": 0.2303,
      "num_input_tokens_seen": 22715712,
      "step": 107635
    },
    {
      "epoch": 11.841584158415841,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012804936736606494,
      "loss": 0.2319,
      "num_input_tokens_seen": 22716704,
      "step": 107640
    },
    {
      "epoch": 11.842134213421343,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.012803512208535131,
      "loss": 0.2335,
      "num_input_tokens_seen": 22717792,
      "step": 107645
    },
    {
      "epoch": 11.842684268426842,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012802087700707742,
      "loss": 0.2314,
      "num_input_tokens_seen": 22718848,
      "step": 107650
    },
    {
      "epoch": 11.843234323432343,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.012800663213137456,
      "loss": 0.2303,
      "num_input_tokens_seen": 22719936,
      "step": 107655
    },
    {
      "epoch": 11.843784378437844,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.012799238745837401,
      "loss": 0.2324,
      "num_input_tokens_seen": 22721024,
      "step": 107660
    },
    {
      "epoch": 11.844334433443345,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012797814298820716,
      "loss": 0.2324,
      "num_input_tokens_seen": 22722112,
      "step": 107665
    },
    {
      "epoch": 11.844884488448844,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.012796389872100508,
      "loss": 0.2319,
      "num_input_tokens_seen": 22723072,
      "step": 107670
    },
    {
      "epoch": 11.845434543454346,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01279496546568992,
      "loss": 0.2319,
      "num_input_tokens_seen": 22724128,
      "step": 107675
    },
    {
      "epoch": 11.845984598459847,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012793541079602077,
      "loss": 0.2303,
      "num_input_tokens_seen": 22725216,
      "step": 107680
    },
    {
      "epoch": 11.846534653465346,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.012792116713850114,
      "loss": 0.2288,
      "num_input_tokens_seen": 22726240,
      "step": 107685
    },
    {
      "epoch": 11.847084708470847,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.012790692368447147,
      "loss": 0.2314,
      "num_input_tokens_seen": 22727328,
      "step": 107690
    },
    {
      "epoch": 11.847634763476348,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.012789268043406304,
      "loss": 0.2314,
      "num_input_tokens_seen": 22728384,
      "step": 107695
    },
    {
      "epoch": 11.848184818481847,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.012787843738740725,
      "loss": 0.2303,
      "num_input_tokens_seen": 22729408,
      "step": 107700
    },
    {
      "epoch": 11.848734873487349,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012786419454463523,
      "loss": 0.2298,
      "num_input_tokens_seen": 22730464,
      "step": 107705
    },
    {
      "epoch": 11.84928492849285,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.012784995190587832,
      "loss": 0.2319,
      "num_input_tokens_seen": 22731456,
      "step": 107710
    },
    {
      "epoch": 11.84983498349835,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01278357094712678,
      "loss": 0.2329,
      "num_input_tokens_seen": 22732480,
      "step": 107715
    },
    {
      "epoch": 11.85038503850385,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012782146724093484,
      "loss": 0.2314,
      "num_input_tokens_seen": 22733504,
      "step": 107720
    },
    {
      "epoch": 11.850935093509351,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012780722521501087,
      "loss": 0.2308,
      "num_input_tokens_seen": 22734592,
      "step": 107725
    },
    {
      "epoch": 11.851485148514852,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012779298339362699,
      "loss": 0.2298,
      "num_input_tokens_seen": 22735648,
      "step": 107730
    },
    {
      "epoch": 11.852035203520352,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012777874177691452,
      "loss": 0.2299,
      "num_input_tokens_seen": 22736736,
      "step": 107735
    },
    {
      "epoch": 11.852585258525853,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012776450036500478,
      "loss": 0.2314,
      "num_input_tokens_seen": 22737856,
      "step": 107740
    },
    {
      "epoch": 11.853135313531354,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.012775025915802892,
      "loss": 0.2314,
      "num_input_tokens_seen": 22738912,
      "step": 107745
    },
    {
      "epoch": 11.853685368536853,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.012773601815611827,
      "loss": 0.2319,
      "num_input_tokens_seen": 22740000,
      "step": 107750
    },
    {
      "epoch": 11.854235423542354,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.012772177735940407,
      "loss": 0.2303,
      "num_input_tokens_seen": 22741088,
      "step": 107755
    },
    {
      "epoch": 11.854785478547855,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012770753676801751,
      "loss": 0.2314,
      "num_input_tokens_seen": 22742112,
      "step": 107760
    },
    {
      "epoch": 11.855335533553355,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012769329638208992,
      "loss": 0.2314,
      "num_input_tokens_seen": 22743232,
      "step": 107765
    },
    {
      "epoch": 11.855885588558856,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.01276790562017525,
      "loss": 0.2319,
      "num_input_tokens_seen": 22744288,
      "step": 107770
    },
    {
      "epoch": 11.856435643564357,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01276648162271365,
      "loss": 0.233,
      "num_input_tokens_seen": 22745248,
      "step": 107775
    },
    {
      "epoch": 11.856985698569858,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.012765057645837323,
      "loss": 0.2314,
      "num_input_tokens_seen": 22746240,
      "step": 107780
    },
    {
      "epoch": 11.857535753575357,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.01276363368955938,
      "loss": 0.2298,
      "num_input_tokens_seen": 22747264,
      "step": 107785
    },
    {
      "epoch": 11.858085808580858,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.012762209753892955,
      "loss": 0.2309,
      "num_input_tokens_seen": 22748352,
      "step": 107790
    },
    {
      "epoch": 11.85863586358636,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012760785838851166,
      "loss": 0.2324,
      "num_input_tokens_seen": 22749408,
      "step": 107795
    },
    {
      "epoch": 11.859185918591859,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012759361944447148,
      "loss": 0.2303,
      "num_input_tokens_seen": 22750432,
      "step": 107800
    },
    {
      "epoch": 11.85973597359736,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.01275793807069401,
      "loss": 0.2324,
      "num_input_tokens_seen": 22751488,
      "step": 107805
    },
    {
      "epoch": 11.86028602860286,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.012756514217604879,
      "loss": 0.232,
      "num_input_tokens_seen": 22752544,
      "step": 107810
    },
    {
      "epoch": 11.86083608360836,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.012755090385192885,
      "loss": 0.2314,
      "num_input_tokens_seen": 22753600,
      "step": 107815
    },
    {
      "epoch": 11.861386138613861,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01275366657347114,
      "loss": 0.2314,
      "num_input_tokens_seen": 22754656,
      "step": 107820
    },
    {
      "epoch": 11.861936193619362,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012752242782452775,
      "loss": 0.2319,
      "num_input_tokens_seen": 22755680,
      "step": 107825
    },
    {
      "epoch": 11.862486248624862,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.012750819012150914,
      "loss": 0.2309,
      "num_input_tokens_seen": 22756704,
      "step": 107830
    },
    {
      "epoch": 11.863036303630363,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.012749395262578666,
      "loss": 0.2319,
      "num_input_tokens_seen": 22757760,
      "step": 107835
    },
    {
      "epoch": 11.863586358635864,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.01274797153374917,
      "loss": 0.2309,
      "num_input_tokens_seen": 22758880,
      "step": 107840
    },
    {
      "epoch": 11.864136413641365,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.012746547825675534,
      "loss": 0.2303,
      "num_input_tokens_seen": 22759968,
      "step": 107845
    },
    {
      "epoch": 11.864686468646864,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.012745124138370888,
      "loss": 0.2293,
      "num_input_tokens_seen": 22760992,
      "step": 107850
    },
    {
      "epoch": 11.865236523652365,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012743700471848356,
      "loss": 0.2314,
      "num_input_tokens_seen": 22762016,
      "step": 107855
    },
    {
      "epoch": 11.865786578657866,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012742276826121045,
      "loss": 0.2293,
      "num_input_tokens_seen": 22763104,
      "step": 107860
    },
    {
      "epoch": 11.866336633663366,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.012740853201202088,
      "loss": 0.2299,
      "num_input_tokens_seen": 22764096,
      "step": 107865
    },
    {
      "epoch": 11.866886688668867,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.012739429597104602,
      "loss": 0.2309,
      "num_input_tokens_seen": 22765184,
      "step": 107870
    },
    {
      "epoch": 11.867436743674368,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.012738006013841714,
      "loss": 0.2309,
      "num_input_tokens_seen": 22766208,
      "step": 107875
    },
    {
      "epoch": 11.867986798679867,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.012736582451426535,
      "loss": 0.2298,
      "num_input_tokens_seen": 22767264,
      "step": 107880
    },
    {
      "epoch": 11.868536853685368,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.012735158909872188,
      "loss": 0.2283,
      "num_input_tokens_seen": 22768320,
      "step": 107885
    },
    {
      "epoch": 11.86908690869087,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.012733735389191802,
      "loss": 0.2329,
      "num_input_tokens_seen": 22769344,
      "step": 107890
    },
    {
      "epoch": 11.869636963696369,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.012732311889398476,
      "loss": 0.2319,
      "num_input_tokens_seen": 22770464,
      "step": 107895
    },
    {
      "epoch": 11.87018701870187,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.012730888410505355,
      "loss": 0.233,
      "num_input_tokens_seen": 22771488,
      "step": 107900
    },
    {
      "epoch": 11.870737073707371,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.012729464952525544,
      "loss": 0.2319,
      "num_input_tokens_seen": 22772544,
      "step": 107905
    },
    {
      "epoch": 11.871287128712872,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.012728041515472162,
      "loss": 0.234,
      "num_input_tokens_seen": 22773600,
      "step": 107910
    },
    {
      "epoch": 11.871837183718371,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.012726618099358336,
      "loss": 0.2314,
      "num_input_tokens_seen": 22774656,
      "step": 107915
    },
    {
      "epoch": 11.872387238723872,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.012725194704197175,
      "loss": 0.2303,
      "num_input_tokens_seen": 22775744,
      "step": 107920
    },
    {
      "epoch": 11.872937293729374,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.012723771330001805,
      "loss": 0.2319,
      "num_input_tokens_seen": 22776800,
      "step": 107925
    },
    {
      "epoch": 11.873487348734873,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.012722347976785347,
      "loss": 0.2309,
      "num_input_tokens_seen": 22777856,
      "step": 107930
    },
    {
      "epoch": 11.874037403740374,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012720924644560908,
      "loss": 0.2314,
      "num_input_tokens_seen": 22778848,
      "step": 107935
    },
    {
      "epoch": 11.874587458745875,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.012719501333341614,
      "loss": 0.2335,
      "num_input_tokens_seen": 22779904,
      "step": 107940
    },
    {
      "epoch": 11.875137513751374,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012718078043140587,
      "loss": 0.2319,
      "num_input_tokens_seen": 22780928,
      "step": 107945
    },
    {
      "epoch": 11.875687568756875,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.012716654773970933,
      "loss": 0.2314,
      "num_input_tokens_seen": 22781984,
      "step": 107950
    },
    {
      "epoch": 11.876237623762377,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.012715231525845782,
      "loss": 0.2293,
      "num_input_tokens_seen": 22783072,
      "step": 107955
    },
    {
      "epoch": 11.876787678767876,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.01271380829877824,
      "loss": 0.233,
      "num_input_tokens_seen": 22784096,
      "step": 107960
    },
    {
      "epoch": 11.877337733773377,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012712385092781433,
      "loss": 0.234,
      "num_input_tokens_seen": 22785216,
      "step": 107965
    },
    {
      "epoch": 11.877887788778878,
      "grad_norm": 0.0004329681396484375,
      "learning_rate": 0.012710961907868478,
      "loss": 0.2303,
      "num_input_tokens_seen": 22786272,
      "step": 107970
    },
    {
      "epoch": 11.87843784378438,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.012709538744052482,
      "loss": 0.2319,
      "num_input_tokens_seen": 22787360,
      "step": 107975
    },
    {
      "epoch": 11.878987898789878,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.012708115601346572,
      "loss": 0.2324,
      "num_input_tokens_seen": 22788416,
      "step": 107980
    },
    {
      "epoch": 11.87953795379538,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012706692479763857,
      "loss": 0.2314,
      "num_input_tokens_seen": 22789440,
      "step": 107985
    },
    {
      "epoch": 11.88008800880088,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012705269379317465,
      "loss": 0.2324,
      "num_input_tokens_seen": 22790528,
      "step": 107990
    },
    {
      "epoch": 11.88063806380638,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012703846300020501,
      "loss": 0.2309,
      "num_input_tokens_seen": 22791584,
      "step": 107995
    },
    {
      "epoch": 11.881188118811881,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.01270242324188608,
      "loss": 0.2313,
      "num_input_tokens_seen": 22792608,
      "step": 108000
    },
    {
      "epoch": 11.881738173817382,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.012701000204927326,
      "loss": 0.2314,
      "num_input_tokens_seen": 22793696,
      "step": 108005
    },
    {
      "epoch": 11.882288228822881,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.012699577189157346,
      "loss": 0.2335,
      "num_input_tokens_seen": 22794720,
      "step": 108010
    },
    {
      "epoch": 11.882838283828383,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.01269815419458926,
      "loss": 0.2308,
      "num_input_tokens_seen": 22795776,
      "step": 108015
    },
    {
      "epoch": 11.883388338833884,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.012696731221236185,
      "loss": 0.2309,
      "num_input_tokens_seen": 22796896,
      "step": 108020
    },
    {
      "epoch": 11.883938393839383,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012695308269111229,
      "loss": 0.2308,
      "num_input_tokens_seen": 22797920,
      "step": 108025
    },
    {
      "epoch": 11.884488448844884,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.012693885338227517,
      "loss": 0.2309,
      "num_input_tokens_seen": 22799040,
      "step": 108030
    },
    {
      "epoch": 11.885038503850385,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01269246242859815,
      "loss": 0.2304,
      "num_input_tokens_seen": 22800128,
      "step": 108035
    },
    {
      "epoch": 11.885588558855886,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012691039540236253,
      "loss": 0.2298,
      "num_input_tokens_seen": 22801216,
      "step": 108040
    },
    {
      "epoch": 11.886138613861386,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.01268961667315494,
      "loss": 0.2324,
      "num_input_tokens_seen": 22802336,
      "step": 108045
    },
    {
      "epoch": 11.886688668866887,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.012688193827367316,
      "loss": 0.2309,
      "num_input_tokens_seen": 22803424,
      "step": 108050
    },
    {
      "epoch": 11.887238723872388,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012686771002886503,
      "loss": 0.2314,
      "num_input_tokens_seen": 22804512,
      "step": 108055
    },
    {
      "epoch": 11.887788778877887,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.012685348199725607,
      "loss": 0.2309,
      "num_input_tokens_seen": 22805632,
      "step": 108060
    },
    {
      "epoch": 11.888338833883388,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.012683925417897753,
      "loss": 0.2303,
      "num_input_tokens_seen": 22806720,
      "step": 108065
    },
    {
      "epoch": 11.88888888888889,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.012682502657416045,
      "loss": 0.2303,
      "num_input_tokens_seen": 22807776,
      "step": 108070
    },
    {
      "epoch": 11.88943894389439,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.012681079918293595,
      "loss": 0.234,
      "num_input_tokens_seen": 22808864,
      "step": 108075
    },
    {
      "epoch": 11.88998899889989,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.01267965720054352,
      "loss": 0.2314,
      "num_input_tokens_seen": 22809952,
      "step": 108080
    },
    {
      "epoch": 11.89053905390539,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01267823450417893,
      "loss": 0.2319,
      "num_input_tokens_seen": 22811072,
      "step": 108085
    },
    {
      "epoch": 11.891089108910892,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.012676811829212943,
      "loss": 0.2319,
      "num_input_tokens_seen": 22812160,
      "step": 108090
    },
    {
      "epoch": 11.891639163916391,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012675389175658663,
      "loss": 0.2319,
      "num_input_tokens_seen": 22813280,
      "step": 108095
    },
    {
      "epoch": 11.892189218921892,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.012673966543529204,
      "loss": 0.2304,
      "num_input_tokens_seen": 22814336,
      "step": 108100
    },
    {
      "epoch": 11.892739273927393,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.012672543932837685,
      "loss": 0.2303,
      "num_input_tokens_seen": 22815360,
      "step": 108105
    },
    {
      "epoch": 11.893289328932893,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.012671121343597207,
      "loss": 0.2303,
      "num_input_tokens_seen": 22816480,
      "step": 108110
    },
    {
      "epoch": 11.893839383938394,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.012669698775820887,
      "loss": 0.2324,
      "num_input_tokens_seen": 22817536,
      "step": 108115
    },
    {
      "epoch": 11.894389438943895,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.01266827622952184,
      "loss": 0.2303,
      "num_input_tokens_seen": 22818656,
      "step": 108120
    },
    {
      "epoch": 11.894939493949394,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.012666853704713163,
      "loss": 0.2319,
      "num_input_tokens_seen": 22819712,
      "step": 108125
    },
    {
      "epoch": 11.895489548954895,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012665431201407982,
      "loss": 0.2308,
      "num_input_tokens_seen": 22820704,
      "step": 108130
    },
    {
      "epoch": 11.896039603960396,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.012664008719619403,
      "loss": 0.2324,
      "num_input_tokens_seen": 22821760,
      "step": 108135
    },
    {
      "epoch": 11.896589658965897,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.012662586259360526,
      "loss": 0.2319,
      "num_input_tokens_seen": 22822784,
      "step": 108140
    },
    {
      "epoch": 11.897139713971397,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.012661163820644479,
      "loss": 0.2324,
      "num_input_tokens_seen": 22823776,
      "step": 108145
    },
    {
      "epoch": 11.897689768976898,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012659741403484356,
      "loss": 0.2303,
      "num_input_tokens_seen": 22824800,
      "step": 108150
    },
    {
      "epoch": 11.898239823982399,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012658319007893278,
      "loss": 0.2308,
      "num_input_tokens_seen": 22825856,
      "step": 108155
    },
    {
      "epoch": 11.898789878987898,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012656896633884352,
      "loss": 0.2329,
      "num_input_tokens_seen": 22826816,
      "step": 108160
    },
    {
      "epoch": 11.8993399339934,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01265547428147068,
      "loss": 0.2319,
      "num_input_tokens_seen": 22827904,
      "step": 108165
    },
    {
      "epoch": 11.8998899889989,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01265405195066538,
      "loss": 0.2298,
      "num_input_tokens_seen": 22829024,
      "step": 108170
    },
    {
      "epoch": 11.9004400440044,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012652629641481552,
      "loss": 0.2319,
      "num_input_tokens_seen": 22830048,
      "step": 108175
    },
    {
      "epoch": 11.900990099009901,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.012651207353932319,
      "loss": 0.2319,
      "num_input_tokens_seen": 22831104,
      "step": 108180
    },
    {
      "epoch": 11.901540154015402,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.012649785088030776,
      "loss": 0.2324,
      "num_input_tokens_seen": 22832128,
      "step": 108185
    },
    {
      "epoch": 11.902090209020901,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012648362843790032,
      "loss": 0.2314,
      "num_input_tokens_seen": 22833184,
      "step": 108190
    },
    {
      "epoch": 11.902640264026402,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.012646940621223209,
      "loss": 0.2309,
      "num_input_tokens_seen": 22834208,
      "step": 108195
    },
    {
      "epoch": 11.903190319031903,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.012645518420343393,
      "loss": 0.2314,
      "num_input_tokens_seen": 22835360,
      "step": 108200
    },
    {
      "epoch": 11.903740374037405,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.012644096241163715,
      "loss": 0.2303,
      "num_input_tokens_seen": 22836384,
      "step": 108205
    },
    {
      "epoch": 11.904290429042904,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.012642674083697267,
      "loss": 0.2329,
      "num_input_tokens_seen": 22837376,
      "step": 108210
    },
    {
      "epoch": 11.904840484048405,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.012641251947957158,
      "loss": 0.2314,
      "num_input_tokens_seen": 22838400,
      "step": 108215
    },
    {
      "epoch": 11.905390539053906,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.012639829833956505,
      "loss": 0.233,
      "num_input_tokens_seen": 22839552,
      "step": 108220
    },
    {
      "epoch": 11.905940594059405,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0126384077417084,
      "loss": 0.2293,
      "num_input_tokens_seen": 22840608,
      "step": 108225
    },
    {
      "epoch": 11.906490649064907,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012636985671225962,
      "loss": 0.2298,
      "num_input_tokens_seen": 22841696,
      "step": 108230
    },
    {
      "epoch": 11.907040704070408,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.012635563622522297,
      "loss": 0.2314,
      "num_input_tokens_seen": 22842752,
      "step": 108235
    },
    {
      "epoch": 11.907590759075907,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.0126341415956105,
      "loss": 0.2314,
      "num_input_tokens_seen": 22843808,
      "step": 108240
    },
    {
      "epoch": 11.908140814081408,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.01263271959050369,
      "loss": 0.2293,
      "num_input_tokens_seen": 22844864,
      "step": 108245
    },
    {
      "epoch": 11.908690869086909,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.01263129760721496,
      "loss": 0.2324,
      "num_input_tokens_seen": 22845888,
      "step": 108250
    },
    {
      "epoch": 11.909240924092408,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.012629875645757434,
      "loss": 0.2308,
      "num_input_tokens_seen": 22846944,
      "step": 108255
    },
    {
      "epoch": 11.90979097909791,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.012628453706144202,
      "loss": 0.2309,
      "num_input_tokens_seen": 22848000,
      "step": 108260
    },
    {
      "epoch": 11.91034103410341,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01262703178838837,
      "loss": 0.2309,
      "num_input_tokens_seen": 22849056,
      "step": 108265
    },
    {
      "epoch": 11.910891089108912,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012625609892503052,
      "loss": 0.2303,
      "num_input_tokens_seen": 22850176,
      "step": 108270
    },
    {
      "epoch": 11.911441144114411,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.012624188018501346,
      "loss": 0.2313,
      "num_input_tokens_seen": 22851136,
      "step": 108275
    },
    {
      "epoch": 11.911991199119912,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012622766166396363,
      "loss": 0.2313,
      "num_input_tokens_seen": 22852192,
      "step": 108280
    },
    {
      "epoch": 11.912541254125413,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.012621344336201202,
      "loss": 0.2319,
      "num_input_tokens_seen": 22853280,
      "step": 108285
    },
    {
      "epoch": 11.913091309130913,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012619922527928965,
      "loss": 0.2303,
      "num_input_tokens_seen": 22854304,
      "step": 108290
    },
    {
      "epoch": 11.913641364136414,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012618500741592766,
      "loss": 0.2303,
      "num_input_tokens_seen": 22855296,
      "step": 108295
    },
    {
      "epoch": 11.914191419141915,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.012617078977205697,
      "loss": 0.2313,
      "num_input_tokens_seen": 22856320,
      "step": 108300
    },
    {
      "epoch": 11.914741474147414,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01261565723478087,
      "loss": 0.2309,
      "num_input_tokens_seen": 22857376,
      "step": 108305
    },
    {
      "epoch": 11.915291529152915,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012614235514331391,
      "loss": 0.2314,
      "num_input_tokens_seen": 22858464,
      "step": 108310
    },
    {
      "epoch": 11.915841584158416,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.01261281381587035,
      "loss": 0.2303,
      "num_input_tokens_seen": 22859520,
      "step": 108315
    },
    {
      "epoch": 11.916391639163916,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012611392139410866,
      "loss": 0.2308,
      "num_input_tokens_seen": 22860544,
      "step": 108320
    },
    {
      "epoch": 11.916941694169417,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.012609970484966029,
      "loss": 0.2319,
      "num_input_tokens_seen": 22861568,
      "step": 108325
    },
    {
      "epoch": 11.917491749174918,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012608548852548947,
      "loss": 0.2324,
      "num_input_tokens_seen": 22862592,
      "step": 108330
    },
    {
      "epoch": 11.918041804180419,
      "grad_norm": 0.000782012939453125,
      "learning_rate": 0.01260712724217273,
      "loss": 0.2324,
      "num_input_tokens_seen": 22863648,
      "step": 108335
    },
    {
      "epoch": 11.918591859185918,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.012605705653850463,
      "loss": 0.2313,
      "num_input_tokens_seen": 22864768,
      "step": 108340
    },
    {
      "epoch": 11.91914191419142,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.012604284087595265,
      "loss": 0.2319,
      "num_input_tokens_seen": 22865856,
      "step": 108345
    },
    {
      "epoch": 11.91969196919692,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.012602862543420232,
      "loss": 0.2314,
      "num_input_tokens_seen": 22866880,
      "step": 108350
    },
    {
      "epoch": 11.92024202420242,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.012601441021338458,
      "loss": 0.2309,
      "num_input_tokens_seen": 22867872,
      "step": 108355
    },
    {
      "epoch": 11.92079207920792,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.012600019521363057,
      "loss": 0.2314,
      "num_input_tokens_seen": 22868896,
      "step": 108360
    },
    {
      "epoch": 11.921342134213422,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.01259859804350712,
      "loss": 0.2303,
      "num_input_tokens_seen": 22869984,
      "step": 108365
    },
    {
      "epoch": 11.921892189218921,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.012597176587783759,
      "loss": 0.2319,
      "num_input_tokens_seen": 22871040,
      "step": 108370
    },
    {
      "epoch": 11.922442244224422,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.012595755154206065,
      "loss": 0.2303,
      "num_input_tokens_seen": 22872128,
      "step": 108375
    },
    {
      "epoch": 11.922992299229923,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.012594333742787138,
      "loss": 0.2303,
      "num_input_tokens_seen": 22873184,
      "step": 108380
    },
    {
      "epoch": 11.923542354235423,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012592912353540086,
      "loss": 0.2334,
      "num_input_tokens_seen": 22874240,
      "step": 108385
    },
    {
      "epoch": 11.924092409240924,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.012591490986478002,
      "loss": 0.2314,
      "num_input_tokens_seen": 22875392,
      "step": 108390
    },
    {
      "epoch": 11.924642464246425,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012590069641613997,
      "loss": 0.2308,
      "num_input_tokens_seen": 22876384,
      "step": 108395
    },
    {
      "epoch": 11.925192519251926,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.012588648318961162,
      "loss": 0.2309,
      "num_input_tokens_seen": 22877440,
      "step": 108400
    },
    {
      "epoch": 11.925742574257425,
      "grad_norm": 0.005859375,
      "learning_rate": 0.012587227018532593,
      "loss": 0.2319,
      "num_input_tokens_seen": 22878528,
      "step": 108405
    },
    {
      "epoch": 11.926292629262926,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.012585805740341403,
      "loss": 0.2303,
      "num_input_tokens_seen": 22879584,
      "step": 108410
    },
    {
      "epoch": 11.926842684268427,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.012584384484400675,
      "loss": 0.2314,
      "num_input_tokens_seen": 22880640,
      "step": 108415
    },
    {
      "epoch": 11.927392739273927,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.012582963250723521,
      "loss": 0.2314,
      "num_input_tokens_seen": 22881696,
      "step": 108420
    },
    {
      "epoch": 11.927942794279428,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.012581542039323037,
      "loss": 0.2309,
      "num_input_tokens_seen": 22882848,
      "step": 108425
    },
    {
      "epoch": 11.928492849284929,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.012580120850212314,
      "loss": 0.2309,
      "num_input_tokens_seen": 22884032,
      "step": 108430
    },
    {
      "epoch": 11.929042904290428,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.012578699683404459,
      "loss": 0.2308,
      "num_input_tokens_seen": 22885088,
      "step": 108435
    },
    {
      "epoch": 11.92959295929593,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.012577278538912561,
      "loss": 0.2298,
      "num_input_tokens_seen": 22886080,
      "step": 108440
    },
    {
      "epoch": 11.93014301430143,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01257585741674973,
      "loss": 0.2319,
      "num_input_tokens_seen": 22887136,
      "step": 108445
    },
    {
      "epoch": 11.930693069306932,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01257443631692906,
      "loss": 0.2308,
      "num_input_tokens_seen": 22888160,
      "step": 108450
    },
    {
      "epoch": 11.93124312431243,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012573015239463642,
      "loss": 0.2308,
      "num_input_tokens_seen": 22889152,
      "step": 108455
    },
    {
      "epoch": 11.931793179317932,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01257159418436658,
      "loss": 0.2314,
      "num_input_tokens_seen": 22890240,
      "step": 108460
    },
    {
      "epoch": 11.932343234323433,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.012570173151650967,
      "loss": 0.2319,
      "num_input_tokens_seen": 22891328,
      "step": 108465
    },
    {
      "epoch": 11.932893289328932,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012568752141329907,
      "loss": 0.2293,
      "num_input_tokens_seen": 22892384,
      "step": 108470
    },
    {
      "epoch": 11.933443344334433,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012567331153416489,
      "loss": 0.2314,
      "num_input_tokens_seen": 22893472,
      "step": 108475
    },
    {
      "epoch": 11.933993399339935,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.01256591018792381,
      "loss": 0.2298,
      "num_input_tokens_seen": 22894496,
      "step": 108480
    },
    {
      "epoch": 11.934543454345434,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.012564489244864975,
      "loss": 0.2303,
      "num_input_tokens_seen": 22895584,
      "step": 108485
    },
    {
      "epoch": 11.935093509350935,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.012563068324253067,
      "loss": 0.2324,
      "num_input_tokens_seen": 22896608,
      "step": 108490
    },
    {
      "epoch": 11.935643564356436,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.012561647426101193,
      "loss": 0.2304,
      "num_input_tokens_seen": 22897664,
      "step": 108495
    },
    {
      "epoch": 11.936193619361937,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.012560226550422446,
      "loss": 0.2308,
      "num_input_tokens_seen": 22898688,
      "step": 108500
    },
    {
      "epoch": 11.936743674367436,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012558805697229914,
      "loss": 0.2309,
      "num_input_tokens_seen": 22899712,
      "step": 108505
    },
    {
      "epoch": 11.937293729372938,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.012557384866536705,
      "loss": 0.2329,
      "num_input_tokens_seen": 22900800,
      "step": 108510
    },
    {
      "epoch": 11.937843784378439,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.012555964058355906,
      "loss": 0.2319,
      "num_input_tokens_seen": 22901792,
      "step": 108515
    },
    {
      "epoch": 11.938393839383938,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.01255454327270061,
      "loss": 0.2319,
      "num_input_tokens_seen": 22902816,
      "step": 108520
    },
    {
      "epoch": 11.938943894389439,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01255312250958392,
      "loss": 0.2324,
      "num_input_tokens_seen": 22903872,
      "step": 108525
    },
    {
      "epoch": 11.93949394939494,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012551701769018922,
      "loss": 0.2308,
      "num_input_tokens_seen": 22904896,
      "step": 108530
    },
    {
      "epoch": 11.94004400440044,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.012550281051018715,
      "loss": 0.2303,
      "num_input_tokens_seen": 22905952,
      "step": 108535
    },
    {
      "epoch": 11.94059405940594,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.012548860355596396,
      "loss": 0.2345,
      "num_input_tokens_seen": 22907008,
      "step": 108540
    },
    {
      "epoch": 11.941144114411442,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.01254743968276505,
      "loss": 0.234,
      "num_input_tokens_seen": 22908064,
      "step": 108545
    },
    {
      "epoch": 11.941694169416941,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.012546019032537777,
      "loss": 0.2324,
      "num_input_tokens_seen": 22909088,
      "step": 108550
    },
    {
      "epoch": 11.942244224422442,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012544598404927667,
      "loss": 0.233,
      "num_input_tokens_seen": 22910144,
      "step": 108555
    },
    {
      "epoch": 11.942794279427943,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.012543177799947817,
      "loss": 0.2324,
      "num_input_tokens_seen": 22911136,
      "step": 108560
    },
    {
      "epoch": 11.943344334433444,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.012541757217611322,
      "loss": 0.2298,
      "num_input_tokens_seen": 22912160,
      "step": 108565
    },
    {
      "epoch": 11.943894389438944,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012540336657931264,
      "loss": 0.2308,
      "num_input_tokens_seen": 22913248,
      "step": 108570
    },
    {
      "epoch": 11.944444444444445,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.012538916120920748,
      "loss": 0.2298,
      "num_input_tokens_seen": 22914272,
      "step": 108575
    },
    {
      "epoch": 11.944994499449946,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012537495606592857,
      "loss": 0.2309,
      "num_input_tokens_seen": 22915328,
      "step": 108580
    },
    {
      "epoch": 11.945544554455445,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.012536075114960693,
      "loss": 0.2329,
      "num_input_tokens_seen": 22916384,
      "step": 108585
    },
    {
      "epoch": 11.946094609460946,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.012534654646037341,
      "loss": 0.2314,
      "num_input_tokens_seen": 22917440,
      "step": 108590
    },
    {
      "epoch": 11.946644664466447,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.01253323419983589,
      "loss": 0.2319,
      "num_input_tokens_seen": 22918528,
      "step": 108595
    },
    {
      "epoch": 11.947194719471947,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01253181377636944,
      "loss": 0.2309,
      "num_input_tokens_seen": 22919552,
      "step": 108600
    },
    {
      "epoch": 11.947744774477448,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.012530393375651075,
      "loss": 0.2308,
      "num_input_tokens_seen": 22920640,
      "step": 108605
    },
    {
      "epoch": 11.948294829482949,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01252897299769389,
      "loss": 0.2309,
      "num_input_tokens_seen": 22921760,
      "step": 108610
    },
    {
      "epoch": 11.948844884488448,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.01252755264251098,
      "loss": 0.2288,
      "num_input_tokens_seen": 22922848,
      "step": 108615
    },
    {
      "epoch": 11.94939493949395,
      "grad_norm": 0.000797271728515625,
      "learning_rate": 0.012526132310115422,
      "loss": 0.2304,
      "num_input_tokens_seen": 22923840,
      "step": 108620
    },
    {
      "epoch": 11.94994499449945,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.012524712000520322,
      "loss": 0.2304,
      "num_input_tokens_seen": 22924864,
      "step": 108625
    },
    {
      "epoch": 11.950495049504951,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01252329171373876,
      "loss": 0.2298,
      "num_input_tokens_seen": 22925856,
      "step": 108630
    },
    {
      "epoch": 11.95104510451045,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01252187144978383,
      "loss": 0.2309,
      "num_input_tokens_seen": 22926944,
      "step": 108635
    },
    {
      "epoch": 11.951595159515952,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.012520451208668628,
      "loss": 0.2294,
      "num_input_tokens_seen": 22928032,
      "step": 108640
    },
    {
      "epoch": 11.952145214521453,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012519030990406231,
      "loss": 0.2345,
      "num_input_tokens_seen": 22929120,
      "step": 108645
    },
    {
      "epoch": 11.952695269526952,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.012517610795009737,
      "loss": 0.232,
      "num_input_tokens_seen": 22930208,
      "step": 108650
    },
    {
      "epoch": 11.953245324532453,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01251619062249223,
      "loss": 0.2335,
      "num_input_tokens_seen": 22931296,
      "step": 108655
    },
    {
      "epoch": 11.953795379537954,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012514770472866808,
      "loss": 0.2309,
      "num_input_tokens_seen": 22932384,
      "step": 108660
    },
    {
      "epoch": 11.954345434543454,
      "grad_norm": 0.000659942626953125,
      "learning_rate": 0.01251335034614655,
      "loss": 0.2278,
      "num_input_tokens_seen": 22933408,
      "step": 108665
    },
    {
      "epoch": 11.954895489548955,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.012511930242344548,
      "loss": 0.2309,
      "num_input_tokens_seen": 22934496,
      "step": 108670
    },
    {
      "epoch": 11.955445544554456,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.012510510161473896,
      "loss": 0.2288,
      "num_input_tokens_seen": 22935584,
      "step": 108675
    },
    {
      "epoch": 11.955995599559955,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.01250909010354767,
      "loss": 0.233,
      "num_input_tokens_seen": 22936640,
      "step": 108680
    },
    {
      "epoch": 11.956545654565456,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.012507670068578971,
      "loss": 0.2309,
      "num_input_tokens_seen": 22937696,
      "step": 108685
    },
    {
      "epoch": 11.957095709570957,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.012506250056580879,
      "loss": 0.232,
      "num_input_tokens_seen": 22938752,
      "step": 108690
    },
    {
      "epoch": 11.957645764576458,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.012504830067566483,
      "loss": 0.232,
      "num_input_tokens_seen": 22939808,
      "step": 108695
    },
    {
      "epoch": 11.958195819581958,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.012503410101548875,
      "loss": 0.2325,
      "num_input_tokens_seen": 22940832,
      "step": 108700
    },
    {
      "epoch": 11.958745874587459,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.012501990158541136,
      "loss": 0.2341,
      "num_input_tokens_seen": 22941888,
      "step": 108705
    },
    {
      "epoch": 11.95929592959296,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.012500570238556349,
      "loss": 0.2319,
      "num_input_tokens_seen": 22942944,
      "step": 108710
    },
    {
      "epoch": 11.95984598459846,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.012499150341607616,
      "loss": 0.2294,
      "num_input_tokens_seen": 22943968,
      "step": 108715
    },
    {
      "epoch": 11.96039603960396,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.012497730467708008,
      "loss": 0.234,
      "num_input_tokens_seen": 22945024,
      "step": 108720
    },
    {
      "epoch": 11.960946094609461,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01249631061687062,
      "loss": 0.2299,
      "num_input_tokens_seen": 22946016,
      "step": 108725
    },
    {
      "epoch": 11.96149614961496,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.012494890789108539,
      "loss": 0.2325,
      "num_input_tokens_seen": 22947104,
      "step": 108730
    },
    {
      "epoch": 11.962046204620462,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.012493470984434842,
      "loss": 0.2309,
      "num_input_tokens_seen": 22948096,
      "step": 108735
    },
    {
      "epoch": 11.962596259625963,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.012492051202862623,
      "loss": 0.2314,
      "num_input_tokens_seen": 22949152,
      "step": 108740
    },
    {
      "epoch": 11.963146314631462,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.01249063144440496,
      "loss": 0.2314,
      "num_input_tokens_seen": 22950240,
      "step": 108745
    },
    {
      "epoch": 11.963696369636963,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.012489211709074946,
      "loss": 0.2309,
      "num_input_tokens_seen": 22951264,
      "step": 108750
    },
    {
      "epoch": 11.964246424642464,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.012487791996885668,
      "loss": 0.2319,
      "num_input_tokens_seen": 22952256,
      "step": 108755
    },
    {
      "epoch": 11.964796479647966,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.012486372307850198,
      "loss": 0.2298,
      "num_input_tokens_seen": 22953280,
      "step": 108760
    },
    {
      "epoch": 11.965346534653465,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.012484952641981632,
      "loss": 0.2283,
      "num_input_tokens_seen": 22954336,
      "step": 108765
    },
    {
      "epoch": 11.965896589658966,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.012483532999293046,
      "loss": 0.2277,
      "num_input_tokens_seen": 22955328,
      "step": 108770
    },
    {
      "epoch": 11.966446644664467,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.012482113379797538,
      "loss": 0.2304,
      "num_input_tokens_seen": 22956352,
      "step": 108775
    },
    {
      "epoch": 11.966996699669966,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.012480693783508178,
      "loss": 0.2314,
      "num_input_tokens_seen": 22957408,
      "step": 108780
    },
    {
      "epoch": 11.967546754675467,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01247927421043805,
      "loss": 0.2309,
      "num_input_tokens_seen": 22958432,
      "step": 108785
    },
    {
      "epoch": 11.968096809680969,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01247785466060025,
      "loss": 0.2325,
      "num_input_tokens_seen": 22959488,
      "step": 108790
    },
    {
      "epoch": 11.968646864686468,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.012476435134007847,
      "loss": 0.2293,
      "num_input_tokens_seen": 22960512,
      "step": 108795
    },
    {
      "epoch": 11.969196919691969,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.012475015630673932,
      "loss": 0.2309,
      "num_input_tokens_seen": 22961664,
      "step": 108800
    },
    {
      "epoch": 11.96974697469747,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.01247359615061159,
      "loss": 0.2319,
      "num_input_tokens_seen": 22962720,
      "step": 108805
    },
    {
      "epoch": 11.97029702970297,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.012472176693833895,
      "loss": 0.2319,
      "num_input_tokens_seen": 22963840,
      "step": 108810
    },
    {
      "epoch": 11.97084708470847,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01247075726035394,
      "loss": 0.2324,
      "num_input_tokens_seen": 22964960,
      "step": 108815
    },
    {
      "epoch": 11.971397139713972,
      "grad_norm": 0.000812530517578125,
      "learning_rate": 0.012469337850184795,
      "loss": 0.2345,
      "num_input_tokens_seen": 22966016,
      "step": 108820
    },
    {
      "epoch": 11.971947194719473,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.012467918463339552,
      "loss": 0.2293,
      "num_input_tokens_seen": 22967104,
      "step": 108825
    },
    {
      "epoch": 11.972497249724972,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012466499099831295,
      "loss": 0.2304,
      "num_input_tokens_seen": 22968096,
      "step": 108830
    },
    {
      "epoch": 11.973047304730473,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.01246507975967309,
      "loss": 0.2304,
      "num_input_tokens_seen": 22969184,
      "step": 108835
    },
    {
      "epoch": 11.973597359735974,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.012463660442878034,
      "loss": 0.2335,
      "num_input_tokens_seen": 22970240,
      "step": 108840
    },
    {
      "epoch": 11.974147414741473,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.012462241149459199,
      "loss": 0.2314,
      "num_input_tokens_seen": 22971264,
      "step": 108845
    },
    {
      "epoch": 11.974697469746975,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.012460821879429675,
      "loss": 0.2324,
      "num_input_tokens_seen": 22972384,
      "step": 108850
    },
    {
      "epoch": 11.975247524752476,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.012459402632802536,
      "loss": 0.233,
      "num_input_tokens_seen": 22973408,
      "step": 108855
    },
    {
      "epoch": 11.975797579757975,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.01245798340959086,
      "loss": 0.2304,
      "num_input_tokens_seen": 22974496,
      "step": 108860
    },
    {
      "epoch": 11.976347634763476,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.012456564209807734,
      "loss": 0.2325,
      "num_input_tokens_seen": 22975584,
      "step": 108865
    },
    {
      "epoch": 11.976897689768977,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.012455145033466234,
      "loss": 0.2319,
      "num_input_tokens_seen": 22976640,
      "step": 108870
    },
    {
      "epoch": 11.977447744774478,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.012453725880579445,
      "loss": 0.2325,
      "num_input_tokens_seen": 22977728,
      "step": 108875
    },
    {
      "epoch": 11.977997799779978,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.01245230675116044,
      "loss": 0.2293,
      "num_input_tokens_seen": 22978848,
      "step": 108880
    },
    {
      "epoch": 11.978547854785479,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012450887645222298,
      "loss": 0.2298,
      "num_input_tokens_seen": 22979904,
      "step": 108885
    },
    {
      "epoch": 11.97909790979098,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.012449468562778109,
      "loss": 0.2309,
      "num_input_tokens_seen": 22981024,
      "step": 108890
    },
    {
      "epoch": 11.979647964796479,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.012448049503840939,
      "loss": 0.2324,
      "num_input_tokens_seen": 22982080,
      "step": 108895
    },
    {
      "epoch": 11.98019801980198,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.012446630468423871,
      "loss": 0.2324,
      "num_input_tokens_seen": 22983200,
      "step": 108900
    },
    {
      "epoch": 11.980748074807481,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.01244521145653999,
      "loss": 0.2335,
      "num_input_tokens_seen": 22984224,
      "step": 108905
    },
    {
      "epoch": 11.98129812981298,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.012443792468202363,
      "loss": 0.2314,
      "num_input_tokens_seen": 22985248,
      "step": 108910
    },
    {
      "epoch": 11.981848184818482,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.012442373503424078,
      "loss": 0.2319,
      "num_input_tokens_seen": 22986368,
      "step": 108915
    },
    {
      "epoch": 11.982398239823983,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.012440954562218212,
      "loss": 0.2335,
      "num_input_tokens_seen": 22987424,
      "step": 108920
    },
    {
      "epoch": 11.982948294829484,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.012439535644597828,
      "loss": 0.2314,
      "num_input_tokens_seen": 22988448,
      "step": 108925
    },
    {
      "epoch": 11.983498349834983,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01243811675057603,
      "loss": 0.2324,
      "num_input_tokens_seen": 22989536,
      "step": 108930
    },
    {
      "epoch": 11.984048404840484,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.012436697880165869,
      "loss": 0.2314,
      "num_input_tokens_seen": 22990592,
      "step": 108935
    },
    {
      "epoch": 11.984598459845985,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01243527903338044,
      "loss": 0.2324,
      "num_input_tokens_seen": 22991616,
      "step": 108940
    },
    {
      "epoch": 11.985148514851485,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.012433860210232815,
      "loss": 0.2273,
      "num_input_tokens_seen": 22992704,
      "step": 108945
    },
    {
      "epoch": 11.985698569856986,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.012432441410736064,
      "loss": 0.2304,
      "num_input_tokens_seen": 22993792,
      "step": 108950
    },
    {
      "epoch": 11.986248624862487,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01243102263490327,
      "loss": 0.2298,
      "num_input_tokens_seen": 22994880,
      "step": 108955
    },
    {
      "epoch": 11.986798679867986,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012429603882747506,
      "loss": 0.2324,
      "num_input_tokens_seen": 22995936,
      "step": 108960
    },
    {
      "epoch": 11.987348734873487,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.012428185154281856,
      "loss": 0.2314,
      "num_input_tokens_seen": 22997024,
      "step": 108965
    },
    {
      "epoch": 11.987898789878988,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.012426766449519384,
      "loss": 0.2308,
      "num_input_tokens_seen": 22998080,
      "step": 108970
    },
    {
      "epoch": 11.988448844884488,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01242534776847317,
      "loss": 0.2314,
      "num_input_tokens_seen": 22999200,
      "step": 108975
    },
    {
      "epoch": 11.988998899889989,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012423929111156296,
      "loss": 0.2324,
      "num_input_tokens_seen": 23000192,
      "step": 108980
    },
    {
      "epoch": 11.98954895489549,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.012422510477581823,
      "loss": 0.2335,
      "num_input_tokens_seen": 23001280,
      "step": 108985
    },
    {
      "epoch": 11.990099009900991,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.012421091867762841,
      "loss": 0.2278,
      "num_input_tokens_seen": 23002400,
      "step": 108990
    },
    {
      "epoch": 11.99064906490649,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012419673281712416,
      "loss": 0.2293,
      "num_input_tokens_seen": 23003456,
      "step": 108995
    },
    {
      "epoch": 11.991199119911991,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.012418254719443621,
      "loss": 0.2314,
      "num_input_tokens_seen": 23004512,
      "step": 109000
    },
    {
      "epoch": 11.991749174917492,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01241683618096954,
      "loss": 0.2314,
      "num_input_tokens_seen": 23005600,
      "step": 109005
    },
    {
      "epoch": 11.992299229922992,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012415417666303235,
      "loss": 0.2267,
      "num_input_tokens_seen": 23006656,
      "step": 109010
    },
    {
      "epoch": 11.992849284928493,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.012413999175457787,
      "loss": 0.2309,
      "num_input_tokens_seen": 23007680,
      "step": 109015
    },
    {
      "epoch": 11.993399339933994,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.012412580708446272,
      "loss": 0.2325,
      "num_input_tokens_seen": 23008736,
      "step": 109020
    },
    {
      "epoch": 11.993949394939493,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.01241116226528175,
      "loss": 0.2299,
      "num_input_tokens_seen": 23009792,
      "step": 109025
    },
    {
      "epoch": 11.994499449944994,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.012409743845977308,
      "loss": 0.2283,
      "num_input_tokens_seen": 23010816,
      "step": 109030
    },
    {
      "epoch": 11.995049504950495,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012408325450546012,
      "loss": 0.2335,
      "num_input_tokens_seen": 23011904,
      "step": 109035
    },
    {
      "epoch": 11.995599559955995,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01240690707900094,
      "loss": 0.2303,
      "num_input_tokens_seen": 23012928,
      "step": 109040
    },
    {
      "epoch": 11.996149614961496,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01240548873135516,
      "loss": 0.2345,
      "num_input_tokens_seen": 23013984,
      "step": 109045
    },
    {
      "epoch": 11.996699669966997,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.012404070407621742,
      "loss": 0.2324,
      "num_input_tokens_seen": 23015008,
      "step": 109050
    },
    {
      "epoch": 11.997249724972498,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012402652107813765,
      "loss": 0.234,
      "num_input_tokens_seen": 23016032,
      "step": 109055
    },
    {
      "epoch": 11.997799779977997,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.012401233831944294,
      "loss": 0.2314,
      "num_input_tokens_seen": 23017120,
      "step": 109060
    },
    {
      "epoch": 11.998349834983498,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01239981558002641,
      "loss": 0.2335,
      "num_input_tokens_seen": 23018144,
      "step": 109065
    },
    {
      "epoch": 11.998899889989,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.012398397352073175,
      "loss": 0.2298,
      "num_input_tokens_seen": 23019232,
      "step": 109070
    },
    {
      "epoch": 11.999449944994499,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.01239697914809766,
      "loss": 0.2319,
      "num_input_tokens_seen": 23020320,
      "step": 109075
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.012395560968112946,
      "loss": 0.2324,
      "num_input_tokens_seen": 23021312,
      "step": 109080
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.23137040436267853,
      "eval_runtime": 60.5761,
      "eval_samples_per_second": 66.693,
      "eval_steps_per_second": 16.673,
      "num_input_tokens_seen": 23021312,
      "step": 109080
    },
    {
      "epoch": 12.000550055005501,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.012394142812132091,
      "loss": 0.2309,
      "num_input_tokens_seen": 23022368,
      "step": 109085
    },
    {
      "epoch": 12.001100110011,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01239272468016817,
      "loss": 0.2288,
      "num_input_tokens_seen": 23023488,
      "step": 109090
    },
    {
      "epoch": 12.001650165016502,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.012391306572234263,
      "loss": 0.2299,
      "num_input_tokens_seen": 23024448,
      "step": 109095
    },
    {
      "epoch": 12.002200220022003,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.012389888488343418,
      "loss": 0.2335,
      "num_input_tokens_seen": 23025568,
      "step": 109100
    },
    {
      "epoch": 12.002750275027502,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.012388470428508732,
      "loss": 0.2299,
      "num_input_tokens_seen": 23026688,
      "step": 109105
    },
    {
      "epoch": 12.003300330033003,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012387052392743255,
      "loss": 0.2314,
      "num_input_tokens_seen": 23027744,
      "step": 109110
    },
    {
      "epoch": 12.003850385038504,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.01238563438106006,
      "loss": 0.2314,
      "num_input_tokens_seen": 23028800,
      "step": 109115
    },
    {
      "epoch": 12.004400440044005,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.012384216393472224,
      "loss": 0.2319,
      "num_input_tokens_seen": 23029824,
      "step": 109120
    },
    {
      "epoch": 12.004950495049505,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012382798429992804,
      "loss": 0.2319,
      "num_input_tokens_seen": 23030880,
      "step": 109125
    },
    {
      "epoch": 12.005500550055006,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.012381380490634879,
      "loss": 0.2346,
      "num_input_tokens_seen": 23031904,
      "step": 109130
    },
    {
      "epoch": 12.006050605060507,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.012379962575411517,
      "loss": 0.2314,
      "num_input_tokens_seen": 23032960,
      "step": 109135
    },
    {
      "epoch": 12.006600660066006,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.012378544684335776,
      "loss": 0.2309,
      "num_input_tokens_seen": 23033952,
      "step": 109140
    },
    {
      "epoch": 12.007150715071507,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.012377126817420734,
      "loss": 0.232,
      "num_input_tokens_seen": 23035008,
      "step": 109145
    },
    {
      "epoch": 12.007700770077008,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01237570897467945,
      "loss": 0.2298,
      "num_input_tokens_seen": 23036032,
      "step": 109150
    },
    {
      "epoch": 12.008250825082508,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.012374291156125006,
      "loss": 0.2324,
      "num_input_tokens_seen": 23037088,
      "step": 109155
    },
    {
      "epoch": 12.008800880088009,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.012372873361770457,
      "loss": 0.234,
      "num_input_tokens_seen": 23038144,
      "step": 109160
    },
    {
      "epoch": 12.00935093509351,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.01237145559162887,
      "loss": 0.2309,
      "num_input_tokens_seen": 23039168,
      "step": 109165
    },
    {
      "epoch": 12.009900990099009,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.01237003784571332,
      "loss": 0.234,
      "num_input_tokens_seen": 23040224,
      "step": 109170
    },
    {
      "epoch": 12.01045104510451,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.012368620124036864,
      "loss": 0.2309,
      "num_input_tokens_seen": 23041248,
      "step": 109175
    },
    {
      "epoch": 12.011001100110011,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012367202426612581,
      "loss": 0.233,
      "num_input_tokens_seen": 23042304,
      "step": 109180
    },
    {
      "epoch": 12.011551155115512,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.012365784753453526,
      "loss": 0.2314,
      "num_input_tokens_seen": 23043296,
      "step": 109185
    },
    {
      "epoch": 12.012101210121012,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.012364367104572766,
      "loss": 0.2309,
      "num_input_tokens_seen": 23044384,
      "step": 109190
    },
    {
      "epoch": 12.012651265126513,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.012362949479983378,
      "loss": 0.233,
      "num_input_tokens_seen": 23045472,
      "step": 109195
    },
    {
      "epoch": 12.013201320132014,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.012361531879698412,
      "loss": 0.2329,
      "num_input_tokens_seen": 23046496,
      "step": 109200
    },
    {
      "epoch": 12.013751375137513,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.012360114303730943,
      "loss": 0.2335,
      "num_input_tokens_seen": 23047648,
      "step": 109205
    },
    {
      "epoch": 12.014301430143014,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012358696752094037,
      "loss": 0.2303,
      "num_input_tokens_seen": 23048672,
      "step": 109210
    },
    {
      "epoch": 12.014851485148515,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.012357279224800748,
      "loss": 0.2319,
      "num_input_tokens_seen": 23049664,
      "step": 109215
    },
    {
      "epoch": 12.015401540154015,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012355861721864153,
      "loss": 0.2298,
      "num_input_tokens_seen": 23050720,
      "step": 109220
    },
    {
      "epoch": 12.015951595159516,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.012354444243297309,
      "loss": 0.2324,
      "num_input_tokens_seen": 23051744,
      "step": 109225
    },
    {
      "epoch": 12.016501650165017,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.012353026789113286,
      "loss": 0.2309,
      "num_input_tokens_seen": 23052736,
      "step": 109230
    },
    {
      "epoch": 12.017051705170518,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.012351609359325147,
      "loss": 0.233,
      "num_input_tokens_seen": 23053792,
      "step": 109235
    },
    {
      "epoch": 12.017601760176017,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.012350191953945946,
      "loss": 0.2314,
      "num_input_tokens_seen": 23054848,
      "step": 109240
    },
    {
      "epoch": 12.018151815181518,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01234877457298876,
      "loss": 0.2324,
      "num_input_tokens_seen": 23055904,
      "step": 109245
    },
    {
      "epoch": 12.01870187018702,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.012347357216466643,
      "loss": 0.2304,
      "num_input_tokens_seen": 23056928,
      "step": 109250
    },
    {
      "epoch": 12.019251925192519,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.012345939884392669,
      "loss": 0.2303,
      "num_input_tokens_seen": 23057984,
      "step": 109255
    },
    {
      "epoch": 12.01980198019802,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.012344522576779888,
      "loss": 0.2288,
      "num_input_tokens_seen": 23059072,
      "step": 109260
    },
    {
      "epoch": 12.020352035203521,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012343105293641366,
      "loss": 0.2303,
      "num_input_tokens_seen": 23060128,
      "step": 109265
    },
    {
      "epoch": 12.02090209020902,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.012341688034990174,
      "loss": 0.2319,
      "num_input_tokens_seen": 23061184,
      "step": 109270
    },
    {
      "epoch": 12.021452145214521,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012340270800839366,
      "loss": 0.2298,
      "num_input_tokens_seen": 23062240,
      "step": 109275
    },
    {
      "epoch": 12.022002200220022,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.012338853591202001,
      "loss": 0.2329,
      "num_input_tokens_seen": 23063296,
      "step": 109280
    },
    {
      "epoch": 12.022552255225522,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012337436406091152,
      "loss": 0.2319,
      "num_input_tokens_seen": 23064352,
      "step": 109285
    },
    {
      "epoch": 12.023102310231023,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012336019245519865,
      "loss": 0.2319,
      "num_input_tokens_seen": 23065472,
      "step": 109290
    },
    {
      "epoch": 12.023652365236524,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012334602109501219,
      "loss": 0.2309,
      "num_input_tokens_seen": 23066560,
      "step": 109295
    },
    {
      "epoch": 12.024202420242025,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.012333184998048264,
      "loss": 0.2319,
      "num_input_tokens_seen": 23067648,
      "step": 109300
    },
    {
      "epoch": 12.024752475247524,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012331767911174059,
      "loss": 0.2324,
      "num_input_tokens_seen": 23068704,
      "step": 109305
    },
    {
      "epoch": 12.025302530253025,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.012330350848891676,
      "loss": 0.2283,
      "num_input_tokens_seen": 23069760,
      "step": 109310
    },
    {
      "epoch": 12.025852585258527,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.012328933811214162,
      "loss": 0.2304,
      "num_input_tokens_seen": 23070784,
      "step": 109315
    },
    {
      "epoch": 12.026402640264026,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.012327516798154586,
      "loss": 0.2298,
      "num_input_tokens_seen": 23071808,
      "step": 109320
    },
    {
      "epoch": 12.026952695269527,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.01232609980972601,
      "loss": 0.235,
      "num_input_tokens_seen": 23072768,
      "step": 109325
    },
    {
      "epoch": 12.027502750275028,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.01232468284594148,
      "loss": 0.2304,
      "num_input_tokens_seen": 23073888,
      "step": 109330
    },
    {
      "epoch": 12.028052805280527,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.01232326590681407,
      "loss": 0.2329,
      "num_input_tokens_seen": 23074912,
      "step": 109335
    },
    {
      "epoch": 12.028602860286028,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01232184899235683,
      "loss": 0.2293,
      "num_input_tokens_seen": 23076000,
      "step": 109340
    },
    {
      "epoch": 12.02915291529153,
      "grad_norm": 0.0048828125,
      "learning_rate": 0.012320432102582825,
      "loss": 0.2314,
      "num_input_tokens_seen": 23076992,
      "step": 109345
    },
    {
      "epoch": 12.029702970297029,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.012319015237505115,
      "loss": 0.2314,
      "num_input_tokens_seen": 23078048,
      "step": 109350
    },
    {
      "epoch": 12.03025302530253,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.01231759839713675,
      "loss": 0.2325,
      "num_input_tokens_seen": 23079072,
      "step": 109355
    },
    {
      "epoch": 12.030803080308031,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.012316181581490795,
      "loss": 0.2288,
      "num_input_tokens_seen": 23080064,
      "step": 109360
    },
    {
      "epoch": 12.031353135313532,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012314764790580305,
      "loss": 0.2309,
      "num_input_tokens_seen": 23081088,
      "step": 109365
    },
    {
      "epoch": 12.031903190319031,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.012313348024418342,
      "loss": 0.234,
      "num_input_tokens_seen": 23082144,
      "step": 109370
    },
    {
      "epoch": 12.032453245324533,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.012311931283017963,
      "loss": 0.2293,
      "num_input_tokens_seen": 23083200,
      "step": 109375
    },
    {
      "epoch": 12.033003300330034,
      "grad_norm": 0.000614166259765625,
      "learning_rate": 0.012310514566392216,
      "loss": 0.2298,
      "num_input_tokens_seen": 23084224,
      "step": 109380
    },
    {
      "epoch": 12.033553355335533,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.012309097874554175,
      "loss": 0.2319,
      "num_input_tokens_seen": 23085312,
      "step": 109385
    },
    {
      "epoch": 12.034103410341034,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.012307681207516881,
      "loss": 0.234,
      "num_input_tokens_seen": 23086336,
      "step": 109390
    },
    {
      "epoch": 12.034653465346535,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.0123062645652934,
      "loss": 0.2304,
      "num_input_tokens_seen": 23087392,
      "step": 109395
    },
    {
      "epoch": 12.035203520352034,
      "grad_norm": 0.00994873046875,
      "learning_rate": 0.012304847947896788,
      "loss": 0.2309,
      "num_input_tokens_seen": 23088416,
      "step": 109400
    },
    {
      "epoch": 12.035753575357536,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.01230343135534009,
      "loss": 0.233,
      "num_input_tokens_seen": 23089472,
      "step": 109405
    },
    {
      "epoch": 12.036303630363037,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012302014787636381,
      "loss": 0.2335,
      "num_input_tokens_seen": 23090464,
      "step": 109410
    },
    {
      "epoch": 12.036853685368538,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.012300598244798703,
      "loss": 0.2288,
      "num_input_tokens_seen": 23091488,
      "step": 109415
    },
    {
      "epoch": 12.037403740374037,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.012299181726840117,
      "loss": 0.2303,
      "num_input_tokens_seen": 23092512,
      "step": 109420
    },
    {
      "epoch": 12.037953795379538,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01229776523377368,
      "loss": 0.2314,
      "num_input_tokens_seen": 23093600,
      "step": 109425
    },
    {
      "epoch": 12.03850385038504,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.012296348765612438,
      "loss": 0.2324,
      "num_input_tokens_seen": 23094656,
      "step": 109430
    },
    {
      "epoch": 12.039053905390539,
      "grad_norm": 0.000423431396484375,
      "learning_rate": 0.012294932322369454,
      "loss": 0.2304,
      "num_input_tokens_seen": 23095648,
      "step": 109435
    },
    {
      "epoch": 12.03960396039604,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.01229351590405778,
      "loss": 0.2314,
      "num_input_tokens_seen": 23096672,
      "step": 109440
    },
    {
      "epoch": 12.04015401540154,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.012292099510690477,
      "loss": 0.2319,
      "num_input_tokens_seen": 23097760,
      "step": 109445
    },
    {
      "epoch": 12.04070407040704,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.012290683142280589,
      "loss": 0.2314,
      "num_input_tokens_seen": 23098752,
      "step": 109450
    },
    {
      "epoch": 12.041254125412541,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012289266798841174,
      "loss": 0.2345,
      "num_input_tokens_seen": 23099872,
      "step": 109455
    },
    {
      "epoch": 12.041804180418042,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.012287850480385288,
      "loss": 0.2335,
      "num_input_tokens_seen": 23100928,
      "step": 109460
    },
    {
      "epoch": 12.042354235423542,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.012286434186925982,
      "loss": 0.2325,
      "num_input_tokens_seen": 23101984,
      "step": 109465
    },
    {
      "epoch": 12.042904290429043,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.012285017918476308,
      "loss": 0.2309,
      "num_input_tokens_seen": 23103072,
      "step": 109470
    },
    {
      "epoch": 12.043454345434544,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012283601675049324,
      "loss": 0.2308,
      "num_input_tokens_seen": 23104064,
      "step": 109475
    },
    {
      "epoch": 12.044004400440045,
      "grad_norm": 0.000518798828125,
      "learning_rate": 0.012282185456658077,
      "loss": 0.2293,
      "num_input_tokens_seen": 23105088,
      "step": 109480
    },
    {
      "epoch": 12.044554455445544,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012280769263315627,
      "loss": 0.2314,
      "num_input_tokens_seen": 23106112,
      "step": 109485
    },
    {
      "epoch": 12.045104510451045,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.012279353095035018,
      "loss": 0.233,
      "num_input_tokens_seen": 23107168,
      "step": 109490
    },
    {
      "epoch": 12.045654565456546,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.012277936951829306,
      "loss": 0.2314,
      "num_input_tokens_seen": 23108192,
      "step": 109495
    },
    {
      "epoch": 12.046204620462046,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.012276520833711545,
      "loss": 0.2319,
      "num_input_tokens_seen": 23109376,
      "step": 109500
    },
    {
      "epoch": 12.046754675467547,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.012275104740694782,
      "loss": 0.2309,
      "num_input_tokens_seen": 23110368,
      "step": 109505
    },
    {
      "epoch": 12.047304730473048,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.012273688672792073,
      "loss": 0.2314,
      "num_input_tokens_seen": 23111456,
      "step": 109510
    },
    {
      "epoch": 12.047854785478547,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01227227263001647,
      "loss": 0.2298,
      "num_input_tokens_seen": 23112448,
      "step": 109515
    },
    {
      "epoch": 12.048404840484048,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.012270856612381018,
      "loss": 0.2324,
      "num_input_tokens_seen": 23113536,
      "step": 109520
    },
    {
      "epoch": 12.04895489548955,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01226944061989877,
      "loss": 0.2319,
      "num_input_tokens_seen": 23114560,
      "step": 109525
    },
    {
      "epoch": 12.049504950495049,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.012268024652582777,
      "loss": 0.2298,
      "num_input_tokens_seen": 23115648,
      "step": 109530
    },
    {
      "epoch": 12.05005500550055,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012266608710446093,
      "loss": 0.2314,
      "num_input_tokens_seen": 23116704,
      "step": 109535
    },
    {
      "epoch": 12.05060506050605,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.012265192793501765,
      "loss": 0.2319,
      "num_input_tokens_seen": 23117728,
      "step": 109540
    },
    {
      "epoch": 12.051155115511552,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01226377690176284,
      "loss": 0.2319,
      "num_input_tokens_seen": 23118816,
      "step": 109545
    },
    {
      "epoch": 12.051705170517051,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.012262361035242373,
      "loss": 0.2351,
      "num_input_tokens_seen": 23119904,
      "step": 109550
    },
    {
      "epoch": 12.052255225522552,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.012260945193953406,
      "loss": 0.2335,
      "num_input_tokens_seen": 23120928,
      "step": 109555
    },
    {
      "epoch": 12.052805280528053,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.012259529377909,
      "loss": 0.2335,
      "num_input_tokens_seen": 23122016,
      "step": 109560
    },
    {
      "epoch": 12.053355335533553,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.012258113587122194,
      "loss": 0.233,
      "num_input_tokens_seen": 23123040,
      "step": 109565
    },
    {
      "epoch": 12.053905390539054,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.012256697821606035,
      "loss": 0.2303,
      "num_input_tokens_seen": 23124064,
      "step": 109570
    },
    {
      "epoch": 12.054455445544555,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.01225528208137358,
      "loss": 0.2304,
      "num_input_tokens_seen": 23125152,
      "step": 109575
    },
    {
      "epoch": 12.055005500550054,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012253866366437872,
      "loss": 0.2303,
      "num_input_tokens_seen": 23126240,
      "step": 109580
    },
    {
      "epoch": 12.055555555555555,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01225245067681196,
      "loss": 0.2314,
      "num_input_tokens_seen": 23127296,
      "step": 109585
    },
    {
      "epoch": 12.056105610561056,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.012251035012508892,
      "loss": 0.2308,
      "num_input_tokens_seen": 23128288,
      "step": 109590
    },
    {
      "epoch": 12.056655665566556,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.012249619373541713,
      "loss": 0.233,
      "num_input_tokens_seen": 23129408,
      "step": 109595
    },
    {
      "epoch": 12.057205720572057,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.012248203759923477,
      "loss": 0.2325,
      "num_input_tokens_seen": 23130432,
      "step": 109600
    },
    {
      "epoch": 12.057755775577558,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.012246788171667222,
      "loss": 0.2288,
      "num_input_tokens_seen": 23131520,
      "step": 109605
    },
    {
      "epoch": 12.058305830583059,
      "grad_norm": 0.000583648681640625,
      "learning_rate": 0.012245372608786002,
      "loss": 0.2314,
      "num_input_tokens_seen": 23132544,
      "step": 109610
    },
    {
      "epoch": 12.058855885588558,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012243957071292862,
      "loss": 0.2299,
      "num_input_tokens_seen": 23133568,
      "step": 109615
    },
    {
      "epoch": 12.05940594059406,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012242541559200844,
      "loss": 0.2298,
      "num_input_tokens_seen": 23134560,
      "step": 109620
    },
    {
      "epoch": 12.05995599559956,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.012241126072522998,
      "loss": 0.2303,
      "num_input_tokens_seen": 23135616,
      "step": 109625
    },
    {
      "epoch": 12.06050605060506,
      "grad_norm": 0.00077056884765625,
      "learning_rate": 0.012239710611272367,
      "loss": 0.2324,
      "num_input_tokens_seen": 23136640,
      "step": 109630
    },
    {
      "epoch": 12.061056105610561,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012238295175462007,
      "loss": 0.2335,
      "num_input_tokens_seen": 23137696,
      "step": 109635
    },
    {
      "epoch": 12.061606160616062,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.012236879765104949,
      "loss": 0.2324,
      "num_input_tokens_seen": 23138720,
      "step": 109640
    },
    {
      "epoch": 12.062156215621561,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.012235464380214243,
      "loss": 0.2298,
      "num_input_tokens_seen": 23139744,
      "step": 109645
    },
    {
      "epoch": 12.062706270627062,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.012234049020802938,
      "loss": 0.2309,
      "num_input_tokens_seen": 23140864,
      "step": 109650
    },
    {
      "epoch": 12.063256325632564,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.01223263368688407,
      "loss": 0.2324,
      "num_input_tokens_seen": 23142048,
      "step": 109655
    },
    {
      "epoch": 12.063806380638065,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0122312183784707,
      "loss": 0.2329,
      "num_input_tokens_seen": 23143040,
      "step": 109660
    },
    {
      "epoch": 12.064356435643564,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.012229803095575855,
      "loss": 0.2319,
      "num_input_tokens_seen": 23144064,
      "step": 109665
    },
    {
      "epoch": 12.064906490649065,
      "grad_norm": 0.000888824462890625,
      "learning_rate": 0.012228387838212585,
      "loss": 0.2319,
      "num_input_tokens_seen": 23145152,
      "step": 109670
    },
    {
      "epoch": 12.065456545654566,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.012226972606393941,
      "loss": 0.2314,
      "num_input_tokens_seen": 23146176,
      "step": 109675
    },
    {
      "epoch": 12.066006600660065,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.012225557400132956,
      "loss": 0.2319,
      "num_input_tokens_seen": 23147200,
      "step": 109680
    },
    {
      "epoch": 12.066556655665567,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012224142219442673,
      "loss": 0.2319,
      "num_input_tokens_seen": 23148288,
      "step": 109685
    },
    {
      "epoch": 12.067106710671068,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012222727064336145,
      "loss": 0.2309,
      "num_input_tokens_seen": 23149376,
      "step": 109690
    },
    {
      "epoch": 12.067656765676567,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.012221311934826404,
      "loss": 0.2293,
      "num_input_tokens_seen": 23150368,
      "step": 109695
    },
    {
      "epoch": 12.068206820682068,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012219896830926502,
      "loss": 0.2329,
      "num_input_tokens_seen": 23151456,
      "step": 109700
    },
    {
      "epoch": 12.06875687568757,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012218481752649477,
      "loss": 0.2303,
      "num_input_tokens_seen": 23152512,
      "step": 109705
    },
    {
      "epoch": 12.069306930693068,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.012217066700008362,
      "loss": 0.2314,
      "num_input_tokens_seen": 23153600,
      "step": 109710
    },
    {
      "epoch": 12.06985698569857,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01221565167301622,
      "loss": 0.2319,
      "num_input_tokens_seen": 23154688,
      "step": 109715
    },
    {
      "epoch": 12.07040704070407,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.012214236671686073,
      "loss": 0.2293,
      "num_input_tokens_seen": 23155776,
      "step": 109720
    },
    {
      "epoch": 12.070957095709572,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012212821696030971,
      "loss": 0.233,
      "num_input_tokens_seen": 23156864,
      "step": 109725
    },
    {
      "epoch": 12.071507150715071,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.012211406746063958,
      "loss": 0.2324,
      "num_input_tokens_seen": 23157952,
      "step": 109730
    },
    {
      "epoch": 12.072057205720572,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.012209991821798067,
      "loss": 0.2324,
      "num_input_tokens_seen": 23159072,
      "step": 109735
    },
    {
      "epoch": 12.072607260726073,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012208576923246343,
      "loss": 0.2314,
      "num_input_tokens_seen": 23160256,
      "step": 109740
    },
    {
      "epoch": 12.073157315731573,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.012207162050421824,
      "loss": 0.2304,
      "num_input_tokens_seen": 23161408,
      "step": 109745
    },
    {
      "epoch": 12.073707370737074,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.012205747203337558,
      "loss": 0.2319,
      "num_input_tokens_seen": 23162400,
      "step": 109750
    },
    {
      "epoch": 12.074257425742575,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.012204332382006576,
      "loss": 0.2335,
      "num_input_tokens_seen": 23163456,
      "step": 109755
    },
    {
      "epoch": 12.074807480748074,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01220291758644192,
      "loss": 0.2335,
      "num_input_tokens_seen": 23164544,
      "step": 109760
    },
    {
      "epoch": 12.075357535753575,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.012201502816656634,
      "loss": 0.2319,
      "num_input_tokens_seen": 23165600,
      "step": 109765
    },
    {
      "epoch": 12.075907590759076,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.012200088072663744,
      "loss": 0.2319,
      "num_input_tokens_seen": 23166656,
      "step": 109770
    },
    {
      "epoch": 12.076457645764576,
      "grad_norm": 0.000446319580078125,
      "learning_rate": 0.012198673354476311,
      "loss": 0.2303,
      "num_input_tokens_seen": 23167712,
      "step": 109775
    },
    {
      "epoch": 12.077007700770077,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.012197258662107356,
      "loss": 0.2298,
      "num_input_tokens_seen": 23168768,
      "step": 109780
    },
    {
      "epoch": 12.077557755775578,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.012195843995569921,
      "loss": 0.234,
      "num_input_tokens_seen": 23169760,
      "step": 109785
    },
    {
      "epoch": 12.078107810781079,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.012194429354877053,
      "loss": 0.2304,
      "num_input_tokens_seen": 23170848,
      "step": 109790
    },
    {
      "epoch": 12.078657865786578,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.012193014740041776,
      "loss": 0.2335,
      "num_input_tokens_seen": 23171904,
      "step": 109795
    },
    {
      "epoch": 12.07920792079208,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.012191600151077137,
      "loss": 0.2298,
      "num_input_tokens_seen": 23172928,
      "step": 109800
    },
    {
      "epoch": 12.07975797579758,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.012190185587996177,
      "loss": 0.2335,
      "num_input_tokens_seen": 23174016,
      "step": 109805
    },
    {
      "epoch": 12.08030803080308,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.012188771050811922,
      "loss": 0.2309,
      "num_input_tokens_seen": 23175072,
      "step": 109810
    },
    {
      "epoch": 12.08085808580858,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012187356539537418,
      "loss": 0.2309,
      "num_input_tokens_seen": 23176192,
      "step": 109815
    },
    {
      "epoch": 12.081408140814082,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.012185942054185696,
      "loss": 0.2319,
      "num_input_tokens_seen": 23177248,
      "step": 109820
    },
    {
      "epoch": 12.081958195819581,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.012184527594769802,
      "loss": 0.2319,
      "num_input_tokens_seen": 23178272,
      "step": 109825
    },
    {
      "epoch": 12.082508250825082,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.012183113161302763,
      "loss": 0.2324,
      "num_input_tokens_seen": 23179360,
      "step": 109830
    },
    {
      "epoch": 12.083058305830583,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.012181698753797616,
      "loss": 0.2303,
      "num_input_tokens_seen": 23180384,
      "step": 109835
    },
    {
      "epoch": 12.083608360836084,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.0121802843722674,
      "loss": 0.2329,
      "num_input_tokens_seen": 23181504,
      "step": 109840
    },
    {
      "epoch": 12.084158415841584,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012178870016725149,
      "loss": 0.2309,
      "num_input_tokens_seen": 23182528,
      "step": 109845
    },
    {
      "epoch": 12.084708470847085,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012177455687183904,
      "loss": 0.2293,
      "num_input_tokens_seen": 23183552,
      "step": 109850
    },
    {
      "epoch": 12.085258525852586,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.012176041383656694,
      "loss": 0.2309,
      "num_input_tokens_seen": 23184576,
      "step": 109855
    },
    {
      "epoch": 12.085808580858085,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.012174627106156551,
      "loss": 0.2309,
      "num_input_tokens_seen": 23185664,
      "step": 109860
    },
    {
      "epoch": 12.086358635863586,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.012173212854696523,
      "loss": 0.2324,
      "num_input_tokens_seen": 23186720,
      "step": 109865
    },
    {
      "epoch": 12.086908690869087,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01217179862928963,
      "loss": 0.2319,
      "num_input_tokens_seen": 23187744,
      "step": 109870
    },
    {
      "epoch": 12.087458745874587,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.01217038442994891,
      "loss": 0.2309,
      "num_input_tokens_seen": 23188768,
      "step": 109875
    },
    {
      "epoch": 12.088008800880088,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.012168970256687403,
      "loss": 0.2319,
      "num_input_tokens_seen": 23189824,
      "step": 109880
    },
    {
      "epoch": 12.088558855885589,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.012167556109518135,
      "loss": 0.2329,
      "num_input_tokens_seen": 23190912,
      "step": 109885
    },
    {
      "epoch": 12.089108910891088,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012166141988454147,
      "loss": 0.2324,
      "num_input_tokens_seen": 23191936,
      "step": 109890
    },
    {
      "epoch": 12.08965896589659,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.012164727893508467,
      "loss": 0.2299,
      "num_input_tokens_seen": 23192992,
      "step": 109895
    },
    {
      "epoch": 12.09020902090209,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.012163313824694126,
      "loss": 0.2309,
      "num_input_tokens_seen": 23194048,
      "step": 109900
    },
    {
      "epoch": 12.090759075907592,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.012161899782024166,
      "loss": 0.2314,
      "num_input_tokens_seen": 23195072,
      "step": 109905
    },
    {
      "epoch": 12.091309130913091,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.012160485765511608,
      "loss": 0.2314,
      "num_input_tokens_seen": 23196128,
      "step": 109910
    },
    {
      "epoch": 12.091859185918592,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.012159071775169493,
      "loss": 0.2319,
      "num_input_tokens_seen": 23197152,
      "step": 109915
    },
    {
      "epoch": 12.092409240924093,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012157657811010854,
      "loss": 0.2324,
      "num_input_tokens_seen": 23198208,
      "step": 109920
    },
    {
      "epoch": 12.092959295929592,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.012156243873048716,
      "loss": 0.2303,
      "num_input_tokens_seen": 23199328,
      "step": 109925
    },
    {
      "epoch": 12.093509350935093,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.012154829961296113,
      "loss": 0.2319,
      "num_input_tokens_seen": 23200352,
      "step": 109930
    },
    {
      "epoch": 12.094059405940595,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012153416075766074,
      "loss": 0.2303,
      "num_input_tokens_seen": 23201440,
      "step": 109935
    },
    {
      "epoch": 12.094609460946094,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01215200221647164,
      "loss": 0.2319,
      "num_input_tokens_seen": 23202496,
      "step": 109940
    },
    {
      "epoch": 12.095159515951595,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012150588383425831,
      "loss": 0.2324,
      "num_input_tokens_seen": 23203456,
      "step": 109945
    },
    {
      "epoch": 12.095709570957096,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.01214917457664168,
      "loss": 0.2319,
      "num_input_tokens_seen": 23204544,
      "step": 109950
    },
    {
      "epoch": 12.096259625962595,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01214776079613222,
      "loss": 0.2288,
      "num_input_tokens_seen": 23205632,
      "step": 109955
    },
    {
      "epoch": 12.096809680968097,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01214634704191048,
      "loss": 0.2324,
      "num_input_tokens_seen": 23206656,
      "step": 109960
    },
    {
      "epoch": 12.097359735973598,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.012144933313989494,
      "loss": 0.2303,
      "num_input_tokens_seen": 23207712,
      "step": 109965
    },
    {
      "epoch": 12.097909790979099,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.012143519612382286,
      "loss": 0.2324,
      "num_input_tokens_seen": 23208768,
      "step": 109970
    },
    {
      "epoch": 12.098459845984598,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.012142105937101882,
      "loss": 0.2324,
      "num_input_tokens_seen": 23209824,
      "step": 109975
    },
    {
      "epoch": 12.099009900990099,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.012140692288161323,
      "loss": 0.2324,
      "num_input_tokens_seen": 23210912,
      "step": 109980
    },
    {
      "epoch": 12.0995599559956,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.012139278665573626,
      "loss": 0.2319,
      "num_input_tokens_seen": 23212032,
      "step": 109985
    },
    {
      "epoch": 12.1001100110011,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.012137865069351828,
      "loss": 0.2298,
      "num_input_tokens_seen": 23213120,
      "step": 109990
    },
    {
      "epoch": 12.1006600660066,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.012136451499508956,
      "loss": 0.2304,
      "num_input_tokens_seen": 23214272,
      "step": 109995
    },
    {
      "epoch": 12.101210121012102,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.012135037956058033,
      "loss": 0.2309,
      "num_input_tokens_seen": 23215296,
      "step": 110000
    },
    {
      "epoch": 12.101760176017601,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01213362443901209,
      "loss": 0.2319,
      "num_input_tokens_seen": 23216352,
      "step": 110005
    },
    {
      "epoch": 12.102310231023102,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012132210948384153,
      "loss": 0.2324,
      "num_input_tokens_seen": 23217440,
      "step": 110010
    },
    {
      "epoch": 12.102860286028603,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.012130797484187257,
      "loss": 0.2314,
      "num_input_tokens_seen": 23218432,
      "step": 110015
    },
    {
      "epoch": 12.103410341034103,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.012129384046434424,
      "loss": 0.2303,
      "num_input_tokens_seen": 23219424,
      "step": 110020
    },
    {
      "epoch": 12.103960396039604,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.012127970635138676,
      "loss": 0.2303,
      "num_input_tokens_seen": 23220416,
      "step": 110025
    },
    {
      "epoch": 12.104510451045105,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.012126557250313046,
      "loss": 0.2308,
      "num_input_tokens_seen": 23221472,
      "step": 110030
    },
    {
      "epoch": 12.105060506050606,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.012125143891970558,
      "loss": 0.2319,
      "num_input_tokens_seen": 23222432,
      "step": 110035
    },
    {
      "epoch": 12.105610561056105,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012123730560124247,
      "loss": 0.2303,
      "num_input_tokens_seen": 23223456,
      "step": 110040
    },
    {
      "epoch": 12.106160616061606,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012122317254787124,
      "loss": 0.2314,
      "num_input_tokens_seen": 23224544,
      "step": 110045
    },
    {
      "epoch": 12.106710671067107,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.01212090397597222,
      "loss": 0.2319,
      "num_input_tokens_seen": 23225600,
      "step": 110050
    },
    {
      "epoch": 12.107260726072607,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.012119490723692571,
      "loss": 0.2319,
      "num_input_tokens_seen": 23226688,
      "step": 110055
    },
    {
      "epoch": 12.107810781078108,
      "grad_norm": 0.000698089599609375,
      "learning_rate": 0.01211807749796119,
      "loss": 0.2309,
      "num_input_tokens_seen": 23227712,
      "step": 110060
    },
    {
      "epoch": 12.108360836083609,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0121166642987911,
      "loss": 0.2309,
      "num_input_tokens_seen": 23228736,
      "step": 110065
    },
    {
      "epoch": 12.108910891089108,
      "grad_norm": 0.0003509521484375,
      "learning_rate": 0.012115251126195341,
      "loss": 0.2319,
      "num_input_tokens_seen": 23229824,
      "step": 110070
    },
    {
      "epoch": 12.10946094609461,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.012113837980186917,
      "loss": 0.2304,
      "num_input_tokens_seen": 23230944,
      "step": 110075
    },
    {
      "epoch": 12.11001100110011,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.012112424860778872,
      "loss": 0.2319,
      "num_input_tokens_seen": 23231968,
      "step": 110080
    },
    {
      "epoch": 12.110561056105611,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.012111011767984221,
      "loss": 0.2324,
      "num_input_tokens_seen": 23233056,
      "step": 110085
    },
    {
      "epoch": 12.11111111111111,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012109598701815983,
      "loss": 0.2314,
      "num_input_tokens_seen": 23234112,
      "step": 110090
    },
    {
      "epoch": 12.111661166116612,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.012108185662287196,
      "loss": 0.2288,
      "num_input_tokens_seen": 23235168,
      "step": 110095
    },
    {
      "epoch": 12.112211221122113,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.012106772649410865,
      "loss": 0.2298,
      "num_input_tokens_seen": 23236256,
      "step": 110100
    },
    {
      "epoch": 12.112761276127612,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.012105359663200026,
      "loss": 0.2324,
      "num_input_tokens_seen": 23237280,
      "step": 110105
    },
    {
      "epoch": 12.113311331133113,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012103946703667701,
      "loss": 0.2298,
      "num_input_tokens_seen": 23238304,
      "step": 110110
    },
    {
      "epoch": 12.113861386138614,
      "grad_norm": 0.005859375,
      "learning_rate": 0.012102533770826904,
      "loss": 0.233,
      "num_input_tokens_seen": 23239328,
      "step": 110115
    },
    {
      "epoch": 12.114411441144114,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012101120864690664,
      "loss": 0.2319,
      "num_input_tokens_seen": 23240320,
      "step": 110120
    },
    {
      "epoch": 12.114961496149615,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.012099707985272,
      "loss": 0.2314,
      "num_input_tokens_seen": 23241280,
      "step": 110125
    },
    {
      "epoch": 12.115511551155116,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01209829513258394,
      "loss": 0.2319,
      "num_input_tokens_seen": 23242272,
      "step": 110130
    },
    {
      "epoch": 12.116061606160615,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.012096882306639505,
      "loss": 0.2314,
      "num_input_tokens_seen": 23243328,
      "step": 110135
    },
    {
      "epoch": 12.116611661166116,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.012095469507451705,
      "loss": 0.2309,
      "num_input_tokens_seen": 23244352,
      "step": 110140
    },
    {
      "epoch": 12.117161716171617,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.012094056735033573,
      "loss": 0.2314,
      "num_input_tokens_seen": 23245472,
      "step": 110145
    },
    {
      "epoch": 12.117711771177119,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01209264398939812,
      "loss": 0.2303,
      "num_input_tokens_seen": 23246560,
      "step": 110150
    },
    {
      "epoch": 12.118261826182618,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01209123127055838,
      "loss": 0.2314,
      "num_input_tokens_seen": 23247552,
      "step": 110155
    },
    {
      "epoch": 12.118811881188119,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.012089818578527364,
      "loss": 0.2335,
      "num_input_tokens_seen": 23248672,
      "step": 110160
    },
    {
      "epoch": 12.11936193619362,
      "grad_norm": 0.00052642822265625,
      "learning_rate": 0.01208840591331809,
      "loss": 0.2309,
      "num_input_tokens_seen": 23249632,
      "step": 110165
    },
    {
      "epoch": 12.11991199119912,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.012086993274943588,
      "loss": 0.2309,
      "num_input_tokens_seen": 23250656,
      "step": 110170
    },
    {
      "epoch": 12.12046204620462,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.012085580663416863,
      "loss": 0.2313,
      "num_input_tokens_seen": 23251712,
      "step": 110175
    },
    {
      "epoch": 12.121012101210122,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.012084168078750946,
      "loss": 0.2303,
      "num_input_tokens_seen": 23252800,
      "step": 110180
    },
    {
      "epoch": 12.12156215621562,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.012082755520958858,
      "loss": 0.2298,
      "num_input_tokens_seen": 23253824,
      "step": 110185
    },
    {
      "epoch": 12.122112211221122,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.012081342990053602,
      "loss": 0.2325,
      "num_input_tokens_seen": 23254944,
      "step": 110190
    },
    {
      "epoch": 12.122662266226623,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.012079930486048216,
      "loss": 0.2309,
      "num_input_tokens_seen": 23256000,
      "step": 110195
    },
    {
      "epoch": 12.123212321232122,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.012078518008955704,
      "loss": 0.2293,
      "num_input_tokens_seen": 23257024,
      "step": 110200
    },
    {
      "epoch": 12.123762376237623,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.01207710555878909,
      "loss": 0.2319,
      "num_input_tokens_seen": 23258080,
      "step": 110205
    },
    {
      "epoch": 12.124312431243125,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012075693135561399,
      "loss": 0.2324,
      "num_input_tokens_seen": 23259136,
      "step": 110210
    },
    {
      "epoch": 12.124862486248626,
      "grad_norm": 0.000576019287109375,
      "learning_rate": 0.012074280739285632,
      "loss": 0.2324,
      "num_input_tokens_seen": 23260224,
      "step": 110215
    },
    {
      "epoch": 12.125412541254125,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.01207286836997482,
      "loss": 0.2329,
      "num_input_tokens_seen": 23261312,
      "step": 110220
    },
    {
      "epoch": 12.125962596259626,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.012071456027641973,
      "loss": 0.2303,
      "num_input_tokens_seen": 23262336,
      "step": 110225
    },
    {
      "epoch": 12.126512651265127,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.012070043712300116,
      "loss": 0.2345,
      "num_input_tokens_seen": 23263392,
      "step": 110230
    },
    {
      "epoch": 12.127062706270626,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.012068631423962256,
      "loss": 0.2319,
      "num_input_tokens_seen": 23264416,
      "step": 110235
    },
    {
      "epoch": 12.127612761276128,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012067219162641411,
      "loss": 0.2303,
      "num_input_tokens_seen": 23265504,
      "step": 110240
    },
    {
      "epoch": 12.128162816281629,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012065806928350608,
      "loss": 0.2303,
      "num_input_tokens_seen": 23266560,
      "step": 110245
    },
    {
      "epoch": 12.128712871287128,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.01206439472110285,
      "loss": 0.2314,
      "num_input_tokens_seen": 23267616,
      "step": 110250
    },
    {
      "epoch": 12.129262926292629,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012062982540911153,
      "loss": 0.2308,
      "num_input_tokens_seen": 23268640,
      "step": 110255
    },
    {
      "epoch": 12.12981298129813,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.012061570387788541,
      "loss": 0.2308,
      "num_input_tokens_seen": 23269728,
      "step": 110260
    },
    {
      "epoch": 12.130363036303631,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.01206015826174802,
      "loss": 0.2319,
      "num_input_tokens_seen": 23270784,
      "step": 110265
    },
    {
      "epoch": 12.13091309130913,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012058746162802617,
      "loss": 0.2308,
      "num_input_tokens_seen": 23271840,
      "step": 110270
    },
    {
      "epoch": 12.131463146314632,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.012057334090965337,
      "loss": 0.2308,
      "num_input_tokens_seen": 23272928,
      "step": 110275
    },
    {
      "epoch": 12.132013201320133,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.012055922046249192,
      "loss": 0.2329,
      "num_input_tokens_seen": 23273984,
      "step": 110280
    },
    {
      "epoch": 12.132563256325632,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.012054510028667208,
      "loss": 0.2314,
      "num_input_tokens_seen": 23275040,
      "step": 110285
    },
    {
      "epoch": 12.133113311331133,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.012053098038232386,
      "loss": 0.2319,
      "num_input_tokens_seen": 23276096,
      "step": 110290
    },
    {
      "epoch": 12.133663366336634,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.012051686074957747,
      "loss": 0.2304,
      "num_input_tokens_seen": 23277184,
      "step": 110295
    },
    {
      "epoch": 12.134213421342134,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.012050274138856306,
      "loss": 0.2298,
      "num_input_tokens_seen": 23278240,
      "step": 110300
    },
    {
      "epoch": 12.134763476347635,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.012048862229941068,
      "loss": 0.2329,
      "num_input_tokens_seen": 23279264,
      "step": 110305
    },
    {
      "epoch": 12.135313531353136,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.012047450348225054,
      "loss": 0.2304,
      "num_input_tokens_seen": 23280320,
      "step": 110310
    },
    {
      "epoch": 12.135863586358635,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012046038493721269,
      "loss": 0.2314,
      "num_input_tokens_seen": 23281408,
      "step": 110315
    },
    {
      "epoch": 12.136413641364136,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012044626666442734,
      "loss": 0.2304,
      "num_input_tokens_seen": 23282464,
      "step": 110320
    },
    {
      "epoch": 12.136963696369637,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.012043214866402457,
      "loss": 0.2319,
      "num_input_tokens_seen": 23283488,
      "step": 110325
    },
    {
      "epoch": 12.137513751375138,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.012041803093613447,
      "loss": 0.2324,
      "num_input_tokens_seen": 23284544,
      "step": 110330
    },
    {
      "epoch": 12.138063806380638,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.012040391348088719,
      "loss": 0.2325,
      "num_input_tokens_seen": 23285664,
      "step": 110335
    },
    {
      "epoch": 12.138613861386139,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.012038979629841283,
      "loss": 0.2324,
      "num_input_tokens_seen": 23286688,
      "step": 110340
    },
    {
      "epoch": 12.13916391639164,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.012037567938884155,
      "loss": 0.2303,
      "num_input_tokens_seen": 23287712,
      "step": 110345
    },
    {
      "epoch": 12.13971397139714,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.01203615627523034,
      "loss": 0.2314,
      "num_input_tokens_seen": 23288768,
      "step": 110350
    },
    {
      "epoch": 12.14026402640264,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012034744638892847,
      "loss": 0.2319,
      "num_input_tokens_seen": 23289824,
      "step": 110355
    },
    {
      "epoch": 12.140814081408141,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.012033333029884696,
      "loss": 0.2314,
      "num_input_tokens_seen": 23291008,
      "step": 110360
    },
    {
      "epoch": 12.14136413641364,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.012031921448218883,
      "loss": 0.2303,
      "num_input_tokens_seen": 23292096,
      "step": 110365
    },
    {
      "epoch": 12.141914191419142,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.012030509893908433,
      "loss": 0.2298,
      "num_input_tokens_seen": 23293120,
      "step": 110370
    },
    {
      "epoch": 12.142464246424643,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.012029098366966346,
      "loss": 0.2304,
      "num_input_tokens_seen": 23294240,
      "step": 110375
    },
    {
      "epoch": 12.143014301430142,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.01202768686740563,
      "loss": 0.2319,
      "num_input_tokens_seen": 23295232,
      "step": 110380
    },
    {
      "epoch": 12.143564356435643,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.012026275395239304,
      "loss": 0.2298,
      "num_input_tokens_seen": 23296352,
      "step": 110385
    },
    {
      "epoch": 12.144114411441144,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.012024863950480367,
      "loss": 0.2309,
      "num_input_tokens_seen": 23297440,
      "step": 110390
    },
    {
      "epoch": 12.144664466446645,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.012023452533141831,
      "loss": 0.2324,
      "num_input_tokens_seen": 23298592,
      "step": 110395
    },
    {
      "epoch": 12.145214521452145,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.01202204114323671,
      "loss": 0.2319,
      "num_input_tokens_seen": 23299680,
      "step": 110400
    },
    {
      "epoch": 12.145764576457646,
      "grad_norm": 0.000370025634765625,
      "learning_rate": 0.012020629780778001,
      "loss": 0.2308,
      "num_input_tokens_seen": 23300736,
      "step": 110405
    },
    {
      "epoch": 12.146314631463147,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.012019218445778721,
      "loss": 0.2324,
      "num_input_tokens_seen": 23301824,
      "step": 110410
    },
    {
      "epoch": 12.146864686468646,
      "grad_norm": 0.00299072265625,
      "learning_rate": 0.012017807138251868,
      "loss": 0.2319,
      "num_input_tokens_seen": 23302880,
      "step": 110415
    },
    {
      "epoch": 12.147414741474147,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.012016395858210465,
      "loss": 0.2309,
      "num_input_tokens_seen": 23303968,
      "step": 110420
    },
    {
      "epoch": 12.147964796479648,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.012014984605667505,
      "loss": 0.2308,
      "num_input_tokens_seen": 23305088,
      "step": 110425
    },
    {
      "epoch": 12.148514851485148,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.012013573380635996,
      "loss": 0.2319,
      "num_input_tokens_seen": 23306176,
      "step": 110430
    },
    {
      "epoch": 12.149064906490649,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.012012162183128955,
      "loss": 0.2308,
      "num_input_tokens_seen": 23307232,
      "step": 110435
    },
    {
      "epoch": 12.14961496149615,
      "grad_norm": 0.00080108642578125,
      "learning_rate": 0.01201075101315938,
      "loss": 0.2308,
      "num_input_tokens_seen": 23308288,
      "step": 110440
    },
    {
      "epoch": 12.150165016501651,
      "grad_norm": 0.000682830810546875,
      "learning_rate": 0.012009339870740274,
      "loss": 0.2303,
      "num_input_tokens_seen": 23309376,
      "step": 110445
    },
    {
      "epoch": 12.15071507150715,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.012007928755884651,
      "loss": 0.235,
      "num_input_tokens_seen": 23310464,
      "step": 110450
    },
    {
      "epoch": 12.151265126512651,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.012006517668605509,
      "loss": 0.2314,
      "num_input_tokens_seen": 23311616,
      "step": 110455
    },
    {
      "epoch": 12.151815181518153,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.012005106608915865,
      "loss": 0.2303,
      "num_input_tokens_seen": 23312608,
      "step": 110460
    },
    {
      "epoch": 12.152365236523652,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.012003695576828712,
      "loss": 0.2324,
      "num_input_tokens_seen": 23313696,
      "step": 110465
    },
    {
      "epoch": 12.152915291529153,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.012002284572357056,
      "loss": 0.2325,
      "num_input_tokens_seen": 23314784,
      "step": 110470
    },
    {
      "epoch": 12.153465346534654,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.01200087359551391,
      "loss": 0.2314,
      "num_input_tokens_seen": 23315808,
      "step": 110475
    },
    {
      "epoch": 12.154015401540153,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.011999462646312267,
      "loss": 0.2324,
      "num_input_tokens_seen": 23316960,
      "step": 110480
    },
    {
      "epoch": 12.154565456545654,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011998051724765141,
      "loss": 0.2309,
      "num_input_tokens_seen": 23318048,
      "step": 110485
    },
    {
      "epoch": 12.155115511551156,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011996640830885535,
      "loss": 0.2314,
      "num_input_tokens_seen": 23319072,
      "step": 110490
    },
    {
      "epoch": 12.155665566556655,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01199522996468644,
      "loss": 0.2304,
      "num_input_tokens_seen": 23320160,
      "step": 110495
    },
    {
      "epoch": 12.156215621562156,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011993819126180878,
      "loss": 0.2319,
      "num_input_tokens_seen": 23321184,
      "step": 110500
    },
    {
      "epoch": 12.156765676567657,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.011992408315381835,
      "loss": 0.2324,
      "num_input_tokens_seen": 23322240,
      "step": 110505
    },
    {
      "epoch": 12.157315731573158,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.011990997532302325,
      "loss": 0.2298,
      "num_input_tokens_seen": 23323360,
      "step": 110510
    },
    {
      "epoch": 12.157865786578657,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01198958677695535,
      "loss": 0.2298,
      "num_input_tokens_seen": 23324448,
      "step": 110515
    },
    {
      "epoch": 12.158415841584159,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011988176049353903,
      "loss": 0.2309,
      "num_input_tokens_seen": 23325472,
      "step": 110520
    },
    {
      "epoch": 12.15896589658966,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.011986765349510996,
      "loss": 0.2314,
      "num_input_tokens_seen": 23326432,
      "step": 110525
    },
    {
      "epoch": 12.159515951595159,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.011985354677439624,
      "loss": 0.2303,
      "num_input_tokens_seen": 23327520,
      "step": 110530
    },
    {
      "epoch": 12.16006600660066,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.011983944033152796,
      "loss": 0.2324,
      "num_input_tokens_seen": 23328544,
      "step": 110535
    },
    {
      "epoch": 12.160616061606161,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.011982533416663508,
      "loss": 0.2314,
      "num_input_tokens_seen": 23329632,
      "step": 110540
    },
    {
      "epoch": 12.16116611661166,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.011981122827984756,
      "loss": 0.2314,
      "num_input_tokens_seen": 23330688,
      "step": 110545
    },
    {
      "epoch": 12.161716171617162,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.011979712267129553,
      "loss": 0.233,
      "num_input_tokens_seen": 23331712,
      "step": 110550
    },
    {
      "epoch": 12.162266226622663,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.011978301734110883,
      "loss": 0.2314,
      "num_input_tokens_seen": 23332832,
      "step": 110555
    },
    {
      "epoch": 12.162816281628162,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.011976891228941768,
      "loss": 0.2309,
      "num_input_tokens_seen": 23333856,
      "step": 110560
    },
    {
      "epoch": 12.163366336633663,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.011975480751635192,
      "loss": 0.2319,
      "num_input_tokens_seen": 23334880,
      "step": 110565
    },
    {
      "epoch": 12.163916391639164,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.01197407030220415,
      "loss": 0.2319,
      "num_input_tokens_seen": 23335872,
      "step": 110570
    },
    {
      "epoch": 12.164466446644665,
      "grad_norm": 0.0029144287109375,
      "learning_rate": 0.011972659880661663,
      "loss": 0.2309,
      "num_input_tokens_seen": 23336896,
      "step": 110575
    },
    {
      "epoch": 12.165016501650165,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01197124948702071,
      "loss": 0.2308,
      "num_input_tokens_seen": 23337952,
      "step": 110580
    },
    {
      "epoch": 12.165566556655666,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.011969839121294299,
      "loss": 0.2304,
      "num_input_tokens_seen": 23339008,
      "step": 110585
    },
    {
      "epoch": 12.166116611661167,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.011968428783495428,
      "loss": 0.2314,
      "num_input_tokens_seen": 23340096,
      "step": 110590
    },
    {
      "epoch": 12.166666666666666,
      "grad_norm": 0.0004596710205078125,
      "learning_rate": 0.01196701847363709,
      "loss": 0.2298,
      "num_input_tokens_seen": 23341088,
      "step": 110595
    },
    {
      "epoch": 12.167216721672167,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.01196560819173229,
      "loss": 0.2335,
      "num_input_tokens_seen": 23342144,
      "step": 110600
    },
    {
      "epoch": 12.167766776677668,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01196419793779402,
      "loss": 0.2309,
      "num_input_tokens_seen": 23343136,
      "step": 110605
    },
    {
      "epoch": 12.168316831683168,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.01196278771183529,
      "loss": 0.2299,
      "num_input_tokens_seen": 23344160,
      "step": 110610
    },
    {
      "epoch": 12.168866886688669,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.011961377513869081,
      "loss": 0.2314,
      "num_input_tokens_seen": 23345152,
      "step": 110615
    },
    {
      "epoch": 12.16941694169417,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.011959967343908397,
      "loss": 0.2303,
      "num_input_tokens_seen": 23346240,
      "step": 110620
    },
    {
      "epoch": 12.16996699669967,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.011958557201966237,
      "loss": 0.2314,
      "num_input_tokens_seen": 23347296,
      "step": 110625
    },
    {
      "epoch": 12.17051705170517,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.011957147088055599,
      "loss": 0.2319,
      "num_input_tokens_seen": 23348352,
      "step": 110630
    },
    {
      "epoch": 12.171067106710671,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.01195573700218947,
      "loss": 0.2303,
      "num_input_tokens_seen": 23349376,
      "step": 110635
    },
    {
      "epoch": 12.171617161716172,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.011954326944380856,
      "loss": 0.2335,
      "num_input_tokens_seen": 23350432,
      "step": 110640
    },
    {
      "epoch": 12.172167216721672,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.011952916914642744,
      "loss": 0.2314,
      "num_input_tokens_seen": 23351520,
      "step": 110645
    },
    {
      "epoch": 12.172717271727173,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.011951506912988144,
      "loss": 0.2324,
      "num_input_tokens_seen": 23352576,
      "step": 110650
    },
    {
      "epoch": 12.173267326732674,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.011950096939430038,
      "loss": 0.2314,
      "num_input_tokens_seen": 23353600,
      "step": 110655
    },
    {
      "epoch": 12.173817381738173,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01194868699398142,
      "loss": 0.2319,
      "num_input_tokens_seen": 23354688,
      "step": 110660
    },
    {
      "epoch": 12.174367436743674,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.011947277076655298,
      "loss": 0.2314,
      "num_input_tokens_seen": 23355776,
      "step": 110665
    },
    {
      "epoch": 12.174917491749175,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.01194586718746465,
      "loss": 0.2304,
      "num_input_tokens_seen": 23356832,
      "step": 110670
    },
    {
      "epoch": 12.175467546754675,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.011944457326422485,
      "loss": 0.233,
      "num_input_tokens_seen": 23357920,
      "step": 110675
    },
    {
      "epoch": 12.176017601760176,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.011943047493541787,
      "loss": 0.2304,
      "num_input_tokens_seen": 23358944,
      "step": 110680
    },
    {
      "epoch": 12.176567656765677,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.011941637688835553,
      "loss": 0.2314,
      "num_input_tokens_seen": 23360000,
      "step": 110685
    },
    {
      "epoch": 12.177117711771178,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.011940227912316782,
      "loss": 0.2314,
      "num_input_tokens_seen": 23361056,
      "step": 110690
    },
    {
      "epoch": 12.177667766776677,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.011938818163998456,
      "loss": 0.2314,
      "num_input_tokens_seen": 23362112,
      "step": 110695
    },
    {
      "epoch": 12.178217821782178,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.011937408443893578,
      "loss": 0.2319,
      "num_input_tokens_seen": 23363168,
      "step": 110700
    },
    {
      "epoch": 12.17876787678768,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01193599875201514,
      "loss": 0.2314,
      "num_input_tokens_seen": 23364224,
      "step": 110705
    },
    {
      "epoch": 12.179317931793179,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.011934589088376123,
      "loss": 0.2324,
      "num_input_tokens_seen": 23365216,
      "step": 110710
    },
    {
      "epoch": 12.17986798679868,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.011933179452989533,
      "loss": 0.2325,
      "num_input_tokens_seen": 23366304,
      "step": 110715
    },
    {
      "epoch": 12.180418041804181,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.011931769845868351,
      "loss": 0.2319,
      "num_input_tokens_seen": 23367392,
      "step": 110720
    },
    {
      "epoch": 12.18096809680968,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.01193036026702558,
      "loss": 0.2314,
      "num_input_tokens_seen": 23368416,
      "step": 110725
    },
    {
      "epoch": 12.181518151815181,
      "grad_norm": 0.005859375,
      "learning_rate": 0.011928950716474204,
      "loss": 0.2304,
      "num_input_tokens_seen": 23369472,
      "step": 110730
    },
    {
      "epoch": 12.182068206820682,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011927541194227212,
      "loss": 0.2335,
      "num_input_tokens_seen": 23370560,
      "step": 110735
    },
    {
      "epoch": 12.182618261826182,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011926131700297602,
      "loss": 0.2309,
      "num_input_tokens_seen": 23371584,
      "step": 110740
    },
    {
      "epoch": 12.183168316831683,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.011924722234698358,
      "loss": 0.2314,
      "num_input_tokens_seen": 23372640,
      "step": 110745
    },
    {
      "epoch": 12.183718371837184,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01192331279744248,
      "loss": 0.2293,
      "num_input_tokens_seen": 23373696,
      "step": 110750
    },
    {
      "epoch": 12.184268426842685,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.011921903388542947,
      "loss": 0.2309,
      "num_input_tokens_seen": 23374688,
      "step": 110755
    },
    {
      "epoch": 12.184818481848184,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011920494008012752,
      "loss": 0.234,
      "num_input_tokens_seen": 23375744,
      "step": 110760
    },
    {
      "epoch": 12.185368536853685,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011919084655864892,
      "loss": 0.2303,
      "num_input_tokens_seen": 23376864,
      "step": 110765
    },
    {
      "epoch": 12.185918591859187,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.011917675332112344,
      "loss": 0.2303,
      "num_input_tokens_seen": 23377888,
      "step": 110770
    },
    {
      "epoch": 12.186468646864686,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011916266036768104,
      "loss": 0.2298,
      "num_input_tokens_seen": 23378912,
      "step": 110775
    },
    {
      "epoch": 12.187018701870187,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.011914856769845166,
      "loss": 0.2298,
      "num_input_tokens_seen": 23379968,
      "step": 110780
    },
    {
      "epoch": 12.187568756875688,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011913447531356506,
      "loss": 0.2309,
      "num_input_tokens_seen": 23381024,
      "step": 110785
    },
    {
      "epoch": 12.188118811881187,
      "grad_norm": 0.000598907470703125,
      "learning_rate": 0.011912038321315124,
      "loss": 0.2314,
      "num_input_tokens_seen": 23382016,
      "step": 110790
    },
    {
      "epoch": 12.188668866886688,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.011910629139733998,
      "loss": 0.2324,
      "num_input_tokens_seen": 23383040,
      "step": 110795
    },
    {
      "epoch": 12.18921892189219,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.011909219986626123,
      "loss": 0.2309,
      "num_input_tokens_seen": 23384160,
      "step": 110800
    },
    {
      "epoch": 12.189768976897689,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01190781086200449,
      "loss": 0.2293,
      "num_input_tokens_seen": 23385216,
      "step": 110805
    },
    {
      "epoch": 12.19031903190319,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011906401765882072,
      "loss": 0.2319,
      "num_input_tokens_seen": 23386240,
      "step": 110810
    },
    {
      "epoch": 12.190869086908691,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.011904992698271869,
      "loss": 0.233,
      "num_input_tokens_seen": 23387232,
      "step": 110815
    },
    {
      "epoch": 12.191419141914192,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.011903583659186865,
      "loss": 0.2319,
      "num_input_tokens_seen": 23388288,
      "step": 110820
    },
    {
      "epoch": 12.191969196919691,
      "grad_norm": 0.000774383544921875,
      "learning_rate": 0.011902174648640039,
      "loss": 0.2319,
      "num_input_tokens_seen": 23389312,
      "step": 110825
    },
    {
      "epoch": 12.192519251925193,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.011900765666644386,
      "loss": 0.2314,
      "num_input_tokens_seen": 23390400,
      "step": 110830
    },
    {
      "epoch": 12.193069306930694,
      "grad_norm": 0.00074005126953125,
      "learning_rate": 0.011899356713212886,
      "loss": 0.2304,
      "num_input_tokens_seen": 23391456,
      "step": 110835
    },
    {
      "epoch": 12.193619361936193,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.011897947788358534,
      "loss": 0.2314,
      "num_input_tokens_seen": 23392512,
      "step": 110840
    },
    {
      "epoch": 12.194169416941694,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011896538892094303,
      "loss": 0.2314,
      "num_input_tokens_seen": 23393600,
      "step": 110845
    },
    {
      "epoch": 12.194719471947195,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011895130024433181,
      "loss": 0.2309,
      "num_input_tokens_seen": 23394656,
      "step": 110850
    },
    {
      "epoch": 12.195269526952695,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.011893721185388166,
      "loss": 0.2319,
      "num_input_tokens_seen": 23395648,
      "step": 110855
    },
    {
      "epoch": 12.195819581958196,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011892312374972219,
      "loss": 0.2314,
      "num_input_tokens_seen": 23396704,
      "step": 110860
    },
    {
      "epoch": 12.196369636963697,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.01189090359319835,
      "loss": 0.2309,
      "num_input_tokens_seen": 23397792,
      "step": 110865
    },
    {
      "epoch": 12.196919691969198,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.011889494840079525,
      "loss": 0.2319,
      "num_input_tokens_seen": 23398848,
      "step": 110870
    },
    {
      "epoch": 12.197469746974697,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.011888086115628731,
      "loss": 0.2309,
      "num_input_tokens_seen": 23399840,
      "step": 110875
    },
    {
      "epoch": 12.198019801980198,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01188667741985896,
      "loss": 0.2298,
      "num_input_tokens_seen": 23400928,
      "step": 110880
    },
    {
      "epoch": 12.1985698569857,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.011885268752783184,
      "loss": 0.2293,
      "num_input_tokens_seen": 23402080,
      "step": 110885
    },
    {
      "epoch": 12.199119911991199,
      "grad_norm": 0.00064849853515625,
      "learning_rate": 0.011883860114414394,
      "loss": 0.2324,
      "num_input_tokens_seen": 23403072,
      "step": 110890
    },
    {
      "epoch": 12.1996699669967,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.011882451504765574,
      "loss": 0.2304,
      "num_input_tokens_seen": 23404128,
      "step": 110895
    },
    {
      "epoch": 12.2002200220022,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011881042923849694,
      "loss": 0.2304,
      "num_input_tokens_seen": 23405184,
      "step": 110900
    },
    {
      "epoch": 12.2007700770077,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01187963437167975,
      "loss": 0.2304,
      "num_input_tokens_seen": 23406240,
      "step": 110905
    },
    {
      "epoch": 12.201320132013201,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.011878225848268715,
      "loss": 0.2309,
      "num_input_tokens_seen": 23407296,
      "step": 110910
    },
    {
      "epoch": 12.201870187018702,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01187681735362958,
      "loss": 0.2314,
      "num_input_tokens_seen": 23408352,
      "step": 110915
    },
    {
      "epoch": 12.202420242024202,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.011875408887775319,
      "loss": 0.2298,
      "num_input_tokens_seen": 23409440,
      "step": 110920
    },
    {
      "epoch": 12.202970297029703,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011874000450718911,
      "loss": 0.2335,
      "num_input_tokens_seen": 23410496,
      "step": 110925
    },
    {
      "epoch": 12.203520352035204,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011872592042473344,
      "loss": 0.2314,
      "num_input_tokens_seen": 23411552,
      "step": 110930
    },
    {
      "epoch": 12.204070407040705,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.011871183663051591,
      "loss": 0.2303,
      "num_input_tokens_seen": 23412576,
      "step": 110935
    },
    {
      "epoch": 12.204620462046204,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011869775312466645,
      "loss": 0.2319,
      "num_input_tokens_seen": 23413600,
      "step": 110940
    },
    {
      "epoch": 12.205170517051705,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.011868366990731476,
      "loss": 0.2314,
      "num_input_tokens_seen": 23414624,
      "step": 110945
    },
    {
      "epoch": 12.205720572057206,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011866958697859059,
      "loss": 0.2335,
      "num_input_tokens_seen": 23415744,
      "step": 110950
    },
    {
      "epoch": 12.206270627062706,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.01186555043386239,
      "loss": 0.2319,
      "num_input_tokens_seen": 23416832,
      "step": 110955
    },
    {
      "epoch": 12.206820682068207,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01186414219875443,
      "loss": 0.2298,
      "num_input_tokens_seen": 23417856,
      "step": 110960
    },
    {
      "epoch": 12.207370737073708,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.011862733992548169,
      "loss": 0.2298,
      "num_input_tokens_seen": 23418912,
      "step": 110965
    },
    {
      "epoch": 12.207920792079207,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01186132581525659,
      "loss": 0.2319,
      "num_input_tokens_seen": 23419968,
      "step": 110970
    },
    {
      "epoch": 12.208470847084708,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.011859917666892653,
      "loss": 0.2304,
      "num_input_tokens_seen": 23421024,
      "step": 110975
    },
    {
      "epoch": 12.20902090209021,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.011858509547469358,
      "loss": 0.2314,
      "num_input_tokens_seen": 23422080,
      "step": 110980
    },
    {
      "epoch": 12.209570957095709,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.011857101456999667,
      "loss": 0.2319,
      "num_input_tokens_seen": 23423136,
      "step": 110985
    },
    {
      "epoch": 12.21012101210121,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011855693395496567,
      "loss": 0.2303,
      "num_input_tokens_seen": 23424192,
      "step": 110990
    },
    {
      "epoch": 12.210671067106711,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.011854285362973034,
      "loss": 0.2319,
      "num_input_tokens_seen": 23425248,
      "step": 110995
    },
    {
      "epoch": 12.211221122112212,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01185287735944204,
      "loss": 0.2293,
      "num_input_tokens_seen": 23426272,
      "step": 111000
    },
    {
      "epoch": 12.211771177117711,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.011851469384916566,
      "loss": 0.2308,
      "num_input_tokens_seen": 23427328,
      "step": 111005
    },
    {
      "epoch": 12.212321232123212,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01185006143940959,
      "loss": 0.2319,
      "num_input_tokens_seen": 23428416,
      "step": 111010
    },
    {
      "epoch": 12.212871287128714,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.011848653522934082,
      "loss": 0.2319,
      "num_input_tokens_seen": 23429472,
      "step": 111015
    },
    {
      "epoch": 12.213421342134213,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.011847245635503025,
      "loss": 0.2329,
      "num_input_tokens_seen": 23430528,
      "step": 111020
    },
    {
      "epoch": 12.213971397139714,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.011845837777129388,
      "loss": 0.2288,
      "num_input_tokens_seen": 23431616,
      "step": 111025
    },
    {
      "epoch": 12.214521452145215,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.01184442994782616,
      "loss": 0.2299,
      "num_input_tokens_seen": 23432640,
      "step": 111030
    },
    {
      "epoch": 12.215071507150714,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0118430221476063,
      "loss": 0.2314,
      "num_input_tokens_seen": 23433696,
      "step": 111035
    },
    {
      "epoch": 12.215621562156215,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.01184161437648279,
      "loss": 0.2319,
      "num_input_tokens_seen": 23434752,
      "step": 111040
    },
    {
      "epoch": 12.216171617161717,
      "grad_norm": 0.002593994140625,
      "learning_rate": 0.011840206634468605,
      "loss": 0.2314,
      "num_input_tokens_seen": 23435776,
      "step": 111045
    },
    {
      "epoch": 12.216721672167218,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.011838798921576717,
      "loss": 0.233,
      "num_input_tokens_seen": 23436896,
      "step": 111050
    },
    {
      "epoch": 12.217271727172717,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.01183739123782011,
      "loss": 0.2335,
      "num_input_tokens_seen": 23438048,
      "step": 111055
    },
    {
      "epoch": 12.217821782178218,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.011835983583211746,
      "loss": 0.2319,
      "num_input_tokens_seen": 23439072,
      "step": 111060
    },
    {
      "epoch": 12.218371837183719,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.011834575957764602,
      "loss": 0.2304,
      "num_input_tokens_seen": 23440160,
      "step": 111065
    },
    {
      "epoch": 12.218921892189218,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.011833168361491658,
      "loss": 0.2309,
      "num_input_tokens_seen": 23441248,
      "step": 111070
    },
    {
      "epoch": 12.21947194719472,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011831760794405873,
      "loss": 0.2324,
      "num_input_tokens_seen": 23442304,
      "step": 111075
    },
    {
      "epoch": 12.22002200220022,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.011830353256520234,
      "loss": 0.2324,
      "num_input_tokens_seen": 23443296,
      "step": 111080
    },
    {
      "epoch": 12.22057205720572,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01182894574784771,
      "loss": 0.2298,
      "num_input_tokens_seen": 23444352,
      "step": 111085
    },
    {
      "epoch": 12.221122112211221,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.011827538268401265,
      "loss": 0.2293,
      "num_input_tokens_seen": 23445408,
      "step": 111090
    },
    {
      "epoch": 12.221672167216722,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.011826130818193881,
      "loss": 0.2298,
      "num_input_tokens_seen": 23446496,
      "step": 111095
    },
    {
      "epoch": 12.222222222222221,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011824723397238524,
      "loss": 0.2309,
      "num_input_tokens_seen": 23447552,
      "step": 111100
    },
    {
      "epoch": 12.222772277227723,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.01182331600554817,
      "loss": 0.2314,
      "num_input_tokens_seen": 23448576,
      "step": 111105
    },
    {
      "epoch": 12.223322332233224,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011821908643135792,
      "loss": 0.2309,
      "num_input_tokens_seen": 23449600,
      "step": 111110
    },
    {
      "epoch": 12.223872387238725,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.01182050131001435,
      "loss": 0.2319,
      "num_input_tokens_seen": 23450624,
      "step": 111115
    },
    {
      "epoch": 12.224422442244224,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.011819094006196823,
      "loss": 0.233,
      "num_input_tokens_seen": 23451680,
      "step": 111120
    },
    {
      "epoch": 12.224972497249725,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01181768673169618,
      "loss": 0.2319,
      "num_input_tokens_seen": 23452800,
      "step": 111125
    },
    {
      "epoch": 12.225522552255226,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.011816279486525397,
      "loss": 0.2309,
      "num_input_tokens_seen": 23453856,
      "step": 111130
    },
    {
      "epoch": 12.226072607260726,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.011814872270697436,
      "loss": 0.2304,
      "num_input_tokens_seen": 23454944,
      "step": 111135
    },
    {
      "epoch": 12.226622662266227,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.011813465084225264,
      "loss": 0.2298,
      "num_input_tokens_seen": 23456032,
      "step": 111140
    },
    {
      "epoch": 12.227172717271728,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.011812057927121863,
      "loss": 0.2303,
      "num_input_tokens_seen": 23457152,
      "step": 111145
    },
    {
      "epoch": 12.227722772277227,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.011810650799400187,
      "loss": 0.2309,
      "num_input_tokens_seen": 23458240,
      "step": 111150
    },
    {
      "epoch": 12.228272827282728,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.011809243701073216,
      "loss": 0.2309,
      "num_input_tokens_seen": 23459232,
      "step": 111155
    },
    {
      "epoch": 12.22882288228823,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.011807836632153913,
      "loss": 0.2314,
      "num_input_tokens_seen": 23460288,
      "step": 111160
    },
    {
      "epoch": 12.229372937293729,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.011806429592655248,
      "loss": 0.2309,
      "num_input_tokens_seen": 23461408,
      "step": 111165
    },
    {
      "epoch": 12.22992299229923,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011805022582590193,
      "loss": 0.2314,
      "num_input_tokens_seen": 23462464,
      "step": 111170
    },
    {
      "epoch": 12.23047304730473,
      "grad_norm": 0.0026702880859375,
      "learning_rate": 0.011803615601971705,
      "loss": 0.2309,
      "num_input_tokens_seen": 23463520,
      "step": 111175
    },
    {
      "epoch": 12.231023102310232,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.011802208650812763,
      "loss": 0.2304,
      "num_input_tokens_seen": 23464608,
      "step": 111180
    },
    {
      "epoch": 12.231573157315731,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011800801729126334,
      "loss": 0.2319,
      "num_input_tokens_seen": 23465568,
      "step": 111185
    },
    {
      "epoch": 12.232123212321232,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.01179939483692537,
      "loss": 0.2319,
      "num_input_tokens_seen": 23466624,
      "step": 111190
    },
    {
      "epoch": 12.232673267326733,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011797987974222853,
      "loss": 0.2319,
      "num_input_tokens_seen": 23467680,
      "step": 111195
    },
    {
      "epoch": 12.233223322332233,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.011796581141031742,
      "loss": 0.2309,
      "num_input_tokens_seen": 23468768,
      "step": 111200
    },
    {
      "epoch": 12.233773377337734,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011795174337365011,
      "loss": 0.233,
      "num_input_tokens_seen": 23469856,
      "step": 111205
    },
    {
      "epoch": 12.234323432343235,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.011793767563235618,
      "loss": 0.2319,
      "num_input_tokens_seen": 23470880,
      "step": 111210
    },
    {
      "epoch": 12.234873487348734,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011792360818656526,
      "loss": 0.2329,
      "num_input_tokens_seen": 23471872,
      "step": 111215
    },
    {
      "epoch": 12.235423542354235,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01179095410364071,
      "loss": 0.2319,
      "num_input_tokens_seen": 23472864,
      "step": 111220
    },
    {
      "epoch": 12.235973597359736,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.011789547418201133,
      "loss": 0.2319,
      "num_input_tokens_seen": 23473856,
      "step": 111225
    },
    {
      "epoch": 12.236523652365236,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.01178814076235075,
      "loss": 0.2303,
      "num_input_tokens_seen": 23474976,
      "step": 111230
    },
    {
      "epoch": 12.237073707370737,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.011786734136102536,
      "loss": 0.2324,
      "num_input_tokens_seen": 23476032,
      "step": 111235
    },
    {
      "epoch": 12.237623762376238,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.011785327539469448,
      "loss": 0.2309,
      "num_input_tokens_seen": 23477056,
      "step": 111240
    },
    {
      "epoch": 12.238173817381739,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01178392097246446,
      "loss": 0.2303,
      "num_input_tokens_seen": 23478144,
      "step": 111245
    },
    {
      "epoch": 12.238723872387238,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011782514435100526,
      "loss": 0.2329,
      "num_input_tokens_seen": 23479264,
      "step": 111250
    },
    {
      "epoch": 12.23927392739274,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01178110792739061,
      "loss": 0.2304,
      "num_input_tokens_seen": 23480320,
      "step": 111255
    },
    {
      "epoch": 12.23982398239824,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011779701449347684,
      "loss": 0.2303,
      "num_input_tokens_seen": 23481408,
      "step": 111260
    },
    {
      "epoch": 12.24037403740374,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.011778295000984696,
      "loss": 0.2329,
      "num_input_tokens_seen": 23482496,
      "step": 111265
    },
    {
      "epoch": 12.24092409240924,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.011776888582314623,
      "loss": 0.2319,
      "num_input_tokens_seen": 23483552,
      "step": 111270
    },
    {
      "epoch": 12.241474147414742,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011775482193350423,
      "loss": 0.2319,
      "num_input_tokens_seen": 23484640,
      "step": 111275
    },
    {
      "epoch": 12.242024202420241,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.011774075834105047,
      "loss": 0.2319,
      "num_input_tokens_seen": 23485728,
      "step": 111280
    },
    {
      "epoch": 12.242574257425742,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011772669504591476,
      "loss": 0.2314,
      "num_input_tokens_seen": 23486752,
      "step": 111285
    },
    {
      "epoch": 12.243124312431243,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.011771263204822656,
      "loss": 0.2324,
      "num_input_tokens_seen": 23487872,
      "step": 111290
    },
    {
      "epoch": 12.243674367436745,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011769856934811557,
      "loss": 0.2335,
      "num_input_tokens_seen": 23488896,
      "step": 111295
    },
    {
      "epoch": 12.244224422442244,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011768450694571138,
      "loss": 0.2319,
      "num_input_tokens_seen": 23489920,
      "step": 111300
    },
    {
      "epoch": 12.244774477447745,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.011767044484114352,
      "loss": 0.2324,
      "num_input_tokens_seen": 23491008,
      "step": 111305
    },
    {
      "epoch": 12.245324532453246,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.011765638303454168,
      "loss": 0.2324,
      "num_input_tokens_seen": 23492128,
      "step": 111310
    },
    {
      "epoch": 12.245874587458745,
      "grad_norm": 0.000782012939453125,
      "learning_rate": 0.011764232152603544,
      "loss": 0.2298,
      "num_input_tokens_seen": 23493184,
      "step": 111315
    },
    {
      "epoch": 12.246424642464246,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.011762826031575444,
      "loss": 0.2314,
      "num_input_tokens_seen": 23494176,
      "step": 111320
    },
    {
      "epoch": 12.246974697469748,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.011761419940382819,
      "loss": 0.2293,
      "num_input_tokens_seen": 23495200,
      "step": 111325
    },
    {
      "epoch": 12.247524752475247,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011760013879038632,
      "loss": 0.2308,
      "num_input_tokens_seen": 23496256,
      "step": 111330
    },
    {
      "epoch": 12.248074807480748,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.011758607847555846,
      "loss": 0.2308,
      "num_input_tokens_seen": 23497280,
      "step": 111335
    },
    {
      "epoch": 12.248624862486249,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011757201845947406,
      "loss": 0.2314,
      "num_input_tokens_seen": 23498304,
      "step": 111340
    },
    {
      "epoch": 12.249174917491748,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.011755795874226293,
      "loss": 0.2324,
      "num_input_tokens_seen": 23499392,
      "step": 111345
    },
    {
      "epoch": 12.24972497249725,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011754389932405448,
      "loss": 0.234,
      "num_input_tokens_seen": 23500384,
      "step": 111350
    },
    {
      "epoch": 12.25027502750275,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.01175298402049783,
      "loss": 0.2319,
      "num_input_tokens_seen": 23501504,
      "step": 111355
    },
    {
      "epoch": 12.250825082508252,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.011751578138516409,
      "loss": 0.2329,
      "num_input_tokens_seen": 23502528,
      "step": 111360
    },
    {
      "epoch": 12.251375137513751,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.011750172286474125,
      "loss": 0.234,
      "num_input_tokens_seen": 23503584,
      "step": 111365
    },
    {
      "epoch": 12.251925192519252,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011748766464383945,
      "loss": 0.2308,
      "num_input_tokens_seen": 23504640,
      "step": 111370
    },
    {
      "epoch": 12.252475247524753,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.01174736067225883,
      "loss": 0.2314,
      "num_input_tokens_seen": 23505664,
      "step": 111375
    },
    {
      "epoch": 12.253025302530252,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.011745954910111725,
      "loss": 0.2314,
      "num_input_tokens_seen": 23506752,
      "step": 111380
    },
    {
      "epoch": 12.253575357535754,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.011744549177955595,
      "loss": 0.2298,
      "num_input_tokens_seen": 23507744,
      "step": 111385
    },
    {
      "epoch": 12.254125412541255,
      "grad_norm": 0.000492095947265625,
      "learning_rate": 0.01174314347580339,
      "loss": 0.2329,
      "num_input_tokens_seen": 23508768,
      "step": 111390
    },
    {
      "epoch": 12.254675467546754,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011741737803668075,
      "loss": 0.2324,
      "num_input_tokens_seen": 23509856,
      "step": 111395
    },
    {
      "epoch": 12.255225522552255,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.011740332161562594,
      "loss": 0.2324,
      "num_input_tokens_seen": 23510912,
      "step": 111400
    },
    {
      "epoch": 12.255775577557756,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.011738926549499908,
      "loss": 0.2314,
      "num_input_tokens_seen": 23511968,
      "step": 111405
    },
    {
      "epoch": 12.256325632563255,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01173752096749297,
      "loss": 0.2309,
      "num_input_tokens_seen": 23512992,
      "step": 111410
    },
    {
      "epoch": 12.256875687568757,
      "grad_norm": 0.000701904296875,
      "learning_rate": 0.011736115415554743,
      "loss": 0.2314,
      "num_input_tokens_seen": 23513984,
      "step": 111415
    },
    {
      "epoch": 12.257425742574258,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011734709893698165,
      "loss": 0.2293,
      "num_input_tokens_seen": 23515072,
      "step": 111420
    },
    {
      "epoch": 12.257975797579759,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011733304401936203,
      "loss": 0.2324,
      "num_input_tokens_seen": 23516128,
      "step": 111425
    },
    {
      "epoch": 12.258525852585258,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.011731898940281805,
      "loss": 0.2314,
      "num_input_tokens_seen": 23517216,
      "step": 111430
    },
    {
      "epoch": 12.25907590759076,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.01173049350874793,
      "loss": 0.2314,
      "num_input_tokens_seen": 23518240,
      "step": 111435
    },
    {
      "epoch": 12.25962596259626,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011729088107347527,
      "loss": 0.2319,
      "num_input_tokens_seen": 23519264,
      "step": 111440
    },
    {
      "epoch": 12.26017601760176,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011727682736093543,
      "loss": 0.2319,
      "num_input_tokens_seen": 23520288,
      "step": 111445
    },
    {
      "epoch": 12.26072607260726,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.011726277394998947,
      "loss": 0.2309,
      "num_input_tokens_seen": 23521344,
      "step": 111450
    },
    {
      "epoch": 12.261276127612762,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011724872084076672,
      "loss": 0.2298,
      "num_input_tokens_seen": 23522400,
      "step": 111455
    },
    {
      "epoch": 12.261826182618261,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.011723466803339683,
      "loss": 0.2319,
      "num_input_tokens_seen": 23523424,
      "step": 111460
    },
    {
      "epoch": 12.262376237623762,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.01172206155280093,
      "loss": 0.2314,
      "num_input_tokens_seen": 23524512,
      "step": 111465
    },
    {
      "epoch": 12.262926292629263,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.011720656332473357,
      "loss": 0.2308,
      "num_input_tokens_seen": 23525504,
      "step": 111470
    },
    {
      "epoch": 12.263476347634764,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011719251142369932,
      "loss": 0.2298,
      "num_input_tokens_seen": 23526528,
      "step": 111475
    },
    {
      "epoch": 12.264026402640264,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.011717845982503585,
      "loss": 0.233,
      "num_input_tokens_seen": 23527616,
      "step": 111480
    },
    {
      "epoch": 12.264576457645765,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.01171644085288728,
      "loss": 0.2314,
      "num_input_tokens_seen": 23528640,
      "step": 111485
    },
    {
      "epoch": 12.265126512651266,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011715035753533967,
      "loss": 0.2324,
      "num_input_tokens_seen": 23529696,
      "step": 111490
    },
    {
      "epoch": 12.265676567656765,
      "grad_norm": 0.00070953369140625,
      "learning_rate": 0.011713630684456588,
      "loss": 0.2314,
      "num_input_tokens_seen": 23530720,
      "step": 111495
    },
    {
      "epoch": 12.266226622662266,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0117122256456681,
      "loss": 0.2298,
      "num_input_tokens_seen": 23531776,
      "step": 111500
    },
    {
      "epoch": 12.266776677667767,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.011710820637181448,
      "loss": 0.2314,
      "num_input_tokens_seen": 23532832,
      "step": 111505
    },
    {
      "epoch": 12.267326732673267,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.01170941565900959,
      "loss": 0.2329,
      "num_input_tokens_seen": 23533856,
      "step": 111510
    },
    {
      "epoch": 12.267876787678768,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.011708010711165466,
      "loss": 0.2309,
      "num_input_tokens_seen": 23534880,
      "step": 111515
    },
    {
      "epoch": 12.268426842684269,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.011706605793662025,
      "loss": 0.2329,
      "num_input_tokens_seen": 23535904,
      "step": 111520
    },
    {
      "epoch": 12.268976897689768,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.011705200906512222,
      "loss": 0.2324,
      "num_input_tokens_seen": 23536928,
      "step": 111525
    },
    {
      "epoch": 12.26952695269527,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.011703796049728996,
      "loss": 0.2314,
      "num_input_tokens_seen": 23537920,
      "step": 111530
    },
    {
      "epoch": 12.27007700770077,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.011702391223325309,
      "loss": 0.2319,
      "num_input_tokens_seen": 23539040,
      "step": 111535
    },
    {
      "epoch": 12.270627062706271,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.011700986427314094,
      "loss": 0.2309,
      "num_input_tokens_seen": 23540096,
      "step": 111540
    },
    {
      "epoch": 12.27117711771177,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.011699581661708302,
      "loss": 0.2319,
      "num_input_tokens_seen": 23541088,
      "step": 111545
    },
    {
      "epoch": 12.271727172717272,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011698176926520887,
      "loss": 0.2329,
      "num_input_tokens_seen": 23542112,
      "step": 111550
    },
    {
      "epoch": 12.272277227722773,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.011696772221764786,
      "loss": 0.2319,
      "num_input_tokens_seen": 23543168,
      "step": 111555
    },
    {
      "epoch": 12.272827282728272,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.011695367547452953,
      "loss": 0.2313,
      "num_input_tokens_seen": 23544224,
      "step": 111560
    },
    {
      "epoch": 12.273377337733773,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.011693962903598337,
      "loss": 0.2319,
      "num_input_tokens_seen": 23545248,
      "step": 111565
    },
    {
      "epoch": 12.273927392739274,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.011692558290213869,
      "loss": 0.2319,
      "num_input_tokens_seen": 23546240,
      "step": 111570
    },
    {
      "epoch": 12.274477447744774,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.011691153707312508,
      "loss": 0.2293,
      "num_input_tokens_seen": 23547328,
      "step": 111575
    },
    {
      "epoch": 12.275027502750275,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.011689749154907194,
      "loss": 0.2309,
      "num_input_tokens_seen": 23548352,
      "step": 111580
    },
    {
      "epoch": 12.275577557755776,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011688344633010878,
      "loss": 0.2303,
      "num_input_tokens_seen": 23549408,
      "step": 111585
    },
    {
      "epoch": 12.276127612761275,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.0116869401416365,
      "loss": 0.2319,
      "num_input_tokens_seen": 23550432,
      "step": 111590
    },
    {
      "epoch": 12.276677667766776,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.011685535680797,
      "loss": 0.2319,
      "num_input_tokens_seen": 23551392,
      "step": 111595
    },
    {
      "epoch": 12.277227722772277,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01168413125050533,
      "loss": 0.2314,
      "num_input_tokens_seen": 23552416,
      "step": 111600
    },
    {
      "epoch": 12.277777777777779,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.011682726850774436,
      "loss": 0.2303,
      "num_input_tokens_seen": 23553536,
      "step": 111605
    },
    {
      "epoch": 12.278327832783278,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 0.011681322481617249,
      "loss": 0.2319,
      "num_input_tokens_seen": 23554528,
      "step": 111610
    },
    {
      "epoch": 12.278877887788779,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.011679918143046723,
      "loss": 0.2309,
      "num_input_tokens_seen": 23555616,
      "step": 111615
    },
    {
      "epoch": 12.27942794279428,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011678513835075796,
      "loss": 0.2309,
      "num_input_tokens_seen": 23556704,
      "step": 111620
    },
    {
      "epoch": 12.27997799779978,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.011677109557717419,
      "loss": 0.2293,
      "num_input_tokens_seen": 23557824,
      "step": 111625
    },
    {
      "epoch": 12.28052805280528,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.011675705310984526,
      "loss": 0.2303,
      "num_input_tokens_seen": 23558912,
      "step": 111630
    },
    {
      "epoch": 12.281078107810782,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.011674301094890059,
      "loss": 0.2308,
      "num_input_tokens_seen": 23559936,
      "step": 111635
    },
    {
      "epoch": 12.281628162816281,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.01167289690944697,
      "loss": 0.2319,
      "num_input_tokens_seen": 23560928,
      "step": 111640
    },
    {
      "epoch": 12.282178217821782,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.011671492754668182,
      "loss": 0.2298,
      "num_input_tokens_seen": 23561952,
      "step": 111645
    },
    {
      "epoch": 12.282728272827283,
      "grad_norm": 0.000652313232421875,
      "learning_rate": 0.01167008863056666,
      "loss": 0.2329,
      "num_input_tokens_seen": 23563008,
      "step": 111650
    },
    {
      "epoch": 12.283278327832782,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01166868453715533,
      "loss": 0.2303,
      "num_input_tokens_seen": 23564032,
      "step": 111655
    },
    {
      "epoch": 12.283828382838283,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011667280474447131,
      "loss": 0.2308,
      "num_input_tokens_seen": 23565120,
      "step": 111660
    },
    {
      "epoch": 12.284378437843785,
      "grad_norm": 0.00066375732421875,
      "learning_rate": 0.011665876442455016,
      "loss": 0.2319,
      "num_input_tokens_seen": 23566112,
      "step": 111665
    },
    {
      "epoch": 12.284928492849286,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01166447244119191,
      "loss": 0.2303,
      "num_input_tokens_seen": 23567200,
      "step": 111670
    },
    {
      "epoch": 12.285478547854785,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.011663068470670765,
      "loss": 0.2314,
      "num_input_tokens_seen": 23568256,
      "step": 111675
    },
    {
      "epoch": 12.286028602860286,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.011661664530904522,
      "loss": 0.2319,
      "num_input_tokens_seen": 23569312,
      "step": 111680
    },
    {
      "epoch": 12.286578657865787,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011660260621906107,
      "loss": 0.2324,
      "num_input_tokens_seen": 23570400,
      "step": 111685
    },
    {
      "epoch": 12.287128712871286,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01165885674368847,
      "loss": 0.2329,
      "num_input_tokens_seen": 23571488,
      "step": 111690
    },
    {
      "epoch": 12.287678767876788,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.011657452896264546,
      "loss": 0.2329,
      "num_input_tokens_seen": 23572512,
      "step": 111695
    },
    {
      "epoch": 12.288228822882289,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.01165604907964728,
      "loss": 0.2309,
      "num_input_tokens_seen": 23573600,
      "step": 111700
    },
    {
      "epoch": 12.288778877887788,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.0116546452938496,
      "loss": 0.2303,
      "num_input_tokens_seen": 23574688,
      "step": 111705
    },
    {
      "epoch": 12.289328932893289,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01165324153888445,
      "loss": 0.2298,
      "num_input_tokens_seen": 23575744,
      "step": 111710
    },
    {
      "epoch": 12.28987898789879,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.011651837814764766,
      "loss": 0.2303,
      "num_input_tokens_seen": 23576736,
      "step": 111715
    },
    {
      "epoch": 12.290429042904291,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.011650434121503484,
      "loss": 0.2308,
      "num_input_tokens_seen": 23577760,
      "step": 111720
    },
    {
      "epoch": 12.29097909790979,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.01164903045911355,
      "loss": 0.2303,
      "num_input_tokens_seen": 23578784,
      "step": 111725
    },
    {
      "epoch": 12.291529152915292,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.011647626827607893,
      "loss": 0.2319,
      "num_input_tokens_seen": 23579808,
      "step": 111730
    },
    {
      "epoch": 12.292079207920793,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.011646223226999446,
      "loss": 0.2314,
      "num_input_tokens_seen": 23580832,
      "step": 111735
    },
    {
      "epoch": 12.292629262926292,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.011644819657301157,
      "loss": 0.2319,
      "num_input_tokens_seen": 23581856,
      "step": 111740
    },
    {
      "epoch": 12.293179317931793,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.01164341611852595,
      "loss": 0.2288,
      "num_input_tokens_seen": 23582880,
      "step": 111745
    },
    {
      "epoch": 12.293729372937294,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.011642012610686767,
      "loss": 0.2314,
      "num_input_tokens_seen": 23583968,
      "step": 111750
    },
    {
      "epoch": 12.294279427942794,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.011640609133796546,
      "loss": 0.2319,
      "num_input_tokens_seen": 23584992,
      "step": 111755
    },
    {
      "epoch": 12.294829482948295,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.01163920568786821,
      "loss": 0.2293,
      "num_input_tokens_seen": 23585952,
      "step": 111760
    },
    {
      "epoch": 12.295379537953796,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011637802272914713,
      "loss": 0.2303,
      "num_input_tokens_seen": 23587008,
      "step": 111765
    },
    {
      "epoch": 12.295929592959295,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.011636398888948971,
      "loss": 0.2319,
      "num_input_tokens_seen": 23588064,
      "step": 111770
    },
    {
      "epoch": 12.296479647964796,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.01163499553598393,
      "loss": 0.2324,
      "num_input_tokens_seen": 23589152,
      "step": 111775
    },
    {
      "epoch": 12.297029702970297,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.011633592214032524,
      "loss": 0.2314,
      "num_input_tokens_seen": 23590144,
      "step": 111780
    },
    {
      "epoch": 12.297579757975798,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.011632188923107675,
      "loss": 0.2319,
      "num_input_tokens_seen": 23591232,
      "step": 111785
    },
    {
      "epoch": 12.298129812981298,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.011630785663222329,
      "loss": 0.2314,
      "num_input_tokens_seen": 23592320,
      "step": 111790
    },
    {
      "epoch": 12.298679867986799,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.01162938243438942,
      "loss": 0.2303,
      "num_input_tokens_seen": 23593312,
      "step": 111795
    },
    {
      "epoch": 12.2992299229923,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.011627979236621868,
      "loss": 0.2309,
      "num_input_tokens_seen": 23594368,
      "step": 111800
    },
    {
      "epoch": 12.2997799779978,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.011626576069932615,
      "loss": 0.2298,
      "num_input_tokens_seen": 23595392,
      "step": 111805
    },
    {
      "epoch": 12.3003300330033,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01162517293433459,
      "loss": 0.2314,
      "num_input_tokens_seen": 23596480,
      "step": 111810
    },
    {
      "epoch": 12.300880088008801,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.011623769829840733,
      "loss": 0.2324,
      "num_input_tokens_seen": 23597536,
      "step": 111815
    },
    {
      "epoch": 12.3014301430143,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.011622366756463966,
      "loss": 0.2293,
      "num_input_tokens_seen": 23598624,
      "step": 111820
    },
    {
      "epoch": 12.301980198019802,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.01162096371421722,
      "loss": 0.2314,
      "num_input_tokens_seen": 23599680,
      "step": 111825
    },
    {
      "epoch": 12.302530253025303,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011619560703113435,
      "loss": 0.2314,
      "num_input_tokens_seen": 23600672,
      "step": 111830
    },
    {
      "epoch": 12.303080308030804,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.011618157723165531,
      "loss": 0.2313,
      "num_input_tokens_seen": 23601728,
      "step": 111835
    },
    {
      "epoch": 12.303630363036303,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011616754774386454,
      "loss": 0.2329,
      "num_input_tokens_seen": 23602752,
      "step": 111840
    },
    {
      "epoch": 12.304180418041804,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.011615351856789118,
      "loss": 0.2293,
      "num_input_tokens_seen": 23603808,
      "step": 111845
    },
    {
      "epoch": 12.304730473047305,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01161394897038646,
      "loss": 0.2283,
      "num_input_tokens_seen": 23604960,
      "step": 111850
    },
    {
      "epoch": 12.305280528052805,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.011612546115191415,
      "loss": 0.2308,
      "num_input_tokens_seen": 23606048,
      "step": 111855
    },
    {
      "epoch": 12.305830583058306,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0116111432912169,
      "loss": 0.2309,
      "num_input_tokens_seen": 23607104,
      "step": 111860
    },
    {
      "epoch": 12.306380638063807,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011609740498475855,
      "loss": 0.2303,
      "num_input_tokens_seen": 23608128,
      "step": 111865
    },
    {
      "epoch": 12.306930693069306,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011608337736981207,
      "loss": 0.2314,
      "num_input_tokens_seen": 23609216,
      "step": 111870
    },
    {
      "epoch": 12.307480748074807,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.011606935006745879,
      "loss": 0.2314,
      "num_input_tokens_seen": 23610240,
      "step": 111875
    },
    {
      "epoch": 12.308030803080309,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011605532307782803,
      "loss": 0.2308,
      "num_input_tokens_seen": 23611296,
      "step": 111880
    },
    {
      "epoch": 12.308580858085808,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.011604129640104904,
      "loss": 0.2319,
      "num_input_tokens_seen": 23612384,
      "step": 111885
    },
    {
      "epoch": 12.309130913091309,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.011602727003725116,
      "loss": 0.2308,
      "num_input_tokens_seen": 23613408,
      "step": 111890
    },
    {
      "epoch": 12.30968096809681,
      "grad_norm": 0.005859375,
      "learning_rate": 0.011601324398656368,
      "loss": 0.2288,
      "num_input_tokens_seen": 23614464,
      "step": 111895
    },
    {
      "epoch": 12.310231023102311,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.011599921824911576,
      "loss": 0.2319,
      "num_input_tokens_seen": 23615520,
      "step": 111900
    },
    {
      "epoch": 12.31078107810781,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.011598519282503673,
      "loss": 0.2324,
      "num_input_tokens_seen": 23616544,
      "step": 111905
    },
    {
      "epoch": 12.311331133113312,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.011597116771445584,
      "loss": 0.2309,
      "num_input_tokens_seen": 23617568,
      "step": 111910
    },
    {
      "epoch": 12.311881188118813,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.011595714291750244,
      "loss": 0.2319,
      "num_input_tokens_seen": 23618592,
      "step": 111915
    },
    {
      "epoch": 12.312431243124312,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.011594311843430568,
      "loss": 0.2314,
      "num_input_tokens_seen": 23619648,
      "step": 111920
    },
    {
      "epoch": 12.312981298129813,
      "grad_norm": 0.000732421875,
      "learning_rate": 0.011592909426499482,
      "loss": 0.2329,
      "num_input_tokens_seen": 23620672,
      "step": 111925
    },
    {
      "epoch": 12.313531353135314,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.011591507040969923,
      "loss": 0.2293,
      "num_input_tokens_seen": 23621696,
      "step": 111930
    },
    {
      "epoch": 12.314081408140813,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0115901046868548,
      "loss": 0.2308,
      "num_input_tokens_seen": 23622816,
      "step": 111935
    },
    {
      "epoch": 12.314631463146315,
      "grad_norm": 0.0004596710205078125,
      "learning_rate": 0.011588702364167048,
      "loss": 0.2298,
      "num_input_tokens_seen": 23623904,
      "step": 111940
    },
    {
      "epoch": 12.315181518151816,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.011587300072919594,
      "loss": 0.2324,
      "num_input_tokens_seen": 23624992,
      "step": 111945
    },
    {
      "epoch": 12.315731573157315,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.011585897813125351,
      "loss": 0.2303,
      "num_input_tokens_seen": 23626048,
      "step": 111950
    },
    {
      "epoch": 12.316281628162816,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.011584495584797256,
      "loss": 0.2319,
      "num_input_tokens_seen": 23627072,
      "step": 111955
    },
    {
      "epoch": 12.316831683168317,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01158309338794822,
      "loss": 0.2304,
      "num_input_tokens_seen": 23628192,
      "step": 111960
    },
    {
      "epoch": 12.317381738173818,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.011581691222591175,
      "loss": 0.2314,
      "num_input_tokens_seen": 23629216,
      "step": 111965
    },
    {
      "epoch": 12.317931793179318,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011580289088739048,
      "loss": 0.2324,
      "num_input_tokens_seen": 23630208,
      "step": 111970
    },
    {
      "epoch": 12.318481848184819,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.011578886986404744,
      "loss": 0.2319,
      "num_input_tokens_seen": 23631200,
      "step": 111975
    },
    {
      "epoch": 12.31903190319032,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.011577484915601205,
      "loss": 0.2319,
      "num_input_tokens_seen": 23632224,
      "step": 111980
    },
    {
      "epoch": 12.319581958195819,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.011576082876341346,
      "loss": 0.2308,
      "num_input_tokens_seen": 23633280,
      "step": 111985
    },
    {
      "epoch": 12.32013201320132,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.011574680868638084,
      "loss": 0.2309,
      "num_input_tokens_seen": 23634304,
      "step": 111990
    },
    {
      "epoch": 12.320682068206821,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.011573278892504346,
      "loss": 0.2308,
      "num_input_tokens_seen": 23635360,
      "step": 111995
    },
    {
      "epoch": 12.32123212321232,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.01157187694795305,
      "loss": 0.2324,
      "num_input_tokens_seen": 23636416,
      "step": 112000
    },
    {
      "epoch": 12.321782178217822,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.011570475034997123,
      "loss": 0.2324,
      "num_input_tokens_seen": 23637536,
      "step": 112005
    },
    {
      "epoch": 12.322332233223323,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.011569073153649483,
      "loss": 0.2314,
      "num_input_tokens_seen": 23638560,
      "step": 112010
    },
    {
      "epoch": 12.322882288228822,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.011567671303923045,
      "loss": 0.2309,
      "num_input_tokens_seen": 23639552,
      "step": 112015
    },
    {
      "epoch": 12.323432343234323,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011566269485830736,
      "loss": 0.2308,
      "num_input_tokens_seen": 23640544,
      "step": 112020
    },
    {
      "epoch": 12.323982398239824,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.01156486769938547,
      "loss": 0.2314,
      "num_input_tokens_seen": 23641632,
      "step": 112025
    },
    {
      "epoch": 12.324532453245325,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.011563465944600176,
      "loss": 0.2319,
      "num_input_tokens_seen": 23642720,
      "step": 112030
    },
    {
      "epoch": 12.325082508250825,
      "grad_norm": 0.000614166259765625,
      "learning_rate": 0.011562064221487764,
      "loss": 0.2303,
      "num_input_tokens_seen": 23643776,
      "step": 112035
    },
    {
      "epoch": 12.325632563256326,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.011560662530061156,
      "loss": 0.2309,
      "num_input_tokens_seen": 23644864,
      "step": 112040
    },
    {
      "epoch": 12.326182618261827,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011559260870333274,
      "loss": 0.2335,
      "num_input_tokens_seen": 23645984,
      "step": 112045
    },
    {
      "epoch": 12.326732673267326,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01155785924231703,
      "loss": 0.2319,
      "num_input_tokens_seen": 23647104,
      "step": 112050
    },
    {
      "epoch": 12.327282728272827,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.011556457646025347,
      "loss": 0.2319,
      "num_input_tokens_seen": 23648128,
      "step": 112055
    },
    {
      "epoch": 12.327832783278328,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.011555056081471147,
      "loss": 0.2324,
      "num_input_tokens_seen": 23649184,
      "step": 112060
    },
    {
      "epoch": 12.328382838283828,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01155365454866733,
      "loss": 0.2308,
      "num_input_tokens_seen": 23650176,
      "step": 112065
    },
    {
      "epoch": 12.328932893289329,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.011552253047626835,
      "loss": 0.2314,
      "num_input_tokens_seen": 23651200,
      "step": 112070
    },
    {
      "epoch": 12.32948294829483,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.011550851578362564,
      "loss": 0.2308,
      "num_input_tokens_seen": 23652288,
      "step": 112075
    },
    {
      "epoch": 12.33003300330033,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011549450140887442,
      "loss": 0.2314,
      "num_input_tokens_seen": 23653312,
      "step": 112080
    },
    {
      "epoch": 12.33058305830583,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.011548048735214384,
      "loss": 0.2319,
      "num_input_tokens_seen": 23654368,
      "step": 112085
    },
    {
      "epoch": 12.331133113311331,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.0115466473613563,
      "loss": 0.2314,
      "num_input_tokens_seen": 23655456,
      "step": 112090
    },
    {
      "epoch": 12.331683168316832,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.011545246019326112,
      "loss": 0.2329,
      "num_input_tokens_seen": 23656512,
      "step": 112095
    },
    {
      "epoch": 12.332233223322332,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011543844709136731,
      "loss": 0.2324,
      "num_input_tokens_seen": 23657632,
      "step": 112100
    },
    {
      "epoch": 12.332783278327833,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011542443430801082,
      "loss": 0.2329,
      "num_input_tokens_seen": 23658656,
      "step": 112105
    },
    {
      "epoch": 12.333333333333334,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.011541042184332068,
      "loss": 0.233,
      "num_input_tokens_seen": 23659712,
      "step": 112110
    },
    {
      "epoch": 12.333883388338833,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.011539640969742606,
      "loss": 0.2308,
      "num_input_tokens_seen": 23660768,
      "step": 112115
    },
    {
      "epoch": 12.334433443344334,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.01153823978704562,
      "loss": 0.2298,
      "num_input_tokens_seen": 23661888,
      "step": 112120
    },
    {
      "epoch": 12.334983498349835,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011536838636254007,
      "loss": 0.2293,
      "num_input_tokens_seen": 23662880,
      "step": 112125
    },
    {
      "epoch": 12.335533553355335,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0115354375173807,
      "loss": 0.2298,
      "num_input_tokens_seen": 23663936,
      "step": 112130
    },
    {
      "epoch": 12.336083608360836,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.011534036430438599,
      "loss": 0.2319,
      "num_input_tokens_seen": 23664992,
      "step": 112135
    },
    {
      "epoch": 12.336633663366337,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.011532635375440617,
      "loss": 0.2314,
      "num_input_tokens_seen": 23666016,
      "step": 112140
    },
    {
      "epoch": 12.337183718371838,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.01153123435239968,
      "loss": 0.2298,
      "num_input_tokens_seen": 23667072,
      "step": 112145
    },
    {
      "epoch": 12.337733773377337,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.011529833361328684,
      "loss": 0.2319,
      "num_input_tokens_seen": 23668192,
      "step": 112150
    },
    {
      "epoch": 12.338283828382838,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.01152843240224055,
      "loss": 0.233,
      "num_input_tokens_seen": 23669248,
      "step": 112155
    },
    {
      "epoch": 12.33883388338834,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.011527031475148193,
      "loss": 0.2298,
      "num_input_tokens_seen": 23670368,
      "step": 112160
    },
    {
      "epoch": 12.339383938393839,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.011525630580064516,
      "loss": 0.2298,
      "num_input_tokens_seen": 23671424,
      "step": 112165
    },
    {
      "epoch": 12.33993399339934,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.011524229717002437,
      "loss": 0.2345,
      "num_input_tokens_seen": 23672512,
      "step": 112170
    },
    {
      "epoch": 12.340484048404841,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.011522828885974867,
      "loss": 0.2319,
      "num_input_tokens_seen": 23673504,
      "step": 112175
    },
    {
      "epoch": 12.34103410341034,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011521428086994709,
      "loss": 0.2304,
      "num_input_tokens_seen": 23674560,
      "step": 112180
    },
    {
      "epoch": 12.341584158415841,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.011520027320074882,
      "loss": 0.2314,
      "num_input_tokens_seen": 23675552,
      "step": 112185
    },
    {
      "epoch": 12.342134213421343,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011518626585228292,
      "loss": 0.2314,
      "num_input_tokens_seen": 23676608,
      "step": 112190
    },
    {
      "epoch": 12.342684268426842,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.011517225882467851,
      "loss": 0.2309,
      "num_input_tokens_seen": 23677696,
      "step": 112195
    },
    {
      "epoch": 12.343234323432343,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.011515825211806472,
      "loss": 0.2304,
      "num_input_tokens_seen": 23678752,
      "step": 112200
    },
    {
      "epoch": 12.343784378437844,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011514424573257052,
      "loss": 0.2309,
      "num_input_tokens_seen": 23679808,
      "step": 112205
    },
    {
      "epoch": 12.344334433443345,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011513023966832514,
      "loss": 0.234,
      "num_input_tokens_seen": 23680800,
      "step": 112210
    },
    {
      "epoch": 12.344884488448844,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.011511623392545757,
      "loss": 0.2293,
      "num_input_tokens_seen": 23681888,
      "step": 112215
    },
    {
      "epoch": 12.345434543454346,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011510222850409698,
      "loss": 0.2293,
      "num_input_tokens_seen": 23682912,
      "step": 112220
    },
    {
      "epoch": 12.345984598459847,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.01150882234043724,
      "loss": 0.2303,
      "num_input_tokens_seen": 23683936,
      "step": 112225
    },
    {
      "epoch": 12.346534653465346,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.011507421862641287,
      "loss": 0.2325,
      "num_input_tokens_seen": 23684960,
      "step": 112230
    },
    {
      "epoch": 12.347084708470847,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.011506021417034756,
      "loss": 0.2298,
      "num_input_tokens_seen": 23686016,
      "step": 112235
    },
    {
      "epoch": 12.347634763476348,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.011504621003630543,
      "loss": 0.2324,
      "num_input_tokens_seen": 23687040,
      "step": 112240
    },
    {
      "epoch": 12.348184818481847,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.011503220622441563,
      "loss": 0.2324,
      "num_input_tokens_seen": 23688064,
      "step": 112245
    },
    {
      "epoch": 12.348734873487349,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.011501820273480725,
      "loss": 0.2319,
      "num_input_tokens_seen": 23689120,
      "step": 112250
    },
    {
      "epoch": 12.34928492849285,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.011500419956760924,
      "loss": 0.2314,
      "num_input_tokens_seen": 23690112,
      "step": 112255
    },
    {
      "epoch": 12.34983498349835,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.01149901967229508,
      "loss": 0.2314,
      "num_input_tokens_seen": 23691200,
      "step": 112260
    },
    {
      "epoch": 12.35038503850385,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011497619420096085,
      "loss": 0.2298,
      "num_input_tokens_seen": 23692256,
      "step": 112265
    },
    {
      "epoch": 12.350935093509351,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.011496219200176856,
      "loss": 0.2314,
      "num_input_tokens_seen": 23693216,
      "step": 112270
    },
    {
      "epoch": 12.351485148514852,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.011494819012550296,
      "loss": 0.2329,
      "num_input_tokens_seen": 23694304,
      "step": 112275
    },
    {
      "epoch": 12.352035203520352,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0114934188572293,
      "loss": 0.2298,
      "num_input_tokens_seen": 23695328,
      "step": 112280
    },
    {
      "epoch": 12.352585258525853,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.011492018734226783,
      "loss": 0.2319,
      "num_input_tokens_seen": 23696352,
      "step": 112285
    },
    {
      "epoch": 12.353135313531354,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.011490618643555644,
      "loss": 0.2304,
      "num_input_tokens_seen": 23697408,
      "step": 112290
    },
    {
      "epoch": 12.353685368536853,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.011489218585228793,
      "loss": 0.2314,
      "num_input_tokens_seen": 23698400,
      "step": 112295
    },
    {
      "epoch": 12.354235423542354,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.01148781855925913,
      "loss": 0.2303,
      "num_input_tokens_seen": 23699456,
      "step": 112300
    },
    {
      "epoch": 12.354785478547855,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.011486418565659552,
      "loss": 0.2293,
      "num_input_tokens_seen": 23700416,
      "step": 112305
    },
    {
      "epoch": 12.355335533553355,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.011485018604442973,
      "loss": 0.2319,
      "num_input_tokens_seen": 23701504,
      "step": 112310
    },
    {
      "epoch": 12.355885588558856,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.011483618675622287,
      "loss": 0.2329,
      "num_input_tokens_seen": 23702592,
      "step": 112315
    },
    {
      "epoch": 12.356435643564357,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011482218779210405,
      "loss": 0.2314,
      "num_input_tokens_seen": 23703648,
      "step": 112320
    },
    {
      "epoch": 12.356985698569858,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.011480818915220225,
      "loss": 0.2293,
      "num_input_tokens_seen": 23704768,
      "step": 112325
    },
    {
      "epoch": 12.357535753575357,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.011479419083664642,
      "loss": 0.2298,
      "num_input_tokens_seen": 23705760,
      "step": 112330
    },
    {
      "epoch": 12.358085808580858,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.011478019284556574,
      "loss": 0.234,
      "num_input_tokens_seen": 23706912,
      "step": 112335
    },
    {
      "epoch": 12.35863586358636,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0114766195179089,
      "loss": 0.2303,
      "num_input_tokens_seen": 23708000,
      "step": 112340
    },
    {
      "epoch": 12.359185918591859,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01147521978373454,
      "loss": 0.2324,
      "num_input_tokens_seen": 23709024,
      "step": 112345
    },
    {
      "epoch": 12.35973597359736,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011473820082046393,
      "loss": 0.2314,
      "num_input_tokens_seen": 23710048,
      "step": 112350
    },
    {
      "epoch": 12.36028602860286,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.011472420412857346,
      "loss": 0.2324,
      "num_input_tokens_seen": 23711200,
      "step": 112355
    },
    {
      "epoch": 12.36083608360836,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01147102077618031,
      "loss": 0.2298,
      "num_input_tokens_seen": 23712224,
      "step": 112360
    },
    {
      "epoch": 12.361386138613861,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.011469621172028184,
      "loss": 0.2314,
      "num_input_tokens_seen": 23713280,
      "step": 112365
    },
    {
      "epoch": 12.361936193619362,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.011468221600413858,
      "loss": 0.2304,
      "num_input_tokens_seen": 23714336,
      "step": 112370
    },
    {
      "epoch": 12.362486248624862,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.01146682206135025,
      "loss": 0.2324,
      "num_input_tokens_seen": 23715456,
      "step": 112375
    },
    {
      "epoch": 12.363036303630363,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.011465422554850238,
      "loss": 0.2298,
      "num_input_tokens_seen": 23716576,
      "step": 112380
    },
    {
      "epoch": 12.363586358635864,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.011464023080926736,
      "loss": 0.2325,
      "num_input_tokens_seen": 23717600,
      "step": 112385
    },
    {
      "epoch": 12.364136413641365,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.011462623639592637,
      "loss": 0.2314,
      "num_input_tokens_seen": 23718688,
      "step": 112390
    },
    {
      "epoch": 12.364686468646864,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.011461224230860833,
      "loss": 0.2314,
      "num_input_tokens_seen": 23719680,
      "step": 112395
    },
    {
      "epoch": 12.365236523652365,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.011459824854744232,
      "loss": 0.2298,
      "num_input_tokens_seen": 23720768,
      "step": 112400
    },
    {
      "epoch": 12.365786578657866,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.011458425511255723,
      "loss": 0.2319,
      "num_input_tokens_seen": 23721824,
      "step": 112405
    },
    {
      "epoch": 12.366336633663366,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011457026200408213,
      "loss": 0.2314,
      "num_input_tokens_seen": 23722912,
      "step": 112410
    },
    {
      "epoch": 12.366886688668867,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01145562692221459,
      "loss": 0.2304,
      "num_input_tokens_seen": 23724000,
      "step": 112415
    },
    {
      "epoch": 12.367436743674368,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.011454227676687747,
      "loss": 0.2314,
      "num_input_tokens_seen": 23724992,
      "step": 112420
    },
    {
      "epoch": 12.367986798679867,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011452828463840594,
      "loss": 0.2313,
      "num_input_tokens_seen": 23726048,
      "step": 112425
    },
    {
      "epoch": 12.368536853685368,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.01145142928368601,
      "loss": 0.2314,
      "num_input_tokens_seen": 23727072,
      "step": 112430
    },
    {
      "epoch": 12.36908690869087,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.011450030136236907,
      "loss": 0.2308,
      "num_input_tokens_seen": 23728096,
      "step": 112435
    },
    {
      "epoch": 12.369636963696369,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011448631021506171,
      "loss": 0.2335,
      "num_input_tokens_seen": 23729120,
      "step": 112440
    },
    {
      "epoch": 12.37018701870187,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.011447231939506695,
      "loss": 0.2324,
      "num_input_tokens_seen": 23730208,
      "step": 112445
    },
    {
      "epoch": 12.370737073707371,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011445832890251386,
      "loss": 0.2324,
      "num_input_tokens_seen": 23731232,
      "step": 112450
    },
    {
      "epoch": 12.371287128712872,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.01144443387375312,
      "loss": 0.2324,
      "num_input_tokens_seen": 23732256,
      "step": 112455
    },
    {
      "epoch": 12.371837183718371,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.011443034890024805,
      "loss": 0.2324,
      "num_input_tokens_seen": 23733344,
      "step": 112460
    },
    {
      "epoch": 12.372387238723872,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.011441635939079337,
      "loss": 0.2298,
      "num_input_tokens_seen": 23734432,
      "step": 112465
    },
    {
      "epoch": 12.372937293729374,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.011440237020929594,
      "loss": 0.2319,
      "num_input_tokens_seen": 23735488,
      "step": 112470
    },
    {
      "epoch": 12.373487348734873,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.011438838135588481,
      "loss": 0.2303,
      "num_input_tokens_seen": 23736480,
      "step": 112475
    },
    {
      "epoch": 12.374037403740374,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.011437439283068887,
      "loss": 0.2324,
      "num_input_tokens_seen": 23737568,
      "step": 112480
    },
    {
      "epoch": 12.374587458745875,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.01143604046338371,
      "loss": 0.2298,
      "num_input_tokens_seen": 23738624,
      "step": 112485
    },
    {
      "epoch": 12.375137513751374,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.011434641676545834,
      "loss": 0.2309,
      "num_input_tokens_seen": 23739648,
      "step": 112490
    },
    {
      "epoch": 12.375687568756875,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.011433242922568154,
      "loss": 0.2319,
      "num_input_tokens_seen": 23740672,
      "step": 112495
    },
    {
      "epoch": 12.376237623762377,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011431844201463565,
      "loss": 0.2309,
      "num_input_tokens_seen": 23741760,
      "step": 112500
    },
    {
      "epoch": 12.376787678767876,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.011430445513244953,
      "loss": 0.2325,
      "num_input_tokens_seen": 23742816,
      "step": 112505
    },
    {
      "epoch": 12.377337733773377,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.011429046857925217,
      "loss": 0.2324,
      "num_input_tokens_seen": 23743872,
      "step": 112510
    },
    {
      "epoch": 12.377887788778878,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.01142764823551724,
      "loss": 0.2324,
      "num_input_tokens_seen": 23744896,
      "step": 112515
    },
    {
      "epoch": 12.37843784378438,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.011426249646033912,
      "loss": 0.2293,
      "num_input_tokens_seen": 23745920,
      "step": 112520
    },
    {
      "epoch": 12.378987898789878,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.011424851089488132,
      "loss": 0.2314,
      "num_input_tokens_seen": 23747008,
      "step": 112525
    },
    {
      "epoch": 12.37953795379538,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.011423452565892778,
      "loss": 0.2293,
      "num_input_tokens_seen": 23748128,
      "step": 112530
    },
    {
      "epoch": 12.38008800880088,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.011422054075260748,
      "loss": 0.2314,
      "num_input_tokens_seen": 23749248,
      "step": 112535
    },
    {
      "epoch": 12.38063806380638,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.011420655617604933,
      "loss": 0.2309,
      "num_input_tokens_seen": 23750336,
      "step": 112540
    },
    {
      "epoch": 12.381188118811881,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.011419257192938207,
      "loss": 0.2314,
      "num_input_tokens_seen": 23751392,
      "step": 112545
    },
    {
      "epoch": 12.381738173817382,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.01141785880127348,
      "loss": 0.2319,
      "num_input_tokens_seen": 23752416,
      "step": 112550
    },
    {
      "epoch": 12.382288228822881,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.011416460442623627,
      "loss": 0.2309,
      "num_input_tokens_seen": 23753408,
      "step": 112555
    },
    {
      "epoch": 12.382838283828383,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011415062117001534,
      "loss": 0.2314,
      "num_input_tokens_seen": 23754400,
      "step": 112560
    },
    {
      "epoch": 12.383388338833884,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0114136638244201,
      "loss": 0.2335,
      "num_input_tokens_seen": 23755456,
      "step": 112565
    },
    {
      "epoch": 12.383938393839385,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011412265564892199,
      "loss": 0.2319,
      "num_input_tokens_seen": 23756448,
      "step": 112570
    },
    {
      "epoch": 12.384488448844884,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.011410867338430728,
      "loss": 0.2298,
      "num_input_tokens_seen": 23757504,
      "step": 112575
    },
    {
      "epoch": 12.385038503850385,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.011409469145048572,
      "loss": 0.2325,
      "num_input_tokens_seen": 23758528,
      "step": 112580
    },
    {
      "epoch": 12.385588558855886,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.01140807098475861,
      "loss": 0.2314,
      "num_input_tokens_seen": 23759648,
      "step": 112585
    },
    {
      "epoch": 12.386138613861386,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.01140667285757374,
      "loss": 0.2293,
      "num_input_tokens_seen": 23760768,
      "step": 112590
    },
    {
      "epoch": 12.386688668866887,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01140527476350684,
      "loss": 0.2309,
      "num_input_tokens_seen": 23761856,
      "step": 112595
    },
    {
      "epoch": 12.387238723872388,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011403876702570799,
      "loss": 0.2324,
      "num_input_tokens_seen": 23762912,
      "step": 112600
    },
    {
      "epoch": 12.387788778877887,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011402478674778501,
      "loss": 0.2309,
      "num_input_tokens_seen": 23764000,
      "step": 112605
    },
    {
      "epoch": 12.388338833883388,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.011401080680142827,
      "loss": 0.2319,
      "num_input_tokens_seen": 23765088,
      "step": 112610
    },
    {
      "epoch": 12.38888888888889,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.011399682718676668,
      "loss": 0.2314,
      "num_input_tokens_seen": 23766144,
      "step": 112615
    },
    {
      "epoch": 12.389438943894389,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.011398284790392904,
      "loss": 0.2298,
      "num_input_tokens_seen": 23767200,
      "step": 112620
    },
    {
      "epoch": 12.38998899889989,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.011396886895304426,
      "loss": 0.2309,
      "num_input_tokens_seen": 23768256,
      "step": 112625
    },
    {
      "epoch": 12.39053905390539,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.011395489033424109,
      "loss": 0.2309,
      "num_input_tokens_seen": 23769312,
      "step": 112630
    },
    {
      "epoch": 12.391089108910892,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.011394091204764837,
      "loss": 0.2293,
      "num_input_tokens_seen": 23770400,
      "step": 112635
    },
    {
      "epoch": 12.391639163916391,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.011392693409339502,
      "loss": 0.2324,
      "num_input_tokens_seen": 23771424,
      "step": 112640
    },
    {
      "epoch": 12.392189218921892,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.011391295647160977,
      "loss": 0.2319,
      "num_input_tokens_seen": 23772512,
      "step": 112645
    },
    {
      "epoch": 12.392739273927393,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011389897918242149,
      "loss": 0.2309,
      "num_input_tokens_seen": 23773568,
      "step": 112650
    },
    {
      "epoch": 12.393289328932893,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011388500222595903,
      "loss": 0.2335,
      "num_input_tokens_seen": 23774592,
      "step": 112655
    },
    {
      "epoch": 12.393839383938394,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011387102560235113,
      "loss": 0.2319,
      "num_input_tokens_seen": 23775616,
      "step": 112660
    },
    {
      "epoch": 12.394389438943895,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.011385704931172667,
      "loss": 0.2314,
      "num_input_tokens_seen": 23776672,
      "step": 112665
    },
    {
      "epoch": 12.394939493949394,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.01138430733542144,
      "loss": 0.2314,
      "num_input_tokens_seen": 23777760,
      "step": 112670
    },
    {
      "epoch": 12.395489548954895,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.011382909772994321,
      "loss": 0.2319,
      "num_input_tokens_seen": 23778784,
      "step": 112675
    },
    {
      "epoch": 12.396039603960396,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.011381512243904192,
      "loss": 0.2335,
      "num_input_tokens_seen": 23779872,
      "step": 112680
    },
    {
      "epoch": 12.396589658965897,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.01138011474816392,
      "loss": 0.2351,
      "num_input_tokens_seen": 23780960,
      "step": 112685
    },
    {
      "epoch": 12.397139713971397,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.011378717285786398,
      "loss": 0.2304,
      "num_input_tokens_seen": 23781984,
      "step": 112690
    },
    {
      "epoch": 12.397689768976898,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.011377319856784498,
      "loss": 0.233,
      "num_input_tokens_seen": 23783104,
      "step": 112695
    },
    {
      "epoch": 12.398239823982399,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.011375922461171108,
      "loss": 0.2324,
      "num_input_tokens_seen": 23784192,
      "step": 112700
    },
    {
      "epoch": 12.398789878987898,
      "grad_norm": 0.002899169921875,
      "learning_rate": 0.011374525098959098,
      "loss": 0.2309,
      "num_input_tokens_seen": 23785248,
      "step": 112705
    },
    {
      "epoch": 12.3993399339934,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.011373127770161346,
      "loss": 0.2319,
      "num_input_tokens_seen": 23786368,
      "step": 112710
    },
    {
      "epoch": 12.3998899889989,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.011371730474790745,
      "loss": 0.234,
      "num_input_tokens_seen": 23787392,
      "step": 112715
    },
    {
      "epoch": 12.4004400440044,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.011370333212860155,
      "loss": 0.2288,
      "num_input_tokens_seen": 23788416,
      "step": 112720
    },
    {
      "epoch": 12.400990099009901,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.011368935984382466,
      "loss": 0.2309,
      "num_input_tokens_seen": 23789472,
      "step": 112725
    },
    {
      "epoch": 12.401540154015402,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011367538789370553,
      "loss": 0.2314,
      "num_input_tokens_seen": 23790496,
      "step": 112730
    },
    {
      "epoch": 12.402090209020901,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011366141627837285,
      "loss": 0.2303,
      "num_input_tokens_seen": 23791584,
      "step": 112735
    },
    {
      "epoch": 12.402640264026402,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.011364744499795553,
      "loss": 0.2304,
      "num_input_tokens_seen": 23792640,
      "step": 112740
    },
    {
      "epoch": 12.403190319031903,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.011363347405258224,
      "loss": 0.232,
      "num_input_tokens_seen": 23793664,
      "step": 112745
    },
    {
      "epoch": 12.403740374037405,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.011361950344238175,
      "loss": 0.2304,
      "num_input_tokens_seen": 23794720,
      "step": 112750
    },
    {
      "epoch": 12.404290429042904,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.01136055331674829,
      "loss": 0.2314,
      "num_input_tokens_seen": 23795776,
      "step": 112755
    },
    {
      "epoch": 12.404840484048405,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011359156322801432,
      "loss": 0.234,
      "num_input_tokens_seen": 23796864,
      "step": 112760
    },
    {
      "epoch": 12.405390539053906,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.011357759362410486,
      "loss": 0.2314,
      "num_input_tokens_seen": 23797888,
      "step": 112765
    },
    {
      "epoch": 12.405940594059405,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01135636243558833,
      "loss": 0.2309,
      "num_input_tokens_seen": 23798944,
      "step": 112770
    },
    {
      "epoch": 12.406490649064907,
      "grad_norm": 0.001953125,
      "learning_rate": 0.011354965542347823,
      "loss": 0.2299,
      "num_input_tokens_seen": 23800000,
      "step": 112775
    },
    {
      "epoch": 12.407040704070408,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011353568682701855,
      "loss": 0.2319,
      "num_input_tokens_seen": 23801024,
      "step": 112780
    },
    {
      "epoch": 12.407590759075907,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.011352171856663288,
      "loss": 0.2324,
      "num_input_tokens_seen": 23802112,
      "step": 112785
    },
    {
      "epoch": 12.408140814081408,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.011350775064245009,
      "loss": 0.2288,
      "num_input_tokens_seen": 23803168,
      "step": 112790
    },
    {
      "epoch": 12.408690869086909,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.011349378305459889,
      "loss": 0.2298,
      "num_input_tokens_seen": 23804224,
      "step": 112795
    },
    {
      "epoch": 12.409240924092408,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01134798158032079,
      "loss": 0.2325,
      "num_input_tokens_seen": 23805216,
      "step": 112800
    },
    {
      "epoch": 12.40979097909791,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.011346584888840596,
      "loss": 0.2314,
      "num_input_tokens_seen": 23806208,
      "step": 112805
    },
    {
      "epoch": 12.41034103410341,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.01134518823103217,
      "loss": 0.2319,
      "num_input_tokens_seen": 23807200,
      "step": 112810
    },
    {
      "epoch": 12.410891089108912,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0113437916069084,
      "loss": 0.2319,
      "num_input_tokens_seen": 23808288,
      "step": 112815
    },
    {
      "epoch": 12.411441144114411,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011342395016482144,
      "loss": 0.2319,
      "num_input_tokens_seen": 23809344,
      "step": 112820
    },
    {
      "epoch": 12.411991199119912,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.011340998459766275,
      "loss": 0.233,
      "num_input_tokens_seen": 23810432,
      "step": 112825
    },
    {
      "epoch": 12.412541254125413,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.011339601936773674,
      "loss": 0.233,
      "num_input_tokens_seen": 23811488,
      "step": 112830
    },
    {
      "epoch": 12.413091309130913,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0113382054475172,
      "loss": 0.2351,
      "num_input_tokens_seen": 23812544,
      "step": 112835
    },
    {
      "epoch": 12.413641364136414,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.011336808992009728,
      "loss": 0.2319,
      "num_input_tokens_seen": 23813632,
      "step": 112840
    },
    {
      "epoch": 12.414191419141915,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.01133541257026414,
      "loss": 0.2325,
      "num_input_tokens_seen": 23814688,
      "step": 112845
    },
    {
      "epoch": 12.414741474147414,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01133401618229328,
      "loss": 0.2345,
      "num_input_tokens_seen": 23815680,
      "step": 112850
    },
    {
      "epoch": 12.415291529152915,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.011332619828110046,
      "loss": 0.234,
      "num_input_tokens_seen": 23816736,
      "step": 112855
    },
    {
      "epoch": 12.415841584158416,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01133122350772729,
      "loss": 0.2309,
      "num_input_tokens_seen": 23817792,
      "step": 112860
    },
    {
      "epoch": 12.416391639163916,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.011329827221157888,
      "loss": 0.2324,
      "num_input_tokens_seen": 23818880,
      "step": 112865
    },
    {
      "epoch": 12.416941694169417,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.011328430968414712,
      "loss": 0.2324,
      "num_input_tokens_seen": 23819936,
      "step": 112870
    },
    {
      "epoch": 12.417491749174918,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.011327034749510616,
      "loss": 0.2309,
      "num_input_tokens_seen": 23821024,
      "step": 112875
    },
    {
      "epoch": 12.418041804180419,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.011325638564458486,
      "loss": 0.2304,
      "num_input_tokens_seen": 23822048,
      "step": 112880
    },
    {
      "epoch": 12.418591859185918,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011324242413271176,
      "loss": 0.2309,
      "num_input_tokens_seen": 23823072,
      "step": 112885
    },
    {
      "epoch": 12.41914191419142,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.011322846295961569,
      "loss": 0.2309,
      "num_input_tokens_seen": 23824160,
      "step": 112890
    },
    {
      "epoch": 12.41969196919692,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.011321450212542519,
      "loss": 0.2319,
      "num_input_tokens_seen": 23825248,
      "step": 112895
    },
    {
      "epoch": 12.42024202420242,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011320054163026892,
      "loss": 0.2309,
      "num_input_tokens_seen": 23826336,
      "step": 112900
    },
    {
      "epoch": 12.42079207920792,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011318658147427569,
      "loss": 0.234,
      "num_input_tokens_seen": 23827392,
      "step": 112905
    },
    {
      "epoch": 12.421342134213422,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.011317262165757395,
      "loss": 0.2314,
      "num_input_tokens_seen": 23828384,
      "step": 112910
    },
    {
      "epoch": 12.421892189218921,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.01131586621802926,
      "loss": 0.2329,
      "num_input_tokens_seen": 23829472,
      "step": 112915
    },
    {
      "epoch": 12.422442244224422,
      "grad_norm": 0.00051116943359375,
      "learning_rate": 0.011314470304256015,
      "loss": 0.2298,
      "num_input_tokens_seen": 23830496,
      "step": 112920
    },
    {
      "epoch": 12.422992299229923,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.011313074424450524,
      "loss": 0.2298,
      "num_input_tokens_seen": 23831520,
      "step": 112925
    },
    {
      "epoch": 12.423542354235423,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.011311678578625667,
      "loss": 0.2303,
      "num_input_tokens_seen": 23832672,
      "step": 112930
    },
    {
      "epoch": 12.424092409240924,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.01131028276679429,
      "loss": 0.2309,
      "num_input_tokens_seen": 23833728,
      "step": 112935
    },
    {
      "epoch": 12.424642464246425,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.011308886988969268,
      "loss": 0.2319,
      "num_input_tokens_seen": 23834720,
      "step": 112940
    },
    {
      "epoch": 12.425192519251926,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.011307491245163467,
      "loss": 0.2314,
      "num_input_tokens_seen": 23835744,
      "step": 112945
    },
    {
      "epoch": 12.425742574257425,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.011306095535389743,
      "loss": 0.2309,
      "num_input_tokens_seen": 23836736,
      "step": 112950
    },
    {
      "epoch": 12.426292629262926,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.011304699859660965,
      "loss": 0.2319,
      "num_input_tokens_seen": 23837792,
      "step": 112955
    },
    {
      "epoch": 12.426842684268427,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01130330421799,
      "loss": 0.2324,
      "num_input_tokens_seen": 23838816,
      "step": 112960
    },
    {
      "epoch": 12.427392739273927,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.0113019086103897,
      "loss": 0.2304,
      "num_input_tokens_seen": 23839840,
      "step": 112965
    },
    {
      "epoch": 12.427942794279428,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.011300513036872935,
      "loss": 0.2309,
      "num_input_tokens_seen": 23840896,
      "step": 112970
    },
    {
      "epoch": 12.428492849284929,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011299117497452564,
      "loss": 0.2346,
      "num_input_tokens_seen": 23841888,
      "step": 112975
    },
    {
      "epoch": 12.429042904290428,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.011297721992141456,
      "loss": 0.2303,
      "num_input_tokens_seen": 23842944,
      "step": 112980
    },
    {
      "epoch": 12.42959295929593,
      "grad_norm": 0.002716064453125,
      "learning_rate": 0.01129632652095247,
      "loss": 0.2298,
      "num_input_tokens_seen": 23844000,
      "step": 112985
    },
    {
      "epoch": 12.43014301430143,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.011294931083898457,
      "loss": 0.2309,
      "num_input_tokens_seen": 23845088,
      "step": 112990
    },
    {
      "epoch": 12.430693069306932,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011293535680992292,
      "loss": 0.2335,
      "num_input_tokens_seen": 23846176,
      "step": 112995
    },
    {
      "epoch": 12.43124312431243,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011292140312246825,
      "loss": 0.2319,
      "num_input_tokens_seen": 23847232,
      "step": 113000
    },
    {
      "epoch": 12.431793179317932,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.011290744977674929,
      "loss": 0.2288,
      "num_input_tokens_seen": 23848256,
      "step": 113005
    },
    {
      "epoch": 12.432343234323433,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.011289349677289453,
      "loss": 0.233,
      "num_input_tokens_seen": 23849312,
      "step": 113010
    },
    {
      "epoch": 12.432893289328932,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.011287954411103258,
      "loss": 0.2335,
      "num_input_tokens_seen": 23850336,
      "step": 113015
    },
    {
      "epoch": 12.433443344334433,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.011286559179129213,
      "loss": 0.2293,
      "num_input_tokens_seen": 23851392,
      "step": 113020
    },
    {
      "epoch": 12.433993399339935,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.011285163981380162,
      "loss": 0.2314,
      "num_input_tokens_seen": 23852448,
      "step": 113025
    },
    {
      "epoch": 12.434543454345434,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.011283768817868975,
      "loss": 0.2324,
      "num_input_tokens_seen": 23853600,
      "step": 113030
    },
    {
      "epoch": 12.435093509350935,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01128237368860851,
      "loss": 0.2335,
      "num_input_tokens_seen": 23854656,
      "step": 113035
    },
    {
      "epoch": 12.435643564356436,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.011280978593611617,
      "loss": 0.2319,
      "num_input_tokens_seen": 23855680,
      "step": 113040
    },
    {
      "epoch": 12.436193619361935,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.011279583532891168,
      "loss": 0.2319,
      "num_input_tokens_seen": 23856736,
      "step": 113045
    },
    {
      "epoch": 12.436743674367436,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.011278188506460004,
      "loss": 0.2309,
      "num_input_tokens_seen": 23857760,
      "step": 113050
    },
    {
      "epoch": 12.437293729372938,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.011276793514330996,
      "loss": 0.2309,
      "num_input_tokens_seen": 23858816,
      "step": 113055
    },
    {
      "epoch": 12.437843784378439,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.011275398556516998,
      "loss": 0.2304,
      "num_input_tokens_seen": 23859808,
      "step": 113060
    },
    {
      "epoch": 12.438393839383938,
      "grad_norm": 0.000736236572265625,
      "learning_rate": 0.011274003633030858,
      "loss": 0.233,
      "num_input_tokens_seen": 23860864,
      "step": 113065
    },
    {
      "epoch": 12.438943894389439,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.011272608743885441,
      "loss": 0.2319,
      "num_input_tokens_seen": 23861952,
      "step": 113070
    },
    {
      "epoch": 12.43949394939494,
      "grad_norm": 0.0026397705078125,
      "learning_rate": 0.0112712138890936,
      "loss": 0.233,
      "num_input_tokens_seen": 23862976,
      "step": 113075
    },
    {
      "epoch": 12.44004400440044,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.011269819068668193,
      "loss": 0.2304,
      "num_input_tokens_seen": 23864032,
      "step": 113080
    },
    {
      "epoch": 12.44059405940594,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.011268424282622074,
      "loss": 0.2314,
      "num_input_tokens_seen": 23865056,
      "step": 113085
    },
    {
      "epoch": 12.441144114411442,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.011267029530968095,
      "loss": 0.2329,
      "num_input_tokens_seen": 23866144,
      "step": 113090
    },
    {
      "epoch": 12.441694169416941,
      "grad_norm": 0.0005035400390625,
      "learning_rate": 0.011265634813719115,
      "loss": 0.2298,
      "num_input_tokens_seen": 23867168,
      "step": 113095
    },
    {
      "epoch": 12.442244224422442,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.011264240130887986,
      "loss": 0.2304,
      "num_input_tokens_seen": 23868192,
      "step": 113100
    },
    {
      "epoch": 12.442794279427943,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.011262845482487566,
      "loss": 0.2314,
      "num_input_tokens_seen": 23869280,
      "step": 113105
    },
    {
      "epoch": 12.443344334433444,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.011261450868530705,
      "loss": 0.2293,
      "num_input_tokens_seen": 23870368,
      "step": 113110
    },
    {
      "epoch": 12.443894389438944,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.011260056289030254,
      "loss": 0.2309,
      "num_input_tokens_seen": 23871424,
      "step": 113115
    },
    {
      "epoch": 12.444444444444445,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.011258661743999076,
      "loss": 0.2309,
      "num_input_tokens_seen": 23872544,
      "step": 113120
    },
    {
      "epoch": 12.444994499449946,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.01125726723345001,
      "loss": 0.2314,
      "num_input_tokens_seen": 23873600,
      "step": 113125
    },
    {
      "epoch": 12.445544554455445,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.01125587275739592,
      "loss": 0.2304,
      "num_input_tokens_seen": 23874656,
      "step": 113130
    },
    {
      "epoch": 12.446094609460946,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011254478315849659,
      "loss": 0.2309,
      "num_input_tokens_seen": 23875808,
      "step": 113135
    },
    {
      "epoch": 12.446644664466447,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.011253083908824065,
      "loss": 0.2309,
      "num_input_tokens_seen": 23876864,
      "step": 113140
    },
    {
      "epoch": 12.447194719471947,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.011251689536332004,
      "loss": 0.2304,
      "num_input_tokens_seen": 23877920,
      "step": 113145
    },
    {
      "epoch": 12.447744774477448,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011250295198386325,
      "loss": 0.2304,
      "num_input_tokens_seen": 23878912,
      "step": 113150
    },
    {
      "epoch": 12.448294829482949,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011248900894999863,
      "loss": 0.2283,
      "num_input_tokens_seen": 23879936,
      "step": 113155
    },
    {
      "epoch": 12.448844884488448,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.011247506626185495,
      "loss": 0.2293,
      "num_input_tokens_seen": 23880960,
      "step": 113160
    },
    {
      "epoch": 12.44939493949395,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.011246112391956048,
      "loss": 0.2324,
      "num_input_tokens_seen": 23882016,
      "step": 113165
    },
    {
      "epoch": 12.44994499449945,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.011244718192324388,
      "loss": 0.233,
      "num_input_tokens_seen": 23883136,
      "step": 113170
    },
    {
      "epoch": 12.450495049504951,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.011243324027303363,
      "loss": 0.2314,
      "num_input_tokens_seen": 23884160,
      "step": 113175
    },
    {
      "epoch": 12.45104510451045,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.011241929896905808,
      "loss": 0.2283,
      "num_input_tokens_seen": 23885216,
      "step": 113180
    },
    {
      "epoch": 12.451595159515952,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.011240535801144586,
      "loss": 0.2314,
      "num_input_tokens_seen": 23886208,
      "step": 113185
    },
    {
      "epoch": 12.452145214521453,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.011239141740032538,
      "loss": 0.2304,
      "num_input_tokens_seen": 23887296,
      "step": 113190
    },
    {
      "epoch": 12.452695269526952,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.011237747713582522,
      "loss": 0.2314,
      "num_input_tokens_seen": 23888384,
      "step": 113195
    },
    {
      "epoch": 12.453245324532453,
      "grad_norm": 0.0006561279296875,
      "learning_rate": 0.011236353721807379,
      "loss": 0.2346,
      "num_input_tokens_seen": 23889440,
      "step": 113200
    },
    {
      "epoch": 12.453795379537954,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.011234959764719953,
      "loss": 0.2325,
      "num_input_tokens_seen": 23890528,
      "step": 113205
    },
    {
      "epoch": 12.454345434543454,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.011233565842333102,
      "loss": 0.2288,
      "num_input_tokens_seen": 23891616,
      "step": 113210
    },
    {
      "epoch": 12.454895489548955,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.01123217195465966,
      "loss": 0.2319,
      "num_input_tokens_seen": 23892704,
      "step": 113215
    },
    {
      "epoch": 12.455445544554456,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.011230778101712488,
      "loss": 0.2278,
      "num_input_tokens_seen": 23893792,
      "step": 113220
    },
    {
      "epoch": 12.455995599559955,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.011229384283504423,
      "loss": 0.2298,
      "num_input_tokens_seen": 23894848,
      "step": 113225
    },
    {
      "epoch": 12.456545654565456,
      "grad_norm": 0.001953125,
      "learning_rate": 0.011227990500048312,
      "loss": 0.233,
      "num_input_tokens_seen": 23895872,
      "step": 113230
    },
    {
      "epoch": 12.457095709570957,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.011226596751357008,
      "loss": 0.2325,
      "num_input_tokens_seen": 23896864,
      "step": 113235
    },
    {
      "epoch": 12.457645764576458,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.011225203037443343,
      "loss": 0.232,
      "num_input_tokens_seen": 23897920,
      "step": 113240
    },
    {
      "epoch": 12.458195819581958,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.011223809358320177,
      "loss": 0.2304,
      "num_input_tokens_seen": 23899072,
      "step": 113245
    },
    {
      "epoch": 12.458745874587459,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01122241571400035,
      "loss": 0.2288,
      "num_input_tokens_seen": 23900128,
      "step": 113250
    },
    {
      "epoch": 12.45929592959296,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.011221022104496697,
      "loss": 0.2309,
      "num_input_tokens_seen": 23901280,
      "step": 113255
    },
    {
      "epoch": 12.45984598459846,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.011219628529822076,
      "loss": 0.2346,
      "num_input_tokens_seen": 23902304,
      "step": 113260
    },
    {
      "epoch": 12.46039603960396,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011218234989989317,
      "loss": 0.2283,
      "num_input_tokens_seen": 23903424,
      "step": 113265
    },
    {
      "epoch": 12.460946094609461,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.011216841485011281,
      "loss": 0.2319,
      "num_input_tokens_seen": 23904480,
      "step": 113270
    },
    {
      "epoch": 12.46149614961496,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.011215448014900799,
      "loss": 0.2298,
      "num_input_tokens_seen": 23905568,
      "step": 113275
    },
    {
      "epoch": 12.462046204620462,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.011214054579670714,
      "loss": 0.2319,
      "num_input_tokens_seen": 23906624,
      "step": 113280
    },
    {
      "epoch": 12.462596259625963,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.01121266117933387,
      "loss": 0.2314,
      "num_input_tokens_seen": 23907712,
      "step": 113285
    },
    {
      "epoch": 12.463146314631462,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.011211267813903112,
      "loss": 0.2315,
      "num_input_tokens_seen": 23908704,
      "step": 113290
    },
    {
      "epoch": 12.463696369636963,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.011209874483391283,
      "loss": 0.2309,
      "num_input_tokens_seen": 23909824,
      "step": 113295
    },
    {
      "epoch": 12.464246424642464,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.011208481187811222,
      "loss": 0.2335,
      "num_input_tokens_seen": 23910912,
      "step": 113300
    },
    {
      "epoch": 12.464796479647966,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.011207087927175768,
      "loss": 0.233,
      "num_input_tokens_seen": 23912000,
      "step": 113305
    },
    {
      "epoch": 12.465346534653465,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.011205694701497768,
      "loss": 0.2335,
      "num_input_tokens_seen": 23913088,
      "step": 113310
    },
    {
      "epoch": 12.465896589658966,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.011204301510790056,
      "loss": 0.2314,
      "num_input_tokens_seen": 23914176,
      "step": 113315
    },
    {
      "epoch": 12.466446644664467,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011202908355065476,
      "loss": 0.2303,
      "num_input_tokens_seen": 23915296,
      "step": 113320
    },
    {
      "epoch": 12.466996699669966,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.011201515234336873,
      "loss": 0.2304,
      "num_input_tokens_seen": 23916352,
      "step": 113325
    },
    {
      "epoch": 12.467546754675467,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01120012214861707,
      "loss": 0.2309,
      "num_input_tokens_seen": 23917408,
      "step": 113330
    },
    {
      "epoch": 12.468096809680969,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.01119872909791893,
      "loss": 0.2298,
      "num_input_tokens_seen": 23918496,
      "step": 113335
    },
    {
      "epoch": 12.468646864686468,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.011197336082255274,
      "loss": 0.2319,
      "num_input_tokens_seen": 23919488,
      "step": 113340
    },
    {
      "epoch": 12.469196919691969,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.011195943101638944,
      "loss": 0.2319,
      "num_input_tokens_seen": 23920544,
      "step": 113345
    },
    {
      "epoch": 12.46974697469747,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.01119455015608279,
      "loss": 0.2314,
      "num_input_tokens_seen": 23921600,
      "step": 113350
    },
    {
      "epoch": 12.47029702970297,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01119315724559963,
      "loss": 0.2319,
      "num_input_tokens_seen": 23922656,
      "step": 113355
    },
    {
      "epoch": 12.47084708470847,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.011191764370202318,
      "loss": 0.2324,
      "num_input_tokens_seen": 23923712,
      "step": 113360
    },
    {
      "epoch": 12.471397139713972,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.011190371529903691,
      "loss": 0.2319,
      "num_input_tokens_seen": 23924704,
      "step": 113365
    },
    {
      "epoch": 12.471947194719473,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.011188978724716575,
      "loss": 0.2324,
      "num_input_tokens_seen": 23925824,
      "step": 113370
    },
    {
      "epoch": 12.472497249724972,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.011187585954653818,
      "loss": 0.2319,
      "num_input_tokens_seen": 23926880,
      "step": 113375
    },
    {
      "epoch": 12.473047304730473,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.011186193219728245,
      "loss": 0.2324,
      "num_input_tokens_seen": 23927936,
      "step": 113380
    },
    {
      "epoch": 12.473597359735974,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.011184800519952708,
      "loss": 0.2309,
      "num_input_tokens_seen": 23929056,
      "step": 113385
    },
    {
      "epoch": 12.474147414741473,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01118340785534003,
      "loss": 0.2324,
      "num_input_tokens_seen": 23930112,
      "step": 113390
    },
    {
      "epoch": 12.474697469746975,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011182015225903046,
      "loss": 0.2309,
      "num_input_tokens_seen": 23931104,
      "step": 113395
    },
    {
      "epoch": 12.475247524752476,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.011180622631654601,
      "loss": 0.2319,
      "num_input_tokens_seen": 23932160,
      "step": 113400
    },
    {
      "epoch": 12.475797579757975,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.01117923007260752,
      "loss": 0.2303,
      "num_input_tokens_seen": 23933216,
      "step": 113405
    },
    {
      "epoch": 12.476347634763476,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011177837548774648,
      "loss": 0.2309,
      "num_input_tokens_seen": 23934304,
      "step": 113410
    },
    {
      "epoch": 12.476897689768977,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011176445060168813,
      "loss": 0.2319,
      "num_input_tokens_seen": 23935328,
      "step": 113415
    },
    {
      "epoch": 12.477447744774478,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.011175052606802844,
      "loss": 0.2283,
      "num_input_tokens_seen": 23936384,
      "step": 113420
    },
    {
      "epoch": 12.477997799779978,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.011173660188689585,
      "loss": 0.2308,
      "num_input_tokens_seen": 23937440,
      "step": 113425
    },
    {
      "epoch": 12.478547854785479,
      "grad_norm": 0.00286865234375,
      "learning_rate": 0.01117226780584186,
      "loss": 0.2329,
      "num_input_tokens_seen": 23938496,
      "step": 113430
    },
    {
      "epoch": 12.47909790979098,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.011170875458272507,
      "loss": 0.2314,
      "num_input_tokens_seen": 23939552,
      "step": 113435
    },
    {
      "epoch": 12.479647964796479,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01116948314599436,
      "loss": 0.2324,
      "num_input_tokens_seen": 23940576,
      "step": 113440
    },
    {
      "epoch": 12.48019801980198,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.011168090869020243,
      "loss": 0.2299,
      "num_input_tokens_seen": 23941664,
      "step": 113445
    },
    {
      "epoch": 12.480748074807481,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011166698627363,
      "loss": 0.2304,
      "num_input_tokens_seen": 23942720,
      "step": 113450
    },
    {
      "epoch": 12.48129812981298,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.011165306421035451,
      "loss": 0.2309,
      "num_input_tokens_seen": 23943808,
      "step": 113455
    },
    {
      "epoch": 12.481848184818482,
      "grad_norm": 0.000652313232421875,
      "learning_rate": 0.011163914250050434,
      "loss": 0.2303,
      "num_input_tokens_seen": 23944832,
      "step": 113460
    },
    {
      "epoch": 12.482398239823983,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011162522114420783,
      "loss": 0.2335,
      "num_input_tokens_seen": 23945920,
      "step": 113465
    },
    {
      "epoch": 12.482948294829482,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.01116113001415932,
      "loss": 0.2319,
      "num_input_tokens_seen": 23947008,
      "step": 113470
    },
    {
      "epoch": 12.483498349834983,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01115973794927888,
      "loss": 0.2324,
      "num_input_tokens_seen": 23948032,
      "step": 113475
    },
    {
      "epoch": 12.484048404840484,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011158345919792291,
      "loss": 0.2309,
      "num_input_tokens_seen": 23949120,
      "step": 113480
    },
    {
      "epoch": 12.484598459845985,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.011156953925712391,
      "loss": 0.2303,
      "num_input_tokens_seen": 23950144,
      "step": 113485
    },
    {
      "epoch": 12.485148514851485,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.011155561967051999,
      "loss": 0.2324,
      "num_input_tokens_seen": 23951200,
      "step": 113490
    },
    {
      "epoch": 12.485698569856986,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.011154170043823944,
      "loss": 0.2319,
      "num_input_tokens_seen": 23952256,
      "step": 113495
    },
    {
      "epoch": 12.486248624862487,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011152778156041064,
      "loss": 0.2303,
      "num_input_tokens_seen": 23953344,
      "step": 113500
    },
    {
      "epoch": 12.486798679867986,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.011151386303716176,
      "loss": 0.2319,
      "num_input_tokens_seen": 23954368,
      "step": 113505
    },
    {
      "epoch": 12.487348734873487,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011149994486862118,
      "loss": 0.2335,
      "num_input_tokens_seen": 23955456,
      "step": 113510
    },
    {
      "epoch": 12.487898789878988,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.011148602705491716,
      "loss": 0.2314,
      "num_input_tokens_seen": 23956544,
      "step": 113515
    },
    {
      "epoch": 12.488448844884488,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.011147210959617782,
      "loss": 0.2309,
      "num_input_tokens_seen": 23957536,
      "step": 113520
    },
    {
      "epoch": 12.488998899889989,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.01114581924925317,
      "loss": 0.2314,
      "num_input_tokens_seen": 23958592,
      "step": 113525
    },
    {
      "epoch": 12.48954895489549,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.011144427574410686,
      "loss": 0.2309,
      "num_input_tokens_seen": 23959712,
      "step": 113530
    },
    {
      "epoch": 12.490099009900991,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01114303593510316,
      "loss": 0.2319,
      "num_input_tokens_seen": 23960768,
      "step": 113535
    },
    {
      "epoch": 12.49064906490649,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01114164433134343,
      "loss": 0.233,
      "num_input_tokens_seen": 23961824,
      "step": 113540
    },
    {
      "epoch": 12.491199119911991,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.011140252763144304,
      "loss": 0.2329,
      "num_input_tokens_seen": 23962912,
      "step": 113545
    },
    {
      "epoch": 12.491749174917492,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.011138861230518623,
      "loss": 0.233,
      "num_input_tokens_seen": 23964000,
      "step": 113550
    },
    {
      "epoch": 12.492299229922992,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.011137469733479205,
      "loss": 0.2324,
      "num_input_tokens_seen": 23964992,
      "step": 113555
    },
    {
      "epoch": 12.492849284928493,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.01113607827203887,
      "loss": 0.2314,
      "num_input_tokens_seen": 23966048,
      "step": 113560
    },
    {
      "epoch": 12.493399339933994,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01113468684621045,
      "loss": 0.2304,
      "num_input_tokens_seen": 23967136,
      "step": 113565
    },
    {
      "epoch": 12.493949394939493,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.011133295456006763,
      "loss": 0.2309,
      "num_input_tokens_seen": 23968192,
      "step": 113570
    },
    {
      "epoch": 12.494499449944994,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.011131904101440639,
      "loss": 0.234,
      "num_input_tokens_seen": 23969216,
      "step": 113575
    },
    {
      "epoch": 12.495049504950495,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011130512782524905,
      "loss": 0.2335,
      "num_input_tokens_seen": 23970336,
      "step": 113580
    },
    {
      "epoch": 12.495599559955995,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.011129121499272369,
      "loss": 0.2293,
      "num_input_tokens_seen": 23971392,
      "step": 113585
    },
    {
      "epoch": 12.496149614961496,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011127730251695866,
      "loss": 0.2324,
      "num_input_tokens_seen": 23972544,
      "step": 113590
    },
    {
      "epoch": 12.496699669966997,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.011126339039808213,
      "loss": 0.2309,
      "num_input_tokens_seen": 23973600,
      "step": 113595
    },
    {
      "epoch": 12.497249724972498,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.011124947863622241,
      "loss": 0.2293,
      "num_input_tokens_seen": 23974656,
      "step": 113600
    },
    {
      "epoch": 12.497799779977997,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.01112355672315076,
      "loss": 0.2319,
      "num_input_tokens_seen": 23975712,
      "step": 113605
    },
    {
      "epoch": 12.498349834983498,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011122165618406597,
      "loss": 0.2277,
      "num_input_tokens_seen": 23976736,
      "step": 113610
    },
    {
      "epoch": 12.498899889989,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011120774549402575,
      "loss": 0.2314,
      "num_input_tokens_seen": 23977728,
      "step": 113615
    },
    {
      "epoch": 12.499449944994499,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011119383516151507,
      "loss": 0.2298,
      "num_input_tokens_seen": 23978784,
      "step": 113620
    },
    {
      "epoch": 12.5,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.011117992518666225,
      "loss": 0.2324,
      "num_input_tokens_seen": 23979808,
      "step": 113625
    },
    {
      "epoch": 12.500550055005501,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011116601556959543,
      "loss": 0.2314,
      "num_input_tokens_seen": 23980832,
      "step": 113630
    },
    {
      "epoch": 12.501100110011,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.011115210631044274,
      "loss": 0.2329,
      "num_input_tokens_seen": 23981856,
      "step": 113635
    },
    {
      "epoch": 12.501650165016502,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.011113819740933251,
      "loss": 0.234,
      "num_input_tokens_seen": 23982880,
      "step": 113640
    },
    {
      "epoch": 12.502200220022003,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.011112428886639285,
      "loss": 0.2314,
      "num_input_tokens_seen": 23983968,
      "step": 113645
    },
    {
      "epoch": 12.502750275027502,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.011111038068175195,
      "loss": 0.2319,
      "num_input_tokens_seen": 23984992,
      "step": 113650
    },
    {
      "epoch": 12.503300330033003,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.011109647285553807,
      "loss": 0.2314,
      "num_input_tokens_seen": 23986048,
      "step": 113655
    },
    {
      "epoch": 12.503850385038504,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011108256538787924,
      "loss": 0.2329,
      "num_input_tokens_seen": 23987040,
      "step": 113660
    },
    {
      "epoch": 12.504400440044005,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01110686582789038,
      "loss": 0.2319,
      "num_input_tokens_seen": 23988096,
      "step": 113665
    },
    {
      "epoch": 12.504950495049505,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.011105475152873981,
      "loss": 0.2308,
      "num_input_tokens_seen": 23989088,
      "step": 113670
    },
    {
      "epoch": 12.505500550055006,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011104084513751554,
      "loss": 0.2324,
      "num_input_tokens_seen": 23990112,
      "step": 113675
    },
    {
      "epoch": 12.506050605060507,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.011102693910535907,
      "loss": 0.2309,
      "num_input_tokens_seen": 23991136,
      "step": 113680
    },
    {
      "epoch": 12.506600660066006,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01110130334323986,
      "loss": 0.2298,
      "num_input_tokens_seen": 23992192,
      "step": 113685
    },
    {
      "epoch": 12.507150715071507,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011099912811876233,
      "loss": 0.2314,
      "num_input_tokens_seen": 23993216,
      "step": 113690
    },
    {
      "epoch": 12.507700770077008,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.01109852231645783,
      "loss": 0.2298,
      "num_input_tokens_seen": 23994240,
      "step": 113695
    },
    {
      "epoch": 12.508250825082508,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.011097131856997483,
      "loss": 0.2309,
      "num_input_tokens_seen": 23995296,
      "step": 113700
    },
    {
      "epoch": 12.508800880088009,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011095741433507996,
      "loss": 0.2303,
      "num_input_tokens_seen": 23996320,
      "step": 113705
    },
    {
      "epoch": 12.50935093509351,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.011094351046002184,
      "loss": 0.2314,
      "num_input_tokens_seen": 23997312,
      "step": 113710
    },
    {
      "epoch": 12.509900990099009,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.011092960694492872,
      "loss": 0.2303,
      "num_input_tokens_seen": 23998464,
      "step": 113715
    },
    {
      "epoch": 12.51045104510451,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.01109157037899286,
      "loss": 0.2309,
      "num_input_tokens_seen": 23999456,
      "step": 113720
    },
    {
      "epoch": 12.511001100110011,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.011090180099514969,
      "loss": 0.2314,
      "num_input_tokens_seen": 24000512,
      "step": 113725
    },
    {
      "epoch": 12.511551155115512,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.011088789856072015,
      "loss": 0.2293,
      "num_input_tokens_seen": 24001568,
      "step": 113730
    },
    {
      "epoch": 12.512101210121012,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.011087399648676803,
      "loss": 0.2314,
      "num_input_tokens_seen": 24002656,
      "step": 113735
    },
    {
      "epoch": 12.512651265126513,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.011086009477342156,
      "loss": 0.2293,
      "num_input_tokens_seen": 24003712,
      "step": 113740
    },
    {
      "epoch": 12.513201320132014,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.011084619342080881,
      "loss": 0.2314,
      "num_input_tokens_seen": 24004864,
      "step": 113745
    },
    {
      "epoch": 12.513751375137513,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.011083229242905786,
      "loss": 0.2319,
      "num_input_tokens_seen": 24005920,
      "step": 113750
    },
    {
      "epoch": 12.514301430143014,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.011081839179829692,
      "loss": 0.2314,
      "num_input_tokens_seen": 24006976,
      "step": 113755
    },
    {
      "epoch": 12.514851485148515,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.011080449152865402,
      "loss": 0.2319,
      "num_input_tokens_seen": 24008096,
      "step": 113760
    },
    {
      "epoch": 12.515401540154015,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011079059162025735,
      "loss": 0.2319,
      "num_input_tokens_seen": 24009120,
      "step": 113765
    },
    {
      "epoch": 12.515951595159516,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0110776692073235,
      "loss": 0.2324,
      "num_input_tokens_seen": 24010208,
      "step": 113770
    },
    {
      "epoch": 12.516501650165017,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.011076279288771501,
      "loss": 0.2314,
      "num_input_tokens_seen": 24011200,
      "step": 113775
    },
    {
      "epoch": 12.517051705170516,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.011074889406382557,
      "loss": 0.2304,
      "num_input_tokens_seen": 24012320,
      "step": 113780
    },
    {
      "epoch": 12.517601760176017,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.011073499560169469,
      "loss": 0.2309,
      "num_input_tokens_seen": 24013440,
      "step": 113785
    },
    {
      "epoch": 12.518151815181518,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.011072109750145057,
      "loss": 0.2319,
      "num_input_tokens_seen": 24014496,
      "step": 113790
    },
    {
      "epoch": 12.51870187018702,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.011070719976322124,
      "loss": 0.2319,
      "num_input_tokens_seen": 24015520,
      "step": 113795
    },
    {
      "epoch": 12.519251925192519,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.011069330238713475,
      "loss": 0.2335,
      "num_input_tokens_seen": 24016672,
      "step": 113800
    },
    {
      "epoch": 12.51980198019802,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01106794053733193,
      "loss": 0.2319,
      "num_input_tokens_seen": 24017728,
      "step": 113805
    },
    {
      "epoch": 12.520352035203521,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.011066550872190283,
      "loss": 0.2314,
      "num_input_tokens_seen": 24018752,
      "step": 113810
    },
    {
      "epoch": 12.52090209020902,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.011065161243301352,
      "loss": 0.2308,
      "num_input_tokens_seen": 24019744,
      "step": 113815
    },
    {
      "epoch": 12.521452145214521,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.011063771650677942,
      "loss": 0.2308,
      "num_input_tokens_seen": 24020800,
      "step": 113820
    },
    {
      "epoch": 12.522002200220022,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.011062382094332857,
      "loss": 0.2309,
      "num_input_tokens_seen": 24021888,
      "step": 113825
    },
    {
      "epoch": 12.522552255225522,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.011060992574278916,
      "loss": 0.2319,
      "num_input_tokens_seen": 24022912,
      "step": 113830
    },
    {
      "epoch": 12.523102310231023,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.011059603090528905,
      "loss": 0.2324,
      "num_input_tokens_seen": 24023968,
      "step": 113835
    },
    {
      "epoch": 12.523652365236524,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.011058213643095649,
      "loss": 0.2319,
      "num_input_tokens_seen": 24024992,
      "step": 113840
    },
    {
      "epoch": 12.524202420242025,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.011056824231991946,
      "loss": 0.2319,
      "num_input_tokens_seen": 24026016,
      "step": 113845
    },
    {
      "epoch": 12.524752475247524,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.011055434857230598,
      "loss": 0.2314,
      "num_input_tokens_seen": 24027072,
      "step": 113850
    },
    {
      "epoch": 12.525302530253025,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.011054045518824416,
      "loss": 0.2329,
      "num_input_tokens_seen": 24028096,
      "step": 113855
    },
    {
      "epoch": 12.525852585258527,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.011052656216786201,
      "loss": 0.2298,
      "num_input_tokens_seen": 24029088,
      "step": 113860
    },
    {
      "epoch": 12.526402640264026,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011051266951128763,
      "loss": 0.2313,
      "num_input_tokens_seen": 24030208,
      "step": 113865
    },
    {
      "epoch": 12.526952695269527,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.011049877721864904,
      "loss": 0.2319,
      "num_input_tokens_seen": 24031328,
      "step": 113870
    },
    {
      "epoch": 12.527502750275028,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.011048488529007422,
      "loss": 0.2309,
      "num_input_tokens_seen": 24032416,
      "step": 113875
    },
    {
      "epoch": 12.528052805280527,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.011047099372569126,
      "loss": 0.234,
      "num_input_tokens_seen": 24033440,
      "step": 113880
    },
    {
      "epoch": 12.528602860286028,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.011045710252562815,
      "loss": 0.2303,
      "num_input_tokens_seen": 24034592,
      "step": 113885
    },
    {
      "epoch": 12.52915291529153,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.011044321169001305,
      "loss": 0.2324,
      "num_input_tokens_seen": 24035680,
      "step": 113890
    },
    {
      "epoch": 12.52970297029703,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011042932121897383,
      "loss": 0.2329,
      "num_input_tokens_seen": 24036704,
      "step": 113895
    },
    {
      "epoch": 12.53025302530253,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.011041543111263855,
      "loss": 0.2319,
      "num_input_tokens_seen": 24037760,
      "step": 113900
    },
    {
      "epoch": 12.530803080308031,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.011040154137113529,
      "loss": 0.2319,
      "num_input_tokens_seen": 24038816,
      "step": 113905
    },
    {
      "epoch": 12.531353135313532,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.011038765199459202,
      "loss": 0.2314,
      "num_input_tokens_seen": 24039872,
      "step": 113910
    },
    {
      "epoch": 12.531903190319031,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.011037376298313669,
      "loss": 0.2309,
      "num_input_tokens_seen": 24040992,
      "step": 113915
    },
    {
      "epoch": 12.532453245324533,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.011035987433689746,
      "loss": 0.2309,
      "num_input_tokens_seen": 24042048,
      "step": 113920
    },
    {
      "epoch": 12.533003300330034,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.011034598605600217,
      "loss": 0.2319,
      "num_input_tokens_seen": 24043136,
      "step": 113925
    },
    {
      "epoch": 12.533553355335533,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.011033209814057894,
      "loss": 0.2319,
      "num_input_tokens_seen": 24044192,
      "step": 113930
    },
    {
      "epoch": 12.534103410341034,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.011031821059075575,
      "loss": 0.2313,
      "num_input_tokens_seen": 24045248,
      "step": 113935
    },
    {
      "epoch": 12.534653465346535,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.011030432340666049,
      "loss": 0.2314,
      "num_input_tokens_seen": 24046368,
      "step": 113940
    },
    {
      "epoch": 12.535203520352034,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.01102904365884213,
      "loss": 0.2324,
      "num_input_tokens_seen": 24047392,
      "step": 113945
    },
    {
      "epoch": 12.535753575357536,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.011027655013616607,
      "loss": 0.2309,
      "num_input_tokens_seen": 24048544,
      "step": 113950
    },
    {
      "epoch": 12.536303630363037,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.011026266405002283,
      "loss": 0.2324,
      "num_input_tokens_seen": 24049632,
      "step": 113955
    },
    {
      "epoch": 12.536853685368538,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.011024877833011959,
      "loss": 0.2319,
      "num_input_tokens_seen": 24050688,
      "step": 113960
    },
    {
      "epoch": 12.537403740374037,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.01102348929765842,
      "loss": 0.2288,
      "num_input_tokens_seen": 24051680,
      "step": 113965
    },
    {
      "epoch": 12.537953795379538,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.011022100798954477,
      "loss": 0.2319,
      "num_input_tokens_seen": 24052736,
      "step": 113970
    },
    {
      "epoch": 12.53850385038504,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.01102071233691292,
      "loss": 0.2314,
      "num_input_tokens_seen": 24053760,
      "step": 113975
    },
    {
      "epoch": 12.539053905390539,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.011019323911546551,
      "loss": 0.2298,
      "num_input_tokens_seen": 24054720,
      "step": 113980
    },
    {
      "epoch": 12.53960396039604,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.011017935522868163,
      "loss": 0.2329,
      "num_input_tokens_seen": 24055712,
      "step": 113985
    },
    {
      "epoch": 12.54015401540154,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01101654717089055,
      "loss": 0.2335,
      "num_input_tokens_seen": 24056800,
      "step": 113990
    },
    {
      "epoch": 12.54070407040704,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.011015158855626513,
      "loss": 0.2314,
      "num_input_tokens_seen": 24057824,
      "step": 113995
    },
    {
      "epoch": 12.541254125412541,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.011013770577088838,
      "loss": 0.2329,
      "num_input_tokens_seen": 24058880,
      "step": 114000
    },
    {
      "epoch": 12.541804180418042,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.011012382335290337,
      "loss": 0.2314,
      "num_input_tokens_seen": 24059904,
      "step": 114005
    },
    {
      "epoch": 12.542354235423542,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.011010994130243788,
      "loss": 0.2314,
      "num_input_tokens_seen": 24060960,
      "step": 114010
    },
    {
      "epoch": 12.542904290429043,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.011009605961961991,
      "loss": 0.2303,
      "num_input_tokens_seen": 24062048,
      "step": 114015
    },
    {
      "epoch": 12.543454345434544,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.011008217830457748,
      "loss": 0.2319,
      "num_input_tokens_seen": 24063168,
      "step": 114020
    },
    {
      "epoch": 12.544004400440045,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.011006829735743838,
      "loss": 0.2308,
      "num_input_tokens_seen": 24064224,
      "step": 114025
    },
    {
      "epoch": 12.544554455445544,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.011005441677833067,
      "loss": 0.2309,
      "num_input_tokens_seen": 24065248,
      "step": 114030
    },
    {
      "epoch": 12.545104510451045,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.011004053656738226,
      "loss": 0.2324,
      "num_input_tokens_seen": 24066304,
      "step": 114035
    },
    {
      "epoch": 12.545654565456546,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.011002665672472097,
      "loss": 0.2298,
      "num_input_tokens_seen": 24067360,
      "step": 114040
    },
    {
      "epoch": 12.546204620462046,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.011001277725047486,
      "loss": 0.2303,
      "num_input_tokens_seen": 24068416,
      "step": 114045
    },
    {
      "epoch": 12.546754675467547,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.010999889814477175,
      "loss": 0.2303,
      "num_input_tokens_seen": 24069504,
      "step": 114050
    },
    {
      "epoch": 12.547304730473048,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.010998501940773966,
      "loss": 0.2314,
      "num_input_tokens_seen": 24070624,
      "step": 114055
    },
    {
      "epoch": 12.547854785478547,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.010997114103950642,
      "loss": 0.2298,
      "num_input_tokens_seen": 24071648,
      "step": 114060
    },
    {
      "epoch": 12.548404840484048,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.010995726304019994,
      "loss": 0.2309,
      "num_input_tokens_seen": 24072736,
      "step": 114065
    },
    {
      "epoch": 12.54895489548955,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.01099433854099482,
      "loss": 0.2319,
      "num_input_tokens_seen": 24073856,
      "step": 114070
    },
    {
      "epoch": 12.549504950495049,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.010992950814887898,
      "loss": 0.2304,
      "num_input_tokens_seen": 24074912,
      "step": 114075
    },
    {
      "epoch": 12.55005500550055,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.010991563125712034,
      "loss": 0.2319,
      "num_input_tokens_seen": 24075968,
      "step": 114080
    },
    {
      "epoch": 12.55060506050605,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.010990175473480007,
      "loss": 0.2293,
      "num_input_tokens_seen": 24076992,
      "step": 114085
    },
    {
      "epoch": 12.551155115511552,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010988787858204604,
      "loss": 0.2303,
      "num_input_tokens_seen": 24077984,
      "step": 114090
    },
    {
      "epoch": 12.551705170517051,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.010987400279898625,
      "loss": 0.2314,
      "num_input_tokens_seen": 24079072,
      "step": 114095
    },
    {
      "epoch": 12.552255225522552,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.01098601273857485,
      "loss": 0.234,
      "num_input_tokens_seen": 24080160,
      "step": 114100
    },
    {
      "epoch": 12.552805280528053,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.010984625234246065,
      "loss": 0.2293,
      "num_input_tokens_seen": 24081216,
      "step": 114105
    },
    {
      "epoch": 12.553355335533553,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.01098323776692507,
      "loss": 0.2303,
      "num_input_tokens_seen": 24082240,
      "step": 114110
    },
    {
      "epoch": 12.553905390539054,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.010981850336624634,
      "loss": 0.2329,
      "num_input_tokens_seen": 24083264,
      "step": 114115
    },
    {
      "epoch": 12.554455445544555,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.010980462943357567,
      "loss": 0.2309,
      "num_input_tokens_seen": 24084288,
      "step": 114120
    },
    {
      "epoch": 12.555005500550054,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01097907558713664,
      "loss": 0.2308,
      "num_input_tokens_seen": 24085344,
      "step": 114125
    },
    {
      "epoch": 12.555555555555555,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01097768826797464,
      "loss": 0.2314,
      "num_input_tokens_seen": 24086400,
      "step": 114130
    },
    {
      "epoch": 12.556105610561056,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.010976300985884365,
      "loss": 0.2303,
      "num_input_tokens_seen": 24087424,
      "step": 114135
    },
    {
      "epoch": 12.556655665566556,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.010974913740878585,
      "loss": 0.233,
      "num_input_tokens_seen": 24088480,
      "step": 114140
    },
    {
      "epoch": 12.557205720572057,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.010973526532970098,
      "loss": 0.2319,
      "num_input_tokens_seen": 24089472,
      "step": 114145
    },
    {
      "epoch": 12.557755775577558,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010972139362171689,
      "loss": 0.234,
      "num_input_tokens_seen": 24090464,
      "step": 114150
    },
    {
      "epoch": 12.558305830583059,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.01097075222849613,
      "loss": 0.2319,
      "num_input_tokens_seen": 24091456,
      "step": 114155
    },
    {
      "epoch": 12.558855885588558,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.010969365131956219,
      "loss": 0.2309,
      "num_input_tokens_seen": 24092544,
      "step": 114160
    },
    {
      "epoch": 12.55940594059406,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.010967978072564733,
      "loss": 0.2329,
      "num_input_tokens_seen": 24093536,
      "step": 114165
    },
    {
      "epoch": 12.55995599559956,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.010966591050334464,
      "loss": 0.2308,
      "num_input_tokens_seen": 24094624,
      "step": 114170
    },
    {
      "epoch": 12.56050605060506,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.010965204065278187,
      "loss": 0.2314,
      "num_input_tokens_seen": 24095680,
      "step": 114175
    },
    {
      "epoch": 12.561056105610561,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.010963817117408684,
      "loss": 0.2303,
      "num_input_tokens_seen": 24096736,
      "step": 114180
    },
    {
      "epoch": 12.561606160616062,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.010962430206738745,
      "loss": 0.2319,
      "num_input_tokens_seen": 24097728,
      "step": 114185
    },
    {
      "epoch": 12.562156215621561,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010961043333281144,
      "loss": 0.2308,
      "num_input_tokens_seen": 24098816,
      "step": 114190
    },
    {
      "epoch": 12.562706270627062,
      "grad_norm": 0.000888824462890625,
      "learning_rate": 0.010959656497048678,
      "loss": 0.2314,
      "num_input_tokens_seen": 24099936,
      "step": 114195
    },
    {
      "epoch": 12.563256325632564,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.010958269698054116,
      "loss": 0.2324,
      "num_input_tokens_seen": 24101056,
      "step": 114200
    },
    {
      "epoch": 12.563806380638063,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01095688293631024,
      "loss": 0.2308,
      "num_input_tokens_seen": 24102112,
      "step": 114205
    },
    {
      "epoch": 12.564356435643564,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.010955496211829837,
      "loss": 0.2324,
      "num_input_tokens_seen": 24103136,
      "step": 114210
    },
    {
      "epoch": 12.564906490649065,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010954109524625682,
      "loss": 0.2314,
      "num_input_tokens_seen": 24104192,
      "step": 114215
    },
    {
      "epoch": 12.565456545654566,
      "grad_norm": 0.0030364990234375,
      "learning_rate": 0.010952722874710559,
      "loss": 0.2303,
      "num_input_tokens_seen": 24105344,
      "step": 114220
    },
    {
      "epoch": 12.566006600660065,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.010951336262097253,
      "loss": 0.2298,
      "num_input_tokens_seen": 24106400,
      "step": 114225
    },
    {
      "epoch": 12.566556655665567,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.01094994968679853,
      "loss": 0.2314,
      "num_input_tokens_seen": 24107456,
      "step": 114230
    },
    {
      "epoch": 12.567106710671068,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.01094856314882718,
      "loss": 0.2319,
      "num_input_tokens_seen": 24108576,
      "step": 114235
    },
    {
      "epoch": 12.567656765676567,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.010947176648195977,
      "loss": 0.2324,
      "num_input_tokens_seen": 24109600,
      "step": 114240
    },
    {
      "epoch": 12.568206820682068,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.010945790184917705,
      "loss": 0.2309,
      "num_input_tokens_seen": 24110656,
      "step": 114245
    },
    {
      "epoch": 12.56875687568757,
      "grad_norm": 0.000652313232421875,
      "learning_rate": 0.010944403759005144,
      "loss": 0.2308,
      "num_input_tokens_seen": 24111712,
      "step": 114250
    },
    {
      "epoch": 12.569306930693068,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.010943017370471058,
      "loss": 0.2303,
      "num_input_tokens_seen": 24112832,
      "step": 114255
    },
    {
      "epoch": 12.56985698569857,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.010941631019328239,
      "loss": 0.2308,
      "num_input_tokens_seen": 24113888,
      "step": 114260
    },
    {
      "epoch": 12.57040704070407,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010940244705589455,
      "loss": 0.2319,
      "num_input_tokens_seen": 24115008,
      "step": 114265
    },
    {
      "epoch": 12.570957095709572,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.010938858429267498,
      "loss": 0.2314,
      "num_input_tokens_seen": 24116032,
      "step": 114270
    },
    {
      "epoch": 12.571507150715071,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.010937472190375126,
      "loss": 0.2335,
      "num_input_tokens_seen": 24117088,
      "step": 114275
    },
    {
      "epoch": 12.572057205720572,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.010936085988925122,
      "loss": 0.2308,
      "num_input_tokens_seen": 24118112,
      "step": 114280
    },
    {
      "epoch": 12.572607260726073,
      "grad_norm": 0.01007080078125,
      "learning_rate": 0.01093469982493027,
      "loss": 0.2293,
      "num_input_tokens_seen": 24119168,
      "step": 114285
    },
    {
      "epoch": 12.573157315731573,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.010933313698403337,
      "loss": 0.2309,
      "num_input_tokens_seen": 24120160,
      "step": 114290
    },
    {
      "epoch": 12.573707370737074,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.010931927609357095,
      "loss": 0.2319,
      "num_input_tokens_seen": 24121248,
      "step": 114295
    },
    {
      "epoch": 12.574257425742575,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010930541557804332,
      "loss": 0.2324,
      "num_input_tokens_seen": 24122240,
      "step": 114300
    },
    {
      "epoch": 12.574807480748074,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.010929155543757806,
      "loss": 0.2303,
      "num_input_tokens_seen": 24123360,
      "step": 114305
    },
    {
      "epoch": 12.575357535753575,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.010927769567230306,
      "loss": 0.2304,
      "num_input_tokens_seen": 24124448,
      "step": 114310
    },
    {
      "epoch": 12.575907590759076,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.010926383628234601,
      "loss": 0.2319,
      "num_input_tokens_seen": 24125568,
      "step": 114315
    },
    {
      "epoch": 12.576457645764577,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010924997726783457,
      "loss": 0.2309,
      "num_input_tokens_seen": 24126624,
      "step": 114320
    },
    {
      "epoch": 12.577007700770077,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.010923611862889662,
      "loss": 0.2314,
      "num_input_tokens_seen": 24127648,
      "step": 114325
    },
    {
      "epoch": 12.577557755775578,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010922226036565974,
      "loss": 0.2304,
      "num_input_tokens_seen": 24128736,
      "step": 114330
    },
    {
      "epoch": 12.578107810781079,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.010920840247825175,
      "loss": 0.2324,
      "num_input_tokens_seen": 24129728,
      "step": 114335
    },
    {
      "epoch": 12.578657865786578,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.010919454496680034,
      "loss": 0.2309,
      "num_input_tokens_seen": 24130816,
      "step": 114340
    },
    {
      "epoch": 12.57920792079208,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.01091806878314332,
      "loss": 0.2324,
      "num_input_tokens_seen": 24131840,
      "step": 114345
    },
    {
      "epoch": 12.57975797579758,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.01091668310722781,
      "loss": 0.2314,
      "num_input_tokens_seen": 24132896,
      "step": 114350
    },
    {
      "epoch": 12.58030803080308,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.010915297468946268,
      "loss": 0.2308,
      "num_input_tokens_seen": 24133952,
      "step": 114355
    },
    {
      "epoch": 12.58085808580858,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.010913911868311473,
      "loss": 0.233,
      "num_input_tokens_seen": 24135008,
      "step": 114360
    },
    {
      "epoch": 12.581408140814082,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010912526305336194,
      "loss": 0.2309,
      "num_input_tokens_seen": 24136128,
      "step": 114365
    },
    {
      "epoch": 12.581958195819581,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.010911140780033194,
      "loss": 0.2288,
      "num_input_tokens_seen": 24137216,
      "step": 114370
    },
    {
      "epoch": 12.582508250825082,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.01090975529241525,
      "loss": 0.2314,
      "num_input_tokens_seen": 24138208,
      "step": 114375
    },
    {
      "epoch": 12.583058305830583,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010908369842495127,
      "loss": 0.2319,
      "num_input_tokens_seen": 24139264,
      "step": 114380
    },
    {
      "epoch": 12.583608360836084,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0109069844302856,
      "loss": 0.2288,
      "num_input_tokens_seen": 24140288,
      "step": 114385
    },
    {
      "epoch": 12.584158415841584,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.010905599055799431,
      "loss": 0.2304,
      "num_input_tokens_seen": 24141408,
      "step": 114390
    },
    {
      "epoch": 12.584708470847085,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.010904213719049385,
      "loss": 0.2319,
      "num_input_tokens_seen": 24142464,
      "step": 114395
    },
    {
      "epoch": 12.585258525852586,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.010902828420048246,
      "loss": 0.2298,
      "num_input_tokens_seen": 24143488,
      "step": 114400
    },
    {
      "epoch": 12.585808580858085,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010901443158808765,
      "loss": 0.2294,
      "num_input_tokens_seen": 24144608,
      "step": 114405
    },
    {
      "epoch": 12.586358635863586,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010900057935343717,
      "loss": 0.2319,
      "num_input_tokens_seen": 24145728,
      "step": 114410
    },
    {
      "epoch": 12.586908690869087,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010898672749665871,
      "loss": 0.2309,
      "num_input_tokens_seen": 24146720,
      "step": 114415
    },
    {
      "epoch": 12.587458745874587,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.010897287601787983,
      "loss": 0.2346,
      "num_input_tokens_seen": 24147776,
      "step": 114420
    },
    {
      "epoch": 12.588008800880088,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.010895902491722833,
      "loss": 0.2304,
      "num_input_tokens_seen": 24148864,
      "step": 114425
    },
    {
      "epoch": 12.588558855885589,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010894517419483175,
      "loss": 0.2278,
      "num_input_tokens_seen": 24149888,
      "step": 114430
    },
    {
      "epoch": 12.589108910891088,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010893132385081782,
      "loss": 0.2329,
      "num_input_tokens_seen": 24150880,
      "step": 114435
    },
    {
      "epoch": 12.58965896589659,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.010891747388531423,
      "loss": 0.2335,
      "num_input_tokens_seen": 24151968,
      "step": 114440
    },
    {
      "epoch": 12.59020902090209,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.01089036242984485,
      "loss": 0.2299,
      "num_input_tokens_seen": 24152992,
      "step": 114445
    },
    {
      "epoch": 12.590759075907592,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.010888977509034836,
      "loss": 0.2304,
      "num_input_tokens_seen": 24154080,
      "step": 114450
    },
    {
      "epoch": 12.591309130913091,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.010887592626114141,
      "loss": 0.233,
      "num_input_tokens_seen": 24155136,
      "step": 114455
    },
    {
      "epoch": 12.591859185918592,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.010886207781095539,
      "loss": 0.2319,
      "num_input_tokens_seen": 24156224,
      "step": 114460
    },
    {
      "epoch": 12.592409240924093,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.010884822973991782,
      "loss": 0.232,
      "num_input_tokens_seen": 24157344,
      "step": 114465
    },
    {
      "epoch": 12.592959295929592,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.010883438204815634,
      "loss": 0.2309,
      "num_input_tokens_seen": 24158400,
      "step": 114470
    },
    {
      "epoch": 12.593509350935093,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010882053473579867,
      "loss": 0.2303,
      "num_input_tokens_seen": 24159456,
      "step": 114475
    },
    {
      "epoch": 12.594059405940595,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.01088066878029723,
      "loss": 0.2299,
      "num_input_tokens_seen": 24160480,
      "step": 114480
    },
    {
      "epoch": 12.594609460946094,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.010879284124980498,
      "loss": 0.2356,
      "num_input_tokens_seen": 24161536,
      "step": 114485
    },
    {
      "epoch": 12.595159515951595,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.010877899507642425,
      "loss": 0.233,
      "num_input_tokens_seen": 24162624,
      "step": 114490
    },
    {
      "epoch": 12.595709570957096,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01087651492829577,
      "loss": 0.2319,
      "num_input_tokens_seen": 24163680,
      "step": 114495
    },
    {
      "epoch": 12.596259625962595,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010875130386953306,
      "loss": 0.2298,
      "num_input_tokens_seen": 24164672,
      "step": 114500
    },
    {
      "epoch": 12.596809680968097,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.010873745883627783,
      "loss": 0.2319,
      "num_input_tokens_seen": 24165696,
      "step": 114505
    },
    {
      "epoch": 12.597359735973598,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01087236141833196,
      "loss": 0.2319,
      "num_input_tokens_seen": 24166784,
      "step": 114510
    },
    {
      "epoch": 12.597909790979099,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.010870976991078609,
      "loss": 0.2298,
      "num_input_tokens_seen": 24167872,
      "step": 114515
    },
    {
      "epoch": 12.598459845984598,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.010869592601880475,
      "loss": 0.2309,
      "num_input_tokens_seen": 24168960,
      "step": 114520
    },
    {
      "epoch": 12.599009900990099,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.010868208250750326,
      "loss": 0.2314,
      "num_input_tokens_seen": 24169952,
      "step": 114525
    },
    {
      "epoch": 12.5995599559956,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.010866823937700922,
      "loss": 0.2309,
      "num_input_tokens_seen": 24171008,
      "step": 114530
    },
    {
      "epoch": 12.6001100110011,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.010865439662745013,
      "loss": 0.2335,
      "num_input_tokens_seen": 24172160,
      "step": 114535
    },
    {
      "epoch": 12.6006600660066,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.010864055425895365,
      "loss": 0.2293,
      "num_input_tokens_seen": 24173216,
      "step": 114540
    },
    {
      "epoch": 12.601210121012102,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.010862671227164732,
      "loss": 0.2314,
      "num_input_tokens_seen": 24174272,
      "step": 114545
    },
    {
      "epoch": 12.601760176017601,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.010861287066565875,
      "loss": 0.2319,
      "num_input_tokens_seen": 24175360,
      "step": 114550
    },
    {
      "epoch": 12.602310231023102,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.010859902944111554,
      "loss": 0.2293,
      "num_input_tokens_seen": 24176448,
      "step": 114555
    },
    {
      "epoch": 12.602860286028603,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.010858518859814512,
      "loss": 0.2314,
      "num_input_tokens_seen": 24177504,
      "step": 114560
    },
    {
      "epoch": 12.603410341034103,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.01085713481368752,
      "loss": 0.2319,
      "num_input_tokens_seen": 24178560,
      "step": 114565
    },
    {
      "epoch": 12.603960396039604,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.010855750805743324,
      "loss": 0.2319,
      "num_input_tokens_seen": 24179616,
      "step": 114570
    },
    {
      "epoch": 12.604510451045105,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.01085436683599469,
      "loss": 0.2304,
      "num_input_tokens_seen": 24180768,
      "step": 114575
    },
    {
      "epoch": 12.605060506050606,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.010852982904454364,
      "loss": 0.2304,
      "num_input_tokens_seen": 24181856,
      "step": 114580
    },
    {
      "epoch": 12.605610561056105,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.010851599011135105,
      "loss": 0.2319,
      "num_input_tokens_seen": 24182880,
      "step": 114585
    },
    {
      "epoch": 12.606160616061606,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01085021515604967,
      "loss": 0.2335,
      "num_input_tokens_seen": 24183904,
      "step": 114590
    },
    {
      "epoch": 12.606710671067107,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010848831339210805,
      "loss": 0.235,
      "num_input_tokens_seen": 24184992,
      "step": 114595
    },
    {
      "epoch": 12.607260726072607,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.010847447560631274,
      "loss": 0.2299,
      "num_input_tokens_seen": 24186080,
      "step": 114600
    },
    {
      "epoch": 12.607810781078108,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.010846063820323826,
      "loss": 0.2319,
      "num_input_tokens_seen": 24187136,
      "step": 114605
    },
    {
      "epoch": 12.608360836083609,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.010844680118301211,
      "loss": 0.2304,
      "num_input_tokens_seen": 24188192,
      "step": 114610
    },
    {
      "epoch": 12.608910891089108,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.010843296454576191,
      "loss": 0.2288,
      "num_input_tokens_seen": 24189216,
      "step": 114615
    },
    {
      "epoch": 12.60946094609461,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.010841912829161507,
      "loss": 0.2314,
      "num_input_tokens_seen": 24190272,
      "step": 114620
    },
    {
      "epoch": 12.61001100110011,
      "grad_norm": 0.0004425048828125,
      "learning_rate": 0.010840529242069921,
      "loss": 0.2283,
      "num_input_tokens_seen": 24191328,
      "step": 114625
    },
    {
      "epoch": 12.61056105610561,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.010839145693314185,
      "loss": 0.2335,
      "num_input_tokens_seen": 24192384,
      "step": 114630
    },
    {
      "epoch": 12.61111111111111,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.01083776218290704,
      "loss": 0.2314,
      "num_input_tokens_seen": 24193440,
      "step": 114635
    },
    {
      "epoch": 12.611661166116612,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.010836378710861246,
      "loss": 0.2324,
      "num_input_tokens_seen": 24194528,
      "step": 114640
    },
    {
      "epoch": 12.612211221122113,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.010834995277189547,
      "loss": 0.2298,
      "num_input_tokens_seen": 24195616,
      "step": 114645
    },
    {
      "epoch": 12.612761276127612,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010833611881904705,
      "loss": 0.2309,
      "num_input_tokens_seen": 24196704,
      "step": 114650
    },
    {
      "epoch": 12.613311331133113,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.010832228525019461,
      "loss": 0.233,
      "num_input_tokens_seen": 24197760,
      "step": 114655
    },
    {
      "epoch": 12.613861386138614,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.010830845206546563,
      "loss": 0.2303,
      "num_input_tokens_seen": 24198784,
      "step": 114660
    },
    {
      "epoch": 12.614411441144114,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.010829461926498767,
      "loss": 0.2319,
      "num_input_tokens_seen": 24199872,
      "step": 114665
    },
    {
      "epoch": 12.614961496149615,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.010828078684888817,
      "loss": 0.2309,
      "num_input_tokens_seen": 24201024,
      "step": 114670
    },
    {
      "epoch": 12.615511551155116,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010826695481729471,
      "loss": 0.2319,
      "num_input_tokens_seen": 24202112,
      "step": 114675
    },
    {
      "epoch": 12.616061606160617,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.010825312317033466,
      "loss": 0.2303,
      "num_input_tokens_seen": 24203200,
      "step": 114680
    },
    {
      "epoch": 12.616611661166116,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.01082392919081355,
      "loss": 0.2304,
      "num_input_tokens_seen": 24204224,
      "step": 114685
    },
    {
      "epoch": 12.617161716171617,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.010822546103082484,
      "loss": 0.2298,
      "num_input_tokens_seen": 24205312,
      "step": 114690
    },
    {
      "epoch": 12.617711771177119,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.010821163053853,
      "loss": 0.2319,
      "num_input_tokens_seen": 24206432,
      "step": 114695
    },
    {
      "epoch": 12.618261826182618,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.01081978004313785,
      "loss": 0.233,
      "num_input_tokens_seen": 24207520,
      "step": 114700
    },
    {
      "epoch": 12.618811881188119,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.010818397070949785,
      "loss": 0.2309,
      "num_input_tokens_seen": 24208576,
      "step": 114705
    },
    {
      "epoch": 12.61936193619362,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.010817014137301543,
      "loss": 0.2329,
      "num_input_tokens_seen": 24209696,
      "step": 114710
    },
    {
      "epoch": 12.61991199119912,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.010815631242205878,
      "loss": 0.2335,
      "num_input_tokens_seen": 24210752,
      "step": 114715
    },
    {
      "epoch": 12.62046204620462,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010814248385675533,
      "loss": 0.2298,
      "num_input_tokens_seen": 24211712,
      "step": 114720
    },
    {
      "epoch": 12.621012101210122,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.010812865567723245,
      "loss": 0.2335,
      "num_input_tokens_seen": 24212832,
      "step": 114725
    },
    {
      "epoch": 12.62156215621562,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.010811482788361774,
      "loss": 0.2324,
      "num_input_tokens_seen": 24213920,
      "step": 114730
    },
    {
      "epoch": 12.622112211221122,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.010810100047603851,
      "loss": 0.2303,
      "num_input_tokens_seen": 24214944,
      "step": 114735
    },
    {
      "epoch": 12.622662266226623,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.010808717345462229,
      "loss": 0.2314,
      "num_input_tokens_seen": 24215936,
      "step": 114740
    },
    {
      "epoch": 12.623212321232124,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.01080733468194965,
      "loss": 0.2309,
      "num_input_tokens_seen": 24217024,
      "step": 114745
    },
    {
      "epoch": 12.623762376237623,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01080595205707885,
      "loss": 0.2314,
      "num_input_tokens_seen": 24218048,
      "step": 114750
    },
    {
      "epoch": 12.624312431243125,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.010804569470862578,
      "loss": 0.2314,
      "num_input_tokens_seen": 24219072,
      "step": 114755
    },
    {
      "epoch": 12.624862486248626,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010803186923313576,
      "loss": 0.2309,
      "num_input_tokens_seen": 24220160,
      "step": 114760
    },
    {
      "epoch": 12.625412541254125,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.010801804414444591,
      "loss": 0.233,
      "num_input_tokens_seen": 24221248,
      "step": 114765
    },
    {
      "epoch": 12.625962596259626,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.010800421944268357,
      "loss": 0.2314,
      "num_input_tokens_seen": 24222304,
      "step": 114770
    },
    {
      "epoch": 12.626512651265127,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.010799039512797617,
      "loss": 0.2319,
      "num_input_tokens_seen": 24223328,
      "step": 114775
    },
    {
      "epoch": 12.627062706270626,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.010797657120045118,
      "loss": 0.2319,
      "num_input_tokens_seen": 24224416,
      "step": 114780
    },
    {
      "epoch": 12.627612761276128,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.01079627476602359,
      "loss": 0.2314,
      "num_input_tokens_seen": 24225472,
      "step": 114785
    },
    {
      "epoch": 12.628162816281629,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.010794892450745787,
      "loss": 0.2324,
      "num_input_tokens_seen": 24226560,
      "step": 114790
    },
    {
      "epoch": 12.628712871287128,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.010793510174224443,
      "loss": 0.2309,
      "num_input_tokens_seen": 24227648,
      "step": 114795
    },
    {
      "epoch": 12.629262926292629,
      "grad_norm": 0.000606536865234375,
      "learning_rate": 0.01079212793647229,
      "loss": 0.2319,
      "num_input_tokens_seen": 24228672,
      "step": 114800
    },
    {
      "epoch": 12.62981298129813,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010790745737502085,
      "loss": 0.2314,
      "num_input_tokens_seen": 24229664,
      "step": 114805
    },
    {
      "epoch": 12.630363036303631,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.010789363577326547,
      "loss": 0.2298,
      "num_input_tokens_seen": 24230720,
      "step": 114810
    },
    {
      "epoch": 12.63091309130913,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.01078798145595843,
      "loss": 0.233,
      "num_input_tokens_seen": 24231776,
      "step": 114815
    },
    {
      "epoch": 12.631463146314632,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.010786599373410468,
      "loss": 0.234,
      "num_input_tokens_seen": 24232832,
      "step": 114820
    },
    {
      "epoch": 12.632013201320133,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.010785217329695394,
      "loss": 0.2308,
      "num_input_tokens_seen": 24233888,
      "step": 114825
    },
    {
      "epoch": 12.632563256325632,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.010783835324825951,
      "loss": 0.2314,
      "num_input_tokens_seen": 24234944,
      "step": 114830
    },
    {
      "epoch": 12.633113311331133,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.010782453358814871,
      "loss": 0.2298,
      "num_input_tokens_seen": 24236032,
      "step": 114835
    },
    {
      "epoch": 12.633663366336634,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.010781071431674902,
      "loss": 0.2303,
      "num_input_tokens_seen": 24237152,
      "step": 114840
    },
    {
      "epoch": 12.634213421342134,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.010779689543418769,
      "loss": 0.2319,
      "num_input_tokens_seen": 24238208,
      "step": 114845
    },
    {
      "epoch": 12.634763476347635,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.01077830769405921,
      "loss": 0.2314,
      "num_input_tokens_seen": 24239264,
      "step": 114850
    },
    {
      "epoch": 12.635313531353136,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.010776925883608965,
      "loss": 0.2329,
      "num_input_tokens_seen": 24240320,
      "step": 114855
    },
    {
      "epoch": 12.635863586358635,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.010775544112080766,
      "loss": 0.2319,
      "num_input_tokens_seen": 24241280,
      "step": 114860
    },
    {
      "epoch": 12.636413641364136,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.010774162379487357,
      "loss": 0.2319,
      "num_input_tokens_seen": 24242368,
      "step": 114865
    },
    {
      "epoch": 12.636963696369637,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.010772780685841461,
      "loss": 0.2308,
      "num_input_tokens_seen": 24243456,
      "step": 114870
    },
    {
      "epoch": 12.637513751375138,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.010771399031155813,
      "loss": 0.2335,
      "num_input_tokens_seen": 24244544,
      "step": 114875
    },
    {
      "epoch": 12.638063806380638,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.01077001741544316,
      "loss": 0.2309,
      "num_input_tokens_seen": 24245632,
      "step": 114880
    },
    {
      "epoch": 12.638613861386139,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.010768635838716222,
      "loss": 0.2308,
      "num_input_tokens_seen": 24246624,
      "step": 114885
    },
    {
      "epoch": 12.63916391639164,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.010767254300987733,
      "loss": 0.2303,
      "num_input_tokens_seen": 24247648,
      "step": 114890
    },
    {
      "epoch": 12.63971397139714,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.010765872802270437,
      "loss": 0.2303,
      "num_input_tokens_seen": 24248640,
      "step": 114895
    },
    {
      "epoch": 12.64026402640264,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.010764491342577055,
      "loss": 0.2303,
      "num_input_tokens_seen": 24249760,
      "step": 114900
    },
    {
      "epoch": 12.640814081408141,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.010763109921920325,
      "loss": 0.2314,
      "num_input_tokens_seen": 24250784,
      "step": 114905
    },
    {
      "epoch": 12.64136413641364,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010761728540312978,
      "loss": 0.2293,
      "num_input_tokens_seen": 24251840,
      "step": 114910
    },
    {
      "epoch": 12.641914191419142,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.010760347197767743,
      "loss": 0.2314,
      "num_input_tokens_seen": 24252864,
      "step": 114915
    },
    {
      "epoch": 12.642464246424643,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.010758965894297359,
      "loss": 0.2288,
      "num_input_tokens_seen": 24253984,
      "step": 114920
    },
    {
      "epoch": 12.643014301430142,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.010757584629914546,
      "loss": 0.2314,
      "num_input_tokens_seen": 24255072,
      "step": 114925
    },
    {
      "epoch": 12.643564356435643,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.010756203404632042,
      "loss": 0.2283,
      "num_input_tokens_seen": 24256064,
      "step": 114930
    },
    {
      "epoch": 12.644114411441144,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.010754822218462579,
      "loss": 0.2309,
      "num_input_tokens_seen": 24257088,
      "step": 114935
    },
    {
      "epoch": 12.644664466446645,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.010753441071418877,
      "loss": 0.2335,
      "num_input_tokens_seen": 24258208,
      "step": 114940
    },
    {
      "epoch": 12.645214521452145,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.010752059963513671,
      "loss": 0.2298,
      "num_input_tokens_seen": 24259232,
      "step": 114945
    },
    {
      "epoch": 12.645764576457646,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010750678894759688,
      "loss": 0.2309,
      "num_input_tokens_seen": 24260288,
      "step": 114950
    },
    {
      "epoch": 12.646314631463147,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010749297865169666,
      "loss": 0.2283,
      "num_input_tokens_seen": 24261344,
      "step": 114955
    },
    {
      "epoch": 12.646864686468646,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010747916874756323,
      "loss": 0.2309,
      "num_input_tokens_seen": 24262336,
      "step": 114960
    },
    {
      "epoch": 12.647414741474147,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.010746535923532384,
      "loss": 0.2293,
      "num_input_tokens_seen": 24263456,
      "step": 114965
    },
    {
      "epoch": 12.647964796479648,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010745155011510589,
      "loss": 0.2293,
      "num_input_tokens_seen": 24264480,
      "step": 114970
    },
    {
      "epoch": 12.648514851485148,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010743774138703654,
      "loss": 0.2293,
      "num_input_tokens_seen": 24265536,
      "step": 114975
    },
    {
      "epoch": 12.649064906490649,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.010742393305124316,
      "loss": 0.2294,
      "num_input_tokens_seen": 24266592,
      "step": 114980
    },
    {
      "epoch": 12.64961496149615,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.010741012510785294,
      "loss": 0.2283,
      "num_input_tokens_seen": 24267584,
      "step": 114985
    },
    {
      "epoch": 12.65016501650165,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.01073963175569931,
      "loss": 0.2314,
      "num_input_tokens_seen": 24268640,
      "step": 114990
    },
    {
      "epoch": 12.65071507150715,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.010738251039879106,
      "loss": 0.2324,
      "num_input_tokens_seen": 24269632,
      "step": 114995
    },
    {
      "epoch": 12.651265126512651,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.010736870363337389,
      "loss": 0.233,
      "num_input_tokens_seen": 24270688,
      "step": 115000
    },
    {
      "epoch": 12.651815181518153,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010735489726086896,
      "loss": 0.2304,
      "num_input_tokens_seen": 24271776,
      "step": 115005
    },
    {
      "epoch": 12.652365236523652,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.010734109128140351,
      "loss": 0.2309,
      "num_input_tokens_seen": 24272832,
      "step": 115010
    },
    {
      "epoch": 12.652915291529153,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.010732728569510467,
      "loss": 0.2309,
      "num_input_tokens_seen": 24273888,
      "step": 115015
    },
    {
      "epoch": 12.653465346534654,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01073134805020998,
      "loss": 0.232,
      "num_input_tokens_seen": 24274976,
      "step": 115020
    },
    {
      "epoch": 12.654015401540153,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.010729967570251606,
      "loss": 0.2299,
      "num_input_tokens_seen": 24275968,
      "step": 115025
    },
    {
      "epoch": 12.654565456545654,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.010728587129648078,
      "loss": 0.2294,
      "num_input_tokens_seen": 24276960,
      "step": 115030
    },
    {
      "epoch": 12.655115511551156,
      "grad_norm": 0.005859375,
      "learning_rate": 0.010727206728412117,
      "loss": 0.2319,
      "num_input_tokens_seen": 24277952,
      "step": 115035
    },
    {
      "epoch": 12.655665566556655,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.01072582636655643,
      "loss": 0.2309,
      "num_input_tokens_seen": 24279008,
      "step": 115040
    },
    {
      "epoch": 12.656215621562156,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.010724446044093757,
      "loss": 0.2273,
      "num_input_tokens_seen": 24280000,
      "step": 115045
    },
    {
      "epoch": 12.656765676567657,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.010723065761036807,
      "loss": 0.2288,
      "num_input_tokens_seen": 24281024,
      "step": 115050
    },
    {
      "epoch": 12.657315731573158,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010721685517398319,
      "loss": 0.2304,
      "num_input_tokens_seen": 24282080,
      "step": 115055
    },
    {
      "epoch": 12.657865786578657,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.010720305313190993,
      "loss": 0.2284,
      "num_input_tokens_seen": 24283136,
      "step": 115060
    },
    {
      "epoch": 12.658415841584159,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.01071892514842756,
      "loss": 0.233,
      "num_input_tokens_seen": 24284160,
      "step": 115065
    },
    {
      "epoch": 12.65896589658966,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.010717545023120747,
      "loss": 0.2351,
      "num_input_tokens_seen": 24285248,
      "step": 115070
    },
    {
      "epoch": 12.659515951595159,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.010716164937283262,
      "loss": 0.2361,
      "num_input_tokens_seen": 24286272,
      "step": 115075
    },
    {
      "epoch": 12.66006600660066,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.010714784890927826,
      "loss": 0.2293,
      "num_input_tokens_seen": 24287264,
      "step": 115080
    },
    {
      "epoch": 12.660616061606161,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.010713404884067168,
      "loss": 0.2346,
      "num_input_tokens_seen": 24288352,
      "step": 115085
    },
    {
      "epoch": 12.66116611661166,
      "grad_norm": 0.004913330078125,
      "learning_rate": 0.010712024916713992,
      "loss": 0.233,
      "num_input_tokens_seen": 24289408,
      "step": 115090
    },
    {
      "epoch": 12.661716171617162,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.010710644988881031,
      "loss": 0.2309,
      "num_input_tokens_seen": 24290400,
      "step": 115095
    },
    {
      "epoch": 12.662266226622663,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.010709265100580994,
      "loss": 0.2315,
      "num_input_tokens_seen": 24291520,
      "step": 115100
    },
    {
      "epoch": 12.662816281628164,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.010707885251826602,
      "loss": 0.2315,
      "num_input_tokens_seen": 24292640,
      "step": 115105
    },
    {
      "epoch": 12.663366336633663,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.010706505442630576,
      "loss": 0.2299,
      "num_input_tokens_seen": 24293696,
      "step": 115110
    },
    {
      "epoch": 12.663916391639164,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.010705125673005621,
      "loss": 0.2278,
      "num_input_tokens_seen": 24294816,
      "step": 115115
    },
    {
      "epoch": 12.664466446644665,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.01070374594296447,
      "loss": 0.2293,
      "num_input_tokens_seen": 24295872,
      "step": 115120
    },
    {
      "epoch": 12.665016501650165,
      "grad_norm": 0.004852294921875,
      "learning_rate": 0.010702366252519831,
      "loss": 0.2325,
      "num_input_tokens_seen": 24296896,
      "step": 115125
    },
    {
      "epoch": 12.665566556655666,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.010700986601684413,
      "loss": 0.2299,
      "num_input_tokens_seen": 24297920,
      "step": 115130
    },
    {
      "epoch": 12.666116611661167,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.010699606990470944,
      "loss": 0.232,
      "num_input_tokens_seen": 24299008,
      "step": 115135
    },
    {
      "epoch": 12.666666666666666,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010698227418892129,
      "loss": 0.232,
      "num_input_tokens_seen": 24300032,
      "step": 115140
    },
    {
      "epoch": 12.667216721672167,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.010696847886960693,
      "loss": 0.2356,
      "num_input_tokens_seen": 24301088,
      "step": 115145
    },
    {
      "epoch": 12.667766776677668,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.010695468394689345,
      "loss": 0.2346,
      "num_input_tokens_seen": 24302304,
      "step": 115150
    },
    {
      "epoch": 12.668316831683168,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.010694088942090797,
      "loss": 0.2294,
      "num_input_tokens_seen": 24303360,
      "step": 115155
    },
    {
      "epoch": 12.668866886688669,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.010692709529177764,
      "loss": 0.2309,
      "num_input_tokens_seen": 24304384,
      "step": 115160
    },
    {
      "epoch": 12.66941694169417,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.01069133015596296,
      "loss": 0.2293,
      "num_input_tokens_seen": 24305408,
      "step": 115165
    },
    {
      "epoch": 12.66996699669967,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.010689950822459103,
      "loss": 0.2324,
      "num_input_tokens_seen": 24306432,
      "step": 115170
    },
    {
      "epoch": 12.67051705170517,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0106885715286789,
      "loss": 0.234,
      "num_input_tokens_seen": 24307424,
      "step": 115175
    },
    {
      "epoch": 12.671067106710671,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.01068719227463506,
      "loss": 0.2303,
      "num_input_tokens_seen": 24308448,
      "step": 115180
    },
    {
      "epoch": 12.671617161716172,
      "grad_norm": 0.0029144287109375,
      "learning_rate": 0.010685813060340306,
      "loss": 0.2319,
      "num_input_tokens_seen": 24309440,
      "step": 115185
    },
    {
      "epoch": 12.672167216721672,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.010684433885807336,
      "loss": 0.2283,
      "num_input_tokens_seen": 24310528,
      "step": 115190
    },
    {
      "epoch": 12.672717271727173,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.010683054751048872,
      "loss": 0.2335,
      "num_input_tokens_seen": 24311584,
      "step": 115195
    },
    {
      "epoch": 12.673267326732674,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.010681675656077623,
      "loss": 0.2314,
      "num_input_tokens_seen": 24312640,
      "step": 115200
    },
    {
      "epoch": 12.673817381738173,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.010680296600906287,
      "loss": 0.2314,
      "num_input_tokens_seen": 24313696,
      "step": 115205
    },
    {
      "epoch": 12.674367436743674,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.010678917585547596,
      "loss": 0.2298,
      "num_input_tokens_seen": 24314816,
      "step": 115210
    },
    {
      "epoch": 12.674917491749175,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010677538610014243,
      "loss": 0.2314,
      "num_input_tokens_seen": 24315840,
      "step": 115215
    },
    {
      "epoch": 12.675467546754675,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.010676159674318944,
      "loss": 0.2304,
      "num_input_tokens_seen": 24316864,
      "step": 115220
    },
    {
      "epoch": 12.676017601760176,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.01067478077847441,
      "loss": 0.2293,
      "num_input_tokens_seen": 24317920,
      "step": 115225
    },
    {
      "epoch": 12.676567656765677,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.01067340192249334,
      "loss": 0.2304,
      "num_input_tokens_seen": 24318944,
      "step": 115230
    },
    {
      "epoch": 12.677117711771178,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01067202310638845,
      "loss": 0.2351,
      "num_input_tokens_seen": 24320032,
      "step": 115235
    },
    {
      "epoch": 12.677667766776677,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010670644330172443,
      "loss": 0.232,
      "num_input_tokens_seen": 24321056,
      "step": 115240
    },
    {
      "epoch": 12.678217821782178,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.010669265593858036,
      "loss": 0.232,
      "num_input_tokens_seen": 24322016,
      "step": 115245
    },
    {
      "epoch": 12.67876787678768,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.01066788689745793,
      "loss": 0.2304,
      "num_input_tokens_seen": 24323072,
      "step": 115250
    },
    {
      "epoch": 12.679317931793179,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.010666508240984827,
      "loss": 0.233,
      "num_input_tokens_seen": 24324096,
      "step": 115255
    },
    {
      "epoch": 12.67986798679868,
      "grad_norm": 0.00494384765625,
      "learning_rate": 0.010665129624451443,
      "loss": 0.2309,
      "num_input_tokens_seen": 24325184,
      "step": 115260
    },
    {
      "epoch": 12.680418041804181,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.010663751047870476,
      "loss": 0.2299,
      "num_input_tokens_seen": 24326272,
      "step": 115265
    },
    {
      "epoch": 12.68096809680968,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.010662372511254634,
      "loss": 0.2325,
      "num_input_tokens_seen": 24327328,
      "step": 115270
    },
    {
      "epoch": 12.681518151815181,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.010660994014616626,
      "loss": 0.2335,
      "num_input_tokens_seen": 24328288,
      "step": 115275
    },
    {
      "epoch": 12.682068206820682,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.010659615557969149,
      "loss": 0.2288,
      "num_input_tokens_seen": 24329344,
      "step": 115280
    },
    {
      "epoch": 12.682618261826182,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.01065823714132492,
      "loss": 0.2289,
      "num_input_tokens_seen": 24330368,
      "step": 115285
    },
    {
      "epoch": 12.683168316831683,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.010656858764696632,
      "loss": 0.2283,
      "num_input_tokens_seen": 24331424,
      "step": 115290
    },
    {
      "epoch": 12.683718371837184,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.01065548042809699,
      "loss": 0.2309,
      "num_input_tokens_seen": 24332480,
      "step": 115295
    },
    {
      "epoch": 12.684268426842685,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.010654102131538707,
      "loss": 0.233,
      "num_input_tokens_seen": 24333536,
      "step": 115300
    },
    {
      "epoch": 12.684818481848184,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.010652723875034473,
      "loss": 0.2314,
      "num_input_tokens_seen": 24334624,
      "step": 115305
    },
    {
      "epoch": 12.685368536853685,
      "grad_norm": 0.004791259765625,
      "learning_rate": 0.010651345658597001,
      "loss": 0.2278,
      "num_input_tokens_seen": 24335616,
      "step": 115310
    },
    {
      "epoch": 12.685918591859187,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.010649967482238992,
      "loss": 0.2304,
      "num_input_tokens_seen": 24336672,
      "step": 115315
    },
    {
      "epoch": 12.686468646864686,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.010648589345973137,
      "loss": 0.233,
      "num_input_tokens_seen": 24337792,
      "step": 115320
    },
    {
      "epoch": 12.687018701870187,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.01064721124981215,
      "loss": 0.2341,
      "num_input_tokens_seen": 24338976,
      "step": 115325
    },
    {
      "epoch": 12.687568756875688,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.010645833193768727,
      "loss": 0.2293,
      "num_input_tokens_seen": 24340000,
      "step": 115330
    },
    {
      "epoch": 12.688118811881187,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.010644455177855571,
      "loss": 0.2304,
      "num_input_tokens_seen": 24340992,
      "step": 115335
    },
    {
      "epoch": 12.688668866886688,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.010643077202085386,
      "loss": 0.2341,
      "num_input_tokens_seen": 24342080,
      "step": 115340
    },
    {
      "epoch": 12.68921892189219,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.01064169926647086,
      "loss": 0.2325,
      "num_input_tokens_seen": 24343104,
      "step": 115345
    },
    {
      "epoch": 12.689768976897689,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.010640321371024705,
      "loss": 0.2293,
      "num_input_tokens_seen": 24344160,
      "step": 115350
    },
    {
      "epoch": 12.69031903190319,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.010638943515759615,
      "loss": 0.2304,
      "num_input_tokens_seen": 24345248,
      "step": 115355
    },
    {
      "epoch": 12.690869086908691,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.010637565700688294,
      "loss": 0.2299,
      "num_input_tokens_seen": 24346304,
      "step": 115360
    },
    {
      "epoch": 12.691419141914192,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.010636187925823434,
      "loss": 0.2335,
      "num_input_tokens_seen": 24347296,
      "step": 115365
    },
    {
      "epoch": 12.691969196919691,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.010634810191177731,
      "loss": 0.2299,
      "num_input_tokens_seen": 24348320,
      "step": 115370
    },
    {
      "epoch": 12.692519251925193,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010633432496763897,
      "loss": 0.2319,
      "num_input_tokens_seen": 24349408,
      "step": 115375
    },
    {
      "epoch": 12.693069306930694,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.01063205484259461,
      "loss": 0.2283,
      "num_input_tokens_seen": 24350496,
      "step": 115380
    },
    {
      "epoch": 12.693619361936193,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010630677228682585,
      "loss": 0.2314,
      "num_input_tokens_seen": 24351488,
      "step": 115385
    },
    {
      "epoch": 12.694169416941694,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.01062929965504051,
      "loss": 0.2309,
      "num_input_tokens_seen": 24352576,
      "step": 115390
    },
    {
      "epoch": 12.694719471947195,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.010627922121681078,
      "loss": 0.2335,
      "num_input_tokens_seen": 24353600,
      "step": 115395
    },
    {
      "epoch": 12.695269526952695,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.010626544628616997,
      "loss": 0.2267,
      "num_input_tokens_seen": 24354624,
      "step": 115400
    },
    {
      "epoch": 12.695819581958196,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.01062516717586095,
      "loss": 0.2294,
      "num_input_tokens_seen": 24355680,
      "step": 115405
    },
    {
      "epoch": 12.696369636963697,
      "grad_norm": 0.004974365234375,
      "learning_rate": 0.01062378976342564,
      "loss": 0.2325,
      "num_input_tokens_seen": 24356704,
      "step": 115410
    },
    {
      "epoch": 12.696919691969196,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.010622412391323765,
      "loss": 0.2356,
      "num_input_tokens_seen": 24357760,
      "step": 115415
    },
    {
      "epoch": 12.697469746974697,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.010621035059568008,
      "loss": 0.233,
      "num_input_tokens_seen": 24358720,
      "step": 115420
    },
    {
      "epoch": 12.698019801980198,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.01061965776817107,
      "loss": 0.2319,
      "num_input_tokens_seen": 24359808,
      "step": 115425
    },
    {
      "epoch": 12.6985698569857,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.010618280517145643,
      "loss": 0.2299,
      "num_input_tokens_seen": 24360960,
      "step": 115430
    },
    {
      "epoch": 12.699119911991199,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.010616903306504428,
      "loss": 0.2325,
      "num_input_tokens_seen": 24361984,
      "step": 115435
    },
    {
      "epoch": 12.6996699669967,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.01061552613626011,
      "loss": 0.2335,
      "num_input_tokens_seen": 24363008,
      "step": 115440
    },
    {
      "epoch": 12.7002200220022,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.010614149006425377,
      "loss": 0.2288,
      "num_input_tokens_seen": 24364000,
      "step": 115445
    },
    {
      "epoch": 12.7007700770077,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.010612771917012936,
      "loss": 0.233,
      "num_input_tokens_seen": 24365056,
      "step": 115450
    },
    {
      "epoch": 12.701320132013201,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.010611394868035473,
      "loss": 0.2346,
      "num_input_tokens_seen": 24366112,
      "step": 115455
    },
    {
      "epoch": 12.701870187018702,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.010610017859505669,
      "loss": 0.234,
      "num_input_tokens_seen": 24367264,
      "step": 115460
    },
    {
      "epoch": 12.702420242024202,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.010608640891436228,
      "loss": 0.233,
      "num_input_tokens_seen": 24368384,
      "step": 115465
    },
    {
      "epoch": 12.702970297029703,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.010607263963839835,
      "loss": 0.2293,
      "num_input_tokens_seen": 24369472,
      "step": 115470
    },
    {
      "epoch": 12.703520352035204,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.010605887076729187,
      "loss": 0.2319,
      "num_input_tokens_seen": 24370528,
      "step": 115475
    },
    {
      "epoch": 12.704070407040705,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.010604510230116968,
      "loss": 0.2283,
      "num_input_tokens_seen": 24371552,
      "step": 115480
    },
    {
      "epoch": 12.704620462046204,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.010603133424015863,
      "loss": 0.2304,
      "num_input_tokens_seen": 24372608,
      "step": 115485
    },
    {
      "epoch": 12.705170517051705,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.010601756658438577,
      "loss": 0.2278,
      "num_input_tokens_seen": 24373632,
      "step": 115490
    },
    {
      "epoch": 12.705720572057206,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.010600379933397781,
      "loss": 0.2319,
      "num_input_tokens_seen": 24374624,
      "step": 115495
    },
    {
      "epoch": 12.706270627062706,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.010599003248906176,
      "loss": 0.2346,
      "num_input_tokens_seen": 24375648,
      "step": 115500
    },
    {
      "epoch": 12.706820682068207,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.010597626604976448,
      "loss": 0.2309,
      "num_input_tokens_seen": 24376704,
      "step": 115505
    },
    {
      "epoch": 12.707370737073708,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.010596250001621272,
      "loss": 0.2309,
      "num_input_tokens_seen": 24377856,
      "step": 115510
    },
    {
      "epoch": 12.707920792079207,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.010594873438853361,
      "loss": 0.2319,
      "num_input_tokens_seen": 24378880,
      "step": 115515
    },
    {
      "epoch": 12.708470847084708,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.010593496916685378,
      "loss": 0.2283,
      "num_input_tokens_seen": 24379872,
      "step": 115520
    },
    {
      "epoch": 12.70902090209021,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.010592120435130024,
      "loss": 0.2304,
      "num_input_tokens_seen": 24380928,
      "step": 115525
    },
    {
      "epoch": 12.70957095709571,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.010590743994199982,
      "loss": 0.2293,
      "num_input_tokens_seen": 24381984,
      "step": 115530
    },
    {
      "epoch": 12.71012101210121,
      "grad_norm": 0.000701904296875,
      "learning_rate": 0.010589367593907936,
      "loss": 0.233,
      "num_input_tokens_seen": 24383008,
      "step": 115535
    },
    {
      "epoch": 12.710671067106711,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01058799123426657,
      "loss": 0.2319,
      "num_input_tokens_seen": 24384032,
      "step": 115540
    },
    {
      "epoch": 12.711221122112212,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.010586614915288572,
      "loss": 0.2341,
      "num_input_tokens_seen": 24385088,
      "step": 115545
    },
    {
      "epoch": 12.711771177117711,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.01058523863698663,
      "loss": 0.2351,
      "num_input_tokens_seen": 24386112,
      "step": 115550
    },
    {
      "epoch": 12.712321232123212,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.010583862399373423,
      "loss": 0.2283,
      "num_input_tokens_seen": 24387168,
      "step": 115555
    },
    {
      "epoch": 12.712871287128714,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.010582486202461635,
      "loss": 0.232,
      "num_input_tokens_seen": 24388192,
      "step": 115560
    },
    {
      "epoch": 12.713421342134213,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.010581110046263957,
      "loss": 0.2304,
      "num_input_tokens_seen": 24389248,
      "step": 115565
    },
    {
      "epoch": 12.713971397139714,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.010579733930793058,
      "loss": 0.2335,
      "num_input_tokens_seen": 24390368,
      "step": 115570
    },
    {
      "epoch": 12.714521452145215,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01057835785606164,
      "loss": 0.234,
      "num_input_tokens_seen": 24391424,
      "step": 115575
    },
    {
      "epoch": 12.715071507150714,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.010576981822082372,
      "loss": 0.2319,
      "num_input_tokens_seen": 24392512,
      "step": 115580
    },
    {
      "epoch": 12.715621562156215,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01057560582886794,
      "loss": 0.2299,
      "num_input_tokens_seen": 24393632,
      "step": 115585
    },
    {
      "epoch": 12.716171617161717,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01057422987643103,
      "loss": 0.2309,
      "num_input_tokens_seen": 24394656,
      "step": 115590
    },
    {
      "epoch": 12.716721672167218,
      "grad_norm": 0.001953125,
      "learning_rate": 0.010572853964784313,
      "loss": 0.2304,
      "num_input_tokens_seen": 24395712,
      "step": 115595
    },
    {
      "epoch": 12.717271727172717,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.01057147809394048,
      "loss": 0.2345,
      "num_input_tokens_seen": 24396800,
      "step": 115600
    },
    {
      "epoch": 12.717821782178218,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.010570102263912213,
      "loss": 0.2309,
      "num_input_tokens_seen": 24397888,
      "step": 115605
    },
    {
      "epoch": 12.718371837183719,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010568726474712178,
      "loss": 0.2272,
      "num_input_tokens_seen": 24398944,
      "step": 115610
    },
    {
      "epoch": 12.718921892189218,
      "grad_norm": 0.010009765625,
      "learning_rate": 0.010567350726353071,
      "loss": 0.2294,
      "num_input_tokens_seen": 24400032,
      "step": 115615
    },
    {
      "epoch": 12.71947194719472,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.010565975018847562,
      "loss": 0.2325,
      "num_input_tokens_seen": 24401120,
      "step": 115620
    },
    {
      "epoch": 12.72002200220022,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.010564599352208338,
      "loss": 0.232,
      "num_input_tokens_seen": 24402208,
      "step": 115625
    },
    {
      "epoch": 12.72057205720572,
      "grad_norm": 0.000545501708984375,
      "learning_rate": 0.010563223726448074,
      "loss": 0.2319,
      "num_input_tokens_seen": 24403264,
      "step": 115630
    },
    {
      "epoch": 12.721122112211221,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01056184814157944,
      "loss": 0.233,
      "num_input_tokens_seen": 24404288,
      "step": 115635
    },
    {
      "epoch": 12.721672167216722,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.010560472597615129,
      "loss": 0.2288,
      "num_input_tokens_seen": 24405376,
      "step": 115640
    },
    {
      "epoch": 12.722222222222221,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.010559097094567814,
      "loss": 0.2293,
      "num_input_tokens_seen": 24406400,
      "step": 115645
    },
    {
      "epoch": 12.722772277227723,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010557721632450162,
      "loss": 0.2309,
      "num_input_tokens_seen": 24407456,
      "step": 115650
    },
    {
      "epoch": 12.723322332233224,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.010556346211274865,
      "loss": 0.2325,
      "num_input_tokens_seen": 24408512,
      "step": 115655
    },
    {
      "epoch": 12.723872387238725,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.010554970831054585,
      "loss": 0.234,
      "num_input_tokens_seen": 24409536,
      "step": 115660
    },
    {
      "epoch": 12.724422442244224,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.010553595491802015,
      "loss": 0.2319,
      "num_input_tokens_seen": 24410592,
      "step": 115665
    },
    {
      "epoch": 12.724972497249725,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.010552220193529818,
      "loss": 0.2324,
      "num_input_tokens_seen": 24411584,
      "step": 115670
    },
    {
      "epoch": 12.725522552255226,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01055084493625067,
      "loss": 0.233,
      "num_input_tokens_seen": 24412672,
      "step": 115675
    },
    {
      "epoch": 12.726072607260726,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.010549469719977255,
      "loss": 0.2304,
      "num_input_tokens_seen": 24413760,
      "step": 115680
    },
    {
      "epoch": 12.726622662266227,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010548094544722239,
      "loss": 0.2314,
      "num_input_tokens_seen": 24414816,
      "step": 115685
    },
    {
      "epoch": 12.727172717271728,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.010546719410498296,
      "loss": 0.2335,
      "num_input_tokens_seen": 24415904,
      "step": 115690
    },
    {
      "epoch": 12.727722772277227,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.01054534431731811,
      "loss": 0.233,
      "num_input_tokens_seen": 24416896,
      "step": 115695
    },
    {
      "epoch": 12.728272827282728,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01054396926519434,
      "loss": 0.2319,
      "num_input_tokens_seen": 24417984,
      "step": 115700
    },
    {
      "epoch": 12.72882288228823,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.010542594254139677,
      "loss": 0.2314,
      "num_input_tokens_seen": 24419040,
      "step": 115705
    },
    {
      "epoch": 12.729372937293729,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010541219284166776,
      "loss": 0.2324,
      "num_input_tokens_seen": 24420096,
      "step": 115710
    },
    {
      "epoch": 12.72992299229923,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.01053984435528832,
      "loss": 0.2319,
      "num_input_tokens_seen": 24421184,
      "step": 115715
    },
    {
      "epoch": 12.73047304730473,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.010538469467516983,
      "loss": 0.2314,
      "num_input_tokens_seen": 24422240,
      "step": 115720
    },
    {
      "epoch": 12.731023102310232,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.010537094620865424,
      "loss": 0.2293,
      "num_input_tokens_seen": 24423264,
      "step": 115725
    },
    {
      "epoch": 12.731573157315731,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.010535719815346328,
      "loss": 0.2309,
      "num_input_tokens_seen": 24424384,
      "step": 115730
    },
    {
      "epoch": 12.732123212321232,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010534345050972354,
      "loss": 0.2319,
      "num_input_tokens_seen": 24425536,
      "step": 115735
    },
    {
      "epoch": 12.732673267326733,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.010532970327756186,
      "loss": 0.2335,
      "num_input_tokens_seen": 24426592,
      "step": 115740
    },
    {
      "epoch": 12.733223322332233,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.010531595645710485,
      "loss": 0.2324,
      "num_input_tokens_seen": 24427584,
      "step": 115745
    },
    {
      "epoch": 12.733773377337734,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.010530221004847918,
      "loss": 0.2309,
      "num_input_tokens_seen": 24428672,
      "step": 115750
    },
    {
      "epoch": 12.734323432343235,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010528846405181163,
      "loss": 0.232,
      "num_input_tokens_seen": 24429728,
      "step": 115755
    },
    {
      "epoch": 12.734873487348734,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.010527471846722883,
      "loss": 0.2314,
      "num_input_tokens_seen": 24430784,
      "step": 115760
    },
    {
      "epoch": 12.735423542354235,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.010526097329485753,
      "loss": 0.2314,
      "num_input_tokens_seen": 24431904,
      "step": 115765
    },
    {
      "epoch": 12.735973597359736,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.010524722853482438,
      "loss": 0.233,
      "num_input_tokens_seen": 24432960,
      "step": 115770
    },
    {
      "epoch": 12.736523652365236,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.010523348418725598,
      "loss": 0.2309,
      "num_input_tokens_seen": 24433920,
      "step": 115775
    },
    {
      "epoch": 12.737073707370737,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.010521974025227915,
      "loss": 0.2314,
      "num_input_tokens_seen": 24435008,
      "step": 115780
    },
    {
      "epoch": 12.737623762376238,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.010520599673002042,
      "loss": 0.2314,
      "num_input_tokens_seen": 24436000,
      "step": 115785
    },
    {
      "epoch": 12.738173817381739,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.010519225362060656,
      "loss": 0.2298,
      "num_input_tokens_seen": 24437056,
      "step": 115790
    },
    {
      "epoch": 12.738723872387238,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.010517851092416423,
      "loss": 0.2309,
      "num_input_tokens_seen": 24438112,
      "step": 115795
    },
    {
      "epoch": 12.73927392739274,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.010516476864081999,
      "loss": 0.2304,
      "num_input_tokens_seen": 24439168,
      "step": 115800
    },
    {
      "epoch": 12.73982398239824,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.01051510267707006,
      "loss": 0.2309,
      "num_input_tokens_seen": 24440256,
      "step": 115805
    },
    {
      "epoch": 12.74037403740374,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.010513728531393264,
      "loss": 0.2304,
      "num_input_tokens_seen": 24441312,
      "step": 115810
    },
    {
      "epoch": 12.74092409240924,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010512354427064283,
      "loss": 0.2345,
      "num_input_tokens_seen": 24442336,
      "step": 115815
    },
    {
      "epoch": 12.741474147414742,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.010510980364095781,
      "loss": 0.2324,
      "num_input_tokens_seen": 24443360,
      "step": 115820
    },
    {
      "epoch": 12.742024202420241,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.010509606342500413,
      "loss": 0.2309,
      "num_input_tokens_seen": 24444416,
      "step": 115825
    },
    {
      "epoch": 12.742574257425742,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.010508232362290851,
      "loss": 0.233,
      "num_input_tokens_seen": 24445472,
      "step": 115830
    },
    {
      "epoch": 12.743124312431243,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.01050685842347976,
      "loss": 0.2319,
      "num_input_tokens_seen": 24446528,
      "step": 115835
    },
    {
      "epoch": 12.743674367436743,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010505484526079792,
      "loss": 0.2304,
      "num_input_tokens_seen": 24447584,
      "step": 115840
    },
    {
      "epoch": 12.744224422442244,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.010504110670103619,
      "loss": 0.2283,
      "num_input_tokens_seen": 24448608,
      "step": 115845
    },
    {
      "epoch": 12.744774477447745,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.010502736855563898,
      "loss": 0.2314,
      "num_input_tokens_seen": 24449632,
      "step": 115850
    },
    {
      "epoch": 12.745324532453246,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.010501363082473299,
      "loss": 0.2298,
      "num_input_tokens_seen": 24450656,
      "step": 115855
    },
    {
      "epoch": 12.745874587458745,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.010499989350844476,
      "loss": 0.2298,
      "num_input_tokens_seen": 24451744,
      "step": 115860
    },
    {
      "epoch": 12.746424642464246,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.010498615660690087,
      "loss": 0.2314,
      "num_input_tokens_seen": 24452832,
      "step": 115865
    },
    {
      "epoch": 12.746974697469748,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.010497242012022808,
      "loss": 0.2304,
      "num_input_tokens_seen": 24453920,
      "step": 115870
    },
    {
      "epoch": 12.747524752475247,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.010495868404855275,
      "loss": 0.2325,
      "num_input_tokens_seen": 24455008,
      "step": 115875
    },
    {
      "epoch": 12.748074807480748,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.010494494839200171,
      "loss": 0.2309,
      "num_input_tokens_seen": 24456064,
      "step": 115880
    },
    {
      "epoch": 12.748624862486249,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.010493121315070146,
      "loss": 0.2335,
      "num_input_tokens_seen": 24457056,
      "step": 115885
    },
    {
      "epoch": 12.749174917491748,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.010491747832477853,
      "loss": 0.2314,
      "num_input_tokens_seen": 24458144,
      "step": 115890
    },
    {
      "epoch": 12.74972497249725,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010490374391435965,
      "loss": 0.2319,
      "num_input_tokens_seen": 24459168,
      "step": 115895
    },
    {
      "epoch": 12.75027502750275,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.010489000991957126,
      "loss": 0.2319,
      "num_input_tokens_seen": 24460224,
      "step": 115900
    },
    {
      "epoch": 12.750825082508252,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.010487627634054003,
      "loss": 0.233,
      "num_input_tokens_seen": 24461280,
      "step": 115905
    },
    {
      "epoch": 12.751375137513751,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.010486254317739255,
      "loss": 0.2273,
      "num_input_tokens_seen": 24462400,
      "step": 115910
    },
    {
      "epoch": 12.751925192519252,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.010484881043025526,
      "loss": 0.2319,
      "num_input_tokens_seen": 24463456,
      "step": 115915
    },
    {
      "epoch": 12.752475247524753,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.010483507809925487,
      "loss": 0.2293,
      "num_input_tokens_seen": 24464512,
      "step": 115920
    },
    {
      "epoch": 12.753025302530252,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.010482134618451787,
      "loss": 0.234,
      "num_input_tokens_seen": 24465504,
      "step": 115925
    },
    {
      "epoch": 12.753575357535754,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.01048076146861709,
      "loss": 0.232,
      "num_input_tokens_seen": 24466656,
      "step": 115930
    },
    {
      "epoch": 12.754125412541255,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.010479388360434041,
      "loss": 0.2309,
      "num_input_tokens_seen": 24467680,
      "step": 115935
    },
    {
      "epoch": 12.754675467546754,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.010478015293915299,
      "loss": 0.2325,
      "num_input_tokens_seen": 24468832,
      "step": 115940
    },
    {
      "epoch": 12.755225522552255,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.010476642269073523,
      "loss": 0.2298,
      "num_input_tokens_seen": 24469856,
      "step": 115945
    },
    {
      "epoch": 12.755775577557756,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.01047526928592136,
      "loss": 0.2324,
      "num_input_tokens_seen": 24470912,
      "step": 115950
    },
    {
      "epoch": 12.756325632563257,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.010473896344471477,
      "loss": 0.2335,
      "num_input_tokens_seen": 24471904,
      "step": 115955
    },
    {
      "epoch": 12.756875687568757,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010472523444736516,
      "loss": 0.2345,
      "num_input_tokens_seen": 24473024,
      "step": 115960
    },
    {
      "epoch": 12.757425742574258,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.010471150586729133,
      "loss": 0.2283,
      "num_input_tokens_seen": 24474016,
      "step": 115965
    },
    {
      "epoch": 12.757975797579759,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.010469777770461984,
      "loss": 0.2314,
      "num_input_tokens_seen": 24475008,
      "step": 115970
    },
    {
      "epoch": 12.758525852585258,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.010468404995947714,
      "loss": 0.2304,
      "num_input_tokens_seen": 24476064,
      "step": 115975
    },
    {
      "epoch": 12.75907590759076,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010467032263198985,
      "loss": 0.2304,
      "num_input_tokens_seen": 24477152,
      "step": 115980
    },
    {
      "epoch": 12.75962596259626,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.010465659572228447,
      "loss": 0.2304,
      "num_input_tokens_seen": 24478240,
      "step": 115985
    },
    {
      "epoch": 12.76017601760176,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.010464286923048741,
      "loss": 0.2319,
      "num_input_tokens_seen": 24479296,
      "step": 115990
    },
    {
      "epoch": 12.76072607260726,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010462914315672535,
      "loss": 0.2298,
      "num_input_tokens_seen": 24480288,
      "step": 115995
    },
    {
      "epoch": 12.761276127612762,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010461541750112463,
      "loss": 0.2288,
      "num_input_tokens_seen": 24481376,
      "step": 116000
    },
    {
      "epoch": 12.761826182618261,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.010460169226381188,
      "loss": 0.2309,
      "num_input_tokens_seen": 24482400,
      "step": 116005
    },
    {
      "epoch": 12.762376237623762,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010458796744491357,
      "loss": 0.2283,
      "num_input_tokens_seen": 24483424,
      "step": 116010
    },
    {
      "epoch": 12.762926292629263,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01045742430445561,
      "loss": 0.2335,
      "num_input_tokens_seen": 24484416,
      "step": 116015
    },
    {
      "epoch": 12.763476347634764,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010456051906286606,
      "loss": 0.2377,
      "num_input_tokens_seen": 24485568,
      "step": 116020
    },
    {
      "epoch": 12.764026402640264,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.010454679549996997,
      "loss": 0.2314,
      "num_input_tokens_seen": 24486592,
      "step": 116025
    },
    {
      "epoch": 12.764576457645765,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.010453307235599419,
      "loss": 0.233,
      "num_input_tokens_seen": 24487712,
      "step": 116030
    },
    {
      "epoch": 12.765126512651266,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010451934963106526,
      "loss": 0.2319,
      "num_input_tokens_seen": 24488736,
      "step": 116035
    },
    {
      "epoch": 12.765676567656765,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.010450562732530968,
      "loss": 0.2324,
      "num_input_tokens_seen": 24489824,
      "step": 116040
    },
    {
      "epoch": 12.766226622662266,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.010449190543885394,
      "loss": 0.2298,
      "num_input_tokens_seen": 24490816,
      "step": 116045
    },
    {
      "epoch": 12.766776677667767,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.010447818397182444,
      "loss": 0.2325,
      "num_input_tokens_seen": 24491904,
      "step": 116050
    },
    {
      "epoch": 12.767326732673267,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010446446292434764,
      "loss": 0.233,
      "num_input_tokens_seen": 24492928,
      "step": 116055
    },
    {
      "epoch": 12.767876787678768,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.010445074229655007,
      "loss": 0.2303,
      "num_input_tokens_seen": 24493984,
      "step": 116060
    },
    {
      "epoch": 12.768426842684269,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.010443702208855812,
      "loss": 0.2293,
      "num_input_tokens_seen": 24495072,
      "step": 116065
    },
    {
      "epoch": 12.768976897689768,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.010442330230049835,
      "loss": 0.2314,
      "num_input_tokens_seen": 24496064,
      "step": 116070
    },
    {
      "epoch": 12.76952695269527,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01044095829324971,
      "loss": 0.2304,
      "num_input_tokens_seen": 24497120,
      "step": 116075
    },
    {
      "epoch": 12.77007700770077,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.010439586398468079,
      "loss": 0.2309,
      "num_input_tokens_seen": 24498208,
      "step": 116080
    },
    {
      "epoch": 12.770627062706271,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.010438214545717599,
      "loss": 0.2335,
      "num_input_tokens_seen": 24499264,
      "step": 116085
    },
    {
      "epoch": 12.77117711771177,
      "grad_norm": 0.002960205078125,
      "learning_rate": 0.010436842735010901,
      "loss": 0.2304,
      "num_input_tokens_seen": 24500352,
      "step": 116090
    },
    {
      "epoch": 12.771727172717272,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.010435470966360636,
      "loss": 0.2319,
      "num_input_tokens_seen": 24501408,
      "step": 116095
    },
    {
      "epoch": 12.772277227722773,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01043409923977945,
      "loss": 0.2324,
      "num_input_tokens_seen": 24502432,
      "step": 116100
    },
    {
      "epoch": 12.772827282728272,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.010432727555279972,
      "loss": 0.2288,
      "num_input_tokens_seen": 24503520,
      "step": 116105
    },
    {
      "epoch": 12.773377337733773,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010431355912874859,
      "loss": 0.2309,
      "num_input_tokens_seen": 24504544,
      "step": 116110
    },
    {
      "epoch": 12.773927392739274,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.01042998431257674,
      "loss": 0.2319,
      "num_input_tokens_seen": 24505632,
      "step": 116115
    },
    {
      "epoch": 12.774477447744774,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.010428612754398265,
      "loss": 0.2335,
      "num_input_tokens_seen": 24506688,
      "step": 116120
    },
    {
      "epoch": 12.775027502750275,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010427241238352078,
      "loss": 0.2309,
      "num_input_tokens_seen": 24507712,
      "step": 116125
    },
    {
      "epoch": 12.775577557755776,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010425869764450809,
      "loss": 0.2293,
      "num_input_tokens_seen": 24508800,
      "step": 116130
    },
    {
      "epoch": 12.776127612761275,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.010424498332707104,
      "loss": 0.2319,
      "num_input_tokens_seen": 24509856,
      "step": 116135
    },
    {
      "epoch": 12.776677667766776,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0104231269431336,
      "loss": 0.233,
      "num_input_tokens_seen": 24510944,
      "step": 116140
    },
    {
      "epoch": 12.777227722772277,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010421755595742946,
      "loss": 0.2314,
      "num_input_tokens_seen": 24512064,
      "step": 116145
    },
    {
      "epoch": 12.777777777777779,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.010420384290547771,
      "loss": 0.2309,
      "num_input_tokens_seen": 24513120,
      "step": 116150
    },
    {
      "epoch": 12.778327832783278,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.010419013027560712,
      "loss": 0.2304,
      "num_input_tokens_seen": 24514176,
      "step": 116155
    },
    {
      "epoch": 12.778877887788779,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.010417641806794418,
      "loss": 0.2293,
      "num_input_tokens_seen": 24515232,
      "step": 116160
    },
    {
      "epoch": 12.77942794279428,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.010416270628261516,
      "loss": 0.2324,
      "num_input_tokens_seen": 24516288,
      "step": 116165
    },
    {
      "epoch": 12.77997799779978,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.010414899491974649,
      "loss": 0.2335,
      "num_input_tokens_seen": 24517280,
      "step": 116170
    },
    {
      "epoch": 12.78052805280528,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.010413528397946455,
      "loss": 0.2309,
      "num_input_tokens_seen": 24518368,
      "step": 116175
    },
    {
      "epoch": 12.781078107810782,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.010412157346189565,
      "loss": 0.2309,
      "num_input_tokens_seen": 24519424,
      "step": 116180
    },
    {
      "epoch": 12.781628162816281,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.010410786336716627,
      "loss": 0.2303,
      "num_input_tokens_seen": 24520448,
      "step": 116185
    },
    {
      "epoch": 12.782178217821782,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010409415369540261,
      "loss": 0.2314,
      "num_input_tokens_seen": 24521440,
      "step": 116190
    },
    {
      "epoch": 12.782728272827283,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.010408044444673114,
      "loss": 0.2303,
      "num_input_tokens_seen": 24522464,
      "step": 116195
    },
    {
      "epoch": 12.783278327832782,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.010406673562127822,
      "loss": 0.2304,
      "num_input_tokens_seen": 24523584,
      "step": 116200
    },
    {
      "epoch": 12.783828382838283,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.010405302721917009,
      "loss": 0.2324,
      "num_input_tokens_seen": 24524608,
      "step": 116205
    },
    {
      "epoch": 12.784378437843785,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.01040393192405332,
      "loss": 0.2314,
      "num_input_tokens_seen": 24525632,
      "step": 116210
    },
    {
      "epoch": 12.784928492849286,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.010402561168549379,
      "loss": 0.2324,
      "num_input_tokens_seen": 24526688,
      "step": 116215
    },
    {
      "epoch": 12.785478547854785,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.010401190455417835,
      "loss": 0.2329,
      "num_input_tokens_seen": 24527744,
      "step": 116220
    },
    {
      "epoch": 12.786028602860286,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.010399819784671306,
      "loss": 0.2324,
      "num_input_tokens_seen": 24528800,
      "step": 116225
    },
    {
      "epoch": 12.786578657865787,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.010398449156322431,
      "loss": 0.2299,
      "num_input_tokens_seen": 24529856,
      "step": 116230
    },
    {
      "epoch": 12.787128712871286,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.010397078570383843,
      "loss": 0.2324,
      "num_input_tokens_seen": 24530880,
      "step": 116235
    },
    {
      "epoch": 12.787678767876788,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.010395708026868176,
      "loss": 0.2304,
      "num_input_tokens_seen": 24531904,
      "step": 116240
    },
    {
      "epoch": 12.788228822882289,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010394337525788054,
      "loss": 0.2309,
      "num_input_tokens_seen": 24532992,
      "step": 116245
    },
    {
      "epoch": 12.788778877887788,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010392967067156115,
      "loss": 0.2314,
      "num_input_tokens_seen": 24534048,
      "step": 116250
    },
    {
      "epoch": 12.789328932893289,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.010391596650984986,
      "loss": 0.2314,
      "num_input_tokens_seen": 24535136,
      "step": 116255
    },
    {
      "epoch": 12.78987898789879,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.010390226277287305,
      "loss": 0.2335,
      "num_input_tokens_seen": 24536224,
      "step": 116260
    },
    {
      "epoch": 12.79042904290429,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.010388855946075692,
      "loss": 0.2345,
      "num_input_tokens_seen": 24537312,
      "step": 116265
    },
    {
      "epoch": 12.79097909790979,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01038748565736278,
      "loss": 0.2303,
      "num_input_tokens_seen": 24538336,
      "step": 116270
    },
    {
      "epoch": 12.791529152915292,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.010386115411161204,
      "loss": 0.2324,
      "num_input_tokens_seen": 24539328,
      "step": 116275
    },
    {
      "epoch": 12.792079207920793,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010384745207483583,
      "loss": 0.2293,
      "num_input_tokens_seen": 24540384,
      "step": 116280
    },
    {
      "epoch": 12.792629262926292,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.010383375046342553,
      "loss": 0.2314,
      "num_input_tokens_seen": 24541440,
      "step": 116285
    },
    {
      "epoch": 12.793179317931793,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.010382004927750745,
      "loss": 0.2309,
      "num_input_tokens_seen": 24542432,
      "step": 116290
    },
    {
      "epoch": 12.793729372937294,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.01038063485172077,
      "loss": 0.2314,
      "num_input_tokens_seen": 24543488,
      "step": 116295
    },
    {
      "epoch": 12.794279427942794,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01037926481826528,
      "loss": 0.2314,
      "num_input_tokens_seen": 24544512,
      "step": 116300
    },
    {
      "epoch": 12.794829482948295,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.010377894827396877,
      "loss": 0.2309,
      "num_input_tokens_seen": 24545600,
      "step": 116305
    },
    {
      "epoch": 12.795379537953796,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.010376524879128208,
      "loss": 0.2314,
      "num_input_tokens_seen": 24546656,
      "step": 116310
    },
    {
      "epoch": 12.795929592959295,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.01037515497347189,
      "loss": 0.233,
      "num_input_tokens_seen": 24547744,
      "step": 116315
    },
    {
      "epoch": 12.796479647964796,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.010373785110440544,
      "loss": 0.2324,
      "num_input_tokens_seen": 24548768,
      "step": 116320
    },
    {
      "epoch": 12.797029702970297,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.010372415290046804,
      "loss": 0.2314,
      "num_input_tokens_seen": 24549856,
      "step": 116325
    },
    {
      "epoch": 12.797579757975798,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010371045512303289,
      "loss": 0.2319,
      "num_input_tokens_seen": 24550944,
      "step": 116330
    },
    {
      "epoch": 12.798129812981298,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010369675777222634,
      "loss": 0.2298,
      "num_input_tokens_seen": 24552032,
      "step": 116335
    },
    {
      "epoch": 12.798679867986799,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01036830608481745,
      "loss": 0.2319,
      "num_input_tokens_seen": 24553088,
      "step": 116340
    },
    {
      "epoch": 12.7992299229923,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010366936435100364,
      "loss": 0.2314,
      "num_input_tokens_seen": 24554144,
      "step": 116345
    },
    {
      "epoch": 12.7997799779978,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.010365566828084009,
      "loss": 0.2298,
      "num_input_tokens_seen": 24555168,
      "step": 116350
    },
    {
      "epoch": 12.8003300330033,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01036419726378099,
      "loss": 0.2351,
      "num_input_tokens_seen": 24556320,
      "step": 116355
    },
    {
      "epoch": 12.800880088008801,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.010362827742203951,
      "loss": 0.2309,
      "num_input_tokens_seen": 24557408,
      "step": 116360
    },
    {
      "epoch": 12.8014301430143,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.010361458263365498,
      "loss": 0.2293,
      "num_input_tokens_seen": 24558464,
      "step": 116365
    },
    {
      "epoch": 12.801980198019802,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010360088827278258,
      "loss": 0.2319,
      "num_input_tokens_seen": 24559520,
      "step": 116370
    },
    {
      "epoch": 12.802530253025303,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010358719433954857,
      "loss": 0.2319,
      "num_input_tokens_seen": 24560608,
      "step": 116375
    },
    {
      "epoch": 12.803080308030804,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.010357350083407907,
      "loss": 0.2309,
      "num_input_tokens_seen": 24561696,
      "step": 116380
    },
    {
      "epoch": 12.803630363036303,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.010355980775650036,
      "loss": 0.2319,
      "num_input_tokens_seen": 24562752,
      "step": 116385
    },
    {
      "epoch": 12.804180418041804,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.010354611510693867,
      "loss": 0.2319,
      "num_input_tokens_seen": 24563808,
      "step": 116390
    },
    {
      "epoch": 12.804730473047305,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.010353242288552007,
      "loss": 0.2303,
      "num_input_tokens_seen": 24564864,
      "step": 116395
    },
    {
      "epoch": 12.805280528052805,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.010351873109237086,
      "loss": 0.2314,
      "num_input_tokens_seen": 24565920,
      "step": 116400
    },
    {
      "epoch": 12.805830583058306,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.010350503972761715,
      "loss": 0.2319,
      "num_input_tokens_seen": 24566976,
      "step": 116405
    },
    {
      "epoch": 12.806380638063807,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010349134879138527,
      "loss": 0.2303,
      "num_input_tokens_seen": 24568000,
      "step": 116410
    },
    {
      "epoch": 12.806930693069306,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.010347765828380125,
      "loss": 0.2288,
      "num_input_tokens_seen": 24569056,
      "step": 116415
    },
    {
      "epoch": 12.807480748074807,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010346396820499133,
      "loss": 0.2298,
      "num_input_tokens_seen": 24570080,
      "step": 116420
    },
    {
      "epoch": 12.808030803080309,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01034502785550817,
      "loss": 0.2319,
      "num_input_tokens_seen": 24571136,
      "step": 116425
    },
    {
      "epoch": 12.808580858085808,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010343658933419854,
      "loss": 0.2324,
      "num_input_tokens_seen": 24572160,
      "step": 116430
    },
    {
      "epoch": 12.809130913091309,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.010342290054246795,
      "loss": 0.2324,
      "num_input_tokens_seen": 24573184,
      "step": 116435
    },
    {
      "epoch": 12.80968096809681,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.010340921218001616,
      "loss": 0.235,
      "num_input_tokens_seen": 24574240,
      "step": 116440
    },
    {
      "epoch": 12.810231023102311,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.010339552424696927,
      "loss": 0.2319,
      "num_input_tokens_seen": 24575296,
      "step": 116445
    },
    {
      "epoch": 12.81078107810781,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.010338183674345353,
      "loss": 0.2319,
      "num_input_tokens_seen": 24576320,
      "step": 116450
    },
    {
      "epoch": 12.811331133113312,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0103368149669595,
      "loss": 0.2335,
      "num_input_tokens_seen": 24577376,
      "step": 116455
    },
    {
      "epoch": 12.811881188118813,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.010335446302551983,
      "loss": 0.2319,
      "num_input_tokens_seen": 24578464,
      "step": 116460
    },
    {
      "epoch": 12.812431243124312,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.010334077681135428,
      "loss": 0.2309,
      "num_input_tokens_seen": 24579552,
      "step": 116465
    },
    {
      "epoch": 12.812981298129813,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.010332709102722432,
      "loss": 0.2293,
      "num_input_tokens_seen": 24580576,
      "step": 116470
    },
    {
      "epoch": 12.813531353135314,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010331340567325618,
      "loss": 0.2298,
      "num_input_tokens_seen": 24581536,
      "step": 116475
    },
    {
      "epoch": 12.814081408140813,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.010329972074957599,
      "loss": 0.2324,
      "num_input_tokens_seen": 24582592,
      "step": 116480
    },
    {
      "epoch": 12.814631463146315,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010328603625630981,
      "loss": 0.2309,
      "num_input_tokens_seen": 24583584,
      "step": 116485
    },
    {
      "epoch": 12.815181518151816,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.010327235219358393,
      "loss": 0.2309,
      "num_input_tokens_seen": 24584640,
      "step": 116490
    },
    {
      "epoch": 12.815731573157315,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.010325866856152426,
      "loss": 0.2298,
      "num_input_tokens_seen": 24585664,
      "step": 116495
    },
    {
      "epoch": 12.816281628162816,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.010324498536025704,
      "loss": 0.2298,
      "num_input_tokens_seen": 24586752,
      "step": 116500
    },
    {
      "epoch": 12.816831683168317,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.010323130258990842,
      "loss": 0.2319,
      "num_input_tokens_seen": 24587776,
      "step": 116505
    },
    {
      "epoch": 12.817381738173818,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.010321762025060435,
      "loss": 0.2319,
      "num_input_tokens_seen": 24588800,
      "step": 116510
    },
    {
      "epoch": 12.817931793179318,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.010320393834247106,
      "loss": 0.2298,
      "num_input_tokens_seen": 24589792,
      "step": 116515
    },
    {
      "epoch": 12.818481848184819,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010319025686563458,
      "loss": 0.2309,
      "num_input_tokens_seen": 24590816,
      "step": 116520
    },
    {
      "epoch": 12.81903190319032,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.010317657582022114,
      "loss": 0.2303,
      "num_input_tokens_seen": 24591936,
      "step": 116525
    },
    {
      "epoch": 12.819581958195819,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.010316289520635666,
      "loss": 0.2309,
      "num_input_tokens_seen": 24592928,
      "step": 116530
    },
    {
      "epoch": 12.82013201320132,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.01031492150241673,
      "loss": 0.2288,
      "num_input_tokens_seen": 24594016,
      "step": 116535
    },
    {
      "epoch": 12.820682068206821,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.010313553527377915,
      "loss": 0.2324,
      "num_input_tokens_seen": 24595072,
      "step": 116540
    },
    {
      "epoch": 12.82123212321232,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.010312185595531827,
      "loss": 0.2309,
      "num_input_tokens_seen": 24596096,
      "step": 116545
    },
    {
      "epoch": 12.821782178217822,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.010310817706891079,
      "loss": 0.2304,
      "num_input_tokens_seen": 24597152,
      "step": 116550
    },
    {
      "epoch": 12.822332233223323,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.010309449861468272,
      "loss": 0.2319,
      "num_input_tokens_seen": 24598176,
      "step": 116555
    },
    {
      "epoch": 12.822882288228822,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.01030808205927601,
      "loss": 0.2309,
      "num_input_tokens_seen": 24599232,
      "step": 116560
    },
    {
      "epoch": 12.823432343234323,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.010306714300326914,
      "loss": 0.233,
      "num_input_tokens_seen": 24600224,
      "step": 116565
    },
    {
      "epoch": 12.823982398239824,
      "grad_norm": 0.000629425048828125,
      "learning_rate": 0.010305346584633572,
      "loss": 0.2324,
      "num_input_tokens_seen": 24601248,
      "step": 116570
    },
    {
      "epoch": 12.824532453245325,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010303978912208602,
      "loss": 0.2319,
      "num_input_tokens_seen": 24602272,
      "step": 116575
    },
    {
      "epoch": 12.825082508250825,
      "grad_norm": 0.000812530517578125,
      "learning_rate": 0.010302611283064608,
      "loss": 0.2329,
      "num_input_tokens_seen": 24603360,
      "step": 116580
    },
    {
      "epoch": 12.825632563256326,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.010301243697214186,
      "loss": 0.2309,
      "num_input_tokens_seen": 24604448,
      "step": 116585
    },
    {
      "epoch": 12.826182618261827,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.010299876154669948,
      "loss": 0.2319,
      "num_input_tokens_seen": 24605536,
      "step": 116590
    },
    {
      "epoch": 12.826732673267326,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.010298508655444492,
      "loss": 0.2314,
      "num_input_tokens_seen": 24606656,
      "step": 116595
    },
    {
      "epoch": 12.827282728272827,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.010297141199550431,
      "loss": 0.2324,
      "num_input_tokens_seen": 24607712,
      "step": 116600
    },
    {
      "epoch": 12.827832783278328,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.010295773787000365,
      "loss": 0.2319,
      "num_input_tokens_seen": 24608768,
      "step": 116605
    },
    {
      "epoch": 12.828382838283828,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01029440641780689,
      "loss": 0.2314,
      "num_input_tokens_seen": 24609824,
      "step": 116610
    },
    {
      "epoch": 12.828932893289329,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.010293039091982615,
      "loss": 0.2335,
      "num_input_tokens_seen": 24610816,
      "step": 116615
    },
    {
      "epoch": 12.82948294829483,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.010291671809540144,
      "loss": 0.2319,
      "num_input_tokens_seen": 24611808,
      "step": 116620
    },
    {
      "epoch": 12.83003300330033,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.010290304570492065,
      "loss": 0.2314,
      "num_input_tokens_seen": 24612832,
      "step": 116625
    },
    {
      "epoch": 12.83058305830583,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.010288937374850995,
      "loss": 0.2303,
      "num_input_tokens_seen": 24613888,
      "step": 116630
    },
    {
      "epoch": 12.831133113311331,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010287570222629527,
      "loss": 0.2314,
      "num_input_tokens_seen": 24614976,
      "step": 116635
    },
    {
      "epoch": 12.831683168316832,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.010286203113840266,
      "loss": 0.2335,
      "num_input_tokens_seen": 24616032,
      "step": 116640
    },
    {
      "epoch": 12.832233223322332,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010284836048495807,
      "loss": 0.2308,
      "num_input_tokens_seen": 24617088,
      "step": 116645
    },
    {
      "epoch": 12.832783278327833,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01028346902660875,
      "loss": 0.2303,
      "num_input_tokens_seen": 24618176,
      "step": 116650
    },
    {
      "epoch": 12.833333333333334,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0102821020481917,
      "loss": 0.233,
      "num_input_tokens_seen": 24619232,
      "step": 116655
    },
    {
      "epoch": 12.833883388338833,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.010280735113257243,
      "loss": 0.2309,
      "num_input_tokens_seen": 24620320,
      "step": 116660
    },
    {
      "epoch": 12.834433443344334,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.010279368221817995,
      "loss": 0.2335,
      "num_input_tokens_seen": 24621440,
      "step": 116665
    },
    {
      "epoch": 12.834983498349835,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.010278001373886542,
      "loss": 0.2314,
      "num_input_tokens_seen": 24622464,
      "step": 116670
    },
    {
      "epoch": 12.835533553355335,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.010276634569475482,
      "loss": 0.2298,
      "num_input_tokens_seen": 24623488,
      "step": 116675
    },
    {
      "epoch": 12.836083608360836,
      "grad_norm": 0.000766754150390625,
      "learning_rate": 0.01027526780859742,
      "loss": 0.2309,
      "num_input_tokens_seen": 24624480,
      "step": 116680
    },
    {
      "epoch": 12.836633663366337,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.010273901091264942,
      "loss": 0.2298,
      "num_input_tokens_seen": 24625568,
      "step": 116685
    },
    {
      "epoch": 12.837183718371836,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.010272534417490651,
      "loss": 0.2314,
      "num_input_tokens_seen": 24626624,
      "step": 116690
    },
    {
      "epoch": 12.837733773377337,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.010271167787287146,
      "loss": 0.2298,
      "num_input_tokens_seen": 24627680,
      "step": 116695
    },
    {
      "epoch": 12.838283828382838,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.010269801200667013,
      "loss": 0.2304,
      "num_input_tokens_seen": 24628768,
      "step": 116700
    },
    {
      "epoch": 12.83883388338834,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.010268434657642857,
      "loss": 0.234,
      "num_input_tokens_seen": 24629792,
      "step": 116705
    },
    {
      "epoch": 12.839383938393839,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.010267068158227262,
      "loss": 0.2319,
      "num_input_tokens_seen": 24630880,
      "step": 116710
    },
    {
      "epoch": 12.83993399339934,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.010265701702432836,
      "loss": 0.2299,
      "num_input_tokens_seen": 24631968,
      "step": 116715
    },
    {
      "epoch": 12.840484048404841,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.010264335290272163,
      "loss": 0.2298,
      "num_input_tokens_seen": 24632960,
      "step": 116720
    },
    {
      "epoch": 12.84103410341034,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.010262968921757836,
      "loss": 0.2283,
      "num_input_tokens_seen": 24634016,
      "step": 116725
    },
    {
      "epoch": 12.841584158415841,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.010261602596902456,
      "loss": 0.2298,
      "num_input_tokens_seen": 24635040,
      "step": 116730
    },
    {
      "epoch": 12.842134213421343,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.010260236315718606,
      "loss": 0.2304,
      "num_input_tokens_seen": 24636128,
      "step": 116735
    },
    {
      "epoch": 12.842684268426842,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01025887007821889,
      "loss": 0.2314,
      "num_input_tokens_seen": 24637184,
      "step": 116740
    },
    {
      "epoch": 12.843234323432343,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01025750388441589,
      "loss": 0.2288,
      "num_input_tokens_seen": 24638176,
      "step": 116745
    },
    {
      "epoch": 12.843784378437844,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.0102561377343222,
      "loss": 0.2324,
      "num_input_tokens_seen": 24639168,
      "step": 116750
    },
    {
      "epoch": 12.844334433443345,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.010254771627950415,
      "loss": 0.2324,
      "num_input_tokens_seen": 24640192,
      "step": 116755
    },
    {
      "epoch": 12.844884488448844,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.01025340556531312,
      "loss": 0.2351,
      "num_input_tokens_seen": 24641280,
      "step": 116760
    },
    {
      "epoch": 12.845434543454346,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.01025203954642291,
      "loss": 0.2335,
      "num_input_tokens_seen": 24642336,
      "step": 116765
    },
    {
      "epoch": 12.845984598459847,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.010250673571292375,
      "loss": 0.2335,
      "num_input_tokens_seen": 24643360,
      "step": 116770
    },
    {
      "epoch": 12.846534653465346,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010249307639934096,
      "loss": 0.2309,
      "num_input_tokens_seen": 24644416,
      "step": 116775
    },
    {
      "epoch": 12.847084708470847,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.010247941752360679,
      "loss": 0.2329,
      "num_input_tokens_seen": 24645408,
      "step": 116780
    },
    {
      "epoch": 12.847634763476348,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.010246575908584691,
      "loss": 0.2314,
      "num_input_tokens_seen": 24646528,
      "step": 116785
    },
    {
      "epoch": 12.848184818481847,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.01024521010861874,
      "loss": 0.2324,
      "num_input_tokens_seen": 24647616,
      "step": 116790
    },
    {
      "epoch": 12.848734873487349,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010243844352475408,
      "loss": 0.2319,
      "num_input_tokens_seen": 24648672,
      "step": 116795
    },
    {
      "epoch": 12.84928492849285,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.010242478640167273,
      "loss": 0.2293,
      "num_input_tokens_seen": 24649696,
      "step": 116800
    },
    {
      "epoch": 12.84983498349835,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.010241112971706934,
      "loss": 0.2319,
      "num_input_tokens_seen": 24650688,
      "step": 116805
    },
    {
      "epoch": 12.85038503850385,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.010239747347106974,
      "loss": 0.2335,
      "num_input_tokens_seen": 24651744,
      "step": 116810
    },
    {
      "epoch": 12.850935093509351,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010238381766379975,
      "loss": 0.2304,
      "num_input_tokens_seen": 24652736,
      "step": 116815
    },
    {
      "epoch": 12.851485148514852,
      "grad_norm": 0.01025390625,
      "learning_rate": 0.010237016229538527,
      "loss": 0.2309,
      "num_input_tokens_seen": 24653760,
      "step": 116820
    },
    {
      "epoch": 12.852035203520352,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.010235650736595214,
      "loss": 0.2335,
      "num_input_tokens_seen": 24654848,
      "step": 116825
    },
    {
      "epoch": 12.852585258525853,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.01023428528756263,
      "loss": 0.2314,
      "num_input_tokens_seen": 24655840,
      "step": 116830
    },
    {
      "epoch": 12.853135313531354,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.010232919882453343,
      "loss": 0.2335,
      "num_input_tokens_seen": 24656896,
      "step": 116835
    },
    {
      "epoch": 12.853685368536853,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010231554521279947,
      "loss": 0.234,
      "num_input_tokens_seen": 24657984,
      "step": 116840
    },
    {
      "epoch": 12.854235423542354,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.010230189204055025,
      "loss": 0.2314,
      "num_input_tokens_seen": 24659040,
      "step": 116845
    },
    {
      "epoch": 12.854785478547855,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 0.01022882393079116,
      "loss": 0.2303,
      "num_input_tokens_seen": 24660096,
      "step": 116850
    },
    {
      "epoch": 12.855335533553355,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.01022745870150094,
      "loss": 0.2319,
      "num_input_tokens_seen": 24661184,
      "step": 116855
    },
    {
      "epoch": 12.855885588558856,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.010226093516196941,
      "loss": 0.2314,
      "num_input_tokens_seen": 24662240,
      "step": 116860
    },
    {
      "epoch": 12.856435643564357,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010224728374891743,
      "loss": 0.2308,
      "num_input_tokens_seen": 24663200,
      "step": 116865
    },
    {
      "epoch": 12.856985698569858,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.01022336327759794,
      "loss": 0.2319,
      "num_input_tokens_seen": 24664224,
      "step": 116870
    },
    {
      "epoch": 12.857535753575357,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.010221998224328099,
      "loss": 0.2309,
      "num_input_tokens_seen": 24665248,
      "step": 116875
    },
    {
      "epoch": 12.858085808580858,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.01022063321509481,
      "loss": 0.2298,
      "num_input_tokens_seen": 24666272,
      "step": 116880
    },
    {
      "epoch": 12.85863586358636,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.010219268249910655,
      "loss": 0.2319,
      "num_input_tokens_seen": 24667296,
      "step": 116885
    },
    {
      "epoch": 12.859185918591859,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.010217903328788204,
      "loss": 0.2309,
      "num_input_tokens_seen": 24668352,
      "step": 116890
    },
    {
      "epoch": 12.85973597359736,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.010216538451740045,
      "loss": 0.2303,
      "num_input_tokens_seen": 24669408,
      "step": 116895
    },
    {
      "epoch": 12.86028602860286,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.010215173618778755,
      "loss": 0.2298,
      "num_input_tokens_seen": 24670496,
      "step": 116900
    },
    {
      "epoch": 12.86083608360836,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.010213808829916916,
      "loss": 0.2288,
      "num_input_tokens_seen": 24671520,
      "step": 116905
    },
    {
      "epoch": 12.861386138613861,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010212444085167107,
      "loss": 0.2314,
      "num_input_tokens_seen": 24672576,
      "step": 116910
    },
    {
      "epoch": 12.861936193619362,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.010211079384541898,
      "loss": 0.2309,
      "num_input_tokens_seen": 24673664,
      "step": 116915
    },
    {
      "epoch": 12.862486248624862,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.010209714728053876,
      "loss": 0.2293,
      "num_input_tokens_seen": 24674688,
      "step": 116920
    },
    {
      "epoch": 12.863036303630363,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.01020835011571561,
      "loss": 0.2298,
      "num_input_tokens_seen": 24675712,
      "step": 116925
    },
    {
      "epoch": 12.863586358635864,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010206985547539689,
      "loss": 0.2293,
      "num_input_tokens_seen": 24676736,
      "step": 116930
    },
    {
      "epoch": 12.864136413641365,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.01020562102353868,
      "loss": 0.2293,
      "num_input_tokens_seen": 24677824,
      "step": 116935
    },
    {
      "epoch": 12.864686468646864,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010204256543725159,
      "loss": 0.2304,
      "num_input_tokens_seen": 24678880,
      "step": 116940
    },
    {
      "epoch": 12.865236523652365,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010202892108111709,
      "loss": 0.2319,
      "num_input_tokens_seen": 24679936,
      "step": 116945
    },
    {
      "epoch": 12.865786578657866,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.010201527716710892,
      "loss": 0.2329,
      "num_input_tokens_seen": 24680928,
      "step": 116950
    },
    {
      "epoch": 12.866336633663366,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.010200163369535296,
      "loss": 0.2309,
      "num_input_tokens_seen": 24682016,
      "step": 116955
    },
    {
      "epoch": 12.866886688668867,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.010198799066597496,
      "loss": 0.234,
      "num_input_tokens_seen": 24683168,
      "step": 116960
    },
    {
      "epoch": 12.867436743674368,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.010197434807910051,
      "loss": 0.2325,
      "num_input_tokens_seen": 24684224,
      "step": 116965
    },
    {
      "epoch": 12.867986798679867,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.010196070593485552,
      "loss": 0.2309,
      "num_input_tokens_seen": 24685312,
      "step": 116970
    },
    {
      "epoch": 12.868536853685368,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.010194706423336563,
      "loss": 0.2325,
      "num_input_tokens_seen": 24686336,
      "step": 116975
    },
    {
      "epoch": 12.86908690869087,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.01019334229747566,
      "loss": 0.2335,
      "num_input_tokens_seen": 24687392,
      "step": 116980
    },
    {
      "epoch": 12.869636963696369,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010191978215915418,
      "loss": 0.2309,
      "num_input_tokens_seen": 24688480,
      "step": 116985
    },
    {
      "epoch": 12.87018701870187,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.0101906141786684,
      "loss": 0.2319,
      "num_input_tokens_seen": 24689504,
      "step": 116990
    },
    {
      "epoch": 12.870737073707371,
      "grad_norm": 0.0008392333984375,
      "learning_rate": 0.010189250185747185,
      "loss": 0.2304,
      "num_input_tokens_seen": 24690496,
      "step": 116995
    },
    {
      "epoch": 12.871287128712872,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.010187886237164347,
      "loss": 0.2298,
      "num_input_tokens_seen": 24691520,
      "step": 117000
    },
    {
      "epoch": 12.871837183718371,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010186522332932445,
      "loss": 0.2335,
      "num_input_tokens_seen": 24692544,
      "step": 117005
    },
    {
      "epoch": 12.872387238723872,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.010185158473064062,
      "loss": 0.233,
      "num_input_tokens_seen": 24693600,
      "step": 117010
    },
    {
      "epoch": 12.872937293729374,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.010183794657571757,
      "loss": 0.2324,
      "num_input_tokens_seen": 24694656,
      "step": 117015
    },
    {
      "epoch": 12.873487348734873,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010182430886468112,
      "loss": 0.2309,
      "num_input_tokens_seen": 24695712,
      "step": 117020
    },
    {
      "epoch": 12.874037403740374,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.01018106715976569,
      "loss": 0.2304,
      "num_input_tokens_seen": 24696768,
      "step": 117025
    },
    {
      "epoch": 12.874587458745875,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.010179703477477054,
      "loss": 0.2298,
      "num_input_tokens_seen": 24697856,
      "step": 117030
    },
    {
      "epoch": 12.875137513751374,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.010178339839614784,
      "loss": 0.233,
      "num_input_tokens_seen": 24698912,
      "step": 117035
    },
    {
      "epoch": 12.875687568756875,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010176976246191436,
      "loss": 0.2319,
      "num_input_tokens_seen": 24699936,
      "step": 117040
    },
    {
      "epoch": 12.876237623762377,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.010175612697219592,
      "loss": 0.2298,
      "num_input_tokens_seen": 24700928,
      "step": 117045
    },
    {
      "epoch": 12.876787678767876,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010174249192711805,
      "loss": 0.2325,
      "num_input_tokens_seen": 24702016,
      "step": 117050
    },
    {
      "epoch": 12.877337733773377,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010172885732680645,
      "loss": 0.2324,
      "num_input_tokens_seen": 24703008,
      "step": 117055
    },
    {
      "epoch": 12.877887788778878,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.010171522317138689,
      "loss": 0.2309,
      "num_input_tokens_seen": 24704064,
      "step": 117060
    },
    {
      "epoch": 12.87843784378438,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.010170158946098488,
      "loss": 0.2324,
      "num_input_tokens_seen": 24705120,
      "step": 117065
    },
    {
      "epoch": 12.878987898789878,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010168795619572617,
      "loss": 0.2298,
      "num_input_tokens_seen": 24706144,
      "step": 117070
    },
    {
      "epoch": 12.87953795379538,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.010167432337573641,
      "loss": 0.2314,
      "num_input_tokens_seen": 24707200,
      "step": 117075
    },
    {
      "epoch": 12.88008800880088,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010166069100114115,
      "loss": 0.2298,
      "num_input_tokens_seen": 24708192,
      "step": 117080
    },
    {
      "epoch": 12.88063806380638,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.010164705907206618,
      "loss": 0.2309,
      "num_input_tokens_seen": 24709248,
      "step": 117085
    },
    {
      "epoch": 12.881188118811881,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010163342758863702,
      "loss": 0.2324,
      "num_input_tokens_seen": 24710368,
      "step": 117090
    },
    {
      "epoch": 12.881738173817382,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.010161979655097936,
      "loss": 0.2303,
      "num_input_tokens_seen": 24711520,
      "step": 117095
    },
    {
      "epoch": 12.882288228822881,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.010160616595921887,
      "loss": 0.2314,
      "num_input_tokens_seen": 24712512,
      "step": 117100
    },
    {
      "epoch": 12.882838283828383,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.010159253581348105,
      "loss": 0.2314,
      "num_input_tokens_seen": 24713568,
      "step": 117105
    },
    {
      "epoch": 12.883388338833884,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.010157890611389164,
      "loss": 0.2335,
      "num_input_tokens_seen": 24714624,
      "step": 117110
    },
    {
      "epoch": 12.883938393839383,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.010156527686057617,
      "loss": 0.2324,
      "num_input_tokens_seen": 24715680,
      "step": 117115
    },
    {
      "epoch": 12.884488448844884,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.010155164805366037,
      "loss": 0.2319,
      "num_input_tokens_seen": 24716736,
      "step": 117120
    },
    {
      "epoch": 12.885038503850385,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.010153801969326977,
      "loss": 0.2324,
      "num_input_tokens_seen": 24717760,
      "step": 117125
    },
    {
      "epoch": 12.885588558855886,
      "grad_norm": 0.01019287109375,
      "learning_rate": 0.010152439177952994,
      "loss": 0.2319,
      "num_input_tokens_seen": 24718816,
      "step": 117130
    },
    {
      "epoch": 12.886138613861386,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010151076431256659,
      "loss": 0.2304,
      "num_input_tokens_seen": 24719904,
      "step": 117135
    },
    {
      "epoch": 12.886688668866887,
      "grad_norm": 0.000766754150390625,
      "learning_rate": 0.010149713729250517,
      "loss": 0.2303,
      "num_input_tokens_seen": 24720992,
      "step": 117140
    },
    {
      "epoch": 12.887238723872388,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.010148351071947146,
      "loss": 0.2325,
      "num_input_tokens_seen": 24721984,
      "step": 117145
    },
    {
      "epoch": 12.887788778877887,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.010146988459359092,
      "loss": 0.2314,
      "num_input_tokens_seen": 24723040,
      "step": 117150
    },
    {
      "epoch": 12.888338833883388,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.010145625891498913,
      "loss": 0.2309,
      "num_input_tokens_seen": 24724096,
      "step": 117155
    },
    {
      "epoch": 12.88888888888889,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.010144263368379177,
      "loss": 0.2319,
      "num_input_tokens_seen": 24725184,
      "step": 117160
    },
    {
      "epoch": 12.88943894389439,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.010142900890012428,
      "loss": 0.2314,
      "num_input_tokens_seen": 24726240,
      "step": 117165
    },
    {
      "epoch": 12.88998899889989,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.010141538456411237,
      "loss": 0.2324,
      "num_input_tokens_seen": 24727296,
      "step": 117170
    },
    {
      "epoch": 12.89053905390539,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.010140176067588156,
      "loss": 0.2314,
      "num_input_tokens_seen": 24728352,
      "step": 117175
    },
    {
      "epoch": 12.891089108910892,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.010138813723555734,
      "loss": 0.233,
      "num_input_tokens_seen": 24729440,
      "step": 117180
    },
    {
      "epoch": 12.891639163916391,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.010137451424326536,
      "loss": 0.2314,
      "num_input_tokens_seen": 24730496,
      "step": 117185
    },
    {
      "epoch": 12.892189218921892,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01013608916991312,
      "loss": 0.2303,
      "num_input_tokens_seen": 24731584,
      "step": 117190
    },
    {
      "epoch": 12.892739273927393,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010134726960328027,
      "loss": 0.2293,
      "num_input_tokens_seen": 24732704,
      "step": 117195
    },
    {
      "epoch": 12.893289328932893,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.010133364795583827,
      "loss": 0.2293,
      "num_input_tokens_seen": 24733888,
      "step": 117200
    },
    {
      "epoch": 12.893839383938394,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.010132002675693063,
      "loss": 0.2299,
      "num_input_tokens_seen": 24734944,
      "step": 117205
    },
    {
      "epoch": 12.894389438943895,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.010130640600668301,
      "loss": 0.2314,
      "num_input_tokens_seen": 24736000,
      "step": 117210
    },
    {
      "epoch": 12.894939493949394,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.010129278570522088,
      "loss": 0.2303,
      "num_input_tokens_seen": 24737056,
      "step": 117215
    },
    {
      "epoch": 12.895489548954895,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.010127916585266973,
      "loss": 0.2309,
      "num_input_tokens_seen": 24738112,
      "step": 117220
    },
    {
      "epoch": 12.896039603960396,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010126554644915513,
      "loss": 0.2303,
      "num_input_tokens_seen": 24739168,
      "step": 117225
    },
    {
      "epoch": 12.896589658965897,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.01012519274948026,
      "loss": 0.2329,
      "num_input_tokens_seen": 24740224,
      "step": 117230
    },
    {
      "epoch": 12.897139713971397,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.010123830898973772,
      "loss": 0.2314,
      "num_input_tokens_seen": 24741216,
      "step": 117235
    },
    {
      "epoch": 12.897689768976898,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.010122469093408592,
      "loss": 0.2309,
      "num_input_tokens_seen": 24742304,
      "step": 117240
    },
    {
      "epoch": 12.898239823982399,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.010121107332797268,
      "loss": 0.2293,
      "num_input_tokens_seen": 24743456,
      "step": 117245
    },
    {
      "epoch": 12.898789878987898,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010119745617152365,
      "loss": 0.2324,
      "num_input_tokens_seen": 24744448,
      "step": 117250
    },
    {
      "epoch": 12.8993399339934,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.010118383946486419,
      "loss": 0.2314,
      "num_input_tokens_seen": 24745408,
      "step": 117255
    },
    {
      "epoch": 12.8998899889989,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.010117022320811988,
      "loss": 0.234,
      "num_input_tokens_seen": 24746464,
      "step": 117260
    },
    {
      "epoch": 12.9004400440044,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.01011566074014162,
      "loss": 0.2319,
      "num_input_tokens_seen": 24747552,
      "step": 117265
    },
    {
      "epoch": 12.900990099009901,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010114299204487861,
      "loss": 0.2319,
      "num_input_tokens_seen": 24748608,
      "step": 117270
    },
    {
      "epoch": 12.901540154015402,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.010112937713863266,
      "loss": 0.2335,
      "num_input_tokens_seen": 24749696,
      "step": 117275
    },
    {
      "epoch": 12.902090209020901,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.010111576268280372,
      "loss": 0.233,
      "num_input_tokens_seen": 24750784,
      "step": 117280
    },
    {
      "epoch": 12.902640264026402,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01011021486775174,
      "loss": 0.2309,
      "num_input_tokens_seen": 24751808,
      "step": 117285
    },
    {
      "epoch": 12.903190319031903,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010108853512289913,
      "loss": 0.2309,
      "num_input_tokens_seen": 24752864,
      "step": 117290
    },
    {
      "epoch": 12.903740374037405,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.010107492201907431,
      "loss": 0.2319,
      "num_input_tokens_seen": 24753888,
      "step": 117295
    },
    {
      "epoch": 12.904290429042904,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.010106130936616846,
      "loss": 0.2298,
      "num_input_tokens_seen": 24754944,
      "step": 117300
    },
    {
      "epoch": 12.904840484048405,
      "grad_norm": 0.00074005126953125,
      "learning_rate": 0.010104769716430703,
      "loss": 0.2324,
      "num_input_tokens_seen": 24756000,
      "step": 117305
    },
    {
      "epoch": 12.905390539053906,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.010103408541361554,
      "loss": 0.2324,
      "num_input_tokens_seen": 24757056,
      "step": 117310
    },
    {
      "epoch": 12.905940594059405,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.010102047411421934,
      "loss": 0.2288,
      "num_input_tokens_seen": 24758144,
      "step": 117315
    },
    {
      "epoch": 12.906490649064907,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.010100686326624392,
      "loss": 0.2303,
      "num_input_tokens_seen": 24759232,
      "step": 117320
    },
    {
      "epoch": 12.907040704070408,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010099325286981478,
      "loss": 0.2293,
      "num_input_tokens_seen": 24760288,
      "step": 117325
    },
    {
      "epoch": 12.907590759075907,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010097964292505725,
      "loss": 0.2319,
      "num_input_tokens_seen": 24761280,
      "step": 117330
    },
    {
      "epoch": 12.908140814081408,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01009660334320969,
      "loss": 0.2298,
      "num_input_tokens_seen": 24762336,
      "step": 117335
    },
    {
      "epoch": 12.908690869086909,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.010095242439105907,
      "loss": 0.2309,
      "num_input_tokens_seen": 24763360,
      "step": 117340
    },
    {
      "epoch": 12.909240924092408,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.010093881580206916,
      "loss": 0.2319,
      "num_input_tokens_seen": 24764384,
      "step": 117345
    },
    {
      "epoch": 12.90979097909791,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.010092520766525273,
      "loss": 0.2303,
      "num_input_tokens_seen": 24765440,
      "step": 117350
    },
    {
      "epoch": 12.91034103410341,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.010091159998073503,
      "loss": 0.2298,
      "num_input_tokens_seen": 24766592,
      "step": 117355
    },
    {
      "epoch": 12.910891089108912,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.010089799274864157,
      "loss": 0.2304,
      "num_input_tokens_seen": 24767616,
      "step": 117360
    },
    {
      "epoch": 12.911441144114411,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.010088438596909782,
      "loss": 0.2308,
      "num_input_tokens_seen": 24768704,
      "step": 117365
    },
    {
      "epoch": 12.911991199119912,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.010087077964222906,
      "loss": 0.2335,
      "num_input_tokens_seen": 24769792,
      "step": 117370
    },
    {
      "epoch": 12.912541254125413,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.010085717376816076,
      "loss": 0.2319,
      "num_input_tokens_seen": 24770816,
      "step": 117375
    },
    {
      "epoch": 12.913091309130913,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.010084356834701831,
      "loss": 0.2308,
      "num_input_tokens_seen": 24771904,
      "step": 117380
    },
    {
      "epoch": 12.913641364136414,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010082996337892704,
      "loss": 0.2288,
      "num_input_tokens_seen": 24772992,
      "step": 117385
    },
    {
      "epoch": 12.914191419141915,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.01008163588640125,
      "loss": 0.2324,
      "num_input_tokens_seen": 24774080,
      "step": 117390
    },
    {
      "epoch": 12.914741474147414,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.010080275480239992,
      "loss": 0.2319,
      "num_input_tokens_seen": 24775168,
      "step": 117395
    },
    {
      "epoch": 12.915291529152915,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.010078915119421477,
      "loss": 0.2309,
      "num_input_tokens_seen": 24776160,
      "step": 117400
    },
    {
      "epoch": 12.915841584158416,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.010077554803958243,
      "loss": 0.2298,
      "num_input_tokens_seen": 24777216,
      "step": 117405
    },
    {
      "epoch": 12.916391639163916,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.010076194533862817,
      "loss": 0.2309,
      "num_input_tokens_seen": 24778304,
      "step": 117410
    },
    {
      "epoch": 12.916941694169417,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.010074834309147747,
      "loss": 0.2314,
      "num_input_tokens_seen": 24779264,
      "step": 117415
    },
    {
      "epoch": 12.917491749174918,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.010073474129825564,
      "loss": 0.2314,
      "num_input_tokens_seen": 24780256,
      "step": 117420
    },
    {
      "epoch": 12.918041804180419,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.010072113995908811,
      "loss": 0.2298,
      "num_input_tokens_seen": 24781344,
      "step": 117425
    },
    {
      "epoch": 12.918591859185918,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.010070753907410015,
      "loss": 0.2324,
      "num_input_tokens_seen": 24782464,
      "step": 117430
    },
    {
      "epoch": 12.91914191419142,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.010069393864341712,
      "loss": 0.2319,
      "num_input_tokens_seen": 24783488,
      "step": 117435
    },
    {
      "epoch": 12.91969196919692,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010068033866716447,
      "loss": 0.2314,
      "num_input_tokens_seen": 24784512,
      "step": 117440
    },
    {
      "epoch": 12.92024202420242,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.010066673914546737,
      "loss": 0.2278,
      "num_input_tokens_seen": 24785600,
      "step": 117445
    },
    {
      "epoch": 12.92079207920792,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.010065314007845138,
      "loss": 0.2309,
      "num_input_tokens_seen": 24786624,
      "step": 117450
    },
    {
      "epoch": 12.921342134213422,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.010063954146624168,
      "loss": 0.233,
      "num_input_tokens_seen": 24787680,
      "step": 117455
    },
    {
      "epoch": 12.921892189218921,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.010062594330896358,
      "loss": 0.2314,
      "num_input_tokens_seen": 24788736,
      "step": 117460
    },
    {
      "epoch": 12.922442244224422,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.010061234560674254,
      "loss": 0.2293,
      "num_input_tokens_seen": 24789696,
      "step": 117465
    },
    {
      "epoch": 12.922992299229923,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.010059874835970376,
      "loss": 0.2298,
      "num_input_tokens_seen": 24790720,
      "step": 117470
    },
    {
      "epoch": 12.923542354235423,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.010058515156797265,
      "loss": 0.2314,
      "num_input_tokens_seen": 24791744,
      "step": 117475
    },
    {
      "epoch": 12.924092409240924,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.010057155523167452,
      "loss": 0.2319,
      "num_input_tokens_seen": 24792832,
      "step": 117480
    },
    {
      "epoch": 12.924642464246425,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.010055795935093457,
      "loss": 0.2324,
      "num_input_tokens_seen": 24793856,
      "step": 117485
    },
    {
      "epoch": 12.925192519251926,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.010054436392587822,
      "loss": 0.2293,
      "num_input_tokens_seen": 24794944,
      "step": 117490
    },
    {
      "epoch": 12.925742574257425,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.01005307689566307,
      "loss": 0.2324,
      "num_input_tokens_seen": 24796032,
      "step": 117495
    },
    {
      "epoch": 12.926292629262926,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.010051717444331741,
      "loss": 0.2314,
      "num_input_tokens_seen": 24797056,
      "step": 117500
    },
    {
      "epoch": 12.926842684268427,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.010050358038606355,
      "loss": 0.2288,
      "num_input_tokens_seen": 24798112,
      "step": 117505
    },
    {
      "epoch": 12.927392739273927,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.01004899867849944,
      "loss": 0.2329,
      "num_input_tokens_seen": 24799136,
      "step": 117510
    },
    {
      "epoch": 12.927942794279428,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.010047639364023534,
      "loss": 0.2314,
      "num_input_tokens_seen": 24800128,
      "step": 117515
    },
    {
      "epoch": 12.928492849284929,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.010046280095191155,
      "loss": 0.2304,
      "num_input_tokens_seen": 24801248,
      "step": 117520
    },
    {
      "epoch": 12.929042904290428,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.010044920872014841,
      "loss": 0.2309,
      "num_input_tokens_seen": 24802336,
      "step": 117525
    },
    {
      "epoch": 12.92959295929593,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.010043561694507113,
      "loss": 0.2314,
      "num_input_tokens_seen": 24803424,
      "step": 117530
    },
    {
      "epoch": 12.93014301430143,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.010042202562680493,
      "loss": 0.2293,
      "num_input_tokens_seen": 24804544,
      "step": 117535
    },
    {
      "epoch": 12.930693069306932,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01004084347654752,
      "loss": 0.2319,
      "num_input_tokens_seen": 24805600,
      "step": 117540
    },
    {
      "epoch": 12.93124312431243,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010039484436120705,
      "loss": 0.2314,
      "num_input_tokens_seen": 24806656,
      "step": 117545
    },
    {
      "epoch": 12.931793179317932,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.010038125441412587,
      "loss": 0.2319,
      "num_input_tokens_seen": 24807680,
      "step": 117550
    },
    {
      "epoch": 12.932343234323433,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.010036766492435688,
      "loss": 0.233,
      "num_input_tokens_seen": 24808768,
      "step": 117555
    },
    {
      "epoch": 12.932893289328932,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.010035407589202522,
      "loss": 0.2303,
      "num_input_tokens_seen": 24809824,
      "step": 117560
    },
    {
      "epoch": 12.933443344334433,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.01003404873172563,
      "loss": 0.2303,
      "num_input_tokens_seen": 24810848,
      "step": 117565
    },
    {
      "epoch": 12.933993399339935,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.010032689920017527,
      "loss": 0.2303,
      "num_input_tokens_seen": 24811936,
      "step": 117570
    },
    {
      "epoch": 12.934543454345434,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.01003133115409073,
      "loss": 0.2324,
      "num_input_tokens_seen": 24812928,
      "step": 117575
    },
    {
      "epoch": 12.935093509350935,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010029972433957783,
      "loss": 0.2314,
      "num_input_tokens_seen": 24814048,
      "step": 117580
    },
    {
      "epoch": 12.935643564356436,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.010028613759631183,
      "loss": 0.2319,
      "num_input_tokens_seen": 24815104,
      "step": 117585
    },
    {
      "epoch": 12.936193619361937,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.010027255131123469,
      "loss": 0.2319,
      "num_input_tokens_seen": 24816192,
      "step": 117590
    },
    {
      "epoch": 12.936743674367436,
      "grad_norm": 0.0007171630859375,
      "learning_rate": 0.010025896548447162,
      "loss": 0.2298,
      "num_input_tokens_seen": 24817248,
      "step": 117595
    },
    {
      "epoch": 12.937293729372938,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.010024538011614775,
      "loss": 0.2324,
      "num_input_tokens_seen": 24818304,
      "step": 117600
    },
    {
      "epoch": 12.937843784378439,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.010023179520638833,
      "loss": 0.2324,
      "num_input_tokens_seen": 24819392,
      "step": 117605
    },
    {
      "epoch": 12.938393839383938,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.010021821075531857,
      "loss": 0.2309,
      "num_input_tokens_seen": 24820416,
      "step": 117610
    },
    {
      "epoch": 12.938943894389439,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.010020462676306373,
      "loss": 0.2309,
      "num_input_tokens_seen": 24821472,
      "step": 117615
    },
    {
      "epoch": 12.93949394939494,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.01001910432297489,
      "loss": 0.2314,
      "num_input_tokens_seen": 24822528,
      "step": 117620
    },
    {
      "epoch": 12.94004400440044,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.010017746015549932,
      "loss": 0.2319,
      "num_input_tokens_seen": 24823584,
      "step": 117625
    },
    {
      "epoch": 12.94059405940594,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.010016387754044022,
      "loss": 0.2319,
      "num_input_tokens_seen": 24824672,
      "step": 117630
    },
    {
      "epoch": 12.941144114411442,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.01001502953846967,
      "loss": 0.2309,
      "num_input_tokens_seen": 24825728,
      "step": 117635
    },
    {
      "epoch": 12.941694169416941,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010013671368839405,
      "loss": 0.2324,
      "num_input_tokens_seen": 24826752,
      "step": 117640
    },
    {
      "epoch": 12.942244224422442,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.010012313245165733,
      "loss": 0.2309,
      "num_input_tokens_seen": 24827808,
      "step": 117645
    },
    {
      "epoch": 12.942794279427943,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.010010955167461177,
      "loss": 0.2319,
      "num_input_tokens_seen": 24828928,
      "step": 117650
    },
    {
      "epoch": 12.943344334433444,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.010009597135738258,
      "loss": 0.2308,
      "num_input_tokens_seen": 24829920,
      "step": 117655
    },
    {
      "epoch": 12.943894389438944,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.010008239150009483,
      "loss": 0.2314,
      "num_input_tokens_seen": 24830912,
      "step": 117660
    },
    {
      "epoch": 12.944444444444445,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.010006881210287373,
      "loss": 0.2309,
      "num_input_tokens_seen": 24832064,
      "step": 117665
    },
    {
      "epoch": 12.944994499449946,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.010005523316584447,
      "loss": 0.2329,
      "num_input_tokens_seen": 24833120,
      "step": 117670
    },
    {
      "epoch": 12.945544554455445,
      "grad_norm": 0.001953125,
      "learning_rate": 0.010004165468913208,
      "loss": 0.2324,
      "num_input_tokens_seen": 24834208,
      "step": 117675
    },
    {
      "epoch": 12.946094609460946,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.010002807667286185,
      "loss": 0.2298,
      "num_input_tokens_seen": 24835232,
      "step": 117680
    },
    {
      "epoch": 12.946644664466447,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.01000144991171588,
      "loss": 0.2304,
      "num_input_tokens_seen": 24836288,
      "step": 117685
    },
    {
      "epoch": 12.947194719471947,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.010000092202214816,
      "loss": 0.2298,
      "num_input_tokens_seen": 24837344,
      "step": 117690
    },
    {
      "epoch": 12.947744774477448,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009998734538795509,
      "loss": 0.2319,
      "num_input_tokens_seen": 24838464,
      "step": 117695
    },
    {
      "epoch": 12.948294829482949,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00999737692147046,
      "loss": 0.2298,
      "num_input_tokens_seen": 24839488,
      "step": 117700
    },
    {
      "epoch": 12.948844884488448,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009996019350252187,
      "loss": 0.2319,
      "num_input_tokens_seen": 24840544,
      "step": 117705
    },
    {
      "epoch": 12.94939493949395,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.009994661825153201,
      "loss": 0.233,
      "num_input_tokens_seen": 24841568,
      "step": 117710
    },
    {
      "epoch": 12.94994499449945,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.00999330434618602,
      "loss": 0.2314,
      "num_input_tokens_seen": 24842624,
      "step": 117715
    },
    {
      "epoch": 12.950495049504951,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.009991946913363149,
      "loss": 0.2303,
      "num_input_tokens_seen": 24843648,
      "step": 117720
    },
    {
      "epoch": 12.95104510451045,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.009990589526697097,
      "loss": 0.2313,
      "num_input_tokens_seen": 24844640,
      "step": 117725
    },
    {
      "epoch": 12.951595159515952,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.009989232186200384,
      "loss": 0.2309,
      "num_input_tokens_seen": 24845696,
      "step": 117730
    },
    {
      "epoch": 12.952145214521453,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.009987874891885507,
      "loss": 0.2319,
      "num_input_tokens_seen": 24846752,
      "step": 117735
    },
    {
      "epoch": 12.952695269526952,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009986517643764984,
      "loss": 0.2303,
      "num_input_tokens_seen": 24847872,
      "step": 117740
    },
    {
      "epoch": 12.953245324532453,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009985160441851326,
      "loss": 0.2324,
      "num_input_tokens_seen": 24848896,
      "step": 117745
    },
    {
      "epoch": 12.953795379537954,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.009983803286157028,
      "loss": 0.2314,
      "num_input_tokens_seen": 24850016,
      "step": 117750
    },
    {
      "epoch": 12.954345434543454,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00998244617669462,
      "loss": 0.2319,
      "num_input_tokens_seen": 24851008,
      "step": 117755
    },
    {
      "epoch": 12.954895489548955,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.00998108911347659,
      "loss": 0.2329,
      "num_input_tokens_seen": 24852064,
      "step": 117760
    },
    {
      "epoch": 12.955445544554456,
      "grad_norm": 0.000522613525390625,
      "learning_rate": 0.009979732096515455,
      "loss": 0.2309,
      "num_input_tokens_seen": 24853056,
      "step": 117765
    },
    {
      "epoch": 12.955995599559955,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.009978375125823724,
      "loss": 0.2303,
      "num_input_tokens_seen": 24854112,
      "step": 117770
    },
    {
      "epoch": 12.956545654565456,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009977018201413897,
      "loss": 0.2314,
      "num_input_tokens_seen": 24855104,
      "step": 117775
    },
    {
      "epoch": 12.957095709570957,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009975661323298482,
      "loss": 0.2314,
      "num_input_tokens_seen": 24856128,
      "step": 117780
    },
    {
      "epoch": 12.957645764576458,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.00997430449148999,
      "loss": 0.2288,
      "num_input_tokens_seen": 24857216,
      "step": 117785
    },
    {
      "epoch": 12.958195819581958,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009972947706000917,
      "loss": 0.2298,
      "num_input_tokens_seen": 24858240,
      "step": 117790
    },
    {
      "epoch": 12.958745874587459,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009971590966843772,
      "loss": 0.2308,
      "num_input_tokens_seen": 24859232,
      "step": 117795
    },
    {
      "epoch": 12.95929592959296,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.009970234274031061,
      "loss": 0.2294,
      "num_input_tokens_seen": 24860352,
      "step": 117800
    },
    {
      "epoch": 12.95984598459846,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.00996887762757529,
      "loss": 0.2298,
      "num_input_tokens_seen": 24861376,
      "step": 117805
    },
    {
      "epoch": 12.96039603960396,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00996752102748896,
      "loss": 0.2319,
      "num_input_tokens_seen": 24862432,
      "step": 117810
    },
    {
      "epoch": 12.960946094609461,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009966164473784572,
      "loss": 0.2324,
      "num_input_tokens_seen": 24863456,
      "step": 117815
    },
    {
      "epoch": 12.96149614961496,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.00996480796647463,
      "loss": 0.2314,
      "num_input_tokens_seen": 24864576,
      "step": 117820
    },
    {
      "epoch": 12.962046204620462,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009963451505571635,
      "loss": 0.234,
      "num_input_tokens_seen": 24865632,
      "step": 117825
    },
    {
      "epoch": 12.962596259625963,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009962095091088098,
      "loss": 0.2314,
      "num_input_tokens_seen": 24866752,
      "step": 117830
    },
    {
      "epoch": 12.963146314631462,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.00996073872303651,
      "loss": 0.2303,
      "num_input_tokens_seen": 24867808,
      "step": 117835
    },
    {
      "epoch": 12.963696369636963,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.00995938240142937,
      "loss": 0.2314,
      "num_input_tokens_seen": 24868896,
      "step": 117840
    },
    {
      "epoch": 12.964246424642464,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.00995802612627919,
      "loss": 0.2308,
      "num_input_tokens_seen": 24870048,
      "step": 117845
    },
    {
      "epoch": 12.964796479647966,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.009956669897598459,
      "loss": 0.2324,
      "num_input_tokens_seen": 24871072,
      "step": 117850
    },
    {
      "epoch": 12.965346534653465,
      "grad_norm": 0.005859375,
      "learning_rate": 0.009955313715399684,
      "loss": 0.2319,
      "num_input_tokens_seen": 24872160,
      "step": 117855
    },
    {
      "epoch": 12.965896589658966,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.009953957579695366,
      "loss": 0.2314,
      "num_input_tokens_seen": 24873248,
      "step": 117860
    },
    {
      "epoch": 12.966446644664467,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.009952601490497988,
      "loss": 0.2319,
      "num_input_tokens_seen": 24874304,
      "step": 117865
    },
    {
      "epoch": 12.966996699669966,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.009951245447820072,
      "loss": 0.2298,
      "num_input_tokens_seen": 24875328,
      "step": 117870
    },
    {
      "epoch": 12.967546754675467,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.009949889451674096,
      "loss": 0.2309,
      "num_input_tokens_seen": 24876384,
      "step": 117875
    },
    {
      "epoch": 12.968096809680969,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00994853350207257,
      "loss": 0.2303,
      "num_input_tokens_seen": 24877472,
      "step": 117880
    },
    {
      "epoch": 12.968646864686468,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.00994717759902799,
      "loss": 0.2298,
      "num_input_tokens_seen": 24878464,
      "step": 117885
    },
    {
      "epoch": 12.969196919691969,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009945821742552843,
      "loss": 0.2319,
      "num_input_tokens_seen": 24879424,
      "step": 117890
    },
    {
      "epoch": 12.96974697469747,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.009944465932659633,
      "loss": 0.2293,
      "num_input_tokens_seen": 24880512,
      "step": 117895
    },
    {
      "epoch": 12.97029702970297,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009943110169360855,
      "loss": 0.233,
      "num_input_tokens_seen": 24881568,
      "step": 117900
    },
    {
      "epoch": 12.97084708470847,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009941754452669005,
      "loss": 0.2319,
      "num_input_tokens_seen": 24882560,
      "step": 117905
    },
    {
      "epoch": 12.971397139713972,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00994039878259658,
      "loss": 0.2324,
      "num_input_tokens_seen": 24883584,
      "step": 117910
    },
    {
      "epoch": 12.971947194719473,
      "grad_norm": 0.00077056884765625,
      "learning_rate": 0.009939043159156065,
      "loss": 0.2309,
      "num_input_tokens_seen": 24884640,
      "step": 117915
    },
    {
      "epoch": 12.972497249724972,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.009937687582359966,
      "loss": 0.2314,
      "num_input_tokens_seen": 24885728,
      "step": 117920
    },
    {
      "epoch": 12.973047304730473,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.009936332052220763,
      "loss": 0.2308,
      "num_input_tokens_seen": 24886688,
      "step": 117925
    },
    {
      "epoch": 12.973597359735974,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.00993497656875097,
      "loss": 0.2329,
      "num_input_tokens_seen": 24887744,
      "step": 117930
    },
    {
      "epoch": 12.974147414741473,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.009933621131963062,
      "loss": 0.2309,
      "num_input_tokens_seen": 24888864,
      "step": 117935
    },
    {
      "epoch": 12.974697469746975,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.009932265741869534,
      "loss": 0.2319,
      "num_input_tokens_seen": 24889920,
      "step": 117940
    },
    {
      "epoch": 12.975247524752476,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.009930910398482885,
      "loss": 0.2309,
      "num_input_tokens_seen": 24891040,
      "step": 117945
    },
    {
      "epoch": 12.975797579757975,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.0099295551018156,
      "loss": 0.2308,
      "num_input_tokens_seen": 24892064,
      "step": 117950
    },
    {
      "epoch": 12.976347634763476,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009928199851880174,
      "loss": 0.2303,
      "num_input_tokens_seen": 24893152,
      "step": 117955
    },
    {
      "epoch": 12.976897689768977,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.0099268446486891,
      "loss": 0.2329,
      "num_input_tokens_seen": 24894240,
      "step": 117960
    },
    {
      "epoch": 12.977447744774478,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009925489492254858,
      "loss": 0.2345,
      "num_input_tokens_seen": 24895360,
      "step": 117965
    },
    {
      "epoch": 12.977997799779978,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.009924134382589947,
      "loss": 0.2303,
      "num_input_tokens_seen": 24896416,
      "step": 117970
    },
    {
      "epoch": 12.978547854785479,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.009922779319706856,
      "loss": 0.2335,
      "num_input_tokens_seen": 24897440,
      "step": 117975
    },
    {
      "epoch": 12.97909790979098,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009921424303618066,
      "loss": 0.2293,
      "num_input_tokens_seen": 24898496,
      "step": 117980
    },
    {
      "epoch": 12.979647964796479,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009920069334336075,
      "loss": 0.2319,
      "num_input_tokens_seen": 24899520,
      "step": 117985
    },
    {
      "epoch": 12.98019801980198,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009918714411873366,
      "loss": 0.2314,
      "num_input_tokens_seen": 24900576,
      "step": 117990
    },
    {
      "epoch": 12.980748074807481,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009917359536242428,
      "loss": 0.2298,
      "num_input_tokens_seen": 24901632,
      "step": 117995
    },
    {
      "epoch": 12.98129812981298,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009916004707455753,
      "loss": 0.2319,
      "num_input_tokens_seen": 24902784,
      "step": 118000
    },
    {
      "epoch": 12.981848184818482,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.009914649925525819,
      "loss": 0.2309,
      "num_input_tokens_seen": 24903872,
      "step": 118005
    },
    {
      "epoch": 12.982398239823983,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.009913295190465115,
      "loss": 0.2324,
      "num_input_tokens_seen": 24905024,
      "step": 118010
    },
    {
      "epoch": 12.982948294829484,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.009911940502286129,
      "loss": 0.2324,
      "num_input_tokens_seen": 24906048,
      "step": 118015
    },
    {
      "epoch": 12.983498349834983,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.00991058586100135,
      "loss": 0.2314,
      "num_input_tokens_seen": 24907040,
      "step": 118020
    },
    {
      "epoch": 12.984048404840484,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009909231266623256,
      "loss": 0.2309,
      "num_input_tokens_seen": 24908128,
      "step": 118025
    },
    {
      "epoch": 12.984598459845985,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.009907876719164332,
      "loss": 0.2283,
      "num_input_tokens_seen": 24909216,
      "step": 118030
    },
    {
      "epoch": 12.985148514851485,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.009906522218637071,
      "loss": 0.2303,
      "num_input_tokens_seen": 24910304,
      "step": 118035
    },
    {
      "epoch": 12.985698569856986,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.009905167765053943,
      "loss": 0.2304,
      "num_input_tokens_seen": 24911328,
      "step": 118040
    },
    {
      "epoch": 12.986248624862487,
      "grad_norm": 0.000690460205078125,
      "learning_rate": 0.009903813358427445,
      "loss": 0.2314,
      "num_input_tokens_seen": 24912480,
      "step": 118045
    },
    {
      "epoch": 12.986798679867986,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.009902458998770052,
      "loss": 0.2314,
      "num_input_tokens_seen": 24913600,
      "step": 118050
    },
    {
      "epoch": 12.987348734873487,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009901104686094245,
      "loss": 0.2304,
      "num_input_tokens_seen": 24914656,
      "step": 118055
    },
    {
      "epoch": 12.987898789878988,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009899750420412518,
      "loss": 0.2324,
      "num_input_tokens_seen": 24915776,
      "step": 118060
    },
    {
      "epoch": 12.988448844884488,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.009898396201737335,
      "loss": 0.2319,
      "num_input_tokens_seen": 24916832,
      "step": 118065
    },
    {
      "epoch": 12.988998899889989,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.009897042030081191,
      "loss": 0.2314,
      "num_input_tokens_seen": 24917856,
      "step": 118070
    },
    {
      "epoch": 12.98954895489549,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.009895687905456564,
      "loss": 0.2319,
      "num_input_tokens_seen": 24918848,
      "step": 118075
    },
    {
      "epoch": 12.990099009900991,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.009894333827875926,
      "loss": 0.2319,
      "num_input_tokens_seen": 24919968,
      "step": 118080
    },
    {
      "epoch": 12.99064906490649,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.009892979797351768,
      "loss": 0.2324,
      "num_input_tokens_seen": 24920992,
      "step": 118085
    },
    {
      "epoch": 12.991199119911991,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00989162581389656,
      "loss": 0.2303,
      "num_input_tokens_seen": 24922144,
      "step": 118090
    },
    {
      "epoch": 12.991749174917492,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.009890271877522791,
      "loss": 0.2314,
      "num_input_tokens_seen": 24923232,
      "step": 118095
    },
    {
      "epoch": 12.992299229922992,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.009888917988242934,
      "loss": 0.2309,
      "num_input_tokens_seen": 24924256,
      "step": 118100
    },
    {
      "epoch": 12.992849284928493,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009887564146069462,
      "loss": 0.2319,
      "num_input_tokens_seen": 24925312,
      "step": 118105
    },
    {
      "epoch": 12.993399339933994,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.009886210351014862,
      "loss": 0.233,
      "num_input_tokens_seen": 24926272,
      "step": 118110
    },
    {
      "epoch": 12.993949394939493,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.009884856603091604,
      "loss": 0.233,
      "num_input_tokens_seen": 24927328,
      "step": 118115
    },
    {
      "epoch": 12.994499449944994,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.009883502902312174,
      "loss": 0.2324,
      "num_input_tokens_seen": 24928352,
      "step": 118120
    },
    {
      "epoch": 12.995049504950495,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.009882149248689041,
      "loss": 0.2314,
      "num_input_tokens_seen": 24929440,
      "step": 118125
    },
    {
      "epoch": 12.995599559955995,
      "grad_norm": 0.005859375,
      "learning_rate": 0.009880795642234678,
      "loss": 0.2304,
      "num_input_tokens_seen": 24930560,
      "step": 118130
    },
    {
      "epoch": 12.996149614961496,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.009879442082961572,
      "loss": 0.2324,
      "num_input_tokens_seen": 24931584,
      "step": 118135
    },
    {
      "epoch": 12.996699669966997,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.009878088570882184,
      "loss": 0.2314,
      "num_input_tokens_seen": 24932544,
      "step": 118140
    },
    {
      "epoch": 12.997249724972498,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.009876735106009001,
      "loss": 0.2329,
      "num_input_tokens_seen": 24933632,
      "step": 118145
    },
    {
      "epoch": 12.997799779977997,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.009875381688354494,
      "loss": 0.2293,
      "num_input_tokens_seen": 24934656,
      "step": 118150
    },
    {
      "epoch": 12.998349834983498,
      "grad_norm": 0.000537872314453125,
      "learning_rate": 0.009874028317931128,
      "loss": 0.2335,
      "num_input_tokens_seen": 24935680,
      "step": 118155
    },
    {
      "epoch": 12.998899889989,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009872674994751389,
      "loss": 0.2303,
      "num_input_tokens_seen": 24936704,
      "step": 118160
    },
    {
      "epoch": 12.999449944994499,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009871321718827743,
      "loss": 0.2324,
      "num_input_tokens_seen": 24937792,
      "step": 118165
    },
    {
      "epoch": 13.0,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.009869968490172657,
      "loss": 0.2324,
      "num_input_tokens_seen": 24938688,
      "step": 118170
    },
    {
      "epoch": 13.0,
      "eval_loss": 0.23142243921756744,
      "eval_runtime": 60.6375,
      "eval_samples_per_second": 66.625,
      "eval_steps_per_second": 16.656,
      "num_input_tokens_seen": 24938688,
      "step": 118170
    },
    {
      "epoch": 13.000550055005501,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009868615308798621,
      "loss": 0.2319,
      "num_input_tokens_seen": 24939744,
      "step": 118175
    },
    {
      "epoch": 13.001100110011,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009867262174718087,
      "loss": 0.2319,
      "num_input_tokens_seen": 24940736,
      "step": 118180
    },
    {
      "epoch": 13.001650165016502,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.009865909087943537,
      "loss": 0.2293,
      "num_input_tokens_seen": 24941824,
      "step": 118185
    },
    {
      "epoch": 13.002200220022003,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009864556048487443,
      "loss": 0.2324,
      "num_input_tokens_seen": 24942848,
      "step": 118190
    },
    {
      "epoch": 13.002750275027502,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.009863203056362266,
      "loss": 0.2319,
      "num_input_tokens_seen": 24943904,
      "step": 118195
    },
    {
      "epoch": 13.003300330033003,
      "grad_norm": 0.002716064453125,
      "learning_rate": 0.009861850111580486,
      "loss": 0.2308,
      "num_input_tokens_seen": 24944960,
      "step": 118200
    },
    {
      "epoch": 13.003850385038504,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009860497214154563,
      "loss": 0.2314,
      "num_input_tokens_seen": 24946016,
      "step": 118205
    },
    {
      "epoch": 13.004400440044005,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.009859144364096979,
      "loss": 0.2329,
      "num_input_tokens_seen": 24947072,
      "step": 118210
    },
    {
      "epoch": 13.004950495049505,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.009857791561420191,
      "loss": 0.2308,
      "num_input_tokens_seen": 24948096,
      "step": 118215
    },
    {
      "epoch": 13.005500550055006,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.009856438806136665,
      "loss": 0.2303,
      "num_input_tokens_seen": 24949184,
      "step": 118220
    },
    {
      "epoch": 13.006050605060507,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.009855086098258883,
      "loss": 0.2303,
      "num_input_tokens_seen": 24950176,
      "step": 118225
    },
    {
      "epoch": 13.006600660066006,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009853733437799295,
      "loss": 0.2293,
      "num_input_tokens_seen": 24951296,
      "step": 118230
    },
    {
      "epoch": 13.007150715071507,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009852380824770383,
      "loss": 0.2319,
      "num_input_tokens_seen": 24952352,
      "step": 118235
    },
    {
      "epoch": 13.007700770077008,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.009851028259184606,
      "loss": 0.2314,
      "num_input_tokens_seen": 24953408,
      "step": 118240
    },
    {
      "epoch": 13.008250825082508,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.009849675741054427,
      "loss": 0.2313,
      "num_input_tokens_seen": 24954432,
      "step": 118245
    },
    {
      "epoch": 13.008800880088009,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00984832327039232,
      "loss": 0.2288,
      "num_input_tokens_seen": 24955552,
      "step": 118250
    },
    {
      "epoch": 13.00935093509351,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.00984697084721074,
      "loss": 0.2345,
      "num_input_tokens_seen": 24956608,
      "step": 118255
    },
    {
      "epoch": 13.009900990099009,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009845618471522164,
      "loss": 0.2303,
      "num_input_tokens_seen": 24957600,
      "step": 118260
    },
    {
      "epoch": 13.01045104510451,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.009844266143339048,
      "loss": 0.2314,
      "num_input_tokens_seen": 24958624,
      "step": 118265
    },
    {
      "epoch": 13.011001100110011,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.009842913862673853,
      "loss": 0.2319,
      "num_input_tokens_seen": 24959680,
      "step": 118270
    },
    {
      "epoch": 13.011551155115512,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009841561629539046,
      "loss": 0.2329,
      "num_input_tokens_seen": 24960704,
      "step": 118275
    },
    {
      "epoch": 13.012101210121012,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.00984020944394709,
      "loss": 0.234,
      "num_input_tokens_seen": 24961824,
      "step": 118280
    },
    {
      "epoch": 13.012651265126513,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.009838857305910454,
      "loss": 0.2314,
      "num_input_tokens_seen": 24962944,
      "step": 118285
    },
    {
      "epoch": 13.013201320132014,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.00983750521544159,
      "loss": 0.2319,
      "num_input_tokens_seen": 24964000,
      "step": 118290
    },
    {
      "epoch": 13.013751375137513,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00983615317255296,
      "loss": 0.2303,
      "num_input_tokens_seen": 24965056,
      "step": 118295
    },
    {
      "epoch": 13.014301430143014,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009834801177257033,
      "loss": 0.2309,
      "num_input_tokens_seen": 24966144,
      "step": 118300
    },
    {
      "epoch": 13.014851485148515,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00983344922956626,
      "loss": 0.2288,
      "num_input_tokens_seen": 24967168,
      "step": 118305
    },
    {
      "epoch": 13.015401540154015,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.009832097329493113,
      "loss": 0.2319,
      "num_input_tokens_seen": 24968224,
      "step": 118310
    },
    {
      "epoch": 13.015951595159516,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009830745477050042,
      "loss": 0.2309,
      "num_input_tokens_seen": 24969248,
      "step": 118315
    },
    {
      "epoch": 13.016501650165017,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009829393672249505,
      "loss": 0.2319,
      "num_input_tokens_seen": 24970336,
      "step": 118320
    },
    {
      "epoch": 13.017051705170518,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.009828041915103974,
      "loss": 0.2319,
      "num_input_tokens_seen": 24971392,
      "step": 118325
    },
    {
      "epoch": 13.017601760176017,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.009826690205625892,
      "loss": 0.2304,
      "num_input_tokens_seen": 24972512,
      "step": 118330
    },
    {
      "epoch": 13.018151815181518,
      "grad_norm": 0.005859375,
      "learning_rate": 0.009825338543827727,
      "loss": 0.2319,
      "num_input_tokens_seen": 24973600,
      "step": 118335
    },
    {
      "epoch": 13.01870187018702,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.009823986929721936,
      "loss": 0.2308,
      "num_input_tokens_seen": 24974560,
      "step": 118340
    },
    {
      "epoch": 13.019251925192519,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009822635363320965,
      "loss": 0.2293,
      "num_input_tokens_seen": 24975584,
      "step": 118345
    },
    {
      "epoch": 13.01980198019802,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.009821283844637289,
      "loss": 0.2319,
      "num_input_tokens_seen": 24976640,
      "step": 118350
    },
    {
      "epoch": 13.020352035203521,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.00981993237368335,
      "loss": 0.2304,
      "num_input_tokens_seen": 24977664,
      "step": 118355
    },
    {
      "epoch": 13.02090209020902,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.009818580950471608,
      "loss": 0.2314,
      "num_input_tokens_seen": 24978720,
      "step": 118360
    },
    {
      "epoch": 13.021452145214521,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.009817229575014526,
      "loss": 0.2309,
      "num_input_tokens_seen": 24979776,
      "step": 118365
    },
    {
      "epoch": 13.022002200220022,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009815878247324543,
      "loss": 0.2308,
      "num_input_tokens_seen": 24980832,
      "step": 118370
    },
    {
      "epoch": 13.022552255225522,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.009814526967414127,
      "loss": 0.2278,
      "num_input_tokens_seen": 24981856,
      "step": 118375
    },
    {
      "epoch": 13.023102310231023,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.009813175735295731,
      "loss": 0.2314,
      "num_input_tokens_seen": 24982912,
      "step": 118380
    },
    {
      "epoch": 13.023652365236524,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.009811824550981799,
      "loss": 0.2309,
      "num_input_tokens_seen": 24984000,
      "step": 118385
    },
    {
      "epoch": 13.024202420242025,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.009810473414484794,
      "loss": 0.2319,
      "num_input_tokens_seen": 24985088,
      "step": 118390
    },
    {
      "epoch": 13.024752475247524,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.009809122325817161,
      "loss": 0.233,
      "num_input_tokens_seen": 24986080,
      "step": 118395
    },
    {
      "epoch": 13.025302530253025,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009807771284991365,
      "loss": 0.2298,
      "num_input_tokens_seen": 24987168,
      "step": 118400
    },
    {
      "epoch": 13.025852585258527,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.009806420292019846,
      "loss": 0.2324,
      "num_input_tokens_seen": 24988192,
      "step": 118405
    },
    {
      "epoch": 13.026402640264026,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.009805069346915054,
      "loss": 0.2335,
      "num_input_tokens_seen": 24989216,
      "step": 118410
    },
    {
      "epoch": 13.026952695269527,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.00980371844968945,
      "loss": 0.2329,
      "num_input_tokens_seen": 24990272,
      "step": 118415
    },
    {
      "epoch": 13.027502750275028,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.009802367600355475,
      "loss": 0.2324,
      "num_input_tokens_seen": 24991328,
      "step": 118420
    },
    {
      "epoch": 13.028052805280527,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00980101679892559,
      "loss": 0.2303,
      "num_input_tokens_seen": 24992480,
      "step": 118425
    },
    {
      "epoch": 13.028602860286028,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009799666045412238,
      "loss": 0.2329,
      "num_input_tokens_seen": 24993568,
      "step": 118430
    },
    {
      "epoch": 13.02915291529153,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.009798315339827862,
      "loss": 0.2314,
      "num_input_tokens_seen": 24994624,
      "step": 118435
    },
    {
      "epoch": 13.029702970297029,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009796964682184927,
      "loss": 0.2303,
      "num_input_tokens_seen": 24995616,
      "step": 118440
    },
    {
      "epoch": 13.03025302530253,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009795614072495865,
      "loss": 0.2319,
      "num_input_tokens_seen": 24996736,
      "step": 118445
    },
    {
      "epoch": 13.030803080308031,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.009794263510773131,
      "loss": 0.2299,
      "num_input_tokens_seen": 24997792,
      "step": 118450
    },
    {
      "epoch": 13.031353135313532,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.009792912997029179,
      "loss": 0.2314,
      "num_input_tokens_seen": 24998880,
      "step": 118455
    },
    {
      "epoch": 13.031903190319031,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009791562531276441,
      "loss": 0.2319,
      "num_input_tokens_seen": 24999968,
      "step": 118460
    },
    {
      "epoch": 13.032453245324533,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009790212113527376,
      "loss": 0.2308,
      "num_input_tokens_seen": 25000992,
      "step": 118465
    },
    {
      "epoch": 13.033003300330034,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.009788861743794423,
      "loss": 0.2319,
      "num_input_tokens_seen": 25002080,
      "step": 118470
    },
    {
      "epoch": 13.033553355335533,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.009787511422090032,
      "loss": 0.2309,
      "num_input_tokens_seen": 25003072,
      "step": 118475
    },
    {
      "epoch": 13.034103410341034,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.009786161148426653,
      "loss": 0.2324,
      "num_input_tokens_seen": 25004160,
      "step": 118480
    },
    {
      "epoch": 13.034653465346535,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.009784810922816715,
      "loss": 0.2314,
      "num_input_tokens_seen": 25005248,
      "step": 118485
    },
    {
      "epoch": 13.035203520352034,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00978346074527268,
      "loss": 0.2319,
      "num_input_tokens_seen": 25006304,
      "step": 118490
    },
    {
      "epoch": 13.035753575357536,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.009782110615806977,
      "loss": 0.2319,
      "num_input_tokens_seen": 25007360,
      "step": 118495
    },
    {
      "epoch": 13.036303630363037,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009780760534432066,
      "loss": 0.2314,
      "num_input_tokens_seen": 25008416,
      "step": 118500
    },
    {
      "epoch": 13.036853685368538,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009779410501160376,
      "loss": 0.2314,
      "num_input_tokens_seen": 25009440,
      "step": 118505
    },
    {
      "epoch": 13.037403740374037,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.00977806051600435,
      "loss": 0.2325,
      "num_input_tokens_seen": 25010464,
      "step": 118510
    },
    {
      "epoch": 13.037953795379538,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.009776710578976442,
      "loss": 0.2308,
      "num_input_tokens_seen": 25011520,
      "step": 118515
    },
    {
      "epoch": 13.03850385038504,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00977536069008908,
      "loss": 0.2298,
      "num_input_tokens_seen": 25012608,
      "step": 118520
    },
    {
      "epoch": 13.039053905390539,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.009774010849354713,
      "loss": 0.2324,
      "num_input_tokens_seen": 25013664,
      "step": 118525
    },
    {
      "epoch": 13.03960396039604,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009772661056785786,
      "loss": 0.2314,
      "num_input_tokens_seen": 25014720,
      "step": 118530
    },
    {
      "epoch": 13.04015401540154,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009771311312394724,
      "loss": 0.2298,
      "num_input_tokens_seen": 25015808,
      "step": 118535
    },
    {
      "epoch": 13.04070407040704,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.009769961616193985,
      "loss": 0.2314,
      "num_input_tokens_seen": 25016800,
      "step": 118540
    },
    {
      "epoch": 13.041254125412541,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.009768611968195997,
      "loss": 0.2319,
      "num_input_tokens_seen": 25017888,
      "step": 118545
    },
    {
      "epoch": 13.041804180418042,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.0097672623684132,
      "loss": 0.2314,
      "num_input_tokens_seen": 25018912,
      "step": 118550
    },
    {
      "epoch": 13.042354235423542,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.009765912816858041,
      "loss": 0.2324,
      "num_input_tokens_seen": 25020000,
      "step": 118555
    },
    {
      "epoch": 13.042904290429043,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.009764563313542945,
      "loss": 0.2298,
      "num_input_tokens_seen": 25021056,
      "step": 118560
    },
    {
      "epoch": 13.043454345434544,
      "grad_norm": 0.00506591796875,
      "learning_rate": 0.009763213858480361,
      "loss": 0.2324,
      "num_input_tokens_seen": 25022112,
      "step": 118565
    },
    {
      "epoch": 13.044004400440045,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.009761864451682723,
      "loss": 0.233,
      "num_input_tokens_seen": 25023168,
      "step": 118570
    },
    {
      "epoch": 13.044554455445544,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.009760515093162463,
      "loss": 0.2314,
      "num_input_tokens_seen": 25024224,
      "step": 118575
    },
    {
      "epoch": 13.045104510451045,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.009759165782932023,
      "loss": 0.2298,
      "num_input_tokens_seen": 25025280,
      "step": 118580
    },
    {
      "epoch": 13.045654565456546,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.009757816521003835,
      "loss": 0.2314,
      "num_input_tokens_seen": 25026368,
      "step": 118585
    },
    {
      "epoch": 13.046204620462046,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009756467307390338,
      "loss": 0.2293,
      "num_input_tokens_seen": 25027456,
      "step": 118590
    },
    {
      "epoch": 13.046754675467547,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009755118142103971,
      "loss": 0.2298,
      "num_input_tokens_seen": 25028448,
      "step": 118595
    },
    {
      "epoch": 13.047304730473048,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.009753769025157157,
      "loss": 0.2298,
      "num_input_tokens_seen": 25029504,
      "step": 118600
    },
    {
      "epoch": 13.047854785478547,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.009752419956562338,
      "loss": 0.2324,
      "num_input_tokens_seen": 25030592,
      "step": 118605
    },
    {
      "epoch": 13.048404840484048,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.009751070936331943,
      "loss": 0.2324,
      "num_input_tokens_seen": 25031680,
      "step": 118610
    },
    {
      "epoch": 13.04895489548955,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.009749721964478415,
      "loss": 0.2314,
      "num_input_tokens_seen": 25032704,
      "step": 118615
    },
    {
      "epoch": 13.049504950495049,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.00974837304101418,
      "loss": 0.2335,
      "num_input_tokens_seen": 25033792,
      "step": 118620
    },
    {
      "epoch": 13.05005500550055,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.009747024165951666,
      "loss": 0.2335,
      "num_input_tokens_seen": 25034848,
      "step": 118625
    },
    {
      "epoch": 13.05060506050605,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.009745675339303315,
      "loss": 0.2314,
      "num_input_tokens_seen": 25035904,
      "step": 118630
    },
    {
      "epoch": 13.051155115511552,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009744326561081547,
      "loss": 0.2314,
      "num_input_tokens_seen": 25036960,
      "step": 118635
    },
    {
      "epoch": 13.051705170517051,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0097429778312988,
      "loss": 0.2304,
      "num_input_tokens_seen": 25038016,
      "step": 118640
    },
    {
      "epoch": 13.052255225522552,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.00974162914996751,
      "loss": 0.2324,
      "num_input_tokens_seen": 25039040,
      "step": 118645
    },
    {
      "epoch": 13.052805280528053,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00974028051710009,
      "loss": 0.2324,
      "num_input_tokens_seen": 25040096,
      "step": 118650
    },
    {
      "epoch": 13.053355335533553,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009738931932708993,
      "loss": 0.2314,
      "num_input_tokens_seen": 25041056,
      "step": 118655
    },
    {
      "epoch": 13.053905390539054,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.009737583396806627,
      "loss": 0.2314,
      "num_input_tokens_seen": 25042112,
      "step": 118660
    },
    {
      "epoch": 13.054455445544555,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.009736234909405434,
      "loss": 0.2314,
      "num_input_tokens_seen": 25043200,
      "step": 118665
    },
    {
      "epoch": 13.055005500550054,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.009734886470517842,
      "loss": 0.2319,
      "num_input_tokens_seen": 25044256,
      "step": 118670
    },
    {
      "epoch": 13.055555555555555,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.009733538080156265,
      "loss": 0.2319,
      "num_input_tokens_seen": 25045312,
      "step": 118675
    },
    {
      "epoch": 13.056105610561056,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009732189738333145,
      "loss": 0.2314,
      "num_input_tokens_seen": 25046368,
      "step": 118680
    },
    {
      "epoch": 13.056655665566556,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.009730841445060904,
      "loss": 0.2309,
      "num_input_tokens_seen": 25047456,
      "step": 118685
    },
    {
      "epoch": 13.057205720572057,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.009729493200351973,
      "loss": 0.2324,
      "num_input_tokens_seen": 25048544,
      "step": 118690
    },
    {
      "epoch": 13.057755775577558,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.009728145004218772,
      "loss": 0.2319,
      "num_input_tokens_seen": 25049600,
      "step": 118695
    },
    {
      "epoch": 13.058305830583059,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.009726796856673724,
      "loss": 0.2303,
      "num_input_tokens_seen": 25050688,
      "step": 118700
    },
    {
      "epoch": 13.058855885588558,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009725448757729268,
      "loss": 0.2319,
      "num_input_tokens_seen": 25051744,
      "step": 118705
    },
    {
      "epoch": 13.05940594059406,
      "grad_norm": 0.000843048095703125,
      "learning_rate": 0.009724100707397806,
      "loss": 0.2309,
      "num_input_tokens_seen": 25052768,
      "step": 118710
    },
    {
      "epoch": 13.05995599559956,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.009722752705691791,
      "loss": 0.234,
      "num_input_tokens_seen": 25053792,
      "step": 118715
    },
    {
      "epoch": 13.06050605060506,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.009721404752623626,
      "loss": 0.2308,
      "num_input_tokens_seen": 25054880,
      "step": 118720
    },
    {
      "epoch": 13.061056105610561,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.009720056848205738,
      "loss": 0.2298,
      "num_input_tokens_seen": 25056000,
      "step": 118725
    },
    {
      "epoch": 13.061606160616062,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.009718708992450557,
      "loss": 0.2308,
      "num_input_tokens_seen": 25057056,
      "step": 118730
    },
    {
      "epoch": 13.062156215621561,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.0097173611853705,
      "loss": 0.235,
      "num_input_tokens_seen": 25058112,
      "step": 118735
    },
    {
      "epoch": 13.062706270627062,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.009716013426977985,
      "loss": 0.2319,
      "num_input_tokens_seen": 25059136,
      "step": 118740
    },
    {
      "epoch": 13.063256325632564,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009714665717285447,
      "loss": 0.2314,
      "num_input_tokens_seen": 25060224,
      "step": 118745
    },
    {
      "epoch": 13.063806380638065,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009713318056305292,
      "loss": 0.2309,
      "num_input_tokens_seen": 25061184,
      "step": 118750
    },
    {
      "epoch": 13.064356435643564,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.00971197044404995,
      "loss": 0.2314,
      "num_input_tokens_seen": 25062176,
      "step": 118755
    },
    {
      "epoch": 13.064906490649065,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009710622880531842,
      "loss": 0.2324,
      "num_input_tokens_seen": 25063168,
      "step": 118760
    },
    {
      "epoch": 13.065456545654566,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.009709275365763381,
      "loss": 0.2324,
      "num_input_tokens_seen": 25064224,
      "step": 118765
    },
    {
      "epoch": 13.066006600660065,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.009707927899756991,
      "loss": 0.2324,
      "num_input_tokens_seen": 25065312,
      "step": 118770
    },
    {
      "epoch": 13.066556655665567,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009706580482525087,
      "loss": 0.2304,
      "num_input_tokens_seen": 25066400,
      "step": 118775
    },
    {
      "epoch": 13.067106710671068,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.009705233114080095,
      "loss": 0.2303,
      "num_input_tokens_seen": 25067456,
      "step": 118780
    },
    {
      "epoch": 13.067656765676567,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00970388579443443,
      "loss": 0.2314,
      "num_input_tokens_seen": 25068544,
      "step": 118785
    },
    {
      "epoch": 13.068206820682068,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009702538523600503,
      "loss": 0.2298,
      "num_input_tokens_seen": 25069568,
      "step": 118790
    },
    {
      "epoch": 13.06875687568757,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.009701191301590742,
      "loss": 0.2298,
      "num_input_tokens_seen": 25070560,
      "step": 118795
    },
    {
      "epoch": 13.069306930693068,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.009699844128417552,
      "loss": 0.2309,
      "num_input_tokens_seen": 25071648,
      "step": 118800
    },
    {
      "epoch": 13.06985698569857,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.009698497004093363,
      "loss": 0.2314,
      "num_input_tokens_seen": 25072704,
      "step": 118805
    },
    {
      "epoch": 13.07040704070407,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.009697149928630579,
      "loss": 0.2319,
      "num_input_tokens_seen": 25073696,
      "step": 118810
    },
    {
      "epoch": 13.070957095709572,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009695802902041617,
      "loss": 0.2319,
      "num_input_tokens_seen": 25074816,
      "step": 118815
    },
    {
      "epoch": 13.071507150715071,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.009694455924338902,
      "loss": 0.2319,
      "num_input_tokens_seen": 25075904,
      "step": 118820
    },
    {
      "epoch": 13.072057205720572,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009693108995534832,
      "loss": 0.2314,
      "num_input_tokens_seen": 25076992,
      "step": 118825
    },
    {
      "epoch": 13.072607260726073,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.009691762115641836,
      "loss": 0.2319,
      "num_input_tokens_seen": 25078080,
      "step": 118830
    },
    {
      "epoch": 13.073157315731573,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.00969041528467232,
      "loss": 0.2314,
      "num_input_tokens_seen": 25079136,
      "step": 118835
    },
    {
      "epoch": 13.073707370737074,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.009689068502638695,
      "loss": 0.2314,
      "num_input_tokens_seen": 25080128,
      "step": 118840
    },
    {
      "epoch": 13.074257425742575,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009687721769553383,
      "loss": 0.2319,
      "num_input_tokens_seen": 25081184,
      "step": 118845
    },
    {
      "epoch": 13.074807480748074,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009686375085428784,
      "loss": 0.233,
      "num_input_tokens_seen": 25082272,
      "step": 118850
    },
    {
      "epoch": 13.075357535753575,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009685028450277322,
      "loss": 0.2314,
      "num_input_tokens_seen": 25083296,
      "step": 118855
    },
    {
      "epoch": 13.075907590759076,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.009683681864111403,
      "loss": 0.2298,
      "num_input_tokens_seen": 25084288,
      "step": 118860
    },
    {
      "epoch": 13.076457645764576,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009682335326943431,
      "loss": 0.2324,
      "num_input_tokens_seen": 25085344,
      "step": 118865
    },
    {
      "epoch": 13.077007700770077,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009680988838785827,
      "loss": 0.2308,
      "num_input_tokens_seen": 25086400,
      "step": 118870
    },
    {
      "epoch": 13.077557755775578,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009679642399650993,
      "loss": 0.2303,
      "num_input_tokens_seen": 25087488,
      "step": 118875
    },
    {
      "epoch": 13.078107810781079,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009678296009551349,
      "loss": 0.2314,
      "num_input_tokens_seen": 25088512,
      "step": 118880
    },
    {
      "epoch": 13.078657865786578,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.009676949668499293,
      "loss": 0.2324,
      "num_input_tokens_seen": 25089536,
      "step": 118885
    },
    {
      "epoch": 13.07920792079208,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.009675603376507235,
      "loss": 0.2308,
      "num_input_tokens_seen": 25090528,
      "step": 118890
    },
    {
      "epoch": 13.07975797579758,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.00967425713358759,
      "loss": 0.2308,
      "num_input_tokens_seen": 25091616,
      "step": 118895
    },
    {
      "epoch": 13.08030803080308,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.009672910939752755,
      "loss": 0.2319,
      "num_input_tokens_seen": 25092640,
      "step": 118900
    },
    {
      "epoch": 13.08085808580858,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00967156479501515,
      "loss": 0.2324,
      "num_input_tokens_seen": 25093760,
      "step": 118905
    },
    {
      "epoch": 13.081408140814082,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009670218699387176,
      "loss": 0.234,
      "num_input_tokens_seen": 25094816,
      "step": 118910
    },
    {
      "epoch": 13.081958195819581,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.009668872652881234,
      "loss": 0.2303,
      "num_input_tokens_seen": 25095776,
      "step": 118915
    },
    {
      "epoch": 13.082508250825082,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.00966752665550974,
      "loss": 0.2308,
      "num_input_tokens_seen": 25096832,
      "step": 118920
    },
    {
      "epoch": 13.083058305830583,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009666180707285092,
      "loss": 0.2303,
      "num_input_tokens_seen": 25097888,
      "step": 118925
    },
    {
      "epoch": 13.083608360836084,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.009664834808219693,
      "loss": 0.2314,
      "num_input_tokens_seen": 25098880,
      "step": 118930
    },
    {
      "epoch": 13.084158415841584,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.00966348895832596,
      "loss": 0.2298,
      "num_input_tokens_seen": 25099904,
      "step": 118935
    },
    {
      "epoch": 13.084708470847085,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00966214315761628,
      "loss": 0.2309,
      "num_input_tokens_seen": 25100992,
      "step": 118940
    },
    {
      "epoch": 13.085258525852586,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.009660797406103068,
      "loss": 0.233,
      "num_input_tokens_seen": 25102016,
      "step": 118945
    },
    {
      "epoch": 13.085808580858085,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009659451703798729,
      "loss": 0.2308,
      "num_input_tokens_seen": 25103072,
      "step": 118950
    },
    {
      "epoch": 13.086358635863586,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009658106050715652,
      "loss": 0.2303,
      "num_input_tokens_seen": 25104160,
      "step": 118955
    },
    {
      "epoch": 13.086908690869087,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.009656760446866256,
      "loss": 0.2309,
      "num_input_tokens_seen": 25105216,
      "step": 118960
    },
    {
      "epoch": 13.087458745874587,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009655414892262928,
      "loss": 0.2309,
      "num_input_tokens_seen": 25106272,
      "step": 118965
    },
    {
      "epoch": 13.088008800880088,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.00965406938691808,
      "loss": 0.2314,
      "num_input_tokens_seen": 25107328,
      "step": 118970
    },
    {
      "epoch": 13.088558855885589,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.009652723930844114,
      "loss": 0.2303,
      "num_input_tokens_seen": 25108352,
      "step": 118975
    },
    {
      "epoch": 13.089108910891088,
      "grad_norm": 0.001953125,
      "learning_rate": 0.009651378524053417,
      "loss": 0.2308,
      "num_input_tokens_seen": 25109376,
      "step": 118980
    },
    {
      "epoch": 13.08965896589659,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.009650033166558403,
      "loss": 0.2314,
      "num_input_tokens_seen": 25110432,
      "step": 118985
    },
    {
      "epoch": 13.09020902090209,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.009648687858371462,
      "loss": 0.2319,
      "num_input_tokens_seen": 25111520,
      "step": 118990
    },
    {
      "epoch": 13.090759075907592,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.009647342599505004,
      "loss": 0.2309,
      "num_input_tokens_seen": 25112544,
      "step": 118995
    },
    {
      "epoch": 13.091309130913091,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.009645997389971418,
      "loss": 0.2309,
      "num_input_tokens_seen": 25113664,
      "step": 119000
    },
    {
      "epoch": 13.091859185918592,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009644652229783102,
      "loss": 0.2324,
      "num_input_tokens_seen": 25114688,
      "step": 119005
    },
    {
      "epoch": 13.092409240924093,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.009643307118952461,
      "loss": 0.2309,
      "num_input_tokens_seen": 25115680,
      "step": 119010
    },
    {
      "epoch": 13.092959295929592,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.00964196205749188,
      "loss": 0.2319,
      "num_input_tokens_seen": 25116736,
      "step": 119015
    },
    {
      "epoch": 13.093509350935093,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009640617045413772,
      "loss": 0.2319,
      "num_input_tokens_seen": 25117760,
      "step": 119020
    },
    {
      "epoch": 13.094059405940595,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.009639272082730523,
      "loss": 0.2314,
      "num_input_tokens_seen": 25118816,
      "step": 119025
    },
    {
      "epoch": 13.094609460946094,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009637927169454528,
      "loss": 0.2309,
      "num_input_tokens_seen": 25119872,
      "step": 119030
    },
    {
      "epoch": 13.095159515951595,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.00963658230559819,
      "loss": 0.2303,
      "num_input_tokens_seen": 25120864,
      "step": 119035
    },
    {
      "epoch": 13.095709570957096,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009635237491173895,
      "loss": 0.2329,
      "num_input_tokens_seen": 25121856,
      "step": 119040
    },
    {
      "epoch": 13.096259625962595,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009633892726194042,
      "loss": 0.2314,
      "num_input_tokens_seen": 25122912,
      "step": 119045
    },
    {
      "epoch": 13.096809680968097,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009632548010671031,
      "loss": 0.2308,
      "num_input_tokens_seen": 25123904,
      "step": 119050
    },
    {
      "epoch": 13.097359735973598,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.009631203344617242,
      "loss": 0.2335,
      "num_input_tokens_seen": 25125056,
      "step": 119055
    },
    {
      "epoch": 13.097909790979099,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.009629858728045079,
      "loss": 0.2293,
      "num_input_tokens_seen": 25126112,
      "step": 119060
    },
    {
      "epoch": 13.098459845984598,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.009628514160966927,
      "loss": 0.2324,
      "num_input_tokens_seen": 25127136,
      "step": 119065
    },
    {
      "epoch": 13.099009900990099,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00962716964339519,
      "loss": 0.2304,
      "num_input_tokens_seen": 25128224,
      "step": 119070
    },
    {
      "epoch": 13.0995599559956,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.009625825175342249,
      "loss": 0.2313,
      "num_input_tokens_seen": 25129248,
      "step": 119075
    },
    {
      "epoch": 13.1001100110011,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009624480756820496,
      "loss": 0.2324,
      "num_input_tokens_seen": 25130304,
      "step": 119080
    },
    {
      "epoch": 13.1006600660066,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.009623136387842325,
      "loss": 0.2303,
      "num_input_tokens_seen": 25131360,
      "step": 119085
    },
    {
      "epoch": 13.101210121012102,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009621792068420124,
      "loss": 0.2319,
      "num_input_tokens_seen": 25132416,
      "step": 119090
    },
    {
      "epoch": 13.101760176017601,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.009620447798566294,
      "loss": 0.2298,
      "num_input_tokens_seen": 25133536,
      "step": 119095
    },
    {
      "epoch": 13.102310231023102,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.009619103578293209,
      "loss": 0.2314,
      "num_input_tokens_seen": 25134560,
      "step": 119100
    },
    {
      "epoch": 13.102860286028603,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.009617759407613263,
      "loss": 0.2324,
      "num_input_tokens_seen": 25135584,
      "step": 119105
    },
    {
      "epoch": 13.103410341034103,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00961641528653885,
      "loss": 0.2303,
      "num_input_tokens_seen": 25136640,
      "step": 119110
    },
    {
      "epoch": 13.103960396039604,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00961507121508235,
      "loss": 0.2314,
      "num_input_tokens_seen": 25137696,
      "step": 119115
    },
    {
      "epoch": 13.104510451045105,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.009613727193256155,
      "loss": 0.233,
      "num_input_tokens_seen": 25138816,
      "step": 119120
    },
    {
      "epoch": 13.105060506050606,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009612383221072657,
      "loss": 0.2293,
      "num_input_tokens_seen": 25139808,
      "step": 119125
    },
    {
      "epoch": 13.105610561056105,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.009611039298544228,
      "loss": 0.2319,
      "num_input_tokens_seen": 25140864,
      "step": 119130
    },
    {
      "epoch": 13.106160616061606,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009609695425683273,
      "loss": 0.2293,
      "num_input_tokens_seen": 25141888,
      "step": 119135
    },
    {
      "epoch": 13.106710671067107,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009608351602502164,
      "loss": 0.2324,
      "num_input_tokens_seen": 25142912,
      "step": 119140
    },
    {
      "epoch": 13.107260726072607,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009607007829013291,
      "loss": 0.2293,
      "num_input_tokens_seen": 25143872,
      "step": 119145
    },
    {
      "epoch": 13.107810781078108,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.009605664105229044,
      "loss": 0.2314,
      "num_input_tokens_seen": 25144928,
      "step": 119150
    },
    {
      "epoch": 13.108360836083609,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.009604320431161795,
      "loss": 0.2319,
      "num_input_tokens_seen": 25145984,
      "step": 119155
    },
    {
      "epoch": 13.108910891089108,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.009602976806823942,
      "loss": 0.2319,
      "num_input_tokens_seen": 25147040,
      "step": 119160
    },
    {
      "epoch": 13.10946094609461,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.009601633232227862,
      "loss": 0.2298,
      "num_input_tokens_seen": 25148096,
      "step": 119165
    },
    {
      "epoch": 13.11001100110011,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009600289707385933,
      "loss": 0.2314,
      "num_input_tokens_seen": 25149152,
      "step": 119170
    },
    {
      "epoch": 13.110561056105611,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.009598946232310545,
      "loss": 0.2308,
      "num_input_tokens_seen": 25150208,
      "step": 119175
    },
    {
      "epoch": 13.11111111111111,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.009597602807014076,
      "loss": 0.2309,
      "num_input_tokens_seen": 25151296,
      "step": 119180
    },
    {
      "epoch": 13.111661166116612,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.009596259431508915,
      "loss": 0.2324,
      "num_input_tokens_seen": 25152352,
      "step": 119185
    },
    {
      "epoch": 13.112211221122113,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.009594916105807436,
      "loss": 0.2288,
      "num_input_tokens_seen": 25153344,
      "step": 119190
    },
    {
      "epoch": 13.112761276127612,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.009593572829922015,
      "loss": 0.2319,
      "num_input_tokens_seen": 25154464,
      "step": 119195
    },
    {
      "epoch": 13.113311331133113,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.009592229603865047,
      "loss": 0.2329,
      "num_input_tokens_seen": 25155520,
      "step": 119200
    },
    {
      "epoch": 13.113861386138614,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.009590886427648897,
      "loss": 0.2309,
      "num_input_tokens_seen": 25156576,
      "step": 119205
    },
    {
      "epoch": 13.114411441144114,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.009589543301285959,
      "loss": 0.2324,
      "num_input_tokens_seen": 25157664,
      "step": 119210
    },
    {
      "epoch": 13.114961496149615,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.009588200224788603,
      "loss": 0.2303,
      "num_input_tokens_seen": 25158688,
      "step": 119215
    },
    {
      "epoch": 13.115511551155116,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.009586857198169202,
      "loss": 0.2319,
      "num_input_tokens_seen": 25159744,
      "step": 119220
    },
    {
      "epoch": 13.116061606160615,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.009585514221440149,
      "loss": 0.2319,
      "num_input_tokens_seen": 25160896,
      "step": 119225
    },
    {
      "epoch": 13.116611661166116,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009584171294613806,
      "loss": 0.2319,
      "num_input_tokens_seen": 25162016,
      "step": 119230
    },
    {
      "epoch": 13.117161716171617,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00958282841770256,
      "loss": 0.234,
      "num_input_tokens_seen": 25163104,
      "step": 119235
    },
    {
      "epoch": 13.117711771177119,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00958148559071879,
      "loss": 0.2325,
      "num_input_tokens_seen": 25164160,
      "step": 119240
    },
    {
      "epoch": 13.118261826182618,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.00958014281367486,
      "loss": 0.2303,
      "num_input_tokens_seen": 25165184,
      "step": 119245
    },
    {
      "epoch": 13.118811881188119,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.009578800086583156,
      "loss": 0.2314,
      "num_input_tokens_seen": 25166272,
      "step": 119250
    },
    {
      "epoch": 13.11936193619362,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.009577457409456048,
      "loss": 0.2324,
      "num_input_tokens_seen": 25167360,
      "step": 119255
    },
    {
      "epoch": 13.11991199119912,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009576114782305916,
      "loss": 0.2309,
      "num_input_tokens_seen": 25168384,
      "step": 119260
    },
    {
      "epoch": 13.12046204620462,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009574772205145136,
      "loss": 0.2314,
      "num_input_tokens_seen": 25169472,
      "step": 119265
    },
    {
      "epoch": 13.121012101210122,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.009573429677986068,
      "loss": 0.2324,
      "num_input_tokens_seen": 25170464,
      "step": 119270
    },
    {
      "epoch": 13.12156215621562,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.009572087200841099,
      "loss": 0.2319,
      "num_input_tokens_seen": 25171520,
      "step": 119275
    },
    {
      "epoch": 13.122112211221122,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.009570744773722595,
      "loss": 0.2324,
      "num_input_tokens_seen": 25172640,
      "step": 119280
    },
    {
      "epoch": 13.122662266226623,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.009569402396642937,
      "loss": 0.2303,
      "num_input_tokens_seen": 25173728,
      "step": 119285
    },
    {
      "epoch": 13.123212321232122,
      "grad_norm": 0.0004634857177734375,
      "learning_rate": 0.009568060069614488,
      "loss": 0.2309,
      "num_input_tokens_seen": 25174752,
      "step": 119290
    },
    {
      "epoch": 13.123762376237623,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.00956671779264962,
      "loss": 0.2309,
      "num_input_tokens_seen": 25175776,
      "step": 119295
    },
    {
      "epoch": 13.124312431243125,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.009565375565760714,
      "loss": 0.2319,
      "num_input_tokens_seen": 25176832,
      "step": 119300
    },
    {
      "epoch": 13.124862486248626,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009564033388960126,
      "loss": 0.2314,
      "num_input_tokens_seen": 25177920,
      "step": 119305
    },
    {
      "epoch": 13.125412541254125,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.009562691262260238,
      "loss": 0.2319,
      "num_input_tokens_seen": 25178976,
      "step": 119310
    },
    {
      "epoch": 13.125962596259626,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009561349185673418,
      "loss": 0.2314,
      "num_input_tokens_seen": 25180064,
      "step": 119315
    },
    {
      "epoch": 13.126512651265127,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009560007159212023,
      "loss": 0.2314,
      "num_input_tokens_seen": 25181120,
      "step": 119320
    },
    {
      "epoch": 13.127062706270626,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.009558665182888443,
      "loss": 0.2314,
      "num_input_tokens_seen": 25182112,
      "step": 119325
    },
    {
      "epoch": 13.127612761276128,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00955732325671503,
      "loss": 0.235,
      "num_input_tokens_seen": 25183200,
      "step": 119330
    },
    {
      "epoch": 13.128162816281629,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009555981380704154,
      "loss": 0.2319,
      "num_input_tokens_seen": 25184224,
      "step": 119335
    },
    {
      "epoch": 13.128712871287128,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.009554639554868192,
      "loss": 0.2324,
      "num_input_tokens_seen": 25185248,
      "step": 119340
    },
    {
      "epoch": 13.129262926292629,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.009553297779219498,
      "loss": 0.2324,
      "num_input_tokens_seen": 25186272,
      "step": 119345
    },
    {
      "epoch": 13.12981298129813,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.009551956053770448,
      "loss": 0.2334,
      "num_input_tokens_seen": 25187264,
      "step": 119350
    },
    {
      "epoch": 13.130363036303631,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.009550614378533408,
      "loss": 0.2298,
      "num_input_tokens_seen": 25188352,
      "step": 119355
    },
    {
      "epoch": 13.13091309130913,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009549272753520733,
      "loss": 0.2319,
      "num_input_tokens_seen": 25189440,
      "step": 119360
    },
    {
      "epoch": 13.131463146314632,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.009547931178744801,
      "loss": 0.2303,
      "num_input_tokens_seen": 25190496,
      "step": 119365
    },
    {
      "epoch": 13.132013201320133,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009546589654217967,
      "loss": 0.2293,
      "num_input_tokens_seen": 25191552,
      "step": 119370
    },
    {
      "epoch": 13.132563256325632,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.0095452481799526,
      "loss": 0.2303,
      "num_input_tokens_seen": 25192640,
      "step": 119375
    },
    {
      "epoch": 13.133113311331133,
      "grad_norm": 0.000782012939453125,
      "learning_rate": 0.009543906755961071,
      "loss": 0.2314,
      "num_input_tokens_seen": 25193664,
      "step": 119380
    },
    {
      "epoch": 13.133663366336634,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.009542565382255726,
      "loss": 0.2298,
      "num_input_tokens_seen": 25194688,
      "step": 119385
    },
    {
      "epoch": 13.134213421342134,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.00954122405884894,
      "loss": 0.2314,
      "num_input_tokens_seen": 25195712,
      "step": 119390
    },
    {
      "epoch": 13.134763476347635,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.00953988278575307,
      "loss": 0.2298,
      "num_input_tokens_seen": 25196832,
      "step": 119395
    },
    {
      "epoch": 13.135313531353136,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.009538541562980488,
      "loss": 0.2303,
      "num_input_tokens_seen": 25197952,
      "step": 119400
    },
    {
      "epoch": 13.135863586358635,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009537200390543543,
      "loss": 0.2309,
      "num_input_tokens_seen": 25198912,
      "step": 119405
    },
    {
      "epoch": 13.136413641364136,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.009535859268454598,
      "loss": 0.2293,
      "num_input_tokens_seen": 25200032,
      "step": 119410
    },
    {
      "epoch": 13.136963696369637,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009534518196726021,
      "loss": 0.2319,
      "num_input_tokens_seen": 25201056,
      "step": 119415
    },
    {
      "epoch": 13.137513751375138,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.009533177175370165,
      "loss": 0.2304,
      "num_input_tokens_seen": 25202112,
      "step": 119420
    },
    {
      "epoch": 13.138063806380638,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.00953183620439939,
      "loss": 0.2329,
      "num_input_tokens_seen": 25203200,
      "step": 119425
    },
    {
      "epoch": 13.138613861386139,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.009530495283826062,
      "loss": 0.2319,
      "num_input_tokens_seen": 25204224,
      "step": 119430
    },
    {
      "epoch": 13.13916391639164,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.009529154413662526,
      "loss": 0.2309,
      "num_input_tokens_seen": 25205280,
      "step": 119435
    },
    {
      "epoch": 13.13971397139714,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.009527813593921157,
      "loss": 0.2298,
      "num_input_tokens_seen": 25206336,
      "step": 119440
    },
    {
      "epoch": 13.14026402640264,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009526472824614298,
      "loss": 0.2329,
      "num_input_tokens_seen": 25207360,
      "step": 119445
    },
    {
      "epoch": 13.140814081408141,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009525132105754315,
      "loss": 0.2319,
      "num_input_tokens_seen": 25208352,
      "step": 119450
    },
    {
      "epoch": 13.14136413641364,
      "grad_norm": 0.00072479248046875,
      "learning_rate": 0.009523791437353567,
      "loss": 0.2329,
      "num_input_tokens_seen": 25209312,
      "step": 119455
    },
    {
      "epoch": 13.141914191419142,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.009522450819424398,
      "loss": 0.2309,
      "num_input_tokens_seen": 25210336,
      "step": 119460
    },
    {
      "epoch": 13.142464246424643,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.009521110251979175,
      "loss": 0.234,
      "num_input_tokens_seen": 25211488,
      "step": 119465
    },
    {
      "epoch": 13.143014301430142,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009519769735030248,
      "loss": 0.2319,
      "num_input_tokens_seen": 25212576,
      "step": 119470
    },
    {
      "epoch": 13.143564356435643,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.009518429268589979,
      "loss": 0.2288,
      "num_input_tokens_seen": 25213664,
      "step": 119475
    },
    {
      "epoch": 13.144114411441144,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.009517088852670712,
      "loss": 0.2314,
      "num_input_tokens_seen": 25214720,
      "step": 119480
    },
    {
      "epoch": 13.144664466446645,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.009515748487284805,
      "loss": 0.2314,
      "num_input_tokens_seen": 25215744,
      "step": 119485
    },
    {
      "epoch": 13.145214521452145,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009514408172444618,
      "loss": 0.233,
      "num_input_tokens_seen": 25216800,
      "step": 119490
    },
    {
      "epoch": 13.145764576457646,
      "grad_norm": 0.0006866455078125,
      "learning_rate": 0.009513067908162491,
      "loss": 0.234,
      "num_input_tokens_seen": 25217824,
      "step": 119495
    },
    {
      "epoch": 13.146314631463147,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.009511727694450791,
      "loss": 0.2308,
      "num_input_tokens_seen": 25218912,
      "step": 119500
    },
    {
      "epoch": 13.146864686468646,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00951038753132186,
      "loss": 0.2319,
      "num_input_tokens_seen": 25220000,
      "step": 119505
    },
    {
      "epoch": 13.147414741474147,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.009509047418788054,
      "loss": 0.2314,
      "num_input_tokens_seen": 25220992,
      "step": 119510
    },
    {
      "epoch": 13.147964796479648,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.009507707356861727,
      "loss": 0.2309,
      "num_input_tokens_seen": 25222080,
      "step": 119515
    },
    {
      "epoch": 13.148514851485148,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009506367345555222,
      "loss": 0.2309,
      "num_input_tokens_seen": 25223168,
      "step": 119520
    },
    {
      "epoch": 13.149064906490649,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.00950502738488089,
      "loss": 0.2314,
      "num_input_tokens_seen": 25224288,
      "step": 119525
    },
    {
      "epoch": 13.14961496149615,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.009503687474851091,
      "loss": 0.234,
      "num_input_tokens_seen": 25225312,
      "step": 119530
    },
    {
      "epoch": 13.150165016501651,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.00950234761547816,
      "loss": 0.2303,
      "num_input_tokens_seen": 25226304,
      "step": 119535
    },
    {
      "epoch": 13.15071507150715,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009501007806774457,
      "loss": 0.2319,
      "num_input_tokens_seen": 25227328,
      "step": 119540
    },
    {
      "epoch": 13.151265126512651,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009499668048752328,
      "loss": 0.2303,
      "num_input_tokens_seen": 25228352,
      "step": 119545
    },
    {
      "epoch": 13.151815181518153,
      "grad_norm": 0.005859375,
      "learning_rate": 0.009498328341424113,
      "loss": 0.2319,
      "num_input_tokens_seen": 25229472,
      "step": 119550
    },
    {
      "epoch": 13.152365236523652,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009496988684802171,
      "loss": 0.2293,
      "num_input_tokens_seen": 25230528,
      "step": 119555
    },
    {
      "epoch": 13.152915291529153,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009495649078898839,
      "loss": 0.2329,
      "num_input_tokens_seen": 25231680,
      "step": 119560
    },
    {
      "epoch": 13.153465346534654,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.009494309523726473,
      "loss": 0.2298,
      "num_input_tokens_seen": 25232736,
      "step": 119565
    },
    {
      "epoch": 13.154015401540153,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.009492970019297416,
      "loss": 0.2314,
      "num_input_tokens_seen": 25233856,
      "step": 119570
    },
    {
      "epoch": 13.154565456545654,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.009491630565624005,
      "loss": 0.2314,
      "num_input_tokens_seen": 25234912,
      "step": 119575
    },
    {
      "epoch": 13.155115511551156,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.009490291162718595,
      "loss": 0.2314,
      "num_input_tokens_seen": 25235968,
      "step": 119580
    },
    {
      "epoch": 13.155665566556655,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009488951810593525,
      "loss": 0.2324,
      "num_input_tokens_seen": 25236992,
      "step": 119585
    },
    {
      "epoch": 13.156215621562156,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.009487612509261148,
      "loss": 0.2288,
      "num_input_tokens_seen": 25238144,
      "step": 119590
    },
    {
      "epoch": 13.156765676567657,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009486273258733798,
      "loss": 0.2324,
      "num_input_tokens_seen": 25239200,
      "step": 119595
    },
    {
      "epoch": 13.157315731573158,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.00948493405902382,
      "loss": 0.2304,
      "num_input_tokens_seen": 25240256,
      "step": 119600
    },
    {
      "epoch": 13.157865786578657,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.009483594910143563,
      "loss": 0.2314,
      "num_input_tokens_seen": 25241280,
      "step": 119605
    },
    {
      "epoch": 13.158415841584159,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00948225581210536,
      "loss": 0.2308,
      "num_input_tokens_seen": 25242368,
      "step": 119610
    },
    {
      "epoch": 13.15896589658966,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.009480916764921559,
      "loss": 0.2319,
      "num_input_tokens_seen": 25243424,
      "step": 119615
    },
    {
      "epoch": 13.159515951595159,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0094795777686045,
      "loss": 0.2335,
      "num_input_tokens_seen": 25244480,
      "step": 119620
    },
    {
      "epoch": 13.16006600660066,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.009478238823166521,
      "loss": 0.2319,
      "num_input_tokens_seen": 25245504,
      "step": 119625
    },
    {
      "epoch": 13.160616061606161,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009476899928619973,
      "loss": 0.2309,
      "num_input_tokens_seen": 25246528,
      "step": 119630
    },
    {
      "epoch": 13.16116611661166,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009475561084977178,
      "loss": 0.2298,
      "num_input_tokens_seen": 25247648,
      "step": 119635
    },
    {
      "epoch": 13.161716171617162,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.009474222292250492,
      "loss": 0.2303,
      "num_input_tokens_seen": 25248672,
      "step": 119640
    },
    {
      "epoch": 13.162266226622663,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00947288355045225,
      "loss": 0.2329,
      "num_input_tokens_seen": 25249728,
      "step": 119645
    },
    {
      "epoch": 13.162816281628162,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009471544859594781,
      "loss": 0.2293,
      "num_input_tokens_seen": 25250720,
      "step": 119650
    },
    {
      "epoch": 13.163366336633663,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.009470206219690433,
      "loss": 0.2314,
      "num_input_tokens_seen": 25251744,
      "step": 119655
    },
    {
      "epoch": 13.163916391639164,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00946886763075154,
      "loss": 0.2319,
      "num_input_tokens_seen": 25252800,
      "step": 119660
    },
    {
      "epoch": 13.164466446644665,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.009467529092790444,
      "loss": 0.2319,
      "num_input_tokens_seen": 25253856,
      "step": 119665
    },
    {
      "epoch": 13.165016501650165,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009466190605819477,
      "loss": 0.2319,
      "num_input_tokens_seen": 25254880,
      "step": 119670
    },
    {
      "epoch": 13.165566556655666,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009464852169850971,
      "loss": 0.2303,
      "num_input_tokens_seen": 25255904,
      "step": 119675
    },
    {
      "epoch": 13.166116611661167,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.00946351378489727,
      "loss": 0.2324,
      "num_input_tokens_seen": 25256928,
      "step": 119680
    },
    {
      "epoch": 13.166666666666666,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.009462175450970702,
      "loss": 0.2314,
      "num_input_tokens_seen": 25258016,
      "step": 119685
    },
    {
      "epoch": 13.167216721672167,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009460837168083611,
      "loss": 0.2324,
      "num_input_tokens_seen": 25259136,
      "step": 119690
    },
    {
      "epoch": 13.167766776677668,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.009459498936248324,
      "loss": 0.2303,
      "num_input_tokens_seen": 25260160,
      "step": 119695
    },
    {
      "epoch": 13.168316831683168,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009458160755477174,
      "loss": 0.2298,
      "num_input_tokens_seen": 25261248,
      "step": 119700
    },
    {
      "epoch": 13.168866886688669,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.009456822625782504,
      "loss": 0.2308,
      "num_input_tokens_seen": 25262240,
      "step": 119705
    },
    {
      "epoch": 13.16941694169417,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009455484547176636,
      "loss": 0.2324,
      "num_input_tokens_seen": 25263328,
      "step": 119710
    },
    {
      "epoch": 13.16996699669967,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.009454146519671903,
      "loss": 0.2319,
      "num_input_tokens_seen": 25264352,
      "step": 119715
    },
    {
      "epoch": 13.17051705170517,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009452808543280646,
      "loss": 0.2308,
      "num_input_tokens_seen": 25265312,
      "step": 119720
    },
    {
      "epoch": 13.171067106710671,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.009451470618015184,
      "loss": 0.2319,
      "num_input_tokens_seen": 25266336,
      "step": 119725
    },
    {
      "epoch": 13.171617161716172,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00945013274388786,
      "loss": 0.2314,
      "num_input_tokens_seen": 25267360,
      "step": 119730
    },
    {
      "epoch": 13.172167216721672,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009448794920911,
      "loss": 0.2298,
      "num_input_tokens_seen": 25268416,
      "step": 119735
    },
    {
      "epoch": 13.172717271727173,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009447457149096927,
      "loss": 0.2308,
      "num_input_tokens_seen": 25269440,
      "step": 119740
    },
    {
      "epoch": 13.173267326732674,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.009446119428457985,
      "loss": 0.2303,
      "num_input_tokens_seen": 25270496,
      "step": 119745
    },
    {
      "epoch": 13.173817381738173,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.00944478175900649,
      "loss": 0.2324,
      "num_input_tokens_seen": 25271520,
      "step": 119750
    },
    {
      "epoch": 13.174367436743674,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.00944344414075478,
      "loss": 0.2314,
      "num_input_tokens_seen": 25272544,
      "step": 119755
    },
    {
      "epoch": 13.174917491749175,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.009442106573715178,
      "loss": 0.2299,
      "num_input_tokens_seen": 25273664,
      "step": 119760
    },
    {
      "epoch": 13.175467546754675,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00944076905790001,
      "loss": 0.2324,
      "num_input_tokens_seen": 25274784,
      "step": 119765
    },
    {
      "epoch": 13.176017601760176,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.009439431593321607,
      "loss": 0.2298,
      "num_input_tokens_seen": 25275840,
      "step": 119770
    },
    {
      "epoch": 13.176567656765677,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009438094179992295,
      "loss": 0.2314,
      "num_input_tokens_seen": 25276928,
      "step": 119775
    },
    {
      "epoch": 13.177117711771178,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.009436756817924403,
      "loss": 0.2324,
      "num_input_tokens_seen": 25277920,
      "step": 119780
    },
    {
      "epoch": 13.177667766776677,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009435419507130251,
      "loss": 0.2298,
      "num_input_tokens_seen": 25278976,
      "step": 119785
    },
    {
      "epoch": 13.178217821782178,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.009434082247622166,
      "loss": 0.2313,
      "num_input_tokens_seen": 25280032,
      "step": 119790
    },
    {
      "epoch": 13.17876787678768,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00943274503941248,
      "loss": 0.2319,
      "num_input_tokens_seen": 25281056,
      "step": 119795
    },
    {
      "epoch": 13.179317931793179,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.0094314078825135,
      "loss": 0.2319,
      "num_input_tokens_seen": 25282048,
      "step": 119800
    },
    {
      "epoch": 13.17986798679868,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.009430070776937574,
      "loss": 0.2298,
      "num_input_tokens_seen": 25283072,
      "step": 119805
    },
    {
      "epoch": 13.180418041804181,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009428733722697006,
      "loss": 0.2313,
      "num_input_tokens_seen": 25284096,
      "step": 119810
    },
    {
      "epoch": 13.18096809680968,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.009427396719804127,
      "loss": 0.2309,
      "num_input_tokens_seen": 25285120,
      "step": 119815
    },
    {
      "epoch": 13.181518151815181,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009426059768271263,
      "loss": 0.233,
      "num_input_tokens_seen": 25286208,
      "step": 119820
    },
    {
      "epoch": 13.182068206820682,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009424722868110724,
      "loss": 0.2313,
      "num_input_tokens_seen": 25287200,
      "step": 119825
    },
    {
      "epoch": 13.182618261826182,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.009423386019334845,
      "loss": 0.2308,
      "num_input_tokens_seen": 25288256,
      "step": 119830
    },
    {
      "epoch": 13.183168316831683,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009422049221955942,
      "loss": 0.2329,
      "num_input_tokens_seen": 25289280,
      "step": 119835
    },
    {
      "epoch": 13.183718371837184,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.009420712475986328,
      "loss": 0.2303,
      "num_input_tokens_seen": 25290336,
      "step": 119840
    },
    {
      "epoch": 13.184268426842685,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009419375781438335,
      "loss": 0.2329,
      "num_input_tokens_seen": 25291296,
      "step": 119845
    },
    {
      "epoch": 13.184818481848184,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.009418039138324273,
      "loss": 0.2319,
      "num_input_tokens_seen": 25292288,
      "step": 119850
    },
    {
      "epoch": 13.185368536853685,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009416702546656473,
      "loss": 0.2319,
      "num_input_tokens_seen": 25293376,
      "step": 119855
    },
    {
      "epoch": 13.185918591859187,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.009415366006447242,
      "loss": 0.2288,
      "num_input_tokens_seen": 25294368,
      "step": 119860
    },
    {
      "epoch": 13.186468646864686,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.0094140295177089,
      "loss": 0.2298,
      "num_input_tokens_seen": 25295360,
      "step": 119865
    },
    {
      "epoch": 13.187018701870187,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.00941269308045377,
      "loss": 0.2314,
      "num_input_tokens_seen": 25296416,
      "step": 119870
    },
    {
      "epoch": 13.187568756875688,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.009411356694694168,
      "loss": 0.233,
      "num_input_tokens_seen": 25297504,
      "step": 119875
    },
    {
      "epoch": 13.188118811881187,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.009410020360442412,
      "loss": 0.2304,
      "num_input_tokens_seen": 25298528,
      "step": 119880
    },
    {
      "epoch": 13.188668866886688,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.009408684077710814,
      "loss": 0.2324,
      "num_input_tokens_seen": 25299552,
      "step": 119885
    },
    {
      "epoch": 13.18921892189219,
      "grad_norm": 0.00078582763671875,
      "learning_rate": 0.009407347846511687,
      "loss": 0.2309,
      "num_input_tokens_seen": 25300640,
      "step": 119890
    },
    {
      "epoch": 13.189768976897689,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00940601166685736,
      "loss": 0.2309,
      "num_input_tokens_seen": 25301664,
      "step": 119895
    },
    {
      "epoch": 13.19031903190319,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.009404675538760133,
      "loss": 0.2309,
      "num_input_tokens_seen": 25302784,
      "step": 119900
    },
    {
      "epoch": 13.190869086908691,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.009403339462232327,
      "loss": 0.2278,
      "num_input_tokens_seen": 25303872,
      "step": 119905
    },
    {
      "epoch": 13.191419141914192,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.009402003437286258,
      "loss": 0.2303,
      "num_input_tokens_seen": 25304928,
      "step": 119910
    },
    {
      "epoch": 13.191969196919691,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.009400667463934231,
      "loss": 0.2298,
      "num_input_tokens_seen": 25305952,
      "step": 119915
    },
    {
      "epoch": 13.192519251925193,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.009399331542188574,
      "loss": 0.2309,
      "num_input_tokens_seen": 25307104,
      "step": 119920
    },
    {
      "epoch": 13.193069306930694,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009397995672061588,
      "loss": 0.2319,
      "num_input_tokens_seen": 25308192,
      "step": 119925
    },
    {
      "epoch": 13.193619361936193,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.009396659853565583,
      "loss": 0.234,
      "num_input_tokens_seen": 25309312,
      "step": 119930
    },
    {
      "epoch": 13.194169416941694,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009395324086712883,
      "loss": 0.2319,
      "num_input_tokens_seen": 25310400,
      "step": 119935
    },
    {
      "epoch": 13.194719471947195,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.009393988371515786,
      "loss": 0.2304,
      "num_input_tokens_seen": 25311392,
      "step": 119940
    },
    {
      "epoch": 13.195269526952695,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009392652707986608,
      "loss": 0.2303,
      "num_input_tokens_seen": 25312448,
      "step": 119945
    },
    {
      "epoch": 13.195819581958196,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009391317096137665,
      "loss": 0.2314,
      "num_input_tokens_seen": 25313568,
      "step": 119950
    },
    {
      "epoch": 13.196369636963697,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.009389981535981255,
      "loss": 0.2298,
      "num_input_tokens_seen": 25314528,
      "step": 119955
    },
    {
      "epoch": 13.196919691969198,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009388646027529696,
      "loss": 0.233,
      "num_input_tokens_seen": 25315552,
      "step": 119960
    },
    {
      "epoch": 13.197469746974697,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.009387310570795289,
      "loss": 0.2309,
      "num_input_tokens_seen": 25316608,
      "step": 119965
    },
    {
      "epoch": 13.198019801980198,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.009385975165790357,
      "loss": 0.2309,
      "num_input_tokens_seen": 25317696,
      "step": 119970
    },
    {
      "epoch": 13.1985698569857,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.009384639812527192,
      "loss": 0.2314,
      "num_input_tokens_seen": 25318752,
      "step": 119975
    },
    {
      "epoch": 13.199119911991199,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.009383304511018107,
      "loss": 0.2303,
      "num_input_tokens_seen": 25319872,
      "step": 119980
    },
    {
      "epoch": 13.1996699669967,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00938196926127541,
      "loss": 0.2329,
      "num_input_tokens_seen": 25320928,
      "step": 119985
    },
    {
      "epoch": 13.2002200220022,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.009380634063311404,
      "loss": 0.2298,
      "num_input_tokens_seen": 25321888,
      "step": 119990
    },
    {
      "epoch": 13.2007700770077,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.009379298917138401,
      "loss": 0.2303,
      "num_input_tokens_seen": 25322944,
      "step": 119995
    },
    {
      "epoch": 13.201320132013201,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.009377963822768702,
      "loss": 0.2314,
      "num_input_tokens_seen": 25324000,
      "step": 120000
    },
    {
      "epoch": 13.201870187018702,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.009376628780214611,
      "loss": 0.2303,
      "num_input_tokens_seen": 25325024,
      "step": 120005
    },
    {
      "epoch": 13.202420242024202,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00937529378948844,
      "loss": 0.2324,
      "num_input_tokens_seen": 25326048,
      "step": 120010
    },
    {
      "epoch": 13.202970297029703,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00937395885060248,
      "loss": 0.2314,
      "num_input_tokens_seen": 25327136,
      "step": 120015
    },
    {
      "epoch": 13.203520352035204,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.009372623963569042,
      "loss": 0.2329,
      "num_input_tokens_seen": 25328128,
      "step": 120020
    },
    {
      "epoch": 13.204070407040705,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.009371289128400436,
      "loss": 0.2329,
      "num_input_tokens_seen": 25329184,
      "step": 120025
    },
    {
      "epoch": 13.204620462046204,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.00936995434510895,
      "loss": 0.2298,
      "num_input_tokens_seen": 25330240,
      "step": 120030
    },
    {
      "epoch": 13.205170517051705,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.009368619613706896,
      "loss": 0.2319,
      "num_input_tokens_seen": 25331296,
      "step": 120035
    },
    {
      "epoch": 13.205720572057206,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00936728493420657,
      "loss": 0.2314,
      "num_input_tokens_seen": 25332320,
      "step": 120040
    },
    {
      "epoch": 13.206270627062706,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.00936595030662028,
      "loss": 0.2314,
      "num_input_tokens_seen": 25333376,
      "step": 120045
    },
    {
      "epoch": 13.206820682068207,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.009364615730960323,
      "loss": 0.2319,
      "num_input_tokens_seen": 25334464,
      "step": 120050
    },
    {
      "epoch": 13.207370737073708,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.009363281207238994,
      "loss": 0.2293,
      "num_input_tokens_seen": 25335552,
      "step": 120055
    },
    {
      "epoch": 13.207920792079207,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.009361946735468599,
      "loss": 0.2304,
      "num_input_tokens_seen": 25336576,
      "step": 120060
    },
    {
      "epoch": 13.208470847084708,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.009360612315661434,
      "loss": 0.2298,
      "num_input_tokens_seen": 25337664,
      "step": 120065
    },
    {
      "epoch": 13.20902090209021,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009359277947829807,
      "loss": 0.2324,
      "num_input_tokens_seen": 25338688,
      "step": 120070
    },
    {
      "epoch": 13.209570957095709,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.009357943631986002,
      "loss": 0.2314,
      "num_input_tokens_seen": 25339776,
      "step": 120075
    },
    {
      "epoch": 13.21012101210121,
      "grad_norm": 0.000530242919921875,
      "learning_rate": 0.00935660936814232,
      "loss": 0.2309,
      "num_input_tokens_seen": 25340864,
      "step": 120080
    },
    {
      "epoch": 13.210671067106711,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.009355275156311071,
      "loss": 0.2298,
      "num_input_tokens_seen": 25341888,
      "step": 120085
    },
    {
      "epoch": 13.211221122112212,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009353940996504537,
      "loss": 0.2319,
      "num_input_tokens_seen": 25342912,
      "step": 120090
    },
    {
      "epoch": 13.211771177117711,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.009352606888735019,
      "loss": 0.2314,
      "num_input_tokens_seen": 25343968,
      "step": 120095
    },
    {
      "epoch": 13.212321232123212,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.009351272833014816,
      "loss": 0.2314,
      "num_input_tokens_seen": 25345088,
      "step": 120100
    },
    {
      "epoch": 13.212871287128714,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.009349938829356214,
      "loss": 0.2325,
      "num_input_tokens_seen": 25346208,
      "step": 120105
    },
    {
      "epoch": 13.213421342134213,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.009348604877771524,
      "loss": 0.2303,
      "num_input_tokens_seen": 25347232,
      "step": 120110
    },
    {
      "epoch": 13.213971397139714,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009347270978273026,
      "loss": 0.234,
      "num_input_tokens_seen": 25348288,
      "step": 120115
    },
    {
      "epoch": 13.214521452145215,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.009345937130873016,
      "loss": 0.2298,
      "num_input_tokens_seen": 25349312,
      "step": 120120
    },
    {
      "epoch": 13.215071507150714,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009344603335583799,
      "loss": 0.2309,
      "num_input_tokens_seen": 25350368,
      "step": 120125
    },
    {
      "epoch": 13.215621562156215,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.00934326959241765,
      "loss": 0.2319,
      "num_input_tokens_seen": 25351392,
      "step": 120130
    },
    {
      "epoch": 13.216171617161717,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009341935901386876,
      "loss": 0.2308,
      "num_input_tokens_seen": 25352416,
      "step": 120135
    },
    {
      "epoch": 13.216721672167218,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.009340602262503765,
      "loss": 0.2298,
      "num_input_tokens_seen": 25353472,
      "step": 120140
    },
    {
      "epoch": 13.217271727172717,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.009339268675780605,
      "loss": 0.2314,
      "num_input_tokens_seen": 25354496,
      "step": 120145
    },
    {
      "epoch": 13.217821782178218,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.00933793514122969,
      "loss": 0.2324,
      "num_input_tokens_seen": 25355584,
      "step": 120150
    },
    {
      "epoch": 13.218371837183719,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.009336601658863304,
      "loss": 0.2324,
      "num_input_tokens_seen": 25356640,
      "step": 120155
    },
    {
      "epoch": 13.218921892189218,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.00933526822869375,
      "loss": 0.2324,
      "num_input_tokens_seen": 25357664,
      "step": 120160
    },
    {
      "epoch": 13.21947194719472,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.009333934850733312,
      "loss": 0.2314,
      "num_input_tokens_seen": 25358688,
      "step": 120165
    },
    {
      "epoch": 13.22002200220022,
      "grad_norm": 0.00067901611328125,
      "learning_rate": 0.009332601524994273,
      "loss": 0.2314,
      "num_input_tokens_seen": 25359680,
      "step": 120170
    },
    {
      "epoch": 13.22057205720572,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.009331268251488927,
      "loss": 0.2335,
      "num_input_tokens_seen": 25360672,
      "step": 120175
    },
    {
      "epoch": 13.221122112211221,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.00932993503022956,
      "loss": 0.2308,
      "num_input_tokens_seen": 25361728,
      "step": 120180
    },
    {
      "epoch": 13.221672167216722,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.009328601861228468,
      "loss": 0.2304,
      "num_input_tokens_seen": 25362784,
      "step": 120185
    },
    {
      "epoch": 13.222222222222221,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.009327268744497925,
      "loss": 0.2319,
      "num_input_tokens_seen": 25363840,
      "step": 120190
    },
    {
      "epoch": 13.222772277227723,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.009325935680050221,
      "loss": 0.2314,
      "num_input_tokens_seen": 25364896,
      "step": 120195
    },
    {
      "epoch": 13.223322332233224,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.009324602667897655,
      "loss": 0.2314,
      "num_input_tokens_seen": 25365984,
      "step": 120200
    },
    {
      "epoch": 13.223872387238725,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.009323269708052494,
      "loss": 0.2303,
      "num_input_tokens_seen": 25367040,
      "step": 120205
    },
    {
      "epoch": 13.224422442244224,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.009321936800527036,
      "loss": 0.2319,
      "num_input_tokens_seen": 25368032,
      "step": 120210
    },
    {
      "epoch": 13.224972497249725,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.009320603945333567,
      "loss": 0.2303,
      "num_input_tokens_seen": 25369024,
      "step": 120215
    },
    {
      "epoch": 13.225522552255226,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.009319271142484355,
      "loss": 0.2308,
      "num_input_tokens_seen": 25370048,
      "step": 120220
    },
    {
      "epoch": 13.226072607260726,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.009317938391991703,
      "loss": 0.2303,
      "num_input_tokens_seen": 25371104,
      "step": 120225
    },
    {
      "epoch": 13.226622662266227,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.00931660569386788,
      "loss": 0.2303,
      "num_input_tokens_seen": 25372192,
      "step": 120230
    },
    {
      "epoch": 13.227172717271728,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.00931527304812518,
      "loss": 0.2309,
      "num_input_tokens_seen": 25373344,
      "step": 120235
    },
    {
      "epoch": 13.227722772277227,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.009313940454775885,
      "loss": 0.2324,
      "num_input_tokens_seen": 25374400,
      "step": 120240
    },
    {
      "epoch": 13.228272827282728,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009312607913832264,
      "loss": 0.2303,
      "num_input_tokens_seen": 25375456,
      "step": 120245
    },
    {
      "epoch": 13.22882288228823,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009311275425306608,
      "loss": 0.2314,
      "num_input_tokens_seen": 25376544,
      "step": 120250
    },
    {
      "epoch": 13.229372937293729,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009309942989211197,
      "loss": 0.2309,
      "num_input_tokens_seen": 25377600,
      "step": 120255
    },
    {
      "epoch": 13.22992299229923,
      "grad_norm": 0.000766754150390625,
      "learning_rate": 0.009308610605558315,
      "loss": 0.2298,
      "num_input_tokens_seen": 25378624,
      "step": 120260
    },
    {
      "epoch": 13.23047304730473,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009307278274360238,
      "loss": 0.2329,
      "num_input_tokens_seen": 25379680,
      "step": 120265
    },
    {
      "epoch": 13.231023102310232,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00930594599562924,
      "loss": 0.2335,
      "num_input_tokens_seen": 25380736,
      "step": 120270
    },
    {
      "epoch": 13.231573157315731,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.009304613769377613,
      "loss": 0.2298,
      "num_input_tokens_seen": 25381792,
      "step": 120275
    },
    {
      "epoch": 13.232123212321232,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.009303281595617623,
      "loss": 0.2324,
      "num_input_tokens_seen": 25382848,
      "step": 120280
    },
    {
      "epoch": 13.232673267326733,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.009301949474361552,
      "loss": 0.2304,
      "num_input_tokens_seen": 25383904,
      "step": 120285
    },
    {
      "epoch": 13.233223322332233,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.00930061740562168,
      "loss": 0.2314,
      "num_input_tokens_seen": 25384896,
      "step": 120290
    },
    {
      "epoch": 13.233773377337734,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009299285389410281,
      "loss": 0.2314,
      "num_input_tokens_seen": 25385952,
      "step": 120295
    },
    {
      "epoch": 13.234323432343235,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.009297953425739636,
      "loss": 0.233,
      "num_input_tokens_seen": 25387008,
      "step": 120300
    },
    {
      "epoch": 13.234873487348734,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.00929662151462202,
      "loss": 0.2329,
      "num_input_tokens_seen": 25388000,
      "step": 120305
    },
    {
      "epoch": 13.235423542354235,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0092952896560697,
      "loss": 0.2319,
      "num_input_tokens_seen": 25389024,
      "step": 120310
    },
    {
      "epoch": 13.235973597359736,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009293957850094965,
      "loss": 0.2314,
      "num_input_tokens_seen": 25390016,
      "step": 120315
    },
    {
      "epoch": 13.236523652365236,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.009292626096710076,
      "loss": 0.2314,
      "num_input_tokens_seen": 25391072,
      "step": 120320
    },
    {
      "epoch": 13.237073707370737,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009291294395927319,
      "loss": 0.2288,
      "num_input_tokens_seen": 25392096,
      "step": 120325
    },
    {
      "epoch": 13.237623762376238,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009289962747758963,
      "loss": 0.2314,
      "num_input_tokens_seen": 25393152,
      "step": 120330
    },
    {
      "epoch": 13.238173817381739,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.009288631152217272,
      "loss": 0.2298,
      "num_input_tokens_seen": 25394176,
      "step": 120335
    },
    {
      "epoch": 13.238723872387238,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.009287299609314533,
      "loss": 0.2324,
      "num_input_tokens_seen": 25395200,
      "step": 120340
    },
    {
      "epoch": 13.23927392739274,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.00928596811906301,
      "loss": 0.2309,
      "num_input_tokens_seen": 25396288,
      "step": 120345
    },
    {
      "epoch": 13.23982398239824,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.00928463668147498,
      "loss": 0.2309,
      "num_input_tokens_seen": 25397344,
      "step": 120350
    },
    {
      "epoch": 13.24037403740374,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.009283305296562714,
      "loss": 0.2308,
      "num_input_tokens_seen": 25398336,
      "step": 120355
    },
    {
      "epoch": 13.24092409240924,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.00928197396433847,
      "loss": 0.2298,
      "num_input_tokens_seen": 25399392,
      "step": 120360
    },
    {
      "epoch": 13.241474147414742,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.009280642684814537,
      "loss": 0.2324,
      "num_input_tokens_seen": 25400384,
      "step": 120365
    },
    {
      "epoch": 13.242024202420241,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.009279311458003169,
      "loss": 0.2319,
      "num_input_tokens_seen": 25401440,
      "step": 120370
    },
    {
      "epoch": 13.242574257425742,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00927798028391665,
      "loss": 0.2314,
      "num_input_tokens_seen": 25402528,
      "step": 120375
    },
    {
      "epoch": 13.243124312431243,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009276649162567239,
      "loss": 0.2324,
      "num_input_tokens_seen": 25403584,
      "step": 120380
    },
    {
      "epoch": 13.243674367436745,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0092753180939672,
      "loss": 0.2314,
      "num_input_tokens_seen": 25404640,
      "step": 120385
    },
    {
      "epoch": 13.244224422442244,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.009273987078128817,
      "loss": 0.2309,
      "num_input_tokens_seen": 25405728,
      "step": 120390
    },
    {
      "epoch": 13.244774477447745,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009272656115064339,
      "loss": 0.2308,
      "num_input_tokens_seen": 25406816,
      "step": 120395
    },
    {
      "epoch": 13.245324532453246,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009271325204786044,
      "loss": 0.2298,
      "num_input_tokens_seen": 25407904,
      "step": 120400
    },
    {
      "epoch": 13.245874587458745,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.009269994347306198,
      "loss": 0.2329,
      "num_input_tokens_seen": 25408960,
      "step": 120405
    },
    {
      "epoch": 13.246424642464246,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.00926866354263706,
      "loss": 0.2319,
      "num_input_tokens_seen": 25409984,
      "step": 120410
    },
    {
      "epoch": 13.246974697469748,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.009267332790790906,
      "loss": 0.2308,
      "num_input_tokens_seen": 25411040,
      "step": 120415
    },
    {
      "epoch": 13.247524752475247,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.009266002091779989,
      "loss": 0.2314,
      "num_input_tokens_seen": 25412064,
      "step": 120420
    },
    {
      "epoch": 13.248074807480748,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.009264671445616582,
      "loss": 0.2283,
      "num_input_tokens_seen": 25413184,
      "step": 120425
    },
    {
      "epoch": 13.248624862486249,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00926334085231295,
      "loss": 0.2319,
      "num_input_tokens_seen": 25414208,
      "step": 120430
    },
    {
      "epoch": 13.249174917491748,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009262010311881346,
      "loss": 0.2314,
      "num_input_tokens_seen": 25415232,
      "step": 120435
    },
    {
      "epoch": 13.24972497249725,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.009260679824334043,
      "loss": 0.2314,
      "num_input_tokens_seen": 25416224,
      "step": 120440
    },
    {
      "epoch": 13.25027502750275,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.009259349389683298,
      "loss": 0.2309,
      "num_input_tokens_seen": 25417216,
      "step": 120445
    },
    {
      "epoch": 13.250825082508252,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00925801900794138,
      "loss": 0.2303,
      "num_input_tokens_seen": 25418336,
      "step": 120450
    },
    {
      "epoch": 13.251375137513751,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00925668867912054,
      "loss": 0.2314,
      "num_input_tokens_seen": 25419424,
      "step": 120455
    },
    {
      "epoch": 13.251925192519252,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009255358403233044,
      "loss": 0.2319,
      "num_input_tokens_seen": 25420416,
      "step": 120460
    },
    {
      "epoch": 13.252475247524753,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.009254028180291156,
      "loss": 0.2314,
      "num_input_tokens_seen": 25421472,
      "step": 120465
    },
    {
      "epoch": 13.253025302530252,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009252698010307137,
      "loss": 0.2324,
      "num_input_tokens_seen": 25422464,
      "step": 120470
    },
    {
      "epoch": 13.253575357535754,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.009251367893293236,
      "loss": 0.2314,
      "num_input_tokens_seen": 25423488,
      "step": 120475
    },
    {
      "epoch": 13.254125412541255,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.009250037829261721,
      "loss": 0.2314,
      "num_input_tokens_seen": 25424576,
      "step": 120480
    },
    {
      "epoch": 13.254675467546754,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009248707818224846,
      "loss": 0.2314,
      "num_input_tokens_seen": 25425664,
      "step": 120485
    },
    {
      "epoch": 13.255225522552255,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.009247377860194876,
      "loss": 0.2309,
      "num_input_tokens_seen": 25426752,
      "step": 120490
    },
    {
      "epoch": 13.255775577557756,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.009246047955184064,
      "loss": 0.2314,
      "num_input_tokens_seen": 25427744,
      "step": 120495
    },
    {
      "epoch": 13.256325632563255,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00924471810320466,
      "loss": 0.2298,
      "num_input_tokens_seen": 25428864,
      "step": 120500
    },
    {
      "epoch": 13.256875687568757,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.009243388304268936,
      "loss": 0.2293,
      "num_input_tokens_seen": 25429888,
      "step": 120505
    },
    {
      "epoch": 13.257425742574258,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.009242058558389134,
      "loss": 0.2314,
      "num_input_tokens_seen": 25430976,
      "step": 120510
    },
    {
      "epoch": 13.257975797579759,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009240728865577518,
      "loss": 0.2293,
      "num_input_tokens_seen": 25432000,
      "step": 120515
    },
    {
      "epoch": 13.258525852585258,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.009239399225846342,
      "loss": 0.2319,
      "num_input_tokens_seen": 25433056,
      "step": 120520
    },
    {
      "epoch": 13.25907590759076,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.009238069639207851,
      "loss": 0.2314,
      "num_input_tokens_seen": 25434112,
      "step": 120525
    },
    {
      "epoch": 13.25962596259626,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.009236740105674317,
      "loss": 0.2314,
      "num_input_tokens_seen": 25435168,
      "step": 120530
    },
    {
      "epoch": 13.26017601760176,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009235410625257977,
      "loss": 0.2314,
      "num_input_tokens_seen": 25436256,
      "step": 120535
    },
    {
      "epoch": 13.26072607260726,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009234081197971094,
      "loss": 0.2319,
      "num_input_tokens_seen": 25437312,
      "step": 120540
    },
    {
      "epoch": 13.261276127612762,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00923275182382592,
      "loss": 0.2324,
      "num_input_tokens_seen": 25438368,
      "step": 120545
    },
    {
      "epoch": 13.261826182618261,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.0092314225028347,
      "loss": 0.2309,
      "num_input_tokens_seen": 25439424,
      "step": 120550
    },
    {
      "epoch": 13.262376237623762,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009230093235009696,
      "loss": 0.2314,
      "num_input_tokens_seen": 25440480,
      "step": 120555
    },
    {
      "epoch": 13.262926292629263,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.009228764020363147,
      "loss": 0.2324,
      "num_input_tokens_seen": 25441536,
      "step": 120560
    },
    {
      "epoch": 13.263476347634764,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009227434858907322,
      "loss": 0.2324,
      "num_input_tokens_seen": 25442592,
      "step": 120565
    },
    {
      "epoch": 13.264026402640264,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.009226105750654452,
      "loss": 0.2303,
      "num_input_tokens_seen": 25443648,
      "step": 120570
    },
    {
      "epoch": 13.264576457645765,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009224776695616796,
      "loss": 0.2309,
      "num_input_tokens_seen": 25444768,
      "step": 120575
    },
    {
      "epoch": 13.265126512651266,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009223447693806603,
      "loss": 0.2293,
      "num_input_tokens_seen": 25445856,
      "step": 120580
    },
    {
      "epoch": 13.265676567656765,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.009222118745236114,
      "loss": 0.2303,
      "num_input_tokens_seen": 25446880,
      "step": 120585
    },
    {
      "epoch": 13.266226622662266,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.009220789849917595,
      "loss": 0.2278,
      "num_input_tokens_seen": 25447968,
      "step": 120590
    },
    {
      "epoch": 13.266776677667767,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.009219461007863278,
      "loss": 0.2288,
      "num_input_tokens_seen": 25449088,
      "step": 120595
    },
    {
      "epoch": 13.267326732673267,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00921813221908541,
      "loss": 0.234,
      "num_input_tokens_seen": 25450112,
      "step": 120600
    },
    {
      "epoch": 13.267876787678768,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.009216803483596252,
      "loss": 0.2329,
      "num_input_tokens_seen": 25451136,
      "step": 120605
    },
    {
      "epoch": 13.268426842684269,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.009215474801408035,
      "loss": 0.2309,
      "num_input_tokens_seen": 25452288,
      "step": 120610
    },
    {
      "epoch": 13.268976897689768,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.009214146172533012,
      "loss": 0.2324,
      "num_input_tokens_seen": 25453344,
      "step": 120615
    },
    {
      "epoch": 13.26952695269527,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009212817596983433,
      "loss": 0.2303,
      "num_input_tokens_seen": 25454336,
      "step": 120620
    },
    {
      "epoch": 13.27007700770077,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.009211489074771529,
      "loss": 0.2319,
      "num_input_tokens_seen": 25455424,
      "step": 120625
    },
    {
      "epoch": 13.270627062706271,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.009210160605909555,
      "loss": 0.2308,
      "num_input_tokens_seen": 25456544,
      "step": 120630
    },
    {
      "epoch": 13.27117711771177,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009208832190409754,
      "loss": 0.2324,
      "num_input_tokens_seen": 25457568,
      "step": 120635
    },
    {
      "epoch": 13.271727172717272,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.00920750382828437,
      "loss": 0.2304,
      "num_input_tokens_seen": 25458624,
      "step": 120640
    },
    {
      "epoch": 13.272277227722773,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.009206175519545643,
      "loss": 0.2304,
      "num_input_tokens_seen": 25459712,
      "step": 120645
    },
    {
      "epoch": 13.272827282728272,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.00920484726420581,
      "loss": 0.2304,
      "num_input_tokens_seen": 25460800,
      "step": 120650
    },
    {
      "epoch": 13.273377337733773,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.009203519062277125,
      "loss": 0.2303,
      "num_input_tokens_seen": 25461856,
      "step": 120655
    },
    {
      "epoch": 13.273927392739274,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.009202190913771823,
      "loss": 0.2293,
      "num_input_tokens_seen": 25462944,
      "step": 120660
    },
    {
      "epoch": 13.274477447744774,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009200862818702143,
      "loss": 0.2324,
      "num_input_tokens_seen": 25464032,
      "step": 120665
    },
    {
      "epoch": 13.275027502750275,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.00919953477708033,
      "loss": 0.234,
      "num_input_tokens_seen": 25465056,
      "step": 120670
    },
    {
      "epoch": 13.275577557755776,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.009198206788918618,
      "loss": 0.2304,
      "num_input_tokens_seen": 25466176,
      "step": 120675
    },
    {
      "epoch": 13.276127612761275,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.009196878854229254,
      "loss": 0.2335,
      "num_input_tokens_seen": 25467232,
      "step": 120680
    },
    {
      "epoch": 13.276677667766776,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.009195550973024472,
      "loss": 0.2293,
      "num_input_tokens_seen": 25468352,
      "step": 120685
    },
    {
      "epoch": 13.277227722772277,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.009194223145316508,
      "loss": 0.2329,
      "num_input_tokens_seen": 25469408,
      "step": 120690
    },
    {
      "epoch": 13.277777777777779,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.00919289537111761,
      "loss": 0.2314,
      "num_input_tokens_seen": 25470400,
      "step": 120695
    },
    {
      "epoch": 13.278327832783278,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009191567650440003,
      "loss": 0.2314,
      "num_input_tokens_seen": 25471456,
      "step": 120700
    },
    {
      "epoch": 13.278877887788779,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.009190239983295932,
      "loss": 0.2319,
      "num_input_tokens_seen": 25472480,
      "step": 120705
    },
    {
      "epoch": 13.27942794279428,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.009188912369697632,
      "loss": 0.2309,
      "num_input_tokens_seen": 25473472,
      "step": 120710
    },
    {
      "epoch": 13.27997799779978,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.009187584809657335,
      "loss": 0.2309,
      "num_input_tokens_seen": 25474624,
      "step": 120715
    },
    {
      "epoch": 13.28052805280528,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.009186257303187284,
      "loss": 0.2299,
      "num_input_tokens_seen": 25475680,
      "step": 120720
    },
    {
      "epoch": 13.281078107810782,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.009184929850299705,
      "loss": 0.2304,
      "num_input_tokens_seen": 25476736,
      "step": 120725
    },
    {
      "epoch": 13.281628162816281,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.009183602451006842,
      "loss": 0.2309,
      "num_input_tokens_seen": 25477728,
      "step": 120730
    },
    {
      "epoch": 13.282178217821782,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009182275105320924,
      "loss": 0.2309,
      "num_input_tokens_seen": 25478816,
      "step": 120735
    },
    {
      "epoch": 13.282728272827283,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.009180947813254181,
      "loss": 0.2335,
      "num_input_tokens_seen": 25479840,
      "step": 120740
    },
    {
      "epoch": 13.283278327832782,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.009179620574818853,
      "loss": 0.2277,
      "num_input_tokens_seen": 25480896,
      "step": 120745
    },
    {
      "epoch": 13.283828382838283,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.009178293390027164,
      "loss": 0.2319,
      "num_input_tokens_seen": 25481920,
      "step": 120750
    },
    {
      "epoch": 13.284378437843785,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.009176966258891361,
      "loss": 0.2314,
      "num_input_tokens_seen": 25483008,
      "step": 120755
    },
    {
      "epoch": 13.284928492849286,
      "grad_norm": 0.0028533935546875,
      "learning_rate": 0.00917563918142366,
      "loss": 0.2293,
      "num_input_tokens_seen": 25484032,
      "step": 120760
    },
    {
      "epoch": 13.285478547854785,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009174312157636295,
      "loss": 0.2303,
      "num_input_tokens_seen": 25485120,
      "step": 120765
    },
    {
      "epoch": 13.286028602860286,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.009172985187541503,
      "loss": 0.2309,
      "num_input_tokens_seen": 25486144,
      "step": 120770
    },
    {
      "epoch": 13.286578657865787,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.009171658271151507,
      "loss": 0.2304,
      "num_input_tokens_seen": 25487232,
      "step": 120775
    },
    {
      "epoch": 13.287128712871286,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.009170331408478545,
      "loss": 0.2324,
      "num_input_tokens_seen": 25488352,
      "step": 120780
    },
    {
      "epoch": 13.287678767876788,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009169004599534842,
      "loss": 0.233,
      "num_input_tokens_seen": 25489440,
      "step": 120785
    },
    {
      "epoch": 13.288228822882289,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.00916767784433262,
      "loss": 0.2329,
      "num_input_tokens_seen": 25490528,
      "step": 120790
    },
    {
      "epoch": 13.288778877887788,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.009166351142884119,
      "loss": 0.2304,
      "num_input_tokens_seen": 25491616,
      "step": 120795
    },
    {
      "epoch": 13.289328932893289,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.009165024495201552,
      "loss": 0.2319,
      "num_input_tokens_seen": 25492736,
      "step": 120800
    },
    {
      "epoch": 13.28987898789879,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009163697901297159,
      "loss": 0.2314,
      "num_input_tokens_seen": 25493728,
      "step": 120805
    },
    {
      "epoch": 13.290429042904291,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.009162371361183166,
      "loss": 0.2303,
      "num_input_tokens_seen": 25494752,
      "step": 120810
    },
    {
      "epoch": 13.29097909790979,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.009161044874871788,
      "loss": 0.2303,
      "num_input_tokens_seen": 25495808,
      "step": 120815
    },
    {
      "epoch": 13.291529152915292,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.009159718442375262,
      "loss": 0.234,
      "num_input_tokens_seen": 25496896,
      "step": 120820
    },
    {
      "epoch": 13.292079207920793,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.009158392063705804,
      "loss": 0.2309,
      "num_input_tokens_seen": 25497952,
      "step": 120825
    },
    {
      "epoch": 13.292629262926292,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009157065738875647,
      "loss": 0.2309,
      "num_input_tokens_seen": 25499008,
      "step": 120830
    },
    {
      "epoch": 13.293179317931793,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.009155739467897012,
      "loss": 0.2308,
      "num_input_tokens_seen": 25500000,
      "step": 120835
    },
    {
      "epoch": 13.293729372937294,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00915441325078212,
      "loss": 0.234,
      "num_input_tokens_seen": 25501056,
      "step": 120840
    },
    {
      "epoch": 13.294279427942794,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009153087087543197,
      "loss": 0.2299,
      "num_input_tokens_seen": 25502080,
      "step": 120845
    },
    {
      "epoch": 13.294829482948295,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009151760978192468,
      "loss": 0.2314,
      "num_input_tokens_seen": 25503136,
      "step": 120850
    },
    {
      "epoch": 13.295379537953796,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009150434922742146,
      "loss": 0.2314,
      "num_input_tokens_seen": 25504128,
      "step": 120855
    },
    {
      "epoch": 13.295929592959295,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.009149108921204461,
      "loss": 0.2319,
      "num_input_tokens_seen": 25505152,
      "step": 120860
    },
    {
      "epoch": 13.296479647964796,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.009147782973591628,
      "loss": 0.2309,
      "num_input_tokens_seen": 25506240,
      "step": 120865
    },
    {
      "epoch": 13.297029702970297,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.009146457079915878,
      "loss": 0.2345,
      "num_input_tokens_seen": 25507296,
      "step": 120870
    },
    {
      "epoch": 13.297579757975798,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.009145131240189421,
      "loss": 0.2324,
      "num_input_tokens_seen": 25508320,
      "step": 120875
    },
    {
      "epoch": 13.298129812981298,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.009143805454424475,
      "loss": 0.2299,
      "num_input_tokens_seen": 25509408,
      "step": 120880
    },
    {
      "epoch": 13.298679867986799,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.009142479722633273,
      "loss": 0.2314,
      "num_input_tokens_seen": 25510464,
      "step": 120885
    },
    {
      "epoch": 13.2992299229923,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009141154044828015,
      "loss": 0.2324,
      "num_input_tokens_seen": 25511552,
      "step": 120890
    },
    {
      "epoch": 13.2997799779978,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.009139828421020938,
      "loss": 0.2314,
      "num_input_tokens_seen": 25512704,
      "step": 120895
    },
    {
      "epoch": 13.3003300330033,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009138502851224247,
      "loss": 0.2299,
      "num_input_tokens_seen": 25513792,
      "step": 120900
    },
    {
      "epoch": 13.300880088008801,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00913717733545016,
      "loss": 0.2308,
      "num_input_tokens_seen": 25514848,
      "step": 120905
    },
    {
      "epoch": 13.3014301430143,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.009135851873710901,
      "loss": 0.2329,
      "num_input_tokens_seen": 25515936,
      "step": 120910
    },
    {
      "epoch": 13.301980198019802,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.009134526466018675,
      "loss": 0.2319,
      "num_input_tokens_seen": 25516960,
      "step": 120915
    },
    {
      "epoch": 13.302530253025303,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00913320111238571,
      "loss": 0.233,
      "num_input_tokens_seen": 25518048,
      "step": 120920
    },
    {
      "epoch": 13.303080308030804,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009131875812824217,
      "loss": 0.2314,
      "num_input_tokens_seen": 25519168,
      "step": 120925
    },
    {
      "epoch": 13.303630363036303,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.009130550567346404,
      "loss": 0.2304,
      "num_input_tokens_seen": 25520224,
      "step": 120930
    },
    {
      "epoch": 13.304180418041804,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.00912922537596449,
      "loss": 0.2325,
      "num_input_tokens_seen": 25521312,
      "step": 120935
    },
    {
      "epoch": 13.304730473047305,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.009127900238690687,
      "loss": 0.2309,
      "num_input_tokens_seen": 25522336,
      "step": 120940
    },
    {
      "epoch": 13.305280528052805,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.009126575155537213,
      "loss": 0.2303,
      "num_input_tokens_seen": 25523424,
      "step": 120945
    },
    {
      "epoch": 13.305830583058306,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.00912525012651628,
      "loss": 0.2329,
      "num_input_tokens_seen": 25524448,
      "step": 120950
    },
    {
      "epoch": 13.306380638063807,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.00912392515164009,
      "loss": 0.2298,
      "num_input_tokens_seen": 25525504,
      "step": 120955
    },
    {
      "epoch": 13.306930693069306,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.009122600230920867,
      "loss": 0.2335,
      "num_input_tokens_seen": 25526560,
      "step": 120960
    },
    {
      "epoch": 13.307480748074807,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.009121275364370816,
      "loss": 0.2319,
      "num_input_tokens_seen": 25527584,
      "step": 120965
    },
    {
      "epoch": 13.308030803080309,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.009119950552002152,
      "loss": 0.2293,
      "num_input_tokens_seen": 25528672,
      "step": 120970
    },
    {
      "epoch": 13.308580858085808,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.009118625793827082,
      "loss": 0.2304,
      "num_input_tokens_seen": 25529760,
      "step": 120975
    },
    {
      "epoch": 13.309130913091309,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.00911730108985781,
      "loss": 0.2304,
      "num_input_tokens_seen": 25530848,
      "step": 120980
    },
    {
      "epoch": 13.30968096809681,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.00911597644010656,
      "loss": 0.2288,
      "num_input_tokens_seen": 25531936,
      "step": 120985
    },
    {
      "epoch": 13.310231023102311,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.009114651844585524,
      "loss": 0.2314,
      "num_input_tokens_seen": 25532928,
      "step": 120990
    },
    {
      "epoch": 13.31078107810781,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.009113327303306922,
      "loss": 0.2319,
      "num_input_tokens_seen": 25533952,
      "step": 120995
    },
    {
      "epoch": 13.311331133113312,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.00911200281628296,
      "loss": 0.2324,
      "num_input_tokens_seen": 25535072,
      "step": 121000
    },
    {
      "epoch": 13.311881188118813,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.009110678383525833,
      "loss": 0.2303,
      "num_input_tokens_seen": 25536096,
      "step": 121005
    },
    {
      "epoch": 13.312431243124312,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.009109354005047768,
      "loss": 0.2298,
      "num_input_tokens_seen": 25537184,
      "step": 121010
    },
    {
      "epoch": 13.312981298129813,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009108029680860953,
      "loss": 0.2314,
      "num_input_tokens_seen": 25538176,
      "step": 121015
    },
    {
      "epoch": 13.313531353135314,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.009106705410977604,
      "loss": 0.2303,
      "num_input_tokens_seen": 25539232,
      "step": 121020
    },
    {
      "epoch": 13.314081408140813,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.009105381195409927,
      "loss": 0.2314,
      "num_input_tokens_seen": 25540288,
      "step": 121025
    },
    {
      "epoch": 13.314631463146315,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009104057034170119,
      "loss": 0.2288,
      "num_input_tokens_seen": 25541376,
      "step": 121030
    },
    {
      "epoch": 13.315181518151816,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.009102732927270388,
      "loss": 0.2309,
      "num_input_tokens_seen": 25542400,
      "step": 121035
    },
    {
      "epoch": 13.315731573157315,
      "grad_norm": 0.000782012939453125,
      "learning_rate": 0.009101408874722937,
      "loss": 0.235,
      "num_input_tokens_seen": 25543392,
      "step": 121040
    },
    {
      "epoch": 13.316281628162816,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.009100084876539974,
      "loss": 0.2308,
      "num_input_tokens_seen": 25544416,
      "step": 121045
    },
    {
      "epoch": 13.316831683168317,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.009098760932733697,
      "loss": 0.2309,
      "num_input_tokens_seen": 25545408,
      "step": 121050
    },
    {
      "epoch": 13.317381738173818,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.009097437043316304,
      "loss": 0.233,
      "num_input_tokens_seen": 25546464,
      "step": 121055
    },
    {
      "epoch": 13.317931793179318,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.009096113208300009,
      "loss": 0.233,
      "num_input_tokens_seen": 25547584,
      "step": 121060
    },
    {
      "epoch": 13.318481848184819,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.009094789427697,
      "loss": 0.2293,
      "num_input_tokens_seen": 25548640,
      "step": 121065
    },
    {
      "epoch": 13.31903190319032,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.009093465701519482,
      "loss": 0.2298,
      "num_input_tokens_seen": 25549600,
      "step": 121070
    },
    {
      "epoch": 13.319581958195819,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.009092142029779659,
      "loss": 0.2303,
      "num_input_tokens_seen": 25550624,
      "step": 121075
    },
    {
      "epoch": 13.32013201320132,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.009090818412489725,
      "loss": 0.2329,
      "num_input_tokens_seen": 25551744,
      "step": 121080
    },
    {
      "epoch": 13.320682068206821,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.009089494849661888,
      "loss": 0.2309,
      "num_input_tokens_seen": 25552768,
      "step": 121085
    },
    {
      "epoch": 13.32123212321232,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00908817134130834,
      "loss": 0.2298,
      "num_input_tokens_seen": 25553792,
      "step": 121090
    },
    {
      "epoch": 13.321782178217822,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.009086847887441272,
      "loss": 0.2308,
      "num_input_tokens_seen": 25554880,
      "step": 121095
    },
    {
      "epoch": 13.322332233223323,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.009085524488072901,
      "loss": 0.2314,
      "num_input_tokens_seen": 25555904,
      "step": 121100
    },
    {
      "epoch": 13.322882288228822,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009084201143215403,
      "loss": 0.2303,
      "num_input_tokens_seen": 25556896,
      "step": 121105
    },
    {
      "epoch": 13.323432343234323,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.00908287785288099,
      "loss": 0.2308,
      "num_input_tokens_seen": 25557984,
      "step": 121110
    },
    {
      "epoch": 13.323982398239824,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.009081554617081855,
      "loss": 0.2314,
      "num_input_tokens_seen": 25559072,
      "step": 121115
    },
    {
      "epoch": 13.324532453245325,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.009080231435830184,
      "loss": 0.2319,
      "num_input_tokens_seen": 25560096,
      "step": 121120
    },
    {
      "epoch": 13.325082508250825,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.009078908309138182,
      "loss": 0.2319,
      "num_input_tokens_seen": 25561184,
      "step": 121125
    },
    {
      "epoch": 13.325632563256326,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00907758523701804,
      "loss": 0.2319,
      "num_input_tokens_seen": 25562208,
      "step": 121130
    },
    {
      "epoch": 13.326182618261827,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.009076262219481953,
      "loss": 0.2309,
      "num_input_tokens_seen": 25563200,
      "step": 121135
    },
    {
      "epoch": 13.326732673267326,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009074939256542121,
      "loss": 0.2325,
      "num_input_tokens_seen": 25564288,
      "step": 121140
    },
    {
      "epoch": 13.327282728272827,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.009073616348210722,
      "loss": 0.2293,
      "num_input_tokens_seen": 25565280,
      "step": 121145
    },
    {
      "epoch": 13.327832783278328,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.009072293494499962,
      "loss": 0.2314,
      "num_input_tokens_seen": 25566272,
      "step": 121150
    },
    {
      "epoch": 13.328382838283828,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.009070970695422025,
      "loss": 0.2293,
      "num_input_tokens_seen": 25567392,
      "step": 121155
    },
    {
      "epoch": 13.328932893289329,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.009069647950989114,
      "loss": 0.2324,
      "num_input_tokens_seen": 25568512,
      "step": 121160
    },
    {
      "epoch": 13.32948294829483,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.009068325261213407,
      "loss": 0.2335,
      "num_input_tokens_seen": 25569568,
      "step": 121165
    },
    {
      "epoch": 13.33003300330033,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.009067002626107096,
      "loss": 0.2319,
      "num_input_tokens_seen": 25570624,
      "step": 121170
    },
    {
      "epoch": 13.33058305830583,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.009065680045682381,
      "loss": 0.2325,
      "num_input_tokens_seen": 25571648,
      "step": 121175
    },
    {
      "epoch": 13.331133113311331,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.009064357519951442,
      "loss": 0.2361,
      "num_input_tokens_seen": 25572736,
      "step": 121180
    },
    {
      "epoch": 13.331683168316832,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.009063035048926471,
      "loss": 0.2288,
      "num_input_tokens_seen": 25573856,
      "step": 121185
    },
    {
      "epoch": 13.332233223322332,
      "grad_norm": 0.0103759765625,
      "learning_rate": 0.009061712632619661,
      "loss": 0.2298,
      "num_input_tokens_seen": 25574912,
      "step": 121190
    },
    {
      "epoch": 13.332783278327833,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00906039027104319,
      "loss": 0.234,
      "num_input_tokens_seen": 25575968,
      "step": 121195
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00905906796420926,
      "loss": 0.2309,
      "num_input_tokens_seen": 25577024,
      "step": 121200
    },
    {
      "epoch": 13.333883388338833,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.009057745712130041,
      "loss": 0.2308,
      "num_input_tokens_seen": 25578016,
      "step": 121205
    },
    {
      "epoch": 13.334433443344334,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.009056423514817735,
      "loss": 0.2314,
      "num_input_tokens_seen": 25579104,
      "step": 121210
    },
    {
      "epoch": 13.334983498349835,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.009055101372284522,
      "loss": 0.2309,
      "num_input_tokens_seen": 25580160,
      "step": 121215
    },
    {
      "epoch": 13.335533553355335,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.009053779284542582,
      "loss": 0.2309,
      "num_input_tokens_seen": 25581184,
      "step": 121220
    },
    {
      "epoch": 13.336083608360836,
      "grad_norm": 0.000843048095703125,
      "learning_rate": 0.009052457251604109,
      "loss": 0.2288,
      "num_input_tokens_seen": 25582240,
      "step": 121225
    },
    {
      "epoch": 13.336633663366337,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.00905113527348128,
      "loss": 0.2319,
      "num_input_tokens_seen": 25583360,
      "step": 121230
    },
    {
      "epoch": 13.337183718371838,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.009049813350186287,
      "loss": 0.233,
      "num_input_tokens_seen": 25584352,
      "step": 121235
    },
    {
      "epoch": 13.337733773377337,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009048491481731306,
      "loss": 0.2303,
      "num_input_tokens_seen": 25585408,
      "step": 121240
    },
    {
      "epoch": 13.338283828382838,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.009047169668128522,
      "loss": 0.2303,
      "num_input_tokens_seen": 25586496,
      "step": 121245
    },
    {
      "epoch": 13.33883388338834,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.009045847909390121,
      "loss": 0.2319,
      "num_input_tokens_seen": 25587584,
      "step": 121250
    },
    {
      "epoch": 13.339383938393839,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.009044526205528286,
      "loss": 0.2324,
      "num_input_tokens_seen": 25588608,
      "step": 121255
    },
    {
      "epoch": 13.33993399339934,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.009043204556555188,
      "loss": 0.2304,
      "num_input_tokens_seen": 25589664,
      "step": 121260
    },
    {
      "epoch": 13.340484048404841,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.00904188296248302,
      "loss": 0.2325,
      "num_input_tokens_seen": 25590752,
      "step": 121265
    },
    {
      "epoch": 13.34103410341034,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.009040561423323953,
      "loss": 0.2288,
      "num_input_tokens_seen": 25591840,
      "step": 121270
    },
    {
      "epoch": 13.341584158415841,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.00903923993909018,
      "loss": 0.2304,
      "num_input_tokens_seen": 25592864,
      "step": 121275
    },
    {
      "epoch": 13.342134213421343,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00903791850979387,
      "loss": 0.2298,
      "num_input_tokens_seen": 25593920,
      "step": 121280
    },
    {
      "epoch": 13.342684268426842,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.009036597135447197,
      "loss": 0.2314,
      "num_input_tokens_seen": 25594944,
      "step": 121285
    },
    {
      "epoch": 13.343234323432343,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.009035275816062357,
      "loss": 0.2309,
      "num_input_tokens_seen": 25596000,
      "step": 121290
    },
    {
      "epoch": 13.343784378437844,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.00903395455165151,
      "loss": 0.2304,
      "num_input_tokens_seen": 25597056,
      "step": 121295
    },
    {
      "epoch": 13.344334433443345,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.009032633342226842,
      "loss": 0.2324,
      "num_input_tokens_seen": 25598112,
      "step": 121300
    },
    {
      "epoch": 13.344884488448844,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.009031312187800535,
      "loss": 0.2309,
      "num_input_tokens_seen": 25599200,
      "step": 121305
    },
    {
      "epoch": 13.345434543454346,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.009029991088384749,
      "loss": 0.2324,
      "num_input_tokens_seen": 25600224,
      "step": 121310
    },
    {
      "epoch": 13.345984598459847,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.00902867004399168,
      "loss": 0.2324,
      "num_input_tokens_seen": 25601216,
      "step": 121315
    },
    {
      "epoch": 13.346534653465346,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.009027349054633487,
      "loss": 0.2314,
      "num_input_tokens_seen": 25602240,
      "step": 121320
    },
    {
      "epoch": 13.347084708470847,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.009026028120322355,
      "loss": 0.2298,
      "num_input_tokens_seen": 25603328,
      "step": 121325
    },
    {
      "epoch": 13.347634763476348,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00902470724107046,
      "loss": 0.2309,
      "num_input_tokens_seen": 25604416,
      "step": 121330
    },
    {
      "epoch": 13.348184818481847,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.009023386416889963,
      "loss": 0.2309,
      "num_input_tokens_seen": 25605408,
      "step": 121335
    },
    {
      "epoch": 13.348734873487349,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00902206564779305,
      "loss": 0.2304,
      "num_input_tokens_seen": 25606464,
      "step": 121340
    },
    {
      "epoch": 13.34928492849285,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.009020744933791885,
      "loss": 0.2325,
      "num_input_tokens_seen": 25607584,
      "step": 121345
    },
    {
      "epoch": 13.34983498349835,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.00901942427489865,
      "loss": 0.234,
      "num_input_tokens_seen": 25608672,
      "step": 121350
    },
    {
      "epoch": 13.35038503850385,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.009018103671125512,
      "loss": 0.233,
      "num_input_tokens_seen": 25609696,
      "step": 121355
    },
    {
      "epoch": 13.350935093509351,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.009016783122484636,
      "loss": 0.2319,
      "num_input_tokens_seen": 25610752,
      "step": 121360
    },
    {
      "epoch": 13.351485148514852,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.009015462628988207,
      "loss": 0.2319,
      "num_input_tokens_seen": 25611808,
      "step": 121365
    },
    {
      "epoch": 13.352035203520352,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.009014142190648378,
      "loss": 0.2319,
      "num_input_tokens_seen": 25612928,
      "step": 121370
    },
    {
      "epoch": 13.352585258525853,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.009012821807477336,
      "loss": 0.2293,
      "num_input_tokens_seen": 25613984,
      "step": 121375
    },
    {
      "epoch": 13.353135313531354,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.009011501479487241,
      "loss": 0.2298,
      "num_input_tokens_seen": 25615008,
      "step": 121380
    },
    {
      "epoch": 13.353685368536853,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.00901018120669026,
      "loss": 0.2319,
      "num_input_tokens_seen": 25616000,
      "step": 121385
    },
    {
      "epoch": 13.354235423542354,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.009008860989098571,
      "loss": 0.2314,
      "num_input_tokens_seen": 25617088,
      "step": 121390
    },
    {
      "epoch": 13.354785478547855,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.009007540826724328,
      "loss": 0.2319,
      "num_input_tokens_seen": 25618112,
      "step": 121395
    },
    {
      "epoch": 13.355335533553355,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.009006220719579709,
      "loss": 0.2329,
      "num_input_tokens_seen": 25619168,
      "step": 121400
    },
    {
      "epoch": 13.355885588558856,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00900490066767688,
      "loss": 0.2335,
      "num_input_tokens_seen": 25620288,
      "step": 121405
    },
    {
      "epoch": 13.356435643564357,
      "grad_norm": 0.005859375,
      "learning_rate": 0.009003580671028,
      "loss": 0.2293,
      "num_input_tokens_seen": 25621344,
      "step": 121410
    },
    {
      "epoch": 13.356985698569858,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.009002260729645242,
      "loss": 0.2319,
      "num_input_tokens_seen": 25622432,
      "step": 121415
    },
    {
      "epoch": 13.357535753575357,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.009000940843540766,
      "loss": 0.2308,
      "num_input_tokens_seen": 25623456,
      "step": 121420
    },
    {
      "epoch": 13.358085808580858,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.008999621012726748,
      "loss": 0.2298,
      "num_input_tokens_seen": 25624480,
      "step": 121425
    },
    {
      "epoch": 13.35863586358636,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.008998301237215336,
      "loss": 0.2314,
      "num_input_tokens_seen": 25625504,
      "step": 121430
    },
    {
      "epoch": 13.359185918591859,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.008996981517018703,
      "loss": 0.233,
      "num_input_tokens_seen": 25626560,
      "step": 121435
    },
    {
      "epoch": 13.35973597359736,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008995661852149011,
      "loss": 0.2324,
      "num_input_tokens_seen": 25627648,
      "step": 121440
    },
    {
      "epoch": 13.36028602860286,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.008994342242618426,
      "loss": 0.2319,
      "num_input_tokens_seen": 25628640,
      "step": 121445
    },
    {
      "epoch": 13.36083608360836,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008993022688439102,
      "loss": 0.2324,
      "num_input_tokens_seen": 25629728,
      "step": 121450
    },
    {
      "epoch": 13.361386138613861,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.008991703189623208,
      "loss": 0.2324,
      "num_input_tokens_seen": 25630784,
      "step": 121455
    },
    {
      "epoch": 13.361936193619362,
      "grad_norm": 0.0005340576171875,
      "learning_rate": 0.0089903837461829,
      "loss": 0.2308,
      "num_input_tokens_seen": 25631808,
      "step": 121460
    },
    {
      "epoch": 13.362486248624862,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008989064358130346,
      "loss": 0.2298,
      "num_input_tokens_seen": 25632832,
      "step": 121465
    },
    {
      "epoch": 13.363036303630363,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0089877450254777,
      "loss": 0.2346,
      "num_input_tokens_seen": 25633920,
      "step": 121470
    },
    {
      "epoch": 13.363586358635864,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.008986425748237117,
      "loss": 0.2309,
      "num_input_tokens_seen": 25635008,
      "step": 121475
    },
    {
      "epoch": 13.364136413641365,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.008985106526420773,
      "loss": 0.2298,
      "num_input_tokens_seen": 25636032,
      "step": 121480
    },
    {
      "epoch": 13.364686468646864,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.008983787360040806,
      "loss": 0.2298,
      "num_input_tokens_seen": 25637088,
      "step": 121485
    },
    {
      "epoch": 13.365236523652365,
      "grad_norm": 0.0007171630859375,
      "learning_rate": 0.00898246824910939,
      "loss": 0.2298,
      "num_input_tokens_seen": 25638176,
      "step": 121490
    },
    {
      "epoch": 13.365786578657866,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.008981149193638674,
      "loss": 0.2304,
      "num_input_tokens_seen": 25639328,
      "step": 121495
    },
    {
      "epoch": 13.366336633663366,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008979830193640816,
      "loss": 0.2319,
      "num_input_tokens_seen": 25640448,
      "step": 121500
    },
    {
      "epoch": 13.366886688668867,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.00897851124912798,
      "loss": 0.2288,
      "num_input_tokens_seen": 25641472,
      "step": 121505
    },
    {
      "epoch": 13.367436743674368,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.008977192360112312,
      "loss": 0.2325,
      "num_input_tokens_seen": 25642528,
      "step": 121510
    },
    {
      "epoch": 13.367986798679867,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.008975873526605973,
      "loss": 0.2335,
      "num_input_tokens_seen": 25643552,
      "step": 121515
    },
    {
      "epoch": 13.368536853685368,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00897455474862112,
      "loss": 0.2283,
      "num_input_tokens_seen": 25644576,
      "step": 121520
    },
    {
      "epoch": 13.36908690869087,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.008973236026169899,
      "loss": 0.2288,
      "num_input_tokens_seen": 25645568,
      "step": 121525
    },
    {
      "epoch": 13.369636963696369,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.008971917359264473,
      "loss": 0.2324,
      "num_input_tokens_seen": 25646592,
      "step": 121530
    },
    {
      "epoch": 13.37018701870187,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.00897059874791699,
      "loss": 0.2319,
      "num_input_tokens_seen": 25647584,
      "step": 121535
    },
    {
      "epoch": 13.370737073707371,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.008969280192139611,
      "loss": 0.2309,
      "num_input_tokens_seen": 25648608,
      "step": 121540
    },
    {
      "epoch": 13.371287128712872,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008967961691944481,
      "loss": 0.2314,
      "num_input_tokens_seen": 25649664,
      "step": 121545
    },
    {
      "epoch": 13.371837183718371,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.008966643247343746,
      "loss": 0.2329,
      "num_input_tokens_seen": 25650720,
      "step": 121550
    },
    {
      "epoch": 13.372387238723872,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.008965324858349572,
      "loss": 0.2288,
      "num_input_tokens_seen": 25651776,
      "step": 121555
    },
    {
      "epoch": 13.372937293729374,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.008964006524974097,
      "loss": 0.2335,
      "num_input_tokens_seen": 25652832,
      "step": 121560
    },
    {
      "epoch": 13.373487348734873,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.008962688247229487,
      "loss": 0.2314,
      "num_input_tokens_seen": 25653920,
      "step": 121565
    },
    {
      "epoch": 13.374037403740374,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008961370025127877,
      "loss": 0.2309,
      "num_input_tokens_seen": 25654976,
      "step": 121570
    },
    {
      "epoch": 13.374587458745875,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.00896005185868142,
      "loss": 0.2329,
      "num_input_tokens_seen": 25655968,
      "step": 121575
    },
    {
      "epoch": 13.375137513751374,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.008958733747902273,
      "loss": 0.2288,
      "num_input_tokens_seen": 25656992,
      "step": 121580
    },
    {
      "epoch": 13.375687568756875,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.00895741569280257,
      "loss": 0.2293,
      "num_input_tokens_seen": 25658048,
      "step": 121585
    },
    {
      "epoch": 13.376237623762377,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.008956097693394473,
      "loss": 0.2304,
      "num_input_tokens_seen": 25659104,
      "step": 121590
    },
    {
      "epoch": 13.376787678767876,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.008954779749690124,
      "loss": 0.234,
      "num_input_tokens_seen": 25660192,
      "step": 121595
    },
    {
      "epoch": 13.377337733773377,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.008953461861701663,
      "loss": 0.2304,
      "num_input_tokens_seen": 25661216,
      "step": 121600
    },
    {
      "epoch": 13.377887788778878,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.008952144029441248,
      "loss": 0.2309,
      "num_input_tokens_seen": 25662240,
      "step": 121605
    },
    {
      "epoch": 13.37843784378438,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.008950826252921016,
      "loss": 0.233,
      "num_input_tokens_seen": 25663296,
      "step": 121610
    },
    {
      "epoch": 13.378987898789878,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.008949508532153117,
      "loss": 0.2309,
      "num_input_tokens_seen": 25664352,
      "step": 121615
    },
    {
      "epoch": 13.37953795379538,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0089481908671497,
      "loss": 0.2303,
      "num_input_tokens_seen": 25665376,
      "step": 121620
    },
    {
      "epoch": 13.38008800880088,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.008946873257922898,
      "loss": 0.2298,
      "num_input_tokens_seen": 25666464,
      "step": 121625
    },
    {
      "epoch": 13.38063806380638,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008945555704484858,
      "loss": 0.2309,
      "num_input_tokens_seen": 25667520,
      "step": 121630
    },
    {
      "epoch": 13.381188118811881,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.008944238206847734,
      "loss": 0.2303,
      "num_input_tokens_seen": 25668512,
      "step": 121635
    },
    {
      "epoch": 13.381738173817382,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008942920765023654,
      "loss": 0.2324,
      "num_input_tokens_seen": 25669664,
      "step": 121640
    },
    {
      "epoch": 13.382288228822881,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.00894160337902477,
      "loss": 0.234,
      "num_input_tokens_seen": 25670816,
      "step": 121645
    },
    {
      "epoch": 13.382838283828383,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008940286048863218,
      "loss": 0.2324,
      "num_input_tokens_seen": 25671840,
      "step": 121650
    },
    {
      "epoch": 13.383388338833884,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.008938968774551144,
      "loss": 0.2303,
      "num_input_tokens_seen": 25673024,
      "step": 121655
    },
    {
      "epoch": 13.383938393839385,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.008937651556100688,
      "loss": 0.2324,
      "num_input_tokens_seen": 25674112,
      "step": 121660
    },
    {
      "epoch": 13.384488448844884,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.008936334393523981,
      "loss": 0.2319,
      "num_input_tokens_seen": 25675168,
      "step": 121665
    },
    {
      "epoch": 13.385038503850385,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008935017286833182,
      "loss": 0.2303,
      "num_input_tokens_seen": 25676192,
      "step": 121670
    },
    {
      "epoch": 13.385588558855886,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.008933700236040405,
      "loss": 0.2335,
      "num_input_tokens_seen": 25677216,
      "step": 121675
    },
    {
      "epoch": 13.386138613861386,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.008932383241157815,
      "loss": 0.233,
      "num_input_tokens_seen": 25678304,
      "step": 121680
    },
    {
      "epoch": 13.386688668866887,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008931066302197529,
      "loss": 0.2335,
      "num_input_tokens_seen": 25679360,
      "step": 121685
    },
    {
      "epoch": 13.387238723872388,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.008929749419171691,
      "loss": 0.2288,
      "num_input_tokens_seen": 25680448,
      "step": 121690
    },
    {
      "epoch": 13.387788778877887,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.00892843259209245,
      "loss": 0.2314,
      "num_input_tokens_seen": 25681504,
      "step": 121695
    },
    {
      "epoch": 13.388338833883388,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.008927115820971922,
      "loss": 0.2324,
      "num_input_tokens_seen": 25682528,
      "step": 121700
    },
    {
      "epoch": 13.38888888888889,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.008925799105822259,
      "loss": 0.234,
      "num_input_tokens_seen": 25683616,
      "step": 121705
    },
    {
      "epoch": 13.389438943894389,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.008924482446655592,
      "loss": 0.2319,
      "num_input_tokens_seen": 25684640,
      "step": 121710
    },
    {
      "epoch": 13.38998899889989,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00892316584348405,
      "loss": 0.2324,
      "num_input_tokens_seen": 25685664,
      "step": 121715
    },
    {
      "epoch": 13.39053905390539,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008921849296319775,
      "loss": 0.2308,
      "num_input_tokens_seen": 25686752,
      "step": 121720
    },
    {
      "epoch": 13.391089108910892,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.008920532805174896,
      "loss": 0.2314,
      "num_input_tokens_seen": 25687808,
      "step": 121725
    },
    {
      "epoch": 13.391639163916391,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.008919216370061555,
      "loss": 0.2319,
      "num_input_tokens_seen": 25688864,
      "step": 121730
    },
    {
      "epoch": 13.392189218921892,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.008917899990991875,
      "loss": 0.2314,
      "num_input_tokens_seen": 25689856,
      "step": 121735
    },
    {
      "epoch": 13.392739273927393,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.00891658366797799,
      "loss": 0.2319,
      "num_input_tokens_seen": 25690912,
      "step": 121740
    },
    {
      "epoch": 13.393289328932893,
      "grad_norm": 0.0029144287109375,
      "learning_rate": 0.008915267401032037,
      "loss": 0.2308,
      "num_input_tokens_seen": 25691872,
      "step": 121745
    },
    {
      "epoch": 13.393839383938394,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008913951190166143,
      "loss": 0.2298,
      "num_input_tokens_seen": 25692928,
      "step": 121750
    },
    {
      "epoch": 13.394389438943895,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.008912635035392445,
      "loss": 0.2324,
      "num_input_tokens_seen": 25693952,
      "step": 121755
    },
    {
      "epoch": 13.394939493949394,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008911318936723065,
      "loss": 0.233,
      "num_input_tokens_seen": 25695040,
      "step": 121760
    },
    {
      "epoch": 13.395489548954895,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.008910002894170134,
      "loss": 0.2314,
      "num_input_tokens_seen": 25696096,
      "step": 121765
    },
    {
      "epoch": 13.396039603960396,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.008908686907745792,
      "loss": 0.2304,
      "num_input_tokens_seen": 25697152,
      "step": 121770
    },
    {
      "epoch": 13.396589658965897,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.008907370977462152,
      "loss": 0.2319,
      "num_input_tokens_seen": 25698272,
      "step": 121775
    },
    {
      "epoch": 13.397139713971397,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.008906055103331355,
      "loss": 0.2319,
      "num_input_tokens_seen": 25699360,
      "step": 121780
    },
    {
      "epoch": 13.397689768976898,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.008904739285365526,
      "loss": 0.2309,
      "num_input_tokens_seen": 25700352,
      "step": 121785
    },
    {
      "epoch": 13.398239823982399,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.008903423523576781,
      "loss": 0.233,
      "num_input_tokens_seen": 25701408,
      "step": 121790
    },
    {
      "epoch": 13.398789878987898,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.008902107817977265,
      "loss": 0.2298,
      "num_input_tokens_seen": 25702432,
      "step": 121795
    },
    {
      "epoch": 13.3993399339934,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.00890079216857909,
      "loss": 0.2298,
      "num_input_tokens_seen": 25703456,
      "step": 121800
    },
    {
      "epoch": 13.3998899889989,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.008899476575394388,
      "loss": 0.2304,
      "num_input_tokens_seen": 25704544,
      "step": 121805
    },
    {
      "epoch": 13.4004400440044,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.008898161038435288,
      "loss": 0.2304,
      "num_input_tokens_seen": 25705632,
      "step": 121810
    },
    {
      "epoch": 13.400990099009901,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.008896845557713902,
      "loss": 0.2319,
      "num_input_tokens_seen": 25706656,
      "step": 121815
    },
    {
      "epoch": 13.401540154015402,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.008895530133242367,
      "loss": 0.2309,
      "num_input_tokens_seen": 25707744,
      "step": 121820
    },
    {
      "epoch": 13.402090209020901,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008894214765032804,
      "loss": 0.2319,
      "num_input_tokens_seen": 25708800,
      "step": 121825
    },
    {
      "epoch": 13.402640264026402,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008892899453097326,
      "loss": 0.2298,
      "num_input_tokens_seen": 25709888,
      "step": 121830
    },
    {
      "epoch": 13.403190319031903,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.008891584197448066,
      "loss": 0.2314,
      "num_input_tokens_seen": 25711008,
      "step": 121835
    },
    {
      "epoch": 13.403740374037405,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.008890268998097143,
      "loss": 0.2309,
      "num_input_tokens_seen": 25712032,
      "step": 121840
    },
    {
      "epoch": 13.404290429042904,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.008888953855056683,
      "loss": 0.2309,
      "num_input_tokens_seen": 25713088,
      "step": 121845
    },
    {
      "epoch": 13.404840484048405,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.008887638768338802,
      "loss": 0.2319,
      "num_input_tokens_seen": 25714112,
      "step": 121850
    },
    {
      "epoch": 13.405390539053906,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.008886323737955617,
      "loss": 0.2324,
      "num_input_tokens_seen": 25715200,
      "step": 121855
    },
    {
      "epoch": 13.405940594059405,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008885008763919256,
      "loss": 0.2324,
      "num_input_tokens_seen": 25716256,
      "step": 121860
    },
    {
      "epoch": 13.406490649064907,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008883693846241832,
      "loss": 0.2298,
      "num_input_tokens_seen": 25717376,
      "step": 121865
    },
    {
      "epoch": 13.407040704070408,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.008882378984935472,
      "loss": 0.2319,
      "num_input_tokens_seen": 25718464,
      "step": 121870
    },
    {
      "epoch": 13.407590759075907,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.008881064180012286,
      "loss": 0.2293,
      "num_input_tokens_seen": 25719584,
      "step": 121875
    },
    {
      "epoch": 13.408140814081408,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.008879749431484394,
      "loss": 0.2298,
      "num_input_tokens_seen": 25720608,
      "step": 121880
    },
    {
      "epoch": 13.408690869086909,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.00887843473936392,
      "loss": 0.2324,
      "num_input_tokens_seen": 25721632,
      "step": 121885
    },
    {
      "epoch": 13.409240924092408,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.00887712010366297,
      "loss": 0.2345,
      "num_input_tokens_seen": 25722720,
      "step": 121890
    },
    {
      "epoch": 13.40979097909791,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.008875805524393668,
      "loss": 0.2319,
      "num_input_tokens_seen": 25723776,
      "step": 121895
    },
    {
      "epoch": 13.41034103410341,
      "grad_norm": 0.000698089599609375,
      "learning_rate": 0.008874491001568131,
      "loss": 0.2309,
      "num_input_tokens_seen": 25724832,
      "step": 121900
    },
    {
      "epoch": 13.410891089108912,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008873176535198464,
      "loss": 0.2319,
      "num_input_tokens_seen": 25725888,
      "step": 121905
    },
    {
      "epoch": 13.411441144114411,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.008871862125296793,
      "loss": 0.2314,
      "num_input_tokens_seen": 25726976,
      "step": 121910
    },
    {
      "epoch": 13.411991199119912,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.008870547771875228,
      "loss": 0.2298,
      "num_input_tokens_seen": 25727968,
      "step": 121915
    },
    {
      "epoch": 13.412541254125413,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.008869233474945882,
      "loss": 0.2319,
      "num_input_tokens_seen": 25729120,
      "step": 121920
    },
    {
      "epoch": 13.413091309130913,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008867919234520875,
      "loss": 0.2293,
      "num_input_tokens_seen": 25730144,
      "step": 121925
    },
    {
      "epoch": 13.413641364136414,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.008866605050612304,
      "loss": 0.2304,
      "num_input_tokens_seen": 25731200,
      "step": 121930
    },
    {
      "epoch": 13.414191419141915,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0088652909232323,
      "loss": 0.2309,
      "num_input_tokens_seen": 25732256,
      "step": 121935
    },
    {
      "epoch": 13.414741474147414,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008863976852392957,
      "loss": 0.2314,
      "num_input_tokens_seen": 25733312,
      "step": 121940
    },
    {
      "epoch": 13.415291529152915,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.008862662838106406,
      "loss": 0.2319,
      "num_input_tokens_seen": 25734368,
      "step": 121945
    },
    {
      "epoch": 13.415841584158416,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00886134888038474,
      "loss": 0.233,
      "num_input_tokens_seen": 25735424,
      "step": 121950
    },
    {
      "epoch": 13.416391639163916,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.008860034979240074,
      "loss": 0.2303,
      "num_input_tokens_seen": 25736448,
      "step": 121955
    },
    {
      "epoch": 13.416941694169417,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.008858721134684527,
      "loss": 0.2324,
      "num_input_tokens_seen": 25737504,
      "step": 121960
    },
    {
      "epoch": 13.417491749174918,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.008857407346730193,
      "loss": 0.233,
      "num_input_tokens_seen": 25738560,
      "step": 121965
    },
    {
      "epoch": 13.418041804180419,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.008856093615389192,
      "loss": 0.2319,
      "num_input_tokens_seen": 25739680,
      "step": 121970
    },
    {
      "epoch": 13.418591859185918,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.008854779940673626,
      "loss": 0.2314,
      "num_input_tokens_seen": 25740768,
      "step": 121975
    },
    {
      "epoch": 13.41914191419142,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008853466322595602,
      "loss": 0.2298,
      "num_input_tokens_seen": 25741856,
      "step": 121980
    },
    {
      "epoch": 13.41969196919692,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.008852152761167235,
      "loss": 0.2309,
      "num_input_tokens_seen": 25742848,
      "step": 121985
    },
    {
      "epoch": 13.42024202420242,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00885083925640062,
      "loss": 0.2298,
      "num_input_tokens_seen": 25743904,
      "step": 121990
    },
    {
      "epoch": 13.42079207920792,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008849525808307874,
      "loss": 0.2303,
      "num_input_tokens_seen": 25744992,
      "step": 121995
    },
    {
      "epoch": 13.421342134213422,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0088482124169011,
      "loss": 0.2298,
      "num_input_tokens_seen": 25746016,
      "step": 122000
    },
    {
      "epoch": 13.421892189218921,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.008846899082192392,
      "loss": 0.2314,
      "num_input_tokens_seen": 25747104,
      "step": 122005
    },
    {
      "epoch": 13.422442244224422,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.008845585804193867,
      "loss": 0.2324,
      "num_input_tokens_seen": 25748192,
      "step": 122010
    },
    {
      "epoch": 13.422992299229923,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.00884427258291763,
      "loss": 0.2319,
      "num_input_tokens_seen": 25749280,
      "step": 122015
    },
    {
      "epoch": 13.423542354235423,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.008842959418375767,
      "loss": 0.2309,
      "num_input_tokens_seen": 25750368,
      "step": 122020
    },
    {
      "epoch": 13.424092409240924,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.0088416463105804,
      "loss": 0.2303,
      "num_input_tokens_seen": 25751392,
      "step": 122025
    },
    {
      "epoch": 13.424642464246425,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.00884033325954362,
      "loss": 0.2288,
      "num_input_tokens_seen": 25752448,
      "step": 122030
    },
    {
      "epoch": 13.425192519251926,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.008839020265277536,
      "loss": 0.2324,
      "num_input_tokens_seen": 25753504,
      "step": 122035
    },
    {
      "epoch": 13.425742574257425,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.008837707327794249,
      "loss": 0.2324,
      "num_input_tokens_seen": 25754528,
      "step": 122040
    },
    {
      "epoch": 13.426292629262926,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008836394447105848,
      "loss": 0.2309,
      "num_input_tokens_seen": 25755584,
      "step": 122045
    },
    {
      "epoch": 13.426842684268427,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.008835081623224448,
      "loss": 0.2319,
      "num_input_tokens_seen": 25756640,
      "step": 122050
    },
    {
      "epoch": 13.427392739273927,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.008833768856162138,
      "loss": 0.2303,
      "num_input_tokens_seen": 25757664,
      "step": 122055
    },
    {
      "epoch": 13.427942794279428,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.008832456145931027,
      "loss": 0.2324,
      "num_input_tokens_seen": 25758688,
      "step": 122060
    },
    {
      "epoch": 13.428492849284929,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.008831143492543207,
      "loss": 0.2293,
      "num_input_tokens_seen": 25759776,
      "step": 122065
    },
    {
      "epoch": 13.429042904290428,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.008829830896010771,
      "loss": 0.233,
      "num_input_tokens_seen": 25760864,
      "step": 122070
    },
    {
      "epoch": 13.42959295929593,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008828518356345834,
      "loss": 0.2303,
      "num_input_tokens_seen": 25761888,
      "step": 122075
    },
    {
      "epoch": 13.43014301430143,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.008827205873560473,
      "loss": 0.2303,
      "num_input_tokens_seen": 25762976,
      "step": 122080
    },
    {
      "epoch": 13.430693069306932,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.008825893447666797,
      "loss": 0.2324,
      "num_input_tokens_seen": 25764000,
      "step": 122085
    },
    {
      "epoch": 13.43124312431243,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0088245810786769,
      "loss": 0.2304,
      "num_input_tokens_seen": 25765056,
      "step": 122090
    },
    {
      "epoch": 13.431793179317932,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.008823268766602867,
      "loss": 0.2324,
      "num_input_tokens_seen": 25766144,
      "step": 122095
    },
    {
      "epoch": 13.432343234323433,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.008821956511456813,
      "loss": 0.2303,
      "num_input_tokens_seen": 25767232,
      "step": 122100
    },
    {
      "epoch": 13.432893289328932,
      "grad_norm": 0.005859375,
      "learning_rate": 0.008820644313250813,
      "loss": 0.2298,
      "num_input_tokens_seen": 25768320,
      "step": 122105
    },
    {
      "epoch": 13.433443344334433,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.008819332171996975,
      "loss": 0.2314,
      "num_input_tokens_seen": 25769376,
      "step": 122110
    },
    {
      "epoch": 13.433993399339935,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 0.00881802008770739,
      "loss": 0.2293,
      "num_input_tokens_seen": 25770400,
      "step": 122115
    },
    {
      "epoch": 13.434543454345434,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.008816708060394139,
      "loss": 0.2309,
      "num_input_tokens_seen": 25771456,
      "step": 122120
    },
    {
      "epoch": 13.435093509350935,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008815396090069325,
      "loss": 0.2345,
      "num_input_tokens_seen": 25772544,
      "step": 122125
    },
    {
      "epoch": 13.435643564356436,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.008814084176745035,
      "loss": 0.2308,
      "num_input_tokens_seen": 25773568,
      "step": 122130
    },
    {
      "epoch": 13.436193619361935,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.008812772320433369,
      "loss": 0.2335,
      "num_input_tokens_seen": 25774624,
      "step": 122135
    },
    {
      "epoch": 13.436743674367436,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.008811460521146411,
      "loss": 0.233,
      "num_input_tokens_seen": 25775680,
      "step": 122140
    },
    {
      "epoch": 13.437293729372938,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.008810148778896246,
      "loss": 0.2293,
      "num_input_tokens_seen": 25776736,
      "step": 122145
    },
    {
      "epoch": 13.437843784378439,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008808837093694977,
      "loss": 0.2309,
      "num_input_tokens_seen": 25777856,
      "step": 122150
    },
    {
      "epoch": 13.438393839383938,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 0.008807525465554676,
      "loss": 0.2335,
      "num_input_tokens_seen": 25778912,
      "step": 122155
    },
    {
      "epoch": 13.438943894389439,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.008806213894487451,
      "loss": 0.2324,
      "num_input_tokens_seen": 25779872,
      "step": 122160
    },
    {
      "epoch": 13.43949394939494,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.008804902380505379,
      "loss": 0.2324,
      "num_input_tokens_seen": 25780864,
      "step": 122165
    },
    {
      "epoch": 13.44004400440044,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.008803590923620541,
      "loss": 0.233,
      "num_input_tokens_seen": 25781888,
      "step": 122170
    },
    {
      "epoch": 13.44059405940594,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008802279523845043,
      "loss": 0.2298,
      "num_input_tokens_seen": 25782912,
      "step": 122175
    },
    {
      "epoch": 13.441144114411442,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.008800968181190954,
      "loss": 0.2304,
      "num_input_tokens_seen": 25784000,
      "step": 122180
    },
    {
      "epoch": 13.441694169416941,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008799656895670369,
      "loss": 0.2335,
      "num_input_tokens_seen": 25785056,
      "step": 122185
    },
    {
      "epoch": 13.442244224422442,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008798345667295374,
      "loss": 0.2303,
      "num_input_tokens_seen": 25786112,
      "step": 122190
    },
    {
      "epoch": 13.442794279427943,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.008797034496078044,
      "loss": 0.2303,
      "num_input_tokens_seen": 25787168,
      "step": 122195
    },
    {
      "epoch": 13.443344334433444,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.008795723382030475,
      "loss": 0.234,
      "num_input_tokens_seen": 25788224,
      "step": 122200
    },
    {
      "epoch": 13.443894389438944,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.008794412325164749,
      "loss": 0.2324,
      "num_input_tokens_seen": 25789312,
      "step": 122205
    },
    {
      "epoch": 13.444444444444445,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008793101325492941,
      "loss": 0.2324,
      "num_input_tokens_seen": 25790400,
      "step": 122210
    },
    {
      "epoch": 13.444994499449946,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.008791790383027144,
      "loss": 0.233,
      "num_input_tokens_seen": 25791488,
      "step": 122215
    },
    {
      "epoch": 13.445544554455445,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.008790479497779431,
      "loss": 0.2314,
      "num_input_tokens_seen": 25792544,
      "step": 122220
    },
    {
      "epoch": 13.446094609460946,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008789168669761892,
      "loss": 0.2303,
      "num_input_tokens_seen": 25793632,
      "step": 122225
    },
    {
      "epoch": 13.446644664466447,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008787857898986609,
      "loss": 0.233,
      "num_input_tokens_seen": 25794688,
      "step": 122230
    },
    {
      "epoch": 13.447194719471947,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008786547185465652,
      "loss": 0.2309,
      "num_input_tokens_seen": 25795808,
      "step": 122235
    },
    {
      "epoch": 13.447744774477448,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.008785236529211113,
      "loss": 0.2293,
      "num_input_tokens_seen": 25796928,
      "step": 122240
    },
    {
      "epoch": 13.448294829482949,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.008783925930235064,
      "loss": 0.2303,
      "num_input_tokens_seen": 25798048,
      "step": 122245
    },
    {
      "epoch": 13.448844884488448,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.00878261538854959,
      "loss": 0.2277,
      "num_input_tokens_seen": 25799072,
      "step": 122250
    },
    {
      "epoch": 13.44939493949395,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.008781304904166764,
      "loss": 0.2351,
      "num_input_tokens_seen": 25800128,
      "step": 122255
    },
    {
      "epoch": 13.44994499449945,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.008779994477098665,
      "loss": 0.2324,
      "num_input_tokens_seen": 25801216,
      "step": 122260
    },
    {
      "epoch": 13.450495049504951,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.008778684107357378,
      "loss": 0.2309,
      "num_input_tokens_seen": 25802240,
      "step": 122265
    },
    {
      "epoch": 13.45104510451045,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008777373794954968,
      "loss": 0.2303,
      "num_input_tokens_seen": 25803232,
      "step": 122270
    },
    {
      "epoch": 13.451595159515952,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.008776063539903522,
      "loss": 0.2309,
      "num_input_tokens_seen": 25804256,
      "step": 122275
    },
    {
      "epoch": 13.452145214521453,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00877475334221511,
      "loss": 0.2309,
      "num_input_tokens_seen": 25805312,
      "step": 122280
    },
    {
      "epoch": 13.452695269526952,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.008773443201901808,
      "loss": 0.2293,
      "num_input_tokens_seen": 25806368,
      "step": 122285
    },
    {
      "epoch": 13.453245324532453,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.008772133118975697,
      "loss": 0.2324,
      "num_input_tokens_seen": 25807488,
      "step": 122290
    },
    {
      "epoch": 13.453795379537954,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.008770823093448841,
      "loss": 0.2304,
      "num_input_tokens_seen": 25808480,
      "step": 122295
    },
    {
      "epoch": 13.454345434543454,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.008769513125333323,
      "loss": 0.2304,
      "num_input_tokens_seen": 25809568,
      "step": 122300
    },
    {
      "epoch": 13.454895489548955,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008768203214641215,
      "loss": 0.2293,
      "num_input_tokens_seen": 25810656,
      "step": 122305
    },
    {
      "epoch": 13.455445544554456,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.008766893361384582,
      "loss": 0.2298,
      "num_input_tokens_seen": 25811744,
      "step": 122310
    },
    {
      "epoch": 13.455995599559955,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.008765583565575505,
      "loss": 0.233,
      "num_input_tokens_seen": 25812832,
      "step": 122315
    },
    {
      "epoch": 13.456545654565456,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008764273827226051,
      "loss": 0.2329,
      "num_input_tokens_seen": 25813920,
      "step": 122320
    },
    {
      "epoch": 13.457095709570957,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.008762964146348296,
      "loss": 0.2314,
      "num_input_tokens_seen": 25815008,
      "step": 122325
    },
    {
      "epoch": 13.457645764576458,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008761654522954306,
      "loss": 0.2303,
      "num_input_tokens_seen": 25816064,
      "step": 122330
    },
    {
      "epoch": 13.458195819581958,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.008760344957056151,
      "loss": 0.2324,
      "num_input_tokens_seen": 25817088,
      "step": 122335
    },
    {
      "epoch": 13.458745874587459,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008759035448665903,
      "loss": 0.2303,
      "num_input_tokens_seen": 25818144,
      "step": 122340
    },
    {
      "epoch": 13.45929592959296,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00875772599779563,
      "loss": 0.2303,
      "num_input_tokens_seen": 25819232,
      "step": 122345
    },
    {
      "epoch": 13.45984598459846,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.008756416604457406,
      "loss": 0.2319,
      "num_input_tokens_seen": 25820320,
      "step": 122350
    },
    {
      "epoch": 13.46039603960396,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.008755107268663295,
      "loss": 0.2309,
      "num_input_tokens_seen": 25821376,
      "step": 122355
    },
    {
      "epoch": 13.460946094609461,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.008753797990425357,
      "loss": 0.2319,
      "num_input_tokens_seen": 25822464,
      "step": 122360
    },
    {
      "epoch": 13.46149614961496,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.008752488769755666,
      "loss": 0.2293,
      "num_input_tokens_seen": 25823456,
      "step": 122365
    },
    {
      "epoch": 13.462046204620462,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.008751179606666288,
      "loss": 0.2314,
      "num_input_tokens_seen": 25824544,
      "step": 122370
    },
    {
      "epoch": 13.462596259625963,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.008749870501169295,
      "loss": 0.2303,
      "num_input_tokens_seen": 25825536,
      "step": 122375
    },
    {
      "epoch": 13.463146314631462,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.008748561453276746,
      "loss": 0.2314,
      "num_input_tokens_seen": 25826592,
      "step": 122380
    },
    {
      "epoch": 13.463696369636963,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0087472524630007,
      "loss": 0.2304,
      "num_input_tokens_seen": 25827712,
      "step": 122385
    },
    {
      "epoch": 13.464246424642464,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008745943530353238,
      "loss": 0.2288,
      "num_input_tokens_seen": 25828736,
      "step": 122390
    },
    {
      "epoch": 13.464796479647966,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.008744634655346402,
      "loss": 0.2324,
      "num_input_tokens_seen": 25829760,
      "step": 122395
    },
    {
      "epoch": 13.465346534653465,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00874332583799227,
      "loss": 0.2293,
      "num_input_tokens_seen": 25830848,
      "step": 122400
    },
    {
      "epoch": 13.465896589658966,
      "grad_norm": 0.0027923583984375,
      "learning_rate": 0.008742017078302905,
      "loss": 0.2314,
      "num_input_tokens_seen": 25831904,
      "step": 122405
    },
    {
      "epoch": 13.466446644664467,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008740708376290362,
      "loss": 0.2314,
      "num_input_tokens_seen": 25832992,
      "step": 122410
    },
    {
      "epoch": 13.466996699669966,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.008739399731966711,
      "loss": 0.2299,
      "num_input_tokens_seen": 25834080,
      "step": 122415
    },
    {
      "epoch": 13.467546754675467,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.008738091145344004,
      "loss": 0.2335,
      "num_input_tokens_seen": 25835136,
      "step": 122420
    },
    {
      "epoch": 13.468096809680969,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.008736782616434305,
      "loss": 0.2309,
      "num_input_tokens_seen": 25836192,
      "step": 122425
    },
    {
      "epoch": 13.468646864686468,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00873547414524968,
      "loss": 0.2319,
      "num_input_tokens_seen": 25837248,
      "step": 122430
    },
    {
      "epoch": 13.469196919691969,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00873416573180218,
      "loss": 0.2298,
      "num_input_tokens_seen": 25838304,
      "step": 122435
    },
    {
      "epoch": 13.46974697469747,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.008732857376103874,
      "loss": 0.2325,
      "num_input_tokens_seen": 25839392,
      "step": 122440
    },
    {
      "epoch": 13.47029702970297,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.008731549078166813,
      "loss": 0.2314,
      "num_input_tokens_seen": 25840448,
      "step": 122445
    },
    {
      "epoch": 13.47084708470847,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.00873024083800305,
      "loss": 0.2319,
      "num_input_tokens_seen": 25841472,
      "step": 122450
    },
    {
      "epoch": 13.471397139713972,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.00872893265562465,
      "loss": 0.2304,
      "num_input_tokens_seen": 25842560,
      "step": 122455
    },
    {
      "epoch": 13.471947194719473,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008727624531043669,
      "loss": 0.2324,
      "num_input_tokens_seen": 25843616,
      "step": 122460
    },
    {
      "epoch": 13.472497249724972,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.008726316464272165,
      "loss": 0.2324,
      "num_input_tokens_seen": 25844576,
      "step": 122465
    },
    {
      "epoch": 13.473047304730473,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.008725008455322198,
      "loss": 0.2304,
      "num_input_tokens_seen": 25845664,
      "step": 122470
    },
    {
      "epoch": 13.473597359735974,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.008723700504205806,
      "loss": 0.2298,
      "num_input_tokens_seen": 25846720,
      "step": 122475
    },
    {
      "epoch": 13.474147414741473,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008722392610935057,
      "loss": 0.2319,
      "num_input_tokens_seen": 25847712,
      "step": 122480
    },
    {
      "epoch": 13.474697469746975,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008721084775522001,
      "loss": 0.2325,
      "num_input_tokens_seen": 25848800,
      "step": 122485
    },
    {
      "epoch": 13.475247524752476,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.008719776997978702,
      "loss": 0.2309,
      "num_input_tokens_seen": 25849856,
      "step": 122490
    },
    {
      "epoch": 13.475797579757975,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.008718469278317202,
      "loss": 0.2324,
      "num_input_tokens_seen": 25850880,
      "step": 122495
    },
    {
      "epoch": 13.476347634763476,
      "grad_norm": 0.002593994140625,
      "learning_rate": 0.008717161616549554,
      "loss": 0.2314,
      "num_input_tokens_seen": 25851968,
      "step": 122500
    },
    {
      "epoch": 13.476897689768977,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008715854012687816,
      "loss": 0.2319,
      "num_input_tokens_seen": 25852992,
      "step": 122505
    },
    {
      "epoch": 13.477447744774478,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008714546466744032,
      "loss": 0.2324,
      "num_input_tokens_seen": 25854048,
      "step": 122510
    },
    {
      "epoch": 13.477997799779978,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008713238978730256,
      "loss": 0.2324,
      "num_input_tokens_seen": 25855040,
      "step": 122515
    },
    {
      "epoch": 13.478547854785479,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008711931548658547,
      "loss": 0.2314,
      "num_input_tokens_seen": 25856064,
      "step": 122520
    },
    {
      "epoch": 13.47909790979098,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.00871062417654094,
      "loss": 0.2314,
      "num_input_tokens_seen": 25857120,
      "step": 122525
    },
    {
      "epoch": 13.479647964796479,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.008709316862389498,
      "loss": 0.2309,
      "num_input_tokens_seen": 25858208,
      "step": 122530
    },
    {
      "epoch": 13.48019801980198,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.008708009606216259,
      "loss": 0.2304,
      "num_input_tokens_seen": 25859232,
      "step": 122535
    },
    {
      "epoch": 13.480748074807481,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.008706702408033274,
      "loss": 0.2298,
      "num_input_tokens_seen": 25860256,
      "step": 122540
    },
    {
      "epoch": 13.48129812981298,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008705395267852602,
      "loss": 0.234,
      "num_input_tokens_seen": 25861280,
      "step": 122545
    },
    {
      "epoch": 13.481848184818482,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.008704088185686274,
      "loss": 0.2319,
      "num_input_tokens_seen": 25862336,
      "step": 122550
    },
    {
      "epoch": 13.482398239823983,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00870278116154635,
      "loss": 0.2324,
      "num_input_tokens_seen": 25863488,
      "step": 122555
    },
    {
      "epoch": 13.482948294829482,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.008701474195444864,
      "loss": 0.2304,
      "num_input_tokens_seen": 25864512,
      "step": 122560
    },
    {
      "epoch": 13.483498349834983,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.008700167287393874,
      "loss": 0.2319,
      "num_input_tokens_seen": 25865600,
      "step": 122565
    },
    {
      "epoch": 13.484048404840484,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008698860437405417,
      "loss": 0.2309,
      "num_input_tokens_seen": 25866720,
      "step": 122570
    },
    {
      "epoch": 13.484598459845985,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.008697553645491535,
      "loss": 0.2309,
      "num_input_tokens_seen": 25867712,
      "step": 122575
    },
    {
      "epoch": 13.485148514851485,
      "grad_norm": 0.000720977783203125,
      "learning_rate": 0.008696246911664285,
      "loss": 0.2319,
      "num_input_tokens_seen": 25868736,
      "step": 122580
    },
    {
      "epoch": 13.485698569856986,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.008694940235935696,
      "loss": 0.2319,
      "num_input_tokens_seen": 25869792,
      "step": 122585
    },
    {
      "epoch": 13.486248624862487,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.008693633618317822,
      "loss": 0.2298,
      "num_input_tokens_seen": 25870880,
      "step": 122590
    },
    {
      "epoch": 13.486798679867986,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.008692327058822696,
      "loss": 0.2304,
      "num_input_tokens_seen": 25872064,
      "step": 122595
    },
    {
      "epoch": 13.487348734873487,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.008691020557462365,
      "loss": 0.2303,
      "num_input_tokens_seen": 25873056,
      "step": 122600
    },
    {
      "epoch": 13.487898789878988,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008689714114248875,
      "loss": 0.2314,
      "num_input_tokens_seen": 25874112,
      "step": 122605
    },
    {
      "epoch": 13.488448844884488,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.008688407729194266,
      "loss": 0.2303,
      "num_input_tokens_seen": 25875168,
      "step": 122610
    },
    {
      "epoch": 13.488998899889989,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.008687101402310564,
      "loss": 0.2293,
      "num_input_tokens_seen": 25876192,
      "step": 122615
    },
    {
      "epoch": 13.48954895489549,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.008685795133609826,
      "loss": 0.2304,
      "num_input_tokens_seen": 25877248,
      "step": 122620
    },
    {
      "epoch": 13.490099009900991,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00868448892310408,
      "loss": 0.2319,
      "num_input_tokens_seen": 25878336,
      "step": 122625
    },
    {
      "epoch": 13.49064906490649,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008683182770805369,
      "loss": 0.2293,
      "num_input_tokens_seen": 25879360,
      "step": 122630
    },
    {
      "epoch": 13.491199119911991,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008681876676725734,
      "loss": 0.2308,
      "num_input_tokens_seen": 25880448,
      "step": 122635
    },
    {
      "epoch": 13.491749174917492,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.008680570640877208,
      "loss": 0.2319,
      "num_input_tokens_seen": 25881504,
      "step": 122640
    },
    {
      "epoch": 13.492299229922992,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.008679264663271834,
      "loss": 0.2293,
      "num_input_tokens_seen": 25882560,
      "step": 122645
    },
    {
      "epoch": 13.492849284928493,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008677958743921639,
      "loss": 0.2319,
      "num_input_tokens_seen": 25883552,
      "step": 122650
    },
    {
      "epoch": 13.493399339933994,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.008676652882838665,
      "loss": 0.2314,
      "num_input_tokens_seen": 25884576,
      "step": 122655
    },
    {
      "epoch": 13.493949394939493,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.008675347080034953,
      "loss": 0.2319,
      "num_input_tokens_seen": 25885664,
      "step": 122660
    },
    {
      "epoch": 13.494499449944994,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.008674041335522525,
      "loss": 0.2319,
      "num_input_tokens_seen": 25886656,
      "step": 122665
    },
    {
      "epoch": 13.495049504950495,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00867273564931343,
      "loss": 0.2314,
      "num_input_tokens_seen": 25887680,
      "step": 122670
    },
    {
      "epoch": 13.495599559955995,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008671430021419686,
      "loss": 0.2325,
      "num_input_tokens_seen": 25888736,
      "step": 122675
    },
    {
      "epoch": 13.496149614961496,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.008670124451853344,
      "loss": 0.2298,
      "num_input_tokens_seen": 25889760,
      "step": 122680
    },
    {
      "epoch": 13.496699669966997,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00866881894062642,
      "loss": 0.2309,
      "num_input_tokens_seen": 25890784,
      "step": 122685
    },
    {
      "epoch": 13.497249724972498,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.008667513487750954,
      "loss": 0.233,
      "num_input_tokens_seen": 25891872,
      "step": 122690
    },
    {
      "epoch": 13.497799779977997,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.008666208093238984,
      "loss": 0.233,
      "num_input_tokens_seen": 25892960,
      "step": 122695
    },
    {
      "epoch": 13.498349834983498,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008664902757102529,
      "loss": 0.233,
      "num_input_tokens_seen": 25893984,
      "step": 122700
    },
    {
      "epoch": 13.498899889989,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00866359747935363,
      "loss": 0.2319,
      "num_input_tokens_seen": 25895072,
      "step": 122705
    },
    {
      "epoch": 13.499449944994499,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.008662292260004308,
      "loss": 0.2335,
      "num_input_tokens_seen": 25896128,
      "step": 122710
    },
    {
      "epoch": 13.5,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.008660987099066595,
      "loss": 0.2335,
      "num_input_tokens_seen": 25897216,
      "step": 122715
    },
    {
      "epoch": 13.500550055005501,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00865968199655253,
      "loss": 0.2314,
      "num_input_tokens_seen": 25898272,
      "step": 122720
    },
    {
      "epoch": 13.501100110011,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008658376952474126,
      "loss": 0.2309,
      "num_input_tokens_seen": 25899328,
      "step": 122725
    },
    {
      "epoch": 13.501650165016502,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.008657071966843427,
      "loss": 0.2303,
      "num_input_tokens_seen": 25900384,
      "step": 122730
    },
    {
      "epoch": 13.502200220022003,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.008655767039672451,
      "loss": 0.233,
      "num_input_tokens_seen": 25901536,
      "step": 122735
    },
    {
      "epoch": 13.502750275027502,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00865446217097322,
      "loss": 0.2314,
      "num_input_tokens_seen": 25902528,
      "step": 122740
    },
    {
      "epoch": 13.503300330033003,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.008653157360757766,
      "loss": 0.2298,
      "num_input_tokens_seen": 25903616,
      "step": 122745
    },
    {
      "epoch": 13.503850385038504,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.008651852609038113,
      "loss": 0.2324,
      "num_input_tokens_seen": 25904640,
      "step": 122750
    },
    {
      "epoch": 13.504400440044005,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.008650547915826298,
      "loss": 0.2309,
      "num_input_tokens_seen": 25905696,
      "step": 122755
    },
    {
      "epoch": 13.504950495049505,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.008649243281134334,
      "loss": 0.2309,
      "num_input_tokens_seen": 25906752,
      "step": 122760
    },
    {
      "epoch": 13.505500550055006,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.008647938704974241,
      "loss": 0.234,
      "num_input_tokens_seen": 25907776,
      "step": 122765
    },
    {
      "epoch": 13.506050605060507,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008646634187358051,
      "loss": 0.2324,
      "num_input_tokens_seen": 25908800,
      "step": 122770
    },
    {
      "epoch": 13.506600660066006,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.008645329728297783,
      "loss": 0.2304,
      "num_input_tokens_seen": 25909856,
      "step": 122775
    },
    {
      "epoch": 13.507150715071507,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00864402532780547,
      "loss": 0.2314,
      "num_input_tokens_seen": 25910880,
      "step": 122780
    },
    {
      "epoch": 13.507700770077008,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008642720985893125,
      "loss": 0.233,
      "num_input_tokens_seen": 25911936,
      "step": 122785
    },
    {
      "epoch": 13.508250825082508,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.008641416702572761,
      "loss": 0.2293,
      "num_input_tokens_seen": 25912992,
      "step": 122790
    },
    {
      "epoch": 13.508800880088009,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008640112477856418,
      "loss": 0.2309,
      "num_input_tokens_seen": 25914048,
      "step": 122795
    },
    {
      "epoch": 13.50935093509351,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0086388083117561,
      "loss": 0.2314,
      "num_input_tokens_seen": 25915072,
      "step": 122800
    },
    {
      "epoch": 13.509900990099009,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008637504204283832,
      "loss": 0.2293,
      "num_input_tokens_seen": 25916160,
      "step": 122805
    },
    {
      "epoch": 13.51045104510451,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.008636200155451641,
      "loss": 0.2319,
      "num_input_tokens_seen": 25917248,
      "step": 122810
    },
    {
      "epoch": 13.511001100110011,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.008634896165271532,
      "loss": 0.2314,
      "num_input_tokens_seen": 25918400,
      "step": 122815
    },
    {
      "epoch": 13.511551155115512,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.008633592233755538,
      "loss": 0.2309,
      "num_input_tokens_seen": 25919488,
      "step": 122820
    },
    {
      "epoch": 13.512101210121012,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.008632288360915664,
      "loss": 0.2314,
      "num_input_tokens_seen": 25920480,
      "step": 122825
    },
    {
      "epoch": 13.512651265126513,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.008630984546763934,
      "loss": 0.2319,
      "num_input_tokens_seen": 25921504,
      "step": 122830
    },
    {
      "epoch": 13.513201320132014,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008629680791312365,
      "loss": 0.2319,
      "num_input_tokens_seen": 25922528,
      "step": 122835
    },
    {
      "epoch": 13.513751375137513,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008628377094572969,
      "loss": 0.2309,
      "num_input_tokens_seen": 25923552,
      "step": 122840
    },
    {
      "epoch": 13.514301430143014,
      "grad_norm": 0.0028533935546875,
      "learning_rate": 0.00862707345655777,
      "loss": 0.2314,
      "num_input_tokens_seen": 25924640,
      "step": 122845
    },
    {
      "epoch": 13.514851485148515,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.008625769877278778,
      "loss": 0.233,
      "num_input_tokens_seen": 25925696,
      "step": 122850
    },
    {
      "epoch": 13.515401540154015,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.008624466356747996,
      "loss": 0.2314,
      "num_input_tokens_seen": 25926784,
      "step": 122855
    },
    {
      "epoch": 13.515951595159516,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008623162894977452,
      "loss": 0.2309,
      "num_input_tokens_seen": 25927872,
      "step": 122860
    },
    {
      "epoch": 13.516501650165017,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008621859491979154,
      "loss": 0.2324,
      "num_input_tokens_seen": 25928928,
      "step": 122865
    },
    {
      "epoch": 13.517051705170516,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008620556147765123,
      "loss": 0.2319,
      "num_input_tokens_seen": 25929888,
      "step": 122870
    },
    {
      "epoch": 13.517601760176017,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.008619252862347365,
      "loss": 0.2309,
      "num_input_tokens_seen": 25931040,
      "step": 122875
    },
    {
      "epoch": 13.518151815181518,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008617949635737886,
      "loss": 0.2304,
      "num_input_tokens_seen": 25932096,
      "step": 122880
    },
    {
      "epoch": 13.51870187018702,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0086166464679487,
      "loss": 0.2319,
      "num_input_tokens_seen": 25933216,
      "step": 122885
    },
    {
      "epoch": 13.519251925192519,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.008615343358991822,
      "loss": 0.2335,
      "num_input_tokens_seen": 25934240,
      "step": 122890
    },
    {
      "epoch": 13.51980198019802,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008614040308879268,
      "loss": 0.2314,
      "num_input_tokens_seen": 25935264,
      "step": 122895
    },
    {
      "epoch": 13.520352035203521,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00861273731762304,
      "loss": 0.2309,
      "num_input_tokens_seen": 25936288,
      "step": 122900
    },
    {
      "epoch": 13.52090209020902,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00861143438523514,
      "loss": 0.2314,
      "num_input_tokens_seen": 25937312,
      "step": 122905
    },
    {
      "epoch": 13.521452145214521,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008610131511727593,
      "loss": 0.2298,
      "num_input_tokens_seen": 25938272,
      "step": 122910
    },
    {
      "epoch": 13.522002200220022,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008608828697112386,
      "loss": 0.2319,
      "num_input_tokens_seen": 25939328,
      "step": 122915
    },
    {
      "epoch": 13.522552255225522,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.008607525941401541,
      "loss": 0.2293,
      "num_input_tokens_seen": 25940352,
      "step": 122920
    },
    {
      "epoch": 13.523102310231023,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008606223244607069,
      "loss": 0.2319,
      "num_input_tokens_seen": 25941440,
      "step": 122925
    },
    {
      "epoch": 13.523652365236524,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.008604920606740963,
      "loss": 0.2293,
      "num_input_tokens_seen": 25942496,
      "step": 122930
    },
    {
      "epoch": 13.524202420242025,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00860361802781524,
      "loss": 0.2298,
      "num_input_tokens_seen": 25943552,
      "step": 122935
    },
    {
      "epoch": 13.524752475247524,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008602315507841896,
      "loss": 0.2309,
      "num_input_tokens_seen": 25944608,
      "step": 122940
    },
    {
      "epoch": 13.525302530253025,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.00860101304683294,
      "loss": 0.2319,
      "num_input_tokens_seen": 25945664,
      "step": 122945
    },
    {
      "epoch": 13.525852585258527,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00859971064480038,
      "loss": 0.2309,
      "num_input_tokens_seen": 25946720,
      "step": 122950
    },
    {
      "epoch": 13.526402640264026,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.00859840830175621,
      "loss": 0.234,
      "num_input_tokens_seen": 25947808,
      "step": 122955
    },
    {
      "epoch": 13.526952695269527,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008597106017712448,
      "loss": 0.2288,
      "num_input_tokens_seen": 25948832,
      "step": 122960
    },
    {
      "epoch": 13.527502750275028,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.008595803792681078,
      "loss": 0.2309,
      "num_input_tokens_seen": 25949888,
      "step": 122965
    },
    {
      "epoch": 13.528052805280527,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.008594501626674118,
      "loss": 0.2293,
      "num_input_tokens_seen": 25951008,
      "step": 122970
    },
    {
      "epoch": 13.528602860286028,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.008593199519703557,
      "loss": 0.2335,
      "num_input_tokens_seen": 25952032,
      "step": 122975
    },
    {
      "epoch": 13.52915291529153,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.008591897471781405,
      "loss": 0.2309,
      "num_input_tokens_seen": 25953088,
      "step": 122980
    },
    {
      "epoch": 13.52970297029703,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00859059548291966,
      "loss": 0.233,
      "num_input_tokens_seen": 25954144,
      "step": 122985
    },
    {
      "epoch": 13.53025302530253,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.008589293553130326,
      "loss": 0.2314,
      "num_input_tokens_seen": 25955200,
      "step": 122990
    },
    {
      "epoch": 13.530803080308031,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008587991682425387,
      "loss": 0.2335,
      "num_input_tokens_seen": 25956256,
      "step": 122995
    },
    {
      "epoch": 13.531353135313532,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.00858668987081686,
      "loss": 0.2299,
      "num_input_tokens_seen": 25957312,
      "step": 123000
    },
    {
      "epoch": 13.531903190319031,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008585388118316722,
      "loss": 0.2303,
      "num_input_tokens_seen": 25958400,
      "step": 123005
    },
    {
      "epoch": 13.532453245324533,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.008584086424936997,
      "loss": 0.2304,
      "num_input_tokens_seen": 25959424,
      "step": 123010
    },
    {
      "epoch": 13.533003300330034,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.008582784790689668,
      "loss": 0.2314,
      "num_input_tokens_seen": 25960480,
      "step": 123015
    },
    {
      "epoch": 13.533553355335533,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.008581483215586725,
      "loss": 0.2293,
      "num_input_tokens_seen": 25961600,
      "step": 123020
    },
    {
      "epoch": 13.534103410341034,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.00858018169964018,
      "loss": 0.2314,
      "num_input_tokens_seen": 25962624,
      "step": 123025
    },
    {
      "epoch": 13.534653465346535,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00857888024286201,
      "loss": 0.233,
      "num_input_tokens_seen": 25963648,
      "step": 123030
    },
    {
      "epoch": 13.535203520352034,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.008577578845264224,
      "loss": 0.2314,
      "num_input_tokens_seen": 25964672,
      "step": 123035
    },
    {
      "epoch": 13.535753575357536,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008576277506858816,
      "loss": 0.2263,
      "num_input_tokens_seen": 25965760,
      "step": 123040
    },
    {
      "epoch": 13.536303630363037,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.00857497622765777,
      "loss": 0.2298,
      "num_input_tokens_seen": 25966784,
      "step": 123045
    },
    {
      "epoch": 13.536853685368538,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.00857367500767309,
      "loss": 0.2319,
      "num_input_tokens_seen": 25967744,
      "step": 123050
    },
    {
      "epoch": 13.537403740374037,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.008572373846916758,
      "loss": 0.2304,
      "num_input_tokens_seen": 25968832,
      "step": 123055
    },
    {
      "epoch": 13.537953795379538,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.008571072745400773,
      "loss": 0.2325,
      "num_input_tokens_seen": 25969856,
      "step": 123060
    },
    {
      "epoch": 13.53850385038504,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.00856977170313713,
      "loss": 0.233,
      "num_input_tokens_seen": 25970848,
      "step": 123065
    },
    {
      "epoch": 13.539053905390539,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.008568470720137814,
      "loss": 0.2319,
      "num_input_tokens_seen": 25971936,
      "step": 123070
    },
    {
      "epoch": 13.53960396039604,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.008567169796414819,
      "loss": 0.2304,
      "num_input_tokens_seen": 25972992,
      "step": 123075
    },
    {
      "epoch": 13.54015401540154,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008565868931980128,
      "loss": 0.233,
      "num_input_tokens_seen": 25974112,
      "step": 123080
    },
    {
      "epoch": 13.54070407040704,
      "grad_norm": 0.002655029296875,
      "learning_rate": 0.008564568126845742,
      "loss": 0.2304,
      "num_input_tokens_seen": 25975136,
      "step": 123085
    },
    {
      "epoch": 13.541254125412541,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008563267381023637,
      "loss": 0.2293,
      "num_input_tokens_seen": 25976160,
      "step": 123090
    },
    {
      "epoch": 13.541804180418042,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.008561966694525809,
      "loss": 0.2293,
      "num_input_tokens_seen": 25977216,
      "step": 123095
    },
    {
      "epoch": 13.542354235423542,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.00856066606736425,
      "loss": 0.2319,
      "num_input_tokens_seen": 25978240,
      "step": 123100
    },
    {
      "epoch": 13.542904290429043,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.008559365499550936,
      "loss": 0.234,
      "num_input_tokens_seen": 25979328,
      "step": 123105
    },
    {
      "epoch": 13.543454345434544,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.008558064991097866,
      "loss": 0.2309,
      "num_input_tokens_seen": 25980384,
      "step": 123110
    },
    {
      "epoch": 13.544004400440045,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008556764542017021,
      "loss": 0.2304,
      "num_input_tokens_seen": 25981408,
      "step": 123115
    },
    {
      "epoch": 13.544554455445544,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.008555464152320372,
      "loss": 0.2288,
      "num_input_tokens_seen": 25982496,
      "step": 123120
    },
    {
      "epoch": 13.545104510451045,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.00855416382201993,
      "loss": 0.2303,
      "num_input_tokens_seen": 25983552,
      "step": 123125
    },
    {
      "epoch": 13.545654565456546,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00855286355112766,
      "loss": 0.2304,
      "num_input_tokens_seen": 25984640,
      "step": 123130
    },
    {
      "epoch": 13.546204620462046,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.008551563339655559,
      "loss": 0.2335,
      "num_input_tokens_seen": 25985664,
      "step": 123135
    },
    {
      "epoch": 13.546754675467547,
      "grad_norm": 0.005859375,
      "learning_rate": 0.008550263187615604,
      "loss": 0.2278,
      "num_input_tokens_seen": 25986784,
      "step": 123140
    },
    {
      "epoch": 13.547304730473048,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.008548963095019775,
      "loss": 0.2309,
      "num_input_tokens_seen": 25987840,
      "step": 123145
    },
    {
      "epoch": 13.547854785478547,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.008547663061880055,
      "loss": 0.2319,
      "num_input_tokens_seen": 25988896,
      "step": 123150
    },
    {
      "epoch": 13.548404840484048,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.00854636308820843,
      "loss": 0.233,
      "num_input_tokens_seen": 25989888,
      "step": 123155
    },
    {
      "epoch": 13.54895489548955,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.008545063174016886,
      "loss": 0.2298,
      "num_input_tokens_seen": 25991008,
      "step": 123160
    },
    {
      "epoch": 13.549504950495049,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.008543763319317395,
      "loss": 0.2325,
      "num_input_tokens_seen": 25992064,
      "step": 123165
    },
    {
      "epoch": 13.55005500550055,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.008542463524121937,
      "loss": 0.2325,
      "num_input_tokens_seen": 25993120,
      "step": 123170
    },
    {
      "epoch": 13.55060506050605,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.008541163788442496,
      "loss": 0.234,
      "num_input_tokens_seen": 25994208,
      "step": 123175
    },
    {
      "epoch": 13.551155115511552,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.008539864112291046,
      "loss": 0.2319,
      "num_input_tokens_seen": 25995200,
      "step": 123180
    },
    {
      "epoch": 13.551705170517051,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008538564495679569,
      "loss": 0.2314,
      "num_input_tokens_seen": 25996288,
      "step": 123185
    },
    {
      "epoch": 13.552255225522552,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008537264938620046,
      "loss": 0.2304,
      "num_input_tokens_seen": 25997312,
      "step": 123190
    },
    {
      "epoch": 13.552805280528053,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.008535965441124447,
      "loss": 0.2314,
      "num_input_tokens_seen": 25998432,
      "step": 123195
    },
    {
      "epoch": 13.553355335533553,
      "grad_norm": 0.005859375,
      "learning_rate": 0.008534666003204757,
      "loss": 0.2309,
      "num_input_tokens_seen": 25999488,
      "step": 123200
    },
    {
      "epoch": 13.553905390539054,
      "grad_norm": 0.000720977783203125,
      "learning_rate": 0.008533366624872944,
      "loss": 0.2293,
      "num_input_tokens_seen": 26000544,
      "step": 123205
    },
    {
      "epoch": 13.554455445544555,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008532067306140987,
      "loss": 0.2293,
      "num_input_tokens_seen": 26001600,
      "step": 123210
    },
    {
      "epoch": 13.555005500550054,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.00853076804702087,
      "loss": 0.2293,
      "num_input_tokens_seen": 26002656,
      "step": 123215
    },
    {
      "epoch": 13.555555555555555,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.00852946884752455,
      "loss": 0.233,
      "num_input_tokens_seen": 26003680,
      "step": 123220
    },
    {
      "epoch": 13.556105610561056,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008528169707664018,
      "loss": 0.2293,
      "num_input_tokens_seen": 26004768,
      "step": 123225
    },
    {
      "epoch": 13.556655665566556,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.008526870627451241,
      "loss": 0.2304,
      "num_input_tokens_seen": 26005792,
      "step": 123230
    },
    {
      "epoch": 13.557205720572057,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008525571606898184,
      "loss": 0.2314,
      "num_input_tokens_seen": 26006784,
      "step": 123235
    },
    {
      "epoch": 13.557755775577558,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.008524272646016828,
      "loss": 0.2304,
      "num_input_tokens_seen": 26007840,
      "step": 123240
    },
    {
      "epoch": 13.558305830583059,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.008522973744819144,
      "loss": 0.2324,
      "num_input_tokens_seen": 26008960,
      "step": 123245
    },
    {
      "epoch": 13.558855885588558,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.008521674903317109,
      "loss": 0.233,
      "num_input_tokens_seen": 26010016,
      "step": 123250
    },
    {
      "epoch": 13.55940594059406,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.008520376121522687,
      "loss": 0.2314,
      "num_input_tokens_seen": 26011136,
      "step": 123255
    },
    {
      "epoch": 13.55995599559956,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008519077399447843,
      "loss": 0.2319,
      "num_input_tokens_seen": 26012256,
      "step": 123260
    },
    {
      "epoch": 13.56050605060506,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.008517778737104553,
      "loss": 0.2293,
      "num_input_tokens_seen": 26013344,
      "step": 123265
    },
    {
      "epoch": 13.561056105610561,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.008516480134504786,
      "loss": 0.2335,
      "num_input_tokens_seen": 26014432,
      "step": 123270
    },
    {
      "epoch": 13.561606160616062,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.008515181591660516,
      "loss": 0.2319,
      "num_input_tokens_seen": 26015456,
      "step": 123275
    },
    {
      "epoch": 13.562156215621561,
      "grad_norm": 0.005859375,
      "learning_rate": 0.008513883108583708,
      "loss": 0.2314,
      "num_input_tokens_seen": 26016512,
      "step": 123280
    },
    {
      "epoch": 13.562706270627062,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008512584685286317,
      "loss": 0.2299,
      "num_input_tokens_seen": 26017536,
      "step": 123285
    },
    {
      "epoch": 13.563256325632564,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.00851128632178033,
      "loss": 0.2335,
      "num_input_tokens_seen": 26018592,
      "step": 123290
    },
    {
      "epoch": 13.563806380638063,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.008509988018077693,
      "loss": 0.2325,
      "num_input_tokens_seen": 26019648,
      "step": 123295
    },
    {
      "epoch": 13.564356435643564,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.008508689774190387,
      "loss": 0.2325,
      "num_input_tokens_seen": 26020736,
      "step": 123300
    },
    {
      "epoch": 13.564906490649065,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.008507391590130376,
      "loss": 0.2325,
      "num_input_tokens_seen": 26021888,
      "step": 123305
    },
    {
      "epoch": 13.565456545654566,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008506093465909616,
      "loss": 0.2303,
      "num_input_tokens_seen": 26022944,
      "step": 123310
    },
    {
      "epoch": 13.566006600660065,
      "grad_norm": 0.003326416015625,
      "learning_rate": 0.008504795401540081,
      "loss": 0.2309,
      "num_input_tokens_seen": 26024032,
      "step": 123315
    },
    {
      "epoch": 13.566556655665567,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.008503497397033726,
      "loss": 0.2283,
      "num_input_tokens_seen": 26025056,
      "step": 123320
    },
    {
      "epoch": 13.567106710671068,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.008502199452402517,
      "loss": 0.233,
      "num_input_tokens_seen": 26026080,
      "step": 123325
    },
    {
      "epoch": 13.567656765676567,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008500901567658424,
      "loss": 0.2324,
      "num_input_tokens_seen": 26027136,
      "step": 123330
    },
    {
      "epoch": 13.568206820682068,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.008499603742813396,
      "loss": 0.2314,
      "num_input_tokens_seen": 26028224,
      "step": 123335
    },
    {
      "epoch": 13.56875687568757,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.008498305977879409,
      "loss": 0.233,
      "num_input_tokens_seen": 26029248,
      "step": 123340
    },
    {
      "epoch": 13.569306930693068,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.008497008272868407,
      "loss": 0.2314,
      "num_input_tokens_seen": 26030368,
      "step": 123345
    },
    {
      "epoch": 13.56985698569857,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.008495710627792367,
      "loss": 0.2319,
      "num_input_tokens_seen": 26031392,
      "step": 123350
    },
    {
      "epoch": 13.57040704070407,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008494413042663235,
      "loss": 0.2325,
      "num_input_tokens_seen": 26032416,
      "step": 123355
    },
    {
      "epoch": 13.570957095709572,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008493115517492978,
      "loss": 0.2293,
      "num_input_tokens_seen": 26033472,
      "step": 123360
    },
    {
      "epoch": 13.571507150715071,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.008491818052293555,
      "loss": 0.2299,
      "num_input_tokens_seen": 26034528,
      "step": 123365
    },
    {
      "epoch": 13.572057205720572,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008490520647076925,
      "loss": 0.2293,
      "num_input_tokens_seen": 26035584,
      "step": 123370
    },
    {
      "epoch": 13.572607260726073,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.008489223301855038,
      "loss": 0.2325,
      "num_input_tokens_seen": 26036608,
      "step": 123375
    },
    {
      "epoch": 13.573157315731573,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.008487926016639852,
      "loss": 0.2314,
      "num_input_tokens_seen": 26037696,
      "step": 123380
    },
    {
      "epoch": 13.573707370737074,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.008486628791443329,
      "loss": 0.2335,
      "num_input_tokens_seen": 26038752,
      "step": 123385
    },
    {
      "epoch": 13.574257425742575,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00848533162627743,
      "loss": 0.2319,
      "num_input_tokens_seen": 26039840,
      "step": 123390
    },
    {
      "epoch": 13.574807480748074,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.008484034521154102,
      "loss": 0.2314,
      "num_input_tokens_seen": 26040896,
      "step": 123395
    },
    {
      "epoch": 13.575357535753575,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008482737476085295,
      "loss": 0.2324,
      "num_input_tokens_seen": 26041984,
      "step": 123400
    },
    {
      "epoch": 13.575907590759076,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.008481440491082976,
      "loss": 0.2324,
      "num_input_tokens_seen": 26043040,
      "step": 123405
    },
    {
      "epoch": 13.576457645764577,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.008480143566159086,
      "loss": 0.2335,
      "num_input_tokens_seen": 26044064,
      "step": 123410
    },
    {
      "epoch": 13.577007700770077,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.008478846701325584,
      "loss": 0.2293,
      "num_input_tokens_seen": 26045152,
      "step": 123415
    },
    {
      "epoch": 13.577557755775578,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.008477549896594428,
      "loss": 0.2324,
      "num_input_tokens_seen": 26046144,
      "step": 123420
    },
    {
      "epoch": 13.578107810781079,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008476253151977559,
      "loss": 0.2288,
      "num_input_tokens_seen": 26047232,
      "step": 123425
    },
    {
      "epoch": 13.578657865786578,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.008474956467486942,
      "loss": 0.2319,
      "num_input_tokens_seen": 26048352,
      "step": 123430
    },
    {
      "epoch": 13.57920792079208,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.008473659843134513,
      "loss": 0.2309,
      "num_input_tokens_seen": 26049376,
      "step": 123435
    },
    {
      "epoch": 13.57975797579758,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.00847236327893223,
      "loss": 0.2335,
      "num_input_tokens_seen": 26050368,
      "step": 123440
    },
    {
      "epoch": 13.58030803080308,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.00847106677489205,
      "loss": 0.2314,
      "num_input_tokens_seen": 26051424,
      "step": 123445
    },
    {
      "epoch": 13.58085808580858,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.008469770331025904,
      "loss": 0.2293,
      "num_input_tokens_seen": 26052448,
      "step": 123450
    },
    {
      "epoch": 13.581408140814082,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008468473947345762,
      "loss": 0.2293,
      "num_input_tokens_seen": 26053536,
      "step": 123455
    },
    {
      "epoch": 13.581958195819581,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.008467177623863554,
      "loss": 0.2335,
      "num_input_tokens_seen": 26054624,
      "step": 123460
    },
    {
      "epoch": 13.582508250825082,
      "grad_norm": 0.001953125,
      "learning_rate": 0.008465881360591239,
      "loss": 0.2304,
      "num_input_tokens_seen": 26055648,
      "step": 123465
    },
    {
      "epoch": 13.583058305830583,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.008464585157540755,
      "loss": 0.2314,
      "num_input_tokens_seen": 26056704,
      "step": 123470
    },
    {
      "epoch": 13.583608360836084,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.008463289014724056,
      "loss": 0.2335,
      "num_input_tokens_seen": 26057760,
      "step": 123475
    },
    {
      "epoch": 13.584158415841584,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.008461992932153089,
      "loss": 0.2335,
      "num_input_tokens_seen": 26058912,
      "step": 123480
    },
    {
      "epoch": 13.584708470847085,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.008460696909839792,
      "loss": 0.2299,
      "num_input_tokens_seen": 26059904,
      "step": 123485
    },
    {
      "epoch": 13.585258525852586,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008459400947796118,
      "loss": 0.2314,
      "num_input_tokens_seen": 26060992,
      "step": 123490
    },
    {
      "epoch": 13.585808580858085,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.008458105046034002,
      "loss": 0.2288,
      "num_input_tokens_seen": 26062080,
      "step": 123495
    },
    {
      "epoch": 13.586358635863586,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008456809204565393,
      "loss": 0.2335,
      "num_input_tokens_seen": 26063104,
      "step": 123500
    },
    {
      "epoch": 13.586908690869087,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.00845551342340224,
      "loss": 0.2314,
      "num_input_tokens_seen": 26064160,
      "step": 123505
    },
    {
      "epoch": 13.587458745874587,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.008454217702556471,
      "loss": 0.2325,
      "num_input_tokens_seen": 26065248,
      "step": 123510
    },
    {
      "epoch": 13.588008800880088,
      "grad_norm": 0.000682830810546875,
      "learning_rate": 0.008452922042040043,
      "loss": 0.2319,
      "num_input_tokens_seen": 26066336,
      "step": 123515
    },
    {
      "epoch": 13.588558855885589,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.008451626441864894,
      "loss": 0.2319,
      "num_input_tokens_seen": 26067328,
      "step": 123520
    },
    {
      "epoch": 13.589108910891088,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.008450330902042953,
      "loss": 0.233,
      "num_input_tokens_seen": 26068320,
      "step": 123525
    },
    {
      "epoch": 13.58965896589659,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00844903542258617,
      "loss": 0.2345,
      "num_input_tokens_seen": 26069376,
      "step": 123530
    },
    {
      "epoch": 13.59020902090209,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.008447740003506484,
      "loss": 0.2345,
      "num_input_tokens_seen": 26070400,
      "step": 123535
    },
    {
      "epoch": 13.590759075907592,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008446444644815842,
      "loss": 0.2319,
      "num_input_tokens_seen": 26071392,
      "step": 123540
    },
    {
      "epoch": 13.591309130913091,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.008445149346526171,
      "loss": 0.2304,
      "num_input_tokens_seen": 26072416,
      "step": 123545
    },
    {
      "epoch": 13.591859185918592,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.008443854108649411,
      "loss": 0.2304,
      "num_input_tokens_seen": 26073472,
      "step": 123550
    },
    {
      "epoch": 13.592409240924093,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.0084425589311975,
      "loss": 0.2314,
      "num_input_tokens_seen": 26074560,
      "step": 123555
    },
    {
      "epoch": 13.592959295929592,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.008441263814182384,
      "loss": 0.2319,
      "num_input_tokens_seen": 26075616,
      "step": 123560
    },
    {
      "epoch": 13.593509350935093,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.008439968757615986,
      "loss": 0.2319,
      "num_input_tokens_seen": 26076640,
      "step": 123565
    },
    {
      "epoch": 13.594059405940595,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.008438673761510253,
      "loss": 0.2314,
      "num_input_tokens_seen": 26077696,
      "step": 123570
    },
    {
      "epoch": 13.594609460946094,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.00843737882587711,
      "loss": 0.2314,
      "num_input_tokens_seen": 26078752,
      "step": 123575
    },
    {
      "epoch": 13.595159515951595,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.008436083950728502,
      "loss": 0.2314,
      "num_input_tokens_seen": 26079776,
      "step": 123580
    },
    {
      "epoch": 13.595709570957096,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.008434789136076355,
      "loss": 0.2324,
      "num_input_tokens_seen": 26080832,
      "step": 123585
    },
    {
      "epoch": 13.596259625962595,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.008433494381932605,
      "loss": 0.2314,
      "num_input_tokens_seen": 26081920,
      "step": 123590
    },
    {
      "epoch": 13.596809680968097,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.008432199688309192,
      "loss": 0.233,
      "num_input_tokens_seen": 26082944,
      "step": 123595
    },
    {
      "epoch": 13.597359735973598,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.008430905055218037,
      "loss": 0.2309,
      "num_input_tokens_seen": 26083936,
      "step": 123600
    },
    {
      "epoch": 13.597909790979099,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.008429610482671086,
      "loss": 0.2335,
      "num_input_tokens_seen": 26084960,
      "step": 123605
    },
    {
      "epoch": 13.598459845984598,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.008428315970680254,
      "loss": 0.2314,
      "num_input_tokens_seen": 26086048,
      "step": 123610
    },
    {
      "epoch": 13.599009900990099,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00842702151925748,
      "loss": 0.2309,
      "num_input_tokens_seen": 26087040,
      "step": 123615
    },
    {
      "epoch": 13.5995599559956,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.008425727128414703,
      "loss": 0.2324,
      "num_input_tokens_seen": 26088128,
      "step": 123620
    },
    {
      "epoch": 13.6001100110011,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008424432798163836,
      "loss": 0.2329,
      "num_input_tokens_seen": 26089152,
      "step": 123625
    },
    {
      "epoch": 13.6006600660066,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.008423138528516826,
      "loss": 0.2303,
      "num_input_tokens_seen": 26090208,
      "step": 123630
    },
    {
      "epoch": 13.601210121012102,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.008421844319485589,
      "loss": 0.2319,
      "num_input_tokens_seen": 26091232,
      "step": 123635
    },
    {
      "epoch": 13.601760176017601,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.00842055017108205,
      "loss": 0.2303,
      "num_input_tokens_seen": 26092320,
      "step": 123640
    },
    {
      "epoch": 13.602310231023102,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.008419256083318146,
      "loss": 0.2314,
      "num_input_tokens_seen": 26093440,
      "step": 123645
    },
    {
      "epoch": 13.602860286028603,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008417962056205798,
      "loss": 0.2324,
      "num_input_tokens_seen": 26094528,
      "step": 123650
    },
    {
      "epoch": 13.603410341034103,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.008416668089756944,
      "loss": 0.2304,
      "num_input_tokens_seen": 26095552,
      "step": 123655
    },
    {
      "epoch": 13.603960396039604,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0084153741839835,
      "loss": 0.2308,
      "num_input_tokens_seen": 26096544,
      "step": 123660
    },
    {
      "epoch": 13.604510451045105,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.008414080338897387,
      "loss": 0.2314,
      "num_input_tokens_seen": 26097632,
      "step": 123665
    },
    {
      "epoch": 13.605060506050606,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 0.008412786554510533,
      "loss": 0.2314,
      "num_input_tokens_seen": 26098656,
      "step": 123670
    },
    {
      "epoch": 13.605610561056105,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.008411492830834867,
      "loss": 0.233,
      "num_input_tokens_seen": 26099776,
      "step": 123675
    },
    {
      "epoch": 13.606160616061606,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008410199167882316,
      "loss": 0.2319,
      "num_input_tokens_seen": 26100864,
      "step": 123680
    },
    {
      "epoch": 13.606710671067107,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008408905565664794,
      "loss": 0.2314,
      "num_input_tokens_seen": 26101888,
      "step": 123685
    },
    {
      "epoch": 13.607260726072607,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.008407612024194225,
      "loss": 0.2319,
      "num_input_tokens_seen": 26102944,
      "step": 123690
    },
    {
      "epoch": 13.607810781078108,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.008406318543482536,
      "loss": 0.2303,
      "num_input_tokens_seen": 26104000,
      "step": 123695
    },
    {
      "epoch": 13.608360836083609,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.00840502512354164,
      "loss": 0.2319,
      "num_input_tokens_seen": 26105056,
      "step": 123700
    },
    {
      "epoch": 13.608910891089108,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008403731764383464,
      "loss": 0.2319,
      "num_input_tokens_seen": 26106176,
      "step": 123705
    },
    {
      "epoch": 13.60946094609461,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.008402438466019933,
      "loss": 0.2309,
      "num_input_tokens_seen": 26107200,
      "step": 123710
    },
    {
      "epoch": 13.61001100110011,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.008401145228462952,
      "loss": 0.2303,
      "num_input_tokens_seen": 26108224,
      "step": 123715
    },
    {
      "epoch": 13.61056105610561,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008399852051724457,
      "loss": 0.2314,
      "num_input_tokens_seen": 26109280,
      "step": 123720
    },
    {
      "epoch": 13.61111111111111,
      "grad_norm": 0.000812530517578125,
      "learning_rate": 0.008398558935816352,
      "loss": 0.2319,
      "num_input_tokens_seen": 26110240,
      "step": 123725
    },
    {
      "epoch": 13.611661166116612,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.00839726588075056,
      "loss": 0.2319,
      "num_input_tokens_seen": 26111296,
      "step": 123730
    },
    {
      "epoch": 13.612211221122113,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.00839597288653901,
      "loss": 0.2298,
      "num_input_tokens_seen": 26112448,
      "step": 123735
    },
    {
      "epoch": 13.612761276127612,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0083946799531936,
      "loss": 0.2329,
      "num_input_tokens_seen": 26113472,
      "step": 123740
    },
    {
      "epoch": 13.613311331133113,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.008393387080726263,
      "loss": 0.2303,
      "num_input_tokens_seen": 26114592,
      "step": 123745
    },
    {
      "epoch": 13.613861386138614,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.008392094269148906,
      "loss": 0.2314,
      "num_input_tokens_seen": 26115584,
      "step": 123750
    },
    {
      "epoch": 13.614411441144114,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.008390801518473441,
      "loss": 0.2325,
      "num_input_tokens_seen": 26116608,
      "step": 123755
    },
    {
      "epoch": 13.614961496149615,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.008389508828711784,
      "loss": 0.2319,
      "num_input_tokens_seen": 26117760,
      "step": 123760
    },
    {
      "epoch": 13.615511551155116,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.008388216199875854,
      "loss": 0.2319,
      "num_input_tokens_seen": 26118784,
      "step": 123765
    },
    {
      "epoch": 13.616061606160617,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00838692363197757,
      "loss": 0.2303,
      "num_input_tokens_seen": 26119840,
      "step": 123770
    },
    {
      "epoch": 13.616611661166116,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.008385631125028834,
      "loss": 0.2319,
      "num_input_tokens_seen": 26120864,
      "step": 123775
    },
    {
      "epoch": 13.617161716171617,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.008384338679041559,
      "loss": 0.2329,
      "num_input_tokens_seen": 26121952,
      "step": 123780
    },
    {
      "epoch": 13.617711771177119,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.008383046294027663,
      "loss": 0.2324,
      "num_input_tokens_seen": 26123040,
      "step": 123785
    },
    {
      "epoch": 13.618261826182618,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.008381753969999047,
      "loss": 0.2324,
      "num_input_tokens_seen": 26124096,
      "step": 123790
    },
    {
      "epoch": 13.618811881188119,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.008380461706967638,
      "loss": 0.2309,
      "num_input_tokens_seen": 26125152,
      "step": 123795
    },
    {
      "epoch": 13.61936193619362,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.008379169504945336,
      "loss": 0.2308,
      "num_input_tokens_seen": 26126208,
      "step": 123800
    },
    {
      "epoch": 13.61991199119912,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.008377877363944046,
      "loss": 0.2324,
      "num_input_tokens_seen": 26127232,
      "step": 123805
    },
    {
      "epoch": 13.62046204620462,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.00837658528397569,
      "loss": 0.2329,
      "num_input_tokens_seen": 26128320,
      "step": 123810
    },
    {
      "epoch": 13.621012101210122,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008375293265052163,
      "loss": 0.233,
      "num_input_tokens_seen": 26129408,
      "step": 123815
    },
    {
      "epoch": 13.62156215621562,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.00837400130718538,
      "loss": 0.2293,
      "num_input_tokens_seen": 26130496,
      "step": 123820
    },
    {
      "epoch": 13.622112211221122,
      "grad_norm": 0.000644683837890625,
      "learning_rate": 0.008372709410387251,
      "loss": 0.2298,
      "num_input_tokens_seen": 26131552,
      "step": 123825
    },
    {
      "epoch": 13.622662266226623,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.008371417574669675,
      "loss": 0.2298,
      "num_input_tokens_seen": 26132576,
      "step": 123830
    },
    {
      "epoch": 13.623212321232124,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.008370125800044568,
      "loss": 0.2319,
      "num_input_tokens_seen": 26133632,
      "step": 123835
    },
    {
      "epoch": 13.623762376237623,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008368834086523827,
      "loss": 0.2314,
      "num_input_tokens_seen": 26134688,
      "step": 123840
    },
    {
      "epoch": 13.624312431243125,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008367542434119362,
      "loss": 0.2324,
      "num_input_tokens_seen": 26135776,
      "step": 123845
    },
    {
      "epoch": 13.624862486248626,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.008366250842843072,
      "loss": 0.2319,
      "num_input_tokens_seen": 26136864,
      "step": 123850
    },
    {
      "epoch": 13.625412541254125,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008364959312706864,
      "loss": 0.2309,
      "num_input_tokens_seen": 26137888,
      "step": 123855
    },
    {
      "epoch": 13.625962596259626,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.008363667843722649,
      "loss": 0.2304,
      "num_input_tokens_seen": 26138944,
      "step": 123860
    },
    {
      "epoch": 13.626512651265127,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.008362376435902315,
      "loss": 0.2314,
      "num_input_tokens_seen": 26139968,
      "step": 123865
    },
    {
      "epoch": 13.627062706270626,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.00836108508925778,
      "loss": 0.2335,
      "num_input_tokens_seen": 26141024,
      "step": 123870
    },
    {
      "epoch": 13.627612761276128,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.008359793803800933,
      "loss": 0.2351,
      "num_input_tokens_seen": 26142016,
      "step": 123875
    },
    {
      "epoch": 13.628162816281629,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.008358502579543679,
      "loss": 0.234,
      "num_input_tokens_seen": 26142976,
      "step": 123880
    },
    {
      "epoch": 13.628712871287128,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008357211416497926,
      "loss": 0.2303,
      "num_input_tokens_seen": 26144064,
      "step": 123885
    },
    {
      "epoch": 13.629262926292629,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008355920314675563,
      "loss": 0.2304,
      "num_input_tokens_seen": 26145120,
      "step": 123890
    },
    {
      "epoch": 13.62981298129813,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.008354629274088499,
      "loss": 0.2319,
      "num_input_tokens_seen": 26146208,
      "step": 123895
    },
    {
      "epoch": 13.630363036303631,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.008353338294748627,
      "loss": 0.2303,
      "num_input_tokens_seen": 26147264,
      "step": 123900
    },
    {
      "epoch": 13.63091309130913,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.008352047376667837,
      "loss": 0.2319,
      "num_input_tokens_seen": 26148320,
      "step": 123905
    },
    {
      "epoch": 13.631463146314632,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008350756519858048,
      "loss": 0.2314,
      "num_input_tokens_seen": 26149376,
      "step": 123910
    },
    {
      "epoch": 13.632013201320133,
      "grad_norm": 0.000812530517578125,
      "learning_rate": 0.008349465724331137,
      "loss": 0.2325,
      "num_input_tokens_seen": 26150432,
      "step": 123915
    },
    {
      "epoch": 13.632563256325632,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.008348174990099017,
      "loss": 0.2304,
      "num_input_tokens_seen": 26151520,
      "step": 123920
    },
    {
      "epoch": 13.633113311331133,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.008346884317173578,
      "loss": 0.2314,
      "num_input_tokens_seen": 26152544,
      "step": 123925
    },
    {
      "epoch": 13.633663366336634,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.008345593705566706,
      "loss": 0.234,
      "num_input_tokens_seen": 26153536,
      "step": 123930
    },
    {
      "epoch": 13.634213421342134,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.008344303155290304,
      "loss": 0.2308,
      "num_input_tokens_seen": 26154592,
      "step": 123935
    },
    {
      "epoch": 13.634763476347635,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00834301266635627,
      "loss": 0.2335,
      "num_input_tokens_seen": 26155616,
      "step": 123940
    },
    {
      "epoch": 13.635313531353136,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008341722238776493,
      "loss": 0.2319,
      "num_input_tokens_seen": 26156672,
      "step": 123945
    },
    {
      "epoch": 13.635863586358635,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008340431872562868,
      "loss": 0.2319,
      "num_input_tokens_seen": 26157792,
      "step": 123950
    },
    {
      "epoch": 13.636413641364136,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.008339141567727283,
      "loss": 0.2293,
      "num_input_tokens_seen": 26158848,
      "step": 123955
    },
    {
      "epoch": 13.636963696369637,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00833785132428164,
      "loss": 0.2298,
      "num_input_tokens_seen": 26159840,
      "step": 123960
    },
    {
      "epoch": 13.637513751375138,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.008336561142237818,
      "loss": 0.2324,
      "num_input_tokens_seen": 26160864,
      "step": 123965
    },
    {
      "epoch": 13.638063806380638,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.008335271021607716,
      "loss": 0.2314,
      "num_input_tokens_seen": 26161952,
      "step": 123970
    },
    {
      "epoch": 13.638613861386139,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.008333980962403228,
      "loss": 0.2319,
      "num_input_tokens_seen": 26163040,
      "step": 123975
    },
    {
      "epoch": 13.63916391639164,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008332690964636232,
      "loss": 0.233,
      "num_input_tokens_seen": 26164096,
      "step": 123980
    },
    {
      "epoch": 13.63971397139714,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.00833140102831863,
      "loss": 0.2345,
      "num_input_tokens_seen": 26165152,
      "step": 123985
    },
    {
      "epoch": 13.64026402640264,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0083301111534623,
      "loss": 0.2314,
      "num_input_tokens_seen": 26166176,
      "step": 123990
    },
    {
      "epoch": 13.640814081408141,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.008328821340079133,
      "loss": 0.2319,
      "num_input_tokens_seen": 26167328,
      "step": 123995
    },
    {
      "epoch": 13.64136413641364,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.008327531588181027,
      "loss": 0.2324,
      "num_input_tokens_seen": 26168352,
      "step": 124000
    },
    {
      "epoch": 13.641914191419142,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.008326241897779856,
      "loss": 0.2324,
      "num_input_tokens_seen": 26169408,
      "step": 124005
    },
    {
      "epoch": 13.642464246424643,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.008324952268887513,
      "loss": 0.2319,
      "num_input_tokens_seen": 26170432,
      "step": 124010
    },
    {
      "epoch": 13.643014301430142,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.008323662701515884,
      "loss": 0.2309,
      "num_input_tokens_seen": 26171520,
      "step": 124015
    },
    {
      "epoch": 13.643564356435643,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.008322373195676845,
      "loss": 0.2329,
      "num_input_tokens_seen": 26172640,
      "step": 124020
    },
    {
      "epoch": 13.644114411441144,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.00832108375138229,
      "loss": 0.233,
      "num_input_tokens_seen": 26173696,
      "step": 124025
    },
    {
      "epoch": 13.644664466446645,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0083197943686441,
      "loss": 0.2303,
      "num_input_tokens_seen": 26174752,
      "step": 124030
    },
    {
      "epoch": 13.645214521452145,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.008318505047474166,
      "loss": 0.2298,
      "num_input_tokens_seen": 26175744,
      "step": 124035
    },
    {
      "epoch": 13.645764576457646,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.008317215787884365,
      "loss": 0.2314,
      "num_input_tokens_seen": 26176864,
      "step": 124040
    },
    {
      "epoch": 13.646314631463147,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.008315926589886573,
      "loss": 0.2309,
      "num_input_tokens_seen": 26178016,
      "step": 124045
    },
    {
      "epoch": 13.646864686468646,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.008314637453492677,
      "loss": 0.2314,
      "num_input_tokens_seen": 26179072,
      "step": 124050
    },
    {
      "epoch": 13.647414741474147,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.008313348378714563,
      "loss": 0.2319,
      "num_input_tokens_seen": 26180128,
      "step": 124055
    },
    {
      "epoch": 13.647964796479648,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.008312059365564112,
      "loss": 0.2293,
      "num_input_tokens_seen": 26181184,
      "step": 124060
    },
    {
      "epoch": 13.648514851485148,
      "grad_norm": 0.000728607177734375,
      "learning_rate": 0.008310770414053202,
      "loss": 0.2309,
      "num_input_tokens_seen": 26182176,
      "step": 124065
    },
    {
      "epoch": 13.649064906490649,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.008309481524193706,
      "loss": 0.2329,
      "num_input_tokens_seen": 26183264,
      "step": 124070
    },
    {
      "epoch": 13.64961496149615,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.008308192695997512,
      "loss": 0.2319,
      "num_input_tokens_seen": 26184320,
      "step": 124075
    },
    {
      "epoch": 13.65016501650165,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.00830690392947649,
      "loss": 0.2314,
      "num_input_tokens_seen": 26185376,
      "step": 124080
    },
    {
      "epoch": 13.65071507150715,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.008305615224642522,
      "loss": 0.2324,
      "num_input_tokens_seen": 26186368,
      "step": 124085
    },
    {
      "epoch": 13.651265126512651,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.008304326581507492,
      "loss": 0.2324,
      "num_input_tokens_seen": 26187456,
      "step": 124090
    },
    {
      "epoch": 13.651815181518153,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.008303038000083265,
      "loss": 0.2303,
      "num_input_tokens_seen": 26188448,
      "step": 124095
    },
    {
      "epoch": 13.652365236523652,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.008301749480381732,
      "loss": 0.2319,
      "num_input_tokens_seen": 26189536,
      "step": 124100
    },
    {
      "epoch": 13.652915291529153,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008300461022414753,
      "loss": 0.2324,
      "num_input_tokens_seen": 26190656,
      "step": 124105
    },
    {
      "epoch": 13.653465346534654,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008299172626194208,
      "loss": 0.2304,
      "num_input_tokens_seen": 26191776,
      "step": 124110
    },
    {
      "epoch": 13.654015401540153,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.008297884291731979,
      "loss": 0.2303,
      "num_input_tokens_seen": 26192832,
      "step": 124115
    },
    {
      "epoch": 13.654565456545654,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.008296596019039931,
      "loss": 0.2303,
      "num_input_tokens_seen": 26193856,
      "step": 124120
    },
    {
      "epoch": 13.655115511551156,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.008295307808129947,
      "loss": 0.2304,
      "num_input_tokens_seen": 26194912,
      "step": 124125
    },
    {
      "epoch": 13.655665566556655,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.008294019659013892,
      "loss": 0.2329,
      "num_input_tokens_seen": 26195968,
      "step": 124130
    },
    {
      "epoch": 13.656215621562156,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.008292731571703635,
      "loss": 0.2298,
      "num_input_tokens_seen": 26196960,
      "step": 124135
    },
    {
      "epoch": 13.656765676567657,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.008291443546211053,
      "loss": 0.2309,
      "num_input_tokens_seen": 26198048,
      "step": 124140
    },
    {
      "epoch": 13.657315731573158,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008290155582548014,
      "loss": 0.2304,
      "num_input_tokens_seen": 26199072,
      "step": 124145
    },
    {
      "epoch": 13.657865786578657,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008288867680726399,
      "loss": 0.2303,
      "num_input_tokens_seen": 26200192,
      "step": 124150
    },
    {
      "epoch": 13.658415841584159,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.00828757984075807,
      "loss": 0.2298,
      "num_input_tokens_seen": 26201248,
      "step": 124155
    },
    {
      "epoch": 13.65896589658966,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008286292062654892,
      "loss": 0.2335,
      "num_input_tokens_seen": 26202368,
      "step": 124160
    },
    {
      "epoch": 13.659515951595159,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.008285004346428737,
      "loss": 0.233,
      "num_input_tokens_seen": 26203424,
      "step": 124165
    },
    {
      "epoch": 13.66006600660066,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008283716692091478,
      "loss": 0.2309,
      "num_input_tokens_seen": 26204512,
      "step": 124170
    },
    {
      "epoch": 13.660616061606161,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.008282429099654981,
      "loss": 0.2309,
      "num_input_tokens_seen": 26205600,
      "step": 124175
    },
    {
      "epoch": 13.66116611661166,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.008281141569131114,
      "loss": 0.2293,
      "num_input_tokens_seen": 26206560,
      "step": 124180
    },
    {
      "epoch": 13.661716171617162,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.008279854100531734,
      "loss": 0.2314,
      "num_input_tokens_seen": 26207616,
      "step": 124185
    },
    {
      "epoch": 13.662266226622663,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008278566693868722,
      "loss": 0.2324,
      "num_input_tokens_seen": 26208736,
      "step": 124190
    },
    {
      "epoch": 13.662816281628164,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.008277279349153928,
      "loss": 0.2329,
      "num_input_tokens_seen": 26209728,
      "step": 124195
    },
    {
      "epoch": 13.663366336633663,
      "grad_norm": 0.005859375,
      "learning_rate": 0.008275992066399224,
      "loss": 0.2319,
      "num_input_tokens_seen": 26210848,
      "step": 124200
    },
    {
      "epoch": 13.663916391639164,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00827470484561648,
      "loss": 0.234,
      "num_input_tokens_seen": 26211872,
      "step": 124205
    },
    {
      "epoch": 13.664466446644665,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00827341768681755,
      "loss": 0.2319,
      "num_input_tokens_seen": 26212928,
      "step": 124210
    },
    {
      "epoch": 13.665016501650165,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.008272130590014309,
      "loss": 0.2319,
      "num_input_tokens_seen": 26213920,
      "step": 124215
    },
    {
      "epoch": 13.665566556655666,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008270843555218604,
      "loss": 0.2294,
      "num_input_tokens_seen": 26214912,
      "step": 124220
    },
    {
      "epoch": 13.666116611661167,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.008269556582442305,
      "loss": 0.2309,
      "num_input_tokens_seen": 26215968,
      "step": 124225
    },
    {
      "epoch": 13.666666666666666,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.008268269671697278,
      "loss": 0.2299,
      "num_input_tokens_seen": 26216992,
      "step": 124230
    },
    {
      "epoch": 13.667216721672167,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008266982822995376,
      "loss": 0.2293,
      "num_input_tokens_seen": 26218016,
      "step": 124235
    },
    {
      "epoch": 13.667766776677668,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.008265696036348467,
      "loss": 0.2298,
      "num_input_tokens_seen": 26219040,
      "step": 124240
    },
    {
      "epoch": 13.668316831683168,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.008264409311768401,
      "loss": 0.2319,
      "num_input_tokens_seen": 26220096,
      "step": 124245
    },
    {
      "epoch": 13.668866886688669,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.008263122649267048,
      "loss": 0.233,
      "num_input_tokens_seen": 26221120,
      "step": 124250
    },
    {
      "epoch": 13.66941694169417,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.008261836048856254,
      "loss": 0.2293,
      "num_input_tokens_seen": 26222176,
      "step": 124255
    },
    {
      "epoch": 13.66996699669967,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008260549510547884,
      "loss": 0.2314,
      "num_input_tokens_seen": 26223232,
      "step": 124260
    },
    {
      "epoch": 13.67051705170517,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0082592630343538,
      "loss": 0.2309,
      "num_input_tokens_seen": 26224256,
      "step": 124265
    },
    {
      "epoch": 13.671067106710671,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00825797662028585,
      "loss": 0.2314,
      "num_input_tokens_seen": 26225344,
      "step": 124270
    },
    {
      "epoch": 13.671617161716172,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008256690268355902,
      "loss": 0.2303,
      "num_input_tokens_seen": 26226432,
      "step": 124275
    },
    {
      "epoch": 13.672167216721672,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.008255403978575796,
      "loss": 0.2288,
      "num_input_tokens_seen": 26227456,
      "step": 124280
    },
    {
      "epoch": 13.672717271727173,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.008254117750957395,
      "loss": 0.2319,
      "num_input_tokens_seen": 26228480,
      "step": 124285
    },
    {
      "epoch": 13.673267326732674,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.00825283158551256,
      "loss": 0.2304,
      "num_input_tokens_seen": 26229504,
      "step": 124290
    },
    {
      "epoch": 13.673817381738173,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.008251545482253134,
      "loss": 0.233,
      "num_input_tokens_seen": 26230592,
      "step": 124295
    },
    {
      "epoch": 13.674367436743674,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.008250259441190979,
      "loss": 0.2309,
      "num_input_tokens_seen": 26231744,
      "step": 124300
    },
    {
      "epoch": 13.674917491749175,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.008248973462337945,
      "loss": 0.2293,
      "num_input_tokens_seen": 26232768,
      "step": 124305
    },
    {
      "epoch": 13.675467546754675,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.008247687545705876,
      "loss": 0.2325,
      "num_input_tokens_seen": 26233760,
      "step": 124310
    },
    {
      "epoch": 13.676017601760176,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.008246401691306632,
      "loss": 0.2288,
      "num_input_tokens_seen": 26234880,
      "step": 124315
    },
    {
      "epoch": 13.676567656765677,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.008245115899152063,
      "loss": 0.2314,
      "num_input_tokens_seen": 26236000,
      "step": 124320
    },
    {
      "epoch": 13.677117711771178,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.008243830169254025,
      "loss": 0.2319,
      "num_input_tokens_seen": 26237120,
      "step": 124325
    },
    {
      "epoch": 13.677667766776677,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.008242544501624363,
      "loss": 0.2299,
      "num_input_tokens_seen": 26238208,
      "step": 124330
    },
    {
      "epoch": 13.678217821782178,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008241258896274918,
      "loss": 0.2298,
      "num_input_tokens_seen": 26239296,
      "step": 124335
    },
    {
      "epoch": 13.67876787678768,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.008239973353217548,
      "loss": 0.2319,
      "num_input_tokens_seen": 26240352,
      "step": 124340
    },
    {
      "epoch": 13.679317931793179,
      "grad_norm": 0.005859375,
      "learning_rate": 0.008238687872464106,
      "loss": 0.2314,
      "num_input_tokens_seen": 26241440,
      "step": 124345
    },
    {
      "epoch": 13.67986798679868,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.008237402454026427,
      "loss": 0.2304,
      "num_input_tokens_seen": 26242496,
      "step": 124350
    },
    {
      "epoch": 13.680418041804181,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.008236117097916372,
      "loss": 0.2304,
      "num_input_tokens_seen": 26243584,
      "step": 124355
    },
    {
      "epoch": 13.68096809680968,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.00823483180414577,
      "loss": 0.2319,
      "num_input_tokens_seen": 26244608,
      "step": 124360
    },
    {
      "epoch": 13.681518151815181,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008233546572726487,
      "loss": 0.2345,
      "num_input_tokens_seen": 26245664,
      "step": 124365
    },
    {
      "epoch": 13.682068206820682,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.008232261403670352,
      "loss": 0.2314,
      "num_input_tokens_seen": 26246784,
      "step": 124370
    },
    {
      "epoch": 13.682618261826182,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.008230976296989215,
      "loss": 0.2319,
      "num_input_tokens_seen": 26247840,
      "step": 124375
    },
    {
      "epoch": 13.683168316831683,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.008229691252694927,
      "loss": 0.234,
      "num_input_tokens_seen": 26248864,
      "step": 124380
    },
    {
      "epoch": 13.683718371837184,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.008228406270799321,
      "loss": 0.2303,
      "num_input_tokens_seen": 26249952,
      "step": 124385
    },
    {
      "epoch": 13.684268426842685,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00822712135131425,
      "loss": 0.233,
      "num_input_tokens_seen": 26251008,
      "step": 124390
    },
    {
      "epoch": 13.684818481848184,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.008225836494251548,
      "loss": 0.2324,
      "num_input_tokens_seen": 26252032,
      "step": 124395
    },
    {
      "epoch": 13.685368536853685,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00822455169962306,
      "loss": 0.2309,
      "num_input_tokens_seen": 26253088,
      "step": 124400
    },
    {
      "epoch": 13.685918591859187,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.008223266967440631,
      "loss": 0.2319,
      "num_input_tokens_seen": 26254144,
      "step": 124405
    },
    {
      "epoch": 13.686468646864686,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008221982297716094,
      "loss": 0.2314,
      "num_input_tokens_seen": 26255200,
      "step": 124410
    },
    {
      "epoch": 13.687018701870187,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.008220697690461302,
      "loss": 0.2304,
      "num_input_tokens_seen": 26256288,
      "step": 124415
    },
    {
      "epoch": 13.687568756875688,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.008219413145688081,
      "loss": 0.2298,
      "num_input_tokens_seen": 26257376,
      "step": 124420
    },
    {
      "epoch": 13.688118811881187,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.008218128663408274,
      "loss": 0.2304,
      "num_input_tokens_seen": 26258432,
      "step": 124425
    },
    {
      "epoch": 13.688668866886688,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.00821684424363372,
      "loss": 0.2293,
      "num_input_tokens_seen": 26259424,
      "step": 124430
    },
    {
      "epoch": 13.68921892189219,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.008215559886376258,
      "loss": 0.2314,
      "num_input_tokens_seen": 26260416,
      "step": 124435
    },
    {
      "epoch": 13.689768976897689,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.00821427559164773,
      "loss": 0.2298,
      "num_input_tokens_seen": 26261472,
      "step": 124440
    },
    {
      "epoch": 13.69031903190319,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.008212991359459968,
      "loss": 0.2293,
      "num_input_tokens_seen": 26262528,
      "step": 124445
    },
    {
      "epoch": 13.690869086908691,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.008211707189824802,
      "loss": 0.2319,
      "num_input_tokens_seen": 26263584,
      "step": 124450
    },
    {
      "epoch": 13.691419141914192,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.008210423082754072,
      "loss": 0.2314,
      "num_input_tokens_seen": 26264672,
      "step": 124455
    },
    {
      "epoch": 13.691969196919691,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008209139038259617,
      "loss": 0.2309,
      "num_input_tokens_seen": 26265792,
      "step": 124460
    },
    {
      "epoch": 13.692519251925193,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.008207855056353273,
      "loss": 0.2314,
      "num_input_tokens_seen": 26266848,
      "step": 124465
    },
    {
      "epoch": 13.693069306930694,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008206571137046871,
      "loss": 0.2299,
      "num_input_tokens_seen": 26267904,
      "step": 124470
    },
    {
      "epoch": 13.693619361936193,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.008205287280352237,
      "loss": 0.2309,
      "num_input_tokens_seen": 26268960,
      "step": 124475
    },
    {
      "epoch": 13.694169416941694,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.008204003486281216,
      "loss": 0.2314,
      "num_input_tokens_seen": 26270048,
      "step": 124480
    },
    {
      "epoch": 13.694719471947195,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.008202719754845626,
      "loss": 0.2314,
      "num_input_tokens_seen": 26271168,
      "step": 124485
    },
    {
      "epoch": 13.695269526952695,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.008201436086057308,
      "loss": 0.2319,
      "num_input_tokens_seen": 26272256,
      "step": 124490
    },
    {
      "epoch": 13.695819581958196,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.008200152479928096,
      "loss": 0.2309,
      "num_input_tokens_seen": 26273312,
      "step": 124495
    },
    {
      "epoch": 13.696369636963697,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008198868936469813,
      "loss": 0.2335,
      "num_input_tokens_seen": 26274400,
      "step": 124500
    },
    {
      "epoch": 13.696919691969196,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.008197585455694293,
      "loss": 0.2319,
      "num_input_tokens_seen": 26275392,
      "step": 124505
    },
    {
      "epoch": 13.697469746974697,
      "grad_norm": 0.00079345703125,
      "learning_rate": 0.008196302037613361,
      "loss": 0.2324,
      "num_input_tokens_seen": 26276416,
      "step": 124510
    },
    {
      "epoch": 13.698019801980198,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.008195018682238847,
      "loss": 0.2319,
      "num_input_tokens_seen": 26277472,
      "step": 124515
    },
    {
      "epoch": 13.6985698569857,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008193735389582587,
      "loss": 0.2329,
      "num_input_tokens_seen": 26278560,
      "step": 124520
    },
    {
      "epoch": 13.699119911991199,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.008192452159656397,
      "loss": 0.2304,
      "num_input_tokens_seen": 26279648,
      "step": 124525
    },
    {
      "epoch": 13.6996699669967,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.008191168992472114,
      "loss": 0.2319,
      "num_input_tokens_seen": 26280800,
      "step": 124530
    },
    {
      "epoch": 13.7002200220022,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008189885888041558,
      "loss": 0.2309,
      "num_input_tokens_seen": 26281856,
      "step": 124535
    },
    {
      "epoch": 13.7007700770077,
      "grad_norm": 0.005859375,
      "learning_rate": 0.008188602846376548,
      "loss": 0.2324,
      "num_input_tokens_seen": 26282912,
      "step": 124540
    },
    {
      "epoch": 13.701320132013201,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008187319867488921,
      "loss": 0.2308,
      "num_input_tokens_seen": 26284000,
      "step": 124545
    },
    {
      "epoch": 13.701870187018702,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008186036951390498,
      "loss": 0.2319,
      "num_input_tokens_seen": 26285056,
      "step": 124550
    },
    {
      "epoch": 13.702420242024202,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008184754098093107,
      "loss": 0.2298,
      "num_input_tokens_seen": 26286176,
      "step": 124555
    },
    {
      "epoch": 13.702970297029703,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.008183471307608568,
      "loss": 0.2345,
      "num_input_tokens_seen": 26287232,
      "step": 124560
    },
    {
      "epoch": 13.703520352035204,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.008182188579948697,
      "loss": 0.2335,
      "num_input_tokens_seen": 26288256,
      "step": 124565
    },
    {
      "epoch": 13.704070407040705,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.008180905915125327,
      "loss": 0.2319,
      "num_input_tokens_seen": 26289312,
      "step": 124570
    },
    {
      "epoch": 13.704620462046204,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008179623313150265,
      "loss": 0.2351,
      "num_input_tokens_seen": 26290400,
      "step": 124575
    },
    {
      "epoch": 13.705170517051705,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.008178340774035352,
      "loss": 0.2319,
      "num_input_tokens_seen": 26291488,
      "step": 124580
    },
    {
      "epoch": 13.705720572057206,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.008177058297792401,
      "loss": 0.2319,
      "num_input_tokens_seen": 26292544,
      "step": 124585
    },
    {
      "epoch": 13.706270627062706,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008175775884433223,
      "loss": 0.234,
      "num_input_tokens_seen": 26293600,
      "step": 124590
    },
    {
      "epoch": 13.706820682068207,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00817449353396965,
      "loss": 0.2335,
      "num_input_tokens_seen": 26294720,
      "step": 124595
    },
    {
      "epoch": 13.707370737073708,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008173211246413487,
      "loss": 0.2314,
      "num_input_tokens_seen": 26295776,
      "step": 124600
    },
    {
      "epoch": 13.707920792079207,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.008171929021776563,
      "loss": 0.2309,
      "num_input_tokens_seen": 26296864,
      "step": 124605
    },
    {
      "epoch": 13.708470847084708,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008170646860070697,
      "loss": 0.2309,
      "num_input_tokens_seen": 26297984,
      "step": 124610
    },
    {
      "epoch": 13.70902090209021,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.008169364761307696,
      "loss": 0.2319,
      "num_input_tokens_seen": 26299136,
      "step": 124615
    },
    {
      "epoch": 13.70957095709571,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.00816808272549939,
      "loss": 0.2303,
      "num_input_tokens_seen": 26300160,
      "step": 124620
    },
    {
      "epoch": 13.71012101210121,
      "grad_norm": 0.0004119873046875,
      "learning_rate": 0.008166800752657579,
      "loss": 0.2298,
      "num_input_tokens_seen": 26301184,
      "step": 124625
    },
    {
      "epoch": 13.710671067106711,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008165518842794095,
      "loss": 0.2313,
      "num_input_tokens_seen": 26302240,
      "step": 124630
    },
    {
      "epoch": 13.711221122112212,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.008164236995920735,
      "loss": 0.2314,
      "num_input_tokens_seen": 26303328,
      "step": 124635
    },
    {
      "epoch": 13.711771177117711,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.008162955212049326,
      "loss": 0.2319,
      "num_input_tokens_seen": 26304480,
      "step": 124640
    },
    {
      "epoch": 13.712321232123212,
      "grad_norm": 0.0006256103515625,
      "learning_rate": 0.00816167349119168,
      "loss": 0.2329,
      "num_input_tokens_seen": 26305504,
      "step": 124645
    },
    {
      "epoch": 13.712871287128714,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008160391833359606,
      "loss": 0.2324,
      "num_input_tokens_seen": 26306560,
      "step": 124650
    },
    {
      "epoch": 13.713421342134213,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.008159110238564924,
      "loss": 0.2319,
      "num_input_tokens_seen": 26307616,
      "step": 124655
    },
    {
      "epoch": 13.713971397139714,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.008157828706819432,
      "loss": 0.2304,
      "num_input_tokens_seen": 26308640,
      "step": 124660
    },
    {
      "epoch": 13.714521452145215,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.008156547238134951,
      "loss": 0.2303,
      "num_input_tokens_seen": 26309760,
      "step": 124665
    },
    {
      "epoch": 13.715071507150714,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.008155265832523297,
      "loss": 0.2309,
      "num_input_tokens_seen": 26310816,
      "step": 124670
    },
    {
      "epoch": 13.715621562156215,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008153984489996267,
      "loss": 0.233,
      "num_input_tokens_seen": 26311840,
      "step": 124675
    },
    {
      "epoch": 13.716171617161717,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008152703210565682,
      "loss": 0.2324,
      "num_input_tokens_seen": 26312864,
      "step": 124680
    },
    {
      "epoch": 13.716721672167218,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.008151421994243346,
      "loss": 0.2324,
      "num_input_tokens_seen": 26313920,
      "step": 124685
    },
    {
      "epoch": 13.717271727172717,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.008150140841041058,
      "loss": 0.2309,
      "num_input_tokens_seen": 26315040,
      "step": 124690
    },
    {
      "epoch": 13.717821782178218,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008148859750970643,
      "loss": 0.2298,
      "num_input_tokens_seen": 26316128,
      "step": 124695
    },
    {
      "epoch": 13.718371837183719,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.008147578724043895,
      "loss": 0.2298,
      "num_input_tokens_seen": 26317248,
      "step": 124700
    },
    {
      "epoch": 13.718921892189218,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.008146297760272633,
      "loss": 0.2298,
      "num_input_tokens_seen": 26318336,
      "step": 124705
    },
    {
      "epoch": 13.71947194719472,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.008145016859668657,
      "loss": 0.2334,
      "num_input_tokens_seen": 26319392,
      "step": 124710
    },
    {
      "epoch": 13.72002200220022,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.008143736022243763,
      "loss": 0.2298,
      "num_input_tokens_seen": 26320384,
      "step": 124715
    },
    {
      "epoch": 13.72057205720572,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.008142455248009765,
      "loss": 0.2293,
      "num_input_tokens_seen": 26321504,
      "step": 124720
    },
    {
      "epoch": 13.721122112211221,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.008141174536978473,
      "loss": 0.2329,
      "num_input_tokens_seen": 26322496,
      "step": 124725
    },
    {
      "epoch": 13.721672167216722,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.008139893889161677,
      "loss": 0.2277,
      "num_input_tokens_seen": 26323520,
      "step": 124730
    },
    {
      "epoch": 13.722222222222221,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.008138613304571192,
      "loss": 0.2324,
      "num_input_tokens_seen": 26324608,
      "step": 124735
    },
    {
      "epoch": 13.722772277227723,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.008137332783218813,
      "loss": 0.2329,
      "num_input_tokens_seen": 26325696,
      "step": 124740
    },
    {
      "epoch": 13.723322332233224,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00813605232511635,
      "loss": 0.2319,
      "num_input_tokens_seen": 26326816,
      "step": 124745
    },
    {
      "epoch": 13.723872387238725,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.00813477193027559,
      "loss": 0.2319,
      "num_input_tokens_seen": 26327840,
      "step": 124750
    },
    {
      "epoch": 13.724422442244224,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.008133491598708346,
      "loss": 0.2298,
      "num_input_tokens_seen": 26328896,
      "step": 124755
    },
    {
      "epoch": 13.724972497249725,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.008132211330426421,
      "loss": 0.2319,
      "num_input_tokens_seen": 26329984,
      "step": 124760
    },
    {
      "epoch": 13.725522552255226,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.0081309311254416,
      "loss": 0.233,
      "num_input_tokens_seen": 26331040,
      "step": 124765
    },
    {
      "epoch": 13.726072607260726,
      "grad_norm": 0.0029144287109375,
      "learning_rate": 0.0081296509837657,
      "loss": 0.2293,
      "num_input_tokens_seen": 26332160,
      "step": 124770
    },
    {
      "epoch": 13.726622662266227,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.008128370905410503,
      "loss": 0.2283,
      "num_input_tokens_seen": 26333248,
      "step": 124775
    },
    {
      "epoch": 13.727172717271728,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008127090890387814,
      "loss": 0.2309,
      "num_input_tokens_seen": 26334240,
      "step": 124780
    },
    {
      "epoch": 13.727722772277227,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.008125810938709434,
      "loss": 0.233,
      "num_input_tokens_seen": 26335328,
      "step": 124785
    },
    {
      "epoch": 13.728272827282728,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008124531050387152,
      "loss": 0.2314,
      "num_input_tokens_seen": 26336384,
      "step": 124790
    },
    {
      "epoch": 13.72882288228823,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008123251225432775,
      "loss": 0.2314,
      "num_input_tokens_seen": 26337408,
      "step": 124795
    },
    {
      "epoch": 13.729372937293729,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.008121971463858091,
      "loss": 0.2309,
      "num_input_tokens_seen": 26338432,
      "step": 124800
    },
    {
      "epoch": 13.72992299229923,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00812069176567489,
      "loss": 0.2324,
      "num_input_tokens_seen": 26339456,
      "step": 124805
    },
    {
      "epoch": 13.73047304730473,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.008119412130894971,
      "loss": 0.2309,
      "num_input_tokens_seen": 26340448,
      "step": 124810
    },
    {
      "epoch": 13.731023102310232,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.008118132559530129,
      "loss": 0.2293,
      "num_input_tokens_seen": 26341536,
      "step": 124815
    },
    {
      "epoch": 13.731573157315731,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008116853051592163,
      "loss": 0.2314,
      "num_input_tokens_seen": 26342528,
      "step": 124820
    },
    {
      "epoch": 13.732123212321232,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.00811557360709286,
      "loss": 0.2324,
      "num_input_tokens_seen": 26343616,
      "step": 124825
    },
    {
      "epoch": 13.732673267326733,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008114294226044006,
      "loss": 0.2335,
      "num_input_tokens_seen": 26344640,
      "step": 124830
    },
    {
      "epoch": 13.733223322332233,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008113014908457398,
      "loss": 0.2319,
      "num_input_tokens_seen": 26345696,
      "step": 124835
    },
    {
      "epoch": 13.733773377337734,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.008111735654344825,
      "loss": 0.2303,
      "num_input_tokens_seen": 26346720,
      "step": 124840
    },
    {
      "epoch": 13.734323432343235,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008110456463718087,
      "loss": 0.2293,
      "num_input_tokens_seen": 26347840,
      "step": 124845
    },
    {
      "epoch": 13.734873487348734,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.008109177336588964,
      "loss": 0.2308,
      "num_input_tokens_seen": 26348896,
      "step": 124850
    },
    {
      "epoch": 13.735423542354235,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.008107898272969243,
      "loss": 0.2303,
      "num_input_tokens_seen": 26349952,
      "step": 124855
    },
    {
      "epoch": 13.735973597359736,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008106619272870721,
      "loss": 0.2319,
      "num_input_tokens_seen": 26351008,
      "step": 124860
    },
    {
      "epoch": 13.736523652365236,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.008105340336305177,
      "loss": 0.2324,
      "num_input_tokens_seen": 26352064,
      "step": 124865
    },
    {
      "epoch": 13.737073707370737,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.008104061463284401,
      "loss": 0.2303,
      "num_input_tokens_seen": 26353120,
      "step": 124870
    },
    {
      "epoch": 13.737623762376238,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.00810278265382019,
      "loss": 0.233,
      "num_input_tokens_seen": 26354112,
      "step": 124875
    },
    {
      "epoch": 13.738173817381739,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.008101503907924313,
      "loss": 0.2309,
      "num_input_tokens_seen": 26355168,
      "step": 124880
    },
    {
      "epoch": 13.738723872387238,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.00810022522560857,
      "loss": 0.2298,
      "num_input_tokens_seen": 26356160,
      "step": 124885
    },
    {
      "epoch": 13.73927392739274,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.008098946606884736,
      "loss": 0.2324,
      "num_input_tokens_seen": 26357248,
      "step": 124890
    },
    {
      "epoch": 13.73982398239824,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0080976680517646,
      "loss": 0.2324,
      "num_input_tokens_seen": 26358304,
      "step": 124895
    },
    {
      "epoch": 13.74037403740374,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00809638956025995,
      "loss": 0.2314,
      "num_input_tokens_seen": 26359360,
      "step": 124900
    },
    {
      "epoch": 13.74092409240924,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.00809511113238256,
      "loss": 0.2298,
      "num_input_tokens_seen": 26360480,
      "step": 124905
    },
    {
      "epoch": 13.741474147414742,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.008093832768144222,
      "loss": 0.2298,
      "num_input_tokens_seen": 26361568,
      "step": 124910
    },
    {
      "epoch": 13.742024202420241,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.008092554467556716,
      "loss": 0.2303,
      "num_input_tokens_seen": 26362656,
      "step": 124915
    },
    {
      "epoch": 13.742574257425742,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.008091276230631814,
      "loss": 0.2324,
      "num_input_tokens_seen": 26363776,
      "step": 124920
    },
    {
      "epoch": 13.743124312431243,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.008089998057381302,
      "loss": 0.2335,
      "num_input_tokens_seen": 26364800,
      "step": 124925
    },
    {
      "epoch": 13.743674367436743,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.008088719947816963,
      "loss": 0.2303,
      "num_input_tokens_seen": 26365888,
      "step": 124930
    },
    {
      "epoch": 13.744224422442244,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.008087441901950583,
      "loss": 0.2303,
      "num_input_tokens_seen": 26366976,
      "step": 124935
    },
    {
      "epoch": 13.744774477447745,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.008086163919793932,
      "loss": 0.2319,
      "num_input_tokens_seen": 26368000,
      "step": 124940
    },
    {
      "epoch": 13.745324532453246,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.008084886001358786,
      "loss": 0.2314,
      "num_input_tokens_seen": 26369024,
      "step": 124945
    },
    {
      "epoch": 13.745874587458745,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.008083608146656929,
      "loss": 0.2335,
      "num_input_tokens_seen": 26370080,
      "step": 124950
    },
    {
      "epoch": 13.746424642464246,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.008082330355700135,
      "loss": 0.2324,
      "num_input_tokens_seen": 26371168,
      "step": 124955
    },
    {
      "epoch": 13.746974697469748,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.008081052628500188,
      "loss": 0.2293,
      "num_input_tokens_seen": 26372160,
      "step": 124960
    },
    {
      "epoch": 13.747524752475247,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.00807977496506886,
      "loss": 0.2309,
      "num_input_tokens_seen": 26373216,
      "step": 124965
    },
    {
      "epoch": 13.748074807480748,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008078497365417919,
      "loss": 0.2314,
      "num_input_tokens_seen": 26374272,
      "step": 124970
    },
    {
      "epoch": 13.748624862486249,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008077219829559152,
      "loss": 0.2304,
      "num_input_tokens_seen": 26375296,
      "step": 124975
    },
    {
      "epoch": 13.749174917491748,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.008075942357504322,
      "loss": 0.2314,
      "num_input_tokens_seen": 26376320,
      "step": 124980
    },
    {
      "epoch": 13.74972497249725,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00807466494926521,
      "loss": 0.2303,
      "num_input_tokens_seen": 26377408,
      "step": 124985
    },
    {
      "epoch": 13.75027502750275,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008073387604853596,
      "loss": 0.2313,
      "num_input_tokens_seen": 26378464,
      "step": 124990
    },
    {
      "epoch": 13.750825082508252,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.008072110324281234,
      "loss": 0.233,
      "num_input_tokens_seen": 26379584,
      "step": 124995
    },
    {
      "epoch": 13.751375137513751,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.008070833107559915,
      "loss": 0.2298,
      "num_input_tokens_seen": 26380640,
      "step": 125000
    },
    {
      "epoch": 13.751925192519252,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.008069555954701397,
      "loss": 0.2319,
      "num_input_tokens_seen": 26381664,
      "step": 125005
    },
    {
      "epoch": 13.752475247524753,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.008068278865717456,
      "loss": 0.2303,
      "num_input_tokens_seen": 26382720,
      "step": 125010
    },
    {
      "epoch": 13.753025302530252,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.00806700184061987,
      "loss": 0.2314,
      "num_input_tokens_seen": 26383872,
      "step": 125015
    },
    {
      "epoch": 13.753575357535754,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.008065724879420392,
      "loss": 0.2303,
      "num_input_tokens_seen": 26384928,
      "step": 125020
    },
    {
      "epoch": 13.754125412541255,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.00806444798213081,
      "loss": 0.2309,
      "num_input_tokens_seen": 26385952,
      "step": 125025
    },
    {
      "epoch": 13.754675467546754,
      "grad_norm": 0.000652313232421875,
      "learning_rate": 0.008063171148762873,
      "loss": 0.2324,
      "num_input_tokens_seen": 26386944,
      "step": 125030
    },
    {
      "epoch": 13.755225522552255,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.008061894379328366,
      "loss": 0.2319,
      "num_input_tokens_seen": 26388064,
      "step": 125035
    },
    {
      "epoch": 13.755775577557756,
      "grad_norm": 0.000782012939453125,
      "learning_rate": 0.008060617673839044,
      "loss": 0.2293,
      "num_input_tokens_seen": 26389120,
      "step": 125040
    },
    {
      "epoch": 13.756325632563257,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.00805934103230668,
      "loss": 0.2304,
      "num_input_tokens_seen": 26390208,
      "step": 125045
    },
    {
      "epoch": 13.756875687568757,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.008058064454743044,
      "loss": 0.2314,
      "num_input_tokens_seen": 26391264,
      "step": 125050
    },
    {
      "epoch": 13.757425742574258,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00805678794115989,
      "loss": 0.2314,
      "num_input_tokens_seen": 26392352,
      "step": 125055
    },
    {
      "epoch": 13.757975797579759,
      "grad_norm": 0.00069427490234375,
      "learning_rate": 0.008055511491568999,
      "loss": 0.2309,
      "num_input_tokens_seen": 26393376,
      "step": 125060
    },
    {
      "epoch": 13.758525852585258,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.008054235105982119,
      "loss": 0.2329,
      "num_input_tokens_seen": 26394400,
      "step": 125065
    },
    {
      "epoch": 13.75907590759076,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.008052958784411021,
      "loss": 0.2335,
      "num_input_tokens_seen": 26395456,
      "step": 125070
    },
    {
      "epoch": 13.75962596259626,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.008051682526867473,
      "loss": 0.2309,
      "num_input_tokens_seen": 26396480,
      "step": 125075
    },
    {
      "epoch": 13.76017601760176,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.008050406333363227,
      "loss": 0.2314,
      "num_input_tokens_seen": 26397472,
      "step": 125080
    },
    {
      "epoch": 13.76072607260726,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.008049130203910057,
      "loss": 0.2303,
      "num_input_tokens_seen": 26398528,
      "step": 125085
    },
    {
      "epoch": 13.761276127612762,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.008047854138519722,
      "loss": 0.2319,
      "num_input_tokens_seen": 26399648,
      "step": 125090
    },
    {
      "epoch": 13.761826182618261,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.008046578137203969,
      "loss": 0.2298,
      "num_input_tokens_seen": 26400704,
      "step": 125095
    },
    {
      "epoch": 13.762376237623762,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.00804530219997457,
      "loss": 0.2319,
      "num_input_tokens_seen": 26401728,
      "step": 125100
    },
    {
      "epoch": 13.762926292629263,
      "grad_norm": 0.00055694580078125,
      "learning_rate": 0.008044026326843291,
      "loss": 0.2319,
      "num_input_tokens_seen": 26402816,
      "step": 125105
    },
    {
      "epoch": 13.763476347634764,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.008042750517821877,
      "loss": 0.2303,
      "num_input_tokens_seen": 26403936,
      "step": 125110
    },
    {
      "epoch": 13.764026402640264,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0080414747729221,
      "loss": 0.2314,
      "num_input_tokens_seen": 26404928,
      "step": 125115
    },
    {
      "epoch": 13.764576457645765,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.008040199092155701,
      "loss": 0.2303,
      "num_input_tokens_seen": 26405952,
      "step": 125120
    },
    {
      "epoch": 13.765126512651266,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.008038923475534452,
      "loss": 0.2309,
      "num_input_tokens_seen": 26406976,
      "step": 125125
    },
    {
      "epoch": 13.765676567656765,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.00803764792307011,
      "loss": 0.2298,
      "num_input_tokens_seen": 26408000,
      "step": 125130
    },
    {
      "epoch": 13.766226622662266,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.008036372434774418,
      "loss": 0.2335,
      "num_input_tokens_seen": 26409088,
      "step": 125135
    },
    {
      "epoch": 13.766776677667767,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.008035097010659147,
      "loss": 0.2324,
      "num_input_tokens_seen": 26410112,
      "step": 125140
    },
    {
      "epoch": 13.767326732673267,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.00803382165073604,
      "loss": 0.2319,
      "num_input_tokens_seen": 26411168,
      "step": 125145
    },
    {
      "epoch": 13.767876787678768,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.008032546355016861,
      "loss": 0.2283,
      "num_input_tokens_seen": 26412160,
      "step": 125150
    },
    {
      "epoch": 13.768426842684269,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.008031271123513354,
      "loss": 0.2303,
      "num_input_tokens_seen": 26413184,
      "step": 125155
    },
    {
      "epoch": 13.768976897689768,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.008029995956237275,
      "loss": 0.2303,
      "num_input_tokens_seen": 26414176,
      "step": 125160
    },
    {
      "epoch": 13.76952695269527,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008028720853200387,
      "loss": 0.2324,
      "num_input_tokens_seen": 26415232,
      "step": 125165
    },
    {
      "epoch": 13.77007700770077,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.008027445814414426,
      "loss": 0.2308,
      "num_input_tokens_seen": 26416320,
      "step": 125170
    },
    {
      "epoch": 13.770627062706271,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.008026170839891157,
      "loss": 0.2314,
      "num_input_tokens_seen": 26417440,
      "step": 125175
    },
    {
      "epoch": 13.77117711771177,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.00802489592964232,
      "loss": 0.233,
      "num_input_tokens_seen": 26418560,
      "step": 125180
    },
    {
      "epoch": 13.771727172717272,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00802362108367967,
      "loss": 0.2299,
      "num_input_tokens_seen": 26419616,
      "step": 125185
    },
    {
      "epoch": 13.772277227722773,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.008022346302014963,
      "loss": 0.2314,
      "num_input_tokens_seen": 26420704,
      "step": 125190
    },
    {
      "epoch": 13.772827282728272,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.008021071584659936,
      "loss": 0.233,
      "num_input_tokens_seen": 26421696,
      "step": 125195
    },
    {
      "epoch": 13.773377337733773,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00801979693162635,
      "loss": 0.2304,
      "num_input_tokens_seen": 26422752,
      "step": 125200
    },
    {
      "epoch": 13.773927392739274,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.008018522342925945,
      "loss": 0.2324,
      "num_input_tokens_seen": 26423776,
      "step": 125205
    },
    {
      "epoch": 13.774477447744774,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.008017247818570465,
      "loss": 0.2288,
      "num_input_tokens_seen": 26424928,
      "step": 125210
    },
    {
      "epoch": 13.775027502750275,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.00801597335857166,
      "loss": 0.233,
      "num_input_tokens_seen": 26426016,
      "step": 125215
    },
    {
      "epoch": 13.775577557755776,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.008014698962941278,
      "loss": 0.2309,
      "num_input_tokens_seen": 26427008,
      "step": 125220
    },
    {
      "epoch": 13.776127612761275,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.00801342463169107,
      "loss": 0.2319,
      "num_input_tokens_seen": 26428064,
      "step": 125225
    },
    {
      "epoch": 13.776677667766776,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.00801215036483277,
      "loss": 0.2319,
      "num_input_tokens_seen": 26429056,
      "step": 125230
    },
    {
      "epoch": 13.777227722772277,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.008010876162378127,
      "loss": 0.2314,
      "num_input_tokens_seen": 26430176,
      "step": 125235
    },
    {
      "epoch": 13.777777777777779,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.008009602024338884,
      "loss": 0.2314,
      "num_input_tokens_seen": 26431200,
      "step": 125240
    },
    {
      "epoch": 13.778327832783278,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.008008327950726783,
      "loss": 0.2335,
      "num_input_tokens_seen": 26432256,
      "step": 125245
    },
    {
      "epoch": 13.778877887788779,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.008007053941553575,
      "loss": 0.234,
      "num_input_tokens_seen": 26433216,
      "step": 125250
    },
    {
      "epoch": 13.77942794279428,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.008005779996830996,
      "loss": 0.2303,
      "num_input_tokens_seen": 26434304,
      "step": 125255
    },
    {
      "epoch": 13.77997799779978,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.008004506116570777,
      "loss": 0.2319,
      "num_input_tokens_seen": 26435328,
      "step": 125260
    },
    {
      "epoch": 13.78052805280528,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00800323230078468,
      "loss": 0.2314,
      "num_input_tokens_seen": 26436448,
      "step": 125265
    },
    {
      "epoch": 13.781078107810782,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.008001958549484423,
      "loss": 0.2329,
      "num_input_tokens_seen": 26437504,
      "step": 125270
    },
    {
      "epoch": 13.781628162816281,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.008000684862681757,
      "loss": 0.2293,
      "num_input_tokens_seen": 26438560,
      "step": 125275
    },
    {
      "epoch": 13.782178217821782,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.007999411240388428,
      "loss": 0.2319,
      "num_input_tokens_seen": 26439520,
      "step": 125280
    },
    {
      "epoch": 13.782728272827283,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.00799813768261616,
      "loss": 0.2319,
      "num_input_tokens_seen": 26440544,
      "step": 125285
    },
    {
      "epoch": 13.783278327832782,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.007996864189376701,
      "loss": 0.2319,
      "num_input_tokens_seen": 26441632,
      "step": 125290
    },
    {
      "epoch": 13.783828382838283,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.007995590760681785,
      "loss": 0.2309,
      "num_input_tokens_seen": 26442624,
      "step": 125295
    },
    {
      "epoch": 13.784378437843785,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.007994317396543136,
      "loss": 0.2335,
      "num_input_tokens_seen": 26443712,
      "step": 125300
    },
    {
      "epoch": 13.784928492849286,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007993044096972516,
      "loss": 0.2319,
      "num_input_tokens_seen": 26444768,
      "step": 125305
    },
    {
      "epoch": 13.785478547854785,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.00799177086198164,
      "loss": 0.2309,
      "num_input_tokens_seen": 26445792,
      "step": 125310
    },
    {
      "epoch": 13.786028602860286,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.007990497691582252,
      "loss": 0.2303,
      "num_input_tokens_seen": 26446848,
      "step": 125315
    },
    {
      "epoch": 13.786578657865787,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.007989224585786084,
      "loss": 0.2319,
      "num_input_tokens_seen": 26447872,
      "step": 125320
    },
    {
      "epoch": 13.787128712871286,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.007987951544604866,
      "loss": 0.2298,
      "num_input_tokens_seen": 26448960,
      "step": 125325
    },
    {
      "epoch": 13.787678767876788,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.007986678568050333,
      "loss": 0.2324,
      "num_input_tokens_seen": 26450016,
      "step": 125330
    },
    {
      "epoch": 13.788228822882289,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.007985405656134217,
      "loss": 0.2314,
      "num_input_tokens_seen": 26451040,
      "step": 125335
    },
    {
      "epoch": 13.788778877887788,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.007984132808868255,
      "loss": 0.2314,
      "num_input_tokens_seen": 26452032,
      "step": 125340
    },
    {
      "epoch": 13.789328932893289,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.007982860026264178,
      "loss": 0.2308,
      "num_input_tokens_seen": 26453056,
      "step": 125345
    },
    {
      "epoch": 13.78987898789879,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007981587308333706,
      "loss": 0.2314,
      "num_input_tokens_seen": 26454112,
      "step": 125350
    },
    {
      "epoch": 13.79042904290429,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007980314655088581,
      "loss": 0.2319,
      "num_input_tokens_seen": 26455168,
      "step": 125355
    },
    {
      "epoch": 13.79097909790979,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.007979042066540517,
      "loss": 0.2298,
      "num_input_tokens_seen": 26456192,
      "step": 125360
    },
    {
      "epoch": 13.791529152915292,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007977769542701263,
      "loss": 0.2309,
      "num_input_tokens_seen": 26457280,
      "step": 125365
    },
    {
      "epoch": 13.792079207920793,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.00797649708358254,
      "loss": 0.2324,
      "num_input_tokens_seen": 26458272,
      "step": 125370
    },
    {
      "epoch": 13.792629262926292,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.007975224689196066,
      "loss": 0.2324,
      "num_input_tokens_seen": 26459424,
      "step": 125375
    },
    {
      "epoch": 13.793179317931793,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00797395235955358,
      "loss": 0.2314,
      "num_input_tokens_seen": 26460512,
      "step": 125380
    },
    {
      "epoch": 13.793729372937294,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.0079726800946668,
      "loss": 0.2298,
      "num_input_tokens_seen": 26461600,
      "step": 125385
    },
    {
      "epoch": 13.794279427942794,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.007971407894547457,
      "loss": 0.2303,
      "num_input_tokens_seen": 26462688,
      "step": 125390
    },
    {
      "epoch": 13.794829482948295,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.007970135759207277,
      "loss": 0.2335,
      "num_input_tokens_seen": 26463744,
      "step": 125395
    },
    {
      "epoch": 13.795379537953796,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.007968863688657978,
      "loss": 0.2324,
      "num_input_tokens_seen": 26464832,
      "step": 125400
    },
    {
      "epoch": 13.795929592959295,
      "grad_norm": 0.005859375,
      "learning_rate": 0.007967591682911295,
      "loss": 0.2324,
      "num_input_tokens_seen": 26465888,
      "step": 125405
    },
    {
      "epoch": 13.796479647964796,
      "grad_norm": 0.005859375,
      "learning_rate": 0.007966319741978936,
      "loss": 0.2314,
      "num_input_tokens_seen": 26467008,
      "step": 125410
    },
    {
      "epoch": 13.797029702970297,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.007965047865872642,
      "loss": 0.2314,
      "num_input_tokens_seen": 26468096,
      "step": 125415
    },
    {
      "epoch": 13.797579757975798,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007963776054604119,
      "loss": 0.2319,
      "num_input_tokens_seen": 26469120,
      "step": 125420
    },
    {
      "epoch": 13.798129812981298,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007962504308185098,
      "loss": 0.2314,
      "num_input_tokens_seen": 26470144,
      "step": 125425
    },
    {
      "epoch": 13.798679867986799,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0079612326266273,
      "loss": 0.2313,
      "num_input_tokens_seen": 26471232,
      "step": 125430
    },
    {
      "epoch": 13.7992299229923,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.007959961009942439,
      "loss": 0.2308,
      "num_input_tokens_seen": 26472320,
      "step": 125435
    },
    {
      "epoch": 13.7997799779978,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007958689458142248,
      "loss": 0.2345,
      "num_input_tokens_seen": 26473344,
      "step": 125440
    },
    {
      "epoch": 13.8003300330033,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.007957417971238428,
      "loss": 0.2324,
      "num_input_tokens_seen": 26474432,
      "step": 125445
    },
    {
      "epoch": 13.800880088008801,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007956146549242708,
      "loss": 0.2324,
      "num_input_tokens_seen": 26475488,
      "step": 125450
    },
    {
      "epoch": 13.8014301430143,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.00795487519216681,
      "loss": 0.2303,
      "num_input_tokens_seen": 26476544,
      "step": 125455
    },
    {
      "epoch": 13.801980198019802,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.007953603900022444,
      "loss": 0.2303,
      "num_input_tokens_seen": 26477600,
      "step": 125460
    },
    {
      "epoch": 13.802530253025303,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.007952332672821333,
      "loss": 0.2319,
      "num_input_tokens_seen": 26478624,
      "step": 125465
    },
    {
      "epoch": 13.803080308030804,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007951061510575192,
      "loss": 0.2319,
      "num_input_tokens_seen": 26479584,
      "step": 125470
    },
    {
      "epoch": 13.803630363036303,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.007949790413295722,
      "loss": 0.2288,
      "num_input_tokens_seen": 26480608,
      "step": 125475
    },
    {
      "epoch": 13.804180418041804,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.007948519380994662,
      "loss": 0.2324,
      "num_input_tokens_seen": 26481728,
      "step": 125480
    },
    {
      "epoch": 13.804730473047305,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.007947248413683716,
      "loss": 0.2324,
      "num_input_tokens_seen": 26482816,
      "step": 125485
    },
    {
      "epoch": 13.805280528052805,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00794597751137459,
      "loss": 0.2288,
      "num_input_tokens_seen": 26483840,
      "step": 125490
    },
    {
      "epoch": 13.805830583058306,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00794470667407901,
      "loss": 0.2303,
      "num_input_tokens_seen": 26484864,
      "step": 125495
    },
    {
      "epoch": 13.806380638063807,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00794343590180868,
      "loss": 0.2283,
      "num_input_tokens_seen": 26485920,
      "step": 125500
    },
    {
      "epoch": 13.806930693069306,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.007942165194575313,
      "loss": 0.2314,
      "num_input_tokens_seen": 26487040,
      "step": 125505
    },
    {
      "epoch": 13.807480748074807,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.00794089455239063,
      "loss": 0.2278,
      "num_input_tokens_seen": 26488128,
      "step": 125510
    },
    {
      "epoch": 13.808030803080309,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.007939623975266328,
      "loss": 0.2309,
      "num_input_tokens_seen": 26489152,
      "step": 125515
    },
    {
      "epoch": 13.808580858085808,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.007938353463214129,
      "loss": 0.2314,
      "num_input_tokens_seen": 26490240,
      "step": 125520
    },
    {
      "epoch": 13.809130913091309,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.007937083016245733,
      "loss": 0.2319,
      "num_input_tokens_seen": 26491360,
      "step": 125525
    },
    {
      "epoch": 13.80968096809681,
      "grad_norm": 0.0030670166015625,
      "learning_rate": 0.007935812634372858,
      "loss": 0.2314,
      "num_input_tokens_seen": 26492384,
      "step": 125530
    },
    {
      "epoch": 13.810231023102311,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007934542317607204,
      "loss": 0.2314,
      "num_input_tokens_seen": 26493376,
      "step": 125535
    },
    {
      "epoch": 13.81078107810781,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.007933272065960482,
      "loss": 0.2303,
      "num_input_tokens_seen": 26494400,
      "step": 125540
    },
    {
      "epoch": 13.811331133113312,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.007932001879444407,
      "loss": 0.2319,
      "num_input_tokens_seen": 26495424,
      "step": 125545
    },
    {
      "epoch": 13.811881188118813,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.007930731758070674,
      "loss": 0.2303,
      "num_input_tokens_seen": 26496512,
      "step": 125550
    },
    {
      "epoch": 13.812431243124312,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.007929461701850996,
      "loss": 0.2298,
      "num_input_tokens_seen": 26497536,
      "step": 125555
    },
    {
      "epoch": 13.812981298129813,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.007928191710797077,
      "loss": 0.2319,
      "num_input_tokens_seen": 26498560,
      "step": 125560
    },
    {
      "epoch": 13.813531353135314,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.007926921784920618,
      "loss": 0.2303,
      "num_input_tokens_seen": 26499584,
      "step": 125565
    },
    {
      "epoch": 13.814081408140813,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.007925651924233333,
      "loss": 0.2314,
      "num_input_tokens_seen": 26500640,
      "step": 125570
    },
    {
      "epoch": 13.814631463146315,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007924382128746915,
      "loss": 0.2308,
      "num_input_tokens_seen": 26501696,
      "step": 125575
    },
    {
      "epoch": 13.815181518151816,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.007923112398473075,
      "loss": 0.2293,
      "num_input_tokens_seen": 26502720,
      "step": 125580
    },
    {
      "epoch": 13.815731573157315,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.007921842733423515,
      "loss": 0.2298,
      "num_input_tokens_seen": 26503744,
      "step": 125585
    },
    {
      "epoch": 13.816281628162816,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.007920573133609929,
      "loss": 0.2319,
      "num_input_tokens_seen": 26504800,
      "step": 125590
    },
    {
      "epoch": 13.816831683168317,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00791930359904402,
      "loss": 0.2309,
      "num_input_tokens_seen": 26505792,
      "step": 125595
    },
    {
      "epoch": 13.817381738173818,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.007918034129737492,
      "loss": 0.2329,
      "num_input_tokens_seen": 26506784,
      "step": 125600
    },
    {
      "epoch": 13.817931793179318,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.007916764725702054,
      "loss": 0.2314,
      "num_input_tokens_seen": 26507808,
      "step": 125605
    },
    {
      "epoch": 13.818481848184819,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.007915495386949396,
      "loss": 0.2303,
      "num_input_tokens_seen": 26508864,
      "step": 125610
    },
    {
      "epoch": 13.81903190319032,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00791422611349121,
      "loss": 0.2329,
      "num_input_tokens_seen": 26509952,
      "step": 125615
    },
    {
      "epoch": 13.819581958195819,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.007912956905339203,
      "loss": 0.2314,
      "num_input_tokens_seen": 26511008,
      "step": 125620
    },
    {
      "epoch": 13.82013201320132,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.007911687762505072,
      "loss": 0.2335,
      "num_input_tokens_seen": 26512064,
      "step": 125625
    },
    {
      "epoch": 13.820682068206821,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007910418685000517,
      "loss": 0.2314,
      "num_input_tokens_seen": 26513152,
      "step": 125630
    },
    {
      "epoch": 13.82123212321232,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.007909149672837235,
      "loss": 0.2324,
      "num_input_tokens_seen": 26514144,
      "step": 125635
    },
    {
      "epoch": 13.821782178217822,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.00790788072602691,
      "loss": 0.2314,
      "num_input_tokens_seen": 26515200,
      "step": 125640
    },
    {
      "epoch": 13.822332233223323,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.00790661184458125,
      "loss": 0.2309,
      "num_input_tokens_seen": 26516224,
      "step": 125645
    },
    {
      "epoch": 13.822882288228822,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00790534302851194,
      "loss": 0.2319,
      "num_input_tokens_seen": 26517312,
      "step": 125650
    },
    {
      "epoch": 13.823432343234323,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.007904074277830677,
      "loss": 0.2308,
      "num_input_tokens_seen": 26518368,
      "step": 125655
    },
    {
      "epoch": 13.823982398239824,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.007902805592549167,
      "loss": 0.2309,
      "num_input_tokens_seen": 26519424,
      "step": 125660
    },
    {
      "epoch": 13.824532453245325,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007901536972679084,
      "loss": 0.2324,
      "num_input_tokens_seen": 26520512,
      "step": 125665
    },
    {
      "epoch": 13.825082508250825,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.007900268418232133,
      "loss": 0.2319,
      "num_input_tokens_seen": 26521536,
      "step": 125670
    },
    {
      "epoch": 13.825632563256326,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.007898999929219996,
      "loss": 0.2309,
      "num_input_tokens_seen": 26522592,
      "step": 125675
    },
    {
      "epoch": 13.826182618261827,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.007897731505654369,
      "loss": 0.233,
      "num_input_tokens_seen": 26523712,
      "step": 125680
    },
    {
      "epoch": 13.826732673267326,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.00789646314754695,
      "loss": 0.2298,
      "num_input_tokens_seen": 26524736,
      "step": 125685
    },
    {
      "epoch": 13.827282728272827,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007895194854909415,
      "loss": 0.2319,
      "num_input_tokens_seen": 26525760,
      "step": 125690
    },
    {
      "epoch": 13.827832783278328,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.007893926627753466,
      "loss": 0.2319,
      "num_input_tokens_seen": 26526784,
      "step": 125695
    },
    {
      "epoch": 13.828382838283828,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007892658466090786,
      "loss": 0.2324,
      "num_input_tokens_seen": 26527872,
      "step": 125700
    },
    {
      "epoch": 13.828932893289329,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.007891390369933055,
      "loss": 0.2309,
      "num_input_tokens_seen": 26528928,
      "step": 125705
    },
    {
      "epoch": 13.82948294829483,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.007890122339291969,
      "loss": 0.2314,
      "num_input_tokens_seen": 26530016,
      "step": 125710
    },
    {
      "epoch": 13.83003300330033,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.007888854374179213,
      "loss": 0.2313,
      "num_input_tokens_seen": 26531072,
      "step": 125715
    },
    {
      "epoch": 13.83058305830583,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.00788758647460648,
      "loss": 0.2314,
      "num_input_tokens_seen": 26532128,
      "step": 125720
    },
    {
      "epoch": 13.831133113311331,
      "grad_norm": 0.0027923583984375,
      "learning_rate": 0.00788631864058545,
      "loss": 0.2303,
      "num_input_tokens_seen": 26533120,
      "step": 125725
    },
    {
      "epoch": 13.831683168316832,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.007885050872127802,
      "loss": 0.2319,
      "num_input_tokens_seen": 26534144,
      "step": 125730
    },
    {
      "epoch": 13.832233223322332,
      "grad_norm": 0.00078582763671875,
      "learning_rate": 0.007883783169245226,
      "loss": 0.2319,
      "num_input_tokens_seen": 26535104,
      "step": 125735
    },
    {
      "epoch": 13.832783278327833,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.007882515531949404,
      "loss": 0.2298,
      "num_input_tokens_seen": 26536224,
      "step": 125740
    },
    {
      "epoch": 13.833333333333334,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.007881247960252028,
      "loss": 0.2298,
      "num_input_tokens_seen": 26537280,
      "step": 125745
    },
    {
      "epoch": 13.833883388338833,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.007879980454164773,
      "loss": 0.2313,
      "num_input_tokens_seen": 26538336,
      "step": 125750
    },
    {
      "epoch": 13.834433443344334,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.007878713013699318,
      "loss": 0.2308,
      "num_input_tokens_seen": 26539456,
      "step": 125755
    },
    {
      "epoch": 13.834983498349835,
      "grad_norm": 0.000423431396484375,
      "learning_rate": 0.007877445638867353,
      "loss": 0.2308,
      "num_input_tokens_seen": 26540544,
      "step": 125760
    },
    {
      "epoch": 13.835533553355335,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.007876178329680548,
      "loss": 0.233,
      "num_input_tokens_seen": 26541568,
      "step": 125765
    },
    {
      "epoch": 13.836083608360836,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.007874911086150588,
      "loss": 0.2324,
      "num_input_tokens_seen": 26542592,
      "step": 125770
    },
    {
      "epoch": 13.836633663366337,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.00787364390828916,
      "loss": 0.2308,
      "num_input_tokens_seen": 26543680,
      "step": 125775
    },
    {
      "epoch": 13.837183718371836,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.00787237679610793,
      "loss": 0.2329,
      "num_input_tokens_seen": 26544800,
      "step": 125780
    },
    {
      "epoch": 13.837733773377337,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.007871109749618592,
      "loss": 0.2335,
      "num_input_tokens_seen": 26545888,
      "step": 125785
    },
    {
      "epoch": 13.838283828382838,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.007869842768832807,
      "loss": 0.2324,
      "num_input_tokens_seen": 26546944,
      "step": 125790
    },
    {
      "epoch": 13.83883388338834,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.007868575853762258,
      "loss": 0.2303,
      "num_input_tokens_seen": 26547968,
      "step": 125795
    },
    {
      "epoch": 13.839383938393839,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.007867309004418628,
      "loss": 0.2308,
      "num_input_tokens_seen": 26549024,
      "step": 125800
    },
    {
      "epoch": 13.83993399339934,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.007866042220813586,
      "loss": 0.2298,
      "num_input_tokens_seen": 26550016,
      "step": 125805
    },
    {
      "epoch": 13.840484048404841,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007864775502958814,
      "loss": 0.2324,
      "num_input_tokens_seen": 26551072,
      "step": 125810
    },
    {
      "epoch": 13.84103410341034,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.007863508850865977,
      "loss": 0.2319,
      "num_input_tokens_seen": 26552064,
      "step": 125815
    },
    {
      "epoch": 13.841584158415841,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00786224226454676,
      "loss": 0.2324,
      "num_input_tokens_seen": 26553088,
      "step": 125820
    },
    {
      "epoch": 13.842134213421343,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.007860975744012826,
      "loss": 0.2314,
      "num_input_tokens_seen": 26554144,
      "step": 125825
    },
    {
      "epoch": 13.842684268426842,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.007859709289275853,
      "loss": 0.2308,
      "num_input_tokens_seen": 26555232,
      "step": 125830
    },
    {
      "epoch": 13.843234323432343,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007858442900347517,
      "loss": 0.2324,
      "num_input_tokens_seen": 26556320,
      "step": 125835
    },
    {
      "epoch": 13.843784378437844,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.007857176577239482,
      "loss": 0.2329,
      "num_input_tokens_seen": 26557376,
      "step": 125840
    },
    {
      "epoch": 13.844334433443345,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007855910319963426,
      "loss": 0.2303,
      "num_input_tokens_seen": 26558496,
      "step": 125845
    },
    {
      "epoch": 13.844884488448844,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.007854644128531013,
      "loss": 0.2298,
      "num_input_tokens_seen": 26559584,
      "step": 125850
    },
    {
      "epoch": 13.845434543454346,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00785337800295392,
      "loss": 0.2303,
      "num_input_tokens_seen": 26560640,
      "step": 125855
    },
    {
      "epoch": 13.845984598459847,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.007852111943243811,
      "loss": 0.2314,
      "num_input_tokens_seen": 26561760,
      "step": 125860
    },
    {
      "epoch": 13.846534653465346,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.007850845949412356,
      "loss": 0.2319,
      "num_input_tokens_seen": 26562784,
      "step": 125865
    },
    {
      "epoch": 13.847084708470847,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.007849580021471228,
      "loss": 0.2314,
      "num_input_tokens_seen": 26563808,
      "step": 125870
    },
    {
      "epoch": 13.847634763476348,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00784831415943209,
      "loss": 0.2303,
      "num_input_tokens_seen": 26564928,
      "step": 125875
    },
    {
      "epoch": 13.848184818481847,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.0078470483633066,
      "loss": 0.2324,
      "num_input_tokens_seen": 26565952,
      "step": 125880
    },
    {
      "epoch": 13.848734873487349,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.007845782633106438,
      "loss": 0.2324,
      "num_input_tokens_seen": 26566944,
      "step": 125885
    },
    {
      "epoch": 13.84928492849285,
      "grad_norm": 0.005859375,
      "learning_rate": 0.007844516968843265,
      "loss": 0.2309,
      "num_input_tokens_seen": 26567968,
      "step": 125890
    },
    {
      "epoch": 13.84983498349835,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007843251370528744,
      "loss": 0.2314,
      "num_input_tokens_seen": 26568992,
      "step": 125895
    },
    {
      "epoch": 13.85038503850385,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.007841985838174545,
      "loss": 0.2303,
      "num_input_tokens_seen": 26570048,
      "step": 125900
    },
    {
      "epoch": 13.850935093509351,
      "grad_norm": 0.005859375,
      "learning_rate": 0.007840720371792322,
      "loss": 0.2314,
      "num_input_tokens_seen": 26571136,
      "step": 125905
    },
    {
      "epoch": 13.851485148514852,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.007839454971393744,
      "loss": 0.2298,
      "num_input_tokens_seen": 26572192,
      "step": 125910
    },
    {
      "epoch": 13.852035203520352,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00783818963699048,
      "loss": 0.2313,
      "num_input_tokens_seen": 26573184,
      "step": 125915
    },
    {
      "epoch": 13.852585258525853,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007836924368594178,
      "loss": 0.2308,
      "num_input_tokens_seen": 26574304,
      "step": 125920
    },
    {
      "epoch": 13.853135313531354,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.007835659166216515,
      "loss": 0.2329,
      "num_input_tokens_seen": 26575392,
      "step": 125925
    },
    {
      "epoch": 13.853685368536853,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.007834394029869136,
      "loss": 0.2308,
      "num_input_tokens_seen": 26576448,
      "step": 125930
    },
    {
      "epoch": 13.854235423542354,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.007833128959563715,
      "loss": 0.2293,
      "num_input_tokens_seen": 26577440,
      "step": 125935
    },
    {
      "epoch": 13.854785478547855,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0078318639553119,
      "loss": 0.2298,
      "num_input_tokens_seen": 26578560,
      "step": 125940
    },
    {
      "epoch": 13.855335533553355,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.007830599017125356,
      "loss": 0.2313,
      "num_input_tokens_seen": 26579552,
      "step": 125945
    },
    {
      "epoch": 13.855885588558856,
      "grad_norm": 0.005859375,
      "learning_rate": 0.007829334145015744,
      "loss": 0.2298,
      "num_input_tokens_seen": 26580608,
      "step": 125950
    },
    {
      "epoch": 13.856435643564357,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.007828069338994716,
      "loss": 0.2308,
      "num_input_tokens_seen": 26581632,
      "step": 125955
    },
    {
      "epoch": 13.856985698569858,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.007826804599073934,
      "loss": 0.2314,
      "num_input_tokens_seen": 26582720,
      "step": 125960
    },
    {
      "epoch": 13.857535753575357,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007825539925265048,
      "loss": 0.2309,
      "num_input_tokens_seen": 26583744,
      "step": 125965
    },
    {
      "epoch": 13.858085808580858,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007824275317579718,
      "loss": 0.2314,
      "num_input_tokens_seen": 26584832,
      "step": 125970
    },
    {
      "epoch": 13.85863586358636,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.007823010776029605,
      "loss": 0.2314,
      "num_input_tokens_seen": 26585856,
      "step": 125975
    },
    {
      "epoch": 13.859185918591859,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.007821746300626351,
      "loss": 0.2309,
      "num_input_tokens_seen": 26586880,
      "step": 125980
    },
    {
      "epoch": 13.85973597359736,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007820481891381624,
      "loss": 0.2314,
      "num_input_tokens_seen": 26587904,
      "step": 125985
    },
    {
      "epoch": 13.86028602860286,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.00781921754830707,
      "loss": 0.2319,
      "num_input_tokens_seen": 26588928,
      "step": 125990
    },
    {
      "epoch": 13.86083608360836,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.007817953271414336,
      "loss": 0.2319,
      "num_input_tokens_seen": 26589984,
      "step": 125995
    },
    {
      "epoch": 13.861386138613861,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00781668906071508,
      "loss": 0.2298,
      "num_input_tokens_seen": 26591040,
      "step": 126000
    },
    {
      "epoch": 13.861936193619362,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.007815424916220955,
      "loss": 0.2324,
      "num_input_tokens_seen": 26592128,
      "step": 126005
    },
    {
      "epoch": 13.862486248624862,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.007814160837943616,
      "loss": 0.2319,
      "num_input_tokens_seen": 26593184,
      "step": 126010
    },
    {
      "epoch": 13.863036303630363,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.007812896825894708,
      "loss": 0.2314,
      "num_input_tokens_seen": 26594208,
      "step": 126015
    },
    {
      "epoch": 13.863586358635864,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.007811632880085876,
      "loss": 0.2335,
      "num_input_tokens_seen": 26595232,
      "step": 126020
    },
    {
      "epoch": 13.864136413641365,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.007810369000528775,
      "loss": 0.2298,
      "num_input_tokens_seen": 26596288,
      "step": 126025
    },
    {
      "epoch": 13.864686468646864,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.0078091051872350515,
      "loss": 0.2298,
      "num_input_tokens_seen": 26597376,
      "step": 126030
    },
    {
      "epoch": 13.865236523652365,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.007807841440216363,
      "loss": 0.234,
      "num_input_tokens_seen": 26598432,
      "step": 126035
    },
    {
      "epoch": 13.865786578657866,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.007806577759484347,
      "loss": 0.2298,
      "num_input_tokens_seen": 26599488,
      "step": 126040
    },
    {
      "epoch": 13.866336633663366,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.0078053141450506475,
      "loss": 0.2308,
      "num_input_tokens_seen": 26600480,
      "step": 126045
    },
    {
      "epoch": 13.866886688668867,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.007804050596926921,
      "loss": 0.2335,
      "num_input_tokens_seen": 26601536,
      "step": 126050
    },
    {
      "epoch": 13.867436743674368,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0078027871151248,
      "loss": 0.2303,
      "num_input_tokens_seen": 26602592,
      "step": 126055
    },
    {
      "epoch": 13.867986798679867,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.007801523699655938,
      "loss": 0.2324,
      "num_input_tokens_seen": 26603616,
      "step": 126060
    },
    {
      "epoch": 13.868536853685368,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0078002603505319835,
      "loss": 0.2329,
      "num_input_tokens_seen": 26604672,
      "step": 126065
    },
    {
      "epoch": 13.86908690869087,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.007798997067764568,
      "loss": 0.2308,
      "num_input_tokens_seen": 26605696,
      "step": 126070
    },
    {
      "epoch": 13.869636963696369,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007797733851365346,
      "loss": 0.2303,
      "num_input_tokens_seen": 26606816,
      "step": 126075
    },
    {
      "epoch": 13.87018701870187,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.007796470701345958,
      "loss": 0.2314,
      "num_input_tokens_seen": 26607840,
      "step": 126080
    },
    {
      "epoch": 13.870737073707371,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007795207617718028,
      "loss": 0.2308,
      "num_input_tokens_seen": 26608896,
      "step": 126085
    },
    {
      "epoch": 13.871287128712872,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.007793944600493227,
      "loss": 0.2324,
      "num_input_tokens_seen": 26609920,
      "step": 126090
    },
    {
      "epoch": 13.871837183718371,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007792681649683173,
      "loss": 0.2319,
      "num_input_tokens_seen": 26610944,
      "step": 126095
    },
    {
      "epoch": 13.872387238723872,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.007791418765299521,
      "loss": 0.2325,
      "num_input_tokens_seen": 26612000,
      "step": 126100
    },
    {
      "epoch": 13.872937293729374,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.007790155947353901,
      "loss": 0.2314,
      "num_input_tokens_seen": 26613056,
      "step": 126105
    },
    {
      "epoch": 13.873487348734873,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00778889319585795,
      "loss": 0.234,
      "num_input_tokens_seen": 26614048,
      "step": 126110
    },
    {
      "epoch": 13.874037403740374,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.00778763051082331,
      "loss": 0.2303,
      "num_input_tokens_seen": 26615136,
      "step": 126115
    },
    {
      "epoch": 13.874587458745875,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007786367892261619,
      "loss": 0.233,
      "num_input_tokens_seen": 26616224,
      "step": 126120
    },
    {
      "epoch": 13.875137513751374,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00778510534018452,
      "loss": 0.2308,
      "num_input_tokens_seen": 26617280,
      "step": 126125
    },
    {
      "epoch": 13.875687568756875,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.007783842854603641,
      "loss": 0.2308,
      "num_input_tokens_seen": 26618368,
      "step": 126130
    },
    {
      "epoch": 13.876237623762377,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.007782580435530615,
      "loss": 0.2303,
      "num_input_tokens_seen": 26619424,
      "step": 126135
    },
    {
      "epoch": 13.876787678767876,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0077813180829770885,
      "loss": 0.2309,
      "num_input_tokens_seen": 26620480,
      "step": 126140
    },
    {
      "epoch": 13.877337733773377,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.007780055796954677,
      "loss": 0.2308,
      "num_input_tokens_seen": 26621504,
      "step": 126145
    },
    {
      "epoch": 13.877887788778878,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007778793577475039,
      "loss": 0.2335,
      "num_input_tokens_seen": 26622560,
      "step": 126150
    },
    {
      "epoch": 13.87843784378438,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.0077775314245497945,
      "loss": 0.2319,
      "num_input_tokens_seen": 26623616,
      "step": 126155
    },
    {
      "epoch": 13.878987898789878,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007776269338190572,
      "loss": 0.2319,
      "num_input_tokens_seen": 26624672,
      "step": 126160
    },
    {
      "epoch": 13.87953795379538,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.007775007318409015,
      "loss": 0.2303,
      "num_input_tokens_seen": 26625728,
      "step": 126165
    },
    {
      "epoch": 13.88008800880088,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.007773745365216741,
      "loss": 0.2314,
      "num_input_tokens_seen": 26626784,
      "step": 126170
    },
    {
      "epoch": 13.88063806380638,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0077724834786253895,
      "loss": 0.2288,
      "num_input_tokens_seen": 26627840,
      "step": 126175
    },
    {
      "epoch": 13.881188118811881,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.007771221658646596,
      "loss": 0.2329,
      "num_input_tokens_seen": 26628896,
      "step": 126180
    },
    {
      "epoch": 13.881738173817382,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.007769959905291977,
      "loss": 0.2303,
      "num_input_tokens_seen": 26629952,
      "step": 126185
    },
    {
      "epoch": 13.882288228822881,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007768698218573175,
      "loss": 0.2303,
      "num_input_tokens_seen": 26631008,
      "step": 126190
    },
    {
      "epoch": 13.882838283828383,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.007767436598501805,
      "loss": 0.2314,
      "num_input_tokens_seen": 26632128,
      "step": 126195
    },
    {
      "epoch": 13.883388338833884,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007766175045089507,
      "loss": 0.2314,
      "num_input_tokens_seen": 26633216,
      "step": 126200
    },
    {
      "epoch": 13.883938393839383,
      "grad_norm": 0.002838134765625,
      "learning_rate": 0.0077649135583478975,
      "loss": 0.2324,
      "num_input_tokens_seen": 26634336,
      "step": 126205
    },
    {
      "epoch": 13.884488448844884,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.007763652138288608,
      "loss": 0.2319,
      "num_input_tokens_seen": 26635360,
      "step": 126210
    },
    {
      "epoch": 13.885038503850385,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.007762390784923271,
      "loss": 0.2329,
      "num_input_tokens_seen": 26636416,
      "step": 126215
    },
    {
      "epoch": 13.885588558855886,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0077611294982634985,
      "loss": 0.2303,
      "num_input_tokens_seen": 26637440,
      "step": 126220
    },
    {
      "epoch": 13.886138613861386,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.007759868278320927,
      "loss": 0.2303,
      "num_input_tokens_seen": 26638432,
      "step": 126225
    },
    {
      "epoch": 13.886688668866887,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.007758607125107169,
      "loss": 0.2314,
      "num_input_tokens_seen": 26639488,
      "step": 126230
    },
    {
      "epoch": 13.887238723872388,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0077573460386338566,
      "loss": 0.2324,
      "num_input_tokens_seen": 26640512,
      "step": 126235
    },
    {
      "epoch": 13.887788778877887,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007756085018912615,
      "loss": 0.234,
      "num_input_tokens_seen": 26641600,
      "step": 126240
    },
    {
      "epoch": 13.888338833883388,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.007754824065955056,
      "loss": 0.2324,
      "num_input_tokens_seen": 26642656,
      "step": 126245
    },
    {
      "epoch": 13.88888888888889,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007753563179772814,
      "loss": 0.2314,
      "num_input_tokens_seen": 26643680,
      "step": 126250
    },
    {
      "epoch": 13.88943894389439,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0077523023603775,
      "loss": 0.2314,
      "num_input_tokens_seen": 26644672,
      "step": 126255
    },
    {
      "epoch": 13.88998899889989,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00775104160778073,
      "loss": 0.2309,
      "num_input_tokens_seen": 26645792,
      "step": 126260
    },
    {
      "epoch": 13.89053905390539,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.007749780921994139,
      "loss": 0.2329,
      "num_input_tokens_seen": 26646816,
      "step": 126265
    },
    {
      "epoch": 13.891089108910892,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.00774852030302934,
      "loss": 0.2314,
      "num_input_tokens_seen": 26647904,
      "step": 126270
    },
    {
      "epoch": 13.891639163916391,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.007747259750897943,
      "loss": 0.2298,
      "num_input_tokens_seen": 26648960,
      "step": 126275
    },
    {
      "epoch": 13.892189218921892,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.0077459992656115795,
      "loss": 0.2303,
      "num_input_tokens_seen": 26650016,
      "step": 126280
    },
    {
      "epoch": 13.892739273927393,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.007744738847181855,
      "loss": 0.2324,
      "num_input_tokens_seen": 26651040,
      "step": 126285
    },
    {
      "epoch": 13.893289328932893,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.007743478495620389,
      "loss": 0.2298,
      "num_input_tokens_seen": 26652160,
      "step": 126290
    },
    {
      "epoch": 13.893839383938394,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007742218210938808,
      "loss": 0.2314,
      "num_input_tokens_seen": 26653184,
      "step": 126295
    },
    {
      "epoch": 13.894389438943895,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0077409579931487105,
      "loss": 0.2303,
      "num_input_tokens_seen": 26654272,
      "step": 126300
    },
    {
      "epoch": 13.894939493949394,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.007739697842261727,
      "loss": 0.2293,
      "num_input_tokens_seen": 26655360,
      "step": 126305
    },
    {
      "epoch": 13.895489548954895,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007738437758289458,
      "loss": 0.2329,
      "num_input_tokens_seen": 26656448,
      "step": 126310
    },
    {
      "epoch": 13.896039603960396,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.0077371777412435306,
      "loss": 0.233,
      "num_input_tokens_seen": 26657536,
      "step": 126315
    },
    {
      "epoch": 13.896589658965897,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.007735917791135544,
      "loss": 0.2309,
      "num_input_tokens_seen": 26658560,
      "step": 126320
    },
    {
      "epoch": 13.897139713971397,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0077346579079771195,
      "loss": 0.2319,
      "num_input_tokens_seen": 26659648,
      "step": 126325
    },
    {
      "epoch": 13.897689768976898,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.007733398091779872,
      "loss": 0.2303,
      "num_input_tokens_seen": 26660768,
      "step": 126330
    },
    {
      "epoch": 13.898239823982399,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007732138342555402,
      "loss": 0.2303,
      "num_input_tokens_seen": 26661824,
      "step": 126335
    },
    {
      "epoch": 13.898789878987898,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007730878660315329,
      "loss": 0.2324,
      "num_input_tokens_seen": 26662880,
      "step": 126340
    },
    {
      "epoch": 13.8993399339934,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007729619045071255,
      "loss": 0.2324,
      "num_input_tokens_seen": 26663968,
      "step": 126345
    },
    {
      "epoch": 13.8998899889989,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0077283594968347935,
      "loss": 0.2329,
      "num_input_tokens_seen": 26665056,
      "step": 126350
    },
    {
      "epoch": 13.9004400440044,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.007727100015617559,
      "loss": 0.2329,
      "num_input_tokens_seen": 26666144,
      "step": 126355
    },
    {
      "epoch": 13.900990099009901,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007725840601431147,
      "loss": 0.2314,
      "num_input_tokens_seen": 26667200,
      "step": 126360
    },
    {
      "epoch": 13.901540154015402,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00772458125428718,
      "loss": 0.2314,
      "num_input_tokens_seen": 26668224,
      "step": 126365
    },
    {
      "epoch": 13.902090209020901,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007723321974197256,
      "loss": 0.2319,
      "num_input_tokens_seen": 26669280,
      "step": 126370
    },
    {
      "epoch": 13.902640264026402,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.007722062761172974,
      "loss": 0.2314,
      "num_input_tokens_seen": 26670368,
      "step": 126375
    },
    {
      "epoch": 13.903190319031903,
      "grad_norm": 0.00078582763671875,
      "learning_rate": 0.0077208036152259475,
      "loss": 0.2309,
      "num_input_tokens_seen": 26671456,
      "step": 126380
    },
    {
      "epoch": 13.903740374037405,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.007719544536367783,
      "loss": 0.2319,
      "num_input_tokens_seen": 26672576,
      "step": 126385
    },
    {
      "epoch": 13.904290429042904,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007718285524610087,
      "loss": 0.2293,
      "num_input_tokens_seen": 26673696,
      "step": 126390
    },
    {
      "epoch": 13.904840484048405,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.0077170265799644585,
      "loss": 0.2314,
      "num_input_tokens_seen": 26674784,
      "step": 126395
    },
    {
      "epoch": 13.905390539053906,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.0077157677024424975,
      "loss": 0.2319,
      "num_input_tokens_seen": 26675840,
      "step": 126400
    },
    {
      "epoch": 13.905940594059405,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.007714508892055811,
      "loss": 0.2319,
      "num_input_tokens_seen": 26676928,
      "step": 126405
    },
    {
      "epoch": 13.906490649064907,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.007713250148815997,
      "loss": 0.2313,
      "num_input_tokens_seen": 26677984,
      "step": 126410
    },
    {
      "epoch": 13.907040704070408,
      "grad_norm": 0.005859375,
      "learning_rate": 0.007711991472734669,
      "loss": 0.2293,
      "num_input_tokens_seen": 26679008,
      "step": 126415
    },
    {
      "epoch": 13.907590759075907,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007710732863823417,
      "loss": 0.2308,
      "num_input_tokens_seen": 26680032,
      "step": 126420
    },
    {
      "epoch": 13.908140814081408,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.007709474322093835,
      "loss": 0.2319,
      "num_input_tokens_seen": 26681152,
      "step": 126425
    },
    {
      "epoch": 13.908690869086909,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007708215847557538,
      "loss": 0.2314,
      "num_input_tokens_seen": 26682272,
      "step": 126430
    },
    {
      "epoch": 13.909240924092408,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0077069574402261086,
      "loss": 0.233,
      "num_input_tokens_seen": 26683360,
      "step": 126435
    },
    {
      "epoch": 13.90979097909791,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.007705699100111155,
      "loss": 0.2303,
      "num_input_tokens_seen": 26684384,
      "step": 126440
    },
    {
      "epoch": 13.91034103410341,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.007704440827224278,
      "loss": 0.2319,
      "num_input_tokens_seen": 26685408,
      "step": 126445
    },
    {
      "epoch": 13.910891089108912,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.007703182621577063,
      "loss": 0.2303,
      "num_input_tokens_seen": 26686432,
      "step": 126450
    },
    {
      "epoch": 13.911441144114411,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007701924483181119,
      "loss": 0.2314,
      "num_input_tokens_seen": 26687456,
      "step": 126455
    },
    {
      "epoch": 13.911991199119912,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.007700666412048028,
      "loss": 0.2329,
      "num_input_tokens_seen": 26688480,
      "step": 126460
    },
    {
      "epoch": 13.912541254125413,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.007699408408189392,
      "loss": 0.2329,
      "num_input_tokens_seen": 26689568,
      "step": 126465
    },
    {
      "epoch": 13.913091309130913,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007698150471616812,
      "loss": 0.2313,
      "num_input_tokens_seen": 26690592,
      "step": 126470
    },
    {
      "epoch": 13.913641364136414,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 0.007696892602341869,
      "loss": 0.232,
      "num_input_tokens_seen": 26691616,
      "step": 126475
    },
    {
      "epoch": 13.914191419141915,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00769563480037617,
      "loss": 0.2309,
      "num_input_tokens_seen": 26692704,
      "step": 126480
    },
    {
      "epoch": 13.914741474147414,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0076943770657312985,
      "loss": 0.2314,
      "num_input_tokens_seen": 26693792,
      "step": 126485
    },
    {
      "epoch": 13.915291529152915,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.007693119398418844,
      "loss": 0.2314,
      "num_input_tokens_seen": 26694816,
      "step": 126490
    },
    {
      "epoch": 13.915841584158416,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.007691861798450399,
      "loss": 0.2298,
      "num_input_tokens_seen": 26695904,
      "step": 126495
    },
    {
      "epoch": 13.916391639163916,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00769060426583756,
      "loss": 0.2319,
      "num_input_tokens_seen": 26696928,
      "step": 126500
    },
    {
      "epoch": 13.916941694169417,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007689346800591916,
      "loss": 0.233,
      "num_input_tokens_seen": 26697952,
      "step": 126505
    },
    {
      "epoch": 13.917491749174918,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0076880894027250585,
      "loss": 0.2303,
      "num_input_tokens_seen": 26698976,
      "step": 126510
    },
    {
      "epoch": 13.918041804180419,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.007686832072248565,
      "loss": 0.2324,
      "num_input_tokens_seen": 26700000,
      "step": 126515
    },
    {
      "epoch": 13.918591859185918,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.007685574809174031,
      "loss": 0.2309,
      "num_input_tokens_seen": 26701120,
      "step": 126520
    },
    {
      "epoch": 13.91914191419142,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.007684317613513043,
      "loss": 0.2309,
      "num_input_tokens_seen": 26702176,
      "step": 126525
    },
    {
      "epoch": 13.91969196919692,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.007683060485277196,
      "loss": 0.2319,
      "num_input_tokens_seen": 26703200,
      "step": 126530
    },
    {
      "epoch": 13.92024202420242,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007681803424478071,
      "loss": 0.2324,
      "num_input_tokens_seen": 26704288,
      "step": 126535
    },
    {
      "epoch": 13.92079207920792,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007680546431127246,
      "loss": 0.2303,
      "num_input_tokens_seen": 26705344,
      "step": 126540
    },
    {
      "epoch": 13.921342134213422,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.007679289505236317,
      "loss": 0.2303,
      "num_input_tokens_seen": 26706432,
      "step": 126545
    },
    {
      "epoch": 13.921892189218921,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.007678032646816857,
      "loss": 0.2319,
      "num_input_tokens_seen": 26707488,
      "step": 126550
    },
    {
      "epoch": 13.922442244224422,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007676775855880459,
      "loss": 0.2303,
      "num_input_tokens_seen": 26708544,
      "step": 126555
    },
    {
      "epoch": 13.922992299229923,
      "grad_norm": 0.005859375,
      "learning_rate": 0.007675519132438707,
      "loss": 0.2314,
      "num_input_tokens_seen": 26709600,
      "step": 126560
    },
    {
      "epoch": 13.923542354235423,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0076742624765031775,
      "loss": 0.2329,
      "num_input_tokens_seen": 26710656,
      "step": 126565
    },
    {
      "epoch": 13.924092409240924,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0076730058880854594,
      "loss": 0.2324,
      "num_input_tokens_seen": 26711680,
      "step": 126570
    },
    {
      "epoch": 13.924642464246425,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007671749367197125,
      "loss": 0.2288,
      "num_input_tokens_seen": 26712768,
      "step": 126575
    },
    {
      "epoch": 13.925192519251926,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007670492913849761,
      "loss": 0.233,
      "num_input_tokens_seen": 26713760,
      "step": 126580
    },
    {
      "epoch": 13.925742574257425,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.007669236528054953,
      "loss": 0.2293,
      "num_input_tokens_seen": 26714816,
      "step": 126585
    },
    {
      "epoch": 13.926292629262926,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007667980209824266,
      "loss": 0.2324,
      "num_input_tokens_seen": 26715840,
      "step": 126590
    },
    {
      "epoch": 13.926842684268427,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.007666723959169295,
      "loss": 0.2309,
      "num_input_tokens_seen": 26716896,
      "step": 126595
    },
    {
      "epoch": 13.927392739273927,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.007665467776101605,
      "loss": 0.2308,
      "num_input_tokens_seen": 26717984,
      "step": 126600
    },
    {
      "epoch": 13.927942794279428,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007664211660632784,
      "loss": 0.2303,
      "num_input_tokens_seen": 26719072,
      "step": 126605
    },
    {
      "epoch": 13.928492849284929,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.007662955612774396,
      "loss": 0.2303,
      "num_input_tokens_seen": 26720160,
      "step": 126610
    },
    {
      "epoch": 13.929042904290428,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007661699632538029,
      "loss": 0.2303,
      "num_input_tokens_seen": 26721216,
      "step": 126615
    },
    {
      "epoch": 13.92959295929593,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00766044371993526,
      "loss": 0.2303,
      "num_input_tokens_seen": 26722336,
      "step": 126620
    },
    {
      "epoch": 13.93014301430143,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007659187874977654,
      "loss": 0.2303,
      "num_input_tokens_seen": 26723392,
      "step": 126625
    },
    {
      "epoch": 13.930693069306932,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007657932097676795,
      "loss": 0.2309,
      "num_input_tokens_seen": 26724416,
      "step": 126630
    },
    {
      "epoch": 13.93124312431243,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.007656676388044248,
      "loss": 0.2324,
      "num_input_tokens_seen": 26725440,
      "step": 126635
    },
    {
      "epoch": 13.931793179317932,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.007655420746091592,
      "loss": 0.2303,
      "num_input_tokens_seen": 26726496,
      "step": 126640
    },
    {
      "epoch": 13.932343234323433,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.0076541651718304024,
      "loss": 0.2319,
      "num_input_tokens_seen": 26727520,
      "step": 126645
    },
    {
      "epoch": 13.932893289328932,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.00765290966527225,
      "loss": 0.2319,
      "num_input_tokens_seen": 26728512,
      "step": 126650
    },
    {
      "epoch": 13.933443344334433,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.007651654226428696,
      "loss": 0.2329,
      "num_input_tokens_seen": 26729632,
      "step": 126655
    },
    {
      "epoch": 13.933993399339935,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.007650398855311325,
      "loss": 0.2314,
      "num_input_tokens_seen": 26730688,
      "step": 126660
    },
    {
      "epoch": 13.934543454345434,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.007649143551931695,
      "loss": 0.2304,
      "num_input_tokens_seen": 26731776,
      "step": 126665
    },
    {
      "epoch": 13.935093509350935,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.007647888316301384,
      "loss": 0.2298,
      "num_input_tokens_seen": 26732800,
      "step": 126670
    },
    {
      "epoch": 13.935643564356436,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.007646633148431962,
      "loss": 0.2329,
      "num_input_tokens_seen": 26733824,
      "step": 126675
    },
    {
      "epoch": 13.936193619361937,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0076453780483349875,
      "loss": 0.2324,
      "num_input_tokens_seen": 26734880,
      "step": 126680
    },
    {
      "epoch": 13.936743674367436,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007644123016022043,
      "loss": 0.2303,
      "num_input_tokens_seen": 26735936,
      "step": 126685
    },
    {
      "epoch": 13.937293729372938,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007642868051504677,
      "loss": 0.2319,
      "num_input_tokens_seen": 26736960,
      "step": 126690
    },
    {
      "epoch": 13.937843784378439,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007641613154794468,
      "loss": 0.2309,
      "num_input_tokens_seen": 26737920,
      "step": 126695
    },
    {
      "epoch": 13.938393839383938,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.007640358325902987,
      "loss": 0.2309,
      "num_input_tokens_seen": 26738976,
      "step": 126700
    },
    {
      "epoch": 13.938943894389439,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.007639103564841785,
      "loss": 0.2324,
      "num_input_tokens_seen": 26740000,
      "step": 126705
    },
    {
      "epoch": 13.93949394939494,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.007637848871622439,
      "loss": 0.2319,
      "num_input_tokens_seen": 26741088,
      "step": 126710
    },
    {
      "epoch": 13.94004400440044,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.0076365942462565005,
      "loss": 0.2324,
      "num_input_tokens_seen": 26742112,
      "step": 126715
    },
    {
      "epoch": 13.94059405940594,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007635339688755545,
      "loss": 0.2309,
      "num_input_tokens_seen": 26743168,
      "step": 126720
    },
    {
      "epoch": 13.941144114411442,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.007634085199131127,
      "loss": 0.2298,
      "num_input_tokens_seen": 26744160,
      "step": 126725
    },
    {
      "epoch": 13.941694169416941,
      "grad_norm": 0.000621795654296875,
      "learning_rate": 0.00763283077739481,
      "loss": 0.2314,
      "num_input_tokens_seen": 26745216,
      "step": 126730
    },
    {
      "epoch": 13.942244224422442,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00763157642355816,
      "loss": 0.2345,
      "num_input_tokens_seen": 26746304,
      "step": 126735
    },
    {
      "epoch": 13.942794279427943,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.007630322137632731,
      "loss": 0.2319,
      "num_input_tokens_seen": 26747360,
      "step": 126740
    },
    {
      "epoch": 13.943344334433444,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007629067919630092,
      "loss": 0.2298,
      "num_input_tokens_seen": 26748416,
      "step": 126745
    },
    {
      "epoch": 13.943894389438944,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00762781376956179,
      "loss": 0.2308,
      "num_input_tokens_seen": 26749440,
      "step": 126750
    },
    {
      "epoch": 13.944444444444445,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.007626559687439392,
      "loss": 0.2303,
      "num_input_tokens_seen": 26750496,
      "step": 126755
    },
    {
      "epoch": 13.944994499449946,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.007625305673274461,
      "loss": 0.2309,
      "num_input_tokens_seen": 26751520,
      "step": 126760
    },
    {
      "epoch": 13.945544554455445,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.007624051727078543,
      "loss": 0.2319,
      "num_input_tokens_seen": 26752544,
      "step": 126765
    },
    {
      "epoch": 13.946094609460946,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007622797848863206,
      "loss": 0.2329,
      "num_input_tokens_seen": 26753568,
      "step": 126770
    },
    {
      "epoch": 13.946644664466447,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.007621544038640003,
      "loss": 0.2314,
      "num_input_tokens_seen": 26754560,
      "step": 126775
    },
    {
      "epoch": 13.947194719471947,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.00762029029642048,
      "loss": 0.2319,
      "num_input_tokens_seen": 26755712,
      "step": 126780
    },
    {
      "epoch": 13.947744774477448,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.007619036622216203,
      "loss": 0.233,
      "num_input_tokens_seen": 26756768,
      "step": 126785
    },
    {
      "epoch": 13.948294829482949,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.007617783016038721,
      "loss": 0.2319,
      "num_input_tokens_seen": 26757824,
      "step": 126790
    },
    {
      "epoch": 13.948844884488448,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007616529477899598,
      "loss": 0.2319,
      "num_input_tokens_seen": 26758912,
      "step": 126795
    },
    {
      "epoch": 13.94939493949395,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.007615276007810379,
      "loss": 0.233,
      "num_input_tokens_seen": 26759936,
      "step": 126800
    },
    {
      "epoch": 13.94994499449945,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.007614022605782613,
      "loss": 0.2293,
      "num_input_tokens_seen": 26760960,
      "step": 126805
    },
    {
      "epoch": 13.950495049504951,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.0076127692718278555,
      "loss": 0.2308,
      "num_input_tokens_seen": 26762048,
      "step": 126810
    },
    {
      "epoch": 13.95104510451045,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.0076115160059576584,
      "loss": 0.2308,
      "num_input_tokens_seen": 26763104,
      "step": 126815
    },
    {
      "epoch": 13.951595159515952,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0076102628081835795,
      "loss": 0.2329,
      "num_input_tokens_seen": 26764128,
      "step": 126820
    },
    {
      "epoch": 13.952145214521453,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0076090096785171625,
      "loss": 0.2319,
      "num_input_tokens_seen": 26765184,
      "step": 126825
    },
    {
      "epoch": 13.952695269526952,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00760775661696995,
      "loss": 0.2309,
      "num_input_tokens_seen": 26766208,
      "step": 126830
    },
    {
      "epoch": 13.953245324532453,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007606503623553505,
      "loss": 0.2324,
      "num_input_tokens_seen": 26767328,
      "step": 126835
    },
    {
      "epoch": 13.953795379537954,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.007605250698279361,
      "loss": 0.2314,
      "num_input_tokens_seen": 26768416,
      "step": 126840
    },
    {
      "epoch": 13.954345434543454,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.007603997841159074,
      "loss": 0.2308,
      "num_input_tokens_seen": 26769472,
      "step": 126845
    },
    {
      "epoch": 13.954895489548955,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.007602745052204193,
      "loss": 0.2314,
      "num_input_tokens_seen": 26770496,
      "step": 126850
    },
    {
      "epoch": 13.955445544554456,
      "grad_norm": 0.0029449462890625,
      "learning_rate": 0.007601492331426259,
      "loss": 0.2303,
      "num_input_tokens_seen": 26771584,
      "step": 126855
    },
    {
      "epoch": 13.955995599559955,
      "grad_norm": 0.000732421875,
      "learning_rate": 0.007600239678836826,
      "loss": 0.2329,
      "num_input_tokens_seen": 26772576,
      "step": 126860
    },
    {
      "epoch": 13.956545654565456,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.007598987094447432,
      "loss": 0.2314,
      "num_input_tokens_seen": 26773568,
      "step": 126865
    },
    {
      "epoch": 13.957095709570957,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007597734578269611,
      "loss": 0.2319,
      "num_input_tokens_seen": 26774560,
      "step": 126870
    },
    {
      "epoch": 13.957645764576458,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.007596482130314929,
      "loss": 0.2304,
      "num_input_tokens_seen": 26775616,
      "step": 126875
    },
    {
      "epoch": 13.958195819581958,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.0075952297505949146,
      "loss": 0.2324,
      "num_input_tokens_seen": 26776640,
      "step": 126880
    },
    {
      "epoch": 13.958745874587459,
      "grad_norm": 0.000736236572265625,
      "learning_rate": 0.007593977439121118,
      "loss": 0.2319,
      "num_input_tokens_seen": 26777664,
      "step": 126885
    },
    {
      "epoch": 13.95929592959296,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00759272519590508,
      "loss": 0.2314,
      "num_input_tokens_seen": 26778656,
      "step": 126890
    },
    {
      "epoch": 13.95984598459846,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007591473020958332,
      "loss": 0.2319,
      "num_input_tokens_seen": 26779744,
      "step": 126895
    },
    {
      "epoch": 13.96039603960396,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.007590220914292422,
      "loss": 0.2314,
      "num_input_tokens_seen": 26780832,
      "step": 126900
    },
    {
      "epoch": 13.960946094609461,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007588968875918891,
      "loss": 0.2314,
      "num_input_tokens_seen": 26781856,
      "step": 126905
    },
    {
      "epoch": 13.96149614961496,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007587716905849285,
      "loss": 0.2298,
      "num_input_tokens_seen": 26782912,
      "step": 126910
    },
    {
      "epoch": 13.962046204620462,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.007586465004095133,
      "loss": 0.2303,
      "num_input_tokens_seen": 26783936,
      "step": 126915
    },
    {
      "epoch": 13.962596259625963,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.0075852131706679714,
      "loss": 0.2314,
      "num_input_tokens_seen": 26785024,
      "step": 126920
    },
    {
      "epoch": 13.963146314631462,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.007583961405579348,
      "loss": 0.2303,
      "num_input_tokens_seen": 26786080,
      "step": 126925
    },
    {
      "epoch": 13.963696369636963,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.007582709708840781,
      "loss": 0.2324,
      "num_input_tokens_seen": 26787136,
      "step": 126930
    },
    {
      "epoch": 13.964246424642464,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.007581458080463832,
      "loss": 0.2308,
      "num_input_tokens_seen": 26788224,
      "step": 126935
    },
    {
      "epoch": 13.964796479647966,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.007580206520460024,
      "loss": 0.2324,
      "num_input_tokens_seen": 26789312,
      "step": 126940
    },
    {
      "epoch": 13.965346534653465,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.007578955028840886,
      "loss": 0.2324,
      "num_input_tokens_seen": 26790400,
      "step": 126945
    },
    {
      "epoch": 13.965896589658966,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.007577703605617965,
      "loss": 0.2324,
      "num_input_tokens_seen": 26791456,
      "step": 126950
    },
    {
      "epoch": 13.966446644664467,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007576452250802782,
      "loss": 0.2335,
      "num_input_tokens_seen": 26792480,
      "step": 126955
    },
    {
      "epoch": 13.966996699669966,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.007575200964406877,
      "loss": 0.2319,
      "num_input_tokens_seen": 26793632,
      "step": 126960
    },
    {
      "epoch": 13.967546754675467,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007573949746441787,
      "loss": 0.2309,
      "num_input_tokens_seen": 26794688,
      "step": 126965
    },
    {
      "epoch": 13.968096809680969,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.007572698596919032,
      "loss": 0.2324,
      "num_input_tokens_seen": 26795776,
      "step": 126970
    },
    {
      "epoch": 13.968646864686468,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.007571447515850157,
      "loss": 0.2314,
      "num_input_tokens_seen": 26796832,
      "step": 126975
    },
    {
      "epoch": 13.969196919691969,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00757019650324668,
      "loss": 0.2314,
      "num_input_tokens_seen": 26797920,
      "step": 126980
    },
    {
      "epoch": 13.96974697469747,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.007568945559120143,
      "loss": 0.2319,
      "num_input_tokens_seen": 26798976,
      "step": 126985
    },
    {
      "epoch": 13.97029702970297,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.007567694683482062,
      "loss": 0.2319,
      "num_input_tokens_seen": 26800064,
      "step": 126990
    },
    {
      "epoch": 13.97084708470847,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007566443876343973,
      "loss": 0.2314,
      "num_input_tokens_seen": 26801216,
      "step": 126995
    },
    {
      "epoch": 13.971397139713972,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.007565193137717412,
      "loss": 0.2324,
      "num_input_tokens_seen": 26802272,
      "step": 127000
    },
    {
      "epoch": 13.971947194719473,
      "grad_norm": 0.000629425048828125,
      "learning_rate": 0.007563942467613889,
      "loss": 0.2298,
      "num_input_tokens_seen": 26803392,
      "step": 127005
    },
    {
      "epoch": 13.972497249724972,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.007562691866044948,
      "loss": 0.2324,
      "num_input_tokens_seen": 26804544,
      "step": 127010
    },
    {
      "epoch": 13.973047304730473,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.007561441333022101,
      "loss": 0.2303,
      "num_input_tokens_seen": 26805600,
      "step": 127015
    },
    {
      "epoch": 13.973597359735974,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00756019086855688,
      "loss": 0.2308,
      "num_input_tokens_seen": 26806592,
      "step": 127020
    },
    {
      "epoch": 13.974147414741473,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007558940472660815,
      "loss": 0.2314,
      "num_input_tokens_seen": 26807648,
      "step": 127025
    },
    {
      "epoch": 13.974697469746975,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.0075576901453454255,
      "loss": 0.2319,
      "num_input_tokens_seen": 26808640,
      "step": 127030
    },
    {
      "epoch": 13.975247524752476,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.007556439886622228,
      "loss": 0.2314,
      "num_input_tokens_seen": 26809664,
      "step": 127035
    },
    {
      "epoch": 13.975797579757975,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0075551896965027595,
      "loss": 0.2329,
      "num_input_tokens_seen": 26810720,
      "step": 127040
    },
    {
      "epoch": 13.976347634763476,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.007553939574998523,
      "loss": 0.2298,
      "num_input_tokens_seen": 26811776,
      "step": 127045
    },
    {
      "epoch": 13.976897689768977,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.007552689522121064,
      "loss": 0.2314,
      "num_input_tokens_seen": 26812800,
      "step": 127050
    },
    {
      "epoch": 13.977447744774478,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.007551439537881892,
      "loss": 0.2319,
      "num_input_tokens_seen": 26813888,
      "step": 127055
    },
    {
      "epoch": 13.977997799779978,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.007550189622292522,
      "loss": 0.2319,
      "num_input_tokens_seen": 26814944,
      "step": 127060
    },
    {
      "epoch": 13.978547854785479,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0075489397753644855,
      "loss": 0.2309,
      "num_input_tokens_seen": 26815968,
      "step": 127065
    },
    {
      "epoch": 13.97909790979098,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00754768999710929,
      "loss": 0.2314,
      "num_input_tokens_seen": 26816960,
      "step": 127070
    },
    {
      "epoch": 13.979647964796479,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.007546440287538459,
      "loss": 0.2314,
      "num_input_tokens_seen": 26818016,
      "step": 127075
    },
    {
      "epoch": 13.98019801980198,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.007545190646663519,
      "loss": 0.2314,
      "num_input_tokens_seen": 26819008,
      "step": 127080
    },
    {
      "epoch": 13.980748074807481,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007543941074495972,
      "loss": 0.2308,
      "num_input_tokens_seen": 26820064,
      "step": 127085
    },
    {
      "epoch": 13.98129812981298,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.007542691571047349,
      "loss": 0.2303,
      "num_input_tokens_seen": 26821184,
      "step": 127090
    },
    {
      "epoch": 13.981848184818482,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0075414421363291535,
      "loss": 0.2309,
      "num_input_tokens_seen": 26822176,
      "step": 127095
    },
    {
      "epoch": 13.982398239823983,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.007540192770352912,
      "loss": 0.2324,
      "num_input_tokens_seen": 26823264,
      "step": 127100
    },
    {
      "epoch": 13.982948294829484,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.007538943473130131,
      "loss": 0.2303,
      "num_input_tokens_seen": 26824256,
      "step": 127105
    },
    {
      "epoch": 13.983498349834983,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.007537694244672324,
      "loss": 0.2314,
      "num_input_tokens_seen": 26825344,
      "step": 127110
    },
    {
      "epoch": 13.984048404840484,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007536445084991015,
      "loss": 0.2319,
      "num_input_tokens_seen": 26826400,
      "step": 127115
    },
    {
      "epoch": 13.984598459845985,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0075351959940977075,
      "loss": 0.2314,
      "num_input_tokens_seen": 26827520,
      "step": 127120
    },
    {
      "epoch": 13.985148514851485,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007533946972003918,
      "loss": 0.2313,
      "num_input_tokens_seen": 26828480,
      "step": 127125
    },
    {
      "epoch": 13.985698569856986,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.007532698018721155,
      "loss": 0.2293,
      "num_input_tokens_seen": 26829568,
      "step": 127130
    },
    {
      "epoch": 13.986248624862487,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.007531449134260929,
      "loss": 0.2314,
      "num_input_tokens_seen": 26830592,
      "step": 127135
    },
    {
      "epoch": 13.986798679867986,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.00753020031863476,
      "loss": 0.2319,
      "num_input_tokens_seen": 26831648,
      "step": 127140
    },
    {
      "epoch": 13.987348734873487,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007528951571854143,
      "loss": 0.2308,
      "num_input_tokens_seen": 26832736,
      "step": 127145
    },
    {
      "epoch": 13.987898789878988,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.0075277028939305995,
      "loss": 0.2298,
      "num_input_tokens_seen": 26833792,
      "step": 127150
    },
    {
      "epoch": 13.988448844884488,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.0075264542848756345,
      "loss": 0.2324,
      "num_input_tokens_seen": 26834880,
      "step": 127155
    },
    {
      "epoch": 13.988998899889989,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0075252057447007465,
      "loss": 0.2314,
      "num_input_tokens_seen": 26835936,
      "step": 127160
    },
    {
      "epoch": 13.98954895489549,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.007523957273417451,
      "loss": 0.2324,
      "num_input_tokens_seen": 26837024,
      "step": 127165
    },
    {
      "epoch": 13.990099009900991,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.0075227088710372566,
      "loss": 0.2335,
      "num_input_tokens_seen": 26838144,
      "step": 127170
    },
    {
      "epoch": 13.99064906490649,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.007521460537571669,
      "loss": 0.2309,
      "num_input_tokens_seen": 26839200,
      "step": 127175
    },
    {
      "epoch": 13.991199119911991,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.007520212273032192,
      "loss": 0.2319,
      "num_input_tokens_seen": 26840288,
      "step": 127180
    },
    {
      "epoch": 13.991749174917492,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.0075189640774303255,
      "loss": 0.2314,
      "num_input_tokens_seen": 26841312,
      "step": 127185
    },
    {
      "epoch": 13.992299229922992,
      "grad_norm": 0.0027313232421875,
      "learning_rate": 0.007517715950777574,
      "loss": 0.2335,
      "num_input_tokens_seen": 26842304,
      "step": 127190
    },
    {
      "epoch": 13.992849284928493,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007516467893085447,
      "loss": 0.2319,
      "num_input_tokens_seen": 26843392,
      "step": 127195
    },
    {
      "epoch": 13.993399339933994,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.007515219904365449,
      "loss": 0.2319,
      "num_input_tokens_seen": 26844448,
      "step": 127200
    },
    {
      "epoch": 13.993949394939493,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.007513971984629078,
      "loss": 0.2293,
      "num_input_tokens_seen": 26845568,
      "step": 127205
    },
    {
      "epoch": 13.994499449944994,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00751272413388783,
      "loss": 0.2324,
      "num_input_tokens_seen": 26846656,
      "step": 127210
    },
    {
      "epoch": 13.995049504950495,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.007511476352153214,
      "loss": 0.2314,
      "num_input_tokens_seen": 26847744,
      "step": 127215
    },
    {
      "epoch": 13.995599559955995,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.007510228639436726,
      "loss": 0.2298,
      "num_input_tokens_seen": 26848832,
      "step": 127220
    },
    {
      "epoch": 13.996149614961496,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.007508980995749862,
      "loss": 0.2293,
      "num_input_tokens_seen": 26849920,
      "step": 127225
    },
    {
      "epoch": 13.996699669966997,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.007507733421104134,
      "loss": 0.2329,
      "num_input_tokens_seen": 26850976,
      "step": 127230
    },
    {
      "epoch": 13.997249724972498,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.007506485915511026,
      "loss": 0.2319,
      "num_input_tokens_seen": 26851968,
      "step": 127235
    },
    {
      "epoch": 13.997799779977997,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007505238478982046,
      "loss": 0.2308,
      "num_input_tokens_seen": 26852960,
      "step": 127240
    },
    {
      "epoch": 13.998349834983498,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.007503991111528682,
      "loss": 0.2303,
      "num_input_tokens_seen": 26853984,
      "step": 127245
    },
    {
      "epoch": 13.998899889989,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007502743813162436,
      "loss": 0.2324,
      "num_input_tokens_seen": 26855104,
      "step": 127250
    },
    {
      "epoch": 13.999449944994499,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.007501496583894806,
      "loss": 0.2309,
      "num_input_tokens_seen": 26856160,
      "step": 127255
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.003448486328125,
      "learning_rate": 0.007500249423737278,
      "loss": 0.2309,
      "num_input_tokens_seen": 26857088,
      "step": 127260
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.23143287003040314,
      "eval_runtime": 60.5896,
      "eval_samples_per_second": 66.678,
      "eval_steps_per_second": 16.67,
      "num_input_tokens_seen": 26857088,
      "step": 127260
    },
    {
      "epoch": 14.000550055005501,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007499002332701361,
      "loss": 0.2304,
      "num_input_tokens_seen": 26858208,
      "step": 127265
    },
    {
      "epoch": 14.001100110011,
      "grad_norm": 0.00081634521484375,
      "learning_rate": 0.007497755310798538,
      "loss": 0.233,
      "num_input_tokens_seen": 26859232,
      "step": 127270
    },
    {
      "epoch": 14.001650165016502,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.007496508358040299,
      "loss": 0.2314,
      "num_input_tokens_seen": 26860288,
      "step": 127275
    },
    {
      "epoch": 14.002200220022003,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007495261474438141,
      "loss": 0.2309,
      "num_input_tokens_seen": 26861312,
      "step": 127280
    },
    {
      "epoch": 14.002750275027502,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.007494014660003558,
      "loss": 0.2298,
      "num_input_tokens_seen": 26862400,
      "step": 127285
    },
    {
      "epoch": 14.003300330033003,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007492767914748047,
      "loss": 0.2298,
      "num_input_tokens_seen": 26863456,
      "step": 127290
    },
    {
      "epoch": 14.003850385038504,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.00749152123868309,
      "loss": 0.2324,
      "num_input_tokens_seen": 26864512,
      "step": 127295
    },
    {
      "epoch": 14.004400440044005,
      "grad_norm": 0.0028076171875,
      "learning_rate": 0.007490274631820174,
      "loss": 0.2308,
      "num_input_tokens_seen": 26865600,
      "step": 127300
    },
    {
      "epoch": 14.004950495049505,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.007489028094170793,
      "loss": 0.2314,
      "num_input_tokens_seen": 26866592,
      "step": 127305
    },
    {
      "epoch": 14.005500550055006,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.007487781625746436,
      "loss": 0.2309,
      "num_input_tokens_seen": 26867680,
      "step": 127310
    },
    {
      "epoch": 14.006050605060507,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.007486535226558596,
      "loss": 0.2335,
      "num_input_tokens_seen": 26868736,
      "step": 127315
    },
    {
      "epoch": 14.006600660066006,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.007485288896618757,
      "loss": 0.2314,
      "num_input_tokens_seen": 26869792,
      "step": 127320
    },
    {
      "epoch": 14.007150715071507,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007484042635938398,
      "loss": 0.2335,
      "num_input_tokens_seen": 26870944,
      "step": 127325
    },
    {
      "epoch": 14.007700770077008,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.007482796444529016,
      "loss": 0.2314,
      "num_input_tokens_seen": 26871968,
      "step": 127330
    },
    {
      "epoch": 14.008250825082508,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.007481550322402087,
      "loss": 0.233,
      "num_input_tokens_seen": 26872960,
      "step": 127335
    },
    {
      "epoch": 14.008800880088009,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007480304269569102,
      "loss": 0.2314,
      "num_input_tokens_seen": 26873920,
      "step": 127340
    },
    {
      "epoch": 14.00935093509351,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.007479058286041548,
      "loss": 0.2324,
      "num_input_tokens_seen": 26875040,
      "step": 127345
    },
    {
      "epoch": 14.009900990099009,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.007477812371830902,
      "loss": 0.2293,
      "num_input_tokens_seen": 26876032,
      "step": 127350
    },
    {
      "epoch": 14.01045104510451,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.007476566526948654,
      "loss": 0.2324,
      "num_input_tokens_seen": 26876992,
      "step": 127355
    },
    {
      "epoch": 14.011001100110011,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.007475320751406276,
      "loss": 0.2314,
      "num_input_tokens_seen": 26877984,
      "step": 127360
    },
    {
      "epoch": 14.011551155115512,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007474075045215256,
      "loss": 0.2314,
      "num_input_tokens_seen": 26879008,
      "step": 127365
    },
    {
      "epoch": 14.012101210121012,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.007472829408387082,
      "loss": 0.2319,
      "num_input_tokens_seen": 26880064,
      "step": 127370
    },
    {
      "epoch": 14.012651265126513,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00747158384093322,
      "loss": 0.2319,
      "num_input_tokens_seen": 26881120,
      "step": 127375
    },
    {
      "epoch": 14.013201320132014,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007470338342865164,
      "loss": 0.2309,
      "num_input_tokens_seen": 26882176,
      "step": 127380
    },
    {
      "epoch": 14.013751375137513,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.007469092914194382,
      "loss": 0.2319,
      "num_input_tokens_seen": 26883200,
      "step": 127385
    },
    {
      "epoch": 14.014301430143014,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.00746784755493236,
      "loss": 0.2319,
      "num_input_tokens_seen": 26884256,
      "step": 127390
    },
    {
      "epoch": 14.014851485148515,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007466602265090569,
      "loss": 0.2298,
      "num_input_tokens_seen": 26885312,
      "step": 127395
    },
    {
      "epoch": 14.015401540154015,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.00746535704468049,
      "loss": 0.2324,
      "num_input_tokens_seen": 26886336,
      "step": 127400
    },
    {
      "epoch": 14.015951595159516,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.007464111893713605,
      "loss": 0.2303,
      "num_input_tokens_seen": 26887392,
      "step": 127405
    },
    {
      "epoch": 14.016501650165017,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.007462866812201384,
      "loss": 0.2324,
      "num_input_tokens_seen": 26888384,
      "step": 127410
    },
    {
      "epoch": 14.017051705170518,
      "grad_norm": 0.000713348388671875,
      "learning_rate": 0.007461621800155297,
      "loss": 0.2314,
      "num_input_tokens_seen": 26889440,
      "step": 127415
    },
    {
      "epoch": 14.017601760176017,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.007460376857586826,
      "loss": 0.2329,
      "num_input_tokens_seen": 26890496,
      "step": 127420
    },
    {
      "epoch": 14.018151815181518,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.0074591319845074425,
      "loss": 0.2314,
      "num_input_tokens_seen": 26891520,
      "step": 127425
    },
    {
      "epoch": 14.01870187018702,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.007457887180928626,
      "loss": 0.2314,
      "num_input_tokens_seen": 26892544,
      "step": 127430
    },
    {
      "epoch": 14.019251925192519,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007456642446861846,
      "loss": 0.2309,
      "num_input_tokens_seen": 26893664,
      "step": 127435
    },
    {
      "epoch": 14.01980198019802,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0074553977823185675,
      "loss": 0.2298,
      "num_input_tokens_seen": 26894720,
      "step": 127440
    },
    {
      "epoch": 14.020352035203521,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.007454153187310271,
      "loss": 0.2314,
      "num_input_tokens_seen": 26895712,
      "step": 127445
    },
    {
      "epoch": 14.02090209020902,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.007452908661848419,
      "loss": 0.2319,
      "num_input_tokens_seen": 26896736,
      "step": 127450
    },
    {
      "epoch": 14.021452145214521,
      "grad_norm": 0.000774383544921875,
      "learning_rate": 0.0074516642059444865,
      "loss": 0.2319,
      "num_input_tokens_seen": 26897792,
      "step": 127455
    },
    {
      "epoch": 14.022002200220022,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007450419819609949,
      "loss": 0.2319,
      "num_input_tokens_seen": 26898880,
      "step": 127460
    },
    {
      "epoch": 14.022552255225522,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.007449175502856263,
      "loss": 0.2319,
      "num_input_tokens_seen": 26899936,
      "step": 127465
    },
    {
      "epoch": 14.023102310231023,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00744793125569491,
      "loss": 0.2319,
      "num_input_tokens_seen": 26900992,
      "step": 127470
    },
    {
      "epoch": 14.023652365236524,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.007446687078137343,
      "loss": 0.2335,
      "num_input_tokens_seen": 26902112,
      "step": 127475
    },
    {
      "epoch": 14.024202420242025,
      "grad_norm": 0.00072479248046875,
      "learning_rate": 0.007445442970195039,
      "loss": 0.2319,
      "num_input_tokens_seen": 26903168,
      "step": 127480
    },
    {
      "epoch": 14.024752475247524,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.007444198931879467,
      "loss": 0.2309,
      "num_input_tokens_seen": 26904224,
      "step": 127485
    },
    {
      "epoch": 14.025302530253025,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.007442954963202083,
      "loss": 0.2319,
      "num_input_tokens_seen": 26905344,
      "step": 127490
    },
    {
      "epoch": 14.025852585258527,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00744171106417436,
      "loss": 0.2304,
      "num_input_tokens_seen": 26906400,
      "step": 127495
    },
    {
      "epoch": 14.026402640264026,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.007440467234807757,
      "loss": 0.2319,
      "num_input_tokens_seen": 26907520,
      "step": 127500
    },
    {
      "epoch": 14.026952695269527,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.007439223475113744,
      "loss": 0.2314,
      "num_input_tokens_seen": 26908544,
      "step": 127505
    },
    {
      "epoch": 14.027502750275028,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0074379797851037745,
      "loss": 0.2303,
      "num_input_tokens_seen": 26909568,
      "step": 127510
    },
    {
      "epoch": 14.028052805280527,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007436736164789316,
      "loss": 0.2324,
      "num_input_tokens_seen": 26910560,
      "step": 127515
    },
    {
      "epoch": 14.028602860286028,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007435492614181838,
      "loss": 0.2309,
      "num_input_tokens_seen": 26911552,
      "step": 127520
    },
    {
      "epoch": 14.02915291529153,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.0074342491332927895,
      "loss": 0.2324,
      "num_input_tokens_seen": 26912512,
      "step": 127525
    },
    {
      "epoch": 14.029702970297029,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.0074330057221336406,
      "loss": 0.2319,
      "num_input_tokens_seen": 26913568,
      "step": 127530
    },
    {
      "epoch": 14.03025302530253,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.007431762380715841,
      "loss": 0.2324,
      "num_input_tokens_seen": 26914528,
      "step": 127535
    },
    {
      "epoch": 14.030803080308031,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.007430519109050858,
      "loss": 0.2303,
      "num_input_tokens_seen": 26915552,
      "step": 127540
    },
    {
      "epoch": 14.031353135313532,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.007429275907150153,
      "loss": 0.2319,
      "num_input_tokens_seen": 26916544,
      "step": 127545
    },
    {
      "epoch": 14.031903190319031,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.007428032775025172,
      "loss": 0.2314,
      "num_input_tokens_seen": 26917632,
      "step": 127550
    },
    {
      "epoch": 14.032453245324533,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.007426789712687386,
      "loss": 0.2314,
      "num_input_tokens_seen": 26918720,
      "step": 127555
    },
    {
      "epoch": 14.033003300330034,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0074255467201482465,
      "loss": 0.2324,
      "num_input_tokens_seen": 26919776,
      "step": 127560
    },
    {
      "epoch": 14.033553355335533,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.007424303797419201,
      "loss": 0.2314,
      "num_input_tokens_seen": 26920768,
      "step": 127565
    },
    {
      "epoch": 14.034103410341034,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.007423060944511714,
      "loss": 0.2324,
      "num_input_tokens_seen": 26921856,
      "step": 127570
    },
    {
      "epoch": 14.034653465346535,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.007421818161437236,
      "loss": 0.2314,
      "num_input_tokens_seen": 26922976,
      "step": 127575
    },
    {
      "epoch": 14.035203520352034,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.007420575448207232,
      "loss": 0.2329,
      "num_input_tokens_seen": 26924032,
      "step": 127580
    },
    {
      "epoch": 14.035753575357536,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007419332804833145,
      "loss": 0.2293,
      "num_input_tokens_seen": 26925024,
      "step": 127585
    },
    {
      "epoch": 14.036303630363037,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007418090231326426,
      "loss": 0.2298,
      "num_input_tokens_seen": 26926048,
      "step": 127590
    },
    {
      "epoch": 14.036853685368538,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.007416847727698529,
      "loss": 0.2314,
      "num_input_tokens_seen": 26927040,
      "step": 127595
    },
    {
      "epoch": 14.037403740374037,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.007415605293960909,
      "loss": 0.2308,
      "num_input_tokens_seen": 26928128,
      "step": 127600
    },
    {
      "epoch": 14.037953795379538,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.007414362930125022,
      "loss": 0.2319,
      "num_input_tokens_seen": 26929152,
      "step": 127605
    },
    {
      "epoch": 14.03850385038504,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.007413120636202312,
      "loss": 0.2314,
      "num_input_tokens_seen": 26930240,
      "step": 127610
    },
    {
      "epoch": 14.039053905390539,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.0074118784122042215,
      "loss": 0.2293,
      "num_input_tokens_seen": 26931296,
      "step": 127615
    },
    {
      "epoch": 14.03960396039604,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.007410636258142214,
      "loss": 0.2303,
      "num_input_tokens_seen": 26932352,
      "step": 127620
    },
    {
      "epoch": 14.04015401540154,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0074093941740277225,
      "loss": 0.2324,
      "num_input_tokens_seen": 26933376,
      "step": 127625
    },
    {
      "epoch": 14.04070407040704,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.0074081521598722035,
      "loss": 0.2303,
      "num_input_tokens_seen": 26934432,
      "step": 127630
    },
    {
      "epoch": 14.041254125412541,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.007406910215687109,
      "loss": 0.2314,
      "num_input_tokens_seen": 26935456,
      "step": 127635
    },
    {
      "epoch": 14.041804180418042,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.007405668341483875,
      "loss": 0.2303,
      "num_input_tokens_seen": 26936544,
      "step": 127640
    },
    {
      "epoch": 14.042354235423542,
      "grad_norm": 0.005859375,
      "learning_rate": 0.007404426537273957,
      "loss": 0.2319,
      "num_input_tokens_seen": 26937600,
      "step": 127645
    },
    {
      "epoch": 14.042904290429043,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.007403184803068793,
      "loss": 0.2298,
      "num_input_tokens_seen": 26938624,
      "step": 127650
    },
    {
      "epoch": 14.043454345434544,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00740194313887982,
      "loss": 0.2298,
      "num_input_tokens_seen": 26939616,
      "step": 127655
    },
    {
      "epoch": 14.044004400440045,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.007400701544718502,
      "loss": 0.2319,
      "num_input_tokens_seen": 26940672,
      "step": 127660
    },
    {
      "epoch": 14.044554455445544,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.007399460020596265,
      "loss": 0.2309,
      "num_input_tokens_seen": 26941728,
      "step": 127665
    },
    {
      "epoch": 14.045104510451045,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.007398218566524563,
      "loss": 0.2314,
      "num_input_tokens_seen": 26942784,
      "step": 127670
    },
    {
      "epoch": 14.045654565456546,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.007396977182514835,
      "loss": 0.2324,
      "num_input_tokens_seen": 26943840,
      "step": 127675
    },
    {
      "epoch": 14.046204620462046,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0073957358685785134,
      "loss": 0.2309,
      "num_input_tokens_seen": 26944896,
      "step": 127680
    },
    {
      "epoch": 14.046754675467547,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.007394494624727045,
      "loss": 0.2303,
      "num_input_tokens_seen": 26945984,
      "step": 127685
    },
    {
      "epoch": 14.047304730473048,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00739325345097187,
      "loss": 0.2314,
      "num_input_tokens_seen": 26946976,
      "step": 127690
    },
    {
      "epoch": 14.047854785478547,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.007392012347324435,
      "loss": 0.2335,
      "num_input_tokens_seen": 26948000,
      "step": 127695
    },
    {
      "epoch": 14.048404840484048,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.00739077131379617,
      "loss": 0.2314,
      "num_input_tokens_seen": 26949024,
      "step": 127700
    },
    {
      "epoch": 14.04895489548955,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.007389530350398508,
      "loss": 0.2314,
      "num_input_tokens_seen": 26950080,
      "step": 127705
    },
    {
      "epoch": 14.049504950495049,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.0073882894571429,
      "loss": 0.2319,
      "num_input_tokens_seen": 26951136,
      "step": 127710
    },
    {
      "epoch": 14.05005500550055,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007387048634040765,
      "loss": 0.2319,
      "num_input_tokens_seen": 26952256,
      "step": 127715
    },
    {
      "epoch": 14.05060506050605,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.007385807881103559,
      "loss": 0.2298,
      "num_input_tokens_seen": 26953312,
      "step": 127720
    },
    {
      "epoch": 14.051155115511552,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007384567198342709,
      "loss": 0.2308,
      "num_input_tokens_seen": 26954368,
      "step": 127725
    },
    {
      "epoch": 14.051705170517051,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007383326585769645,
      "loss": 0.2288,
      "num_input_tokens_seen": 26955424,
      "step": 127730
    },
    {
      "epoch": 14.052255225522552,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.007382086043395809,
      "loss": 0.2293,
      "num_input_tokens_seen": 26956448,
      "step": 127735
    },
    {
      "epoch": 14.052805280528053,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.007380845571232626,
      "loss": 0.2308,
      "num_input_tokens_seen": 26957504,
      "step": 127740
    },
    {
      "epoch": 14.053355335533553,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.007379605169291532,
      "loss": 0.2303,
      "num_input_tokens_seen": 26958528,
      "step": 127745
    },
    {
      "epoch": 14.053905390539054,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.007378364837583965,
      "loss": 0.2304,
      "num_input_tokens_seen": 26959616,
      "step": 127750
    },
    {
      "epoch": 14.054455445544555,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.007377124576121348,
      "loss": 0.2314,
      "num_input_tokens_seen": 26960704,
      "step": 127755
    },
    {
      "epoch": 14.055005500550054,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.007375884384915122,
      "loss": 0.2309,
      "num_input_tokens_seen": 26961792,
      "step": 127760
    },
    {
      "epoch": 14.055555555555555,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0073746442639767065,
      "loss": 0.2314,
      "num_input_tokens_seen": 26962848,
      "step": 127765
    },
    {
      "epoch": 14.056105610561056,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.00737340421331754,
      "loss": 0.2303,
      "num_input_tokens_seen": 26963840,
      "step": 127770
    },
    {
      "epoch": 14.056655665566556,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.007372164232949041,
      "loss": 0.2314,
      "num_input_tokens_seen": 26964832,
      "step": 127775
    },
    {
      "epoch": 14.057205720572057,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007370924322882644,
      "loss": 0.2319,
      "num_input_tokens_seen": 26965888,
      "step": 127780
    },
    {
      "epoch": 14.057755775577558,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.007369684483129785,
      "loss": 0.2313,
      "num_input_tokens_seen": 26966944,
      "step": 127785
    },
    {
      "epoch": 14.058305830583059,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.0073684447137018745,
      "loss": 0.2319,
      "num_input_tokens_seen": 26968000,
      "step": 127790
    },
    {
      "epoch": 14.058855885588558,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0073672050146103545,
      "loss": 0.2319,
      "num_input_tokens_seen": 26969056,
      "step": 127795
    },
    {
      "epoch": 14.05940594059406,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007365965385866637,
      "loss": 0.2335,
      "num_input_tokens_seen": 26970112,
      "step": 127800
    },
    {
      "epoch": 14.05995599559956,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.007364725827482152,
      "loss": 0.2314,
      "num_input_tokens_seen": 26971136,
      "step": 127805
    },
    {
      "epoch": 14.06050605060506,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.007363486339468333,
      "loss": 0.2329,
      "num_input_tokens_seen": 26972160,
      "step": 127810
    },
    {
      "epoch": 14.061056105610561,
      "grad_norm": 0.002716064453125,
      "learning_rate": 0.007362246921836594,
      "loss": 0.2303,
      "num_input_tokens_seen": 26973248,
      "step": 127815
    },
    {
      "epoch": 14.061606160616062,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007361007574598355,
      "loss": 0.2308,
      "num_input_tokens_seen": 26974304,
      "step": 127820
    },
    {
      "epoch": 14.062156215621561,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00735976829776505,
      "loss": 0.2303,
      "num_input_tokens_seen": 26975264,
      "step": 127825
    },
    {
      "epoch": 14.062706270627062,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007358529091348089,
      "loss": 0.2303,
      "num_input_tokens_seen": 26976288,
      "step": 127830
    },
    {
      "epoch": 14.063256325632564,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007357289955358898,
      "loss": 0.2324,
      "num_input_tokens_seen": 26977408,
      "step": 127835
    },
    {
      "epoch": 14.063806380638065,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007356050889808906,
      "loss": 0.2329,
      "num_input_tokens_seen": 26978464,
      "step": 127840
    },
    {
      "epoch": 14.064356435643564,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.007354811894709517,
      "loss": 0.2308,
      "num_input_tokens_seen": 26979488,
      "step": 127845
    },
    {
      "epoch": 14.064906490649065,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007353572970072164,
      "loss": 0.2309,
      "num_input_tokens_seen": 26980544,
      "step": 127850
    },
    {
      "epoch": 14.065456545654566,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.007352334115908255,
      "loss": 0.2298,
      "num_input_tokens_seen": 26981664,
      "step": 127855
    },
    {
      "epoch": 14.066006600660065,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007351095332229212,
      "loss": 0.2303,
      "num_input_tokens_seen": 26982720,
      "step": 127860
    },
    {
      "epoch": 14.066556655665567,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.007349856619046459,
      "loss": 0.2329,
      "num_input_tokens_seen": 26983776,
      "step": 127865
    },
    {
      "epoch": 14.067106710671068,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0073486179763714,
      "loss": 0.2319,
      "num_input_tokens_seen": 26984832,
      "step": 127870
    },
    {
      "epoch": 14.067656765676567,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.007347379404215464,
      "loss": 0.2329,
      "num_input_tokens_seen": 26985856,
      "step": 127875
    },
    {
      "epoch": 14.068206820682068,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007346140902590056,
      "loss": 0.2309,
      "num_input_tokens_seen": 26986976,
      "step": 127880
    },
    {
      "epoch": 14.06875687568757,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.007344902471506599,
      "loss": 0.2309,
      "num_input_tokens_seen": 26988000,
      "step": 127885
    },
    {
      "epoch": 14.069306930693068,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.007343664110976497,
      "loss": 0.2314,
      "num_input_tokens_seen": 26989056,
      "step": 127890
    },
    {
      "epoch": 14.06985698569857,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.007342425821011169,
      "loss": 0.2324,
      "num_input_tokens_seen": 26990144,
      "step": 127895
    },
    {
      "epoch": 14.07040704070407,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0073411876016220325,
      "loss": 0.2345,
      "num_input_tokens_seen": 26991200,
      "step": 127900
    },
    {
      "epoch": 14.070957095709572,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00733994945282049,
      "loss": 0.2309,
      "num_input_tokens_seen": 26992224,
      "step": 127905
    },
    {
      "epoch": 14.071507150715071,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.007338711374617963,
      "loss": 0.2288,
      "num_input_tokens_seen": 26993184,
      "step": 127910
    },
    {
      "epoch": 14.072057205720572,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.007337473367025851,
      "loss": 0.2308,
      "num_input_tokens_seen": 26994208,
      "step": 127915
    },
    {
      "epoch": 14.072607260726073,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007336235430055572,
      "loss": 0.2314,
      "num_input_tokens_seen": 26995328,
      "step": 127920
    },
    {
      "epoch": 14.073157315731573,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.007334997563718538,
      "loss": 0.2324,
      "num_input_tokens_seen": 26996384,
      "step": 127925
    },
    {
      "epoch": 14.073707370737074,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.0073337597680261475,
      "loss": 0.233,
      "num_input_tokens_seen": 26997472,
      "step": 127930
    },
    {
      "epoch": 14.074257425742575,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.007332522042989822,
      "loss": 0.2303,
      "num_input_tokens_seen": 26998464,
      "step": 127935
    },
    {
      "epoch": 14.074807480748074,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.007331284388620959,
      "loss": 0.2314,
      "num_input_tokens_seen": 26999520,
      "step": 127940
    },
    {
      "epoch": 14.075357535753575,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0073300468049309645,
      "loss": 0.2303,
      "num_input_tokens_seen": 27000608,
      "step": 127945
    },
    {
      "epoch": 14.075907590759076,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.007328809291931247,
      "loss": 0.2298,
      "num_input_tokens_seen": 27001696,
      "step": 127950
    },
    {
      "epoch": 14.076457645764576,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.007327571849633213,
      "loss": 0.2298,
      "num_input_tokens_seen": 27002752,
      "step": 127955
    },
    {
      "epoch": 14.077007700770077,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.007326334478048274,
      "loss": 0.2309,
      "num_input_tokens_seen": 27003776,
      "step": 127960
    },
    {
      "epoch": 14.077557755775578,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007325097177187827,
      "loss": 0.2314,
      "num_input_tokens_seen": 27004864,
      "step": 127965
    },
    {
      "epoch": 14.078107810781079,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.007323859947063273,
      "loss": 0.2314,
      "num_input_tokens_seen": 27005920,
      "step": 127970
    },
    {
      "epoch": 14.078657865786578,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.007322622787686016,
      "loss": 0.2324,
      "num_input_tokens_seen": 27007008,
      "step": 127975
    },
    {
      "epoch": 14.07920792079208,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0073213856990674625,
      "loss": 0.2298,
      "num_input_tokens_seen": 27008096,
      "step": 127980
    },
    {
      "epoch": 14.07975797579758,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.007320148681219018,
      "loss": 0.2319,
      "num_input_tokens_seen": 27009120,
      "step": 127985
    },
    {
      "epoch": 14.08030803080308,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007318911734152078,
      "loss": 0.2309,
      "num_input_tokens_seen": 27010240,
      "step": 127990
    },
    {
      "epoch": 14.08085808580858,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0073176748578780365,
      "loss": 0.2313,
      "num_input_tokens_seen": 27011328,
      "step": 127995
    },
    {
      "epoch": 14.081408140814082,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007316438052408305,
      "loss": 0.2309,
      "num_input_tokens_seen": 27012320,
      "step": 128000
    },
    {
      "epoch": 14.081958195819581,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.007315201317754272,
      "loss": 0.2324,
      "num_input_tokens_seen": 27013408,
      "step": 128005
    },
    {
      "epoch": 14.082508250825082,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007313964653927339,
      "loss": 0.2308,
      "num_input_tokens_seen": 27014400,
      "step": 128010
    },
    {
      "epoch": 14.083058305830583,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007312728060938915,
      "loss": 0.2319,
      "num_input_tokens_seen": 27015424,
      "step": 128015
    },
    {
      "epoch": 14.083608360836084,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.007311491538800379,
      "loss": 0.2319,
      "num_input_tokens_seen": 27016512,
      "step": 128020
    },
    {
      "epoch": 14.084158415841584,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.007310255087523143,
      "loss": 0.2324,
      "num_input_tokens_seen": 27017600,
      "step": 128025
    },
    {
      "epoch": 14.084708470847085,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00730901870711859,
      "loss": 0.2314,
      "num_input_tokens_seen": 27018592,
      "step": 128030
    },
    {
      "epoch": 14.085258525852586,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.007307782397598122,
      "loss": 0.2303,
      "num_input_tokens_seen": 27019648,
      "step": 128035
    },
    {
      "epoch": 14.085808580858085,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0073065461589731364,
      "loss": 0.2329,
      "num_input_tokens_seen": 27020640,
      "step": 128040
    },
    {
      "epoch": 14.086358635863586,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.00730530999125502,
      "loss": 0.2308,
      "num_input_tokens_seen": 27021600,
      "step": 128045
    },
    {
      "epoch": 14.086908690869087,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.007304073894455173,
      "loss": 0.2298,
      "num_input_tokens_seen": 27022688,
      "step": 128050
    },
    {
      "epoch": 14.087458745874587,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007302837868584985,
      "loss": 0.2309,
      "num_input_tokens_seen": 27023712,
      "step": 128055
    },
    {
      "epoch": 14.088008800880088,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00730160191365584,
      "loss": 0.2314,
      "num_input_tokens_seen": 27024832,
      "step": 128060
    },
    {
      "epoch": 14.088558855885589,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.007300366029679137,
      "loss": 0.2304,
      "num_input_tokens_seen": 27025952,
      "step": 128065
    },
    {
      "epoch": 14.089108910891088,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.007299130216666265,
      "loss": 0.2319,
      "num_input_tokens_seen": 27027008,
      "step": 128070
    },
    {
      "epoch": 14.08965896589659,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007297894474628621,
      "loss": 0.2293,
      "num_input_tokens_seen": 27028000,
      "step": 128075
    },
    {
      "epoch": 14.09020902090209,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007296658803577588,
      "loss": 0.2308,
      "num_input_tokens_seen": 27029024,
      "step": 128080
    },
    {
      "epoch": 14.090759075907592,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0072954232035245484,
      "loss": 0.2314,
      "num_input_tokens_seen": 27030080,
      "step": 128085
    },
    {
      "epoch": 14.091309130913091,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.007294187674480895,
      "loss": 0.2314,
      "num_input_tokens_seen": 27031136,
      "step": 128090
    },
    {
      "epoch": 14.091859185918592,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.007292952216458017,
      "loss": 0.2303,
      "num_input_tokens_seen": 27032256,
      "step": 128095
    },
    {
      "epoch": 14.092409240924093,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0072917168294673075,
      "loss": 0.2308,
      "num_input_tokens_seen": 27033248,
      "step": 128100
    },
    {
      "epoch": 14.092959295929592,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.007290481513520143,
      "loss": 0.2324,
      "num_input_tokens_seen": 27034336,
      "step": 128105
    },
    {
      "epoch": 14.093509350935093,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0072892462686279065,
      "loss": 0.2324,
      "num_input_tokens_seen": 27035360,
      "step": 128110
    },
    {
      "epoch": 14.094059405940595,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.007288011094801992,
      "loss": 0.2319,
      "num_input_tokens_seen": 27036352,
      "step": 128115
    },
    {
      "epoch": 14.094609460946094,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.007286775992053772,
      "loss": 0.2314,
      "num_input_tokens_seen": 27037408,
      "step": 128120
    },
    {
      "epoch": 14.095159515951595,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.00728554096039464,
      "loss": 0.2303,
      "num_input_tokens_seen": 27038464,
      "step": 128125
    },
    {
      "epoch": 14.095709570957096,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.007284305999835978,
      "loss": 0.2309,
      "num_input_tokens_seen": 27039552,
      "step": 128130
    },
    {
      "epoch": 14.096259625962595,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007283071110389161,
      "loss": 0.2303,
      "num_input_tokens_seen": 27040608,
      "step": 128135
    },
    {
      "epoch": 14.096809680968097,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.007281836292065578,
      "loss": 0.2314,
      "num_input_tokens_seen": 27041632,
      "step": 128140
    },
    {
      "epoch": 14.097359735973598,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.007280601544876603,
      "loss": 0.2314,
      "num_input_tokens_seen": 27042656,
      "step": 128145
    },
    {
      "epoch": 14.097909790979099,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007279366868833619,
      "loss": 0.2308,
      "num_input_tokens_seen": 27043744,
      "step": 128150
    },
    {
      "epoch": 14.098459845984598,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0072781322639480115,
      "loss": 0.2298,
      "num_input_tokens_seen": 27044800,
      "step": 128155
    },
    {
      "epoch": 14.099009900990099,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007276897730231147,
      "loss": 0.2303,
      "num_input_tokens_seen": 27045856,
      "step": 128160
    },
    {
      "epoch": 14.0995599559956,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007275663267694417,
      "loss": 0.2293,
      "num_input_tokens_seen": 27046880,
      "step": 128165
    },
    {
      "epoch": 14.1001100110011,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.007274428876349185,
      "loss": 0.2314,
      "num_input_tokens_seen": 27047936,
      "step": 128170
    },
    {
      "epoch": 14.1006600660066,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007273194556206843,
      "loss": 0.2303,
      "num_input_tokens_seen": 27049024,
      "step": 128175
    },
    {
      "epoch": 14.101210121012102,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007271960307278752,
      "loss": 0.2303,
      "num_input_tokens_seen": 27050080,
      "step": 128180
    },
    {
      "epoch": 14.101760176017601,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.007270726129576296,
      "loss": 0.2303,
      "num_input_tokens_seen": 27051104,
      "step": 128185
    },
    {
      "epoch": 14.102310231023102,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0072694920231108524,
      "loss": 0.2303,
      "num_input_tokens_seen": 27052192,
      "step": 128190
    },
    {
      "epoch": 14.102860286028603,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.007268257987893791,
      "loss": 0.2324,
      "num_input_tokens_seen": 27053248,
      "step": 128195
    },
    {
      "epoch": 14.103410341034103,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.007267024023936481,
      "loss": 0.2308,
      "num_input_tokens_seen": 27054368,
      "step": 128200
    },
    {
      "epoch": 14.103960396039604,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0072657901312503,
      "loss": 0.2319,
      "num_input_tokens_seen": 27055424,
      "step": 128205
    },
    {
      "epoch": 14.104510451045105,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00726455630984662,
      "loss": 0.2298,
      "num_input_tokens_seen": 27056512,
      "step": 128210
    },
    {
      "epoch": 14.105060506050606,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0072633225597368185,
      "loss": 0.2314,
      "num_input_tokens_seen": 27057536,
      "step": 128215
    },
    {
      "epoch": 14.105610561056105,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00726208888093226,
      "loss": 0.2314,
      "num_input_tokens_seen": 27058624,
      "step": 128220
    },
    {
      "epoch": 14.106160616061606,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.00726085527344431,
      "loss": 0.2335,
      "num_input_tokens_seen": 27059648,
      "step": 128225
    },
    {
      "epoch": 14.106710671067107,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007259621737284348,
      "loss": 0.2298,
      "num_input_tokens_seen": 27060768,
      "step": 128230
    },
    {
      "epoch": 14.107260726072607,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.007258388272463733,
      "loss": 0.2329,
      "num_input_tokens_seen": 27061792,
      "step": 128235
    },
    {
      "epoch": 14.107810781078108,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0072571548789938396,
      "loss": 0.2314,
      "num_input_tokens_seen": 27062848,
      "step": 128240
    },
    {
      "epoch": 14.108360836083609,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.007255921556886039,
      "loss": 0.2335,
      "num_input_tokens_seen": 27063904,
      "step": 128245
    },
    {
      "epoch": 14.108910891089108,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007254688306151689,
      "loss": 0.2329,
      "num_input_tokens_seen": 27065088,
      "step": 128250
    },
    {
      "epoch": 14.10946094609461,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.007253455126802164,
      "loss": 0.2303,
      "num_input_tokens_seen": 27066144,
      "step": 128255
    },
    {
      "epoch": 14.11001100110011,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007252222018848821,
      "loss": 0.2329,
      "num_input_tokens_seen": 27067168,
      "step": 128260
    },
    {
      "epoch": 14.110561056105611,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.0072509889823030305,
      "loss": 0.2319,
      "num_input_tokens_seen": 27068256,
      "step": 128265
    },
    {
      "epoch": 14.11111111111111,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.007249756017176163,
      "loss": 0.2309,
      "num_input_tokens_seen": 27069312,
      "step": 128270
    },
    {
      "epoch": 14.111661166116612,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.007248523123479567,
      "loss": 0.2319,
      "num_input_tokens_seen": 27070304,
      "step": 128275
    },
    {
      "epoch": 14.112211221122113,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.007247290301224622,
      "loss": 0.2314,
      "num_input_tokens_seen": 27071360,
      "step": 128280
    },
    {
      "epoch": 14.112761276127612,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0072460575504226755,
      "loss": 0.2303,
      "num_input_tokens_seen": 27072448,
      "step": 128285
    },
    {
      "epoch": 14.113311331133113,
      "grad_norm": 0.002655029296875,
      "learning_rate": 0.007244824871085101,
      "loss": 0.2324,
      "num_input_tokens_seen": 27073536,
      "step": 128290
    },
    {
      "epoch": 14.113861386138614,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00724359226322325,
      "loss": 0.2309,
      "num_input_tokens_seen": 27074624,
      "step": 128295
    },
    {
      "epoch": 14.114411441144114,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007242359726848485,
      "loss": 0.2314,
      "num_input_tokens_seen": 27075616,
      "step": 128300
    },
    {
      "epoch": 14.114961496149615,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.007241127261972176,
      "loss": 0.2335,
      "num_input_tokens_seen": 27076640,
      "step": 128305
    },
    {
      "epoch": 14.115511551155116,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.007239894868605667,
      "loss": 0.2324,
      "num_input_tokens_seen": 27077632,
      "step": 128310
    },
    {
      "epoch": 14.116061606160615,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.007238662546760329,
      "loss": 0.2319,
      "num_input_tokens_seen": 27078720,
      "step": 128315
    },
    {
      "epoch": 14.116611661166116,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.007237430296447507,
      "loss": 0.2324,
      "num_input_tokens_seen": 27079808,
      "step": 128320
    },
    {
      "epoch": 14.117161716171617,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.007236198117678568,
      "loss": 0.2314,
      "num_input_tokens_seen": 27080864,
      "step": 128325
    },
    {
      "epoch": 14.117711771177119,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.007234966010464869,
      "loss": 0.2308,
      "num_input_tokens_seen": 27081952,
      "step": 128330
    },
    {
      "epoch": 14.118261826182618,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.007233733974817758,
      "loss": 0.2308,
      "num_input_tokens_seen": 27082976,
      "step": 128335
    },
    {
      "epoch": 14.118811881188119,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.007232502010748598,
      "loss": 0.2303,
      "num_input_tokens_seen": 27084032,
      "step": 128340
    },
    {
      "epoch": 14.11936193619362,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.007231270118268742,
      "loss": 0.2314,
      "num_input_tokens_seen": 27085120,
      "step": 128345
    },
    {
      "epoch": 14.11991199119912,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.007230038297389536,
      "loss": 0.2303,
      "num_input_tokens_seen": 27086208,
      "step": 128350
    },
    {
      "epoch": 14.12046204620462,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.007228806548122337,
      "loss": 0.2309,
      "num_input_tokens_seen": 27087232,
      "step": 128355
    },
    {
      "epoch": 14.121012101210122,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0072275748704785,
      "loss": 0.2309,
      "num_input_tokens_seen": 27088256,
      "step": 128360
    },
    {
      "epoch": 14.12156215621562,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.007226343264469381,
      "loss": 0.2329,
      "num_input_tokens_seen": 27089280,
      "step": 128365
    },
    {
      "epoch": 14.122112211221122,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.007225111730106327,
      "loss": 0.2288,
      "num_input_tokens_seen": 27090368,
      "step": 128370
    },
    {
      "epoch": 14.122662266226623,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.00722388026740068,
      "loss": 0.2319,
      "num_input_tokens_seen": 27091360,
      "step": 128375
    },
    {
      "epoch": 14.123212321232122,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.007222648876363799,
      "loss": 0.2319,
      "num_input_tokens_seen": 27092416,
      "step": 128380
    },
    {
      "epoch": 14.123762376237623,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.007221417557007036,
      "loss": 0.2298,
      "num_input_tokens_seen": 27093472,
      "step": 128385
    },
    {
      "epoch": 14.124312431243125,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.00722018630934173,
      "loss": 0.2314,
      "num_input_tokens_seen": 27094592,
      "step": 128390
    },
    {
      "epoch": 14.124862486248626,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.007218955133379239,
      "loss": 0.2298,
      "num_input_tokens_seen": 27095616,
      "step": 128395
    },
    {
      "epoch": 14.125412541254125,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.007217724029130899,
      "loss": 0.2303,
      "num_input_tokens_seen": 27096704,
      "step": 128400
    },
    {
      "epoch": 14.125962596259626,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.007216492996608067,
      "loss": 0.2308,
      "num_input_tokens_seen": 27097792,
      "step": 128405
    },
    {
      "epoch": 14.126512651265127,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.00721526203582208,
      "loss": 0.2314,
      "num_input_tokens_seen": 27098816,
      "step": 128410
    },
    {
      "epoch": 14.127062706270626,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.007214031146784286,
      "loss": 0.2319,
      "num_input_tokens_seen": 27099872,
      "step": 128415
    },
    {
      "epoch": 14.127612761276128,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007212800329506037,
      "loss": 0.2309,
      "num_input_tokens_seen": 27100896,
      "step": 128420
    },
    {
      "epoch": 14.128162816281629,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.007211569583998666,
      "loss": 0.2329,
      "num_input_tokens_seen": 27102016,
      "step": 128425
    },
    {
      "epoch": 14.128712871287128,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.007210338910273525,
      "loss": 0.2319,
      "num_input_tokens_seen": 27103072,
      "step": 128430
    },
    {
      "epoch": 14.129262926292629,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007209108308341953,
      "loss": 0.2324,
      "num_input_tokens_seen": 27104160,
      "step": 128435
    },
    {
      "epoch": 14.12981298129813,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.007207877778215281,
      "loss": 0.2303,
      "num_input_tokens_seen": 27105184,
      "step": 128440
    },
    {
      "epoch": 14.130363036303631,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.007206647319904871,
      "loss": 0.2335,
      "num_input_tokens_seen": 27106272,
      "step": 128445
    },
    {
      "epoch": 14.13091309130913,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.007205416933422048,
      "loss": 0.2314,
      "num_input_tokens_seen": 27107360,
      "step": 128450
    },
    {
      "epoch": 14.131463146314632,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.007204186618778161,
      "loss": 0.2309,
      "num_input_tokens_seen": 27108480,
      "step": 128455
    },
    {
      "epoch": 14.132013201320133,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.007202956375984547,
      "loss": 0.2314,
      "num_input_tokens_seen": 27109472,
      "step": 128460
    },
    {
      "epoch": 14.132563256325632,
      "grad_norm": 0.000484466552734375,
      "learning_rate": 0.007201726205052537,
      "loss": 0.2314,
      "num_input_tokens_seen": 27110528,
      "step": 128465
    },
    {
      "epoch": 14.133113311331133,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.007200496105993474,
      "loss": 0.2314,
      "num_input_tokens_seen": 27111552,
      "step": 128470
    },
    {
      "epoch": 14.133663366336634,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.007199266078818694,
      "loss": 0.2303,
      "num_input_tokens_seen": 27112544,
      "step": 128475
    },
    {
      "epoch": 14.134213421342134,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.007198036123539543,
      "loss": 0.2324,
      "num_input_tokens_seen": 27113632,
      "step": 128480
    },
    {
      "epoch": 14.134763476347635,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00719680624016735,
      "loss": 0.2319,
      "num_input_tokens_seen": 27114752,
      "step": 128485
    },
    {
      "epoch": 14.135313531353136,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.007195576428713443,
      "loss": 0.2293,
      "num_input_tokens_seen": 27115840,
      "step": 128490
    },
    {
      "epoch": 14.135863586358635,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.007194346689189168,
      "loss": 0.2335,
      "num_input_tokens_seen": 27116864,
      "step": 128495
    },
    {
      "epoch": 14.136413641364136,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.0071931170216058445,
      "loss": 0.2329,
      "num_input_tokens_seen": 27117920,
      "step": 128500
    },
    {
      "epoch": 14.136963696369637,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.007191887425974823,
      "loss": 0.2329,
      "num_input_tokens_seen": 27119040,
      "step": 128505
    },
    {
      "epoch": 14.137513751375138,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0071906579023074315,
      "loss": 0.2303,
      "num_input_tokens_seen": 27120096,
      "step": 128510
    },
    {
      "epoch": 14.138063806380638,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.0071894284506149935,
      "loss": 0.2319,
      "num_input_tokens_seen": 27121152,
      "step": 128515
    },
    {
      "epoch": 14.138613861386139,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.00718819907090885,
      "loss": 0.2303,
      "num_input_tokens_seen": 27122240,
      "step": 128520
    },
    {
      "epoch": 14.13916391639164,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.007186969763200322,
      "loss": 0.2314,
      "num_input_tokens_seen": 27123232,
      "step": 128525
    },
    {
      "epoch": 14.13971397139714,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.007185740527500744,
      "loss": 0.2303,
      "num_input_tokens_seen": 27124256,
      "step": 128530
    },
    {
      "epoch": 14.14026402640264,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007184511363821453,
      "loss": 0.2314,
      "num_input_tokens_seen": 27125248,
      "step": 128535
    },
    {
      "epoch": 14.140814081408141,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007183282272173763,
      "loss": 0.2304,
      "num_input_tokens_seen": 27126336,
      "step": 128540
    },
    {
      "epoch": 14.14136413641364,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.007182053252569016,
      "loss": 0.2309,
      "num_input_tokens_seen": 27127392,
      "step": 128545
    },
    {
      "epoch": 14.141914191419142,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007180824305018528,
      "loss": 0.2314,
      "num_input_tokens_seen": 27128448,
      "step": 128550
    },
    {
      "epoch": 14.142464246424643,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.007179595429533636,
      "loss": 0.2335,
      "num_input_tokens_seen": 27129504,
      "step": 128555
    },
    {
      "epoch": 14.143014301430142,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0071783666261256525,
      "loss": 0.2303,
      "num_input_tokens_seen": 27130528,
      "step": 128560
    },
    {
      "epoch": 14.143564356435643,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.007177137894805913,
      "loss": 0.2293,
      "num_input_tokens_seen": 27131616,
      "step": 128565
    },
    {
      "epoch": 14.144114411441144,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007175909235585744,
      "loss": 0.2314,
      "num_input_tokens_seen": 27132640,
      "step": 128570
    },
    {
      "epoch": 14.144664466446645,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.007174680648476465,
      "loss": 0.2309,
      "num_input_tokens_seen": 27133728,
      "step": 128575
    },
    {
      "epoch": 14.145214521452145,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.007173452133489395,
      "loss": 0.2324,
      "num_input_tokens_seen": 27134784,
      "step": 128580
    },
    {
      "epoch": 14.145764576457646,
      "grad_norm": 0.000537872314453125,
      "learning_rate": 0.00717222369063586,
      "loss": 0.2308,
      "num_input_tokens_seen": 27135776,
      "step": 128585
    },
    {
      "epoch": 14.146314631463147,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.007170995319927185,
      "loss": 0.2314,
      "num_input_tokens_seen": 27136832,
      "step": 128590
    },
    {
      "epoch": 14.146864686468646,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0071697670213746924,
      "loss": 0.2335,
      "num_input_tokens_seen": 27137888,
      "step": 128595
    },
    {
      "epoch": 14.147414741474147,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007168538794989702,
      "loss": 0.2319,
      "num_input_tokens_seen": 27138944,
      "step": 128600
    },
    {
      "epoch": 14.147964796479648,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.007167310640783524,
      "loss": 0.234,
      "num_input_tokens_seen": 27139968,
      "step": 128605
    },
    {
      "epoch": 14.148514851485148,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007166082558767494,
      "loss": 0.2313,
      "num_input_tokens_seen": 27141024,
      "step": 128610
    },
    {
      "epoch": 14.149064906490649,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.0071648545489529135,
      "loss": 0.2303,
      "num_input_tokens_seen": 27142080,
      "step": 128615
    },
    {
      "epoch": 14.14961496149615,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0071636266113511106,
      "loss": 0.2298,
      "num_input_tokens_seen": 27143136,
      "step": 128620
    },
    {
      "epoch": 14.150165016501651,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.007162398745973405,
      "loss": 0.2324,
      "num_input_tokens_seen": 27144224,
      "step": 128625
    },
    {
      "epoch": 14.15071507150715,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0071611709528311045,
      "loss": 0.2308,
      "num_input_tokens_seen": 27145280,
      "step": 128630
    },
    {
      "epoch": 14.151265126512651,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007159943231935536,
      "loss": 0.2324,
      "num_input_tokens_seen": 27146304,
      "step": 128635
    },
    {
      "epoch": 14.151815181518153,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007158715583298004,
      "loss": 0.2313,
      "num_input_tokens_seen": 27147392,
      "step": 128640
    },
    {
      "epoch": 14.152365236523652,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.007157488006929826,
      "loss": 0.2298,
      "num_input_tokens_seen": 27148384,
      "step": 128645
    },
    {
      "epoch": 14.152915291529153,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0071562605028423236,
      "loss": 0.2303,
      "num_input_tokens_seen": 27149344,
      "step": 128650
    },
    {
      "epoch": 14.153465346534654,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0071550330710468,
      "loss": 0.2308,
      "num_input_tokens_seen": 27150432,
      "step": 128655
    },
    {
      "epoch": 14.154015401540153,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0071538057115545775,
      "loss": 0.2324,
      "num_input_tokens_seen": 27151520,
      "step": 128660
    },
    {
      "epoch": 14.154565456545654,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007152578424376955,
      "loss": 0.2303,
      "num_input_tokens_seen": 27152576,
      "step": 128665
    },
    {
      "epoch": 14.155115511551156,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.00715135120952526,
      "loss": 0.2293,
      "num_input_tokens_seen": 27153664,
      "step": 128670
    },
    {
      "epoch": 14.155665566556655,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.007150124067010788,
      "loss": 0.2319,
      "num_input_tokens_seen": 27154816,
      "step": 128675
    },
    {
      "epoch": 14.156215621562156,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.007148896996844857,
      "loss": 0.2324,
      "num_input_tokens_seen": 27155904,
      "step": 128680
    },
    {
      "epoch": 14.156765676567657,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.007147669999038778,
      "loss": 0.2303,
      "num_input_tokens_seen": 27156928,
      "step": 128685
    },
    {
      "epoch": 14.157315731573158,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.007146443073603854,
      "loss": 0.2303,
      "num_input_tokens_seen": 27157952,
      "step": 128690
    },
    {
      "epoch": 14.157865786578657,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0071452162205513986,
      "loss": 0.2324,
      "num_input_tokens_seen": 27158944,
      "step": 128695
    },
    {
      "epoch": 14.158415841584159,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.007143989439892712,
      "loss": 0.2303,
      "num_input_tokens_seen": 27160000,
      "step": 128700
    },
    {
      "epoch": 14.15896589658966,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007142762731639106,
      "loss": 0.2293,
      "num_input_tokens_seen": 27161088,
      "step": 128705
    },
    {
      "epoch": 14.159515951595159,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0071415360958018885,
      "loss": 0.2314,
      "num_input_tokens_seen": 27162112,
      "step": 128710
    },
    {
      "epoch": 14.16006600660066,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0071403095323923586,
      "loss": 0.2314,
      "num_input_tokens_seen": 27163200,
      "step": 128715
    },
    {
      "epoch": 14.160616061606161,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.0071390830414218304,
      "loss": 0.2293,
      "num_input_tokens_seen": 27164224,
      "step": 128720
    },
    {
      "epoch": 14.16116611661166,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.007137856622901599,
      "loss": 0.2319,
      "num_input_tokens_seen": 27165344,
      "step": 128725
    },
    {
      "epoch": 14.161716171617162,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.007136630276842967,
      "loss": 0.2298,
      "num_input_tokens_seen": 27166368,
      "step": 128730
    },
    {
      "epoch": 14.162266226622663,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00713540400325724,
      "loss": 0.2293,
      "num_input_tokens_seen": 27167424,
      "step": 128735
    },
    {
      "epoch": 14.162816281628162,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.0071341778021557205,
      "loss": 0.2324,
      "num_input_tokens_seen": 27168544,
      "step": 128740
    },
    {
      "epoch": 14.163366336633663,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007132951673549715,
      "loss": 0.2329,
      "num_input_tokens_seen": 27169632,
      "step": 128745
    },
    {
      "epoch": 14.163916391639164,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.007131725617450521,
      "loss": 0.2283,
      "num_input_tokens_seen": 27170624,
      "step": 128750
    },
    {
      "epoch": 14.164466446644665,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007130499633869428,
      "loss": 0.2309,
      "num_input_tokens_seen": 27171648,
      "step": 128755
    },
    {
      "epoch": 14.165016501650165,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.007129273722817746,
      "loss": 0.2303,
      "num_input_tokens_seen": 27172640,
      "step": 128760
    },
    {
      "epoch": 14.165566556655666,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.007128047884306776,
      "loss": 0.2308,
      "num_input_tokens_seen": 27173728,
      "step": 128765
    },
    {
      "epoch": 14.166116611661167,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.007126822118347804,
      "loss": 0.2303,
      "num_input_tokens_seen": 27174720,
      "step": 128770
    },
    {
      "epoch": 14.166666666666666,
      "grad_norm": 0.0031280517578125,
      "learning_rate": 0.007125596424952143,
      "loss": 0.2304,
      "num_input_tokens_seen": 27175840,
      "step": 128775
    },
    {
      "epoch": 14.167216721672167,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.007124370804131075,
      "loss": 0.2303,
      "num_input_tokens_seen": 27176896,
      "step": 128780
    },
    {
      "epoch": 14.167766776677668,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007123145255895906,
      "loss": 0.2309,
      "num_input_tokens_seen": 27177888,
      "step": 128785
    },
    {
      "epoch": 14.168316831683168,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007121919780257924,
      "loss": 0.2303,
      "num_input_tokens_seen": 27178912,
      "step": 128790
    },
    {
      "epoch": 14.168866886688669,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.007120694377228426,
      "loss": 0.2314,
      "num_input_tokens_seen": 27179936,
      "step": 128795
    },
    {
      "epoch": 14.16941694169417,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.007119469046818714,
      "loss": 0.2309,
      "num_input_tokens_seen": 27181120,
      "step": 128800
    },
    {
      "epoch": 14.16996699669967,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0071182437890400675,
      "loss": 0.2319,
      "num_input_tokens_seen": 27182176,
      "step": 128805
    },
    {
      "epoch": 14.17051705170517,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007117018603903791,
      "loss": 0.2314,
      "num_input_tokens_seen": 27183200,
      "step": 128810
    },
    {
      "epoch": 14.171067106710671,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007115793491421166,
      "loss": 0.2324,
      "num_input_tokens_seen": 27184320,
      "step": 128815
    },
    {
      "epoch": 14.171617161716172,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.007114568451603491,
      "loss": 0.2308,
      "num_input_tokens_seen": 27185376,
      "step": 128820
    },
    {
      "epoch": 14.172167216721672,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.007113343484462058,
      "loss": 0.2324,
      "num_input_tokens_seen": 27186464,
      "step": 128825
    },
    {
      "epoch": 14.172717271727173,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00711211859000815,
      "loss": 0.2303,
      "num_input_tokens_seen": 27187520,
      "step": 128830
    },
    {
      "epoch": 14.173267326732674,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007110893768253065,
      "loss": 0.2319,
      "num_input_tokens_seen": 27188672,
      "step": 128835
    },
    {
      "epoch": 14.173817381738173,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007109669019208087,
      "loss": 0.2309,
      "num_input_tokens_seen": 27189696,
      "step": 128840
    },
    {
      "epoch": 14.174367436743674,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.007108444342884496,
      "loss": 0.2319,
      "num_input_tokens_seen": 27190784,
      "step": 128845
    },
    {
      "epoch": 14.174917491749175,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.007107219739293589,
      "loss": 0.2303,
      "num_input_tokens_seen": 27191744,
      "step": 128850
    },
    {
      "epoch": 14.175467546754675,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007105995208446649,
      "loss": 0.2304,
      "num_input_tokens_seen": 27192832,
      "step": 128855
    },
    {
      "epoch": 14.176017601760176,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.007104770750354969,
      "loss": 0.2303,
      "num_input_tokens_seen": 27193856,
      "step": 128860
    },
    {
      "epoch": 14.176567656765677,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007103546365029829,
      "loss": 0.2308,
      "num_input_tokens_seen": 27194912,
      "step": 128865
    },
    {
      "epoch": 14.177117711771178,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.007102322052482507,
      "loss": 0.2308,
      "num_input_tokens_seen": 27195936,
      "step": 128870
    },
    {
      "epoch": 14.177667766776677,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.007101097812724294,
      "loss": 0.2309,
      "num_input_tokens_seen": 27197088,
      "step": 128875
    },
    {
      "epoch": 14.178217821782178,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.007099873645766471,
      "loss": 0.2314,
      "num_input_tokens_seen": 27198176,
      "step": 128880
    },
    {
      "epoch": 14.17876787678768,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.007098649551620328,
      "loss": 0.2298,
      "num_input_tokens_seen": 27199264,
      "step": 128885
    },
    {
      "epoch": 14.179317931793179,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00709742553029714,
      "loss": 0.2308,
      "num_input_tokens_seen": 27200288,
      "step": 128890
    },
    {
      "epoch": 14.17986798679868,
      "grad_norm": 0.0031890869140625,
      "learning_rate": 0.007096201581808183,
      "loss": 0.2298,
      "num_input_tokens_seen": 27201344,
      "step": 128895
    },
    {
      "epoch": 14.180418041804181,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.007094977706164751,
      "loss": 0.2308,
      "num_input_tokens_seen": 27202368,
      "step": 128900
    },
    {
      "epoch": 14.18096809680968,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.007093753903378109,
      "loss": 0.233,
      "num_input_tokens_seen": 27203392,
      "step": 128905
    },
    {
      "epoch": 14.181518151815181,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007092530173459544,
      "loss": 0.2313,
      "num_input_tokens_seen": 27204416,
      "step": 128910
    },
    {
      "epoch": 14.182068206820682,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00709130651642034,
      "loss": 0.2298,
      "num_input_tokens_seen": 27205504,
      "step": 128915
    },
    {
      "epoch": 14.182618261826182,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.007090082932271762,
      "loss": 0.2293,
      "num_input_tokens_seen": 27206560,
      "step": 128920
    },
    {
      "epoch": 14.183168316831683,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.0070888594210251005,
      "loss": 0.2329,
      "num_input_tokens_seen": 27207648,
      "step": 128925
    },
    {
      "epoch": 14.183718371837184,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.007087635982691621,
      "loss": 0.2298,
      "num_input_tokens_seen": 27208704,
      "step": 128930
    },
    {
      "epoch": 14.184268426842685,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007086412617282602,
      "loss": 0.2304,
      "num_input_tokens_seen": 27209824,
      "step": 128935
    },
    {
      "epoch": 14.184818481848184,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007085189324809327,
      "loss": 0.2313,
      "num_input_tokens_seen": 27210880,
      "step": 128940
    },
    {
      "epoch": 14.185368536853685,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.007083966105283059,
      "loss": 0.2303,
      "num_input_tokens_seen": 27211968,
      "step": 128945
    },
    {
      "epoch": 14.185918591859187,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0070827429587150825,
      "loss": 0.2298,
      "num_input_tokens_seen": 27213056,
      "step": 128950
    },
    {
      "epoch": 14.186468646864686,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0070815198851166645,
      "loss": 0.2314,
      "num_input_tokens_seen": 27214112,
      "step": 128955
    },
    {
      "epoch": 14.187018701870187,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.007080296884499073,
      "loss": 0.2335,
      "num_input_tokens_seen": 27215104,
      "step": 128960
    },
    {
      "epoch": 14.187568756875688,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.007079073956873583,
      "loss": 0.2303,
      "num_input_tokens_seen": 27216160,
      "step": 128965
    },
    {
      "epoch": 14.188118811881187,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007077851102251469,
      "loss": 0.2309,
      "num_input_tokens_seen": 27217280,
      "step": 128970
    },
    {
      "epoch": 14.188668866886688,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.007076628320644005,
      "loss": 0.2308,
      "num_input_tokens_seen": 27218272,
      "step": 128975
    },
    {
      "epoch": 14.18921892189219,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.007075405612062457,
      "loss": 0.2308,
      "num_input_tokens_seen": 27219328,
      "step": 128980
    },
    {
      "epoch": 14.189768976897689,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.007074182976518087,
      "loss": 0.2298,
      "num_input_tokens_seen": 27220384,
      "step": 128985
    },
    {
      "epoch": 14.19031903190319,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0070729604140221695,
      "loss": 0.2324,
      "num_input_tokens_seen": 27221376,
      "step": 128990
    },
    {
      "epoch": 14.190869086908691,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00707173792458597,
      "loss": 0.2324,
      "num_input_tokens_seen": 27222368,
      "step": 128995
    },
    {
      "epoch": 14.191419141914192,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007070515508220766,
      "loss": 0.233,
      "num_input_tokens_seen": 27223360,
      "step": 129000
    },
    {
      "epoch": 14.191969196919691,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007069293164937814,
      "loss": 0.2309,
      "num_input_tokens_seen": 27224448,
      "step": 129005
    },
    {
      "epoch": 14.192519251925193,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.007068070894748376,
      "loss": 0.2319,
      "num_input_tokens_seen": 27225472,
      "step": 129010
    },
    {
      "epoch": 14.193069306930694,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0070668486976637294,
      "loss": 0.2351,
      "num_input_tokens_seen": 27226496,
      "step": 129015
    },
    {
      "epoch": 14.193619361936193,
      "grad_norm": 0.005859375,
      "learning_rate": 0.007065626573695125,
      "loss": 0.2335,
      "num_input_tokens_seen": 27227584,
      "step": 129020
    },
    {
      "epoch": 14.194169416941694,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.007064404522853832,
      "loss": 0.2309,
      "num_input_tokens_seen": 27228736,
      "step": 129025
    },
    {
      "epoch": 14.194719471947195,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.007063182545151122,
      "loss": 0.2324,
      "num_input_tokens_seen": 27229760,
      "step": 129030
    },
    {
      "epoch": 14.195269526952695,
      "grad_norm": 0.005859375,
      "learning_rate": 0.007061960640598244,
      "loss": 0.2313,
      "num_input_tokens_seen": 27230816,
      "step": 129035
    },
    {
      "epoch": 14.195819581958196,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.007060738809206471,
      "loss": 0.2319,
      "num_input_tokens_seen": 27231872,
      "step": 129040
    },
    {
      "epoch": 14.196369636963697,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.007059517050987053,
      "loss": 0.2293,
      "num_input_tokens_seen": 27232928,
      "step": 129045
    },
    {
      "epoch": 14.196919691969198,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.007058295365951255,
      "loss": 0.2298,
      "num_input_tokens_seen": 27234016,
      "step": 129050
    },
    {
      "epoch": 14.197469746974697,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.007057073754110346,
      "loss": 0.2335,
      "num_input_tokens_seen": 27235040,
      "step": 129055
    },
    {
      "epoch": 14.198019801980198,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.007055852215475571,
      "loss": 0.2309,
      "num_input_tokens_seen": 27236064,
      "step": 129060
    },
    {
      "epoch": 14.1985698569857,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.007054630750058198,
      "loss": 0.2324,
      "num_input_tokens_seen": 27237088,
      "step": 129065
    },
    {
      "epoch": 14.199119911991199,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007053409357869476,
      "loss": 0.2308,
      "num_input_tokens_seen": 27238112,
      "step": 129070
    },
    {
      "epoch": 14.1996699669967,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.007052188038920671,
      "loss": 0.2324,
      "num_input_tokens_seen": 27239104,
      "step": 129075
    },
    {
      "epoch": 14.2002200220022,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.007050966793223033,
      "loss": 0.2293,
      "num_input_tokens_seen": 27240128,
      "step": 129080
    },
    {
      "epoch": 14.2007700770077,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.007049745620787817,
      "loss": 0.2335,
      "num_input_tokens_seen": 27241184,
      "step": 129085
    },
    {
      "epoch": 14.201320132013201,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.007048524521626288,
      "loss": 0.2303,
      "num_input_tokens_seen": 27242208,
      "step": 129090
    },
    {
      "epoch": 14.201870187018702,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.007047303495749685,
      "loss": 0.2298,
      "num_input_tokens_seen": 27243264,
      "step": 129095
    },
    {
      "epoch": 14.202420242024202,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.007046082543169278,
      "loss": 0.2324,
      "num_input_tokens_seen": 27244352,
      "step": 129100
    },
    {
      "epoch": 14.202970297029703,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007044861663896305,
      "loss": 0.2308,
      "num_input_tokens_seen": 27245408,
      "step": 129105
    },
    {
      "epoch": 14.203520352035204,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.007043640857942023,
      "loss": 0.2324,
      "num_input_tokens_seen": 27246464,
      "step": 129110
    },
    {
      "epoch": 14.204070407040705,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.007042420125317692,
      "loss": 0.2329,
      "num_input_tokens_seen": 27247520,
      "step": 129115
    },
    {
      "epoch": 14.204620462046204,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.007041199466034552,
      "loss": 0.2314,
      "num_input_tokens_seen": 27248544,
      "step": 129120
    },
    {
      "epoch": 14.205170517051705,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.007039978880103862,
      "loss": 0.2298,
      "num_input_tokens_seen": 27249600,
      "step": 129125
    },
    {
      "epoch": 14.205720572057206,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.007038758367536867,
      "loss": 0.2314,
      "num_input_tokens_seen": 27250688,
      "step": 129130
    },
    {
      "epoch": 14.206270627062706,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.007037537928344811,
      "loss": 0.2319,
      "num_input_tokens_seen": 27251744,
      "step": 129135
    },
    {
      "epoch": 14.206820682068207,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.007036317562538946,
      "loss": 0.2303,
      "num_input_tokens_seen": 27252832,
      "step": 129140
    },
    {
      "epoch": 14.207370737073708,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.007035097270130521,
      "loss": 0.2314,
      "num_input_tokens_seen": 27253824,
      "step": 129145
    },
    {
      "epoch": 14.207920792079207,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.007033877051130789,
      "loss": 0.2309,
      "num_input_tokens_seen": 27254816,
      "step": 129150
    },
    {
      "epoch": 14.208470847084708,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.00703265690555099,
      "loss": 0.2303,
      "num_input_tokens_seen": 27255840,
      "step": 129155
    },
    {
      "epoch": 14.20902090209021,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.007031436833402362,
      "loss": 0.2319,
      "num_input_tokens_seen": 27256896,
      "step": 129160
    },
    {
      "epoch": 14.209570957095709,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.007030216834696158,
      "loss": 0.2308,
      "num_input_tokens_seen": 27257888,
      "step": 129165
    },
    {
      "epoch": 14.21012101210121,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0070289969094436255,
      "loss": 0.2309,
      "num_input_tokens_seen": 27258944,
      "step": 129170
    },
    {
      "epoch": 14.210671067106711,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.007027777057656,
      "loss": 0.2324,
      "num_input_tokens_seen": 27259904,
      "step": 129175
    },
    {
      "epoch": 14.211221122112212,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.007026557279344533,
      "loss": 0.2304,
      "num_input_tokens_seen": 27260992,
      "step": 129180
    },
    {
      "epoch": 14.211771177117711,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.007025337574520456,
      "loss": 0.2308,
      "num_input_tokens_seen": 27262080,
      "step": 129185
    },
    {
      "epoch": 14.212321232123212,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.007024117943195021,
      "loss": 0.2293,
      "num_input_tokens_seen": 27263136,
      "step": 129190
    },
    {
      "epoch": 14.212871287128714,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.007022898385379458,
      "loss": 0.2298,
      "num_input_tokens_seen": 27264160,
      "step": 129195
    },
    {
      "epoch": 14.213421342134213,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.007021678901085012,
      "loss": 0.2303,
      "num_input_tokens_seen": 27265248,
      "step": 129200
    },
    {
      "epoch": 14.213971397139714,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0070204594903229296,
      "loss": 0.2303,
      "num_input_tokens_seen": 27266272,
      "step": 129205
    },
    {
      "epoch": 14.214521452145215,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.007019240153104439,
      "loss": 0.2309,
      "num_input_tokens_seen": 27267360,
      "step": 129210
    },
    {
      "epoch": 14.215071507150714,
      "grad_norm": 0.000690460205078125,
      "learning_rate": 0.0070180208894407865,
      "loss": 0.2329,
      "num_input_tokens_seen": 27268352,
      "step": 129215
    },
    {
      "epoch": 14.215621562156215,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.007016801699343206,
      "loss": 0.234,
      "num_input_tokens_seen": 27269408,
      "step": 129220
    },
    {
      "epoch": 14.216171617161717,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.007015582582822921,
      "loss": 0.2319,
      "num_input_tokens_seen": 27270400,
      "step": 129225
    },
    {
      "epoch": 14.216721672167218,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.0070143635398911925,
      "loss": 0.2303,
      "num_input_tokens_seen": 27271456,
      "step": 129230
    },
    {
      "epoch": 14.217271727172717,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.007013144570559238,
      "loss": 0.2308,
      "num_input_tokens_seen": 27272512,
      "step": 129235
    },
    {
      "epoch": 14.217821782178218,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.007011925674838302,
      "loss": 0.2308,
      "num_input_tokens_seen": 27273600,
      "step": 129240
    },
    {
      "epoch": 14.218371837183719,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.007010706852739614,
      "loss": 0.233,
      "num_input_tokens_seen": 27274656,
      "step": 129245
    },
    {
      "epoch": 14.218921892189218,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.007009488104274403,
      "loss": 0.2298,
      "num_input_tokens_seen": 27275712,
      "step": 129250
    },
    {
      "epoch": 14.21947194719472,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.007008269429453905,
      "loss": 0.2329,
      "num_input_tokens_seen": 27276800,
      "step": 129255
    },
    {
      "epoch": 14.22002200220022,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.007007050828289353,
      "loss": 0.2309,
      "num_input_tokens_seen": 27277920,
      "step": 129260
    },
    {
      "epoch": 14.22057205720572,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.007005832300791984,
      "loss": 0.2304,
      "num_input_tokens_seen": 27279008,
      "step": 129265
    },
    {
      "epoch": 14.221122112211221,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.007004613846973022,
      "loss": 0.2314,
      "num_input_tokens_seen": 27280096,
      "step": 129270
    },
    {
      "epoch": 14.221672167216722,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.007003395466843692,
      "loss": 0.233,
      "num_input_tokens_seen": 27281184,
      "step": 129275
    },
    {
      "epoch": 14.222222222222221,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.007002177160415236,
      "loss": 0.2314,
      "num_input_tokens_seen": 27282208,
      "step": 129280
    },
    {
      "epoch": 14.222772277227723,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.007000958927698863,
      "loss": 0.2293,
      "num_input_tokens_seen": 27283296,
      "step": 129285
    },
    {
      "epoch": 14.223322332233224,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.006999740768705824,
      "loss": 0.2324,
      "num_input_tokens_seen": 27284384,
      "step": 129290
    },
    {
      "epoch": 14.223872387238725,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.006998522683447335,
      "loss": 0.2314,
      "num_input_tokens_seen": 27285376,
      "step": 129295
    },
    {
      "epoch": 14.224422442244224,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.006997304671934619,
      "loss": 0.2319,
      "num_input_tokens_seen": 27286464,
      "step": 129300
    },
    {
      "epoch": 14.224972497249725,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.006996086734178911,
      "loss": 0.2324,
      "num_input_tokens_seen": 27287552,
      "step": 129305
    },
    {
      "epoch": 14.225522552255226,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.006994868870191425,
      "loss": 0.2303,
      "num_input_tokens_seen": 27288608,
      "step": 129310
    },
    {
      "epoch": 14.226072607260726,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.006993651079983391,
      "loss": 0.2324,
      "num_input_tokens_seen": 27289664,
      "step": 129315
    },
    {
      "epoch": 14.226622662266227,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.00699243336356604,
      "loss": 0.2319,
      "num_input_tokens_seen": 27290752,
      "step": 129320
    },
    {
      "epoch": 14.227172717271728,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006991215720950582,
      "loss": 0.2298,
      "num_input_tokens_seen": 27291840,
      "step": 129325
    },
    {
      "epoch": 14.227722772277227,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.006989998152148251,
      "loss": 0.2324,
      "num_input_tokens_seen": 27292896,
      "step": 129330
    },
    {
      "epoch": 14.228272827282728,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.006988780657170258,
      "loss": 0.2314,
      "num_input_tokens_seen": 27293888,
      "step": 129335
    },
    {
      "epoch": 14.22882288228823,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.006987563236027836,
      "loss": 0.2319,
      "num_input_tokens_seen": 27295008,
      "step": 129340
    },
    {
      "epoch": 14.229372937293729,
      "grad_norm": 0.00081634521484375,
      "learning_rate": 0.006986345888732191,
      "loss": 0.2319,
      "num_input_tokens_seen": 27296000,
      "step": 129345
    },
    {
      "epoch": 14.22992299229923,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0069851286152945525,
      "loss": 0.2298,
      "num_input_tokens_seen": 27297024,
      "step": 129350
    },
    {
      "epoch": 14.23047304730473,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.006983911415726142,
      "loss": 0.2314,
      "num_input_tokens_seen": 27298080,
      "step": 129355
    },
    {
      "epoch": 14.231023102310232,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.006982694290038175,
      "loss": 0.2298,
      "num_input_tokens_seen": 27299104,
      "step": 129360
    },
    {
      "epoch": 14.231573157315731,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006981477238241862,
      "loss": 0.2303,
      "num_input_tokens_seen": 27300160,
      "step": 129365
    },
    {
      "epoch": 14.232123212321232,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.0069802602603484245,
      "loss": 0.2324,
      "num_input_tokens_seen": 27301248,
      "step": 129370
    },
    {
      "epoch": 14.232673267326733,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.006979043356369081,
      "loss": 0.2319,
      "num_input_tokens_seen": 27302272,
      "step": 129375
    },
    {
      "epoch": 14.233223322332233,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006977826526315051,
      "loss": 0.2314,
      "num_input_tokens_seen": 27303328,
      "step": 129380
    },
    {
      "epoch": 14.233773377337734,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.006976609770197544,
      "loss": 0.2309,
      "num_input_tokens_seen": 27304384,
      "step": 129385
    },
    {
      "epoch": 14.234323432343235,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.006975393088027771,
      "loss": 0.2314,
      "num_input_tokens_seen": 27305440,
      "step": 129390
    },
    {
      "epoch": 14.234873487348734,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.006974176479816953,
      "loss": 0.2304,
      "num_input_tokens_seen": 27306464,
      "step": 129395
    },
    {
      "epoch": 14.235423542354235,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.006972959945576293,
      "loss": 0.2303,
      "num_input_tokens_seen": 27307456,
      "step": 129400
    },
    {
      "epoch": 14.235973597359736,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.006971743485317011,
      "loss": 0.2329,
      "num_input_tokens_seen": 27308544,
      "step": 129405
    },
    {
      "epoch": 14.236523652365236,
      "grad_norm": 0.000598907470703125,
      "learning_rate": 0.006970527099050323,
      "loss": 0.2308,
      "num_input_tokens_seen": 27309568,
      "step": 129410
    },
    {
      "epoch": 14.237073707370737,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006969310786787427,
      "loss": 0.2308,
      "num_input_tokens_seen": 27310592,
      "step": 129415
    },
    {
      "epoch": 14.237623762376238,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006968094548539546,
      "loss": 0.2309,
      "num_input_tokens_seen": 27311616,
      "step": 129420
    },
    {
      "epoch": 14.238173817381739,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.006966878384317879,
      "loss": 0.2319,
      "num_input_tokens_seen": 27312736,
      "step": 129425
    },
    {
      "epoch": 14.238723872387238,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.006965662294133638,
      "loss": 0.2319,
      "num_input_tokens_seen": 27313824,
      "step": 129430
    },
    {
      "epoch": 14.23927392739274,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.006964446277998039,
      "loss": 0.2314,
      "num_input_tokens_seen": 27314880,
      "step": 129435
    },
    {
      "epoch": 14.23982398239824,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.006963230335922274,
      "loss": 0.2308,
      "num_input_tokens_seen": 27315936,
      "step": 129440
    },
    {
      "epoch": 14.24037403740374,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.006962014467917568,
      "loss": 0.2314,
      "num_input_tokens_seen": 27316928,
      "step": 129445
    },
    {
      "epoch": 14.24092409240924,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00696079867399511,
      "loss": 0.2288,
      "num_input_tokens_seen": 27317952,
      "step": 129450
    },
    {
      "epoch": 14.241474147414742,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.006959582954166119,
      "loss": 0.2293,
      "num_input_tokens_seen": 27319008,
      "step": 129455
    },
    {
      "epoch": 14.242024202420241,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00695836730844179,
      "loss": 0.2308,
      "num_input_tokens_seen": 27319968,
      "step": 129460
    },
    {
      "epoch": 14.242574257425742,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.006957151736833328,
      "loss": 0.2324,
      "num_input_tokens_seen": 27321120,
      "step": 129465
    },
    {
      "epoch": 14.243124312431243,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0069559362393519465,
      "loss": 0.2308,
      "num_input_tokens_seen": 27322144,
      "step": 129470
    },
    {
      "epoch": 14.243674367436745,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0069547208160088345,
      "loss": 0.2319,
      "num_input_tokens_seen": 27323200,
      "step": 129475
    },
    {
      "epoch": 14.244224422442244,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.006953505466815206,
      "loss": 0.2298,
      "num_input_tokens_seen": 27324192,
      "step": 129480
    },
    {
      "epoch": 14.244774477447745,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.0069522901917822505,
      "loss": 0.2298,
      "num_input_tokens_seen": 27325280,
      "step": 129485
    },
    {
      "epoch": 14.245324532453246,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.006951074990921175,
      "loss": 0.2308,
      "num_input_tokens_seen": 27326432,
      "step": 129490
    },
    {
      "epoch": 14.245874587458745,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006949859864243185,
      "loss": 0.2319,
      "num_input_tokens_seen": 27327424,
      "step": 129495
    },
    {
      "epoch": 14.246424642464246,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.006948644811759467,
      "loss": 0.2324,
      "num_input_tokens_seen": 27328512,
      "step": 129500
    },
    {
      "epoch": 14.246974697469748,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0069474298334812334,
      "loss": 0.2298,
      "num_input_tokens_seen": 27329568,
      "step": 129505
    },
    {
      "epoch": 14.247524752475247,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.006946214929419674,
      "loss": 0.2319,
      "num_input_tokens_seen": 27330624,
      "step": 129510
    },
    {
      "epoch": 14.248074807480748,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0069450000995859825,
      "loss": 0.233,
      "num_input_tokens_seen": 27331712,
      "step": 129515
    },
    {
      "epoch": 14.248624862486249,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.006943785343991358,
      "loss": 0.2324,
      "num_input_tokens_seen": 27332768,
      "step": 129520
    },
    {
      "epoch": 14.249174917491748,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.006942570662647,
      "loss": 0.2293,
      "num_input_tokens_seen": 27333856,
      "step": 129525
    },
    {
      "epoch": 14.24972497249725,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.006941356055564107,
      "loss": 0.2303,
      "num_input_tokens_seen": 27334912,
      "step": 129530
    },
    {
      "epoch": 14.25027502750275,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006940141522753869,
      "loss": 0.2308,
      "num_input_tokens_seen": 27335936,
      "step": 129535
    },
    {
      "epoch": 14.250825082508252,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.0069389270642274715,
      "loss": 0.2324,
      "num_input_tokens_seen": 27336992,
      "step": 129540
    },
    {
      "epoch": 14.251375137513751,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.006937712679996118,
      "loss": 0.234,
      "num_input_tokens_seen": 27338048,
      "step": 129545
    },
    {
      "epoch": 14.251925192519252,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.006936498370071002,
      "loss": 0.2319,
      "num_input_tokens_seen": 27339040,
      "step": 129550
    },
    {
      "epoch": 14.252475247524753,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.006935284134463305,
      "loss": 0.2293,
      "num_input_tokens_seen": 27340128,
      "step": 129555
    },
    {
      "epoch": 14.253025302530252,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00693406997318423,
      "loss": 0.2293,
      "num_input_tokens_seen": 27341120,
      "step": 129560
    },
    {
      "epoch": 14.253575357535754,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.0069328558862449575,
      "loss": 0.2314,
      "num_input_tokens_seen": 27342144,
      "step": 129565
    },
    {
      "epoch": 14.254125412541255,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006931641873656685,
      "loss": 0.2319,
      "num_input_tokens_seen": 27343200,
      "step": 129570
    },
    {
      "epoch": 14.254675467546754,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.006930427935430594,
      "loss": 0.2293,
      "num_input_tokens_seen": 27344224,
      "step": 129575
    },
    {
      "epoch": 14.255225522552255,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006929214071577875,
      "loss": 0.2319,
      "num_input_tokens_seen": 27345280,
      "step": 129580
    },
    {
      "epoch": 14.255775577557756,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.0069280002821097255,
      "loss": 0.2319,
      "num_input_tokens_seen": 27346432,
      "step": 129585
    },
    {
      "epoch": 14.256325632563255,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.006926786567037316,
      "loss": 0.2303,
      "num_input_tokens_seen": 27347456,
      "step": 129590
    },
    {
      "epoch": 14.256875687568757,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.006925572926371846,
      "loss": 0.2324,
      "num_input_tokens_seen": 27348640,
      "step": 129595
    },
    {
      "epoch": 14.257425742574258,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0069243593601244924,
      "loss": 0.2319,
      "num_input_tokens_seen": 27349728,
      "step": 129600
    },
    {
      "epoch": 14.257975797579759,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.006923145868306442,
      "loss": 0.2309,
      "num_input_tokens_seen": 27350752,
      "step": 129605
    },
    {
      "epoch": 14.258525852585258,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.006921932450928887,
      "loss": 0.233,
      "num_input_tokens_seen": 27351840,
      "step": 129610
    },
    {
      "epoch": 14.25907590759076,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.0069207191080029986,
      "loss": 0.2324,
      "num_input_tokens_seen": 27352896,
      "step": 129615
    },
    {
      "epoch": 14.25962596259626,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00691950583953997,
      "loss": 0.2324,
      "num_input_tokens_seen": 27353984,
      "step": 129620
    },
    {
      "epoch": 14.26017601760176,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.006918292645550979,
      "loss": 0.2319,
      "num_input_tokens_seen": 27355040,
      "step": 129625
    },
    {
      "epoch": 14.26072607260726,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.006917079526047202,
      "loss": 0.2319,
      "num_input_tokens_seen": 27356064,
      "step": 129630
    },
    {
      "epoch": 14.261276127612762,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.006915866481039823,
      "loss": 0.2329,
      "num_input_tokens_seen": 27357184,
      "step": 129635
    },
    {
      "epoch": 14.261826182618261,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.006914653510540024,
      "loss": 0.2303,
      "num_input_tokens_seen": 27358240,
      "step": 129640
    },
    {
      "epoch": 14.262376237623762,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0069134406145589895,
      "loss": 0.2314,
      "num_input_tokens_seen": 27359328,
      "step": 129645
    },
    {
      "epoch": 14.262926292629263,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.006912227793107894,
      "loss": 0.2314,
      "num_input_tokens_seen": 27360384,
      "step": 129650
    },
    {
      "epoch": 14.263476347634764,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006911015046197907,
      "loss": 0.2309,
      "num_input_tokens_seen": 27361472,
      "step": 129655
    },
    {
      "epoch": 14.264026402640264,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.006909802373840213,
      "loss": 0.2303,
      "num_input_tokens_seen": 27362496,
      "step": 129660
    },
    {
      "epoch": 14.264576457645765,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0069085897760459885,
      "loss": 0.2329,
      "num_input_tokens_seen": 27363520,
      "step": 129665
    },
    {
      "epoch": 14.265126512651266,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0069073772528264145,
      "loss": 0.2324,
      "num_input_tokens_seen": 27364608,
      "step": 129670
    },
    {
      "epoch": 14.265676567656765,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.006906164804192663,
      "loss": 0.2324,
      "num_input_tokens_seen": 27365600,
      "step": 129675
    },
    {
      "epoch": 14.266226622662266,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.0069049524301559,
      "loss": 0.2293,
      "num_input_tokens_seen": 27366656,
      "step": 129680
    },
    {
      "epoch": 14.266776677667767,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006903740130727311,
      "loss": 0.2319,
      "num_input_tokens_seen": 27367680,
      "step": 129685
    },
    {
      "epoch": 14.267326732673267,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.006902527905918059,
      "loss": 0.2319,
      "num_input_tokens_seen": 27368800,
      "step": 129690
    },
    {
      "epoch": 14.267876787678768,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.006901315755739321,
      "loss": 0.2309,
      "num_input_tokens_seen": 27369888,
      "step": 129695
    },
    {
      "epoch": 14.268426842684269,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.006900103680202276,
      "loss": 0.2314,
      "num_input_tokens_seen": 27370944,
      "step": 129700
    },
    {
      "epoch": 14.268976897689768,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.006898891679318083,
      "loss": 0.2283,
      "num_input_tokens_seen": 27371936,
      "step": 129705
    },
    {
      "epoch": 14.26952695269527,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.006897679753097925,
      "loss": 0.2319,
      "num_input_tokens_seen": 27372992,
      "step": 129710
    },
    {
      "epoch": 14.27007700770077,
      "grad_norm": 0.001953125,
      "learning_rate": 0.006896467901552958,
      "loss": 0.2314,
      "num_input_tokens_seen": 27373984,
      "step": 129715
    },
    {
      "epoch": 14.270627062706271,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.006895256124694358,
      "loss": 0.2303,
      "num_input_tokens_seen": 27375040,
      "step": 129720
    },
    {
      "epoch": 14.27117711771177,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0068940444225332985,
      "loss": 0.2319,
      "num_input_tokens_seen": 27376032,
      "step": 129725
    },
    {
      "epoch": 14.271727172717272,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.006892832795080937,
      "loss": 0.2345,
      "num_input_tokens_seen": 27377152,
      "step": 129730
    },
    {
      "epoch": 14.272277227722773,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006891621242348451,
      "loss": 0.2319,
      "num_input_tokens_seen": 27378176,
      "step": 129735
    },
    {
      "epoch": 14.272827282728272,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006890409764347001,
      "loss": 0.2303,
      "num_input_tokens_seen": 27379232,
      "step": 129740
    },
    {
      "epoch": 14.273377337733773,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.006889198361087746,
      "loss": 0.2324,
      "num_input_tokens_seen": 27380288,
      "step": 129745
    },
    {
      "epoch": 14.273927392739274,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00688798703258186,
      "loss": 0.2319,
      "num_input_tokens_seen": 27381344,
      "step": 129750
    },
    {
      "epoch": 14.274477447744774,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.006886775778840503,
      "loss": 0.2319,
      "num_input_tokens_seen": 27382400,
      "step": 129755
    },
    {
      "epoch": 14.275027502750275,
      "grad_norm": 0.000614166259765625,
      "learning_rate": 0.006885564599874847,
      "loss": 0.2314,
      "num_input_tokens_seen": 27383424,
      "step": 129760
    },
    {
      "epoch": 14.275577557755776,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0068843534956960495,
      "loss": 0.2308,
      "num_input_tokens_seen": 27384448,
      "step": 129765
    },
    {
      "epoch": 14.276127612761275,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006883142466315264,
      "loss": 0.2309,
      "num_input_tokens_seen": 27385440,
      "step": 129770
    },
    {
      "epoch": 14.276677667766776,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.00688193151174366,
      "loss": 0.2324,
      "num_input_tokens_seen": 27386560,
      "step": 129775
    },
    {
      "epoch": 14.277227722772277,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.006880720631992399,
      "loss": 0.2293,
      "num_input_tokens_seen": 27387616,
      "step": 129780
    },
    {
      "epoch": 14.277777777777779,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.006879509827072645,
      "loss": 0.2324,
      "num_input_tokens_seen": 27388704,
      "step": 129785
    },
    {
      "epoch": 14.278327832783278,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.006878299096995552,
      "loss": 0.2319,
      "num_input_tokens_seen": 27389792,
      "step": 129790
    },
    {
      "epoch": 14.278877887788779,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.006877088441772274,
      "loss": 0.2324,
      "num_input_tokens_seen": 27390880,
      "step": 129795
    },
    {
      "epoch": 14.27942794279428,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.006875877861413978,
      "loss": 0.2314,
      "num_input_tokens_seen": 27391936,
      "step": 129800
    },
    {
      "epoch": 14.27997799779978,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0068746673559318145,
      "loss": 0.2309,
      "num_input_tokens_seen": 27393024,
      "step": 129805
    },
    {
      "epoch": 14.28052805280528,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006873456925336941,
      "loss": 0.2309,
      "num_input_tokens_seen": 27394080,
      "step": 129810
    },
    {
      "epoch": 14.281078107810782,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0068722465696405225,
      "loss": 0.2304,
      "num_input_tokens_seen": 27395072,
      "step": 129815
    },
    {
      "epoch": 14.281628162816281,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.006871036288853701,
      "loss": 0.2324,
      "num_input_tokens_seen": 27396128,
      "step": 129820
    },
    {
      "epoch": 14.282178217821782,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.006869826082987643,
      "loss": 0.2319,
      "num_input_tokens_seen": 27397184,
      "step": 129825
    },
    {
      "epoch": 14.282728272827283,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.00686861595205349,
      "loss": 0.2324,
      "num_input_tokens_seen": 27398240,
      "step": 129830
    },
    {
      "epoch": 14.283278327832782,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.006867405896062404,
      "loss": 0.2303,
      "num_input_tokens_seen": 27399296,
      "step": 129835
    },
    {
      "epoch": 14.283828382838283,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006866195915025539,
      "loss": 0.2309,
      "num_input_tokens_seen": 27400480,
      "step": 129840
    },
    {
      "epoch": 14.284378437843785,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.00686498600895404,
      "loss": 0.2278,
      "num_input_tokens_seen": 27401536,
      "step": 129845
    },
    {
      "epoch": 14.284928492849286,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.006863776177859065,
      "loss": 0.2329,
      "num_input_tokens_seen": 27402624,
      "step": 129850
    },
    {
      "epoch": 14.285478547854785,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.006862566421751757,
      "loss": 0.2309,
      "num_input_tokens_seen": 27403712,
      "step": 129855
    },
    {
      "epoch": 14.286028602860286,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.006861356740643274,
      "loss": 0.2293,
      "num_input_tokens_seen": 27404832,
      "step": 129860
    },
    {
      "epoch": 14.286578657865787,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.006860147134544755,
      "loss": 0.2324,
      "num_input_tokens_seen": 27405888,
      "step": 129865
    },
    {
      "epoch": 14.287128712871286,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.006858937603467355,
      "loss": 0.2319,
      "num_input_tokens_seen": 27406944,
      "step": 129870
    },
    {
      "epoch": 14.287678767876788,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.006857728147422225,
      "loss": 0.2314,
      "num_input_tokens_seen": 27407968,
      "step": 129875
    },
    {
      "epoch": 14.288228822882289,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0068565187664205015,
      "loss": 0.2319,
      "num_input_tokens_seen": 27409024,
      "step": 129880
    },
    {
      "epoch": 14.288778877887788,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006855309460473344,
      "loss": 0.2314,
      "num_input_tokens_seen": 27410144,
      "step": 129885
    },
    {
      "epoch": 14.289328932893289,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.006854100229591891,
      "loss": 0.2314,
      "num_input_tokens_seen": 27411200,
      "step": 129890
    },
    {
      "epoch": 14.28987898789879,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.006852891073787275,
      "loss": 0.2324,
      "num_input_tokens_seen": 27412256,
      "step": 129895
    },
    {
      "epoch": 14.290429042904291,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.006851681993070665,
      "loss": 0.2303,
      "num_input_tokens_seen": 27413248,
      "step": 129900
    },
    {
      "epoch": 14.29097909790979,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.006850472987453185,
      "loss": 0.2293,
      "num_input_tokens_seen": 27414304,
      "step": 129905
    },
    {
      "epoch": 14.291529152915292,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.006849264056945992,
      "loss": 0.2314,
      "num_input_tokens_seen": 27415392,
      "step": 129910
    },
    {
      "epoch": 14.292079207920793,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006848055201560221,
      "loss": 0.2314,
      "num_input_tokens_seen": 27416416,
      "step": 129915
    },
    {
      "epoch": 14.292629262926292,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.006846846421307007,
      "loss": 0.2304,
      "num_input_tokens_seen": 27417504,
      "step": 129920
    },
    {
      "epoch": 14.293179317931793,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.006845637716197498,
      "loss": 0.2303,
      "num_input_tokens_seen": 27418528,
      "step": 129925
    },
    {
      "epoch": 14.293729372937294,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.006844429086242839,
      "loss": 0.2324,
      "num_input_tokens_seen": 27419648,
      "step": 129930
    },
    {
      "epoch": 14.294279427942794,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.006843220531454157,
      "loss": 0.2313,
      "num_input_tokens_seen": 27420704,
      "step": 129935
    },
    {
      "epoch": 14.294829482948295,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.006842012051842605,
      "loss": 0.2298,
      "num_input_tokens_seen": 27421824,
      "step": 129940
    },
    {
      "epoch": 14.295379537953796,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.006840803647419306,
      "loss": 0.2314,
      "num_input_tokens_seen": 27422976,
      "step": 129945
    },
    {
      "epoch": 14.295929592959295,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.006839595318195406,
      "loss": 0.2308,
      "num_input_tokens_seen": 27424000,
      "step": 129950
    },
    {
      "epoch": 14.296479647964796,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.006838387064182044,
      "loss": 0.2319,
      "num_input_tokens_seen": 27425024,
      "step": 129955
    },
    {
      "epoch": 14.297029702970297,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.006837178885390348,
      "loss": 0.2314,
      "num_input_tokens_seen": 27426080,
      "step": 129960
    },
    {
      "epoch": 14.297579757975798,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006835970781831462,
      "loss": 0.2335,
      "num_input_tokens_seen": 27427200,
      "step": 129965
    },
    {
      "epoch": 14.298129812981298,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006834762753516511,
      "loss": 0.2319,
      "num_input_tokens_seen": 27428288,
      "step": 129970
    },
    {
      "epoch": 14.298679867986799,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.006833554800456639,
      "loss": 0.2329,
      "num_input_tokens_seen": 27429376,
      "step": 129975
    },
    {
      "epoch": 14.2992299229923,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.006832346922662967,
      "loss": 0.2314,
      "num_input_tokens_seen": 27430464,
      "step": 129980
    },
    {
      "epoch": 14.2997799779978,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0068311391201466365,
      "loss": 0.2309,
      "num_input_tokens_seen": 27431584,
      "step": 129985
    },
    {
      "epoch": 14.3003300330033,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.006829931392918781,
      "loss": 0.2304,
      "num_input_tokens_seen": 27432608,
      "step": 129990
    },
    {
      "epoch": 14.300880088008801,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.006828723740990522,
      "loss": 0.2319,
      "num_input_tokens_seen": 27433632,
      "step": 129995
    },
    {
      "epoch": 14.3014301430143,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.006827516164373001,
      "loss": 0.2319,
      "num_input_tokens_seen": 27434656,
      "step": 130000
    },
    {
      "epoch": 14.301980198019802,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006826308663077342,
      "loss": 0.2308,
      "num_input_tokens_seen": 27435712,
      "step": 130005
    },
    {
      "epoch": 14.302530253025303,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.006825101237114662,
      "loss": 0.2319,
      "num_input_tokens_seen": 27436704,
      "step": 130010
    },
    {
      "epoch": 14.303080308030804,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.006823893886496114,
      "loss": 0.2319,
      "num_input_tokens_seen": 27437728,
      "step": 130015
    },
    {
      "epoch": 14.303630363036303,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006822686611232806,
      "loss": 0.2309,
      "num_input_tokens_seen": 27438784,
      "step": 130020
    },
    {
      "epoch": 14.304180418041804,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006821479411335878,
      "loss": 0.2319,
      "num_input_tokens_seen": 27439840,
      "step": 130025
    },
    {
      "epoch": 14.304730473047305,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.006820272286816449,
      "loss": 0.2319,
      "num_input_tokens_seen": 27440864,
      "step": 130030
    },
    {
      "epoch": 14.305280528052805,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.00681906523768564,
      "loss": 0.2335,
      "num_input_tokens_seen": 27441888,
      "step": 130035
    },
    {
      "epoch": 14.305830583058306,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.006817858263954581,
      "loss": 0.2319,
      "num_input_tokens_seen": 27442944,
      "step": 130040
    },
    {
      "epoch": 14.306380638063807,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006816651365634397,
      "loss": 0.2314,
      "num_input_tokens_seen": 27444000,
      "step": 130045
    },
    {
      "epoch": 14.306930693069306,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.006815444542736216,
      "loss": 0.2319,
      "num_input_tokens_seen": 27445056,
      "step": 130050
    },
    {
      "epoch": 14.307480748074807,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.006814237795271154,
      "loss": 0.2309,
      "num_input_tokens_seen": 27446176,
      "step": 130055
    },
    {
      "epoch": 14.308030803080309,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.006813031123250331,
      "loss": 0.2324,
      "num_input_tokens_seen": 27447232,
      "step": 130060
    },
    {
      "epoch": 14.308580858085808,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.006811824526684876,
      "loss": 0.2314,
      "num_input_tokens_seen": 27448256,
      "step": 130065
    },
    {
      "epoch": 14.309130913091309,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.006810618005585894,
      "loss": 0.2324,
      "num_input_tokens_seen": 27449376,
      "step": 130070
    },
    {
      "epoch": 14.30968096809681,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.006809411559964526,
      "loss": 0.2324,
      "num_input_tokens_seen": 27450496,
      "step": 130075
    },
    {
      "epoch": 14.310231023102311,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006808205189831883,
      "loss": 0.2324,
      "num_input_tokens_seen": 27451488,
      "step": 130080
    },
    {
      "epoch": 14.31078107810781,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.006806998895199077,
      "loss": 0.2314,
      "num_input_tokens_seen": 27452512,
      "step": 130085
    },
    {
      "epoch": 14.311331133113312,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.006805792676077235,
      "loss": 0.2309,
      "num_input_tokens_seen": 27453600,
      "step": 130090
    },
    {
      "epoch": 14.311881188118813,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.0068045865324774645,
      "loss": 0.2309,
      "num_input_tokens_seen": 27454688,
      "step": 130095
    },
    {
      "epoch": 14.312431243124312,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.006803380464410888,
      "loss": 0.2298,
      "num_input_tokens_seen": 27455744,
      "step": 130100
    },
    {
      "epoch": 14.312981298129813,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0068021744718886265,
      "loss": 0.2309,
      "num_input_tokens_seen": 27456800,
      "step": 130105
    },
    {
      "epoch": 14.313531353135314,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006800968554921782,
      "loss": 0.234,
      "num_input_tokens_seen": 27457760,
      "step": 130110
    },
    {
      "epoch": 14.314081408140813,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.006799762713521484,
      "loss": 0.2293,
      "num_input_tokens_seen": 27458752,
      "step": 130115
    },
    {
      "epoch": 14.314631463146315,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.006798556947698837,
      "loss": 0.2309,
      "num_input_tokens_seen": 27459808,
      "step": 130120
    },
    {
      "epoch": 14.315181518151816,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.006797351257464949,
      "loss": 0.2319,
      "num_input_tokens_seen": 27460864,
      "step": 130125
    },
    {
      "epoch": 14.315731573157315,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0067961456428309395,
      "loss": 0.2314,
      "num_input_tokens_seen": 27461952,
      "step": 130130
    },
    {
      "epoch": 14.316281628162816,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.006794940103807918,
      "loss": 0.2314,
      "num_input_tokens_seen": 27462976,
      "step": 130135
    },
    {
      "epoch": 14.316831683168317,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0067937346404070035,
      "loss": 0.2309,
      "num_input_tokens_seen": 27464032,
      "step": 130140
    },
    {
      "epoch": 14.317381738173818,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006792529252639297,
      "loss": 0.2319,
      "num_input_tokens_seen": 27465088,
      "step": 130145
    },
    {
      "epoch": 14.317931793179318,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.006791323940515907,
      "loss": 0.2293,
      "num_input_tokens_seen": 27466176,
      "step": 130150
    },
    {
      "epoch": 14.318481848184819,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.006790118704047943,
      "loss": 0.2324,
      "num_input_tokens_seen": 27467264,
      "step": 130155
    },
    {
      "epoch": 14.31903190319032,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.006788913543246517,
      "loss": 0.2324,
      "num_input_tokens_seen": 27468288,
      "step": 130160
    },
    {
      "epoch": 14.319581958195819,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0067877084581227385,
      "loss": 0.2308,
      "num_input_tokens_seen": 27469280,
      "step": 130165
    },
    {
      "epoch": 14.32013201320132,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.006786503448687712,
      "loss": 0.2298,
      "num_input_tokens_seen": 27470368,
      "step": 130170
    },
    {
      "epoch": 14.320682068206821,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.006785298514952536,
      "loss": 0.2314,
      "num_input_tokens_seen": 27471488,
      "step": 130175
    },
    {
      "epoch": 14.32123212321232,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.006784093656928328,
      "loss": 0.2304,
      "num_input_tokens_seen": 27472608,
      "step": 130180
    },
    {
      "epoch": 14.321782178217822,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00678288887462618,
      "loss": 0.2319,
      "num_input_tokens_seen": 27473664,
      "step": 130185
    },
    {
      "epoch": 14.322332233223323,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0067816841680572015,
      "loss": 0.2293,
      "num_input_tokens_seen": 27474752,
      "step": 130190
    },
    {
      "epoch": 14.322882288228822,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.006780479537232504,
      "loss": 0.2303,
      "num_input_tokens_seen": 27475840,
      "step": 130195
    },
    {
      "epoch": 14.323432343234323,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0067792749821631755,
      "loss": 0.2335,
      "num_input_tokens_seen": 27476896,
      "step": 130200
    },
    {
      "epoch": 14.323982398239824,
      "grad_norm": 0.00067901611328125,
      "learning_rate": 0.006778070502860332,
      "loss": 0.2324,
      "num_input_tokens_seen": 27478016,
      "step": 130205
    },
    {
      "epoch": 14.324532453245325,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.0067768660993350605,
      "loss": 0.2298,
      "num_input_tokens_seen": 27479104,
      "step": 130210
    },
    {
      "epoch": 14.325082508250825,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006775661771598468,
      "loss": 0.2319,
      "num_input_tokens_seen": 27480192,
      "step": 130215
    },
    {
      "epoch": 14.325632563256326,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006774457519661661,
      "loss": 0.2335,
      "num_input_tokens_seen": 27481216,
      "step": 130220
    },
    {
      "epoch": 14.326182618261827,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006773253343535727,
      "loss": 0.2309,
      "num_input_tokens_seen": 27482368,
      "step": 130225
    },
    {
      "epoch": 14.326732673267326,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.006772049243231774,
      "loss": 0.2314,
      "num_input_tokens_seen": 27483424,
      "step": 130230
    },
    {
      "epoch": 14.327282728272827,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.00677084521876089,
      "loss": 0.2309,
      "num_input_tokens_seen": 27484480,
      "step": 130235
    },
    {
      "epoch": 14.327832783278328,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006769641270134181,
      "loss": 0.2324,
      "num_input_tokens_seen": 27485536,
      "step": 130240
    },
    {
      "epoch": 14.328382838283828,
      "grad_norm": 0.00057220458984375,
      "learning_rate": 0.006768437397362734,
      "loss": 0.2303,
      "num_input_tokens_seen": 27486624,
      "step": 130245
    },
    {
      "epoch": 14.328932893289329,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006767233600457652,
      "loss": 0.2308,
      "num_input_tokens_seen": 27487648,
      "step": 130250
    },
    {
      "epoch": 14.32948294829483,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.006766029879430031,
      "loss": 0.2308,
      "num_input_tokens_seen": 27488704,
      "step": 130255
    },
    {
      "epoch": 14.33003300330033,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006764826234290955,
      "loss": 0.233,
      "num_input_tokens_seen": 27489792,
      "step": 130260
    },
    {
      "epoch": 14.33058305830583,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.006763622665051531,
      "loss": 0.2324,
      "num_input_tokens_seen": 27490880,
      "step": 130265
    },
    {
      "epoch": 14.331133113311331,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.006762419171722838,
      "loss": 0.2298,
      "num_input_tokens_seen": 27491968,
      "step": 130270
    },
    {
      "epoch": 14.331683168316832,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006761215754315976,
      "loss": 0.233,
      "num_input_tokens_seen": 27493056,
      "step": 130275
    },
    {
      "epoch": 14.332233223322332,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.00676001241284204,
      "loss": 0.2324,
      "num_input_tokens_seen": 27494144,
      "step": 130280
    },
    {
      "epoch": 14.332783278327833,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.0067588091473121115,
      "loss": 0.2324,
      "num_input_tokens_seen": 27495232,
      "step": 130285
    },
    {
      "epoch": 14.333333333333334,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.006757605957737288,
      "loss": 0.2303,
      "num_input_tokens_seen": 27496320,
      "step": 130290
    },
    {
      "epoch": 14.333883388338833,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006756402844128655,
      "loss": 0.2314,
      "num_input_tokens_seen": 27497312,
      "step": 130295
    },
    {
      "epoch": 14.334433443344334,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.006755199806497298,
      "loss": 0.2308,
      "num_input_tokens_seen": 27498400,
      "step": 130300
    },
    {
      "epoch": 14.334983498349835,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.006753996844854307,
      "loss": 0.2293,
      "num_input_tokens_seen": 27499488,
      "step": 130305
    },
    {
      "epoch": 14.335533553355335,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.006752793959210775,
      "loss": 0.2303,
      "num_input_tokens_seen": 27500512,
      "step": 130310
    },
    {
      "epoch": 14.336083608360836,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.006751591149577779,
      "loss": 0.2313,
      "num_input_tokens_seen": 27501536,
      "step": 130315
    },
    {
      "epoch": 14.336633663366337,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006750388415966416,
      "loss": 0.2324,
      "num_input_tokens_seen": 27502624,
      "step": 130320
    },
    {
      "epoch": 14.337183718371838,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.0067491857583877575,
      "loss": 0.2319,
      "num_input_tokens_seen": 27503616,
      "step": 130325
    },
    {
      "epoch": 14.337733773377337,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.006747983176852896,
      "loss": 0.2329,
      "num_input_tokens_seen": 27504608,
      "step": 130330
    },
    {
      "epoch": 14.338283828382838,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006746780671372917,
      "loss": 0.2319,
      "num_input_tokens_seen": 27505664,
      "step": 130335
    },
    {
      "epoch": 14.33883388338834,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006745578241958897,
      "loss": 0.2324,
      "num_input_tokens_seen": 27506752,
      "step": 130340
    },
    {
      "epoch": 14.339383938393839,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.006744375888621926,
      "loss": 0.2309,
      "num_input_tokens_seen": 27507840,
      "step": 130345
    },
    {
      "epoch": 14.33993399339934,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006743173611373076,
      "loss": 0.2329,
      "num_input_tokens_seen": 27508864,
      "step": 130350
    },
    {
      "epoch": 14.340484048404841,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.006741971410223439,
      "loss": 0.2314,
      "num_input_tokens_seen": 27509952,
      "step": 130355
    },
    {
      "epoch": 14.34103410341034,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006740769285184082,
      "loss": 0.2303,
      "num_input_tokens_seen": 27511040,
      "step": 130360
    },
    {
      "epoch": 14.341584158415841,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.006739567236266092,
      "loss": 0.2298,
      "num_input_tokens_seen": 27512128,
      "step": 130365
    },
    {
      "epoch": 14.342134213421343,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.006738365263480552,
      "loss": 0.2303,
      "num_input_tokens_seen": 27513184,
      "step": 130370
    },
    {
      "epoch": 14.342684268426842,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.006737163366838529,
      "loss": 0.2314,
      "num_input_tokens_seen": 27514208,
      "step": 130375
    },
    {
      "epoch": 14.343234323432343,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.006735961546351114,
      "loss": 0.2314,
      "num_input_tokens_seen": 27515296,
      "step": 130380
    },
    {
      "epoch": 14.343784378437844,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006734759802029369,
      "loss": 0.2304,
      "num_input_tokens_seen": 27516352,
      "step": 130385
    },
    {
      "epoch": 14.344334433443345,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.006733558133884377,
      "loss": 0.2324,
      "num_input_tokens_seen": 27517376,
      "step": 130390
    },
    {
      "epoch": 14.344884488448844,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.0067323565419272185,
      "loss": 0.2308,
      "num_input_tokens_seen": 27518400,
      "step": 130395
    },
    {
      "epoch": 14.345434543454346,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.006731155026168957,
      "loss": 0.2298,
      "num_input_tokens_seen": 27519488,
      "step": 130400
    },
    {
      "epoch": 14.345984598459847,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.006729953586620678,
      "loss": 0.2314,
      "num_input_tokens_seen": 27520544,
      "step": 130405
    },
    {
      "epoch": 14.346534653465346,
      "grad_norm": 0.002593994140625,
      "learning_rate": 0.006728752223293447,
      "loss": 0.2329,
      "num_input_tokens_seen": 27521536,
      "step": 130410
    },
    {
      "epoch": 14.347084708470847,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.006727550936198334,
      "loss": 0.2319,
      "num_input_tokens_seen": 27522592,
      "step": 130415
    },
    {
      "epoch": 14.347634763476348,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006726349725346414,
      "loss": 0.2319,
      "num_input_tokens_seen": 27523584,
      "step": 130420
    },
    {
      "epoch": 14.348184818481847,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.0067251485907487574,
      "loss": 0.2314,
      "num_input_tokens_seen": 27524640,
      "step": 130425
    },
    {
      "epoch": 14.348734873487349,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.006723947532416443,
      "loss": 0.2303,
      "num_input_tokens_seen": 27525664,
      "step": 130430
    },
    {
      "epoch": 14.34928492849285,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.006722746550360531,
      "loss": 0.2319,
      "num_input_tokens_seen": 27526720,
      "step": 130435
    },
    {
      "epoch": 14.34983498349835,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.006721545644592088,
      "loss": 0.2303,
      "num_input_tokens_seen": 27527776,
      "step": 130440
    },
    {
      "epoch": 14.35038503850385,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0067203448151221855,
      "loss": 0.2319,
      "num_input_tokens_seen": 27528832,
      "step": 130445
    },
    {
      "epoch": 14.350935093509351,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.006719144061961892,
      "loss": 0.2319,
      "num_input_tokens_seen": 27529856,
      "step": 130450
    },
    {
      "epoch": 14.351485148514852,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.006717943385122282,
      "loss": 0.2313,
      "num_input_tokens_seen": 27530880,
      "step": 130455
    },
    {
      "epoch": 14.352035203520352,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00671674278461441,
      "loss": 0.2304,
      "num_input_tokens_seen": 27531936,
      "step": 130460
    },
    {
      "epoch": 14.352585258525853,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.006715542260449343,
      "loss": 0.233,
      "num_input_tokens_seen": 27533024,
      "step": 130465
    },
    {
      "epoch": 14.353135313531354,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.00671434181263815,
      "loss": 0.2319,
      "num_input_tokens_seen": 27534016,
      "step": 130470
    },
    {
      "epoch": 14.353685368536853,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.00671314144119189,
      "loss": 0.2303,
      "num_input_tokens_seen": 27535104,
      "step": 130475
    },
    {
      "epoch": 14.354235423542354,
      "grad_norm": 0.0004749298095703125,
      "learning_rate": 0.006711941146121628,
      "loss": 0.234,
      "num_input_tokens_seen": 27536192,
      "step": 130480
    },
    {
      "epoch": 14.354785478547855,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.006710740927438431,
      "loss": 0.2314,
      "num_input_tokens_seen": 27537216,
      "step": 130485
    },
    {
      "epoch": 14.355335533553355,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.006709540785153355,
      "loss": 0.2335,
      "num_input_tokens_seen": 27538304,
      "step": 130490
    },
    {
      "epoch": 14.355885588558856,
      "grad_norm": 0.005859375,
      "learning_rate": 0.006708340719277468,
      "loss": 0.2314,
      "num_input_tokens_seen": 27539456,
      "step": 130495
    },
    {
      "epoch": 14.356435643564357,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006707140729821818,
      "loss": 0.2303,
      "num_input_tokens_seen": 27540544,
      "step": 130500
    },
    {
      "epoch": 14.356985698569858,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.006705940816797474,
      "loss": 0.2319,
      "num_input_tokens_seen": 27541600,
      "step": 130505
    },
    {
      "epoch": 14.357535753575357,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0067047409802155,
      "loss": 0.2303,
      "num_input_tokens_seen": 27542720,
      "step": 130510
    },
    {
      "epoch": 14.358085808580858,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.006703541220086941,
      "loss": 0.2288,
      "num_input_tokens_seen": 27543840,
      "step": 130515
    },
    {
      "epoch": 14.35863586358636,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006702341536422869,
      "loss": 0.2308,
      "num_input_tokens_seen": 27544928,
      "step": 130520
    },
    {
      "epoch": 14.359185918591859,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.006701141929234332,
      "loss": 0.2293,
      "num_input_tokens_seen": 27546048,
      "step": 130525
    },
    {
      "epoch": 14.35973597359736,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.006699942398532382,
      "loss": 0.2319,
      "num_input_tokens_seen": 27547072,
      "step": 130530
    },
    {
      "epoch": 14.36028602860286,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006698742944328082,
      "loss": 0.2324,
      "num_input_tokens_seen": 27548192,
      "step": 130535
    },
    {
      "epoch": 14.36083608360836,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006697543566632483,
      "loss": 0.2319,
      "num_input_tokens_seen": 27549184,
      "step": 130540
    },
    {
      "epoch": 14.361386138613861,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.006696344265456647,
      "loss": 0.2314,
      "num_input_tokens_seen": 27550240,
      "step": 130545
    },
    {
      "epoch": 14.361936193619362,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006695145040811621,
      "loss": 0.2314,
      "num_input_tokens_seen": 27551296,
      "step": 130550
    },
    {
      "epoch": 14.362486248624862,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006693945892708453,
      "loss": 0.2314,
      "num_input_tokens_seen": 27552320,
      "step": 130555
    },
    {
      "epoch": 14.363036303630363,
      "grad_norm": 0.0008392333984375,
      "learning_rate": 0.006692746821158201,
      "loss": 0.2314,
      "num_input_tokens_seen": 27553408,
      "step": 130560
    },
    {
      "epoch": 14.363586358635864,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.006691547826171915,
      "loss": 0.2288,
      "num_input_tokens_seen": 27554432,
      "step": 130565
    },
    {
      "epoch": 14.364136413641365,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.006690348907760651,
      "loss": 0.2298,
      "num_input_tokens_seen": 27555520,
      "step": 130570
    },
    {
      "epoch": 14.364686468646864,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.006689150065935452,
      "loss": 0.2324,
      "num_input_tokens_seen": 27556576,
      "step": 130575
    },
    {
      "epoch": 14.365236523652365,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.006687951300707365,
      "loss": 0.2314,
      "num_input_tokens_seen": 27557600,
      "step": 130580
    },
    {
      "epoch": 14.365786578657866,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.006686752612087449,
      "loss": 0.2314,
      "num_input_tokens_seen": 27558656,
      "step": 130585
    },
    {
      "epoch": 14.366336633663366,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.006685554000086739,
      "loss": 0.2303,
      "num_input_tokens_seen": 27559712,
      "step": 130590
    },
    {
      "epoch": 14.366886688668867,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006684355464716288,
      "loss": 0.2314,
      "num_input_tokens_seen": 27560800,
      "step": 130595
    },
    {
      "epoch": 14.367436743674368,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.006683157005987148,
      "loss": 0.2314,
      "num_input_tokens_seen": 27561792,
      "step": 130600
    },
    {
      "epoch": 14.367986798679867,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.006681958623910352,
      "loss": 0.2319,
      "num_input_tokens_seen": 27562816,
      "step": 130605
    },
    {
      "epoch": 14.368536853685368,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.006680760318496959,
      "loss": 0.2288,
      "num_input_tokens_seen": 27563872,
      "step": 130610
    },
    {
      "epoch": 14.36908690869087,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.0066795620897580005,
      "loss": 0.2319,
      "num_input_tokens_seen": 27564992,
      "step": 130615
    },
    {
      "epoch": 14.369636963696369,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.006678363937704525,
      "loss": 0.2324,
      "num_input_tokens_seen": 27566048,
      "step": 130620
    },
    {
      "epoch": 14.37018701870187,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006677165862347582,
      "loss": 0.2324,
      "num_input_tokens_seen": 27567104,
      "step": 130625
    },
    {
      "epoch": 14.370737073707371,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.006675967863698203,
      "loss": 0.2303,
      "num_input_tokens_seen": 27568128,
      "step": 130630
    },
    {
      "epoch": 14.371287128712872,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.006674769941767438,
      "loss": 0.234,
      "num_input_tokens_seen": 27569152,
      "step": 130635
    },
    {
      "epoch": 14.371837183718371,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006673572096566319,
      "loss": 0.2309,
      "num_input_tokens_seen": 27570208,
      "step": 130640
    },
    {
      "epoch": 14.372387238723872,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006672374328105897,
      "loss": 0.2314,
      "num_input_tokens_seen": 27571264,
      "step": 130645
    },
    {
      "epoch": 14.372937293729374,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.006671176636397197,
      "loss": 0.2288,
      "num_input_tokens_seen": 27572288,
      "step": 130650
    },
    {
      "epoch": 14.373487348734873,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0066699790214512665,
      "loss": 0.2288,
      "num_input_tokens_seen": 27573376,
      "step": 130655
    },
    {
      "epoch": 14.374037403740374,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.006668781483279147,
      "loss": 0.2314,
      "num_input_tokens_seen": 27574464,
      "step": 130660
    },
    {
      "epoch": 14.374587458745875,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0066675840218918675,
      "loss": 0.233,
      "num_input_tokens_seen": 27575520,
      "step": 130665
    },
    {
      "epoch": 14.375137513751374,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.006666386637300471,
      "loss": 0.2303,
      "num_input_tokens_seen": 27576576,
      "step": 130670
    },
    {
      "epoch": 14.375687568756875,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00666518932951599,
      "loss": 0.2309,
      "num_input_tokens_seen": 27577600,
      "step": 130675
    },
    {
      "epoch": 14.376237623762377,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0066639920985494495,
      "loss": 0.2335,
      "num_input_tokens_seen": 27578688,
      "step": 130680
    },
    {
      "epoch": 14.376787678767876,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.006662794944411906,
      "loss": 0.2293,
      "num_input_tokens_seen": 27579776,
      "step": 130685
    },
    {
      "epoch": 14.377337733773377,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0066615978671143794,
      "loss": 0.2298,
      "num_input_tokens_seen": 27580768,
      "step": 130690
    },
    {
      "epoch": 14.377887788778878,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.006660400866667899,
      "loss": 0.2303,
      "num_input_tokens_seen": 27581824,
      "step": 130695
    },
    {
      "epoch": 14.37843784378438,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006659203943083507,
      "loss": 0.2324,
      "num_input_tokens_seen": 27582848,
      "step": 130700
    },
    {
      "epoch": 14.378987898789878,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.006658007096372224,
      "loss": 0.2324,
      "num_input_tokens_seen": 27583840,
      "step": 130705
    },
    {
      "epoch": 14.37953795379538,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.006656810326545089,
      "loss": 0.2304,
      "num_input_tokens_seen": 27584864,
      "step": 130710
    },
    {
      "epoch": 14.38008800880088,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.006655613633613134,
      "loss": 0.2324,
      "num_input_tokens_seen": 27585920,
      "step": 130715
    },
    {
      "epoch": 14.38063806380638,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.006654417017587378,
      "loss": 0.2293,
      "num_input_tokens_seen": 27586976,
      "step": 130720
    },
    {
      "epoch": 14.381188118811881,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.006653220478478862,
      "loss": 0.2319,
      "num_input_tokens_seen": 27588032,
      "step": 130725
    },
    {
      "epoch": 14.381738173817382,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006652024016298602,
      "loss": 0.2303,
      "num_input_tokens_seen": 27589056,
      "step": 130730
    },
    {
      "epoch": 14.382288228822881,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00665082763105763,
      "loss": 0.2314,
      "num_input_tokens_seen": 27590048,
      "step": 130735
    },
    {
      "epoch": 14.382838283828383,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.006649631322766979,
      "loss": 0.2319,
      "num_input_tokens_seen": 27591136,
      "step": 130740
    },
    {
      "epoch": 14.383388338833884,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.006648435091437665,
      "loss": 0.2298,
      "num_input_tokens_seen": 27592128,
      "step": 130745
    },
    {
      "epoch": 14.383938393839385,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.006647238937080722,
      "loss": 0.2293,
      "num_input_tokens_seen": 27593152,
      "step": 130750
    },
    {
      "epoch": 14.384488448844884,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0066460428597071635,
      "loss": 0.2319,
      "num_input_tokens_seen": 27594176,
      "step": 130755
    },
    {
      "epoch": 14.385038503850385,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006644846859328027,
      "loss": 0.2303,
      "num_input_tokens_seen": 27595200,
      "step": 130760
    },
    {
      "epoch": 14.385588558855886,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.006643650935954321,
      "loss": 0.2309,
      "num_input_tokens_seen": 27596256,
      "step": 130765
    },
    {
      "epoch": 14.386138613861386,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006642455089597075,
      "loss": 0.2314,
      "num_input_tokens_seen": 27597344,
      "step": 130770
    },
    {
      "epoch": 14.386688668866887,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.006641259320267314,
      "loss": 0.2324,
      "num_input_tokens_seen": 27598432,
      "step": 130775
    },
    {
      "epoch": 14.387238723872388,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.006640063627976051,
      "loss": 0.2324,
      "num_input_tokens_seen": 27599424,
      "step": 130780
    },
    {
      "epoch": 14.387788778877887,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0066388680127343145,
      "loss": 0.2293,
      "num_input_tokens_seen": 27600480,
      "step": 130785
    },
    {
      "epoch": 14.388338833883388,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.006637672474553121,
      "loss": 0.2293,
      "num_input_tokens_seen": 27601504,
      "step": 130790
    },
    {
      "epoch": 14.38888888888889,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.006636477013443475,
      "loss": 0.2319,
      "num_input_tokens_seen": 27602496,
      "step": 130795
    },
    {
      "epoch": 14.389438943894389,
      "grad_norm": 0.0050048828125,
      "learning_rate": 0.006635281629416419,
      "loss": 0.2324,
      "num_input_tokens_seen": 27603616,
      "step": 130800
    },
    {
      "epoch": 14.38998899889989,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0066340863224829535,
      "loss": 0.2309,
      "num_input_tokens_seen": 27604672,
      "step": 130805
    },
    {
      "epoch": 14.39053905390539,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.006632891092654104,
      "loss": 0.2288,
      "num_input_tokens_seen": 27605728,
      "step": 130810
    },
    {
      "epoch": 14.391089108910892,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.006631695939940885,
      "loss": 0.2324,
      "num_input_tokens_seen": 27606816,
      "step": 130815
    },
    {
      "epoch": 14.391639163916391,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0066305008643543,
      "loss": 0.2283,
      "num_input_tokens_seen": 27607904,
      "step": 130820
    },
    {
      "epoch": 14.392189218921892,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.006629305865905375,
      "loss": 0.2298,
      "num_input_tokens_seen": 27608928,
      "step": 130825
    },
    {
      "epoch": 14.392739273927393,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.00662811094460512,
      "loss": 0.2293,
      "num_input_tokens_seen": 27609952,
      "step": 130830
    },
    {
      "epoch": 14.393289328932893,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.006626916100464555,
      "loss": 0.2293,
      "num_input_tokens_seen": 27610944,
      "step": 130835
    },
    {
      "epoch": 14.393839383938394,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.006625721333494687,
      "loss": 0.2319,
      "num_input_tokens_seen": 27611968,
      "step": 130840
    },
    {
      "epoch": 14.394389438943895,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.006624526643706521,
      "loss": 0.2324,
      "num_input_tokens_seen": 27613024,
      "step": 130845
    },
    {
      "epoch": 14.394939493949394,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00662333203111108,
      "loss": 0.2314,
      "num_input_tokens_seen": 27614112,
      "step": 130850
    },
    {
      "epoch": 14.395489548954895,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.006622137495719356,
      "loss": 0.2325,
      "num_input_tokens_seen": 27615168,
      "step": 130855
    },
    {
      "epoch": 14.396039603960396,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.006620943037542382,
      "loss": 0.2303,
      "num_input_tokens_seen": 27616224,
      "step": 130860
    },
    {
      "epoch": 14.396589658965897,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.006619748656591157,
      "loss": 0.2324,
      "num_input_tokens_seen": 27617312,
      "step": 130865
    },
    {
      "epoch": 14.397139713971397,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0066185543528766815,
      "loss": 0.233,
      "num_input_tokens_seen": 27618336,
      "step": 130870
    },
    {
      "epoch": 14.397689768976898,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.006617360126409973,
      "loss": 0.233,
      "num_input_tokens_seen": 27619360,
      "step": 130875
    },
    {
      "epoch": 14.398239823982399,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.006616165977202028,
      "loss": 0.2309,
      "num_input_tokens_seen": 27620448,
      "step": 130880
    },
    {
      "epoch": 14.398789878987898,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.006614971905263859,
      "loss": 0.2309,
      "num_input_tokens_seen": 27621440,
      "step": 130885
    },
    {
      "epoch": 14.3993399339934,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.006613777910606477,
      "loss": 0.2309,
      "num_input_tokens_seen": 27622528,
      "step": 130890
    },
    {
      "epoch": 14.3998899889989,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0066125839932408715,
      "loss": 0.2293,
      "num_input_tokens_seen": 27623552,
      "step": 130895
    },
    {
      "epoch": 14.4004400440044,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006611390153178062,
      "loss": 0.2314,
      "num_input_tokens_seen": 27624640,
      "step": 130900
    },
    {
      "epoch": 14.400990099009901,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0066101963904290455,
      "loss": 0.2303,
      "num_input_tokens_seen": 27625664,
      "step": 130905
    },
    {
      "epoch": 14.401540154015402,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006609002705004817,
      "loss": 0.2288,
      "num_input_tokens_seen": 27626688,
      "step": 130910
    },
    {
      "epoch": 14.402090209020901,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.006607809096916382,
      "loss": 0.2324,
      "num_input_tokens_seen": 27627712,
      "step": 130915
    },
    {
      "epoch": 14.402640264026402,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.006606615566174745,
      "loss": 0.2324,
      "num_input_tokens_seen": 27628768,
      "step": 130920
    },
    {
      "epoch": 14.403190319031903,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.006605422112790911,
      "loss": 0.2309,
      "num_input_tokens_seen": 27629792,
      "step": 130925
    },
    {
      "epoch": 14.403740374037405,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006604228736775872,
      "loss": 0.2335,
      "num_input_tokens_seen": 27630848,
      "step": 130930
    },
    {
      "epoch": 14.404290429042904,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.006603035438140624,
      "loss": 0.2309,
      "num_input_tokens_seen": 27631872,
      "step": 130935
    },
    {
      "epoch": 14.404840484048405,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.006601842216896168,
      "loss": 0.2288,
      "num_input_tokens_seen": 27632928,
      "step": 130940
    },
    {
      "epoch": 14.405390539053906,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.006600649073053502,
      "loss": 0.2329,
      "num_input_tokens_seen": 27633888,
      "step": 130945
    },
    {
      "epoch": 14.405940594059405,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006599456006623629,
      "loss": 0.2319,
      "num_input_tokens_seen": 27634912,
      "step": 130950
    },
    {
      "epoch": 14.406490649064907,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.006598263017617539,
      "loss": 0.2309,
      "num_input_tokens_seen": 27635968,
      "step": 130955
    },
    {
      "epoch": 14.407040704070408,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006597070106046222,
      "loss": 0.2329,
      "num_input_tokens_seen": 27636992,
      "step": 130960
    },
    {
      "epoch": 14.407590759075907,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.006595877271920684,
      "loss": 0.2304,
      "num_input_tokens_seen": 27637952,
      "step": 130965
    },
    {
      "epoch": 14.408140814081408,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006594684515251905,
      "loss": 0.233,
      "num_input_tokens_seen": 27639040,
      "step": 130970
    },
    {
      "epoch": 14.408690869086909,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006593491836050888,
      "loss": 0.2319,
      "num_input_tokens_seen": 27640096,
      "step": 130975
    },
    {
      "epoch": 14.409240924092408,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.006592299234328627,
      "loss": 0.2319,
      "num_input_tokens_seen": 27641152,
      "step": 130980
    },
    {
      "epoch": 14.40979097909791,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0065911067100961045,
      "loss": 0.2319,
      "num_input_tokens_seen": 27642176,
      "step": 130985
    },
    {
      "epoch": 14.41034103410341,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.006589914263364322,
      "loss": 0.2335,
      "num_input_tokens_seen": 27643296,
      "step": 130990
    },
    {
      "epoch": 14.410891089108912,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.006588721894144257,
      "loss": 0.233,
      "num_input_tokens_seen": 27644384,
      "step": 130995
    },
    {
      "epoch": 14.411441144114411,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.006587529602446909,
      "loss": 0.2309,
      "num_input_tokens_seen": 27645408,
      "step": 131000
    },
    {
      "epoch": 14.411991199119912,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006586337388283269,
      "loss": 0.2314,
      "num_input_tokens_seen": 27646528,
      "step": 131005
    },
    {
      "epoch": 14.412541254125413,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.006585145251664312,
      "loss": 0.2288,
      "num_input_tokens_seen": 27647520,
      "step": 131010
    },
    {
      "epoch": 14.413091309130913,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.00658395319260104,
      "loss": 0.2298,
      "num_input_tokens_seen": 27648608,
      "step": 131015
    },
    {
      "epoch": 14.413641364136414,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.006582761211104429,
      "loss": 0.234,
      "num_input_tokens_seen": 27649600,
      "step": 131020
    },
    {
      "epoch": 14.414191419141915,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.006581569307185473,
      "loss": 0.2314,
      "num_input_tokens_seen": 27650656,
      "step": 131025
    },
    {
      "epoch": 14.414741474147414,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.006580377480855148,
      "loss": 0.2303,
      "num_input_tokens_seen": 27651680,
      "step": 131030
    },
    {
      "epoch": 14.415291529152915,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.006579185732124443,
      "loss": 0.2303,
      "num_input_tokens_seen": 27652736,
      "step": 131035
    },
    {
      "epoch": 14.415841584158416,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.006577994061004349,
      "loss": 0.2293,
      "num_input_tokens_seen": 27653792,
      "step": 131040
    },
    {
      "epoch": 14.416391639163916,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.006576802467505837,
      "loss": 0.2298,
      "num_input_tokens_seen": 27654816,
      "step": 131045
    },
    {
      "epoch": 14.416941694169417,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.006575610951639899,
      "loss": 0.2324,
      "num_input_tokens_seen": 27655872,
      "step": 131050
    },
    {
      "epoch": 14.417491749174918,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.006574419513417508,
      "loss": 0.2324,
      "num_input_tokens_seen": 27656928,
      "step": 131055
    },
    {
      "epoch": 14.418041804180419,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.00657322815284965,
      "loss": 0.2319,
      "num_input_tokens_seen": 27657952,
      "step": 131060
    },
    {
      "epoch": 14.418591859185918,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006572036869947309,
      "loss": 0.2319,
      "num_input_tokens_seen": 27659072,
      "step": 131065
    },
    {
      "epoch": 14.41914191419142,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006570845664721455,
      "loss": 0.2314,
      "num_input_tokens_seen": 27660160,
      "step": 131070
    },
    {
      "epoch": 14.41969196919692,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.006569654537183076,
      "loss": 0.2335,
      "num_input_tokens_seen": 27661216,
      "step": 131075
    },
    {
      "epoch": 14.42024202420242,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006568463487343148,
      "loss": 0.2319,
      "num_input_tokens_seen": 27662176,
      "step": 131080
    },
    {
      "epoch": 14.42079207920792,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.006567272515212639,
      "loss": 0.2319,
      "num_input_tokens_seen": 27663232,
      "step": 131085
    },
    {
      "epoch": 14.421342134213422,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.006566081620802533,
      "loss": 0.2319,
      "num_input_tokens_seen": 27664288,
      "step": 131090
    },
    {
      "epoch": 14.421892189218921,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.006564890804123811,
      "loss": 0.2314,
      "num_input_tokens_seen": 27665408,
      "step": 131095
    },
    {
      "epoch": 14.422442244224422,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.00656370006518744,
      "loss": 0.2288,
      "num_input_tokens_seen": 27666496,
      "step": 131100
    },
    {
      "epoch": 14.422992299229923,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0065625094040044,
      "loss": 0.2324,
      "num_input_tokens_seen": 27667584,
      "step": 131105
    },
    {
      "epoch": 14.423542354235423,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.006561318820585657,
      "loss": 0.2329,
      "num_input_tokens_seen": 27668640,
      "step": 131110
    },
    {
      "epoch": 14.424092409240924,
      "grad_norm": 0.01043701171875,
      "learning_rate": 0.00656012831494219,
      "loss": 0.2303,
      "num_input_tokens_seen": 27669760,
      "step": 131115
    },
    {
      "epoch": 14.424642464246425,
      "grad_norm": 0.005859375,
      "learning_rate": 0.006558937887084976,
      "loss": 0.2324,
      "num_input_tokens_seen": 27670752,
      "step": 131120
    },
    {
      "epoch": 14.425192519251926,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.006557747537024975,
      "loss": 0.2298,
      "num_input_tokens_seen": 27671840,
      "step": 131125
    },
    {
      "epoch": 14.425742574257425,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.006556557264773171,
      "loss": 0.2319,
      "num_input_tokens_seen": 27672928,
      "step": 131130
    },
    {
      "epoch": 14.426292629262926,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.006555367070340519,
      "loss": 0.2309,
      "num_input_tokens_seen": 27674016,
      "step": 131135
    },
    {
      "epoch": 14.426842684268427,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.006554176953738004,
      "loss": 0.2314,
      "num_input_tokens_seen": 27675072,
      "step": 131140
    },
    {
      "epoch": 14.427392739273927,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006552986914976581,
      "loss": 0.2298,
      "num_input_tokens_seen": 27676096,
      "step": 131145
    },
    {
      "epoch": 14.427942794279428,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.006551796954067224,
      "loss": 0.2324,
      "num_input_tokens_seen": 27677152,
      "step": 131150
    },
    {
      "epoch": 14.428492849284929,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006550607071020907,
      "loss": 0.2293,
      "num_input_tokens_seen": 27678272,
      "step": 131155
    },
    {
      "epoch": 14.429042904290428,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.0065494172658485846,
      "loss": 0.2335,
      "num_input_tokens_seen": 27679328,
      "step": 131160
    },
    {
      "epoch": 14.42959295929593,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.006548227538561234,
      "loss": 0.2319,
      "num_input_tokens_seen": 27680384,
      "step": 131165
    },
    {
      "epoch": 14.43014301430143,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.0065470378891698084,
      "loss": 0.2298,
      "num_input_tokens_seen": 27681440,
      "step": 131170
    },
    {
      "epoch": 14.430693069306932,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.006545848317685278,
      "loss": 0.2293,
      "num_input_tokens_seen": 27682528,
      "step": 131175
    },
    {
      "epoch": 14.43124312431243,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006544658824118614,
      "loss": 0.2298,
      "num_input_tokens_seen": 27683584,
      "step": 131180
    },
    {
      "epoch": 14.431793179317932,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.006543469408480765,
      "loss": 0.2283,
      "num_input_tokens_seen": 27684608,
      "step": 131185
    },
    {
      "epoch": 14.432343234323433,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006542280070782708,
      "loss": 0.2319,
      "num_input_tokens_seen": 27685696,
      "step": 131190
    },
    {
      "epoch": 14.432893289328932,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.006541090811035397,
      "loss": 0.2314,
      "num_input_tokens_seen": 27686688,
      "step": 131195
    },
    {
      "epoch": 14.433443344334433,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006539901629249787,
      "loss": 0.2319,
      "num_input_tokens_seen": 27687776,
      "step": 131200
    },
    {
      "epoch": 14.433993399339935,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.006538712525436843,
      "loss": 0.2324,
      "num_input_tokens_seen": 27688832,
      "step": 131205
    },
    {
      "epoch": 14.434543454345434,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.006537523499607527,
      "loss": 0.2314,
      "num_input_tokens_seen": 27689920,
      "step": 131210
    },
    {
      "epoch": 14.435093509350935,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.006536334551772801,
      "loss": 0.233,
      "num_input_tokens_seen": 27690976,
      "step": 131215
    },
    {
      "epoch": 14.435643564356436,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.006535145681943618,
      "loss": 0.2298,
      "num_input_tokens_seen": 27692000,
      "step": 131220
    },
    {
      "epoch": 14.436193619361935,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.006533956890130932,
      "loss": 0.2293,
      "num_input_tokens_seen": 27693120,
      "step": 131225
    },
    {
      "epoch": 14.436743674367436,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0065327681763457,
      "loss": 0.2325,
      "num_input_tokens_seen": 27694176,
      "step": 131230
    },
    {
      "epoch": 14.437293729372938,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.0065315795405988825,
      "loss": 0.2298,
      "num_input_tokens_seen": 27695232,
      "step": 131235
    },
    {
      "epoch": 14.437843784378439,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.006530390982901438,
      "loss": 0.2319,
      "num_input_tokens_seen": 27696288,
      "step": 131240
    },
    {
      "epoch": 14.438393839383938,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.006529202503264315,
      "loss": 0.2298,
      "num_input_tokens_seen": 27697344,
      "step": 131245
    },
    {
      "epoch": 14.438943894389439,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0065280141016984625,
      "loss": 0.2314,
      "num_input_tokens_seen": 27698400,
      "step": 131250
    },
    {
      "epoch": 14.43949394939494,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.006526825778214845,
      "loss": 0.2309,
      "num_input_tokens_seen": 27699456,
      "step": 131255
    },
    {
      "epoch": 14.44004400440044,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.006525637532824403,
      "loss": 0.2298,
      "num_input_tokens_seen": 27700576,
      "step": 131260
    },
    {
      "epoch": 14.44059405940594,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.006524449365538092,
      "loss": 0.2335,
      "num_input_tokens_seen": 27701568,
      "step": 131265
    },
    {
      "epoch": 14.441144114411442,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.006523261276366873,
      "loss": 0.235,
      "num_input_tokens_seen": 27702656,
      "step": 131270
    },
    {
      "epoch": 14.441694169416941,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.0065220732653216784,
      "loss": 0.2335,
      "num_input_tokens_seen": 27703744,
      "step": 131275
    },
    {
      "epoch": 14.442244224422442,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.006520885332413473,
      "loss": 0.234,
      "num_input_tokens_seen": 27704832,
      "step": 131280
    },
    {
      "epoch": 14.442794279427943,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.006519697477653194,
      "loss": 0.2324,
      "num_input_tokens_seen": 27705920,
      "step": 131285
    },
    {
      "epoch": 14.443344334433444,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0065185097010517935,
      "loss": 0.233,
      "num_input_tokens_seen": 27706976,
      "step": 131290
    },
    {
      "epoch": 14.443894389438944,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006517322002620224,
      "loss": 0.2324,
      "num_input_tokens_seen": 27707968,
      "step": 131295
    },
    {
      "epoch": 14.444444444444445,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.006516134382369424,
      "loss": 0.2319,
      "num_input_tokens_seen": 27709024,
      "step": 131300
    },
    {
      "epoch": 14.444994499449946,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.006514946840310346,
      "loss": 0.2298,
      "num_input_tokens_seen": 27710080,
      "step": 131305
    },
    {
      "epoch": 14.445544554455445,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0065137593764539314,
      "loss": 0.2298,
      "num_input_tokens_seen": 27711168,
      "step": 131310
    },
    {
      "epoch": 14.446094609460946,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00651257199081112,
      "loss": 0.2314,
      "num_input_tokens_seen": 27712224,
      "step": 131315
    },
    {
      "epoch": 14.446644664466447,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.006511384683392859,
      "loss": 0.2309,
      "num_input_tokens_seen": 27713280,
      "step": 131320
    },
    {
      "epoch": 14.447194719471947,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.006510197454210093,
      "loss": 0.2335,
      "num_input_tokens_seen": 27714368,
      "step": 131325
    },
    {
      "epoch": 14.447744774477448,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006509010303273768,
      "loss": 0.2314,
      "num_input_tokens_seen": 27715424,
      "step": 131330
    },
    {
      "epoch": 14.448294829482949,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006507823230594822,
      "loss": 0.2319,
      "num_input_tokens_seen": 27716416,
      "step": 131335
    },
    {
      "epoch": 14.448844884488448,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0065066362361841865,
      "loss": 0.2309,
      "num_input_tokens_seen": 27717408,
      "step": 131340
    },
    {
      "epoch": 14.44939493949395,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006505449320052812,
      "loss": 0.2293,
      "num_input_tokens_seen": 27718368,
      "step": 131345
    },
    {
      "epoch": 14.44994499449945,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006504262482211634,
      "loss": 0.2319,
      "num_input_tokens_seen": 27719392,
      "step": 131350
    },
    {
      "epoch": 14.450495049504951,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006503075722671596,
      "loss": 0.2314,
      "num_input_tokens_seen": 27720448,
      "step": 131355
    },
    {
      "epoch": 14.45104510451045,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0065018890414436335,
      "loss": 0.2319,
      "num_input_tokens_seen": 27721536,
      "step": 131360
    },
    {
      "epoch": 14.451595159515952,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0065007024385386756,
      "loss": 0.2329,
      "num_input_tokens_seen": 27722688,
      "step": 131365
    },
    {
      "epoch": 14.452145214521453,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.006499515913967672,
      "loss": 0.2303,
      "num_input_tokens_seen": 27723744,
      "step": 131370
    },
    {
      "epoch": 14.452695269526952,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.006498329467741544,
      "loss": 0.2299,
      "num_input_tokens_seen": 27724832,
      "step": 131375
    },
    {
      "epoch": 14.453245324532453,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.006497143099871235,
      "loss": 0.2304,
      "num_input_tokens_seen": 27725920,
      "step": 131380
    },
    {
      "epoch": 14.453795379537954,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.006495956810367683,
      "loss": 0.2303,
      "num_input_tokens_seen": 27726976,
      "step": 131385
    },
    {
      "epoch": 14.454345434543454,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.006494770599241811,
      "loss": 0.2293,
      "num_input_tokens_seen": 27727968,
      "step": 131390
    },
    {
      "epoch": 14.454895489548955,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.006493584466504562,
      "loss": 0.2288,
      "num_input_tokens_seen": 27728928,
      "step": 131395
    },
    {
      "epoch": 14.455445544554456,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.006492398412166859,
      "loss": 0.2324,
      "num_input_tokens_seen": 27730016,
      "step": 131400
    },
    {
      "epoch": 14.455995599559955,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.006491212436239635,
      "loss": 0.2314,
      "num_input_tokens_seen": 27731072,
      "step": 131405
    },
    {
      "epoch": 14.456545654565456,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.006490026538733831,
      "loss": 0.2324,
      "num_input_tokens_seen": 27732160,
      "step": 131410
    },
    {
      "epoch": 14.457095709570957,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.0064888407196603635,
      "loss": 0.2298,
      "num_input_tokens_seen": 27733216,
      "step": 131415
    },
    {
      "epoch": 14.457645764576458,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.006487654979030172,
      "loss": 0.2278,
      "num_input_tokens_seen": 27734304,
      "step": 131420
    },
    {
      "epoch": 14.458195819581958,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.006486469316854174,
      "loss": 0.2314,
      "num_input_tokens_seen": 27735392,
      "step": 131425
    },
    {
      "epoch": 14.458745874587459,
      "grad_norm": 0.000591278076171875,
      "learning_rate": 0.006485283733143309,
      "loss": 0.2314,
      "num_input_tokens_seen": 27736416,
      "step": 131430
    },
    {
      "epoch": 14.45929592959296,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.006484098227908492,
      "loss": 0.2309,
      "num_input_tokens_seen": 27737504,
      "step": 131435
    },
    {
      "epoch": 14.45984598459846,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0064829128011606564,
      "loss": 0.2308,
      "num_input_tokens_seen": 27738624,
      "step": 131440
    },
    {
      "epoch": 14.46039603960396,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006481727452910731,
      "loss": 0.2309,
      "num_input_tokens_seen": 27739680,
      "step": 131445
    },
    {
      "epoch": 14.460946094609461,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.006480542183169631,
      "loss": 0.2277,
      "num_input_tokens_seen": 27740800,
      "step": 131450
    },
    {
      "epoch": 14.46149614961496,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006479356991948291,
      "loss": 0.2319,
      "num_input_tokens_seen": 27741888,
      "step": 131455
    },
    {
      "epoch": 14.462046204620462,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.006478171879257629,
      "loss": 0.2314,
      "num_input_tokens_seen": 27743040,
      "step": 131460
    },
    {
      "epoch": 14.462596259625963,
      "grad_norm": 0.005859375,
      "learning_rate": 0.006476986845108558,
      "loss": 0.2299,
      "num_input_tokens_seen": 27744160,
      "step": 131465
    },
    {
      "epoch": 14.463146314631462,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0064758018895120185,
      "loss": 0.2324,
      "num_input_tokens_seen": 27745184,
      "step": 131470
    },
    {
      "epoch": 14.463696369636963,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.006474617012478924,
      "loss": 0.233,
      "num_input_tokens_seen": 27746176,
      "step": 131475
    },
    {
      "epoch": 14.464246424642464,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.006473432214020187,
      "loss": 0.2314,
      "num_input_tokens_seen": 27747296,
      "step": 131480
    },
    {
      "epoch": 14.464796479647966,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.00647224749414674,
      "loss": 0.2314,
      "num_input_tokens_seen": 27748384,
      "step": 131485
    },
    {
      "epoch": 14.465346534653465,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006471062852869491,
      "loss": 0.2314,
      "num_input_tokens_seen": 27749504,
      "step": 131490
    },
    {
      "epoch": 14.465896589658966,
      "grad_norm": 0.005859375,
      "learning_rate": 0.006469878290199362,
      "loss": 0.2329,
      "num_input_tokens_seen": 27750624,
      "step": 131495
    },
    {
      "epoch": 14.466446644664467,
      "grad_norm": 0.0004634857177734375,
      "learning_rate": 0.006468693806147278,
      "loss": 0.2272,
      "num_input_tokens_seen": 27751648,
      "step": 131500
    },
    {
      "epoch": 14.466996699669966,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.006467509400724145,
      "loss": 0.2314,
      "num_input_tokens_seen": 27752736,
      "step": 131505
    },
    {
      "epoch": 14.467546754675467,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006466325073940886,
      "loss": 0.2293,
      "num_input_tokens_seen": 27753824,
      "step": 131510
    },
    {
      "epoch": 14.468096809680969,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00646514082580841,
      "loss": 0.2298,
      "num_input_tokens_seen": 27754848,
      "step": 131515
    },
    {
      "epoch": 14.468646864686468,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.006463956656337636,
      "loss": 0.2319,
      "num_input_tokens_seen": 27755936,
      "step": 131520
    },
    {
      "epoch": 14.469196919691969,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.006462772565539483,
      "loss": 0.2283,
      "num_input_tokens_seen": 27757056,
      "step": 131525
    },
    {
      "epoch": 14.46974697469747,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.006461588553424852,
      "loss": 0.2345,
      "num_input_tokens_seen": 27758112,
      "step": 131530
    },
    {
      "epoch": 14.47029702970297,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0064604046200046695,
      "loss": 0.2309,
      "num_input_tokens_seen": 27759200,
      "step": 131535
    },
    {
      "epoch": 14.47084708470847,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006459220765289833,
      "loss": 0.2288,
      "num_input_tokens_seen": 27760256,
      "step": 131540
    },
    {
      "epoch": 14.471397139713972,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.006458036989291266,
      "loss": 0.2309,
      "num_input_tokens_seen": 27761312,
      "step": 131545
    },
    {
      "epoch": 14.471947194719473,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.006456853292019869,
      "loss": 0.2303,
      "num_input_tokens_seen": 27762400,
      "step": 131550
    },
    {
      "epoch": 14.472497249724972,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.006455669673486555,
      "loss": 0.2298,
      "num_input_tokens_seen": 27763456,
      "step": 131555
    },
    {
      "epoch": 14.473047304730473,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.00645448613370224,
      "loss": 0.2351,
      "num_input_tokens_seen": 27764512,
      "step": 131560
    },
    {
      "epoch": 14.473597359735974,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.006453302672677818,
      "loss": 0.2319,
      "num_input_tokens_seen": 27765536,
      "step": 131565
    },
    {
      "epoch": 14.474147414741473,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.0064521192904242125,
      "loss": 0.2319,
      "num_input_tokens_seen": 27766592,
      "step": 131570
    },
    {
      "epoch": 14.474697469746975,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.00645093598695232,
      "loss": 0.2335,
      "num_input_tokens_seen": 27767616,
      "step": 131575
    },
    {
      "epoch": 14.475247524752476,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.006449752762273038,
      "loss": 0.2314,
      "num_input_tokens_seen": 27768672,
      "step": 131580
    },
    {
      "epoch": 14.475797579757975,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.006448569616397293,
      "loss": 0.2309,
      "num_input_tokens_seen": 27769728,
      "step": 131585
    },
    {
      "epoch": 14.476347634763476,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.006447386549335971,
      "loss": 0.2314,
      "num_input_tokens_seen": 27770784,
      "step": 131590
    },
    {
      "epoch": 14.476897689768977,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.00644620356109999,
      "loss": 0.2314,
      "num_input_tokens_seen": 27771808,
      "step": 131595
    },
    {
      "epoch": 14.477447744774478,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.006445020651700248,
      "loss": 0.2298,
      "num_input_tokens_seen": 27772864,
      "step": 131600
    },
    {
      "epoch": 14.477997799779978,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.006443837821147639,
      "loss": 0.2293,
      "num_input_tokens_seen": 27773888,
      "step": 131605
    },
    {
      "epoch": 14.478547854785479,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.006442655069453069,
      "loss": 0.2319,
      "num_input_tokens_seen": 27774912,
      "step": 131610
    },
    {
      "epoch": 14.47909790979098,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.006441472396627444,
      "loss": 0.233,
      "num_input_tokens_seen": 27776000,
      "step": 131615
    },
    {
      "epoch": 14.479647964796479,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006440289802681665,
      "loss": 0.2308,
      "num_input_tokens_seen": 27777056,
      "step": 131620
    },
    {
      "epoch": 14.48019801980198,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.006439107287626628,
      "loss": 0.2309,
      "num_input_tokens_seen": 27778144,
      "step": 131625
    },
    {
      "epoch": 14.480748074807481,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.0064379248514732255,
      "loss": 0.2319,
      "num_input_tokens_seen": 27779200,
      "step": 131630
    },
    {
      "epoch": 14.48129812981298,
      "grad_norm": 0.0028076171875,
      "learning_rate": 0.006436742494232368,
      "loss": 0.2309,
      "num_input_tokens_seen": 27780256,
      "step": 131635
    },
    {
      "epoch": 14.481848184818482,
      "grad_norm": 0.00286865234375,
      "learning_rate": 0.006435560215914933,
      "loss": 0.2309,
      "num_input_tokens_seen": 27781312,
      "step": 131640
    },
    {
      "epoch": 14.482398239823983,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006434378016531842,
      "loss": 0.2314,
      "num_input_tokens_seen": 27782336,
      "step": 131645
    },
    {
      "epoch": 14.482948294829482,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.006433195896093979,
      "loss": 0.2319,
      "num_input_tokens_seen": 27783392,
      "step": 131650
    },
    {
      "epoch": 14.483498349834983,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.006432013854612232,
      "loss": 0.2299,
      "num_input_tokens_seen": 27784480,
      "step": 131655
    },
    {
      "epoch": 14.484048404840484,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.0064308318920975075,
      "loss": 0.2309,
      "num_input_tokens_seen": 27785536,
      "step": 131660
    },
    {
      "epoch": 14.484598459845985,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.006429650008560689,
      "loss": 0.2319,
      "num_input_tokens_seen": 27786656,
      "step": 131665
    },
    {
      "epoch": 14.485148514851485,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.006428468204012673,
      "loss": 0.2319,
      "num_input_tokens_seen": 27787616,
      "step": 131670
    },
    {
      "epoch": 14.485698569856986,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.006427286478464359,
      "loss": 0.2298,
      "num_input_tokens_seen": 27788736,
      "step": 131675
    },
    {
      "epoch": 14.486248624862487,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.006426104831926626,
      "loss": 0.2272,
      "num_input_tokens_seen": 27789824,
      "step": 131680
    },
    {
      "epoch": 14.486798679867986,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0064249232644103755,
      "loss": 0.2335,
      "num_input_tokens_seen": 27790880,
      "step": 131685
    },
    {
      "epoch": 14.487348734873487,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0064237417759264925,
      "loss": 0.2319,
      "num_input_tokens_seen": 27791968,
      "step": 131690
    },
    {
      "epoch": 14.487898789878988,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0064225603664858615,
      "loss": 0.233,
      "num_input_tokens_seen": 27793024,
      "step": 131695
    },
    {
      "epoch": 14.488448844884488,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.006421379036099376,
      "loss": 0.2313,
      "num_input_tokens_seen": 27794080,
      "step": 131700
    },
    {
      "epoch": 14.488998899889989,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.006420197784777924,
      "loss": 0.233,
      "num_input_tokens_seen": 27795104,
      "step": 131705
    },
    {
      "epoch": 14.48954895489549,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.006419016612532397,
      "loss": 0.2304,
      "num_input_tokens_seen": 27796256,
      "step": 131710
    },
    {
      "epoch": 14.490099009900991,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.006417835519373675,
      "loss": 0.2309,
      "num_input_tokens_seen": 27797344,
      "step": 131715
    },
    {
      "epoch": 14.49064906490649,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.00641665450531264,
      "loss": 0.2299,
      "num_input_tokens_seen": 27798432,
      "step": 131720
    },
    {
      "epoch": 14.491199119911991,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006415473570360182,
      "loss": 0.2319,
      "num_input_tokens_seen": 27799552,
      "step": 131725
    },
    {
      "epoch": 14.491749174917492,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006414292714527185,
      "loss": 0.2335,
      "num_input_tokens_seen": 27800576,
      "step": 131730
    },
    {
      "epoch": 14.492299229922992,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.006413111937824538,
      "loss": 0.233,
      "num_input_tokens_seen": 27801632,
      "step": 131735
    },
    {
      "epoch": 14.492849284928493,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.006411931240263117,
      "loss": 0.2324,
      "num_input_tokens_seen": 27802752,
      "step": 131740
    },
    {
      "epoch": 14.493399339933994,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.0064107506218538,
      "loss": 0.2293,
      "num_input_tokens_seen": 27803808,
      "step": 131745
    },
    {
      "epoch": 14.493949394939493,
      "grad_norm": 0.01031494140625,
      "learning_rate": 0.006409570082607479,
      "loss": 0.2267,
      "num_input_tokens_seen": 27804864,
      "step": 131750
    },
    {
      "epoch": 14.494499449944994,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.006408389622535022,
      "loss": 0.2298,
      "num_input_tokens_seen": 27805952,
      "step": 131755
    },
    {
      "epoch": 14.495049504950495,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0064072092416473155,
      "loss": 0.2288,
      "num_input_tokens_seen": 27806944,
      "step": 131760
    },
    {
      "epoch": 14.495599559955995,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.006406028939955244,
      "loss": 0.2278,
      "num_input_tokens_seen": 27808000,
      "step": 131765
    },
    {
      "epoch": 14.496149614961496,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.006404848717469674,
      "loss": 0.2309,
      "num_input_tokens_seen": 27809024,
      "step": 131770
    },
    {
      "epoch": 14.496699669966997,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006403668574201495,
      "loss": 0.2319,
      "num_input_tokens_seen": 27810016,
      "step": 131775
    },
    {
      "epoch": 14.497249724972498,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.006402488510161571,
      "loss": 0.2325,
      "num_input_tokens_seen": 27811040,
      "step": 131780
    },
    {
      "epoch": 14.497799779977997,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.006401308525360787,
      "loss": 0.2272,
      "num_input_tokens_seen": 27812160,
      "step": 131785
    },
    {
      "epoch": 14.498349834983498,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00640012861981002,
      "loss": 0.2319,
      "num_input_tokens_seen": 27813152,
      "step": 131790
    },
    {
      "epoch": 14.498899889989,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.0063989487935201365,
      "loss": 0.2324,
      "num_input_tokens_seen": 27814240,
      "step": 131795
    },
    {
      "epoch": 14.499449944994499,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00639776904650202,
      "loss": 0.2324,
      "num_input_tokens_seen": 27815232,
      "step": 131800
    },
    {
      "epoch": 14.5,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006396589378766532,
      "loss": 0.2309,
      "num_input_tokens_seen": 27816384,
      "step": 131805
    },
    {
      "epoch": 14.500550055005501,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.006395409790324559,
      "loss": 0.2314,
      "num_input_tokens_seen": 27817472,
      "step": 131810
    },
    {
      "epoch": 14.501100110011,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.006394230281186957,
      "loss": 0.2345,
      "num_input_tokens_seen": 27818528,
      "step": 131815
    },
    {
      "epoch": 14.501650165016502,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.006393050851364607,
      "loss": 0.2309,
      "num_input_tokens_seen": 27819552,
      "step": 131820
    },
    {
      "epoch": 14.502200220022003,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.006391871500868382,
      "loss": 0.2319,
      "num_input_tokens_seen": 27820608,
      "step": 131825
    },
    {
      "epoch": 14.502750275027502,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006390692229709141,
      "loss": 0.2293,
      "num_input_tokens_seen": 27821632,
      "step": 131830
    },
    {
      "epoch": 14.503300330033003,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.006389513037897765,
      "loss": 0.234,
      "num_input_tokens_seen": 27822720,
      "step": 131835
    },
    {
      "epoch": 14.503850385038504,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006388333925445109,
      "loss": 0.2288,
      "num_input_tokens_seen": 27823776,
      "step": 131840
    },
    {
      "epoch": 14.504400440044005,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006387154892362047,
      "loss": 0.2335,
      "num_input_tokens_seen": 27824800,
      "step": 131845
    },
    {
      "epoch": 14.504950495049505,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006385975938659453,
      "loss": 0.2309,
      "num_input_tokens_seen": 27825792,
      "step": 131850
    },
    {
      "epoch": 14.505500550055006,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.006384797064348183,
      "loss": 0.2314,
      "num_input_tokens_seen": 27826784,
      "step": 131855
    },
    {
      "epoch": 14.506050605060507,
      "grad_norm": 0.0007171630859375,
      "learning_rate": 0.0063836182694391,
      "loss": 0.2314,
      "num_input_tokens_seen": 27827808,
      "step": 131860
    },
    {
      "epoch": 14.506600660066006,
      "grad_norm": 0.00057220458984375,
      "learning_rate": 0.006382439553943078,
      "loss": 0.2329,
      "num_input_tokens_seen": 27828896,
      "step": 131865
    },
    {
      "epoch": 14.507150715071507,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006381260917870971,
      "loss": 0.2329,
      "num_input_tokens_seen": 27830048,
      "step": 131870
    },
    {
      "epoch": 14.507700770077008,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.006380082361233645,
      "loss": 0.2319,
      "num_input_tokens_seen": 27831136,
      "step": 131875
    },
    {
      "epoch": 14.508250825082508,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.006378903884041971,
      "loss": 0.2314,
      "num_input_tokens_seen": 27832160,
      "step": 131880
    },
    {
      "epoch": 14.508800880088009,
      "grad_norm": 0.010498046875,
      "learning_rate": 0.006377725486306795,
      "loss": 0.2298,
      "num_input_tokens_seen": 27833216,
      "step": 131885
    },
    {
      "epoch": 14.50935093509351,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006376547168038993,
      "loss": 0.2309,
      "num_input_tokens_seen": 27834240,
      "step": 131890
    },
    {
      "epoch": 14.509900990099009,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.006375368929249412,
      "loss": 0.2324,
      "num_input_tokens_seen": 27835232,
      "step": 131895
    },
    {
      "epoch": 14.51045104510451,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.006374190769948918,
      "loss": 0.2324,
      "num_input_tokens_seen": 27836256,
      "step": 131900
    },
    {
      "epoch": 14.511001100110011,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.006373012690148372,
      "loss": 0.2335,
      "num_input_tokens_seen": 27837280,
      "step": 131905
    },
    {
      "epoch": 14.511551155115512,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.006371834689858625,
      "loss": 0.2298,
      "num_input_tokens_seen": 27838368,
      "step": 131910
    },
    {
      "epoch": 14.512101210121012,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.006370656769090542,
      "loss": 0.2293,
      "num_input_tokens_seen": 27839456,
      "step": 131915
    },
    {
      "epoch": 14.512651265126513,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.006369478927854969,
      "loss": 0.234,
      "num_input_tokens_seen": 27840448,
      "step": 131920
    },
    {
      "epoch": 14.513201320132014,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006368301166162774,
      "loss": 0.2293,
      "num_input_tokens_seen": 27841504,
      "step": 131925
    },
    {
      "epoch": 14.513751375137513,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.006367123484024798,
      "loss": 0.235,
      "num_input_tokens_seen": 27842528,
      "step": 131930
    },
    {
      "epoch": 14.514301430143014,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.006365945881451905,
      "loss": 0.2314,
      "num_input_tokens_seen": 27843584,
      "step": 131935
    },
    {
      "epoch": 14.514851485148515,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.0063647683584549496,
      "loss": 0.2329,
      "num_input_tokens_seen": 27844640,
      "step": 131940
    },
    {
      "epoch": 14.515401540154015,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.006363590915044774,
      "loss": 0.2324,
      "num_input_tokens_seen": 27845664,
      "step": 131945
    },
    {
      "epoch": 14.515951595159516,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.006362413551232244,
      "loss": 0.2309,
      "num_input_tokens_seen": 27846688,
      "step": 131950
    },
    {
      "epoch": 14.516501650165017,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.006361236267028199,
      "loss": 0.2288,
      "num_input_tokens_seen": 27847776,
      "step": 131955
    },
    {
      "epoch": 14.517051705170516,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.006360059062443492,
      "loss": 0.2324,
      "num_input_tokens_seen": 27848832,
      "step": 131960
    },
    {
      "epoch": 14.517601760176017,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.006358881937488982,
      "loss": 0.234,
      "num_input_tokens_seen": 27849952,
      "step": 131965
    },
    {
      "epoch": 14.518151815181518,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.006357704892175503,
      "loss": 0.2324,
      "num_input_tokens_seen": 27851008,
      "step": 131970
    },
    {
      "epoch": 14.51870187018702,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.006356527926513919,
      "loss": 0.2319,
      "num_input_tokens_seen": 27852064,
      "step": 131975
    },
    {
      "epoch": 14.519251925192519,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.006355351040515067,
      "loss": 0.2319,
      "num_input_tokens_seen": 27853152,
      "step": 131980
    },
    {
      "epoch": 14.51980198019802,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.006354174234189792,
      "loss": 0.2335,
      "num_input_tokens_seen": 27854208,
      "step": 131985
    },
    {
      "epoch": 14.520352035203521,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0063529975075489456,
      "loss": 0.2319,
      "num_input_tokens_seen": 27855296,
      "step": 131990
    },
    {
      "epoch": 14.52090209020902,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.006351820860603372,
      "loss": 0.2314,
      "num_input_tokens_seen": 27856384,
      "step": 131995
    },
    {
      "epoch": 14.521452145214521,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006350644293363919,
      "loss": 0.2314,
      "num_input_tokens_seen": 27857376,
      "step": 132000
    },
    {
      "epoch": 14.522002200220022,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.006349467805841429,
      "loss": 0.2335,
      "num_input_tokens_seen": 27858432,
      "step": 132005
    },
    {
      "epoch": 14.522552255225522,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.006348291398046738,
      "loss": 0.2309,
      "num_input_tokens_seen": 27859520,
      "step": 132010
    },
    {
      "epoch": 14.523102310231023,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.006347115069990692,
      "loss": 0.2293,
      "num_input_tokens_seen": 27860608,
      "step": 132015
    },
    {
      "epoch": 14.523652365236524,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006345938821684135,
      "loss": 0.2308,
      "num_input_tokens_seen": 27861632,
      "step": 132020
    },
    {
      "epoch": 14.524202420242025,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.006344762653137913,
      "loss": 0.2319,
      "num_input_tokens_seen": 27862720,
      "step": 132025
    },
    {
      "epoch": 14.524752475247524,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00634358656436286,
      "loss": 0.2308,
      "num_input_tokens_seen": 27863744,
      "step": 132030
    },
    {
      "epoch": 14.525302530253025,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.0063424105553698115,
      "loss": 0.2314,
      "num_input_tokens_seen": 27864864,
      "step": 132035
    },
    {
      "epoch": 14.525852585258527,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.006341234626169616,
      "loss": 0.2324,
      "num_input_tokens_seen": 27865984,
      "step": 132040
    },
    {
      "epoch": 14.526402640264026,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0063400587767731,
      "loss": 0.2283,
      "num_input_tokens_seen": 27867040,
      "step": 132045
    },
    {
      "epoch": 14.526952695269527,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.006338883007191105,
      "loss": 0.2324,
      "num_input_tokens_seen": 27868096,
      "step": 132050
    },
    {
      "epoch": 14.527502750275028,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.006337707317434478,
      "loss": 0.2324,
      "num_input_tokens_seen": 27869088,
      "step": 132055
    },
    {
      "epoch": 14.528052805280527,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.006336531707514039,
      "loss": 0.2319,
      "num_input_tokens_seen": 27870176,
      "step": 132060
    },
    {
      "epoch": 14.528602860286028,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.0063353561774406346,
      "loss": 0.2309,
      "num_input_tokens_seen": 27871232,
      "step": 132065
    },
    {
      "epoch": 14.52915291529153,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00633418072722509,
      "loss": 0.2314,
      "num_input_tokens_seen": 27872224,
      "step": 132070
    },
    {
      "epoch": 14.52970297029703,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.006333005356878244,
      "loss": 0.2298,
      "num_input_tokens_seen": 27873248,
      "step": 132075
    },
    {
      "epoch": 14.53025302530253,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006331830066410935,
      "loss": 0.2298,
      "num_input_tokens_seen": 27874304,
      "step": 132080
    },
    {
      "epoch": 14.530803080308031,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.006330654855833982,
      "loss": 0.2309,
      "num_input_tokens_seen": 27875328,
      "step": 132085
    },
    {
      "epoch": 14.531353135313532,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.0063294797251582284,
      "loss": 0.2303,
      "num_input_tokens_seen": 27876320,
      "step": 132090
    },
    {
      "epoch": 14.531903190319031,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.006328304674394501,
      "loss": 0.2278,
      "num_input_tokens_seen": 27877376,
      "step": 132095
    },
    {
      "epoch": 14.532453245324533,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0063271297035536215,
      "loss": 0.2324,
      "num_input_tokens_seen": 27878400,
      "step": 132100
    },
    {
      "epoch": 14.533003300330034,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.006325954812646427,
      "loss": 0.233,
      "num_input_tokens_seen": 27879488,
      "step": 132105
    },
    {
      "epoch": 14.533553355335533,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006324780001683745,
      "loss": 0.2314,
      "num_input_tokens_seen": 27880480,
      "step": 132110
    },
    {
      "epoch": 14.534103410341034,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.006323605270676407,
      "loss": 0.2308,
      "num_input_tokens_seen": 27881568,
      "step": 132115
    },
    {
      "epoch": 14.534653465346535,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.006322430619635236,
      "loss": 0.234,
      "num_input_tokens_seen": 27882688,
      "step": 132120
    },
    {
      "epoch": 14.535203520352034,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.006321256048571053,
      "loss": 0.2324,
      "num_input_tokens_seen": 27883744,
      "step": 132125
    },
    {
      "epoch": 14.535753575357536,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.006320081557494691,
      "loss": 0.2309,
      "num_input_tokens_seen": 27884768,
      "step": 132130
    },
    {
      "epoch": 14.536303630363037,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.006318907146416969,
      "loss": 0.2309,
      "num_input_tokens_seen": 27885856,
      "step": 132135
    },
    {
      "epoch": 14.536853685368538,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.006317732815348721,
      "loss": 0.2293,
      "num_input_tokens_seen": 27887008,
      "step": 132140
    },
    {
      "epoch": 14.537403740374037,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.006316558564300762,
      "loss": 0.2314,
      "num_input_tokens_seen": 27888128,
      "step": 132145
    },
    {
      "epoch": 14.537953795379538,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0063153843932839135,
      "loss": 0.2319,
      "num_input_tokens_seen": 27889184,
      "step": 132150
    },
    {
      "epoch": 14.53850385038504,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.006314210302309002,
      "loss": 0.2335,
      "num_input_tokens_seen": 27890208,
      "step": 132155
    },
    {
      "epoch": 14.539053905390539,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.006313036291386843,
      "loss": 0.234,
      "num_input_tokens_seen": 27891360,
      "step": 132160
    },
    {
      "epoch": 14.53960396039604,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.006311862360528259,
      "loss": 0.233,
      "num_input_tokens_seen": 27892416,
      "step": 132165
    },
    {
      "epoch": 14.54015401540154,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.006310688509744077,
      "loss": 0.2319,
      "num_input_tokens_seen": 27893600,
      "step": 132170
    },
    {
      "epoch": 14.54070407040704,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.006309514739045103,
      "loss": 0.2309,
      "num_input_tokens_seen": 27894624,
      "step": 132175
    },
    {
      "epoch": 14.541254125412541,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.006308341048442168,
      "loss": 0.234,
      "num_input_tokens_seen": 27895744,
      "step": 132180
    },
    {
      "epoch": 14.541804180418042,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006307167437946078,
      "loss": 0.2335,
      "num_input_tokens_seen": 27896864,
      "step": 132185
    },
    {
      "epoch": 14.542354235423542,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.006305993907567653,
      "loss": 0.2309,
      "num_input_tokens_seen": 27897888,
      "step": 132190
    },
    {
      "epoch": 14.542904290429043,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.006304820457317717,
      "loss": 0.2288,
      "num_input_tokens_seen": 27898944,
      "step": 132195
    },
    {
      "epoch": 14.543454345434544,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0063036470872070705,
      "loss": 0.2304,
      "num_input_tokens_seen": 27900000,
      "step": 132200
    },
    {
      "epoch": 14.544004400440045,
      "grad_norm": 0.005859375,
      "learning_rate": 0.006302473797246545,
      "loss": 0.2345,
      "num_input_tokens_seen": 27901088,
      "step": 132205
    },
    {
      "epoch": 14.544554455445544,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006301300587446937,
      "loss": 0.2304,
      "num_input_tokens_seen": 27902144,
      "step": 132210
    },
    {
      "epoch": 14.545104510451045,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.006300127457819073,
      "loss": 0.2299,
      "num_input_tokens_seen": 27903232,
      "step": 132215
    },
    {
      "epoch": 14.545654565456546,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.006298954408373755,
      "loss": 0.2304,
      "num_input_tokens_seen": 27904320,
      "step": 132220
    },
    {
      "epoch": 14.546204620462046,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.0062977814391218,
      "loss": 0.2319,
      "num_input_tokens_seen": 27905408,
      "step": 132225
    },
    {
      "epoch": 14.546754675467547,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0062966085500740215,
      "loss": 0.2319,
      "num_input_tokens_seen": 27906400,
      "step": 132230
    },
    {
      "epoch": 14.547304730473048,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.006295435741241225,
      "loss": 0.2309,
      "num_input_tokens_seen": 27907456,
      "step": 132235
    },
    {
      "epoch": 14.547854785478547,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.006294263012634218,
      "loss": 0.2314,
      "num_input_tokens_seen": 27908480,
      "step": 132240
    },
    {
      "epoch": 14.548404840484048,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.006293090364263815,
      "loss": 0.2309,
      "num_input_tokens_seen": 27909664,
      "step": 132245
    },
    {
      "epoch": 14.54895489548955,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.006291917796140808,
      "loss": 0.2345,
      "num_input_tokens_seen": 27910752,
      "step": 132250
    },
    {
      "epoch": 14.549504950495049,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006290745308276028,
      "loss": 0.2334,
      "num_input_tokens_seen": 27911776,
      "step": 132255
    },
    {
      "epoch": 14.55005500550055,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006289572900680271,
      "loss": 0.2313,
      "num_input_tokens_seen": 27912800,
      "step": 132260
    },
    {
      "epoch": 14.55060506050605,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.006288400573364334,
      "loss": 0.2308,
      "num_input_tokens_seen": 27913952,
      "step": 132265
    },
    {
      "epoch": 14.551155115511552,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0062872283263390345,
      "loss": 0.2319,
      "num_input_tokens_seen": 27915040,
      "step": 132270
    },
    {
      "epoch": 14.551705170517051,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.006286056159615165,
      "loss": 0.2351,
      "num_input_tokens_seen": 27916064,
      "step": 132275
    },
    {
      "epoch": 14.552255225522552,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0062848840732035345,
      "loss": 0.2335,
      "num_input_tokens_seen": 27917120,
      "step": 132280
    },
    {
      "epoch": 14.552805280528053,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006283712067114951,
      "loss": 0.2277,
      "num_input_tokens_seen": 27918144,
      "step": 132285
    },
    {
      "epoch": 14.553355335533553,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0062825401413602044,
      "loss": 0.2314,
      "num_input_tokens_seen": 27919232,
      "step": 132290
    },
    {
      "epoch": 14.553905390539054,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00628136829595011,
      "loss": 0.233,
      "num_input_tokens_seen": 27920320,
      "step": 132295
    },
    {
      "epoch": 14.554455445544555,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0062801965308954535,
      "loss": 0.2298,
      "num_input_tokens_seen": 27921344,
      "step": 132300
    },
    {
      "epoch": 14.555005500550054,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.006279024846207043,
      "loss": 0.2298,
      "num_input_tokens_seen": 27922368,
      "step": 132305
    },
    {
      "epoch": 14.555555555555555,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.006277853241895679,
      "loss": 0.2293,
      "num_input_tokens_seen": 27923424,
      "step": 132310
    },
    {
      "epoch": 14.556105610561056,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.006276681717972152,
      "loss": 0.2319,
      "num_input_tokens_seen": 27924448,
      "step": 132315
    },
    {
      "epoch": 14.556655665566556,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.006275510274447269,
      "loss": 0.233,
      "num_input_tokens_seen": 27925504,
      "step": 132320
    },
    {
      "epoch": 14.557205720572057,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.006274338911331817,
      "loss": 0.2298,
      "num_input_tokens_seen": 27926560,
      "step": 132325
    },
    {
      "epoch": 14.557755775577558,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0062731676286366,
      "loss": 0.2308,
      "num_input_tokens_seen": 27927616,
      "step": 132330
    },
    {
      "epoch": 14.558305830583059,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.0062719964263724046,
      "loss": 0.2309,
      "num_input_tokens_seen": 27928736,
      "step": 132335
    },
    {
      "epoch": 14.558855885588558,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.006270825304550031,
      "loss": 0.2298,
      "num_input_tokens_seen": 27929888,
      "step": 132340
    },
    {
      "epoch": 14.55940594059406,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006269654263180275,
      "loss": 0.2319,
      "num_input_tokens_seen": 27930976,
      "step": 132345
    },
    {
      "epoch": 14.55995599559956,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.006268483302273924,
      "loss": 0.2329,
      "num_input_tokens_seen": 27932032,
      "step": 132350
    },
    {
      "epoch": 14.56050605060506,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0062673124218417765,
      "loss": 0.2298,
      "num_input_tokens_seen": 27933024,
      "step": 132355
    },
    {
      "epoch": 14.561056105610561,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.006266141621894619,
      "loss": 0.2303,
      "num_input_tokens_seen": 27934080,
      "step": 132360
    },
    {
      "epoch": 14.561606160616062,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006264970902443234,
      "loss": 0.2298,
      "num_input_tokens_seen": 27935168,
      "step": 132365
    },
    {
      "epoch": 14.562156215621561,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.006263800263498431,
      "loss": 0.2283,
      "num_input_tokens_seen": 27936224,
      "step": 132370
    },
    {
      "epoch": 14.562706270627062,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006262629705070983,
      "loss": 0.2309,
      "num_input_tokens_seen": 27937344,
      "step": 132375
    },
    {
      "epoch": 14.563256325632564,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.006261459227171691,
      "loss": 0.2314,
      "num_input_tokens_seen": 27938400,
      "step": 132380
    },
    {
      "epoch": 14.563806380638063,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006260288829811336,
      "loss": 0.2309,
      "num_input_tokens_seen": 27939424,
      "step": 132385
    },
    {
      "epoch": 14.564356435643564,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.006259118513000698,
      "loss": 0.2324,
      "num_input_tokens_seen": 27940448,
      "step": 132390
    },
    {
      "epoch": 14.564906490649065,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.006257948276750573,
      "loss": 0.2324,
      "num_input_tokens_seen": 27941536,
      "step": 132395
    },
    {
      "epoch": 14.565456545654566,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.006256778121071741,
      "loss": 0.2319,
      "num_input_tokens_seen": 27942624,
      "step": 132400
    },
    {
      "epoch": 14.566006600660065,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.006255608045974997,
      "loss": 0.2319,
      "num_input_tokens_seen": 27943680,
      "step": 132405
    },
    {
      "epoch": 14.566556655665567,
      "grad_norm": 0.00054931640625,
      "learning_rate": 0.006254438051471118,
      "loss": 0.2298,
      "num_input_tokens_seen": 27944672,
      "step": 132410
    },
    {
      "epoch": 14.567106710671068,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006253268137570881,
      "loss": 0.2314,
      "num_input_tokens_seen": 27945792,
      "step": 132415
    },
    {
      "epoch": 14.567656765676567,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006252098304285079,
      "loss": 0.2308,
      "num_input_tokens_seen": 27946816,
      "step": 132420
    },
    {
      "epoch": 14.568206820682068,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.00625092855162448,
      "loss": 0.2308,
      "num_input_tokens_seen": 27947808,
      "step": 132425
    },
    {
      "epoch": 14.56875687568757,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0062497588795998845,
      "loss": 0.2335,
      "num_input_tokens_seen": 27948832,
      "step": 132430
    },
    {
      "epoch": 14.569306930693068,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006248589288222062,
      "loss": 0.2314,
      "num_input_tokens_seen": 27949888,
      "step": 132435
    },
    {
      "epoch": 14.56985698569857,
      "grad_norm": 0.00079345703125,
      "learning_rate": 0.006247419777501788,
      "loss": 0.2298,
      "num_input_tokens_seen": 27950944,
      "step": 132440
    },
    {
      "epoch": 14.57040704070407,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.00624625034744985,
      "loss": 0.2304,
      "num_input_tokens_seen": 27952032,
      "step": 132445
    },
    {
      "epoch": 14.570957095709572,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006245080998077019,
      "loss": 0.2303,
      "num_input_tokens_seen": 27953120,
      "step": 132450
    },
    {
      "epoch": 14.571507150715071,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.006243911729394073,
      "loss": 0.2304,
      "num_input_tokens_seen": 27954272,
      "step": 132455
    },
    {
      "epoch": 14.572057205720572,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006242742541411798,
      "loss": 0.2308,
      "num_input_tokens_seen": 27955264,
      "step": 132460
    },
    {
      "epoch": 14.572607260726073,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006241573434140958,
      "loss": 0.2308,
      "num_input_tokens_seen": 27956320,
      "step": 132465
    },
    {
      "epoch": 14.573157315731573,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.006240404407592337,
      "loss": 0.2309,
      "num_input_tokens_seen": 27957344,
      "step": 132470
    },
    {
      "epoch": 14.573707370737074,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.006239235461776707,
      "loss": 0.2314,
      "num_input_tokens_seen": 27958400,
      "step": 132475
    },
    {
      "epoch": 14.574257425742575,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.006238066596704833,
      "loss": 0.2288,
      "num_input_tokens_seen": 27959520,
      "step": 132480
    },
    {
      "epoch": 14.574807480748074,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.006236897812387494,
      "loss": 0.2324,
      "num_input_tokens_seen": 27960640,
      "step": 132485
    },
    {
      "epoch": 14.575357535753575,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.006235729108835464,
      "loss": 0.2329,
      "num_input_tokens_seen": 27961760,
      "step": 132490
    },
    {
      "epoch": 14.575907590759076,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.00623456048605952,
      "loss": 0.2319,
      "num_input_tokens_seen": 27962816,
      "step": 132495
    },
    {
      "epoch": 14.576457645764577,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006233391944070424,
      "loss": 0.2303,
      "num_input_tokens_seen": 27963840,
      "step": 132500
    },
    {
      "epoch": 14.577007700770077,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.006232223482878944,
      "loss": 0.2319,
      "num_input_tokens_seen": 27964928,
      "step": 132505
    },
    {
      "epoch": 14.577557755775578,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0062310551024958504,
      "loss": 0.2329,
      "num_input_tokens_seen": 27965920,
      "step": 132510
    },
    {
      "epoch": 14.578107810781079,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.006229886802931918,
      "loss": 0.2288,
      "num_input_tokens_seen": 27966912,
      "step": 132515
    },
    {
      "epoch": 14.578657865786578,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006228718584197913,
      "loss": 0.2309,
      "num_input_tokens_seen": 27967936,
      "step": 132520
    },
    {
      "epoch": 14.57920792079208,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0062275504463046,
      "loss": 0.2303,
      "num_input_tokens_seen": 27968992,
      "step": 132525
    },
    {
      "epoch": 14.57975797579758,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.006226382389262741,
      "loss": 0.2329,
      "num_input_tokens_seen": 27970048,
      "step": 132530
    },
    {
      "epoch": 14.58030803080308,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.00622521441308311,
      "loss": 0.2329,
      "num_input_tokens_seen": 27971104,
      "step": 132535
    },
    {
      "epoch": 14.58085808580858,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006224046517776463,
      "loss": 0.2324,
      "num_input_tokens_seen": 27972160,
      "step": 132540
    },
    {
      "epoch": 14.581408140814082,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.006222878703353567,
      "loss": 0.2314,
      "num_input_tokens_seen": 27973280,
      "step": 132545
    },
    {
      "epoch": 14.581958195819581,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.006221710969825192,
      "loss": 0.2335,
      "num_input_tokens_seen": 27974304,
      "step": 132550
    },
    {
      "epoch": 14.582508250825082,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0062205433172020895,
      "loss": 0.2345,
      "num_input_tokens_seen": 27975328,
      "step": 132555
    },
    {
      "epoch": 14.583058305830583,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.006219375745495032,
      "loss": 0.2329,
      "num_input_tokens_seen": 27976352,
      "step": 132560
    },
    {
      "epoch": 14.583608360836084,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.0062182082547147695,
      "loss": 0.2298,
      "num_input_tokens_seen": 27977376,
      "step": 132565
    },
    {
      "epoch": 14.584158415841584,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.006217040844872067,
      "loss": 0.2314,
      "num_input_tokens_seen": 27978464,
      "step": 132570
    },
    {
      "epoch": 14.584708470847085,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.0062158735159776905,
      "loss": 0.2314,
      "num_input_tokens_seen": 27979488,
      "step": 132575
    },
    {
      "epoch": 14.585258525852586,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.006214706268042388,
      "loss": 0.2303,
      "num_input_tokens_seen": 27980512,
      "step": 132580
    },
    {
      "epoch": 14.585808580858085,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.006213539101076926,
      "loss": 0.2319,
      "num_input_tokens_seen": 27981568,
      "step": 132585
    },
    {
      "epoch": 14.586358635863586,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.006212372015092053,
      "loss": 0.2309,
      "num_input_tokens_seen": 27982688,
      "step": 132590
    },
    {
      "epoch": 14.586908690869087,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0062112050100985375,
      "loss": 0.2288,
      "num_input_tokens_seen": 27983776,
      "step": 132595
    },
    {
      "epoch": 14.587458745874587,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.006210038086107122,
      "loss": 0.2303,
      "num_input_tokens_seen": 27984832,
      "step": 132600
    },
    {
      "epoch": 14.588008800880088,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.006208871243128565,
      "loss": 0.2324,
      "num_input_tokens_seen": 27985888,
      "step": 132605
    },
    {
      "epoch": 14.588558855885589,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006207704481173633,
      "loss": 0.2314,
      "num_input_tokens_seen": 27987040,
      "step": 132610
    },
    {
      "epoch": 14.589108910891088,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0062065378002530616,
      "loss": 0.2314,
      "num_input_tokens_seen": 27988128,
      "step": 132615
    },
    {
      "epoch": 14.58965896589659,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0062053712003776185,
      "loss": 0.2313,
      "num_input_tokens_seen": 27989184,
      "step": 132620
    },
    {
      "epoch": 14.59020902090209,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.006204204681558041,
      "loss": 0.2314,
      "num_input_tokens_seen": 27990240,
      "step": 132625
    },
    {
      "epoch": 14.590759075907592,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.0062030382438050926,
      "loss": 0.2335,
      "num_input_tokens_seen": 27991264,
      "step": 132630
    },
    {
      "epoch": 14.591309130913091,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.00620187188712952,
      "loss": 0.2319,
      "num_input_tokens_seen": 27992352,
      "step": 132635
    },
    {
      "epoch": 14.591859185918592,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.006200705611542077,
      "loss": 0.2345,
      "num_input_tokens_seen": 27993440,
      "step": 132640
    },
    {
      "epoch": 14.592409240924093,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.006199539417053501,
      "loss": 0.2324,
      "num_input_tokens_seen": 27994496,
      "step": 132645
    },
    {
      "epoch": 14.592959295929592,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006198373303674553,
      "loss": 0.2283,
      "num_input_tokens_seen": 27995584,
      "step": 132650
    },
    {
      "epoch": 14.593509350935093,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0061972072714159675,
      "loss": 0.2334,
      "num_input_tokens_seen": 27996576,
      "step": 132655
    },
    {
      "epoch": 14.594059405940595,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.006196041320288499,
      "loss": 0.2304,
      "num_input_tokens_seen": 27997696,
      "step": 132660
    },
    {
      "epoch": 14.594609460946094,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006194875450302901,
      "loss": 0.2314,
      "num_input_tokens_seen": 27998720,
      "step": 132665
    },
    {
      "epoch": 14.595159515951595,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.006193709661469904,
      "loss": 0.2319,
      "num_input_tokens_seen": 27999776,
      "step": 132670
    },
    {
      "epoch": 14.595709570957096,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.006192543953800265,
      "loss": 0.2309,
      "num_input_tokens_seen": 28000832,
      "step": 132675
    },
    {
      "epoch": 14.596259625962595,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006191378327304715,
      "loss": 0.2319,
      "num_input_tokens_seen": 28001888,
      "step": 132680
    },
    {
      "epoch": 14.596809680968097,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.006190212781994007,
      "loss": 0.2303,
      "num_input_tokens_seen": 28002944,
      "step": 132685
    },
    {
      "epoch": 14.597359735973598,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.006189047317878884,
      "loss": 0.2324,
      "num_input_tokens_seen": 28004000,
      "step": 132690
    },
    {
      "epoch": 14.597909790979099,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.006187881934970078,
      "loss": 0.2329,
      "num_input_tokens_seen": 28005088,
      "step": 132695
    },
    {
      "epoch": 14.598459845984598,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.006186716633278344,
      "loss": 0.2314,
      "num_input_tokens_seen": 28006176,
      "step": 132700
    },
    {
      "epoch": 14.599009900990099,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.006185551412814405,
      "loss": 0.2314,
      "num_input_tokens_seen": 28007328,
      "step": 132705
    },
    {
      "epoch": 14.5995599559956,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006184386273589016,
      "loss": 0.2314,
      "num_input_tokens_seen": 28008448,
      "step": 132710
    },
    {
      "epoch": 14.6001100110011,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.006183221215612904,
      "loss": 0.2314,
      "num_input_tokens_seen": 28009472,
      "step": 132715
    },
    {
      "epoch": 14.6006600660066,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.006182056238896811,
      "loss": 0.2329,
      "num_input_tokens_seen": 28010464,
      "step": 132720
    },
    {
      "epoch": 14.601210121012102,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006180891343451478,
      "loss": 0.2324,
      "num_input_tokens_seen": 28011488,
      "step": 132725
    },
    {
      "epoch": 14.601760176017601,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.006179726529287635,
      "loss": 0.2324,
      "num_input_tokens_seen": 28012576,
      "step": 132730
    },
    {
      "epoch": 14.602310231023102,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006178561796416023,
      "loss": 0.2319,
      "num_input_tokens_seen": 28013664,
      "step": 132735
    },
    {
      "epoch": 14.602860286028603,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.0061773971448473705,
      "loss": 0.2335,
      "num_input_tokens_seen": 28014720,
      "step": 132740
    },
    {
      "epoch": 14.603410341034103,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006176232574592414,
      "loss": 0.2314,
      "num_input_tokens_seen": 28015712,
      "step": 132745
    },
    {
      "epoch": 14.603960396039604,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006175068085661894,
      "loss": 0.233,
      "num_input_tokens_seen": 28016800,
      "step": 132750
    },
    {
      "epoch": 14.604510451045105,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00617390367806653,
      "loss": 0.2319,
      "num_input_tokens_seen": 28017856,
      "step": 132755
    },
    {
      "epoch": 14.605060506050606,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.006172739351817068,
      "loss": 0.2309,
      "num_input_tokens_seen": 28018976,
      "step": 132760
    },
    {
      "epoch": 14.605610561056105,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.006171575106924229,
      "loss": 0.2314,
      "num_input_tokens_seen": 28020064,
      "step": 132765
    },
    {
      "epoch": 14.606160616061606,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.006170410943398743,
      "loss": 0.2314,
      "num_input_tokens_seen": 28021056,
      "step": 132770
    },
    {
      "epoch": 14.606710671067107,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.006169246861251342,
      "loss": 0.2309,
      "num_input_tokens_seen": 28022112,
      "step": 132775
    },
    {
      "epoch": 14.607260726072607,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.006168082860492755,
      "loss": 0.2314,
      "num_input_tokens_seen": 28023136,
      "step": 132780
    },
    {
      "epoch": 14.607810781078108,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.006166918941133716,
      "loss": 0.2319,
      "num_input_tokens_seen": 28024128,
      "step": 132785
    },
    {
      "epoch": 14.608360836083609,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.006165755103184946,
      "loss": 0.2319,
      "num_input_tokens_seen": 28025184,
      "step": 132790
    },
    {
      "epoch": 14.608910891089108,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.006164591346657167,
      "loss": 0.2304,
      "num_input_tokens_seen": 28026272,
      "step": 132795
    },
    {
      "epoch": 14.60946094609461,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006163427671561109,
      "loss": 0.2319,
      "num_input_tokens_seen": 28027328,
      "step": 132800
    },
    {
      "epoch": 14.61001100110011,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0061622640779075005,
      "loss": 0.2303,
      "num_input_tokens_seen": 28028352,
      "step": 132805
    },
    {
      "epoch": 14.61056105610561,
      "grad_norm": 0.000774383544921875,
      "learning_rate": 0.0061611005657070655,
      "loss": 0.2309,
      "num_input_tokens_seen": 28029408,
      "step": 132810
    },
    {
      "epoch": 14.61111111111111,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.006159937134970527,
      "loss": 0.2308,
      "num_input_tokens_seen": 28030400,
      "step": 132815
    },
    {
      "epoch": 14.611661166116612,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.0061587737857086,
      "loss": 0.2324,
      "num_input_tokens_seen": 28031456,
      "step": 132820
    },
    {
      "epoch": 14.612211221122113,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0061576105179320184,
      "loss": 0.2298,
      "num_input_tokens_seen": 28032512,
      "step": 132825
    },
    {
      "epoch": 14.612761276127612,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.006156447331651492,
      "loss": 0.2319,
      "num_input_tokens_seen": 28033504,
      "step": 132830
    },
    {
      "epoch": 14.613311331133113,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.006155284226877748,
      "loss": 0.2314,
      "num_input_tokens_seen": 28034528,
      "step": 132835
    },
    {
      "epoch": 14.613861386138614,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.006154121203621509,
      "loss": 0.2303,
      "num_input_tokens_seen": 28035616,
      "step": 132840
    },
    {
      "epoch": 14.614411441144114,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006152958261893487,
      "loss": 0.2309,
      "num_input_tokens_seen": 28036672,
      "step": 132845
    },
    {
      "epoch": 14.614961496149615,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.006151795401704407,
      "loss": 0.2303,
      "num_input_tokens_seen": 28037728,
      "step": 132850
    },
    {
      "epoch": 14.615511551155116,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006150632623064979,
      "loss": 0.2293,
      "num_input_tokens_seen": 28038784,
      "step": 132855
    },
    {
      "epoch": 14.616061606160617,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006149469925985921,
      "loss": 0.2304,
      "num_input_tokens_seen": 28039840,
      "step": 132860
    },
    {
      "epoch": 14.616611661166116,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006148307310477961,
      "loss": 0.2309,
      "num_input_tokens_seen": 28040896,
      "step": 132865
    },
    {
      "epoch": 14.617161716171617,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.006147144776551796,
      "loss": 0.2304,
      "num_input_tokens_seen": 28041888,
      "step": 132870
    },
    {
      "epoch": 14.617711771177119,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.006145982324218158,
      "loss": 0.2314,
      "num_input_tokens_seen": 28042912,
      "step": 132875
    },
    {
      "epoch": 14.618261826182618,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00614481995348775,
      "loss": 0.2308,
      "num_input_tokens_seen": 28044000,
      "step": 132880
    },
    {
      "epoch": 14.618811881188119,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006143657664371283,
      "loss": 0.2303,
      "num_input_tokens_seen": 28045088,
      "step": 132885
    },
    {
      "epoch": 14.61936193619362,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006142495456879475,
      "loss": 0.2319,
      "num_input_tokens_seen": 28046080,
      "step": 132890
    },
    {
      "epoch": 14.61991199119912,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.006141333331023033,
      "loss": 0.2308,
      "num_input_tokens_seen": 28047136,
      "step": 132895
    },
    {
      "epoch": 14.62046204620462,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.0061401712868126785,
      "loss": 0.2314,
      "num_input_tokens_seen": 28048224,
      "step": 132900
    },
    {
      "epoch": 14.621012101210122,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.006139009324259114,
      "loss": 0.2319,
      "num_input_tokens_seen": 28049280,
      "step": 132905
    },
    {
      "epoch": 14.62156215621562,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.006137847443373042,
      "loss": 0.2335,
      "num_input_tokens_seen": 28050336,
      "step": 132910
    },
    {
      "epoch": 14.622112211221122,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006136685644165179,
      "loss": 0.2309,
      "num_input_tokens_seen": 28051424,
      "step": 132915
    },
    {
      "epoch": 14.622662266226623,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.0061355239266462315,
      "loss": 0.2304,
      "num_input_tokens_seen": 28052480,
      "step": 132920
    },
    {
      "epoch": 14.623212321232124,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00613436229082691,
      "loss": 0.2335,
      "num_input_tokens_seen": 28053504,
      "step": 132925
    },
    {
      "epoch": 14.623762376237623,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0061332007367179186,
      "loss": 0.2303,
      "num_input_tokens_seen": 28054528,
      "step": 132930
    },
    {
      "epoch": 14.624312431243125,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006132039264329954,
      "loss": 0.2293,
      "num_input_tokens_seen": 28055584,
      "step": 132935
    },
    {
      "epoch": 14.624862486248626,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006130877873673736,
      "loss": 0.2319,
      "num_input_tokens_seen": 28056704,
      "step": 132940
    },
    {
      "epoch": 14.625412541254125,
      "grad_norm": 0.005859375,
      "learning_rate": 0.006129716564759954,
      "loss": 0.2308,
      "num_input_tokens_seen": 28057792,
      "step": 132945
    },
    {
      "epoch": 14.625962596259626,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.006128555337599319,
      "loss": 0.2293,
      "num_input_tokens_seen": 28058848,
      "step": 132950
    },
    {
      "epoch": 14.626512651265127,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.006127394192202537,
      "loss": 0.2319,
      "num_input_tokens_seen": 28059904,
      "step": 132955
    },
    {
      "epoch": 14.627062706270626,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006126233128580302,
      "loss": 0.2303,
      "num_input_tokens_seen": 28060928,
      "step": 132960
    },
    {
      "epoch": 14.627612761276128,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.006125072146743321,
      "loss": 0.2309,
      "num_input_tokens_seen": 28062016,
      "step": 132965
    },
    {
      "epoch": 14.628162816281629,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.006123911246702287,
      "loss": 0.2335,
      "num_input_tokens_seen": 28063008,
      "step": 132970
    },
    {
      "epoch": 14.628712871287128,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.006122750428467902,
      "loss": 0.2298,
      "num_input_tokens_seen": 28064032,
      "step": 132975
    },
    {
      "epoch": 14.629262926292629,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.006121589692050875,
      "loss": 0.2304,
      "num_input_tokens_seen": 28065088,
      "step": 132980
    },
    {
      "epoch": 14.62981298129813,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.00612042903746189,
      "loss": 0.2319,
      "num_input_tokens_seen": 28066176,
      "step": 132985
    },
    {
      "epoch": 14.630363036303631,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0061192684647116535,
      "loss": 0.2303,
      "num_input_tokens_seen": 28067264,
      "step": 132990
    },
    {
      "epoch": 14.63091309130913,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0061181079738108525,
      "loss": 0.2309,
      "num_input_tokens_seen": 28068352,
      "step": 132995
    },
    {
      "epoch": 14.631463146314632,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.006116947564770194,
      "loss": 0.233,
      "num_input_tokens_seen": 28069504,
      "step": 133000
    },
    {
      "epoch": 14.632013201320133,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.006115787237600364,
      "loss": 0.2308,
      "num_input_tokens_seen": 28070528,
      "step": 133005
    },
    {
      "epoch": 14.632563256325632,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006114626992312057,
      "loss": 0.2314,
      "num_input_tokens_seen": 28071552,
      "step": 133010
    },
    {
      "epoch": 14.633113311331133,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006113466828915976,
      "loss": 0.2304,
      "num_input_tokens_seen": 28072640,
      "step": 133015
    },
    {
      "epoch": 14.633663366336634,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.006112306747422808,
      "loss": 0.2314,
      "num_input_tokens_seen": 28073728,
      "step": 133020
    },
    {
      "epoch": 14.634213421342134,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.006111146747843238,
      "loss": 0.2314,
      "num_input_tokens_seen": 28074848,
      "step": 133025
    },
    {
      "epoch": 14.634763476347635,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.006109986830187969,
      "loss": 0.2309,
      "num_input_tokens_seen": 28075968,
      "step": 133030
    },
    {
      "epoch": 14.635313531353136,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006108826994467673,
      "loss": 0.2278,
      "num_input_tokens_seen": 28077024,
      "step": 133035
    },
    {
      "epoch": 14.635863586358635,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006107667240693064,
      "loss": 0.2324,
      "num_input_tokens_seen": 28078112,
      "step": 133040
    },
    {
      "epoch": 14.636413641364136,
      "grad_norm": 0.000621795654296875,
      "learning_rate": 0.00610650756887482,
      "loss": 0.2298,
      "num_input_tokens_seen": 28079168,
      "step": 133045
    },
    {
      "epoch": 14.636963696369637,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.006105347979023621,
      "loss": 0.2324,
      "num_input_tokens_seen": 28080256,
      "step": 133050
    },
    {
      "epoch": 14.637513751375138,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006104188471150168,
      "loss": 0.2314,
      "num_input_tokens_seen": 28081344,
      "step": 133055
    },
    {
      "epoch": 14.638063806380638,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.0061030290452651365,
      "loss": 0.2324,
      "num_input_tokens_seen": 28082400,
      "step": 133060
    },
    {
      "epoch": 14.638613861386139,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.006101869701379218,
      "loss": 0.2324,
      "num_input_tokens_seen": 28083520,
      "step": 133065
    },
    {
      "epoch": 14.63916391639164,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.006100710439503099,
      "loss": 0.2324,
      "num_input_tokens_seen": 28084544,
      "step": 133070
    },
    {
      "epoch": 14.63971397139714,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.006099551259647459,
      "loss": 0.233,
      "num_input_tokens_seen": 28085600,
      "step": 133075
    },
    {
      "epoch": 14.64026402640264,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0060983921618229875,
      "loss": 0.2319,
      "num_input_tokens_seen": 28086656,
      "step": 133080
    },
    {
      "epoch": 14.640814081408141,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.00609723314604036,
      "loss": 0.2329,
      "num_input_tokens_seen": 28087680,
      "step": 133085
    },
    {
      "epoch": 14.64136413641364,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.006096074212310261,
      "loss": 0.2319,
      "num_input_tokens_seen": 28088768,
      "step": 133090
    },
    {
      "epoch": 14.641914191419142,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006094915360643381,
      "loss": 0.2308,
      "num_input_tokens_seen": 28089856,
      "step": 133095
    },
    {
      "epoch": 14.642464246424643,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.006093756591050387,
      "loss": 0.2309,
      "num_input_tokens_seen": 28090912,
      "step": 133100
    },
    {
      "epoch": 14.643014301430142,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00609259790354197,
      "loss": 0.2319,
      "num_input_tokens_seen": 28091936,
      "step": 133105
    },
    {
      "epoch": 14.643564356435643,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.006091439298128799,
      "loss": 0.2314,
      "num_input_tokens_seen": 28092928,
      "step": 133110
    },
    {
      "epoch": 14.644114411441144,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.006090280774821561,
      "loss": 0.2288,
      "num_input_tokens_seen": 28094016,
      "step": 133115
    },
    {
      "epoch": 14.644664466446645,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.006089122333630926,
      "loss": 0.2329,
      "num_input_tokens_seen": 28095008,
      "step": 133120
    },
    {
      "epoch": 14.645214521452145,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.006087963974567573,
      "loss": 0.2319,
      "num_input_tokens_seen": 28096032,
      "step": 133125
    },
    {
      "epoch": 14.645764576457646,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006086805697642187,
      "loss": 0.2298,
      "num_input_tokens_seen": 28097088,
      "step": 133130
    },
    {
      "epoch": 14.646314631463147,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.0060856475028654294,
      "loss": 0.2303,
      "num_input_tokens_seen": 28098144,
      "step": 133135
    },
    {
      "epoch": 14.646864686468646,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.006084489390247987,
      "loss": 0.2319,
      "num_input_tokens_seen": 28099232,
      "step": 133140
    },
    {
      "epoch": 14.647414741474147,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.006083331359800528,
      "loss": 0.2319,
      "num_input_tokens_seen": 28100352,
      "step": 133145
    },
    {
      "epoch": 14.647964796479648,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006082173411533714,
      "loss": 0.2303,
      "num_input_tokens_seen": 28101408,
      "step": 133150
    },
    {
      "epoch": 14.648514851485148,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.006081015545458241,
      "loss": 0.2314,
      "num_input_tokens_seen": 28102464,
      "step": 133155
    },
    {
      "epoch": 14.649064906490649,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.00607985776158476,
      "loss": 0.2308,
      "num_input_tokens_seen": 28103456,
      "step": 133160
    },
    {
      "epoch": 14.64961496149615,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.006078700059923955,
      "loss": 0.2309,
      "num_input_tokens_seen": 28104448,
      "step": 133165
    },
    {
      "epoch": 14.65016501650165,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.006077542440486492,
      "loss": 0.2308,
      "num_input_tokens_seen": 28105472,
      "step": 133170
    },
    {
      "epoch": 14.65071507150715,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006076384903283034,
      "loss": 0.2324,
      "num_input_tokens_seen": 28106464,
      "step": 133175
    },
    {
      "epoch": 14.651265126512651,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0060752274483242535,
      "loss": 0.2314,
      "num_input_tokens_seen": 28107520,
      "step": 133180
    },
    {
      "epoch": 14.651815181518153,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0060740700756208195,
      "loss": 0.2309,
      "num_input_tokens_seen": 28108704,
      "step": 133185
    },
    {
      "epoch": 14.652365236523652,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.006072912785183403,
      "loss": 0.2298,
      "num_input_tokens_seen": 28109792,
      "step": 133190
    },
    {
      "epoch": 14.652915291529153,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006071755577022667,
      "loss": 0.2308,
      "num_input_tokens_seen": 28110848,
      "step": 133195
    },
    {
      "epoch": 14.653465346534654,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.006070598451149271,
      "loss": 0.2303,
      "num_input_tokens_seen": 28111936,
      "step": 133200
    },
    {
      "epoch": 14.654015401540153,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.006069441407573886,
      "loss": 0.2293,
      "num_input_tokens_seen": 28112992,
      "step": 133205
    },
    {
      "epoch": 14.654565456545654,
      "grad_norm": 0.005859375,
      "learning_rate": 0.006068284446307173,
      "loss": 0.2319,
      "num_input_tokens_seen": 28114016,
      "step": 133210
    },
    {
      "epoch": 14.655115511551156,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.006067127567359793,
      "loss": 0.2293,
      "num_input_tokens_seen": 28115104,
      "step": 133215
    },
    {
      "epoch": 14.655665566556655,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00606597077074242,
      "loss": 0.2288,
      "num_input_tokens_seen": 28116160,
      "step": 133220
    },
    {
      "epoch": 14.656215621562156,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.006064814056465701,
      "loss": 0.2329,
      "num_input_tokens_seen": 28117184,
      "step": 133225
    },
    {
      "epoch": 14.656765676567657,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006063657424540308,
      "loss": 0.2319,
      "num_input_tokens_seen": 28118272,
      "step": 133230
    },
    {
      "epoch": 14.657315731573158,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.006062500874976892,
      "loss": 0.2325,
      "num_input_tokens_seen": 28119360,
      "step": 133235
    },
    {
      "epoch": 14.657865786578657,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.006061344407786117,
      "loss": 0.2303,
      "num_input_tokens_seen": 28120448,
      "step": 133240
    },
    {
      "epoch": 14.658415841584159,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.006060188022978645,
      "loss": 0.2298,
      "num_input_tokens_seen": 28121536,
      "step": 133245
    },
    {
      "epoch": 14.65896589658966,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.006059031720565126,
      "loss": 0.2324,
      "num_input_tokens_seen": 28122592,
      "step": 133250
    },
    {
      "epoch": 14.659515951595159,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0060578755005562274,
      "loss": 0.2314,
      "num_input_tokens_seen": 28123648,
      "step": 133255
    },
    {
      "epoch": 14.66006600660066,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0060567193629626,
      "loss": 0.2314,
      "num_input_tokens_seen": 28124800,
      "step": 133260
    },
    {
      "epoch": 14.660616061606161,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006055563307794892,
      "loss": 0.2314,
      "num_input_tokens_seen": 28125856,
      "step": 133265
    },
    {
      "epoch": 14.66116611661166,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.006054407335063764,
      "loss": 0.2329,
      "num_input_tokens_seen": 28126912,
      "step": 133270
    },
    {
      "epoch": 14.661716171617162,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0060532514447798735,
      "loss": 0.2319,
      "num_input_tokens_seen": 28127968,
      "step": 133275
    },
    {
      "epoch": 14.662266226622663,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.0060520956369538755,
      "loss": 0.2308,
      "num_input_tokens_seen": 28129056,
      "step": 133280
    },
    {
      "epoch": 14.662816281628164,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.00605093991159642,
      "loss": 0.2319,
      "num_input_tokens_seen": 28130080,
      "step": 133285
    },
    {
      "epoch": 14.663366336633663,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.006049784268718149,
      "loss": 0.2335,
      "num_input_tokens_seen": 28131104,
      "step": 133290
    },
    {
      "epoch": 14.663916391639164,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.006048628708329724,
      "loss": 0.2314,
      "num_input_tokens_seen": 28132256,
      "step": 133295
    },
    {
      "epoch": 14.664466446644665,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006047473230441792,
      "loss": 0.2319,
      "num_input_tokens_seen": 28133312,
      "step": 133300
    },
    {
      "epoch": 14.665016501650165,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.006046317835065008,
      "loss": 0.2314,
      "num_input_tokens_seen": 28134368,
      "step": 133305
    },
    {
      "epoch": 14.665566556655666,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.006045162522210018,
      "loss": 0.2298,
      "num_input_tokens_seen": 28135392,
      "step": 133310
    },
    {
      "epoch": 14.666116611661167,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.006044007291887461,
      "loss": 0.2319,
      "num_input_tokens_seen": 28136480,
      "step": 133315
    },
    {
      "epoch": 14.666666666666666,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.006042852144107997,
      "loss": 0.2304,
      "num_input_tokens_seen": 28137440,
      "step": 133320
    },
    {
      "epoch": 14.667216721672167,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.006041697078882262,
      "loss": 0.2324,
      "num_input_tokens_seen": 28138592,
      "step": 133325
    },
    {
      "epoch": 14.667766776677668,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0060405420962209065,
      "loss": 0.2288,
      "num_input_tokens_seen": 28139584,
      "step": 133330
    },
    {
      "epoch": 14.668316831683168,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00603938719613458,
      "loss": 0.2303,
      "num_input_tokens_seen": 28140704,
      "step": 133335
    },
    {
      "epoch": 14.668866886688669,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.006038232378633916,
      "loss": 0.2293,
      "num_input_tokens_seen": 28141792,
      "step": 133340
    },
    {
      "epoch": 14.66941694169417,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00603707764372957,
      "loss": 0.233,
      "num_input_tokens_seen": 28142880,
      "step": 133345
    },
    {
      "epoch": 14.66996699669967,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.006035922991432173,
      "loss": 0.2319,
      "num_input_tokens_seen": 28143904,
      "step": 133350
    },
    {
      "epoch": 14.67051705170517,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.006034768421752373,
      "loss": 0.2314,
      "num_input_tokens_seen": 28145024,
      "step": 133355
    },
    {
      "epoch": 14.671067106710671,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006033613934700817,
      "loss": 0.2298,
      "num_input_tokens_seen": 28146080,
      "step": 133360
    },
    {
      "epoch": 14.671617161716172,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0060324595302881315,
      "loss": 0.2314,
      "num_input_tokens_seen": 28147072,
      "step": 133365
    },
    {
      "epoch": 14.672167216721672,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.00603130520852497,
      "loss": 0.2319,
      "num_input_tokens_seen": 28148064,
      "step": 133370
    },
    {
      "epoch": 14.672717271727173,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.006030150969421959,
      "loss": 0.2304,
      "num_input_tokens_seen": 28149152,
      "step": 133375
    },
    {
      "epoch": 14.673267326732674,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.006028996812989746,
      "loss": 0.2309,
      "num_input_tokens_seen": 28150240,
      "step": 133380
    },
    {
      "epoch": 14.673817381738173,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.006027842739238962,
      "loss": 0.233,
      "num_input_tokens_seen": 28151264,
      "step": 133385
    },
    {
      "epoch": 14.674367436743674,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006026688748180244,
      "loss": 0.2309,
      "num_input_tokens_seen": 28152288,
      "step": 133390
    },
    {
      "epoch": 14.674917491749175,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006025534839824238,
      "loss": 0.233,
      "num_input_tokens_seen": 28153376,
      "step": 133395
    },
    {
      "epoch": 14.675467546754675,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.00602438101418157,
      "loss": 0.2335,
      "num_input_tokens_seen": 28154496,
      "step": 133400
    },
    {
      "epoch": 14.676017601760176,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.006023227271262869,
      "loss": 0.2324,
      "num_input_tokens_seen": 28155584,
      "step": 133405
    },
    {
      "epoch": 14.676567656765677,
      "grad_norm": 0.005859375,
      "learning_rate": 0.006022073611078774,
      "loss": 0.2325,
      "num_input_tokens_seen": 28156608,
      "step": 133410
    },
    {
      "epoch": 14.677117711771178,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006020920033639919,
      "loss": 0.2314,
      "num_input_tokens_seen": 28157664,
      "step": 133415
    },
    {
      "epoch": 14.677667766776677,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.006019766538956942,
      "loss": 0.2309,
      "num_input_tokens_seen": 28158720,
      "step": 133420
    },
    {
      "epoch": 14.678217821782178,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.006018613127040468,
      "loss": 0.2314,
      "num_input_tokens_seen": 28159776,
      "step": 133425
    },
    {
      "epoch": 14.67876787678768,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00601745979790112,
      "loss": 0.2309,
      "num_input_tokens_seen": 28160896,
      "step": 133430
    },
    {
      "epoch": 14.679317931793179,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.006016306551549541,
      "loss": 0.2314,
      "num_input_tokens_seen": 28161888,
      "step": 133435
    },
    {
      "epoch": 14.67986798679868,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00601515338799635,
      "loss": 0.2314,
      "num_input_tokens_seen": 28162912,
      "step": 133440
    },
    {
      "epoch": 14.680418041804181,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.006014000307252175,
      "loss": 0.2324,
      "num_input_tokens_seen": 28164000,
      "step": 133445
    },
    {
      "epoch": 14.68096809680968,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.006012847309327656,
      "loss": 0.2303,
      "num_input_tokens_seen": 28165056,
      "step": 133450
    },
    {
      "epoch": 14.681518151815181,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.006011694394233404,
      "loss": 0.2304,
      "num_input_tokens_seen": 28166112,
      "step": 133455
    },
    {
      "epoch": 14.682068206820682,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.006010541561980058,
      "loss": 0.2303,
      "num_input_tokens_seen": 28167104,
      "step": 133460
    },
    {
      "epoch": 14.682618261826182,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.006009388812578231,
      "loss": 0.2329,
      "num_input_tokens_seen": 28168224,
      "step": 133465
    },
    {
      "epoch": 14.683168316831683,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.006008236146038553,
      "loss": 0.2309,
      "num_input_tokens_seen": 28169248,
      "step": 133470
    },
    {
      "epoch": 14.683718371837184,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0060070835623716535,
      "loss": 0.2293,
      "num_input_tokens_seen": 28170368,
      "step": 133475
    },
    {
      "epoch": 14.684268426842685,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.006005931061588145,
      "loss": 0.2309,
      "num_input_tokens_seen": 28171424,
      "step": 133480
    },
    {
      "epoch": 14.684818481848184,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.006004778643698657,
      "loss": 0.2314,
      "num_input_tokens_seen": 28172384,
      "step": 133485
    },
    {
      "epoch": 14.685368536853685,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.006003626308713806,
      "loss": 0.2293,
      "num_input_tokens_seen": 28173408,
      "step": 133490
    },
    {
      "epoch": 14.685918591859187,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.006002474056644218,
      "loss": 0.2308,
      "num_input_tokens_seen": 28174528,
      "step": 133495
    },
    {
      "epoch": 14.686468646864686,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.006001321887500504,
      "loss": 0.2335,
      "num_input_tokens_seen": 28175520,
      "step": 133500
    },
    {
      "epoch": 14.687018701870187,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0060001698012932874,
      "loss": 0.2298,
      "num_input_tokens_seen": 28176576,
      "step": 133505
    },
    {
      "epoch": 14.687568756875688,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.005999017798033192,
      "loss": 0.2324,
      "num_input_tokens_seen": 28177696,
      "step": 133510
    },
    {
      "epoch": 14.688118811881187,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005997865877730828,
      "loss": 0.233,
      "num_input_tokens_seen": 28178784,
      "step": 133515
    },
    {
      "epoch": 14.688668866886688,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.005996714040396817,
      "loss": 0.2319,
      "num_input_tokens_seen": 28179808,
      "step": 133520
    },
    {
      "epoch": 14.68921892189219,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.005995562286041768,
      "loss": 0.2314,
      "num_input_tokens_seen": 28180928,
      "step": 133525
    },
    {
      "epoch": 14.689768976897689,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0059944106146763015,
      "loss": 0.2324,
      "num_input_tokens_seen": 28181984,
      "step": 133530
    },
    {
      "epoch": 14.69031903190319,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005993259026311036,
      "loss": 0.2298,
      "num_input_tokens_seen": 28183104,
      "step": 133535
    },
    {
      "epoch": 14.690869086908691,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.005992107520956574,
      "loss": 0.2303,
      "num_input_tokens_seen": 28184192,
      "step": 133540
    },
    {
      "epoch": 14.691419141914192,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005990956098623542,
      "loss": 0.2303,
      "num_input_tokens_seen": 28185216,
      "step": 133545
    },
    {
      "epoch": 14.691969196919691,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005989804759322543,
      "loss": 0.2319,
      "num_input_tokens_seen": 28186240,
      "step": 133550
    },
    {
      "epoch": 14.692519251925193,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.005988653503064184,
      "loss": 0.2293,
      "num_input_tokens_seen": 28187264,
      "step": 133555
    },
    {
      "epoch": 14.693069306930694,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.005987502329859084,
      "loss": 0.2309,
      "num_input_tokens_seen": 28188320,
      "step": 133560
    },
    {
      "epoch": 14.693619361936193,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005986351239717849,
      "loss": 0.2298,
      "num_input_tokens_seen": 28189472,
      "step": 133565
    },
    {
      "epoch": 14.694169416941694,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.005985200232651095,
      "loss": 0.2298,
      "num_input_tokens_seen": 28190560,
      "step": 133570
    },
    {
      "epoch": 14.694719471947195,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005984049308669424,
      "loss": 0.2324,
      "num_input_tokens_seen": 28191552,
      "step": 133575
    },
    {
      "epoch": 14.695269526952695,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.005982898467783439,
      "loss": 0.2319,
      "num_input_tokens_seen": 28192640,
      "step": 133580
    },
    {
      "epoch": 14.695819581958196,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005981747710003753,
      "loss": 0.2314,
      "num_input_tokens_seen": 28193728,
      "step": 133585
    },
    {
      "epoch": 14.696369636963697,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.005980597035340975,
      "loss": 0.2314,
      "num_input_tokens_seen": 28194752,
      "step": 133590
    },
    {
      "epoch": 14.696919691969196,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005979446443805702,
      "loss": 0.2319,
      "num_input_tokens_seen": 28195840,
      "step": 133595
    },
    {
      "epoch": 14.697469746974697,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.005978295935408548,
      "loss": 0.2319,
      "num_input_tokens_seen": 28196864,
      "step": 133600
    },
    {
      "epoch": 14.698019801980198,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.005977145510160106,
      "loss": 0.2298,
      "num_input_tokens_seen": 28197952,
      "step": 133605
    },
    {
      "epoch": 14.6985698569857,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005975995168070987,
      "loss": 0.2309,
      "num_input_tokens_seen": 28199008,
      "step": 133610
    },
    {
      "epoch": 14.699119911991199,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0059748449091517886,
      "loss": 0.2303,
      "num_input_tokens_seen": 28200032,
      "step": 133615
    },
    {
      "epoch": 14.6996699669967,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.005973694733413113,
      "loss": 0.2298,
      "num_input_tokens_seen": 28201152,
      "step": 133620
    },
    {
      "epoch": 14.7002200220022,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.005972544640865565,
      "loss": 0.2283,
      "num_input_tokens_seen": 28202208,
      "step": 133625
    },
    {
      "epoch": 14.7007700770077,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.005971394631519739,
      "loss": 0.2314,
      "num_input_tokens_seen": 28203168,
      "step": 133630
    },
    {
      "epoch": 14.701320132013201,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.005970244705386241,
      "loss": 0.2304,
      "num_input_tokens_seen": 28204224,
      "step": 133635
    },
    {
      "epoch": 14.701870187018702,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.005969094862475658,
      "loss": 0.2293,
      "num_input_tokens_seen": 28205216,
      "step": 133640
    },
    {
      "epoch": 14.702420242024202,
      "grad_norm": 0.00080108642578125,
      "learning_rate": 0.005967945102798596,
      "loss": 0.2288,
      "num_input_tokens_seen": 28206304,
      "step": 133645
    },
    {
      "epoch": 14.702970297029703,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005966795426365653,
      "loss": 0.2298,
      "num_input_tokens_seen": 28207392,
      "step": 133650
    },
    {
      "epoch": 14.703520352035204,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.005965645833187419,
      "loss": 0.2303,
      "num_input_tokens_seen": 28208416,
      "step": 133655
    },
    {
      "epoch": 14.704070407040705,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.005964496323274495,
      "loss": 0.2298,
      "num_input_tokens_seen": 28209472,
      "step": 133660
    },
    {
      "epoch": 14.704620462046204,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005963346896637475,
      "loss": 0.2314,
      "num_input_tokens_seen": 28210592,
      "step": 133665
    },
    {
      "epoch": 14.705170517051705,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.005962197553286944,
      "loss": 0.2303,
      "num_input_tokens_seen": 28211616,
      "step": 133670
    },
    {
      "epoch": 14.705720572057206,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.005961048293233502,
      "loss": 0.2314,
      "num_input_tokens_seen": 28212672,
      "step": 133675
    },
    {
      "epoch": 14.706270627062706,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.005959899116487741,
      "loss": 0.2293,
      "num_input_tokens_seen": 28213728,
      "step": 133680
    },
    {
      "epoch": 14.706820682068207,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.005958750023060255,
      "loss": 0.2309,
      "num_input_tokens_seen": 28214752,
      "step": 133685
    },
    {
      "epoch": 14.707370737073708,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005957601012961633,
      "loss": 0.2319,
      "num_input_tokens_seen": 28215808,
      "step": 133690
    },
    {
      "epoch": 14.707920792079207,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.0059564520862024585,
      "loss": 0.2309,
      "num_input_tokens_seen": 28216864,
      "step": 133695
    },
    {
      "epoch": 14.708470847084708,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.005955303242793326,
      "loss": 0.2298,
      "num_input_tokens_seen": 28217920,
      "step": 133700
    },
    {
      "epoch": 14.70902090209021,
      "grad_norm": 0.00060272216796875,
      "learning_rate": 0.005954154482744823,
      "loss": 0.2298,
      "num_input_tokens_seen": 28218912,
      "step": 133705
    },
    {
      "epoch": 14.70957095709571,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005953005806067542,
      "loss": 0.2319,
      "num_input_tokens_seen": 28220000,
      "step": 133710
    },
    {
      "epoch": 14.71012101210121,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.005951857212772066,
      "loss": 0.2319,
      "num_input_tokens_seen": 28221120,
      "step": 133715
    },
    {
      "epoch": 14.710671067106711,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0059507087028689756,
      "loss": 0.2303,
      "num_input_tokens_seen": 28222208,
      "step": 133720
    },
    {
      "epoch": 14.711221122112212,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005949560276368865,
      "loss": 0.2329,
      "num_input_tokens_seen": 28223264,
      "step": 133725
    },
    {
      "epoch": 14.711771177117711,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005948411933282311,
      "loss": 0.2314,
      "num_input_tokens_seen": 28224288,
      "step": 133730
    },
    {
      "epoch": 14.712321232123212,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005947263673619899,
      "loss": 0.2288,
      "num_input_tokens_seen": 28225312,
      "step": 133735
    },
    {
      "epoch": 14.712871287128714,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0059461154973922195,
      "loss": 0.2314,
      "num_input_tokens_seen": 28226368,
      "step": 133740
    },
    {
      "epoch": 14.713421342134213,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0059449674046098445,
      "loss": 0.2319,
      "num_input_tokens_seen": 28227392,
      "step": 133745
    },
    {
      "epoch": 14.713971397139714,
      "grad_norm": 0.00060272216796875,
      "learning_rate": 0.0059438193952833646,
      "loss": 0.2309,
      "num_input_tokens_seen": 28228416,
      "step": 133750
    },
    {
      "epoch": 14.714521452145215,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.00594267146942335,
      "loss": 0.2309,
      "num_input_tokens_seen": 28229504,
      "step": 133755
    },
    {
      "epoch": 14.715071507150714,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.005941523627040387,
      "loss": 0.2319,
      "num_input_tokens_seen": 28230592,
      "step": 133760
    },
    {
      "epoch": 14.715621562156215,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005940375868145059,
      "loss": 0.2293,
      "num_input_tokens_seen": 28231744,
      "step": 133765
    },
    {
      "epoch": 14.716171617161717,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.005939228192747934,
      "loss": 0.2314,
      "num_input_tokens_seen": 28232736,
      "step": 133770
    },
    {
      "epoch": 14.716721672167218,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0059380806008596,
      "loss": 0.2324,
      "num_input_tokens_seen": 28233792,
      "step": 133775
    },
    {
      "epoch": 14.717271727172717,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.005936933092490631,
      "loss": 0.2309,
      "num_input_tokens_seen": 28234816,
      "step": 133780
    },
    {
      "epoch": 14.717821782178218,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005935785667651593,
      "loss": 0.2314,
      "num_input_tokens_seen": 28235776,
      "step": 133785
    },
    {
      "epoch": 14.718371837183719,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0059346383263530695,
      "loss": 0.2309,
      "num_input_tokens_seen": 28236832,
      "step": 133790
    },
    {
      "epoch": 14.718921892189218,
      "grad_norm": 0.000698089599609375,
      "learning_rate": 0.005933491068605635,
      "loss": 0.2314,
      "num_input_tokens_seen": 28237888,
      "step": 133795
    },
    {
      "epoch": 14.71947194719472,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0059323438944198685,
      "loss": 0.233,
      "num_input_tokens_seen": 28238976,
      "step": 133800
    },
    {
      "epoch": 14.72002200220022,
      "grad_norm": 0.005859375,
      "learning_rate": 0.005931196803806336,
      "loss": 0.2303,
      "num_input_tokens_seen": 28239968,
      "step": 133805
    },
    {
      "epoch": 14.72057205720572,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.005930049796775608,
      "loss": 0.2314,
      "num_input_tokens_seen": 28240960,
      "step": 133810
    },
    {
      "epoch": 14.721122112211221,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.005928902873338261,
      "loss": 0.2298,
      "num_input_tokens_seen": 28242112,
      "step": 133815
    },
    {
      "epoch": 14.721672167216722,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.005927756033504855,
      "loss": 0.2308,
      "num_input_tokens_seen": 28243136,
      "step": 133820
    },
    {
      "epoch": 14.722222222222221,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.005926609277285978,
      "loss": 0.2329,
      "num_input_tokens_seen": 28244160,
      "step": 133825
    },
    {
      "epoch": 14.722772277227723,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.00592546260469219,
      "loss": 0.2314,
      "num_input_tokens_seen": 28245216,
      "step": 133830
    },
    {
      "epoch": 14.723322332233224,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.005924316015734053,
      "loss": 0.2324,
      "num_input_tokens_seen": 28246272,
      "step": 133835
    },
    {
      "epoch": 14.723872387238725,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.005923169510422146,
      "loss": 0.2309,
      "num_input_tokens_seen": 28247360,
      "step": 133840
    },
    {
      "epoch": 14.724422442244224,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.005922023088767025,
      "loss": 0.233,
      "num_input_tokens_seen": 28248416,
      "step": 133845
    },
    {
      "epoch": 14.724972497249725,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.005920876750779261,
      "loss": 0.2314,
      "num_input_tokens_seen": 28249408,
      "step": 133850
    },
    {
      "epoch": 14.725522552255226,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.005919730496469425,
      "loss": 0.2319,
      "num_input_tokens_seen": 28250496,
      "step": 133855
    },
    {
      "epoch": 14.726072607260726,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00591858432584807,
      "loss": 0.2324,
      "num_input_tokens_seen": 28251520,
      "step": 133860
    },
    {
      "epoch": 14.726622662266227,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.005917438238925771,
      "loss": 0.2324,
      "num_input_tokens_seen": 28252576,
      "step": 133865
    },
    {
      "epoch": 14.727172717271728,
      "grad_norm": 0.000637054443359375,
      "learning_rate": 0.005916292235713079,
      "loss": 0.2324,
      "num_input_tokens_seen": 28253664,
      "step": 133870
    },
    {
      "epoch": 14.727722772277227,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005915146316220571,
      "loss": 0.2298,
      "num_input_tokens_seen": 28254720,
      "step": 133875
    },
    {
      "epoch": 14.728272827282728,
      "grad_norm": 0.0005035400390625,
      "learning_rate": 0.0059140004804587925,
      "loss": 0.2324,
      "num_input_tokens_seen": 28255776,
      "step": 133880
    },
    {
      "epoch": 14.72882288228823,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005912854728438313,
      "loss": 0.2308,
      "num_input_tokens_seen": 28256832,
      "step": 133885
    },
    {
      "epoch": 14.729372937293729,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.005911709060169696,
      "loss": 0.2303,
      "num_input_tokens_seen": 28257888,
      "step": 133890
    },
    {
      "epoch": 14.72992299229923,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.005910563475663488,
      "loss": 0.2293,
      "num_input_tokens_seen": 28258944,
      "step": 133895
    },
    {
      "epoch": 14.73047304730473,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.005909417974930263,
      "loss": 0.2324,
      "num_input_tokens_seen": 28260032,
      "step": 133900
    },
    {
      "epoch": 14.731023102310232,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.005908272557980564,
      "loss": 0.2325,
      "num_input_tokens_seen": 28261152,
      "step": 133905
    },
    {
      "epoch": 14.731573157315731,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.005907127224824955,
      "loss": 0.233,
      "num_input_tokens_seen": 28262208,
      "step": 133910
    },
    {
      "epoch": 14.732123212321232,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.005905981975473995,
      "loss": 0.2314,
      "num_input_tokens_seen": 28263264,
      "step": 133915
    },
    {
      "epoch": 14.732673267326733,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.005904836809938232,
      "loss": 0.2324,
      "num_input_tokens_seen": 28264320,
      "step": 133920
    },
    {
      "epoch": 14.733223322332233,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005903691728228228,
      "loss": 0.233,
      "num_input_tokens_seen": 28265376,
      "step": 133925
    },
    {
      "epoch": 14.733773377337734,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.005902546730354532,
      "loss": 0.2309,
      "num_input_tokens_seen": 28266432,
      "step": 133930
    },
    {
      "epoch": 14.734323432343235,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.005901401816327688,
      "loss": 0.2314,
      "num_input_tokens_seen": 28267456,
      "step": 133935
    },
    {
      "epoch": 14.734873487348734,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.005900256986158268,
      "loss": 0.2293,
      "num_input_tokens_seen": 28268512,
      "step": 133940
    },
    {
      "epoch": 14.735423542354235,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.005899112239856807,
      "loss": 0.2298,
      "num_input_tokens_seen": 28269536,
      "step": 133945
    },
    {
      "epoch": 14.735973597359736,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.005897967577433866,
      "loss": 0.2314,
      "num_input_tokens_seen": 28270592,
      "step": 133950
    },
    {
      "epoch": 14.736523652365236,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.005896822998899992,
      "loss": 0.2293,
      "num_input_tokens_seen": 28271616,
      "step": 133955
    },
    {
      "epoch": 14.737073707370737,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.005895678504265725,
      "loss": 0.2324,
      "num_input_tokens_seen": 28272704,
      "step": 133960
    },
    {
      "epoch": 14.737623762376238,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.005894534093541623,
      "loss": 0.2309,
      "num_input_tokens_seen": 28273792,
      "step": 133965
    },
    {
      "epoch": 14.738173817381739,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005893389766738235,
      "loss": 0.2314,
      "num_input_tokens_seen": 28274848,
      "step": 133970
    },
    {
      "epoch": 14.738723872387238,
      "grad_norm": 0.00074005126953125,
      "learning_rate": 0.005892245523866099,
      "loss": 0.2304,
      "num_input_tokens_seen": 28275936,
      "step": 133975
    },
    {
      "epoch": 14.73927392739274,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.005891101364935769,
      "loss": 0.2309,
      "num_input_tokens_seen": 28276960,
      "step": 133980
    },
    {
      "epoch": 14.73982398239824,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.005889957289957785,
      "loss": 0.2324,
      "num_input_tokens_seen": 28277920,
      "step": 133985
    },
    {
      "epoch": 14.74037403740374,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005888813298942695,
      "loss": 0.2345,
      "num_input_tokens_seen": 28278976,
      "step": 133990
    },
    {
      "epoch": 14.74092409240924,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.005887669391901037,
      "loss": 0.2314,
      "num_input_tokens_seen": 28280000,
      "step": 133995
    },
    {
      "epoch": 14.741474147414742,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.005886525568843359,
      "loss": 0.2324,
      "num_input_tokens_seen": 28281024,
      "step": 134000
    },
    {
      "epoch": 14.742024202420241,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005885381829780205,
      "loss": 0.2314,
      "num_input_tokens_seen": 28282080,
      "step": 134005
    },
    {
      "epoch": 14.742574257425742,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.005884238174722107,
      "loss": 0.2309,
      "num_input_tokens_seen": 28283072,
      "step": 134010
    },
    {
      "epoch": 14.743124312431243,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.0058830946036796195,
      "loss": 0.2308,
      "num_input_tokens_seen": 28284032,
      "step": 134015
    },
    {
      "epoch": 14.743674367436743,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.005881951116663268,
      "loss": 0.2319,
      "num_input_tokens_seen": 28285024,
      "step": 134020
    },
    {
      "epoch": 14.744224422442244,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005880807713683598,
      "loss": 0.2314,
      "num_input_tokens_seen": 28286144,
      "step": 134025
    },
    {
      "epoch": 14.744774477447745,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0058796643947511525,
      "loss": 0.2313,
      "num_input_tokens_seen": 28287200,
      "step": 134030
    },
    {
      "epoch": 14.745324532453246,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0058785211598764596,
      "loss": 0.2319,
      "num_input_tokens_seen": 28288224,
      "step": 134035
    },
    {
      "epoch": 14.745874587458745,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005877378009070064,
      "loss": 0.2314,
      "num_input_tokens_seen": 28289280,
      "step": 134040
    },
    {
      "epoch": 14.746424642464246,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005876234942342498,
      "loss": 0.2298,
      "num_input_tokens_seen": 28290304,
      "step": 134045
    },
    {
      "epoch": 14.746974697469748,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005875091959704292,
      "loss": 0.233,
      "num_input_tokens_seen": 28291328,
      "step": 134050
    },
    {
      "epoch": 14.747524752475247,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0058739490611659844,
      "loss": 0.2309,
      "num_input_tokens_seen": 28292384,
      "step": 134055
    },
    {
      "epoch": 14.748074807480748,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.0058728062467381095,
      "loss": 0.2304,
      "num_input_tokens_seen": 28293472,
      "step": 134060
    },
    {
      "epoch": 14.748624862486249,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005871663516431204,
      "loss": 0.2309,
      "num_input_tokens_seen": 28294496,
      "step": 134065
    },
    {
      "epoch": 14.749174917491748,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.005870520870255798,
      "loss": 0.2309,
      "num_input_tokens_seen": 28295552,
      "step": 134070
    },
    {
      "epoch": 14.74972497249725,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.005869378308222413,
      "loss": 0.2304,
      "num_input_tokens_seen": 28296608,
      "step": 134075
    },
    {
      "epoch": 14.75027502750275,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.005868235830341588,
      "loss": 0.2319,
      "num_input_tokens_seen": 28297568,
      "step": 134080
    },
    {
      "epoch": 14.750825082508252,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005867093436623851,
      "loss": 0.2319,
      "num_input_tokens_seen": 28298624,
      "step": 134085
    },
    {
      "epoch": 14.751375137513751,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.005865951127079738,
      "loss": 0.2319,
      "num_input_tokens_seen": 28299712,
      "step": 134090
    },
    {
      "epoch": 14.751925192519252,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.005864808901719768,
      "loss": 0.2309,
      "num_input_tokens_seen": 28300736,
      "step": 134095
    },
    {
      "epoch": 14.752475247524753,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0058636667605544675,
      "loss": 0.2324,
      "num_input_tokens_seen": 28301792,
      "step": 134100
    },
    {
      "epoch": 14.753025302530252,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00586252470359437,
      "loss": 0.2335,
      "num_input_tokens_seen": 28302848,
      "step": 134105
    },
    {
      "epoch": 14.753575357535754,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005861382730849995,
      "loss": 0.2309,
      "num_input_tokens_seen": 28303904,
      "step": 134110
    },
    {
      "epoch": 14.754125412541255,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.005860240842331868,
      "loss": 0.2319,
      "num_input_tokens_seen": 28304928,
      "step": 134115
    },
    {
      "epoch": 14.754675467546754,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0058590990380505215,
      "loss": 0.2293,
      "num_input_tokens_seen": 28305920,
      "step": 134120
    },
    {
      "epoch": 14.755225522552255,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.005857957318016467,
      "loss": 0.2293,
      "num_input_tokens_seen": 28306976,
      "step": 134125
    },
    {
      "epoch": 14.755775577557756,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00585681568224024,
      "loss": 0.2303,
      "num_input_tokens_seen": 28308064,
      "step": 134130
    },
    {
      "epoch": 14.756325632563257,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.005855674130732348,
      "loss": 0.2304,
      "num_input_tokens_seen": 28309088,
      "step": 134135
    },
    {
      "epoch": 14.756875687568757,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005854532663503322,
      "loss": 0.2324,
      "num_input_tokens_seen": 28310176,
      "step": 134140
    },
    {
      "epoch": 14.757425742574258,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.005853391280563685,
      "loss": 0.2319,
      "num_input_tokens_seen": 28311200,
      "step": 134145
    },
    {
      "epoch": 14.757975797579759,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005852249981923944,
      "loss": 0.2298,
      "num_input_tokens_seen": 28312224,
      "step": 134150
    },
    {
      "epoch": 14.758525852585258,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.005851108767594633,
      "loss": 0.2309,
      "num_input_tokens_seen": 28313280,
      "step": 134155
    },
    {
      "epoch": 14.75907590759076,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.0058499676375862565,
      "loss": 0.2314,
      "num_input_tokens_seen": 28314304,
      "step": 134160
    },
    {
      "epoch": 14.75962596259626,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0058488265919093415,
      "loss": 0.2314,
      "num_input_tokens_seen": 28315296,
      "step": 134165
    },
    {
      "epoch": 14.76017601760176,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005847685630574397,
      "loss": 0.2309,
      "num_input_tokens_seen": 28316384,
      "step": 134170
    },
    {
      "epoch": 14.76072607260726,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.005846544753591943,
      "loss": 0.2314,
      "num_input_tokens_seen": 28317408,
      "step": 134175
    },
    {
      "epoch": 14.761276127612762,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.0058454039609724985,
      "loss": 0.2309,
      "num_input_tokens_seen": 28318432,
      "step": 134180
    },
    {
      "epoch": 14.761826182618261,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.005844263252726572,
      "loss": 0.2303,
      "num_input_tokens_seen": 28319488,
      "step": 134185
    },
    {
      "epoch": 14.762376237623762,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005843122628864672,
      "loss": 0.2304,
      "num_input_tokens_seen": 28320544,
      "step": 134190
    },
    {
      "epoch": 14.762926292629263,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.0058419820893973184,
      "loss": 0.2325,
      "num_input_tokens_seen": 28321728,
      "step": 134195
    },
    {
      "epoch": 14.763476347634764,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0058408416343350205,
      "loss": 0.2319,
      "num_input_tokens_seen": 28322784,
      "step": 134200
    },
    {
      "epoch": 14.764026402640264,
      "grad_norm": 0.0003910064697265625,
      "learning_rate": 0.005839701263688294,
      "loss": 0.2309,
      "num_input_tokens_seen": 28323744,
      "step": 134205
    },
    {
      "epoch": 14.764576457645765,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.005838560977467648,
      "loss": 0.2319,
      "num_input_tokens_seen": 28324800,
      "step": 134210
    },
    {
      "epoch": 14.765126512651266,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0058374207756835845,
      "loss": 0.2314,
      "num_input_tokens_seen": 28325888,
      "step": 134215
    },
    {
      "epoch": 14.765676567656765,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005836280658346621,
      "loss": 0.2298,
      "num_input_tokens_seen": 28326944,
      "step": 134220
    },
    {
      "epoch": 14.766226622662266,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005835140625467256,
      "loss": 0.2329,
      "num_input_tokens_seen": 28328000,
      "step": 134225
    },
    {
      "epoch": 14.766776677667767,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.005834000677056003,
      "loss": 0.2319,
      "num_input_tokens_seen": 28329056,
      "step": 134230
    },
    {
      "epoch": 14.767326732673267,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.005832860813123373,
      "loss": 0.2309,
      "num_input_tokens_seen": 28330112,
      "step": 134235
    },
    {
      "epoch": 14.767876787678768,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.00583172103367986,
      "loss": 0.2293,
      "num_input_tokens_seen": 28331136,
      "step": 134240
    },
    {
      "epoch": 14.768426842684269,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.005830581338735979,
      "loss": 0.2325,
      "num_input_tokens_seen": 28332160,
      "step": 134245
    },
    {
      "epoch": 14.768976897689768,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005829441728302227,
      "loss": 0.2308,
      "num_input_tokens_seen": 28333216,
      "step": 134250
    },
    {
      "epoch": 14.76952695269527,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0058283022023891086,
      "loss": 0.2314,
      "num_input_tokens_seen": 28334240,
      "step": 134255
    },
    {
      "epoch": 14.77007700770077,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005827162761007133,
      "loss": 0.2309,
      "num_input_tokens_seen": 28335264,
      "step": 134260
    },
    {
      "epoch": 14.770627062706271,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0058260234041667925,
      "loss": 0.2303,
      "num_input_tokens_seen": 28336352,
      "step": 134265
    },
    {
      "epoch": 14.77117711771177,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005824884131878597,
      "loss": 0.2298,
      "num_input_tokens_seen": 28337408,
      "step": 134270
    },
    {
      "epoch": 14.771727172717272,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.005823744944153035,
      "loss": 0.2319,
      "num_input_tokens_seen": 28338496,
      "step": 134275
    },
    {
      "epoch": 14.772277227722773,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0058226058410006205,
      "loss": 0.2309,
      "num_input_tokens_seen": 28339552,
      "step": 134280
    },
    {
      "epoch": 14.772827282728272,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005821466822431837,
      "loss": 0.2324,
      "num_input_tokens_seen": 28340576,
      "step": 134285
    },
    {
      "epoch": 14.773377337733773,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.005820327888457191,
      "loss": 0.2324,
      "num_input_tokens_seen": 28341632,
      "step": 134290
    },
    {
      "epoch": 14.773927392739274,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005819189039087182,
      "loss": 0.2324,
      "num_input_tokens_seen": 28342816,
      "step": 134295
    },
    {
      "epoch": 14.774477447744774,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.005818050274332297,
      "loss": 0.2324,
      "num_input_tokens_seen": 28343808,
      "step": 134300
    },
    {
      "epoch": 14.775027502750275,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005816911594203042,
      "loss": 0.2314,
      "num_input_tokens_seen": 28344864,
      "step": 134305
    },
    {
      "epoch": 14.775577557755776,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.0058157729987099015,
      "loss": 0.2319,
      "num_input_tokens_seen": 28345952,
      "step": 134310
    },
    {
      "epoch": 14.776127612761275,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.005814634487863375,
      "loss": 0.2314,
      "num_input_tokens_seen": 28347008,
      "step": 134315
    },
    {
      "epoch": 14.776677667766776,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005813496061673959,
      "loss": 0.2319,
      "num_input_tokens_seen": 28348096,
      "step": 134320
    },
    {
      "epoch": 14.777227722772277,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.005812357720152137,
      "loss": 0.2298,
      "num_input_tokens_seen": 28349120,
      "step": 134325
    },
    {
      "epoch": 14.777777777777779,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00581121946330841,
      "loss": 0.2314,
      "num_input_tokens_seen": 28350176,
      "step": 134330
    },
    {
      "epoch": 14.778327832783278,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.005810081291153265,
      "loss": 0.2314,
      "num_input_tokens_seen": 28351296,
      "step": 134335
    },
    {
      "epoch": 14.778877887788779,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005808943203697185,
      "loss": 0.2309,
      "num_input_tokens_seen": 28352352,
      "step": 134340
    },
    {
      "epoch": 14.77942794279428,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.005807805200950665,
      "loss": 0.2309,
      "num_input_tokens_seen": 28353440,
      "step": 134345
    },
    {
      "epoch": 14.77997799779978,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005806667282924194,
      "loss": 0.2314,
      "num_input_tokens_seen": 28354496,
      "step": 134350
    },
    {
      "epoch": 14.78052805280528,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.0058055294496282646,
      "loss": 0.2324,
      "num_input_tokens_seen": 28355616,
      "step": 134355
    },
    {
      "epoch": 14.781078107810782,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.005804391701073358,
      "loss": 0.2314,
      "num_input_tokens_seen": 28356672,
      "step": 134360
    },
    {
      "epoch": 14.781628162816281,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005803254037269955,
      "loss": 0.2319,
      "num_input_tokens_seen": 28357792,
      "step": 134365
    },
    {
      "epoch": 14.782178217821782,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.0058021164582285475,
      "loss": 0.2298,
      "num_input_tokens_seen": 28358880,
      "step": 134370
    },
    {
      "epoch": 14.782728272827283,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.005800978963959624,
      "loss": 0.2314,
      "num_input_tokens_seen": 28359904,
      "step": 134375
    },
    {
      "epoch": 14.783278327832782,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005799841554473658,
      "loss": 0.233,
      "num_input_tokens_seen": 28360992,
      "step": 134380
    },
    {
      "epoch": 14.783828382838283,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.005798704229781142,
      "loss": 0.2309,
      "num_input_tokens_seen": 28361984,
      "step": 134385
    },
    {
      "epoch": 14.784378437843785,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.005797566989892549,
      "loss": 0.2314,
      "num_input_tokens_seen": 28362944,
      "step": 134390
    },
    {
      "epoch": 14.784928492849286,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.005796429834818371,
      "loss": 0.2314,
      "num_input_tokens_seen": 28363936,
      "step": 134395
    },
    {
      "epoch": 14.785478547854785,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.005795292764569079,
      "loss": 0.2304,
      "num_input_tokens_seen": 28364960,
      "step": 134400
    },
    {
      "epoch": 14.786028602860286,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005794155779155155,
      "loss": 0.2309,
      "num_input_tokens_seen": 28366016,
      "step": 134405
    },
    {
      "epoch": 14.786578657865787,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005793018878587085,
      "loss": 0.2319,
      "num_input_tokens_seen": 28367072,
      "step": 134410
    },
    {
      "epoch": 14.787128712871286,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.005791882062875337,
      "loss": 0.2309,
      "num_input_tokens_seen": 28368128,
      "step": 134415
    },
    {
      "epoch": 14.787678767876788,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.005790745332030399,
      "loss": 0.2319,
      "num_input_tokens_seen": 28369216,
      "step": 134420
    },
    {
      "epoch": 14.788228822882289,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.005789608686062736,
      "loss": 0.2319,
      "num_input_tokens_seen": 28370336,
      "step": 134425
    },
    {
      "epoch": 14.788778877887788,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0057884721249828325,
      "loss": 0.234,
      "num_input_tokens_seen": 28371328,
      "step": 134430
    },
    {
      "epoch": 14.789328932893289,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.005787335648801165,
      "loss": 0.2303,
      "num_input_tokens_seen": 28372352,
      "step": 134435
    },
    {
      "epoch": 14.78987898789879,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005786199257528199,
      "loss": 0.2303,
      "num_input_tokens_seen": 28373376,
      "step": 134440
    },
    {
      "epoch": 14.79042904290429,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.005785062951174418,
      "loss": 0.2329,
      "num_input_tokens_seen": 28374496,
      "step": 134445
    },
    {
      "epoch": 14.79097909790979,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0057839267297502914,
      "loss": 0.2319,
      "num_input_tokens_seen": 28375456,
      "step": 134450
    },
    {
      "epoch": 14.791529152915292,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0057827905932662845,
      "loss": 0.2293,
      "num_input_tokens_seen": 28376448,
      "step": 134455
    },
    {
      "epoch": 14.792079207920793,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.005781654541732872,
      "loss": 0.2314,
      "num_input_tokens_seen": 28377472,
      "step": 134460
    },
    {
      "epoch": 14.792629262926292,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.005780518575160527,
      "loss": 0.2309,
      "num_input_tokens_seen": 28378528,
      "step": 134465
    },
    {
      "epoch": 14.793179317931793,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005779382693559725,
      "loss": 0.233,
      "num_input_tokens_seen": 28379552,
      "step": 134470
    },
    {
      "epoch": 14.793729372937294,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005778246896940927,
      "loss": 0.2309,
      "num_input_tokens_seen": 28380544,
      "step": 134475
    },
    {
      "epoch": 14.794279427942794,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005777111185314597,
      "loss": 0.2304,
      "num_input_tokens_seen": 28381632,
      "step": 134480
    },
    {
      "epoch": 14.794829482948295,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.005775975558691208,
      "loss": 0.2309,
      "num_input_tokens_seen": 28382720,
      "step": 134485
    },
    {
      "epoch": 14.795379537953796,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.005774840017081226,
      "loss": 0.2314,
      "num_input_tokens_seen": 28383712,
      "step": 134490
    },
    {
      "epoch": 14.795929592959295,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005773704560495123,
      "loss": 0.2304,
      "num_input_tokens_seen": 28384736,
      "step": 134495
    },
    {
      "epoch": 14.796479647964796,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.005772569188943357,
      "loss": 0.2288,
      "num_input_tokens_seen": 28385760,
      "step": 134500
    },
    {
      "epoch": 14.797029702970297,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.005771433902436388,
      "loss": 0.233,
      "num_input_tokens_seen": 28386816,
      "step": 134505
    },
    {
      "epoch": 14.797579757975798,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.005770298700984689,
      "loss": 0.2319,
      "num_input_tokens_seen": 28387904,
      "step": 134510
    },
    {
      "epoch": 14.798129812981298,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005769163584598712,
      "loss": 0.2314,
      "num_input_tokens_seen": 28388896,
      "step": 134515
    },
    {
      "epoch": 14.798679867986799,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.005768028553288927,
      "loss": 0.2308,
      "num_input_tokens_seen": 28389952,
      "step": 134520
    },
    {
      "epoch": 14.7992299229923,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005766893607065798,
      "loss": 0.2335,
      "num_input_tokens_seen": 28390976,
      "step": 134525
    },
    {
      "epoch": 14.7997799779978,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.005765758745939774,
      "loss": 0.2298,
      "num_input_tokens_seen": 28392160,
      "step": 134530
    },
    {
      "epoch": 14.8003300330033,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.005764623969921326,
      "loss": 0.2324,
      "num_input_tokens_seen": 28393184,
      "step": 134535
    },
    {
      "epoch": 14.800880088008801,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.005763489279020902,
      "loss": 0.2319,
      "num_input_tokens_seen": 28394208,
      "step": 134540
    },
    {
      "epoch": 14.8014301430143,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.005762354673248965,
      "loss": 0.2335,
      "num_input_tokens_seen": 28395264,
      "step": 134545
    },
    {
      "epoch": 14.801980198019802,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.005761220152615979,
      "loss": 0.2314,
      "num_input_tokens_seen": 28396320,
      "step": 134550
    },
    {
      "epoch": 14.802530253025303,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.005760085717132387,
      "loss": 0.233,
      "num_input_tokens_seen": 28397376,
      "step": 134555
    },
    {
      "epoch": 14.803080308030804,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005758951366808656,
      "loss": 0.2309,
      "num_input_tokens_seen": 28398400,
      "step": 134560
    },
    {
      "epoch": 14.803630363036303,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005757817101655237,
      "loss": 0.2303,
      "num_input_tokens_seen": 28399424,
      "step": 134565
    },
    {
      "epoch": 14.804180418041804,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.005756682921682578,
      "loss": 0.2325,
      "num_input_tokens_seen": 28400480,
      "step": 134570
    },
    {
      "epoch": 14.804730473047305,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005755548826901136,
      "loss": 0.2298,
      "num_input_tokens_seen": 28401472,
      "step": 134575
    },
    {
      "epoch": 14.805280528052805,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0057544148173213645,
      "loss": 0.2319,
      "num_input_tokens_seen": 28402560,
      "step": 134580
    },
    {
      "epoch": 14.805830583058306,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.005753280892953721,
      "loss": 0.2351,
      "num_input_tokens_seen": 28403584,
      "step": 134585
    },
    {
      "epoch": 14.806380638063807,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.005752147053808651,
      "loss": 0.2309,
      "num_input_tokens_seen": 28404608,
      "step": 134590
    },
    {
      "epoch": 14.806930693069306,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005751013299896598,
      "loss": 0.2308,
      "num_input_tokens_seen": 28405632,
      "step": 134595
    },
    {
      "epoch": 14.807480748074807,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.005749879631228023,
      "loss": 0.2329,
      "num_input_tokens_seen": 28406720,
      "step": 134600
    },
    {
      "epoch": 14.808030803080309,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005748746047813358,
      "loss": 0.2314,
      "num_input_tokens_seen": 28407712,
      "step": 134605
    },
    {
      "epoch": 14.808580858085808,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005747612549663072,
      "loss": 0.2324,
      "num_input_tokens_seen": 28408704,
      "step": 134610
    },
    {
      "epoch": 14.809130913091309,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.005746479136787602,
      "loss": 0.234,
      "num_input_tokens_seen": 28409696,
      "step": 134615
    },
    {
      "epoch": 14.80968096809681,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005745345809197387,
      "loss": 0.2324,
      "num_input_tokens_seen": 28410816,
      "step": 134620
    },
    {
      "epoch": 14.810231023102311,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005744212566902886,
      "loss": 0.2288,
      "num_input_tokens_seen": 28411872,
      "step": 134625
    },
    {
      "epoch": 14.81078107810781,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.005743079409914531,
      "loss": 0.233,
      "num_input_tokens_seen": 28412896,
      "step": 134630
    },
    {
      "epoch": 14.811331133113312,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.005741946338242769,
      "loss": 0.2298,
      "num_input_tokens_seen": 28413984,
      "step": 134635
    },
    {
      "epoch": 14.811881188118813,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005740813351898053,
      "loss": 0.2314,
      "num_input_tokens_seen": 28415040,
      "step": 134640
    },
    {
      "epoch": 14.812431243124312,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.00573968045089081,
      "loss": 0.2308,
      "num_input_tokens_seen": 28416032,
      "step": 134645
    },
    {
      "epoch": 14.812981298129813,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.005738547635231495,
      "loss": 0.2314,
      "num_input_tokens_seen": 28417088,
      "step": 134650
    },
    {
      "epoch": 14.813531353135314,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0057374149049305365,
      "loss": 0.2335,
      "num_input_tokens_seen": 28418176,
      "step": 134655
    },
    {
      "epoch": 14.814081408140813,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.005736282259998387,
      "loss": 0.2309,
      "num_input_tokens_seen": 28419168,
      "step": 134660
    },
    {
      "epoch": 14.814631463146315,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.005735149700445471,
      "loss": 0.233,
      "num_input_tokens_seen": 28420160,
      "step": 134665
    },
    {
      "epoch": 14.815181518151816,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005734017226282237,
      "loss": 0.2314,
      "num_input_tokens_seen": 28421216,
      "step": 134670
    },
    {
      "epoch": 14.815731573157315,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.005732884837519123,
      "loss": 0.2288,
      "num_input_tokens_seen": 28422336,
      "step": 134675
    },
    {
      "epoch": 14.816281628162816,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.00573175253416656,
      "loss": 0.2329,
      "num_input_tokens_seen": 28423424,
      "step": 134680
    },
    {
      "epoch": 14.816831683168317,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00573062031623499,
      "loss": 0.2324,
      "num_input_tokens_seen": 28424512,
      "step": 134685
    },
    {
      "epoch": 14.817381738173818,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.005729488183734839,
      "loss": 0.2324,
      "num_input_tokens_seen": 28425536,
      "step": 134690
    },
    {
      "epoch": 14.817931793179318,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005728356136676547,
      "loss": 0.2313,
      "num_input_tokens_seen": 28426624,
      "step": 134695
    },
    {
      "epoch": 14.818481848184819,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005727224175070554,
      "loss": 0.2324,
      "num_input_tokens_seen": 28427648,
      "step": 134700
    },
    {
      "epoch": 14.81903190319032,
      "grad_norm": 0.0007476806640625,
      "learning_rate": 0.005726092298927282,
      "loss": 0.2303,
      "num_input_tokens_seen": 28428704,
      "step": 134705
    },
    {
      "epoch": 14.819581958195819,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005724960508257169,
      "loss": 0.2319,
      "num_input_tokens_seen": 28429824,
      "step": 134710
    },
    {
      "epoch": 14.82013201320132,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0057238288030706476,
      "loss": 0.2345,
      "num_input_tokens_seen": 28430880,
      "step": 134715
    },
    {
      "epoch": 14.820682068206821,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.005722697183378133,
      "loss": 0.2304,
      "num_input_tokens_seen": 28431968,
      "step": 134720
    },
    {
      "epoch": 14.82123212321232,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.005721565649190076,
      "loss": 0.2303,
      "num_input_tokens_seen": 28433024,
      "step": 134725
    },
    {
      "epoch": 14.821782178217822,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005720434200516894,
      "loss": 0.2319,
      "num_input_tokens_seen": 28434080,
      "step": 134730
    },
    {
      "epoch": 14.822332233223323,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.005719302837369021,
      "loss": 0.2314,
      "num_input_tokens_seen": 28435168,
      "step": 134735
    },
    {
      "epoch": 14.822882288228822,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.00571817155975688,
      "loss": 0.2308,
      "num_input_tokens_seen": 28436224,
      "step": 134740
    },
    {
      "epoch": 14.823432343234323,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.005717040367690893,
      "loss": 0.2309,
      "num_input_tokens_seen": 28437312,
      "step": 134745
    },
    {
      "epoch": 14.823982398239824,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.00571590926118149,
      "loss": 0.2293,
      "num_input_tokens_seen": 28438400,
      "step": 134750
    },
    {
      "epoch": 14.824532453245325,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005714778240239102,
      "loss": 0.2335,
      "num_input_tokens_seen": 28439488,
      "step": 134755
    },
    {
      "epoch": 14.825082508250825,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.005713647304874144,
      "loss": 0.2319,
      "num_input_tokens_seen": 28440576,
      "step": 134760
    },
    {
      "epoch": 14.825632563256326,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.005712516455097045,
      "loss": 0.2314,
      "num_input_tokens_seen": 28441632,
      "step": 134765
    },
    {
      "epoch": 14.826182618261827,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.005711385690918222,
      "loss": 0.2324,
      "num_input_tokens_seen": 28442720,
      "step": 134770
    },
    {
      "epoch": 14.826732673267326,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.005710255012348105,
      "loss": 0.2298,
      "num_input_tokens_seen": 28443744,
      "step": 134775
    },
    {
      "epoch": 14.827282728272827,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005709124419397104,
      "loss": 0.2324,
      "num_input_tokens_seen": 28444800,
      "step": 134780
    },
    {
      "epoch": 14.827832783278328,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.005707993912075646,
      "loss": 0.2324,
      "num_input_tokens_seen": 28445856,
      "step": 134785
    },
    {
      "epoch": 14.828382838283828,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.005706863490394154,
      "loss": 0.2335,
      "num_input_tokens_seen": 28446912,
      "step": 134790
    },
    {
      "epoch": 14.828932893289329,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0057057331543630355,
      "loss": 0.2309,
      "num_input_tokens_seen": 28447936,
      "step": 134795
    },
    {
      "epoch": 14.82948294829483,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005704602903992721,
      "loss": 0.2309,
      "num_input_tokens_seen": 28449024,
      "step": 134800
    },
    {
      "epoch": 14.83003300330033,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.005703472739293617,
      "loss": 0.233,
      "num_input_tokens_seen": 28450048,
      "step": 134805
    },
    {
      "epoch": 14.83058305830583,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005702342660276144,
      "loss": 0.2303,
      "num_input_tokens_seen": 28451104,
      "step": 134810
    },
    {
      "epoch": 14.831133113311331,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.005701212666950719,
      "loss": 0.2324,
      "num_input_tokens_seen": 28452128,
      "step": 134815
    },
    {
      "epoch": 14.831683168316832,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.005700082759327752,
      "loss": 0.2293,
      "num_input_tokens_seen": 28453184,
      "step": 134820
    },
    {
      "epoch": 14.832233223322332,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005698952937417665,
      "loss": 0.2314,
      "num_input_tokens_seen": 28454304,
      "step": 134825
    },
    {
      "epoch": 14.832783278327833,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.005697823201230867,
      "loss": 0.2303,
      "num_input_tokens_seen": 28455360,
      "step": 134830
    },
    {
      "epoch": 14.833333333333334,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005696693550777762,
      "loss": 0.2314,
      "num_input_tokens_seen": 28456352,
      "step": 134835
    },
    {
      "epoch": 14.833883388338833,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.005695563986068769,
      "loss": 0.2304,
      "num_input_tokens_seen": 28457440,
      "step": 134840
    },
    {
      "epoch": 14.834433443344334,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.0056944345071142976,
      "loss": 0.2324,
      "num_input_tokens_seen": 28458432,
      "step": 134845
    },
    {
      "epoch": 14.834983498349835,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.005693305113924763,
      "loss": 0.2319,
      "num_input_tokens_seen": 28459552,
      "step": 134850
    },
    {
      "epoch": 14.835533553355335,
      "grad_norm": 0.0028533935546875,
      "learning_rate": 0.00569217580651057,
      "loss": 0.2314,
      "num_input_tokens_seen": 28460608,
      "step": 134855
    },
    {
      "epoch": 14.836083608360836,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.005691046584882121,
      "loss": 0.2298,
      "num_input_tokens_seen": 28461696,
      "step": 134860
    },
    {
      "epoch": 14.836633663366337,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.005689917449049829,
      "loss": 0.2314,
      "num_input_tokens_seen": 28462784,
      "step": 134865
    },
    {
      "epoch": 14.837183718371836,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0056887883990240995,
      "loss": 0.2335,
      "num_input_tokens_seen": 28463872,
      "step": 134870
    },
    {
      "epoch": 14.837733773377337,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.005687659434815345,
      "loss": 0.2335,
      "num_input_tokens_seen": 28464992,
      "step": 134875
    },
    {
      "epoch": 14.838283828382838,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.005686530556433964,
      "loss": 0.2304,
      "num_input_tokens_seen": 28466048,
      "step": 134880
    },
    {
      "epoch": 14.83883388338834,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0056854017638903595,
      "loss": 0.233,
      "num_input_tokens_seen": 28467168,
      "step": 134885
    },
    {
      "epoch": 14.839383938393839,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.00568427305719494,
      "loss": 0.2293,
      "num_input_tokens_seen": 28468224,
      "step": 134890
    },
    {
      "epoch": 14.83993399339934,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.005683144436358102,
      "loss": 0.2309,
      "num_input_tokens_seen": 28469280,
      "step": 134895
    },
    {
      "epoch": 14.840484048404841,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.00568201590139025,
      "loss": 0.2303,
      "num_input_tokens_seen": 28470336,
      "step": 134900
    },
    {
      "epoch": 14.84103410341034,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0056808874523017906,
      "loss": 0.233,
      "num_input_tokens_seen": 28471360,
      "step": 134905
    },
    {
      "epoch": 14.841584158415841,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005679759089103118,
      "loss": 0.2324,
      "num_input_tokens_seen": 28472384,
      "step": 134910
    },
    {
      "epoch": 14.842134213421343,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.005678630811804635,
      "loss": 0.2304,
      "num_input_tokens_seen": 28473408,
      "step": 134915
    },
    {
      "epoch": 14.842684268426842,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.005677502620416736,
      "loss": 0.2314,
      "num_input_tokens_seen": 28474464,
      "step": 134920
    },
    {
      "epoch": 14.843234323432343,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.005676374514949822,
      "loss": 0.2324,
      "num_input_tokens_seen": 28475520,
      "step": 134925
    },
    {
      "epoch": 14.843784378437844,
      "grad_norm": 0.000843048095703125,
      "learning_rate": 0.005675246495414294,
      "loss": 0.2329,
      "num_input_tokens_seen": 28476544,
      "step": 134930
    },
    {
      "epoch": 14.844334433443345,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005674118561820539,
      "loss": 0.2319,
      "num_input_tokens_seen": 28477536,
      "step": 134935
    },
    {
      "epoch": 14.844884488448844,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005672990714178966,
      "loss": 0.2314,
      "num_input_tokens_seen": 28478656,
      "step": 134940
    },
    {
      "epoch": 14.845434543454346,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.00567186295249996,
      "loss": 0.2319,
      "num_input_tokens_seen": 28479744,
      "step": 134945
    },
    {
      "epoch": 14.845984598459847,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005670735276793912,
      "loss": 0.2303,
      "num_input_tokens_seen": 28480832,
      "step": 134950
    },
    {
      "epoch": 14.846534653465346,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.005669607687071222,
      "loss": 0.2314,
      "num_input_tokens_seen": 28481920,
      "step": 134955
    },
    {
      "epoch": 14.847084708470847,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.005668480183342279,
      "loss": 0.2335,
      "num_input_tokens_seen": 28483008,
      "step": 134960
    },
    {
      "epoch": 14.847634763476348,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.005667352765617482,
      "loss": 0.2298,
      "num_input_tokens_seen": 28484128,
      "step": 134965
    },
    {
      "epoch": 14.848184818481847,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005666225433907217,
      "loss": 0.2309,
      "num_input_tokens_seen": 28485184,
      "step": 134970
    },
    {
      "epoch": 14.848734873487349,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005665098188221867,
      "loss": 0.2325,
      "num_input_tokens_seen": 28486208,
      "step": 134975
    },
    {
      "epoch": 14.84928492849285,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005663971028571829,
      "loss": 0.2313,
      "num_input_tokens_seen": 28487264,
      "step": 134980
    },
    {
      "epoch": 14.84983498349835,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00566284395496749,
      "loss": 0.2335,
      "num_input_tokens_seen": 28488288,
      "step": 134985
    },
    {
      "epoch": 14.85038503850385,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005661716967419242,
      "loss": 0.2308,
      "num_input_tokens_seen": 28489248,
      "step": 134990
    },
    {
      "epoch": 14.850935093509351,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.005660590065937469,
      "loss": 0.2314,
      "num_input_tokens_seen": 28490304,
      "step": 134995
    },
    {
      "epoch": 14.851485148514852,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00565946325053255,
      "loss": 0.2324,
      "num_input_tokens_seen": 28491360,
      "step": 135000
    },
    {
      "epoch": 14.852035203520352,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.00565833652121488,
      "loss": 0.2314,
      "num_input_tokens_seen": 28492416,
      "step": 135005
    },
    {
      "epoch": 14.852585258525853,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.005657209877994836,
      "loss": 0.2298,
      "num_input_tokens_seen": 28493504,
      "step": 135010
    },
    {
      "epoch": 14.853135313531354,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0056560833208828066,
      "loss": 0.2335,
      "num_input_tokens_seen": 28494656,
      "step": 135015
    },
    {
      "epoch": 14.853685368536853,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005654956849889177,
      "loss": 0.2324,
      "num_input_tokens_seen": 28495744,
      "step": 135020
    },
    {
      "epoch": 14.854235423542354,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.0056538304650243214,
      "loss": 0.2303,
      "num_input_tokens_seen": 28496832,
      "step": 135025
    },
    {
      "epoch": 14.854785478547855,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.005652704166298631,
      "loss": 0.2314,
      "num_input_tokens_seen": 28497888,
      "step": 135030
    },
    {
      "epoch": 14.855335533553355,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005651577953722476,
      "loss": 0.2314,
      "num_input_tokens_seen": 28498944,
      "step": 135035
    },
    {
      "epoch": 14.855885588558856,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005650451827306242,
      "loss": 0.2304,
      "num_input_tokens_seen": 28499968,
      "step": 135040
    },
    {
      "epoch": 14.856435643564357,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0056493257870603095,
      "loss": 0.2309,
      "num_input_tokens_seen": 28501024,
      "step": 135045
    },
    {
      "epoch": 14.856985698569858,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.005648199832995052,
      "loss": 0.2309,
      "num_input_tokens_seen": 28502112,
      "step": 135050
    },
    {
      "epoch": 14.857535753575357,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.005647073965120852,
      "loss": 0.2309,
      "num_input_tokens_seen": 28503168,
      "step": 135055
    },
    {
      "epoch": 14.858085808580858,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.00564594818344808,
      "loss": 0.2309,
      "num_input_tokens_seen": 28504224,
      "step": 135060
    },
    {
      "epoch": 14.85863586358636,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00564482248798712,
      "loss": 0.2314,
      "num_input_tokens_seen": 28505312,
      "step": 135065
    },
    {
      "epoch": 14.859185918591859,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.005643696878748337,
      "loss": 0.2324,
      "num_input_tokens_seen": 28506400,
      "step": 135070
    },
    {
      "epoch": 14.85973597359736,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005642571355742109,
      "loss": 0.2319,
      "num_input_tokens_seen": 28507424,
      "step": 135075
    },
    {
      "epoch": 14.86028602860286,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.005641445918978817,
      "loss": 0.2308,
      "num_input_tokens_seen": 28508416,
      "step": 135080
    },
    {
      "epoch": 14.86083608360836,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005640320568468821,
      "loss": 0.2309,
      "num_input_tokens_seen": 28509472,
      "step": 135085
    },
    {
      "epoch": 14.861386138613861,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005639195304222506,
      "loss": 0.2319,
      "num_input_tokens_seen": 28510496,
      "step": 135090
    },
    {
      "epoch": 14.861936193619362,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.00563807012625023,
      "loss": 0.2335,
      "num_input_tokens_seen": 28511616,
      "step": 135095
    },
    {
      "epoch": 14.862486248624862,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.005636945034562369,
      "loss": 0.2324,
      "num_input_tokens_seen": 28512672,
      "step": 135100
    },
    {
      "epoch": 14.863036303630363,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005635820029169299,
      "loss": 0.2303,
      "num_input_tokens_seen": 28513696,
      "step": 135105
    },
    {
      "epoch": 14.863586358635864,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005634695110081377,
      "loss": 0.2303,
      "num_input_tokens_seen": 28514752,
      "step": 135110
    },
    {
      "epoch": 14.864136413641365,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.00563357027730898,
      "loss": 0.2335,
      "num_input_tokens_seen": 28515936,
      "step": 135115
    },
    {
      "epoch": 14.864686468646864,
      "grad_norm": 0.005859375,
      "learning_rate": 0.005632445530862472,
      "loss": 0.2324,
      "num_input_tokens_seen": 28516992,
      "step": 135120
    },
    {
      "epoch": 14.865236523652365,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.005631320870752214,
      "loss": 0.2293,
      "num_input_tokens_seen": 28518016,
      "step": 135125
    },
    {
      "epoch": 14.865786578657866,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.005630196296988577,
      "loss": 0.2314,
      "num_input_tokens_seen": 28519072,
      "step": 135130
    },
    {
      "epoch": 14.866336633663366,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.005629071809581927,
      "loss": 0.2345,
      "num_input_tokens_seen": 28520128,
      "step": 135135
    },
    {
      "epoch": 14.866886688668867,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.005627947408542622,
      "loss": 0.2324,
      "num_input_tokens_seen": 28521216,
      "step": 135140
    },
    {
      "epoch": 14.867436743674368,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.0056268230938810344,
      "loss": 0.2314,
      "num_input_tokens_seen": 28522336,
      "step": 135145
    },
    {
      "epoch": 14.867986798679867,
      "grad_norm": 0.005859375,
      "learning_rate": 0.005625698865607515,
      "loss": 0.2329,
      "num_input_tokens_seen": 28523360,
      "step": 135150
    },
    {
      "epoch": 14.868536853685368,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00562457472373243,
      "loss": 0.2324,
      "num_input_tokens_seen": 28524416,
      "step": 135155
    },
    {
      "epoch": 14.86908690869087,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.005623450668266147,
      "loss": 0.2324,
      "num_input_tokens_seen": 28525504,
      "step": 135160
    },
    {
      "epoch": 14.869636963696369,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005622326699219015,
      "loss": 0.234,
      "num_input_tokens_seen": 28526528,
      "step": 135165
    },
    {
      "epoch": 14.87018701870187,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005621202816601402,
      "loss": 0.2298,
      "num_input_tokens_seen": 28527616,
      "step": 135170
    },
    {
      "epoch": 14.870737073707371,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0056200790204236585,
      "loss": 0.2303,
      "num_input_tokens_seen": 28528640,
      "step": 135175
    },
    {
      "epoch": 14.871287128712872,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0056189553106961495,
      "loss": 0.2324,
      "num_input_tokens_seen": 28529696,
      "step": 135180
    },
    {
      "epoch": 14.871837183718371,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.005617831687429223,
      "loss": 0.233,
      "num_input_tokens_seen": 28530720,
      "step": 135185
    },
    {
      "epoch": 14.872387238723872,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.00561670815063324,
      "loss": 0.2303,
      "num_input_tokens_seen": 28531776,
      "step": 135190
    },
    {
      "epoch": 14.872937293729374,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005615584700318562,
      "loss": 0.233,
      "num_input_tokens_seen": 28532800,
      "step": 135195
    },
    {
      "epoch": 14.873487348734873,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.00561446133649553,
      "loss": 0.2324,
      "num_input_tokens_seen": 28533824,
      "step": 135200
    },
    {
      "epoch": 14.874037403740374,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.00561333805917451,
      "loss": 0.233,
      "num_input_tokens_seen": 28534912,
      "step": 135205
    },
    {
      "epoch": 14.874587458745875,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.005612214868365844,
      "loss": 0.2324,
      "num_input_tokens_seen": 28536000,
      "step": 135210
    },
    {
      "epoch": 14.875137513751374,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.005611091764079887,
      "loss": 0.2324,
      "num_input_tokens_seen": 28537088,
      "step": 135215
    },
    {
      "epoch": 14.875687568756875,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.005609968746327,
      "loss": 0.2303,
      "num_input_tokens_seen": 28538176,
      "step": 135220
    },
    {
      "epoch": 14.876237623762377,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005608845815117518,
      "loss": 0.2303,
      "num_input_tokens_seen": 28539232,
      "step": 135225
    },
    {
      "epoch": 14.876787678767876,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.005607722970461804,
      "loss": 0.2314,
      "num_input_tokens_seen": 28540256,
      "step": 135230
    },
    {
      "epoch": 14.877337733773377,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.0056066002123702,
      "loss": 0.2288,
      "num_input_tokens_seen": 28541376,
      "step": 135235
    },
    {
      "epoch": 14.877887788778878,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00560547754085305,
      "loss": 0.2325,
      "num_input_tokens_seen": 28542432,
      "step": 135240
    },
    {
      "epoch": 14.87843784378438,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005604354955920705,
      "loss": 0.233,
      "num_input_tokens_seen": 28543488,
      "step": 135245
    },
    {
      "epoch": 14.878987898789878,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005603232457583512,
      "loss": 0.2324,
      "num_input_tokens_seen": 28544608,
      "step": 135250
    },
    {
      "epoch": 14.87953795379538,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.005602110045851822,
      "loss": 0.2304,
      "num_input_tokens_seen": 28545696,
      "step": 135255
    },
    {
      "epoch": 14.88008800880088,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0056009877207359735,
      "loss": 0.2303,
      "num_input_tokens_seen": 28546752,
      "step": 135260
    },
    {
      "epoch": 14.88063806380638,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.005599865482246306,
      "loss": 0.2329,
      "num_input_tokens_seen": 28547744,
      "step": 135265
    },
    {
      "epoch": 14.881188118811881,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.005598743330393169,
      "loss": 0.2303,
      "num_input_tokens_seen": 28548768,
      "step": 135270
    },
    {
      "epoch": 14.881738173817382,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.005597621265186902,
      "loss": 0.2309,
      "num_input_tokens_seen": 28549856,
      "step": 135275
    },
    {
      "epoch": 14.882288228822881,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.005596499286637854,
      "loss": 0.2303,
      "num_input_tokens_seen": 28550976,
      "step": 135280
    },
    {
      "epoch": 14.882838283828383,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.005595377394756361,
      "loss": 0.2335,
      "num_input_tokens_seen": 28552032,
      "step": 135285
    },
    {
      "epoch": 14.883388338833884,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.005594255589552754,
      "loss": 0.233,
      "num_input_tokens_seen": 28553088,
      "step": 135290
    },
    {
      "epoch": 14.883938393839383,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.005593133871037387,
      "loss": 0.2303,
      "num_input_tokens_seen": 28554208,
      "step": 135295
    },
    {
      "epoch": 14.884488448844884,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0055920122392205856,
      "loss": 0.2314,
      "num_input_tokens_seen": 28555232,
      "step": 135300
    },
    {
      "epoch": 14.885038503850385,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.005590890694112692,
      "loss": 0.2324,
      "num_input_tokens_seen": 28556352,
      "step": 135305
    },
    {
      "epoch": 14.885588558855886,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005589769235724051,
      "loss": 0.2298,
      "num_input_tokens_seen": 28557408,
      "step": 135310
    },
    {
      "epoch": 14.886138613861386,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.005588647864064984,
      "loss": 0.2309,
      "num_input_tokens_seen": 28558432,
      "step": 135315
    },
    {
      "epoch": 14.886688668866887,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00558752657914584,
      "loss": 0.2303,
      "num_input_tokens_seen": 28559488,
      "step": 135320
    },
    {
      "epoch": 14.887238723872388,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.005586405380976947,
      "loss": 0.2314,
      "num_input_tokens_seen": 28560480,
      "step": 135325
    },
    {
      "epoch": 14.887788778877887,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.005585284269568629,
      "loss": 0.2309,
      "num_input_tokens_seen": 28561536,
      "step": 135330
    },
    {
      "epoch": 14.888338833883388,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005584163244931237,
      "loss": 0.2319,
      "num_input_tokens_seen": 28562624,
      "step": 135335
    },
    {
      "epoch": 14.88888888888889,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.005583042307075089,
      "loss": 0.2319,
      "num_input_tokens_seen": 28563712,
      "step": 135340
    },
    {
      "epoch": 14.88943894389439,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005581921456010529,
      "loss": 0.2303,
      "num_input_tokens_seen": 28564736,
      "step": 135345
    },
    {
      "epoch": 14.88998899889989,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.005580800691747877,
      "loss": 0.2319,
      "num_input_tokens_seen": 28565760,
      "step": 135350
    },
    {
      "epoch": 14.89053905390539,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005579680014297462,
      "loss": 0.2335,
      "num_input_tokens_seen": 28566816,
      "step": 135355
    },
    {
      "epoch": 14.891089108910892,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005578559423669615,
      "loss": 0.2314,
      "num_input_tokens_seen": 28567936,
      "step": 135360
    },
    {
      "epoch": 14.891639163916391,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.005577438919874666,
      "loss": 0.2272,
      "num_input_tokens_seen": 28568960,
      "step": 135365
    },
    {
      "epoch": 14.892189218921892,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.005576318502922944,
      "loss": 0.2309,
      "num_input_tokens_seen": 28569984,
      "step": 135370
    },
    {
      "epoch": 14.892739273927393,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0055751981728247735,
      "loss": 0.233,
      "num_input_tokens_seen": 28570976,
      "step": 135375
    },
    {
      "epoch": 14.893289328932893,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005574077929590474,
      "loss": 0.2303,
      "num_input_tokens_seen": 28572000,
      "step": 135380
    },
    {
      "epoch": 14.893839383938394,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.0055729577732303815,
      "loss": 0.2324,
      "num_input_tokens_seen": 28573088,
      "step": 135385
    },
    {
      "epoch": 14.894389438943895,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.005571837703754802,
      "loss": 0.2314,
      "num_input_tokens_seen": 28574144,
      "step": 135390
    },
    {
      "epoch": 14.894939493949394,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005570717721174082,
      "loss": 0.2308,
      "num_input_tokens_seen": 28575104,
      "step": 135395
    },
    {
      "epoch": 14.895489548954895,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005569597825498531,
      "loss": 0.2298,
      "num_input_tokens_seen": 28576224,
      "step": 135400
    },
    {
      "epoch": 14.896039603960396,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0055684780167384665,
      "loss": 0.2314,
      "num_input_tokens_seen": 28577216,
      "step": 135405
    },
    {
      "epoch": 14.896589658965897,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.005567358294904219,
      "loss": 0.2288,
      "num_input_tokens_seen": 28578304,
      "step": 135410
    },
    {
      "epoch": 14.897139713971397,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.0055662386600061,
      "loss": 0.2298,
      "num_input_tokens_seen": 28579264,
      "step": 135415
    },
    {
      "epoch": 14.897689768976898,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.005565119112054431,
      "loss": 0.2293,
      "num_input_tokens_seen": 28580320,
      "step": 135420
    },
    {
      "epoch": 14.898239823982399,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005563999651059538,
      "loss": 0.2309,
      "num_input_tokens_seen": 28581344,
      "step": 135425
    },
    {
      "epoch": 14.898789878987898,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0055628802770317275,
      "loss": 0.2324,
      "num_input_tokens_seen": 28582464,
      "step": 135430
    },
    {
      "epoch": 14.8993399339934,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.005561760989981327,
      "loss": 0.2314,
      "num_input_tokens_seen": 28583520,
      "step": 135435
    },
    {
      "epoch": 14.8998899889989,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.00556064178991864,
      "loss": 0.2319,
      "num_input_tokens_seen": 28584544,
      "step": 135440
    },
    {
      "epoch": 14.9004400440044,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.005559522676853993,
      "loss": 0.233,
      "num_input_tokens_seen": 28585600,
      "step": 135445
    },
    {
      "epoch": 14.900990099009901,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.005558403650797693,
      "loss": 0.2324,
      "num_input_tokens_seen": 28586624,
      "step": 135450
    },
    {
      "epoch": 14.901540154015402,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005557284711760054,
      "loss": 0.2324,
      "num_input_tokens_seen": 28587648,
      "step": 135455
    },
    {
      "epoch": 14.902090209020901,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.005556165859751395,
      "loss": 0.2314,
      "num_input_tokens_seen": 28588736,
      "step": 135460
    },
    {
      "epoch": 14.902640264026402,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00555504709478202,
      "loss": 0.2309,
      "num_input_tokens_seen": 28589760,
      "step": 135465
    },
    {
      "epoch": 14.903190319031903,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005553928416862248,
      "loss": 0.2314,
      "num_input_tokens_seen": 28590784,
      "step": 135470
    },
    {
      "epoch": 14.903740374037405,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00555280982600238,
      "loss": 0.2314,
      "num_input_tokens_seen": 28591840,
      "step": 135475
    },
    {
      "epoch": 14.904290429042904,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.005551691322212732,
      "loss": 0.2309,
      "num_input_tokens_seen": 28592832,
      "step": 135480
    },
    {
      "epoch": 14.904840484048405,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005550572905503616,
      "loss": 0.2304,
      "num_input_tokens_seen": 28593792,
      "step": 135485
    },
    {
      "epoch": 14.905390539053906,
      "grad_norm": 0.005859375,
      "learning_rate": 0.005549454575885329,
      "loss": 0.233,
      "num_input_tokens_seen": 28594880,
      "step": 135490
    },
    {
      "epoch": 14.905940594059405,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.005548336333368191,
      "loss": 0.2303,
      "num_input_tokens_seen": 28595872,
      "step": 135495
    },
    {
      "epoch": 14.906490649064907,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0055472181779625,
      "loss": 0.234,
      "num_input_tokens_seen": 28596864,
      "step": 135500
    },
    {
      "epoch": 14.907040704070408,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0055461001096785536,
      "loss": 0.2309,
      "num_input_tokens_seen": 28597824,
      "step": 135505
    },
    {
      "epoch": 14.907590759075907,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0055449821285266775,
      "loss": 0.2298,
      "num_input_tokens_seen": 28598880,
      "step": 135510
    },
    {
      "epoch": 14.908140814081408,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.0055438642345171625,
      "loss": 0.2314,
      "num_input_tokens_seen": 28599904,
      "step": 135515
    },
    {
      "epoch": 14.908690869086909,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0055427464276603085,
      "loss": 0.2324,
      "num_input_tokens_seen": 28600992,
      "step": 135520
    },
    {
      "epoch": 14.909240924092408,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005541628707966428,
      "loss": 0.2314,
      "num_input_tokens_seen": 28602048,
      "step": 135525
    },
    {
      "epoch": 14.90979097909791,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0055405110754458125,
      "loss": 0.2309,
      "num_input_tokens_seen": 28603168,
      "step": 135530
    },
    {
      "epoch": 14.91034103410341,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.005539393530108768,
      "loss": 0.2329,
      "num_input_tokens_seen": 28604320,
      "step": 135535
    },
    {
      "epoch": 14.910891089108912,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0055382760719655975,
      "loss": 0.2335,
      "num_input_tokens_seen": 28605440,
      "step": 135540
    },
    {
      "epoch": 14.911441144114411,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.005537158701026591,
      "loss": 0.2303,
      "num_input_tokens_seen": 28606464,
      "step": 135545
    },
    {
      "epoch": 14.911991199119912,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.005536041417302058,
      "loss": 0.2303,
      "num_input_tokens_seen": 28607584,
      "step": 135550
    },
    {
      "epoch": 14.912541254125413,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0055349242208022845,
      "loss": 0.2314,
      "num_input_tokens_seen": 28608640,
      "step": 135555
    },
    {
      "epoch": 14.913091309130913,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005533807111537578,
      "loss": 0.2314,
      "num_input_tokens_seen": 28609696,
      "step": 135560
    },
    {
      "epoch": 14.913641364136414,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.0055326900895182225,
      "loss": 0.2314,
      "num_input_tokens_seen": 28610688,
      "step": 135565
    },
    {
      "epoch": 14.914191419141915,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0055315731547545205,
      "loss": 0.2319,
      "num_input_tokens_seen": 28611712,
      "step": 135570
    },
    {
      "epoch": 14.914741474147414,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.005530456307256772,
      "loss": 0.2329,
      "num_input_tokens_seen": 28612736,
      "step": 135575
    },
    {
      "epoch": 14.915291529152915,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.005529339547035257,
      "loss": 0.2324,
      "num_input_tokens_seen": 28613888,
      "step": 135580
    },
    {
      "epoch": 14.915841584158416,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.005528222874100281,
      "loss": 0.2298,
      "num_input_tokens_seen": 28614944,
      "step": 135585
    },
    {
      "epoch": 14.916391639163916,
      "grad_norm": 0.0026702880859375,
      "learning_rate": 0.005527106288462124,
      "loss": 0.2314,
      "num_input_tokens_seen": 28616064,
      "step": 135590
    },
    {
      "epoch": 14.916941694169417,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.005525989790131082,
      "loss": 0.2298,
      "num_input_tokens_seen": 28617184,
      "step": 135595
    },
    {
      "epoch": 14.917491749174918,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.005524873379117451,
      "loss": 0.2314,
      "num_input_tokens_seen": 28618304,
      "step": 135600
    },
    {
      "epoch": 14.918041804180419,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005523757055431513,
      "loss": 0.2303,
      "num_input_tokens_seen": 28619424,
      "step": 135605
    },
    {
      "epoch": 14.918591859185918,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.005522640819083562,
      "loss": 0.2309,
      "num_input_tokens_seen": 28620448,
      "step": 135610
    },
    {
      "epoch": 14.91914191419142,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005521524670083883,
      "loss": 0.234,
      "num_input_tokens_seen": 28621408,
      "step": 135615
    },
    {
      "epoch": 14.91969196919692,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.005520408608442759,
      "loss": 0.2314,
      "num_input_tokens_seen": 28622432,
      "step": 135620
    },
    {
      "epoch": 14.92024202420242,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005519292634170479,
      "loss": 0.2303,
      "num_input_tokens_seen": 28623456,
      "step": 135625
    },
    {
      "epoch": 14.92079207920792,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.005518176747277331,
      "loss": 0.2335,
      "num_input_tokens_seen": 28624448,
      "step": 135630
    },
    {
      "epoch": 14.921342134213422,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.005517060947773602,
      "loss": 0.233,
      "num_input_tokens_seen": 28625472,
      "step": 135635
    },
    {
      "epoch": 14.921892189218921,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005515945235669571,
      "loss": 0.2298,
      "num_input_tokens_seen": 28626464,
      "step": 135640
    },
    {
      "epoch": 14.922442244224422,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005514829610975519,
      "loss": 0.2293,
      "num_input_tokens_seen": 28627488,
      "step": 135645
    },
    {
      "epoch": 14.922992299229923,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.005513714073701729,
      "loss": 0.2325,
      "num_input_tokens_seen": 28628544,
      "step": 135650
    },
    {
      "epoch": 14.923542354235423,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.005512598623858485,
      "loss": 0.2319,
      "num_input_tokens_seen": 28629568,
      "step": 135655
    },
    {
      "epoch": 14.924092409240924,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.005511483261456073,
      "loss": 0.232,
      "num_input_tokens_seen": 28630592,
      "step": 135660
    },
    {
      "epoch": 14.924642464246425,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005510367986504765,
      "loss": 0.2293,
      "num_input_tokens_seen": 28631680,
      "step": 135665
    },
    {
      "epoch": 14.925192519251926,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.005509252799014838,
      "loss": 0.2309,
      "num_input_tokens_seen": 28632768,
      "step": 135670
    },
    {
      "epoch": 14.925742574257425,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005508137698996579,
      "loss": 0.2314,
      "num_input_tokens_seen": 28633792,
      "step": 135675
    },
    {
      "epoch": 14.926292629262926,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005507022686460255,
      "loss": 0.2319,
      "num_input_tokens_seen": 28634784,
      "step": 135680
    },
    {
      "epoch": 14.926842684268427,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.005505907761416147,
      "loss": 0.2324,
      "num_input_tokens_seen": 28635872,
      "step": 135685
    },
    {
      "epoch": 14.927392739273927,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.0055047929238745364,
      "loss": 0.2299,
      "num_input_tokens_seen": 28636896,
      "step": 135690
    },
    {
      "epoch": 14.927942794279428,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.005503678173845689,
      "loss": 0.2303,
      "num_input_tokens_seen": 28637920,
      "step": 135695
    },
    {
      "epoch": 14.928492849284929,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005502563511339887,
      "loss": 0.2314,
      "num_input_tokens_seen": 28639008,
      "step": 135700
    },
    {
      "epoch": 14.929042904290428,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.005501448936367396,
      "loss": 0.233,
      "num_input_tokens_seen": 28640064,
      "step": 135705
    },
    {
      "epoch": 14.92959295929593,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.005500334448938492,
      "loss": 0.2314,
      "num_input_tokens_seen": 28641120,
      "step": 135710
    },
    {
      "epoch": 14.93014301430143,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.005499220049063453,
      "loss": 0.2304,
      "num_input_tokens_seen": 28642144,
      "step": 135715
    },
    {
      "epoch": 14.930693069306932,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.005498105736752537,
      "loss": 0.2314,
      "num_input_tokens_seen": 28643136,
      "step": 135720
    },
    {
      "epoch": 14.93124312431243,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0054969915120160265,
      "loss": 0.2319,
      "num_input_tokens_seen": 28644224,
      "step": 135725
    },
    {
      "epoch": 14.931793179317932,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.005495877374864185,
      "loss": 0.2299,
      "num_input_tokens_seen": 28645312,
      "step": 135730
    },
    {
      "epoch": 14.932343234323433,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.005494763325307278,
      "loss": 0.2325,
      "num_input_tokens_seen": 28646368,
      "step": 135735
    },
    {
      "epoch": 14.932893289328932,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.0054936493633555745,
      "loss": 0.2298,
      "num_input_tokens_seen": 28647424,
      "step": 135740
    },
    {
      "epoch": 14.933443344334433,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.005492535489019344,
      "loss": 0.2319,
      "num_input_tokens_seen": 28648448,
      "step": 135745
    },
    {
      "epoch": 14.933993399339935,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.005491421702308858,
      "loss": 0.2309,
      "num_input_tokens_seen": 28649440,
      "step": 135750
    },
    {
      "epoch": 14.934543454345434,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.005490308003234374,
      "loss": 0.2304,
      "num_input_tokens_seen": 28650560,
      "step": 135755
    },
    {
      "epoch": 14.935093509350935,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.005489194391806152,
      "loss": 0.2298,
      "num_input_tokens_seen": 28651616,
      "step": 135760
    },
    {
      "epoch": 14.935643564356436,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.005488080868034462,
      "loss": 0.233,
      "num_input_tokens_seen": 28652704,
      "step": 135765
    },
    {
      "epoch": 14.936193619361937,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005486967431929568,
      "loss": 0.2304,
      "num_input_tokens_seen": 28653728,
      "step": 135770
    },
    {
      "epoch": 14.936743674367436,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.005485854083501734,
      "loss": 0.2304,
      "num_input_tokens_seen": 28654720,
      "step": 135775
    },
    {
      "epoch": 14.937293729372938,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005484740822761218,
      "loss": 0.2314,
      "num_input_tokens_seen": 28655840,
      "step": 135780
    },
    {
      "epoch": 14.937843784378439,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0054836276497182744,
      "loss": 0.2309,
      "num_input_tokens_seen": 28656928,
      "step": 135785
    },
    {
      "epoch": 14.938393839383938,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.005482514564383173,
      "loss": 0.2319,
      "num_input_tokens_seen": 28657984,
      "step": 135790
    },
    {
      "epoch": 14.938943894389439,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.005481401566766165,
      "loss": 0.2309,
      "num_input_tokens_seen": 28659040,
      "step": 135795
    },
    {
      "epoch": 14.93949394939494,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00548028865687751,
      "loss": 0.2303,
      "num_input_tokens_seen": 28660160,
      "step": 135800
    },
    {
      "epoch": 14.94004400440044,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0054791758347274705,
      "loss": 0.2309,
      "num_input_tokens_seen": 28661216,
      "step": 135805
    },
    {
      "epoch": 14.94059405940594,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.005478063100326295,
      "loss": 0.2319,
      "num_input_tokens_seen": 28662208,
      "step": 135810
    },
    {
      "epoch": 14.941144114411442,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005476950453684248,
      "loss": 0.2303,
      "num_input_tokens_seen": 28663328,
      "step": 135815
    },
    {
      "epoch": 14.941694169416941,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.005475837894811574,
      "loss": 0.2314,
      "num_input_tokens_seen": 28664384,
      "step": 135820
    },
    {
      "epoch": 14.942244224422442,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.005474725423718531,
      "loss": 0.2329,
      "num_input_tokens_seen": 28665472,
      "step": 135825
    },
    {
      "epoch": 14.942794279427943,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.00547361304041538,
      "loss": 0.2319,
      "num_input_tokens_seen": 28666496,
      "step": 135830
    },
    {
      "epoch": 14.943344334433444,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00547250074491236,
      "loss": 0.2299,
      "num_input_tokens_seen": 28667616,
      "step": 135835
    },
    {
      "epoch": 14.943894389438944,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.005471388537219735,
      "loss": 0.2314,
      "num_input_tokens_seen": 28668672,
      "step": 135840
    },
    {
      "epoch": 14.944444444444445,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.005470276417347743,
      "loss": 0.2309,
      "num_input_tokens_seen": 28669792,
      "step": 135845
    },
    {
      "epoch": 14.944994499449946,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.005469164385306645,
      "loss": 0.2309,
      "num_input_tokens_seen": 28670784,
      "step": 135850
    },
    {
      "epoch": 14.945544554455445,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.005468052441106683,
      "loss": 0.2314,
      "num_input_tokens_seen": 28671872,
      "step": 135855
    },
    {
      "epoch": 14.946094609460946,
      "grad_norm": 0.000732421875,
      "learning_rate": 0.005466940584758106,
      "loss": 0.2314,
      "num_input_tokens_seen": 28672928,
      "step": 135860
    },
    {
      "epoch": 14.946644664466447,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0054658288162711685,
      "loss": 0.2335,
      "num_input_tokens_seen": 28674016,
      "step": 135865
    },
    {
      "epoch": 14.947194719471947,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.005464717135656106,
      "loss": 0.2298,
      "num_input_tokens_seen": 28675040,
      "step": 135870
    },
    {
      "epoch": 14.947744774477448,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.005463605542923177,
      "loss": 0.2319,
      "num_input_tokens_seen": 28676096,
      "step": 135875
    },
    {
      "epoch": 14.948294829482949,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.005462494038082614,
      "loss": 0.2314,
      "num_input_tokens_seen": 28677056,
      "step": 135880
    },
    {
      "epoch": 14.948844884488448,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005461382621144668,
      "loss": 0.2324,
      "num_input_tokens_seen": 28678080,
      "step": 135885
    },
    {
      "epoch": 14.94939493949395,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.005460271292119584,
      "loss": 0.2325,
      "num_input_tokens_seen": 28679168,
      "step": 135890
    },
    {
      "epoch": 14.94994499449945,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0054591600510175975,
      "loss": 0.2324,
      "num_input_tokens_seen": 28680192,
      "step": 135895
    },
    {
      "epoch": 14.950495049504951,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.005458048897848961,
      "loss": 0.2319,
      "num_input_tokens_seen": 28681184,
      "step": 135900
    },
    {
      "epoch": 14.95104510451045,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.005456937832623908,
      "loss": 0.2335,
      "num_input_tokens_seen": 28682272,
      "step": 135905
    },
    {
      "epoch": 14.951595159515952,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.005455826855352675,
      "loss": 0.2335,
      "num_input_tokens_seen": 28683264,
      "step": 135910
    },
    {
      "epoch": 14.952145214521453,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.005454715966045507,
      "loss": 0.2298,
      "num_input_tokens_seen": 28684288,
      "step": 135915
    },
    {
      "epoch": 14.952695269526952,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.005453605164712645,
      "loss": 0.2309,
      "num_input_tokens_seen": 28685344,
      "step": 135920
    },
    {
      "epoch": 14.953245324532453,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00545249445136432,
      "loss": 0.2308,
      "num_input_tokens_seen": 28686368,
      "step": 135925
    },
    {
      "epoch": 14.953795379537954,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.005451383826010777,
      "loss": 0.2308,
      "num_input_tokens_seen": 28687424,
      "step": 135930
    },
    {
      "epoch": 14.954345434543454,
      "grad_norm": 0.005859375,
      "learning_rate": 0.005450273288662241,
      "loss": 0.2309,
      "num_input_tokens_seen": 28688448,
      "step": 135935
    },
    {
      "epoch": 14.954895489548955,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.005449162839328955,
      "loss": 0.2309,
      "num_input_tokens_seen": 28689568,
      "step": 135940
    },
    {
      "epoch": 14.955445544554456,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.005448052478021157,
      "loss": 0.233,
      "num_input_tokens_seen": 28690624,
      "step": 135945
    },
    {
      "epoch": 14.955995599559955,
      "grad_norm": 0.0027313232421875,
      "learning_rate": 0.005446942204749071,
      "loss": 0.2319,
      "num_input_tokens_seen": 28691616,
      "step": 135950
    },
    {
      "epoch": 14.956545654565456,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00544583201952294,
      "loss": 0.2288,
      "num_input_tokens_seen": 28692704,
      "step": 135955
    },
    {
      "epoch": 14.957095709570957,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.0054447219223529835,
      "loss": 0.2314,
      "num_input_tokens_seen": 28693760,
      "step": 135960
    },
    {
      "epoch": 14.957645764576458,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.0054436119132494475,
      "loss": 0.2324,
      "num_input_tokens_seen": 28694880,
      "step": 135965
    },
    {
      "epoch": 14.958195819581958,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.005442501992222549,
      "loss": 0.2298,
      "num_input_tokens_seen": 28695936,
      "step": 135970
    },
    {
      "epoch": 14.958745874587459,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.005441392159282524,
      "loss": 0.2314,
      "num_input_tokens_seen": 28696992,
      "step": 135975
    },
    {
      "epoch": 14.95929592959296,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0054402824144396044,
      "loss": 0.2314,
      "num_input_tokens_seen": 28698016,
      "step": 135980
    },
    {
      "epoch": 14.95984598459846,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005439172757704011,
      "loss": 0.2335,
      "num_input_tokens_seen": 28699040,
      "step": 135985
    },
    {
      "epoch": 14.96039603960396,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0054380631890859794,
      "loss": 0.2319,
      "num_input_tokens_seen": 28700064,
      "step": 135990
    },
    {
      "epoch": 14.960946094609461,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005436953708595726,
      "loss": 0.2319,
      "num_input_tokens_seen": 28701152,
      "step": 135995
    },
    {
      "epoch": 14.96149614961496,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005435844316243481,
      "loss": 0.2314,
      "num_input_tokens_seen": 28702208,
      "step": 136000
    },
    {
      "epoch": 14.962046204620462,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.005434735012039475,
      "loss": 0.2314,
      "num_input_tokens_seen": 28703328,
      "step": 136005
    },
    {
      "epoch": 14.962596259625963,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.005433625795993921,
      "loss": 0.2335,
      "num_input_tokens_seen": 28704384,
      "step": 136010
    },
    {
      "epoch": 14.963146314631462,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.005432516668117052,
      "loss": 0.2335,
      "num_input_tokens_seen": 28705440,
      "step": 136015
    },
    {
      "epoch": 14.963696369636963,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0054314076284190885,
      "loss": 0.2309,
      "num_input_tokens_seen": 28706528,
      "step": 136020
    },
    {
      "epoch": 14.964246424642464,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005430298676910242,
      "loss": 0.2319,
      "num_input_tokens_seen": 28707488,
      "step": 136025
    },
    {
      "epoch": 14.964796479647966,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.0054291898136007415,
      "loss": 0.2304,
      "num_input_tokens_seen": 28708512,
      "step": 136030
    },
    {
      "epoch": 14.965346534653465,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.005428081038500806,
      "loss": 0.2319,
      "num_input_tokens_seen": 28709568,
      "step": 136035
    },
    {
      "epoch": 14.965896589658966,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005426972351620661,
      "loss": 0.2309,
      "num_input_tokens_seen": 28710624,
      "step": 136040
    },
    {
      "epoch": 14.966446644664467,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.005425863752970516,
      "loss": 0.2335,
      "num_input_tokens_seen": 28711712,
      "step": 136045
    },
    {
      "epoch": 14.966996699669966,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.005424755242560586,
      "loss": 0.2304,
      "num_input_tokens_seen": 28712736,
      "step": 136050
    },
    {
      "epoch": 14.967546754675467,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.005423646820401094,
      "loss": 0.2309,
      "num_input_tokens_seen": 28713824,
      "step": 136055
    },
    {
      "epoch": 14.968096809680969,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.005422538486502252,
      "loss": 0.2319,
      "num_input_tokens_seen": 28714912,
      "step": 136060
    },
    {
      "epoch": 14.968646864686468,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.005421430240874283,
      "loss": 0.2309,
      "num_input_tokens_seen": 28715936,
      "step": 136065
    },
    {
      "epoch": 14.969196919691969,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.005420322083527394,
      "loss": 0.2303,
      "num_input_tokens_seen": 28716992,
      "step": 136070
    },
    {
      "epoch": 14.96974697469747,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.0054192140144717965,
      "loss": 0.2309,
      "num_input_tokens_seen": 28718080,
      "step": 136075
    },
    {
      "epoch": 14.97029702970297,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.005418106033717711,
      "loss": 0.2325,
      "num_input_tokens_seen": 28719072,
      "step": 136080
    },
    {
      "epoch": 14.97084708470847,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005416998141275339,
      "loss": 0.2345,
      "num_input_tokens_seen": 28720192,
      "step": 136085
    },
    {
      "epoch": 14.971397139713972,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005415890337154895,
      "loss": 0.2335,
      "num_input_tokens_seen": 28721248,
      "step": 136090
    },
    {
      "epoch": 14.971947194719473,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.005414782621366598,
      "loss": 0.2324,
      "num_input_tokens_seen": 28722272,
      "step": 136095
    },
    {
      "epoch": 14.972497249724972,
      "grad_norm": 0.005859375,
      "learning_rate": 0.005413674993920644,
      "loss": 0.2309,
      "num_input_tokens_seen": 28723328,
      "step": 136100
    },
    {
      "epoch": 14.973047304730473,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005412567454827254,
      "loss": 0.2308,
      "num_input_tokens_seen": 28724416,
      "step": 136105
    },
    {
      "epoch": 14.973597359735974,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.005411460004096627,
      "loss": 0.2304,
      "num_input_tokens_seen": 28725472,
      "step": 136110
    },
    {
      "epoch": 14.974147414741473,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.005410352641738963,
      "loss": 0.2314,
      "num_input_tokens_seen": 28726560,
      "step": 136115
    },
    {
      "epoch": 14.974697469746975,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.005409245367764488,
      "loss": 0.2298,
      "num_input_tokens_seen": 28727584,
      "step": 136120
    },
    {
      "epoch": 14.975247524752476,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0054081381821833915,
      "loss": 0.2329,
      "num_input_tokens_seen": 28728576,
      "step": 136125
    },
    {
      "epoch": 14.975797579757975,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0054070310850058865,
      "loss": 0.2314,
      "num_input_tokens_seen": 28729664,
      "step": 136130
    },
    {
      "epoch": 14.976347634763476,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.005405924076242175,
      "loss": 0.2309,
      "num_input_tokens_seen": 28730720,
      "step": 136135
    },
    {
      "epoch": 14.976897689768977,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00540481715590245,
      "loss": 0.233,
      "num_input_tokens_seen": 28731712,
      "step": 136140
    },
    {
      "epoch": 14.977447744774478,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.005403710323996923,
      "loss": 0.233,
      "num_input_tokens_seen": 28732768,
      "step": 136145
    },
    {
      "epoch": 14.977997799779978,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005402603580535791,
      "loss": 0.2288,
      "num_input_tokens_seen": 28733792,
      "step": 136150
    },
    {
      "epoch": 14.978547854785479,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.005401496925529263,
      "loss": 0.233,
      "num_input_tokens_seen": 28734816,
      "step": 136155
    },
    {
      "epoch": 14.97909790979098,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.005400390358987532,
      "loss": 0.2309,
      "num_input_tokens_seen": 28735808,
      "step": 136160
    },
    {
      "epoch": 14.979647964796479,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.0053992838809207916,
      "loss": 0.2293,
      "num_input_tokens_seen": 28736896,
      "step": 136165
    },
    {
      "epoch": 14.98019801980198,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.005398177491339251,
      "loss": 0.2319,
      "num_input_tokens_seen": 28737920,
      "step": 136170
    },
    {
      "epoch": 14.980748074807481,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.005397071190253089,
      "loss": 0.234,
      "num_input_tokens_seen": 28738976,
      "step": 136175
    },
    {
      "epoch": 14.98129812981298,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.005395964977672525,
      "loss": 0.2298,
      "num_input_tokens_seen": 28740032,
      "step": 136180
    },
    {
      "epoch": 14.981848184818482,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.005394858853607742,
      "loss": 0.2319,
      "num_input_tokens_seen": 28741088,
      "step": 136185
    },
    {
      "epoch": 14.982398239823983,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005393752818068934,
      "loss": 0.2309,
      "num_input_tokens_seen": 28742176,
      "step": 136190
    },
    {
      "epoch": 14.982948294829484,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005392646871066299,
      "loss": 0.2304,
      "num_input_tokens_seen": 28743168,
      "step": 136195
    },
    {
      "epoch": 14.983498349834983,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.005391541012610024,
      "loss": 0.2314,
      "num_input_tokens_seen": 28744224,
      "step": 136200
    },
    {
      "epoch": 14.984048404840484,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.005390435242710303,
      "loss": 0.2288,
      "num_input_tokens_seen": 28745280,
      "step": 136205
    },
    {
      "epoch": 14.984598459845985,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.005389329561377337,
      "loss": 0.2309,
      "num_input_tokens_seen": 28746304,
      "step": 136210
    },
    {
      "epoch": 14.985148514851485,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005388223968621301,
      "loss": 0.2309,
      "num_input_tokens_seen": 28747392,
      "step": 136215
    },
    {
      "epoch": 14.985698569856986,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.005387118464452401,
      "loss": 0.233,
      "num_input_tokens_seen": 28748448,
      "step": 136220
    },
    {
      "epoch": 14.986248624862487,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0053860130488808085,
      "loss": 0.2325,
      "num_input_tokens_seen": 28749568,
      "step": 136225
    },
    {
      "epoch": 14.986798679867986,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.005384907721916728,
      "loss": 0.2314,
      "num_input_tokens_seen": 28750624,
      "step": 136230
    },
    {
      "epoch": 14.987348734873487,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.005383802483570335,
      "loss": 0.2319,
      "num_input_tokens_seen": 28751648,
      "step": 136235
    },
    {
      "epoch": 14.987898789878988,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.005382697333851817,
      "loss": 0.2319,
      "num_input_tokens_seen": 28752736,
      "step": 136240
    },
    {
      "epoch": 14.988448844884488,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.005381592272771371,
      "loss": 0.2335,
      "num_input_tokens_seen": 28753792,
      "step": 136245
    },
    {
      "epoch": 14.988998899889989,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.005380487300339167,
      "loss": 0.2345,
      "num_input_tokens_seen": 28754880,
      "step": 136250
    },
    {
      "epoch": 14.98954895489549,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.005379382416565401,
      "loss": 0.2324,
      "num_input_tokens_seen": 28755936,
      "step": 136255
    },
    {
      "epoch": 14.990099009900991,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.005378277621460246,
      "loss": 0.2314,
      "num_input_tokens_seen": 28757024,
      "step": 136260
    },
    {
      "epoch": 14.99064906490649,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0053771729150338884,
      "loss": 0.2324,
      "num_input_tokens_seen": 28758080,
      "step": 136265
    },
    {
      "epoch": 14.991199119911991,
      "grad_norm": 0.005859375,
      "learning_rate": 0.005376068297296517,
      "loss": 0.2314,
      "num_input_tokens_seen": 28759136,
      "step": 136270
    },
    {
      "epoch": 14.991749174917492,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.005374963768258301,
      "loss": 0.2309,
      "num_input_tokens_seen": 28760224,
      "step": 136275
    },
    {
      "epoch": 14.992299229922992,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005373859327929428,
      "loss": 0.2298,
      "num_input_tokens_seen": 28761280,
      "step": 136280
    },
    {
      "epoch": 14.992849284928493,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.005372754976320077,
      "loss": 0.2309,
      "num_input_tokens_seen": 28762368,
      "step": 136285
    },
    {
      "epoch": 14.993399339933994,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.005371650713440412,
      "loss": 0.2298,
      "num_input_tokens_seen": 28763456,
      "step": 136290
    },
    {
      "epoch": 14.993949394939493,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.005370546539300634,
      "loss": 0.2324,
      "num_input_tokens_seen": 28764544,
      "step": 136295
    },
    {
      "epoch": 14.994499449944994,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0053694424539109074,
      "loss": 0.2324,
      "num_input_tokens_seen": 28765600,
      "step": 136300
    },
    {
      "epoch": 14.995049504950495,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.005368338457281403,
      "loss": 0.2309,
      "num_input_tokens_seen": 28766592,
      "step": 136305
    },
    {
      "epoch": 14.995599559955995,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.005367234549422306,
      "loss": 0.2319,
      "num_input_tokens_seen": 28767584,
      "step": 136310
    },
    {
      "epoch": 14.996149614961496,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005366130730343782,
      "loss": 0.2304,
      "num_input_tokens_seen": 28768608,
      "step": 136315
    },
    {
      "epoch": 14.996699669966997,
      "grad_norm": 0.0031585693359375,
      "learning_rate": 0.005365027000056008,
      "loss": 0.2314,
      "num_input_tokens_seen": 28769664,
      "step": 136320
    },
    {
      "epoch": 14.997249724972498,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.005363923358569162,
      "loss": 0.2324,
      "num_input_tokens_seen": 28770752,
      "step": 136325
    },
    {
      "epoch": 14.997799779977997,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0053628198058934045,
      "loss": 0.2319,
      "num_input_tokens_seen": 28771776,
      "step": 136330
    },
    {
      "epoch": 14.998349834983498,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005361716342038919,
      "loss": 0.2319,
      "num_input_tokens_seen": 28772768,
      "step": 136335
    },
    {
      "epoch": 14.998899889989,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.005360612967015862,
      "loss": 0.2324,
      "num_input_tokens_seen": 28773888,
      "step": 136340
    },
    {
      "epoch": 14.999449944994499,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005359509680834417,
      "loss": 0.233,
      "num_input_tokens_seen": 28774880,
      "step": 136345
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0053584064835047384,
      "loss": 0.2319,
      "num_input_tokens_seen": 28775840,
      "step": 136350
    },
    {
      "epoch": 15.0,
      "eval_loss": 0.23128081858158112,
      "eval_runtime": 60.5619,
      "eval_samples_per_second": 66.709,
      "eval_steps_per_second": 16.677,
      "num_input_tokens_seen": 28775840,
      "step": 136350
    },
    {
      "epoch": 15.000550055005501,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005357303375037003,
      "loss": 0.2329,
      "num_input_tokens_seen": 28776896,
      "step": 136355
    },
    {
      "epoch": 15.001100110011,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.005356200355441379,
      "loss": 0.2299,
      "num_input_tokens_seen": 28777952,
      "step": 136360
    },
    {
      "epoch": 15.001650165016502,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.005355097424728023,
      "loss": 0.2319,
      "num_input_tokens_seen": 28778912,
      "step": 136365
    },
    {
      "epoch": 15.002200220022003,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.005353994582907112,
      "loss": 0.2335,
      "num_input_tokens_seen": 28780000,
      "step": 136370
    },
    {
      "epoch": 15.002750275027502,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0053528918299888,
      "loss": 0.2314,
      "num_input_tokens_seen": 28781088,
      "step": 136375
    },
    {
      "epoch": 15.003300330033003,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005351789165983253,
      "loss": 0.2309,
      "num_input_tokens_seen": 28782144,
      "step": 136380
    },
    {
      "epoch": 15.003850385038504,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005350686590900642,
      "loss": 0.2319,
      "num_input_tokens_seen": 28783200,
      "step": 136385
    },
    {
      "epoch": 15.004400440044005,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005349584104751117,
      "loss": 0.2324,
      "num_input_tokens_seen": 28784288,
      "step": 136390
    },
    {
      "epoch": 15.004950495049505,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.0053484817075448495,
      "loss": 0.2314,
      "num_input_tokens_seen": 28785344,
      "step": 136395
    },
    {
      "epoch": 15.005500550055006,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0053473793992919955,
      "loss": 0.2314,
      "num_input_tokens_seen": 28786400,
      "step": 136400
    },
    {
      "epoch": 15.006050605060507,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.005346277180002707,
      "loss": 0.2298,
      "num_input_tokens_seen": 28787488,
      "step": 136405
    },
    {
      "epoch": 15.006600660066006,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.005345175049687151,
      "loss": 0.2329,
      "num_input_tokens_seen": 28788480,
      "step": 136410
    },
    {
      "epoch": 15.007150715071507,
      "grad_norm": 0.005859375,
      "learning_rate": 0.005344073008355483,
      "loss": 0.2303,
      "num_input_tokens_seen": 28789600,
      "step": 136415
    },
    {
      "epoch": 15.007700770077008,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0053429710560178685,
      "loss": 0.2319,
      "num_input_tokens_seen": 28790592,
      "step": 136420
    },
    {
      "epoch": 15.008250825082508,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.005341869192684453,
      "loss": 0.2319,
      "num_input_tokens_seen": 28791584,
      "step": 136425
    },
    {
      "epoch": 15.008800880088009,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00534076741836539,
      "loss": 0.2319,
      "num_input_tokens_seen": 28792704,
      "step": 136430
    },
    {
      "epoch": 15.00935093509351,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00533966573307084,
      "loss": 0.2314,
      "num_input_tokens_seen": 28793728,
      "step": 136435
    },
    {
      "epoch": 15.009900990099009,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.005338564136810954,
      "loss": 0.2298,
      "num_input_tokens_seen": 28794720,
      "step": 136440
    },
    {
      "epoch": 15.01045104510451,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.005337462629595892,
      "loss": 0.2314,
      "num_input_tokens_seen": 28795776,
      "step": 136445
    },
    {
      "epoch": 15.011001100110011,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005336361211435801,
      "loss": 0.2303,
      "num_input_tokens_seen": 28796864,
      "step": 136450
    },
    {
      "epoch": 15.011551155115512,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.005335259882340826,
      "loss": 0.2314,
      "num_input_tokens_seen": 28797888,
      "step": 136455
    },
    {
      "epoch": 15.012101210121012,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.005334158642321126,
      "loss": 0.2303,
      "num_input_tokens_seen": 28799008,
      "step": 136460
    },
    {
      "epoch": 15.012651265126513,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0053330574913868466,
      "loss": 0.2319,
      "num_input_tokens_seen": 28800096,
      "step": 136465
    },
    {
      "epoch": 15.013201320132014,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0053319564295481355,
      "loss": 0.2319,
      "num_input_tokens_seen": 28801152,
      "step": 136470
    },
    {
      "epoch": 15.013751375137513,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.005330855456815148,
      "loss": 0.2314,
      "num_input_tokens_seen": 28802176,
      "step": 136475
    },
    {
      "epoch": 15.014301430143014,
      "grad_norm": 0.0006256103515625,
      "learning_rate": 0.005329754573198021,
      "loss": 0.2304,
      "num_input_tokens_seen": 28803232,
      "step": 136480
    },
    {
      "epoch": 15.014851485148515,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005328653778706911,
      "loss": 0.233,
      "num_input_tokens_seen": 28804288,
      "step": 136485
    },
    {
      "epoch": 15.015401540154015,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.005327553073351953,
      "loss": 0.2319,
      "num_input_tokens_seen": 28805344,
      "step": 136490
    },
    {
      "epoch": 15.015951595159516,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.005326452457143298,
      "loss": 0.2329,
      "num_input_tokens_seen": 28806336,
      "step": 136495
    },
    {
      "epoch": 15.016501650165017,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.005325351930091092,
      "loss": 0.2303,
      "num_input_tokens_seen": 28807424,
      "step": 136500
    },
    {
      "epoch": 15.017051705170518,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.005324251492205472,
      "loss": 0.2314,
      "num_input_tokens_seen": 28808416,
      "step": 136505
    },
    {
      "epoch": 15.017601760176017,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.005323151143496588,
      "loss": 0.2303,
      "num_input_tokens_seen": 28809504,
      "step": 136510
    },
    {
      "epoch": 15.018151815181518,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.005322050883974575,
      "loss": 0.2299,
      "num_input_tokens_seen": 28810528,
      "step": 136515
    },
    {
      "epoch": 15.01870187018702,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0053209507136495705,
      "loss": 0.2319,
      "num_input_tokens_seen": 28811520,
      "step": 136520
    },
    {
      "epoch": 15.019251925192519,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.00531985063253172,
      "loss": 0.233,
      "num_input_tokens_seen": 28812672,
      "step": 136525
    },
    {
      "epoch": 15.01980198019802,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00531875064063116,
      "loss": 0.2309,
      "num_input_tokens_seen": 28813664,
      "step": 136530
    },
    {
      "epoch": 15.020352035203521,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.005317650737958036,
      "loss": 0.2309,
      "num_input_tokens_seen": 28814656,
      "step": 136535
    },
    {
      "epoch": 15.02090209020902,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005316550924522477,
      "loss": 0.2314,
      "num_input_tokens_seen": 28815712,
      "step": 136540
    },
    {
      "epoch": 15.021452145214521,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.005315451200334618,
      "loss": 0.2304,
      "num_input_tokens_seen": 28816832,
      "step": 136545
    },
    {
      "epoch": 15.022002200220022,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.005314351565404599,
      "loss": 0.2319,
      "num_input_tokens_seen": 28817856,
      "step": 136550
    },
    {
      "epoch": 15.022552255225522,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.005313252019742552,
      "loss": 0.2325,
      "num_input_tokens_seen": 28818880,
      "step": 136555
    },
    {
      "epoch": 15.023102310231023,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.005312152563358621,
      "loss": 0.2314,
      "num_input_tokens_seen": 28819968,
      "step": 136560
    },
    {
      "epoch": 15.023652365236524,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.005311053196262929,
      "loss": 0.2303,
      "num_input_tokens_seen": 28820928,
      "step": 136565
    },
    {
      "epoch": 15.024202420242025,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.005309953918465605,
      "loss": 0.2293,
      "num_input_tokens_seen": 28821920,
      "step": 136570
    },
    {
      "epoch": 15.024752475247524,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0053088547299767926,
      "loss": 0.2324,
      "num_input_tokens_seen": 28822944,
      "step": 136575
    },
    {
      "epoch": 15.025302530253025,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.005307755630806609,
      "loss": 0.2335,
      "num_input_tokens_seen": 28823968,
      "step": 136580
    },
    {
      "epoch": 15.025852585258527,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.005306656620965192,
      "loss": 0.2325,
      "num_input_tokens_seen": 28825024,
      "step": 136585
    },
    {
      "epoch": 15.026402640264026,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.005305557700462675,
      "loss": 0.2319,
      "num_input_tokens_seen": 28826080,
      "step": 136590
    },
    {
      "epoch": 15.026952695269527,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.005304458869309176,
      "loss": 0.2314,
      "num_input_tokens_seen": 28827136,
      "step": 136595
    },
    {
      "epoch": 15.027502750275028,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.005303360127514831,
      "loss": 0.2319,
      "num_input_tokens_seen": 28828224,
      "step": 136600
    },
    {
      "epoch": 15.028052805280527,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.005302261475089759,
      "loss": 0.2298,
      "num_input_tokens_seen": 28829248,
      "step": 136605
    },
    {
      "epoch": 15.028602860286028,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005301162912044087,
      "loss": 0.2304,
      "num_input_tokens_seen": 28830240,
      "step": 136610
    },
    {
      "epoch": 15.02915291529153,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.005300064438387951,
      "loss": 0.2319,
      "num_input_tokens_seen": 28831328,
      "step": 136615
    },
    {
      "epoch": 15.029702970297029,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.005298966054131459,
      "loss": 0.2319,
      "num_input_tokens_seen": 28832384,
      "step": 136620
    },
    {
      "epoch": 15.03025302530253,
      "grad_norm": 0.002716064453125,
      "learning_rate": 0.005297867759284747,
      "loss": 0.2345,
      "num_input_tokens_seen": 28833408,
      "step": 136625
    },
    {
      "epoch": 15.030803080308031,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.005296769553857929,
      "loss": 0.2309,
      "num_input_tokens_seen": 28834464,
      "step": 136630
    },
    {
      "epoch": 15.031353135313532,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005295671437861135,
      "loss": 0.2314,
      "num_input_tokens_seen": 28835488,
      "step": 136635
    },
    {
      "epoch": 15.031903190319031,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.005294573411304475,
      "loss": 0.2298,
      "num_input_tokens_seen": 28836576,
      "step": 136640
    },
    {
      "epoch": 15.032453245324533,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0052934754741980755,
      "loss": 0.2314,
      "num_input_tokens_seen": 28837664,
      "step": 136645
    },
    {
      "epoch": 15.033003300330034,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.0052923776265520585,
      "loss": 0.2308,
      "num_input_tokens_seen": 28838720,
      "step": 136650
    },
    {
      "epoch": 15.033553355335533,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.005291279868376533,
      "loss": 0.2309,
      "num_input_tokens_seen": 28839776,
      "step": 136655
    },
    {
      "epoch": 15.034103410341034,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.005290182199681631,
      "loss": 0.2314,
      "num_input_tokens_seen": 28840800,
      "step": 136660
    },
    {
      "epoch": 15.034653465346535,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.005289084620477452,
      "loss": 0.2304,
      "num_input_tokens_seen": 28841888,
      "step": 136665
    },
    {
      "epoch": 15.035203520352034,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00528798713077412,
      "loss": 0.2324,
      "num_input_tokens_seen": 28842944,
      "step": 136670
    },
    {
      "epoch": 15.035753575357536,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005286889730581757,
      "loss": 0.2298,
      "num_input_tokens_seen": 28844000,
      "step": 136675
    },
    {
      "epoch": 15.036303630363037,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005285792419910471,
      "loss": 0.2319,
      "num_input_tokens_seen": 28845024,
      "step": 136680
    },
    {
      "epoch": 15.036853685368538,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.00528469519877037,
      "loss": 0.2319,
      "num_input_tokens_seen": 28846112,
      "step": 136685
    },
    {
      "epoch": 15.037403740374037,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.005283598067171576,
      "loss": 0.233,
      "num_input_tokens_seen": 28847104,
      "step": 136690
    },
    {
      "epoch": 15.037953795379538,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.005282501025124189,
      "loss": 0.2314,
      "num_input_tokens_seen": 28848160,
      "step": 136695
    },
    {
      "epoch": 15.03850385038504,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.00528140407263833,
      "loss": 0.2309,
      "num_input_tokens_seen": 28849152,
      "step": 136700
    },
    {
      "epoch": 15.039053905390539,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.005280307209724111,
      "loss": 0.2319,
      "num_input_tokens_seen": 28850272,
      "step": 136705
    },
    {
      "epoch": 15.03960396039604,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00527921043639163,
      "loss": 0.2319,
      "num_input_tokens_seen": 28851296,
      "step": 136710
    },
    {
      "epoch": 15.04015401540154,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005278113752651008,
      "loss": 0.2309,
      "num_input_tokens_seen": 28852320,
      "step": 136715
    },
    {
      "epoch": 15.04070407040704,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.005277017158512341,
      "loss": 0.2288,
      "num_input_tokens_seen": 28853376,
      "step": 136720
    },
    {
      "epoch": 15.041254125412541,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.005275920653985741,
      "loss": 0.2309,
      "num_input_tokens_seen": 28854400,
      "step": 136725
    },
    {
      "epoch": 15.041804180418042,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005274824239081321,
      "loss": 0.2319,
      "num_input_tokens_seen": 28855424,
      "step": 136730
    },
    {
      "epoch": 15.042354235423542,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.005273727913809172,
      "loss": 0.2319,
      "num_input_tokens_seen": 28856480,
      "step": 136735
    },
    {
      "epoch": 15.042904290429043,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.005272631678179413,
      "loss": 0.2319,
      "num_input_tokens_seen": 28857568,
      "step": 136740
    },
    {
      "epoch": 15.043454345434544,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.005271535532202135,
      "loss": 0.2319,
      "num_input_tokens_seen": 28858688,
      "step": 136745
    },
    {
      "epoch": 15.044004400440045,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.005270439475887448,
      "loss": 0.233,
      "num_input_tokens_seen": 28859712,
      "step": 136750
    },
    {
      "epoch": 15.044554455445544,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.005269343509245449,
      "loss": 0.2324,
      "num_input_tokens_seen": 28860800,
      "step": 136755
    },
    {
      "epoch": 15.045104510451045,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.005268247632286241,
      "loss": 0.2303,
      "num_input_tokens_seen": 28861824,
      "step": 136760
    },
    {
      "epoch": 15.045654565456546,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.00526715184501993,
      "loss": 0.2314,
      "num_input_tokens_seen": 28862880,
      "step": 136765
    },
    {
      "epoch": 15.046204620462046,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.005266056147456605,
      "loss": 0.2319,
      "num_input_tokens_seen": 28863936,
      "step": 136770
    },
    {
      "epoch": 15.046754675467547,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.005264960539606375,
      "loss": 0.2309,
      "num_input_tokens_seen": 28865056,
      "step": 136775
    },
    {
      "epoch": 15.047304730473048,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.005263865021479327,
      "loss": 0.234,
      "num_input_tokens_seen": 28866048,
      "step": 136780
    },
    {
      "epoch": 15.047854785478547,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.005262769593085564,
      "loss": 0.2314,
      "num_input_tokens_seen": 28867072,
      "step": 136785
    },
    {
      "epoch": 15.048404840484048,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0052616742544351865,
      "loss": 0.2314,
      "num_input_tokens_seen": 28868128,
      "step": 136790
    },
    {
      "epoch": 15.04895489548955,
      "grad_norm": 0.00066375732421875,
      "learning_rate": 0.005260579005538278,
      "loss": 0.2319,
      "num_input_tokens_seen": 28869152,
      "step": 136795
    },
    {
      "epoch": 15.049504950495049,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.005259483846404945,
      "loss": 0.233,
      "num_input_tokens_seen": 28870240,
      "step": 136800
    },
    {
      "epoch": 15.05005500550055,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0052583887770452755,
      "loss": 0.2298,
      "num_input_tokens_seen": 28871264,
      "step": 136805
    },
    {
      "epoch": 15.05060506050605,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0052572937974693585,
      "loss": 0.2314,
      "num_input_tokens_seen": 28872288,
      "step": 136810
    },
    {
      "epoch": 15.051155115511552,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.005256198907687288,
      "loss": 0.2335,
      "num_input_tokens_seen": 28873280,
      "step": 136815
    },
    {
      "epoch": 15.051705170517051,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005255104107709158,
      "loss": 0.2314,
      "num_input_tokens_seen": 28874304,
      "step": 136820
    },
    {
      "epoch": 15.052255225522552,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00525400939754506,
      "loss": 0.2293,
      "num_input_tokens_seen": 28875328,
      "step": 136825
    },
    {
      "epoch": 15.052805280528053,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005252914777205083,
      "loss": 0.2314,
      "num_input_tokens_seen": 28876448,
      "step": 136830
    },
    {
      "epoch": 15.053355335533553,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.005251820246699307,
      "loss": 0.2314,
      "num_input_tokens_seen": 28877504,
      "step": 136835
    },
    {
      "epoch": 15.053905390539054,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.005250725806037827,
      "loss": 0.2314,
      "num_input_tokens_seen": 28878624,
      "step": 136840
    },
    {
      "epoch": 15.054455445544555,
      "grad_norm": 0.005859375,
      "learning_rate": 0.005249631455230728,
      "loss": 0.2314,
      "num_input_tokens_seen": 28879648,
      "step": 136845
    },
    {
      "epoch": 15.055005500550054,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.005248537194288104,
      "loss": 0.234,
      "num_input_tokens_seen": 28880672,
      "step": 136850
    },
    {
      "epoch": 15.055555555555555,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.005247443023220033,
      "loss": 0.2319,
      "num_input_tokens_seen": 28881728,
      "step": 136855
    },
    {
      "epoch": 15.056105610561056,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.005246348942036593,
      "loss": 0.2309,
      "num_input_tokens_seen": 28882816,
      "step": 136860
    },
    {
      "epoch": 15.056655665566556,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005245254950747882,
      "loss": 0.2293,
      "num_input_tokens_seen": 28883904,
      "step": 136865
    },
    {
      "epoch": 15.057205720572057,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.005244161049363968,
      "loss": 0.2329,
      "num_input_tokens_seen": 28885024,
      "step": 136870
    },
    {
      "epoch": 15.057755775577558,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.005243067237894942,
      "loss": 0.2319,
      "num_input_tokens_seen": 28886080,
      "step": 136875
    },
    {
      "epoch": 15.058305830583059,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.005241973516350888,
      "loss": 0.2314,
      "num_input_tokens_seen": 28887136,
      "step": 136880
    },
    {
      "epoch": 15.058855885588558,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.005240879884741877,
      "loss": 0.2319,
      "num_input_tokens_seen": 28888256,
      "step": 136885
    },
    {
      "epoch": 15.05940594059406,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005239786343077998,
      "loss": 0.2314,
      "num_input_tokens_seen": 28889344,
      "step": 136890
    },
    {
      "epoch": 15.05995599559956,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.005238692891369324,
      "loss": 0.233,
      "num_input_tokens_seen": 28890432,
      "step": 136895
    },
    {
      "epoch": 15.06050605060506,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.0052375995296259255,
      "loss": 0.2319,
      "num_input_tokens_seen": 28891488,
      "step": 136900
    },
    {
      "epoch": 15.061056105610561,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.005236506257857896,
      "loss": 0.2304,
      "num_input_tokens_seen": 28892544,
      "step": 136905
    },
    {
      "epoch": 15.061606160616062,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.005235413076075299,
      "loss": 0.2324,
      "num_input_tokens_seen": 28893632,
      "step": 136910
    },
    {
      "epoch": 15.062156215621561,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.005234319984288219,
      "loss": 0.2319,
      "num_input_tokens_seen": 28894720,
      "step": 136915
    },
    {
      "epoch": 15.062706270627062,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.005233226982506727,
      "loss": 0.2335,
      "num_input_tokens_seen": 28895776,
      "step": 136920
    },
    {
      "epoch": 15.063256325632564,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005232134070740889,
      "loss": 0.2335,
      "num_input_tokens_seen": 28896864,
      "step": 136925
    },
    {
      "epoch": 15.063806380638065,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.005231041249000785,
      "loss": 0.2314,
      "num_input_tokens_seen": 28897888,
      "step": 136930
    },
    {
      "epoch": 15.064356435643564,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.005229948517296485,
      "loss": 0.233,
      "num_input_tokens_seen": 28898912,
      "step": 136935
    },
    {
      "epoch": 15.064906490649065,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0052288558756380665,
      "loss": 0.2309,
      "num_input_tokens_seen": 28900000,
      "step": 136940
    },
    {
      "epoch": 15.065456545654566,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.005227763324035595,
      "loss": 0.2319,
      "num_input_tokens_seen": 28901088,
      "step": 136945
    },
    {
      "epoch": 15.066006600660065,
      "grad_norm": 0.003265380859375,
      "learning_rate": 0.0052266708624991364,
      "loss": 0.2283,
      "num_input_tokens_seen": 28902144,
      "step": 136950
    },
    {
      "epoch": 15.066556655665567,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.005225578491038766,
      "loss": 0.2308,
      "num_input_tokens_seen": 28903136,
      "step": 136955
    },
    {
      "epoch": 15.067106710671068,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005224486209664539,
      "loss": 0.2329,
      "num_input_tokens_seen": 28904256,
      "step": 136960
    },
    {
      "epoch": 15.067656765676567,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.005223394018386542,
      "loss": 0.2324,
      "num_input_tokens_seen": 28905312,
      "step": 136965
    },
    {
      "epoch": 15.068206820682068,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00522230191721483,
      "loss": 0.2329,
      "num_input_tokens_seen": 28906368,
      "step": 136970
    },
    {
      "epoch": 15.06875687568757,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.005221209906159465,
      "loss": 0.2314,
      "num_input_tokens_seen": 28907424,
      "step": 136975
    },
    {
      "epoch": 15.069306930693068,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.005220117985230522,
      "loss": 0.2319,
      "num_input_tokens_seen": 28908416,
      "step": 136980
    },
    {
      "epoch": 15.06985698569857,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.005219026154438053,
      "loss": 0.2308,
      "num_input_tokens_seen": 28909408,
      "step": 136985
    },
    {
      "epoch": 15.07040704070407,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.005217934413792127,
      "loss": 0.2329,
      "num_input_tokens_seen": 28910496,
      "step": 136990
    },
    {
      "epoch": 15.070957095709572,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.005216842763302809,
      "loss": 0.2303,
      "num_input_tokens_seen": 28911552,
      "step": 136995
    },
    {
      "epoch": 15.071507150715071,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005215751202980152,
      "loss": 0.2293,
      "num_input_tokens_seen": 28912576,
      "step": 137000
    },
    {
      "epoch": 15.072057205720572,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005214659732834228,
      "loss": 0.2308,
      "num_input_tokens_seen": 28913600,
      "step": 137005
    },
    {
      "epoch": 15.072607260726073,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005213568352875083,
      "loss": 0.2298,
      "num_input_tokens_seen": 28914688,
      "step": 137010
    },
    {
      "epoch": 15.073157315731573,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.005212477063112789,
      "loss": 0.2314,
      "num_input_tokens_seen": 28915712,
      "step": 137015
    },
    {
      "epoch": 15.073707370737074,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005211385863557391,
      "loss": 0.2303,
      "num_input_tokens_seen": 28916800,
      "step": 137020
    },
    {
      "epoch": 15.074257425742575,
      "grad_norm": 0.000732421875,
      "learning_rate": 0.0052102947542189545,
      "loss": 0.2314,
      "num_input_tokens_seen": 28917888,
      "step": 137025
    },
    {
      "epoch": 15.074807480748074,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005209203735107537,
      "loss": 0.2308,
      "num_input_tokens_seen": 28918976,
      "step": 137030
    },
    {
      "epoch": 15.075357535753575,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005208112806233185,
      "loss": 0.2335,
      "num_input_tokens_seen": 28920032,
      "step": 137035
    },
    {
      "epoch": 15.075907590759076,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.005207021967605965,
      "loss": 0.2335,
      "num_input_tokens_seen": 28921088,
      "step": 137040
    },
    {
      "epoch": 15.076457645764576,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.00520593121923592,
      "loss": 0.2319,
      "num_input_tokens_seen": 28922176,
      "step": 137045
    },
    {
      "epoch": 15.077007700770077,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.005204840561133107,
      "loss": 0.2319,
      "num_input_tokens_seen": 28923296,
      "step": 137050
    },
    {
      "epoch": 15.077557755775578,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.005203749993307581,
      "loss": 0.2314,
      "num_input_tokens_seen": 28924320,
      "step": 137055
    },
    {
      "epoch": 15.078107810781079,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0052026595157693934,
      "loss": 0.2303,
      "num_input_tokens_seen": 28925344,
      "step": 137060
    },
    {
      "epoch": 15.078657865786578,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005201569128528586,
      "loss": 0.2309,
      "num_input_tokens_seen": 28926400,
      "step": 137065
    },
    {
      "epoch": 15.07920792079208,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0052004788315952195,
      "loss": 0.2303,
      "num_input_tokens_seen": 28927424,
      "step": 137070
    },
    {
      "epoch": 15.07975797579758,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.005199388624979325,
      "loss": 0.2303,
      "num_input_tokens_seen": 28928544,
      "step": 137075
    },
    {
      "epoch": 15.08030803080308,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005198298508690976,
      "loss": 0.2303,
      "num_input_tokens_seen": 28929568,
      "step": 137080
    },
    {
      "epoch": 15.08085808580858,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.0051972084827402055,
      "loss": 0.2314,
      "num_input_tokens_seen": 28930688,
      "step": 137085
    },
    {
      "epoch": 15.081408140814082,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.005196118547137054,
      "loss": 0.2303,
      "num_input_tokens_seen": 28931808,
      "step": 137090
    },
    {
      "epoch": 15.081958195819581,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.005195028701891579,
      "loss": 0.2319,
      "num_input_tokens_seen": 28932768,
      "step": 137095
    },
    {
      "epoch": 15.082508250825082,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.005193938947013816,
      "loss": 0.2303,
      "num_input_tokens_seen": 28933856,
      "step": 137100
    },
    {
      "epoch": 15.083058305830583,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.005192849282513811,
      "loss": 0.2308,
      "num_input_tokens_seen": 28934912,
      "step": 137105
    },
    {
      "epoch": 15.083608360836084,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005191759708401613,
      "loss": 0.2288,
      "num_input_tokens_seen": 28936000,
      "step": 137110
    },
    {
      "epoch": 15.084158415841584,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0051906702246872554,
      "loss": 0.2308,
      "num_input_tokens_seen": 28937088,
      "step": 137115
    },
    {
      "epoch": 15.084708470847085,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.005189580831380788,
      "loss": 0.2309,
      "num_input_tokens_seen": 28938144,
      "step": 137120
    },
    {
      "epoch": 15.085258525852586,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.005188491528492241,
      "loss": 0.2314,
      "num_input_tokens_seen": 28939168,
      "step": 137125
    },
    {
      "epoch": 15.085808580858085,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.005187402316031665,
      "loss": 0.2309,
      "num_input_tokens_seen": 28940288,
      "step": 137130
    },
    {
      "epoch": 15.086358635863586,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005186313194009089,
      "loss": 0.2324,
      "num_input_tokens_seen": 28941408,
      "step": 137135
    },
    {
      "epoch": 15.086908690869087,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005185224162434554,
      "loss": 0.2324,
      "num_input_tokens_seen": 28942496,
      "step": 137140
    },
    {
      "epoch": 15.087458745874587,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005184135221318105,
      "loss": 0.2319,
      "num_input_tokens_seen": 28943520,
      "step": 137145
    },
    {
      "epoch": 15.088008800880088,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.005183046370669764,
      "loss": 0.2309,
      "num_input_tokens_seen": 28944544,
      "step": 137150
    },
    {
      "epoch": 15.088558855885589,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.005181957610499581,
      "loss": 0.2319,
      "num_input_tokens_seen": 28945568,
      "step": 137155
    },
    {
      "epoch": 15.089108910891088,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.005180868940817579,
      "loss": 0.2308,
      "num_input_tokens_seen": 28946624,
      "step": 137160
    },
    {
      "epoch": 15.08965896589659,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.005179780361633794,
      "loss": 0.2319,
      "num_input_tokens_seen": 28947712,
      "step": 137165
    },
    {
      "epoch": 15.09020902090209,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.005178691872958267,
      "loss": 0.2309,
      "num_input_tokens_seen": 28948800,
      "step": 137170
    },
    {
      "epoch": 15.090759075907592,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00517760347480102,
      "loss": 0.2308,
      "num_input_tokens_seen": 28949888,
      "step": 137175
    },
    {
      "epoch": 15.091309130913091,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005176515167172094,
      "loss": 0.2303,
      "num_input_tokens_seen": 28951008,
      "step": 137180
    },
    {
      "epoch": 15.091859185918592,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.005175426950081514,
      "loss": 0.2293,
      "num_input_tokens_seen": 28952096,
      "step": 137185
    },
    {
      "epoch": 15.092409240924093,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005174338823539303,
      "loss": 0.2303,
      "num_input_tokens_seen": 28953184,
      "step": 137190
    },
    {
      "epoch": 15.092959295929592,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.005173250787555497,
      "loss": 0.2329,
      "num_input_tokens_seen": 28954240,
      "step": 137195
    },
    {
      "epoch": 15.093509350935093,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005172162842140122,
      "loss": 0.2319,
      "num_input_tokens_seen": 28955264,
      "step": 137200
    },
    {
      "epoch": 15.094059405940595,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0051710749873032135,
      "loss": 0.2309,
      "num_input_tokens_seen": 28956320,
      "step": 137205
    },
    {
      "epoch": 15.094609460946094,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00516998722305479,
      "loss": 0.2304,
      "num_input_tokens_seen": 28957408,
      "step": 137210
    },
    {
      "epoch": 15.095159515951595,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.00516889954940487,
      "loss": 0.2329,
      "num_input_tokens_seen": 28958496,
      "step": 137215
    },
    {
      "epoch": 15.095709570957096,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005167811966363487,
      "loss": 0.2308,
      "num_input_tokens_seen": 28959520,
      "step": 137220
    },
    {
      "epoch": 15.096259625962595,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.005166724473940664,
      "loss": 0.2335,
      "num_input_tokens_seen": 28960544,
      "step": 137225
    },
    {
      "epoch": 15.096809680968097,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.005165637072146427,
      "loss": 0.2314,
      "num_input_tokens_seen": 28961600,
      "step": 137230
    },
    {
      "epoch": 15.097359735973598,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.005164549760990795,
      "loss": 0.2314,
      "num_input_tokens_seen": 28962688,
      "step": 137235
    },
    {
      "epoch": 15.097909790979099,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.005163462540483783,
      "loss": 0.2303,
      "num_input_tokens_seen": 28963744,
      "step": 137240
    },
    {
      "epoch": 15.098459845984598,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.005162375410635421,
      "loss": 0.2314,
      "num_input_tokens_seen": 28964832,
      "step": 137245
    },
    {
      "epoch": 15.099009900990099,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00516128837145572,
      "loss": 0.2314,
      "num_input_tokens_seen": 28965888,
      "step": 137250
    },
    {
      "epoch": 15.0995599559956,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.005160201422954703,
      "loss": 0.2314,
      "num_input_tokens_seen": 28966976,
      "step": 137255
    },
    {
      "epoch": 15.1001100110011,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.005159114565142392,
      "loss": 0.2298,
      "num_input_tokens_seen": 28968032,
      "step": 137260
    },
    {
      "epoch": 15.1006600660066,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005158027798028797,
      "loss": 0.2324,
      "num_input_tokens_seen": 28969120,
      "step": 137265
    },
    {
      "epoch": 15.101210121012102,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.00515694112162394,
      "loss": 0.2319,
      "num_input_tokens_seen": 28970208,
      "step": 137270
    },
    {
      "epoch": 15.101760176017601,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.005155854535937829,
      "loss": 0.2293,
      "num_input_tokens_seen": 28971264,
      "step": 137275
    },
    {
      "epoch": 15.102310231023102,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005154768040980484,
      "loss": 0.2329,
      "num_input_tokens_seen": 28972320,
      "step": 137280
    },
    {
      "epoch": 15.102860286028603,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.005153681636761921,
      "loss": 0.2324,
      "num_input_tokens_seen": 28973440,
      "step": 137285
    },
    {
      "epoch": 15.103410341034103,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005152595323292147,
      "loss": 0.2309,
      "num_input_tokens_seen": 28974496,
      "step": 137290
    },
    {
      "epoch": 15.103960396039604,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.005151509100581181,
      "loss": 0.2319,
      "num_input_tokens_seen": 28975520,
      "step": 137295
    },
    {
      "epoch": 15.104510451045105,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005150422968639029,
      "loss": 0.2314,
      "num_input_tokens_seen": 28976512,
      "step": 137300
    },
    {
      "epoch": 15.105060506050606,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005149336927475697,
      "loss": 0.2319,
      "num_input_tokens_seen": 28977472,
      "step": 137305
    },
    {
      "epoch": 15.105610561056105,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0051482509771012,
      "loss": 0.2329,
      "num_input_tokens_seen": 28978560,
      "step": 137310
    },
    {
      "epoch": 15.106160616061606,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.005147165117525546,
      "loss": 0.2329,
      "num_input_tokens_seen": 28979552,
      "step": 137315
    },
    {
      "epoch": 15.106710671067107,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.00514607934875875,
      "loss": 0.2309,
      "num_input_tokens_seen": 28980544,
      "step": 137320
    },
    {
      "epoch": 15.107260726072607,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.005144993670810812,
      "loss": 0.2314,
      "num_input_tokens_seen": 28981600,
      "step": 137325
    },
    {
      "epoch": 15.107810781078108,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0051439080836917315,
      "loss": 0.2324,
      "num_input_tokens_seen": 28982624,
      "step": 137330
    },
    {
      "epoch": 15.108360836083609,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.005142822587411523,
      "loss": 0.2319,
      "num_input_tokens_seen": 28983648,
      "step": 137335
    },
    {
      "epoch": 15.108910891089108,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.005141737181980187,
      "loss": 0.2293,
      "num_input_tokens_seen": 28984704,
      "step": 137340
    },
    {
      "epoch": 15.10946094609461,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005140651867407735,
      "loss": 0.2314,
      "num_input_tokens_seen": 28985792,
      "step": 137345
    },
    {
      "epoch": 15.11001100110011,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.005139566643704164,
      "loss": 0.2303,
      "num_input_tokens_seen": 28986880,
      "step": 137350
    },
    {
      "epoch": 15.110561056105611,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.005138481510879469,
      "loss": 0.2335,
      "num_input_tokens_seen": 28987904,
      "step": 137355
    },
    {
      "epoch": 15.11111111111111,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.005137396468943664,
      "loss": 0.2293,
      "num_input_tokens_seen": 28989024,
      "step": 137360
    },
    {
      "epoch": 15.111661166116612,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.005136311517906739,
      "loss": 0.2303,
      "num_input_tokens_seen": 28990144,
      "step": 137365
    },
    {
      "epoch": 15.112211221122113,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005135226657778697,
      "loss": 0.2324,
      "num_input_tokens_seen": 28991200,
      "step": 137370
    },
    {
      "epoch": 15.112761276127612,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005134141888569541,
      "loss": 0.2309,
      "num_input_tokens_seen": 28992256,
      "step": 137375
    },
    {
      "epoch": 15.113311331133113,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.005133057210289262,
      "loss": 0.2319,
      "num_input_tokens_seen": 28993376,
      "step": 137380
    },
    {
      "epoch": 15.113861386138614,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005131972622947864,
      "loss": 0.2309,
      "num_input_tokens_seen": 28994432,
      "step": 137385
    },
    {
      "epoch": 15.114411441144114,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.005130888126555335,
      "loss": 0.2298,
      "num_input_tokens_seen": 28995456,
      "step": 137390
    },
    {
      "epoch": 15.114961496149615,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005129803721121674,
      "loss": 0.2319,
      "num_input_tokens_seen": 28996512,
      "step": 137395
    },
    {
      "epoch": 15.115511551155116,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00512871940665688,
      "loss": 0.2288,
      "num_input_tokens_seen": 28997568,
      "step": 137400
    },
    {
      "epoch": 15.116061606160615,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00512763518317094,
      "loss": 0.2314,
      "num_input_tokens_seen": 28998592,
      "step": 137405
    },
    {
      "epoch": 15.116611661166116,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.005126551050673853,
      "loss": 0.2329,
      "num_input_tokens_seen": 28999680,
      "step": 137410
    },
    {
      "epoch": 15.117161716171617,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.005125467009175603,
      "loss": 0.2324,
      "num_input_tokens_seen": 29000736,
      "step": 137415
    },
    {
      "epoch": 15.117711771177119,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.005124383058686188,
      "loss": 0.2298,
      "num_input_tokens_seen": 29001760,
      "step": 137420
    },
    {
      "epoch": 15.118261826182618,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0051232991992155915,
      "loss": 0.2293,
      "num_input_tokens_seen": 29002848,
      "step": 137425
    },
    {
      "epoch": 15.118811881188119,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005122215430773808,
      "loss": 0.2314,
      "num_input_tokens_seen": 29003872,
      "step": 137430
    },
    {
      "epoch": 15.11936193619362,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00512113175337083,
      "loss": 0.2298,
      "num_input_tokens_seen": 29004896,
      "step": 137435
    },
    {
      "epoch": 15.11991199119912,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005120048167016634,
      "loss": 0.2293,
      "num_input_tokens_seen": 29005984,
      "step": 137440
    },
    {
      "epoch": 15.12046204620462,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00511896467172122,
      "loss": 0.2314,
      "num_input_tokens_seen": 29006976,
      "step": 137445
    },
    {
      "epoch": 15.121012101210122,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00511788126749456,
      "loss": 0.2314,
      "num_input_tokens_seen": 29008064,
      "step": 137450
    },
    {
      "epoch": 15.12156215621562,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.005116797954346648,
      "loss": 0.2324,
      "num_input_tokens_seen": 29009120,
      "step": 137455
    },
    {
      "epoch": 15.122112211221122,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.005115714732287472,
      "loss": 0.2319,
      "num_input_tokens_seen": 29010080,
      "step": 137460
    },
    {
      "epoch": 15.122662266226623,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.0051146316013270104,
      "loss": 0.2319,
      "num_input_tokens_seen": 29011200,
      "step": 137465
    },
    {
      "epoch": 15.123212321232122,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.005113548561475239,
      "loss": 0.2319,
      "num_input_tokens_seen": 29012256,
      "step": 137470
    },
    {
      "epoch": 15.123762376237623,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.005112465612742153,
      "loss": 0.2324,
      "num_input_tokens_seen": 29013312,
      "step": 137475
    },
    {
      "epoch": 15.124312431243125,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005111382755137723,
      "loss": 0.2308,
      "num_input_tokens_seen": 29014368,
      "step": 137480
    },
    {
      "epoch": 15.124862486248626,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.005110299988671932,
      "loss": 0.2335,
      "num_input_tokens_seen": 29015520,
      "step": 137485
    },
    {
      "epoch": 15.125412541254125,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.005109217313354764,
      "loss": 0.2319,
      "num_input_tokens_seen": 29016576,
      "step": 137490
    },
    {
      "epoch": 15.125962596259626,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.0051081347291961916,
      "loss": 0.2293,
      "num_input_tokens_seen": 29017632,
      "step": 137495
    },
    {
      "epoch": 15.126512651265127,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.005107052236206198,
      "loss": 0.2314,
      "num_input_tokens_seen": 29018720,
      "step": 137500
    },
    {
      "epoch": 15.127062706270626,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005105969834394751,
      "loss": 0.2308,
      "num_input_tokens_seen": 29019744,
      "step": 137505
    },
    {
      "epoch": 15.127612761276128,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.005104887523771836,
      "loss": 0.2319,
      "num_input_tokens_seen": 29020800,
      "step": 137510
    },
    {
      "epoch": 15.128162816281629,
      "grad_norm": 0.0006561279296875,
      "learning_rate": 0.005103805304347427,
      "loss": 0.2308,
      "num_input_tokens_seen": 29021856,
      "step": 137515
    },
    {
      "epoch": 15.128712871287128,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.005102723176131491,
      "loss": 0.2319,
      "num_input_tokens_seen": 29022944,
      "step": 137520
    },
    {
      "epoch": 15.129262926292629,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005101641139134011,
      "loss": 0.2314,
      "num_input_tokens_seen": 29024064,
      "step": 137525
    },
    {
      "epoch": 15.12981298129813,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005100559193364951,
      "loss": 0.2314,
      "num_input_tokens_seen": 29025152,
      "step": 137530
    },
    {
      "epoch": 15.130363036303631,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00509947733883429,
      "loss": 0.2335,
      "num_input_tokens_seen": 29026208,
      "step": 137535
    },
    {
      "epoch": 15.13091309130913,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005098395575551993,
      "loss": 0.2304,
      "num_input_tokens_seen": 29027264,
      "step": 137540
    },
    {
      "epoch": 15.131463146314632,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005097313903528032,
      "loss": 0.2314,
      "num_input_tokens_seen": 29028256,
      "step": 137545
    },
    {
      "epoch": 15.132013201320133,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005096232322772382,
      "loss": 0.2314,
      "num_input_tokens_seen": 29029280,
      "step": 137550
    },
    {
      "epoch": 15.132563256325632,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.005095150833295,
      "loss": 0.2309,
      "num_input_tokens_seen": 29030304,
      "step": 137555
    },
    {
      "epoch": 15.133113311331133,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.005094069435105867,
      "loss": 0.2303,
      "num_input_tokens_seen": 29031360,
      "step": 137560
    },
    {
      "epoch": 15.133663366336634,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.005092988128214935,
      "loss": 0.2335,
      "num_input_tokens_seen": 29032448,
      "step": 137565
    },
    {
      "epoch": 15.134213421342134,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005091906912632179,
      "loss": 0.2314,
      "num_input_tokens_seen": 29033472,
      "step": 137570
    },
    {
      "epoch": 15.134763476347635,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0050908257883675685,
      "loss": 0.2319,
      "num_input_tokens_seen": 29034496,
      "step": 137575
    },
    {
      "epoch": 15.135313531353136,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.0050897447554310565,
      "loss": 0.2314,
      "num_input_tokens_seen": 29035456,
      "step": 137580
    },
    {
      "epoch": 15.135863586358635,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.005088663813832616,
      "loss": 0.2329,
      "num_input_tokens_seen": 29036512,
      "step": 137585
    },
    {
      "epoch": 15.136413641364136,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.005087582963582205,
      "loss": 0.2314,
      "num_input_tokens_seen": 29037536,
      "step": 137590
    },
    {
      "epoch": 15.136963696369637,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005086502204689781,
      "loss": 0.2309,
      "num_input_tokens_seen": 29038656,
      "step": 137595
    },
    {
      "epoch": 15.137513751375138,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.00508542153716531,
      "loss": 0.2324,
      "num_input_tokens_seen": 29039680,
      "step": 137600
    },
    {
      "epoch": 15.138063806380638,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0050843409610187495,
      "loss": 0.2293,
      "num_input_tokens_seen": 29040768,
      "step": 137605
    },
    {
      "epoch": 15.138613861386139,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.005083260476260065,
      "loss": 0.2319,
      "num_input_tokens_seen": 29041792,
      "step": 137610
    },
    {
      "epoch": 15.13916391639164,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005082180082899212,
      "loss": 0.2298,
      "num_input_tokens_seen": 29042848,
      "step": 137615
    },
    {
      "epoch": 15.13971397139714,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.005081099780946141,
      "loss": 0.2324,
      "num_input_tokens_seen": 29043904,
      "step": 137620
    },
    {
      "epoch": 15.14026402640264,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.005080019570410813,
      "loss": 0.2319,
      "num_input_tokens_seen": 29044896,
      "step": 137625
    },
    {
      "epoch": 15.140814081408141,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.005078939451303185,
      "loss": 0.2303,
      "num_input_tokens_seen": 29046016,
      "step": 137630
    },
    {
      "epoch": 15.14136413641364,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005077859423633216,
      "loss": 0.2335,
      "num_input_tokens_seen": 29047040,
      "step": 137635
    },
    {
      "epoch": 15.141914191419142,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.0050767794874108565,
      "loss": 0.2319,
      "num_input_tokens_seen": 29048096,
      "step": 137640
    },
    {
      "epoch": 15.142464246424643,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.005075699642646054,
      "loss": 0.2314,
      "num_input_tokens_seen": 29049152,
      "step": 137645
    },
    {
      "epoch": 15.143014301430142,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005074619889348769,
      "loss": 0.233,
      "num_input_tokens_seen": 29050176,
      "step": 137650
    },
    {
      "epoch": 15.143564356435643,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.005073540227528945,
      "loss": 0.2335,
      "num_input_tokens_seen": 29051200,
      "step": 137655
    },
    {
      "epoch": 15.144114411441144,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.005072460657196539,
      "loss": 0.2329,
      "num_input_tokens_seen": 29052256,
      "step": 137660
    },
    {
      "epoch": 15.144664466446645,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.005071381178361504,
      "loss": 0.2314,
      "num_input_tokens_seen": 29053312,
      "step": 137665
    },
    {
      "epoch": 15.145214521452145,
      "grad_norm": 0.005859375,
      "learning_rate": 0.005070301791033779,
      "loss": 0.2319,
      "num_input_tokens_seen": 29054400,
      "step": 137670
    },
    {
      "epoch": 15.145764576457646,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.005069222495223323,
      "loss": 0.2314,
      "num_input_tokens_seen": 29055456,
      "step": 137675
    },
    {
      "epoch": 15.146314631463147,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.005068143290940078,
      "loss": 0.2303,
      "num_input_tokens_seen": 29056512,
      "step": 137680
    },
    {
      "epoch": 15.146864686468646,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.005067064178193981,
      "loss": 0.2309,
      "num_input_tokens_seen": 29057568,
      "step": 137685
    },
    {
      "epoch": 15.147414741474147,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.005065985156994996,
      "loss": 0.2319,
      "num_input_tokens_seen": 29058592,
      "step": 137690
    },
    {
      "epoch": 15.147964796479648,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.005064906227353054,
      "loss": 0.2309,
      "num_input_tokens_seen": 29059648,
      "step": 137695
    },
    {
      "epoch": 15.148514851485148,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0050638273892781105,
      "loss": 0.2319,
      "num_input_tokens_seen": 29060736,
      "step": 137700
    },
    {
      "epoch": 15.149064906490649,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.005062748642780101,
      "loss": 0.2303,
      "num_input_tokens_seen": 29061792,
      "step": 137705
    },
    {
      "epoch": 15.14961496149615,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.005061669987868964,
      "loss": 0.2293,
      "num_input_tokens_seen": 29062848,
      "step": 137710
    },
    {
      "epoch": 15.150165016501651,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.0050605914245546455,
      "loss": 0.2319,
      "num_input_tokens_seen": 29063872,
      "step": 137715
    },
    {
      "epoch": 15.15071507150715,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.005059512952847087,
      "loss": 0.2314,
      "num_input_tokens_seen": 29064960,
      "step": 137720
    },
    {
      "epoch": 15.151265126512651,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.005058434572756233,
      "loss": 0.2298,
      "num_input_tokens_seen": 29065984,
      "step": 137725
    },
    {
      "epoch": 15.151815181518153,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.005057356284292017,
      "loss": 0.2319,
      "num_input_tokens_seen": 29067072,
      "step": 137730
    },
    {
      "epoch": 15.152365236523652,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.005056278087464372,
      "loss": 0.2309,
      "num_input_tokens_seen": 29068032,
      "step": 137735
    },
    {
      "epoch": 15.152915291529153,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.005055199982283246,
      "loss": 0.2319,
      "num_input_tokens_seen": 29069152,
      "step": 137740
    },
    {
      "epoch": 15.153465346534654,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0050541219687585594,
      "loss": 0.2309,
      "num_input_tokens_seen": 29070272,
      "step": 137745
    },
    {
      "epoch": 15.154015401540153,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.005053044046900267,
      "loss": 0.2314,
      "num_input_tokens_seen": 29071328,
      "step": 137750
    },
    {
      "epoch": 15.154565456545654,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005051966216718296,
      "loss": 0.233,
      "num_input_tokens_seen": 29072352,
      "step": 137755
    },
    {
      "epoch": 15.155115511551156,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.005050888478222572,
      "loss": 0.2325,
      "num_input_tokens_seen": 29073440,
      "step": 137760
    },
    {
      "epoch": 15.155665566556655,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0050498108314230425,
      "loss": 0.2309,
      "num_input_tokens_seen": 29074528,
      "step": 137765
    },
    {
      "epoch": 15.156215621562156,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.005048733276329625,
      "loss": 0.2324,
      "num_input_tokens_seen": 29075616,
      "step": 137770
    },
    {
      "epoch": 15.156765676567657,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0050476558129522574,
      "loss": 0.2324,
      "num_input_tokens_seen": 29076640,
      "step": 137775
    },
    {
      "epoch": 15.157315731573158,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.005046578441300876,
      "loss": 0.2314,
      "num_input_tokens_seen": 29077664,
      "step": 137780
    },
    {
      "epoch": 15.157865786578657,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0050455011613854015,
      "loss": 0.2319,
      "num_input_tokens_seen": 29078720,
      "step": 137785
    },
    {
      "epoch": 15.158415841584159,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.00504442397321577,
      "loss": 0.2319,
      "num_input_tokens_seen": 29079776,
      "step": 137790
    },
    {
      "epoch": 15.15896589658966,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.005043346876801901,
      "loss": 0.2314,
      "num_input_tokens_seen": 29080832,
      "step": 137795
    },
    {
      "epoch": 15.159515951595159,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0050422698721537314,
      "loss": 0.2314,
      "num_input_tokens_seen": 29081952,
      "step": 137800
    },
    {
      "epoch": 15.16006600660066,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.005041192959281178,
      "loss": 0.2324,
      "num_input_tokens_seen": 29083040,
      "step": 137805
    },
    {
      "epoch": 15.160616061606161,
      "grad_norm": 0.005859375,
      "learning_rate": 0.005040116138194168,
      "loss": 0.2293,
      "num_input_tokens_seen": 29084064,
      "step": 137810
    },
    {
      "epoch": 15.16116611661166,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.005039039408902635,
      "loss": 0.2314,
      "num_input_tokens_seen": 29085088,
      "step": 137815
    },
    {
      "epoch": 15.161716171617162,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.005037962771416491,
      "loss": 0.2314,
      "num_input_tokens_seen": 29086112,
      "step": 137820
    },
    {
      "epoch": 15.162266226622663,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.005036886225745669,
      "loss": 0.2319,
      "num_input_tokens_seen": 29087168,
      "step": 137825
    },
    {
      "epoch": 15.162816281628162,
      "grad_norm": 0.005859375,
      "learning_rate": 0.005035809771900079,
      "loss": 0.2303,
      "num_input_tokens_seen": 29088192,
      "step": 137830
    },
    {
      "epoch": 15.163366336633663,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.005034733409889651,
      "loss": 0.2309,
      "num_input_tokens_seen": 29089184,
      "step": 137835
    },
    {
      "epoch": 15.163916391639164,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.005033657139724308,
      "loss": 0.2314,
      "num_input_tokens_seen": 29090176,
      "step": 137840
    },
    {
      "epoch": 15.164466446644665,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.005032580961413965,
      "loss": 0.2309,
      "num_input_tokens_seen": 29091200,
      "step": 137845
    },
    {
      "epoch": 15.165016501650165,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.005031504874968534,
      "loss": 0.2303,
      "num_input_tokens_seen": 29092224,
      "step": 137850
    },
    {
      "epoch": 15.165566556655666,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.005030428880397946,
      "loss": 0.2309,
      "num_input_tokens_seen": 29093216,
      "step": 137855
    },
    {
      "epoch": 15.166116611661167,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.005029352977712105,
      "loss": 0.2319,
      "num_input_tokens_seen": 29094272,
      "step": 137860
    },
    {
      "epoch": 15.166666666666666,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.005028277166920933,
      "loss": 0.2319,
      "num_input_tokens_seen": 29095392,
      "step": 137865
    },
    {
      "epoch": 15.167216721672167,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.005027201448034351,
      "loss": 0.2324,
      "num_input_tokens_seen": 29096416,
      "step": 137870
    },
    {
      "epoch": 15.167766776677668,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.0050261258210622635,
      "loss": 0.2298,
      "num_input_tokens_seen": 29097504,
      "step": 137875
    },
    {
      "epoch": 15.168316831683168,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.005025050286014594,
      "loss": 0.2314,
      "num_input_tokens_seen": 29098496,
      "step": 137880
    },
    {
      "epoch": 15.168866886688669,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.0050239748429012415,
      "loss": 0.2319,
      "num_input_tokens_seen": 29099520,
      "step": 137885
    },
    {
      "epoch": 15.16941694169417,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.005022899491732128,
      "loss": 0.2298,
      "num_input_tokens_seen": 29100608,
      "step": 137890
    },
    {
      "epoch": 15.16996699669967,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.005021824232517168,
      "loss": 0.2324,
      "num_input_tokens_seen": 29101632,
      "step": 137895
    },
    {
      "epoch": 15.17051705170517,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.005020749065266259,
      "loss": 0.2324,
      "num_input_tokens_seen": 29102656,
      "step": 137900
    },
    {
      "epoch": 15.171067106710671,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0050196739899893246,
      "loss": 0.2283,
      "num_input_tokens_seen": 29103648,
      "step": 137905
    },
    {
      "epoch": 15.171617161716172,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.005018599006696259,
      "loss": 0.2314,
      "num_input_tokens_seen": 29104640,
      "step": 137910
    },
    {
      "epoch": 15.172167216721672,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.005017524115396984,
      "loss": 0.2308,
      "num_input_tokens_seen": 29105696,
      "step": 137915
    },
    {
      "epoch": 15.172717271727173,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0050164493161013925,
      "loss": 0.2314,
      "num_input_tokens_seen": 29106784,
      "step": 137920
    },
    {
      "epoch": 15.173267326732674,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.005015374608819398,
      "loss": 0.2303,
      "num_input_tokens_seen": 29107808,
      "step": 137925
    },
    {
      "epoch": 15.173817381738173,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.005014299993560908,
      "loss": 0.2303,
      "num_input_tokens_seen": 29108800,
      "step": 137930
    },
    {
      "epoch": 15.174367436743674,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.005013225470335821,
      "loss": 0.2319,
      "num_input_tokens_seen": 29109792,
      "step": 137935
    },
    {
      "epoch": 15.174917491749175,
      "grad_norm": 0.0008392333984375,
      "learning_rate": 0.005012151039154047,
      "loss": 0.2319,
      "num_input_tokens_seen": 29110784,
      "step": 137940
    },
    {
      "epoch": 15.175467546754675,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0050110767000254765,
      "loss": 0.2278,
      "num_input_tokens_seen": 29111872,
      "step": 137945
    },
    {
      "epoch": 15.176017601760176,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.005010002452960021,
      "loss": 0.2325,
      "num_input_tokens_seen": 29112928,
      "step": 137950
    },
    {
      "epoch": 15.176567656765677,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.005008928297967585,
      "loss": 0.2324,
      "num_input_tokens_seen": 29114016,
      "step": 137955
    },
    {
      "epoch": 15.177117711771178,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.005007854235058056,
      "loss": 0.2319,
      "num_input_tokens_seen": 29115040,
      "step": 137960
    },
    {
      "epoch": 15.177667766776677,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0050067802642413456,
      "loss": 0.2314,
      "num_input_tokens_seen": 29116128,
      "step": 137965
    },
    {
      "epoch": 15.178217821782178,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0050057063855273465,
      "loss": 0.2303,
      "num_input_tokens_seen": 29117216,
      "step": 137970
    },
    {
      "epoch": 15.17876787678768,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.005004632598925951,
      "loss": 0.2319,
      "num_input_tokens_seen": 29118240,
      "step": 137975
    },
    {
      "epoch": 15.179317931793179,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00500355890444706,
      "loss": 0.2298,
      "num_input_tokens_seen": 29119264,
      "step": 137980
    },
    {
      "epoch": 15.17986798679868,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.005002485302100571,
      "loss": 0.2303,
      "num_input_tokens_seen": 29120256,
      "step": 137985
    },
    {
      "epoch": 15.180418041804181,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0050014117918963825,
      "loss": 0.2309,
      "num_input_tokens_seen": 29121280,
      "step": 137990
    },
    {
      "epoch": 15.18096809680968,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.0050003383738443825,
      "loss": 0.2314,
      "num_input_tokens_seen": 29122336,
      "step": 137995
    },
    {
      "epoch": 15.181518151815181,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.004999265047954462,
      "loss": 0.2314,
      "num_input_tokens_seen": 29123360,
      "step": 138000
    },
    {
      "epoch": 15.182068206820682,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004998191814236515,
      "loss": 0.2319,
      "num_input_tokens_seen": 29124480,
      "step": 138005
    },
    {
      "epoch": 15.182618261826182,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004997118672700435,
      "loss": 0.2324,
      "num_input_tokens_seen": 29125472,
      "step": 138010
    },
    {
      "epoch": 15.183168316831683,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.004996045623356117,
      "loss": 0.2314,
      "num_input_tokens_seen": 29126560,
      "step": 138015
    },
    {
      "epoch": 15.183718371837184,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.004994972666213448,
      "loss": 0.2298,
      "num_input_tokens_seen": 29127584,
      "step": 138020
    },
    {
      "epoch": 15.184268426842685,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004993899801282308,
      "loss": 0.2314,
      "num_input_tokens_seen": 29128672,
      "step": 138025
    },
    {
      "epoch": 15.184818481848184,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.004992827028572598,
      "loss": 0.2329,
      "num_input_tokens_seen": 29129728,
      "step": 138030
    },
    {
      "epoch": 15.185368536853685,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.004991754348094196,
      "loss": 0.234,
      "num_input_tokens_seen": 29130880,
      "step": 138035
    },
    {
      "epoch": 15.185918591859187,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00499068175985699,
      "loss": 0.2309,
      "num_input_tokens_seen": 29131936,
      "step": 138040
    },
    {
      "epoch": 15.186468646864686,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004989609263870873,
      "loss": 0.2293,
      "num_input_tokens_seen": 29132992,
      "step": 138045
    },
    {
      "epoch": 15.187018701870187,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.004988536860145718,
      "loss": 0.2298,
      "num_input_tokens_seen": 29134112,
      "step": 138050
    },
    {
      "epoch": 15.187568756875688,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.00498746454869142,
      "loss": 0.2319,
      "num_input_tokens_seen": 29135168,
      "step": 138055
    },
    {
      "epoch": 15.188118811881187,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0049863923295178515,
      "loss": 0.2314,
      "num_input_tokens_seen": 29136192,
      "step": 138060
    },
    {
      "epoch": 15.188668866886688,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.004985320202634901,
      "loss": 0.2319,
      "num_input_tokens_seen": 29137312,
      "step": 138065
    },
    {
      "epoch": 15.18921892189219,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.004984248168052453,
      "loss": 0.2303,
      "num_input_tokens_seen": 29138368,
      "step": 138070
    },
    {
      "epoch": 15.189768976897689,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0049831762257803785,
      "loss": 0.2314,
      "num_input_tokens_seen": 29139456,
      "step": 138075
    },
    {
      "epoch": 15.19031903190319,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004982104375828567,
      "loss": 0.2303,
      "num_input_tokens_seen": 29140480,
      "step": 138080
    },
    {
      "epoch": 15.190869086908691,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.004981032618206893,
      "loss": 0.2314,
      "num_input_tokens_seen": 29141472,
      "step": 138085
    },
    {
      "epoch": 15.191419141914192,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.0049799609529252295,
      "loss": 0.2324,
      "num_input_tokens_seen": 29142560,
      "step": 138090
    },
    {
      "epoch": 15.191969196919691,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.004978889379993456,
      "loss": 0.2314,
      "num_input_tokens_seen": 29143584,
      "step": 138095
    },
    {
      "epoch": 15.192519251925193,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.004977817899421451,
      "loss": 0.2314,
      "num_input_tokens_seen": 29144640,
      "step": 138100
    },
    {
      "epoch": 15.193069306930694,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0049767465112190955,
      "loss": 0.2309,
      "num_input_tokens_seen": 29145728,
      "step": 138105
    },
    {
      "epoch": 15.193619361936193,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004975675215396257,
      "loss": 0.2335,
      "num_input_tokens_seen": 29146688,
      "step": 138110
    },
    {
      "epoch": 15.194169416941694,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004974604011962805,
      "loss": 0.2324,
      "num_input_tokens_seen": 29147680,
      "step": 138115
    },
    {
      "epoch": 15.194719471947195,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.004973532900928618,
      "loss": 0.2298,
      "num_input_tokens_seen": 29148736,
      "step": 138120
    },
    {
      "epoch": 15.195269526952695,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.004972461882303566,
      "loss": 0.2309,
      "num_input_tokens_seen": 29149824,
      "step": 138125
    },
    {
      "epoch": 15.195819581958196,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004971390956097526,
      "loss": 0.2298,
      "num_input_tokens_seen": 29150880,
      "step": 138130
    },
    {
      "epoch": 15.196369636963697,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.004970320122320364,
      "loss": 0.2304,
      "num_input_tokens_seen": 29151968,
      "step": 138135
    },
    {
      "epoch": 15.196919691969198,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.004969249380981943,
      "loss": 0.2309,
      "num_input_tokens_seen": 29153024,
      "step": 138140
    },
    {
      "epoch": 15.197469746974697,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004968178732092142,
      "loss": 0.2314,
      "num_input_tokens_seen": 29154144,
      "step": 138145
    },
    {
      "epoch": 15.198019801980198,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.00496710817566082,
      "loss": 0.2298,
      "num_input_tokens_seen": 29155168,
      "step": 138150
    },
    {
      "epoch": 15.1985698569857,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004966037711697847,
      "loss": 0.2319,
      "num_input_tokens_seen": 29156224,
      "step": 138155
    },
    {
      "epoch": 15.199119911991199,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.004964967340213096,
      "loss": 0.2314,
      "num_input_tokens_seen": 29157184,
      "step": 138160
    },
    {
      "epoch": 15.1996699669967,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.004963897061216419,
      "loss": 0.2314,
      "num_input_tokens_seen": 29158208,
      "step": 138165
    },
    {
      "epoch": 15.2002200220022,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004962826874717692,
      "loss": 0.2313,
      "num_input_tokens_seen": 29159232,
      "step": 138170
    },
    {
      "epoch": 15.2007700770077,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00496175678072677,
      "loss": 0.2309,
      "num_input_tokens_seen": 29160224,
      "step": 138175
    },
    {
      "epoch": 15.201320132013201,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.004960686779253518,
      "loss": 0.2314,
      "num_input_tokens_seen": 29161280,
      "step": 138180
    },
    {
      "epoch": 15.201870187018702,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.004959616870307803,
      "loss": 0.2304,
      "num_input_tokens_seen": 29162336,
      "step": 138185
    },
    {
      "epoch": 15.202420242024202,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.004958547053899477,
      "loss": 0.2319,
      "num_input_tokens_seen": 29163424,
      "step": 138190
    },
    {
      "epoch": 15.202970297029703,
      "grad_norm": 0.005859375,
      "learning_rate": 0.004957477330038411,
      "loss": 0.2314,
      "num_input_tokens_seen": 29164512,
      "step": 138195
    },
    {
      "epoch": 15.203520352035204,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.004956407698734449,
      "loss": 0.2314,
      "num_input_tokens_seen": 29165536,
      "step": 138200
    },
    {
      "epoch": 15.204070407040705,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.004955338159997466,
      "loss": 0.2308,
      "num_input_tokens_seen": 29166592,
      "step": 138205
    },
    {
      "epoch": 15.204620462046204,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004954268713837306,
      "loss": 0.2324,
      "num_input_tokens_seen": 29167616,
      "step": 138210
    },
    {
      "epoch": 15.205170517051705,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.004953199360263829,
      "loss": 0.2308,
      "num_input_tokens_seen": 29168672,
      "step": 138215
    },
    {
      "epoch": 15.205720572057206,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004952130099286899,
      "loss": 0.2314,
      "num_input_tokens_seen": 29169728,
      "step": 138220
    },
    {
      "epoch": 15.206270627062706,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.004951060930916363,
      "loss": 0.2329,
      "num_input_tokens_seen": 29170784,
      "step": 138225
    },
    {
      "epoch": 15.206820682068207,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.004949991855162073,
      "loss": 0.2309,
      "num_input_tokens_seen": 29171904,
      "step": 138230
    },
    {
      "epoch": 15.207370737073708,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.004948922872033886,
      "loss": 0.2324,
      "num_input_tokens_seen": 29172896,
      "step": 138235
    },
    {
      "epoch": 15.207920792079207,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.004947853981541653,
      "loss": 0.2314,
      "num_input_tokens_seen": 29173984,
      "step": 138240
    },
    {
      "epoch": 15.208470847084708,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.0049467851836952305,
      "loss": 0.2319,
      "num_input_tokens_seen": 29175008,
      "step": 138245
    },
    {
      "epoch": 15.20902090209021,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.004945716478504467,
      "loss": 0.2298,
      "num_input_tokens_seen": 29176064,
      "step": 138250
    },
    {
      "epoch": 15.209570957095709,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004944647865979204,
      "loss": 0.2314,
      "num_input_tokens_seen": 29177120,
      "step": 138255
    },
    {
      "epoch": 15.21012101210121,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.004943579346129301,
      "loss": 0.2314,
      "num_input_tokens_seen": 29178208,
      "step": 138260
    },
    {
      "epoch": 15.210671067106711,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.004942510918964598,
      "loss": 0.2293,
      "num_input_tokens_seen": 29179264,
      "step": 138265
    },
    {
      "epoch": 15.211221122112212,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0049414425844949445,
      "loss": 0.2335,
      "num_input_tokens_seen": 29180288,
      "step": 138270
    },
    {
      "epoch": 15.211771177117711,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004940374342730194,
      "loss": 0.2335,
      "num_input_tokens_seen": 29181344,
      "step": 138275
    },
    {
      "epoch": 15.212321232123212,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004939306193680181,
      "loss": 0.234,
      "num_input_tokens_seen": 29182432,
      "step": 138280
    },
    {
      "epoch": 15.212871287128714,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00493823813735476,
      "loss": 0.2324,
      "num_input_tokens_seen": 29183488,
      "step": 138285
    },
    {
      "epoch": 15.213421342134213,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.004937170173763764,
      "loss": 0.2303,
      "num_input_tokens_seen": 29184512,
      "step": 138290
    },
    {
      "epoch": 15.213971397139714,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.004936102302917043,
      "loss": 0.2335,
      "num_input_tokens_seen": 29185600,
      "step": 138295
    },
    {
      "epoch": 15.214521452145215,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.004935034524824444,
      "loss": 0.2314,
      "num_input_tokens_seen": 29186656,
      "step": 138300
    },
    {
      "epoch": 15.215071507150714,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0049339668394957955,
      "loss": 0.2303,
      "num_input_tokens_seen": 29187712,
      "step": 138305
    },
    {
      "epoch": 15.215621562156215,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0049328992469409505,
      "loss": 0.2309,
      "num_input_tokens_seen": 29188704,
      "step": 138310
    },
    {
      "epoch": 15.216171617161717,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004931831747169736,
      "loss": 0.2308,
      "num_input_tokens_seen": 29189760,
      "step": 138315
    },
    {
      "epoch": 15.216721672167218,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.004930764340192004,
      "loss": 0.2293,
      "num_input_tokens_seen": 29190784,
      "step": 138320
    },
    {
      "epoch": 15.217271727172717,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.004929697026017579,
      "loss": 0.2345,
      "num_input_tokens_seen": 29191840,
      "step": 138325
    },
    {
      "epoch": 15.217821782178218,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.0049286298046563055,
      "loss": 0.2324,
      "num_input_tokens_seen": 29192928,
      "step": 138330
    },
    {
      "epoch": 15.218371837183719,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004927562676118023,
      "loss": 0.2309,
      "num_input_tokens_seen": 29194016,
      "step": 138335
    },
    {
      "epoch": 15.218921892189218,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.004926495640412556,
      "loss": 0.2308,
      "num_input_tokens_seen": 29195040,
      "step": 138340
    },
    {
      "epoch": 15.21947194719472,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.004925428697549751,
      "loss": 0.2308,
      "num_input_tokens_seen": 29196064,
      "step": 138345
    },
    {
      "epoch": 15.22002200220022,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.004924361847539429,
      "loss": 0.2303,
      "num_input_tokens_seen": 29197056,
      "step": 138350
    },
    {
      "epoch": 15.22057205720572,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.004923295090391431,
      "loss": 0.2298,
      "num_input_tokens_seen": 29198112,
      "step": 138355
    },
    {
      "epoch": 15.221122112211221,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0049222284261155914,
      "loss": 0.2329,
      "num_input_tokens_seen": 29199104,
      "step": 138360
    },
    {
      "epoch": 15.221672167216722,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004921161854721731,
      "loss": 0.2304,
      "num_input_tokens_seen": 29200192,
      "step": 138365
    },
    {
      "epoch": 15.222222222222221,
      "grad_norm": 0.005859375,
      "learning_rate": 0.004920095376219691,
      "loss": 0.2319,
      "num_input_tokens_seen": 29201184,
      "step": 138370
    },
    {
      "epoch": 15.222772277227723,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.004919028990619295,
      "loss": 0.2319,
      "num_input_tokens_seen": 29202240,
      "step": 138375
    },
    {
      "epoch": 15.223322332233224,
      "grad_norm": 0.005859375,
      "learning_rate": 0.004917962697930366,
      "loss": 0.2298,
      "num_input_tokens_seen": 29203328,
      "step": 138380
    },
    {
      "epoch": 15.223872387238725,
      "grad_norm": 0.00078582763671875,
      "learning_rate": 0.004916896498162737,
      "loss": 0.2303,
      "num_input_tokens_seen": 29204352,
      "step": 138385
    },
    {
      "epoch": 15.224422442244224,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.004915830391326236,
      "loss": 0.2314,
      "num_input_tokens_seen": 29205344,
      "step": 138390
    },
    {
      "epoch": 15.224972497249725,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.004914764377430691,
      "loss": 0.2308,
      "num_input_tokens_seen": 29206400,
      "step": 138395
    },
    {
      "epoch": 15.225522552255226,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.004913698456485923,
      "loss": 0.2309,
      "num_input_tokens_seen": 29207424,
      "step": 138400
    },
    {
      "epoch": 15.226072607260726,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0049126326285017525,
      "loss": 0.2319,
      "num_input_tokens_seen": 29208512,
      "step": 138405
    },
    {
      "epoch": 15.226622662266227,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0049115668934880076,
      "loss": 0.2329,
      "num_input_tokens_seen": 29209504,
      "step": 138410
    },
    {
      "epoch": 15.227172717271728,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.004910501251454514,
      "loss": 0.2303,
      "num_input_tokens_seen": 29210560,
      "step": 138415
    },
    {
      "epoch": 15.227722772277227,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.004909435702411083,
      "loss": 0.2308,
      "num_input_tokens_seen": 29211616,
      "step": 138420
    },
    {
      "epoch": 15.228272827282728,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004908370246367547,
      "loss": 0.2309,
      "num_input_tokens_seen": 29212672,
      "step": 138425
    },
    {
      "epoch": 15.22882288228823,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.004907304883333713,
      "loss": 0.2314,
      "num_input_tokens_seen": 29213760,
      "step": 138430
    },
    {
      "epoch": 15.229372937293729,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.004906239613319415,
      "loss": 0.2298,
      "num_input_tokens_seen": 29214848,
      "step": 138435
    },
    {
      "epoch": 15.22992299229923,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.004905174436334456,
      "loss": 0.2319,
      "num_input_tokens_seen": 29215840,
      "step": 138440
    },
    {
      "epoch": 15.23047304730473,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.004904109352388662,
      "loss": 0.2314,
      "num_input_tokens_seen": 29216928,
      "step": 138445
    },
    {
      "epoch": 15.231023102310232,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.004903044361491851,
      "loss": 0.2298,
      "num_input_tokens_seen": 29217984,
      "step": 138450
    },
    {
      "epoch": 15.231573157315731,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0049019794636538305,
      "loss": 0.2303,
      "num_input_tokens_seen": 29219072,
      "step": 138455
    },
    {
      "epoch": 15.232123212321232,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0049009146588844255,
      "loss": 0.2324,
      "num_input_tokens_seen": 29220128,
      "step": 138460
    },
    {
      "epoch": 15.232673267326733,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.004899849947193443,
      "loss": 0.2288,
      "num_input_tokens_seen": 29221248,
      "step": 138465
    },
    {
      "epoch": 15.233223322332233,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004898785328590688,
      "loss": 0.2324,
      "num_input_tokens_seen": 29222304,
      "step": 138470
    },
    {
      "epoch": 15.233773377337734,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.0048977208030859904,
      "loss": 0.2278,
      "num_input_tokens_seen": 29223360,
      "step": 138475
    },
    {
      "epoch": 15.234323432343235,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004896656370689149,
      "loss": 0.2314,
      "num_input_tokens_seen": 29224416,
      "step": 138480
    },
    {
      "epoch": 15.234873487348734,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.004895592031409981,
      "loss": 0.2309,
      "num_input_tokens_seen": 29225504,
      "step": 138485
    },
    {
      "epoch": 15.235423542354235,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.004894527785258294,
      "loss": 0.2308,
      "num_input_tokens_seen": 29226656,
      "step": 138490
    },
    {
      "epoch": 15.235973597359736,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.004893463632243891,
      "loss": 0.2293,
      "num_input_tokens_seen": 29227776,
      "step": 138495
    },
    {
      "epoch": 15.236523652365236,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004892399572376583,
      "loss": 0.2298,
      "num_input_tokens_seen": 29228832,
      "step": 138500
    },
    {
      "epoch": 15.237073707370737,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.004891335605666178,
      "loss": 0.2314,
      "num_input_tokens_seen": 29229920,
      "step": 138505
    },
    {
      "epoch": 15.237623762376238,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0048902717321224875,
      "loss": 0.2309,
      "num_input_tokens_seen": 29230976,
      "step": 138510
    },
    {
      "epoch": 15.238173817381739,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004889207951755311,
      "loss": 0.2324,
      "num_input_tokens_seen": 29231968,
      "step": 138515
    },
    {
      "epoch": 15.238723872387238,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.004888144264574449,
      "loss": 0.2324,
      "num_input_tokens_seen": 29232960,
      "step": 138520
    },
    {
      "epoch": 15.23927392739274,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.004887080670589714,
      "loss": 0.2309,
      "num_input_tokens_seen": 29233984,
      "step": 138525
    },
    {
      "epoch": 15.23982398239824,
      "grad_norm": 0.000659942626953125,
      "learning_rate": 0.004886017169810894,
      "loss": 0.2309,
      "num_input_tokens_seen": 29235040,
      "step": 138530
    },
    {
      "epoch": 15.24037403740374,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.004884953762247809,
      "loss": 0.2303,
      "num_input_tokens_seen": 29236096,
      "step": 138535
    },
    {
      "epoch": 15.24092409240924,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.004883890447910253,
      "loss": 0.2303,
      "num_input_tokens_seen": 29237152,
      "step": 138540
    },
    {
      "epoch": 15.241474147414742,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004882827226808019,
      "loss": 0.2324,
      "num_input_tokens_seen": 29238208,
      "step": 138545
    },
    {
      "epoch": 15.242024202420241,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.004881764098950917,
      "loss": 0.2303,
      "num_input_tokens_seen": 29239264,
      "step": 138550
    },
    {
      "epoch": 15.242574257425742,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.004880701064348734,
      "loss": 0.2303,
      "num_input_tokens_seen": 29240320,
      "step": 138555
    },
    {
      "epoch": 15.243124312431243,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004879638123011275,
      "loss": 0.2308,
      "num_input_tokens_seen": 29241408,
      "step": 138560
    },
    {
      "epoch": 15.243674367436745,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00487857527494834,
      "loss": 0.2324,
      "num_input_tokens_seen": 29242432,
      "step": 138565
    },
    {
      "epoch": 15.244224422442244,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004877512520169715,
      "loss": 0.2298,
      "num_input_tokens_seen": 29243552,
      "step": 138570
    },
    {
      "epoch": 15.244774477447745,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004876449858685203,
      "loss": 0.2324,
      "num_input_tokens_seen": 29244608,
      "step": 138575
    },
    {
      "epoch": 15.245324532453246,
      "grad_norm": 0.002716064453125,
      "learning_rate": 0.004875387290504593,
      "loss": 0.2314,
      "num_input_tokens_seen": 29245728,
      "step": 138580
    },
    {
      "epoch": 15.245874587458745,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.004874324815637683,
      "loss": 0.2298,
      "num_input_tokens_seen": 29246688,
      "step": 138585
    },
    {
      "epoch": 15.246424642464246,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0048732624340942585,
      "loss": 0.2309,
      "num_input_tokens_seen": 29247712,
      "step": 138590
    },
    {
      "epoch": 15.246974697469748,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.004872200145884114,
      "loss": 0.2298,
      "num_input_tokens_seen": 29248800,
      "step": 138595
    },
    {
      "epoch": 15.247524752475247,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.004871137951017046,
      "loss": 0.2303,
      "num_input_tokens_seen": 29249856,
      "step": 138600
    },
    {
      "epoch": 15.248074807480748,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.00487007584950284,
      "loss": 0.2303,
      "num_input_tokens_seen": 29250880,
      "step": 138605
    },
    {
      "epoch": 15.248624862486249,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.004869013841351278,
      "loss": 0.2335,
      "num_input_tokens_seen": 29251936,
      "step": 138610
    },
    {
      "epoch": 15.249174917491748,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.004867951926572155,
      "loss": 0.2303,
      "num_input_tokens_seen": 29252960,
      "step": 138615
    },
    {
      "epoch": 15.24972497249725,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.004866890105175256,
      "loss": 0.2303,
      "num_input_tokens_seen": 29254048,
      "step": 138620
    },
    {
      "epoch": 15.25027502750275,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004865828377170376,
      "loss": 0.2324,
      "num_input_tokens_seen": 29255072,
      "step": 138625
    },
    {
      "epoch": 15.250825082508252,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.004864766742567291,
      "loss": 0.2293,
      "num_input_tokens_seen": 29256128,
      "step": 138630
    },
    {
      "epoch": 15.251375137513751,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.004863705201375782,
      "loss": 0.2309,
      "num_input_tokens_seen": 29257152,
      "step": 138635
    },
    {
      "epoch": 15.251925192519252,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004862643753605645,
      "loss": 0.2319,
      "num_input_tokens_seen": 29258208,
      "step": 138640
    },
    {
      "epoch": 15.252475247524753,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.004861582399266651,
      "loss": 0.2314,
      "num_input_tokens_seen": 29259264,
      "step": 138645
    },
    {
      "epoch": 15.253025302530252,
      "grad_norm": 0.000766754150390625,
      "learning_rate": 0.004860521138368586,
      "loss": 0.2308,
      "num_input_tokens_seen": 29260320,
      "step": 138650
    },
    {
      "epoch": 15.253575357535754,
      "grad_norm": 0.000644683837890625,
      "learning_rate": 0.0048594599709212375,
      "loss": 0.2319,
      "num_input_tokens_seen": 29261440,
      "step": 138655
    },
    {
      "epoch": 15.254125412541255,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0048583988969343765,
      "loss": 0.2319,
      "num_input_tokens_seen": 29262560,
      "step": 138660
    },
    {
      "epoch": 15.254675467546754,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00485733791641779,
      "loss": 0.2303,
      "num_input_tokens_seen": 29263552,
      "step": 138665
    },
    {
      "epoch": 15.255225522552255,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.004856277029381249,
      "loss": 0.2303,
      "num_input_tokens_seen": 29264640,
      "step": 138670
    },
    {
      "epoch": 15.255775577557756,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004855216235834536,
      "loss": 0.2293,
      "num_input_tokens_seen": 29265728,
      "step": 138675
    },
    {
      "epoch": 15.256325632563255,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004854155535787431,
      "loss": 0.2314,
      "num_input_tokens_seen": 29266720,
      "step": 138680
    },
    {
      "epoch": 15.256875687568757,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004853094929249701,
      "loss": 0.2308,
      "num_input_tokens_seen": 29267776,
      "step": 138685
    },
    {
      "epoch": 15.257425742574258,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004852034416231132,
      "loss": 0.2319,
      "num_input_tokens_seen": 29268832,
      "step": 138690
    },
    {
      "epoch": 15.257975797579759,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.004850973996741485,
      "loss": 0.2303,
      "num_input_tokens_seen": 29269952,
      "step": 138695
    },
    {
      "epoch": 15.258525852585258,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.004849913670790547,
      "loss": 0.2319,
      "num_input_tokens_seen": 29271008,
      "step": 138700
    },
    {
      "epoch": 15.25907590759076,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00484885343838808,
      "loss": 0.2329,
      "num_input_tokens_seen": 29272032,
      "step": 138705
    },
    {
      "epoch": 15.25962596259626,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004847793299543859,
      "loss": 0.2309,
      "num_input_tokens_seen": 29273120,
      "step": 138710
    },
    {
      "epoch": 15.26017601760176,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.004846733254267661,
      "loss": 0.2314,
      "num_input_tokens_seen": 29274144,
      "step": 138715
    },
    {
      "epoch": 15.26072607260726,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.004845673302569246,
      "loss": 0.2314,
      "num_input_tokens_seen": 29275232,
      "step": 138720
    },
    {
      "epoch": 15.261276127612762,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004844613444458391,
      "loss": 0.2314,
      "num_input_tokens_seen": 29276288,
      "step": 138725
    },
    {
      "epoch": 15.261826182618261,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004843553679944859,
      "loss": 0.2324,
      "num_input_tokens_seen": 29277344,
      "step": 138730
    },
    {
      "epoch": 15.262376237623762,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.004842494009038416,
      "loss": 0.2319,
      "num_input_tokens_seen": 29278400,
      "step": 138735
    },
    {
      "epoch": 15.262926292629263,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.0048414344317488376,
      "loss": 0.2335,
      "num_input_tokens_seen": 29279456,
      "step": 138740
    },
    {
      "epoch": 15.263476347634764,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0048403749480858795,
      "loss": 0.2293,
      "num_input_tokens_seen": 29280448,
      "step": 138745
    },
    {
      "epoch": 15.264026402640264,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004839315558059314,
      "loss": 0.2308,
      "num_input_tokens_seen": 29281440,
      "step": 138750
    },
    {
      "epoch": 15.264576457645765,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.004838256261678902,
      "loss": 0.2308,
      "num_input_tokens_seen": 29282432,
      "step": 138755
    },
    {
      "epoch": 15.265126512651266,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.004837197058954401,
      "loss": 0.2298,
      "num_input_tokens_seen": 29283488,
      "step": 138760
    },
    {
      "epoch": 15.265676567656765,
      "grad_norm": 0.005859375,
      "learning_rate": 0.004836137949895579,
      "loss": 0.2319,
      "num_input_tokens_seen": 29284544,
      "step": 138765
    },
    {
      "epoch": 15.266226622662266,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0048350789345121955,
      "loss": 0.2319,
      "num_input_tokens_seen": 29285632,
      "step": 138770
    },
    {
      "epoch": 15.266776677667767,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.004834020012814016,
      "loss": 0.2314,
      "num_input_tokens_seen": 29286688,
      "step": 138775
    },
    {
      "epoch": 15.267326732673267,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.004832961184810798,
      "loss": 0.2314,
      "num_input_tokens_seen": 29287776,
      "step": 138780
    },
    {
      "epoch": 15.267876787678768,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.004831902450512291,
      "loss": 0.2298,
      "num_input_tokens_seen": 29288864,
      "step": 138785
    },
    {
      "epoch": 15.268426842684269,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.004830843809928261,
      "loss": 0.2314,
      "num_input_tokens_seen": 29289984,
      "step": 138790
    },
    {
      "epoch": 15.268976897689768,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004829785263068469,
      "loss": 0.2314,
      "num_input_tokens_seen": 29290976,
      "step": 138795
    },
    {
      "epoch": 15.26952695269527,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.004828726809942662,
      "loss": 0.2334,
      "num_input_tokens_seen": 29292064,
      "step": 138800
    },
    {
      "epoch": 15.27007700770077,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004827668450560603,
      "loss": 0.2314,
      "num_input_tokens_seen": 29293184,
      "step": 138805
    },
    {
      "epoch": 15.270627062706271,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.004826610184932038,
      "loss": 0.2308,
      "num_input_tokens_seen": 29294208,
      "step": 138810
    },
    {
      "epoch": 15.27117711771177,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00482555201306673,
      "loss": 0.2303,
      "num_input_tokens_seen": 29295264,
      "step": 138815
    },
    {
      "epoch": 15.271727172717272,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0048244939349744225,
      "loss": 0.2319,
      "num_input_tokens_seen": 29296352,
      "step": 138820
    },
    {
      "epoch": 15.272277227722773,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.004823435950664872,
      "loss": 0.2319,
      "num_input_tokens_seen": 29297440,
      "step": 138825
    },
    {
      "epoch": 15.272827282728272,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004822378060147834,
      "loss": 0.2319,
      "num_input_tokens_seen": 29298528,
      "step": 138830
    },
    {
      "epoch": 15.273377337733773,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.004821320263433048,
      "loss": 0.2319,
      "num_input_tokens_seen": 29299616,
      "step": 138835
    },
    {
      "epoch": 15.273927392739274,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.004820262560530275,
      "loss": 0.2329,
      "num_input_tokens_seen": 29300704,
      "step": 138840
    },
    {
      "epoch": 15.274477447744774,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.004819204951449251,
      "loss": 0.2319,
      "num_input_tokens_seen": 29301792,
      "step": 138845
    },
    {
      "epoch": 15.275027502750275,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.004818147436199731,
      "loss": 0.2309,
      "num_input_tokens_seen": 29302784,
      "step": 138850
    },
    {
      "epoch": 15.275577557755776,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0048170900147914645,
      "loss": 0.2298,
      "num_input_tokens_seen": 29303904,
      "step": 138855
    },
    {
      "epoch": 15.276127612761275,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004816032687234188,
      "loss": 0.2314,
      "num_input_tokens_seen": 29304992,
      "step": 138860
    },
    {
      "epoch": 15.276677667766776,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004814975453537656,
      "loss": 0.2314,
      "num_input_tokens_seen": 29305984,
      "step": 138865
    },
    {
      "epoch": 15.277227722772277,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.004813918313711609,
      "loss": 0.2308,
      "num_input_tokens_seen": 29307072,
      "step": 138870
    },
    {
      "epoch": 15.277777777777779,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.004812861267765783,
      "loss": 0.2309,
      "num_input_tokens_seen": 29308032,
      "step": 138875
    },
    {
      "epoch": 15.278327832783278,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.004811804315709929,
      "loss": 0.2309,
      "num_input_tokens_seen": 29309056,
      "step": 138880
    },
    {
      "epoch": 15.278877887788779,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.0048107474575537846,
      "loss": 0.2319,
      "num_input_tokens_seen": 29310144,
      "step": 138885
    },
    {
      "epoch": 15.27942794279428,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004809690693307095,
      "loss": 0.2314,
      "num_input_tokens_seen": 29311200,
      "step": 138890
    },
    {
      "epoch": 15.27997799779978,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.004808634022979597,
      "loss": 0.235,
      "num_input_tokens_seen": 29312288,
      "step": 138895
    },
    {
      "epoch": 15.28052805280528,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004807577446581025,
      "loss": 0.2314,
      "num_input_tokens_seen": 29313312,
      "step": 138900
    },
    {
      "epoch": 15.281078107810782,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.0048065209641211205,
      "loss": 0.2319,
      "num_input_tokens_seen": 29314336,
      "step": 138905
    },
    {
      "epoch": 15.281628162816281,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004805464575609622,
      "loss": 0.2313,
      "num_input_tokens_seen": 29315392,
      "step": 138910
    },
    {
      "epoch": 15.282178217821782,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.004804408281056269,
      "loss": 0.2319,
      "num_input_tokens_seen": 29316448,
      "step": 138915
    },
    {
      "epoch": 15.282728272827283,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.004803352080470792,
      "loss": 0.2314,
      "num_input_tokens_seen": 29317472,
      "step": 138920
    },
    {
      "epoch": 15.283278327832782,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004802295973862921,
      "loss": 0.2308,
      "num_input_tokens_seen": 29318496,
      "step": 138925
    },
    {
      "epoch": 15.283828382838283,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0048012399612424005,
      "loss": 0.2324,
      "num_input_tokens_seen": 29319520,
      "step": 138930
    },
    {
      "epoch": 15.284378437843785,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.004800184042618951,
      "loss": 0.2314,
      "num_input_tokens_seen": 29320608,
      "step": 138935
    },
    {
      "epoch": 15.284928492849286,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.004799128218002311,
      "loss": 0.2319,
      "num_input_tokens_seen": 29321632,
      "step": 138940
    },
    {
      "epoch": 15.285478547854785,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004798072487402218,
      "loss": 0.2314,
      "num_input_tokens_seen": 29322656,
      "step": 138945
    },
    {
      "epoch": 15.286028602860286,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004797016850828389,
      "loss": 0.2324,
      "num_input_tokens_seen": 29323680,
      "step": 138950
    },
    {
      "epoch": 15.286578657865787,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.004795961308290566,
      "loss": 0.2345,
      "num_input_tokens_seen": 29324736,
      "step": 138955
    },
    {
      "epoch": 15.287128712871286,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.004794905859798465,
      "loss": 0.2314,
      "num_input_tokens_seen": 29325760,
      "step": 138960
    },
    {
      "epoch": 15.287678767876788,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0047938505053618215,
      "loss": 0.2309,
      "num_input_tokens_seen": 29326784,
      "step": 138965
    },
    {
      "epoch": 15.288228822882289,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.004792795244990364,
      "loss": 0.2319,
      "num_input_tokens_seen": 29327840,
      "step": 138970
    },
    {
      "epoch": 15.288778877887788,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.00479174007869381,
      "loss": 0.2308,
      "num_input_tokens_seen": 29328864,
      "step": 138975
    },
    {
      "epoch": 15.289328932893289,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.004790685006481894,
      "loss": 0.2314,
      "num_input_tokens_seen": 29329920,
      "step": 138980
    },
    {
      "epoch": 15.28987898789879,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004789630028364331,
      "loss": 0.2288,
      "num_input_tokens_seen": 29331008,
      "step": 138985
    },
    {
      "epoch": 15.290429042904291,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.004788575144350854,
      "loss": 0.2293,
      "num_input_tokens_seen": 29332128,
      "step": 138990
    },
    {
      "epoch": 15.29097909790979,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004787520354451175,
      "loss": 0.2303,
      "num_input_tokens_seen": 29333184,
      "step": 138995
    },
    {
      "epoch": 15.291529152915292,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00478646565867502,
      "loss": 0.2319,
      "num_input_tokens_seen": 29334208,
      "step": 139000
    },
    {
      "epoch": 15.292079207920793,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.004785411057032115,
      "loss": 0.2329,
      "num_input_tokens_seen": 29335200,
      "step": 139005
    },
    {
      "epoch": 15.292629262926292,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.004784356549532175,
      "loss": 0.2288,
      "num_input_tokens_seen": 29336256,
      "step": 139010
    },
    {
      "epoch": 15.293179317931793,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.004783302136184914,
      "loss": 0.2309,
      "num_input_tokens_seen": 29337280,
      "step": 139015
    },
    {
      "epoch": 15.293729372937294,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0047822478170000534,
      "loss": 0.2308,
      "num_input_tokens_seen": 29338272,
      "step": 139020
    },
    {
      "epoch": 15.294279427942794,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004781193591987314,
      "loss": 0.2314,
      "num_input_tokens_seen": 29339264,
      "step": 139025
    },
    {
      "epoch": 15.294829482948295,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.004780139461156413,
      "loss": 0.2314,
      "num_input_tokens_seen": 29340320,
      "step": 139030
    },
    {
      "epoch": 15.295379537953796,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004779085424517063,
      "loss": 0.2319,
      "num_input_tokens_seen": 29341376,
      "step": 139035
    },
    {
      "epoch": 15.295929592959295,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004778031482078973,
      "loss": 0.2314,
      "num_input_tokens_seen": 29342464,
      "step": 139040
    },
    {
      "epoch": 15.296479647964796,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004776977633851868,
      "loss": 0.233,
      "num_input_tokens_seen": 29343552,
      "step": 139045
    },
    {
      "epoch": 15.297029702970297,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.004775923879845449,
      "loss": 0.2324,
      "num_input_tokens_seen": 29344576,
      "step": 139050
    },
    {
      "epoch": 15.297579757975798,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.004774870220069433,
      "loss": 0.2319,
      "num_input_tokens_seen": 29345600,
      "step": 139055
    },
    {
      "epoch": 15.298129812981298,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.004773816654533538,
      "loss": 0.2308,
      "num_input_tokens_seen": 29346592,
      "step": 139060
    },
    {
      "epoch": 15.298679867986799,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0047727631832474625,
      "loss": 0.2319,
      "num_input_tokens_seen": 29347680,
      "step": 139065
    },
    {
      "epoch": 15.2992299229923,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.004771709806220926,
      "loss": 0.2319,
      "num_input_tokens_seen": 29348800,
      "step": 139070
    },
    {
      "epoch": 15.2997799779978,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.004770656523463626,
      "loss": 0.2314,
      "num_input_tokens_seen": 29349824,
      "step": 139075
    },
    {
      "epoch": 15.3003300330033,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.004769603334985278,
      "loss": 0.2309,
      "num_input_tokens_seen": 29350912,
      "step": 139080
    },
    {
      "epoch": 15.300880088008801,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.004768550240795592,
      "loss": 0.2293,
      "num_input_tokens_seen": 29351936,
      "step": 139085
    },
    {
      "epoch": 15.3014301430143,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.0047674972409042626,
      "loss": 0.2298,
      "num_input_tokens_seen": 29352992,
      "step": 139090
    },
    {
      "epoch": 15.301980198019802,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.004766444335321007,
      "loss": 0.2309,
      "num_input_tokens_seen": 29354016,
      "step": 139095
    },
    {
      "epoch": 15.302530253025303,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.004765391524055518,
      "loss": 0.2329,
      "num_input_tokens_seen": 29355072,
      "step": 139100
    },
    {
      "epoch": 15.303080308030804,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00476433880711751,
      "loss": 0.2324,
      "num_input_tokens_seen": 29356160,
      "step": 139105
    },
    {
      "epoch": 15.303630363036303,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.004763286184516673,
      "loss": 0.2325,
      "num_input_tokens_seen": 29357120,
      "step": 139110
    },
    {
      "epoch": 15.304180418041804,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004762233656262716,
      "loss": 0.2335,
      "num_input_tokens_seen": 29358144,
      "step": 139115
    },
    {
      "epoch": 15.304730473047305,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.004761181222365344,
      "loss": 0.2324,
      "num_input_tokens_seen": 29359296,
      "step": 139120
    },
    {
      "epoch": 15.305280528052805,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.004760128882834248,
      "loss": 0.2314,
      "num_input_tokens_seen": 29360352,
      "step": 139125
    },
    {
      "epoch": 15.305830583058306,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004759076637679134,
      "loss": 0.2324,
      "num_input_tokens_seen": 29361344,
      "step": 139130
    },
    {
      "epoch": 15.306380638063807,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.004758024486909692,
      "loss": 0.2304,
      "num_input_tokens_seen": 29362464,
      "step": 139135
    },
    {
      "epoch": 15.306930693069306,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.0047569724305356244,
      "loss": 0.2303,
      "num_input_tokens_seen": 29363520,
      "step": 139140
    },
    {
      "epoch": 15.307480748074807,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.00475592046856663,
      "loss": 0.2314,
      "num_input_tokens_seen": 29364544,
      "step": 139145
    },
    {
      "epoch": 15.308030803080309,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.004754868601012397,
      "loss": 0.2309,
      "num_input_tokens_seen": 29365664,
      "step": 139150
    },
    {
      "epoch": 15.308580858085808,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.004753816827882631,
      "loss": 0.2298,
      "num_input_tokens_seen": 29366752,
      "step": 139155
    },
    {
      "epoch": 15.309130913091309,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.004752765149187016,
      "loss": 0.2314,
      "num_input_tokens_seen": 29367776,
      "step": 139160
    },
    {
      "epoch": 15.30968096809681,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004751713564935242,
      "loss": 0.2314,
      "num_input_tokens_seen": 29368864,
      "step": 139165
    },
    {
      "epoch": 15.310231023102311,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004750662075137008,
      "loss": 0.2335,
      "num_input_tokens_seen": 29369920,
      "step": 139170
    },
    {
      "epoch": 15.31078107810781,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004749610679802003,
      "loss": 0.2314,
      "num_input_tokens_seen": 29370976,
      "step": 139175
    },
    {
      "epoch": 15.311331133113312,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.004748559378939923,
      "loss": 0.2314,
      "num_input_tokens_seen": 29372000,
      "step": 139180
    },
    {
      "epoch": 15.311881188118813,
      "grad_norm": 0.000640869140625,
      "learning_rate": 0.004747508172560453,
      "loss": 0.2309,
      "num_input_tokens_seen": 29373024,
      "step": 139185
    },
    {
      "epoch": 15.312431243124312,
      "grad_norm": 0.00051116943359375,
      "learning_rate": 0.004746457060673274,
      "loss": 0.2324,
      "num_input_tokens_seen": 29374016,
      "step": 139190
    },
    {
      "epoch": 15.312981298129813,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.004745406043288081,
      "loss": 0.2335,
      "num_input_tokens_seen": 29375040,
      "step": 139195
    },
    {
      "epoch": 15.313531353135314,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.004744355120414565,
      "loss": 0.2319,
      "num_input_tokens_seen": 29376096,
      "step": 139200
    },
    {
      "epoch": 15.314081408140813,
      "grad_norm": 0.000453948974609375,
      "learning_rate": 0.0047433042920624014,
      "loss": 0.2319,
      "num_input_tokens_seen": 29377152,
      "step": 139205
    },
    {
      "epoch": 15.314631463146315,
      "grad_norm": 0.0006103515625,
      "learning_rate": 0.004742253558241286,
      "loss": 0.2319,
      "num_input_tokens_seen": 29378112,
      "step": 139210
    },
    {
      "epoch": 15.315181518151816,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004741202918960892,
      "loss": 0.2319,
      "num_input_tokens_seen": 29379136,
      "step": 139215
    },
    {
      "epoch": 15.315731573157315,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.004740152374230912,
      "loss": 0.2324,
      "num_input_tokens_seen": 29380160,
      "step": 139220
    },
    {
      "epoch": 15.316281628162816,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00473910192406102,
      "loss": 0.2288,
      "num_input_tokens_seen": 29381216,
      "step": 139225
    },
    {
      "epoch": 15.316831683168317,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004738051568460904,
      "loss": 0.2314,
      "num_input_tokens_seen": 29382240,
      "step": 139230
    },
    {
      "epoch": 15.317381738173818,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.004737001307440245,
      "loss": 0.2293,
      "num_input_tokens_seen": 29383296,
      "step": 139235
    },
    {
      "epoch": 15.317931793179318,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.004735951141008716,
      "loss": 0.2303,
      "num_input_tokens_seen": 29384384,
      "step": 139240
    },
    {
      "epoch": 15.318481848184819,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004734901069176005,
      "loss": 0.2293,
      "num_input_tokens_seen": 29385472,
      "step": 139245
    },
    {
      "epoch": 15.31903190319032,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004733851091951786,
      "loss": 0.2329,
      "num_input_tokens_seen": 29386528,
      "step": 139250
    },
    {
      "epoch": 15.319581958195819,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004732801209345726,
      "loss": 0.2319,
      "num_input_tokens_seen": 29387584,
      "step": 139255
    },
    {
      "epoch": 15.32013201320132,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.004731751421367518,
      "loss": 0.2329,
      "num_input_tokens_seen": 29388640,
      "step": 139260
    },
    {
      "epoch": 15.320682068206821,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004730701728026827,
      "loss": 0.2324,
      "num_input_tokens_seen": 29389728,
      "step": 139265
    },
    {
      "epoch": 15.32123212321232,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004729652129333334,
      "loss": 0.2303,
      "num_input_tokens_seen": 29390848,
      "step": 139270
    },
    {
      "epoch": 15.321782178217822,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.0047286026252967095,
      "loss": 0.2308,
      "num_input_tokens_seen": 29391904,
      "step": 139275
    },
    {
      "epoch": 15.322332233223323,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.004727553215926623,
      "loss": 0.2303,
      "num_input_tokens_seen": 29392928,
      "step": 139280
    },
    {
      "epoch": 15.322882288228822,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0047265039012327455,
      "loss": 0.2314,
      "num_input_tokens_seen": 29393952,
      "step": 139285
    },
    {
      "epoch": 15.323432343234323,
      "grad_norm": 0.000522613525390625,
      "learning_rate": 0.004725454681224755,
      "loss": 0.2293,
      "num_input_tokens_seen": 29394944,
      "step": 139290
    },
    {
      "epoch": 15.323982398239824,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.004724405555912322,
      "loss": 0.2308,
      "num_input_tokens_seen": 29396000,
      "step": 139295
    },
    {
      "epoch": 15.324532453245325,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0047233565253051125,
      "loss": 0.2304,
      "num_input_tokens_seen": 29397024,
      "step": 139300
    },
    {
      "epoch": 15.325082508250825,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.004722307589412791,
      "loss": 0.2329,
      "num_input_tokens_seen": 29398080,
      "step": 139305
    },
    {
      "epoch": 15.325632563256326,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.004721258748245033,
      "loss": 0.2335,
      "num_input_tokens_seen": 29399168,
      "step": 139310
    },
    {
      "epoch": 15.326182618261827,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00472021000181149,
      "loss": 0.2319,
      "num_input_tokens_seen": 29400224,
      "step": 139315
    },
    {
      "epoch": 15.326732673267326,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.004719161350121849,
      "loss": 0.2309,
      "num_input_tokens_seen": 29401344,
      "step": 139320
    },
    {
      "epoch": 15.327282728272827,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0047181127931857645,
      "loss": 0.2303,
      "num_input_tokens_seen": 29402432,
      "step": 139325
    },
    {
      "epoch": 15.327832783278328,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004717064331012897,
      "loss": 0.2319,
      "num_input_tokens_seen": 29403520,
      "step": 139330
    },
    {
      "epoch": 15.328382838283828,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.0047160159636129165,
      "loss": 0.2335,
      "num_input_tokens_seen": 29404576,
      "step": 139335
    },
    {
      "epoch": 15.328932893289329,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.004714967690995478,
      "loss": 0.2293,
      "num_input_tokens_seen": 29405664,
      "step": 139340
    },
    {
      "epoch": 15.32948294829483,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.004713919513170247,
      "loss": 0.2309,
      "num_input_tokens_seen": 29406688,
      "step": 139345
    },
    {
      "epoch": 15.33003300330033,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.00471287143014689,
      "loss": 0.2314,
      "num_input_tokens_seen": 29407808,
      "step": 139350
    },
    {
      "epoch": 15.33058305830583,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.004711823441935054,
      "loss": 0.2329,
      "num_input_tokens_seen": 29408896,
      "step": 139355
    },
    {
      "epoch": 15.331133113311331,
      "grad_norm": 0.000644683837890625,
      "learning_rate": 0.004710775548544411,
      "loss": 0.2288,
      "num_input_tokens_seen": 29409952,
      "step": 139360
    },
    {
      "epoch": 15.331683168316832,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004709727749984607,
      "loss": 0.2319,
      "num_input_tokens_seen": 29411008,
      "step": 139365
    },
    {
      "epoch": 15.332233223322332,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004708680046265309,
      "loss": 0.2314,
      "num_input_tokens_seen": 29412064,
      "step": 139370
    },
    {
      "epoch": 15.332783278327833,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004707632437396166,
      "loss": 0.2324,
      "num_input_tokens_seen": 29413120,
      "step": 139375
    },
    {
      "epoch": 15.333333333333334,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.004706584923386834,
      "loss": 0.2319,
      "num_input_tokens_seen": 29414176,
      "step": 139380
    },
    {
      "epoch": 15.333883388338833,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004705537504246974,
      "loss": 0.2314,
      "num_input_tokens_seen": 29415232,
      "step": 139385
    },
    {
      "epoch": 15.334433443344334,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0047044901799862355,
      "loss": 0.2319,
      "num_input_tokens_seen": 29416256,
      "step": 139390
    },
    {
      "epoch": 15.334983498349835,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.004703442950614267,
      "loss": 0.2335,
      "num_input_tokens_seen": 29417312,
      "step": 139395
    },
    {
      "epoch": 15.335533553355335,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.004702395816140723,
      "loss": 0.2329,
      "num_input_tokens_seen": 29418400,
      "step": 139400
    },
    {
      "epoch": 15.336083608360836,
      "grad_norm": 0.000621795654296875,
      "learning_rate": 0.004701348776575256,
      "loss": 0.2319,
      "num_input_tokens_seen": 29419392,
      "step": 139405
    },
    {
      "epoch": 15.336633663366337,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00470030183192752,
      "loss": 0.2309,
      "num_input_tokens_seen": 29420384,
      "step": 139410
    },
    {
      "epoch": 15.337183718371838,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00469925498220716,
      "loss": 0.2329,
      "num_input_tokens_seen": 29421376,
      "step": 139415
    },
    {
      "epoch": 15.337733773377337,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004698208227423821,
      "loss": 0.2314,
      "num_input_tokens_seen": 29422432,
      "step": 139420
    },
    {
      "epoch": 15.338283828382838,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004697161567587154,
      "loss": 0.2309,
      "num_input_tokens_seen": 29423488,
      "step": 139425
    },
    {
      "epoch": 15.33883388338834,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.004696115002706805,
      "loss": 0.2303,
      "num_input_tokens_seen": 29424544,
      "step": 139430
    },
    {
      "epoch": 15.339383938393839,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004695068532792417,
      "loss": 0.2309,
      "num_input_tokens_seen": 29425600,
      "step": 139435
    },
    {
      "epoch": 15.33993399339934,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.0046940221578536424,
      "loss": 0.2319,
      "num_input_tokens_seen": 29426624,
      "step": 139440
    },
    {
      "epoch": 15.340484048404841,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004692975877900117,
      "loss": 0.2303,
      "num_input_tokens_seen": 29427648,
      "step": 139445
    },
    {
      "epoch": 15.34103410341034,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.004691929692941489,
      "loss": 0.2324,
      "num_input_tokens_seen": 29428736,
      "step": 139450
    },
    {
      "epoch": 15.341584158415841,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.004690883602987396,
      "loss": 0.2329,
      "num_input_tokens_seen": 29429824,
      "step": 139455
    },
    {
      "epoch": 15.342134213421343,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.004689837608047483,
      "loss": 0.2308,
      "num_input_tokens_seen": 29430848,
      "step": 139460
    },
    {
      "epoch": 15.342684268426842,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0046887917081313915,
      "loss": 0.2298,
      "num_input_tokens_seen": 29431936,
      "step": 139465
    },
    {
      "epoch": 15.343234323432343,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004687745903248756,
      "loss": 0.2308,
      "num_input_tokens_seen": 29433024,
      "step": 139470
    },
    {
      "epoch": 15.343784378437844,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.004686700193409223,
      "loss": 0.2329,
      "num_input_tokens_seen": 29434080,
      "step": 139475
    },
    {
      "epoch": 15.344334433443345,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004685654578622421,
      "loss": 0.2319,
      "num_input_tokens_seen": 29435136,
      "step": 139480
    },
    {
      "epoch": 15.344884488448844,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004684609058897996,
      "loss": 0.2308,
      "num_input_tokens_seen": 29436256,
      "step": 139485
    },
    {
      "epoch": 15.345434543454346,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.004683563634245574,
      "loss": 0.2303,
      "num_input_tokens_seen": 29437280,
      "step": 139490
    },
    {
      "epoch": 15.345984598459847,
      "grad_norm": 0.000667572021484375,
      "learning_rate": 0.004682518304674796,
      "loss": 0.2303,
      "num_input_tokens_seen": 29438336,
      "step": 139495
    },
    {
      "epoch": 15.346534653465346,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.004681473070195301,
      "loss": 0.2309,
      "num_input_tokens_seen": 29439424,
      "step": 139500
    },
    {
      "epoch": 15.347084708470847,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.004680427930816713,
      "loss": 0.2324,
      "num_input_tokens_seen": 29440512,
      "step": 139505
    },
    {
      "epoch": 15.347634763476348,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.004679382886548674,
      "loss": 0.2298,
      "num_input_tokens_seen": 29441632,
      "step": 139510
    },
    {
      "epoch": 15.348184818481847,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.004678337937400804,
      "loss": 0.2314,
      "num_input_tokens_seen": 29442688,
      "step": 139515
    },
    {
      "epoch": 15.348734873487349,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0046772930833827425,
      "loss": 0.2319,
      "num_input_tokens_seen": 29443744,
      "step": 139520
    },
    {
      "epoch": 15.34928492849285,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.004676248324504121,
      "loss": 0.2303,
      "num_input_tokens_seen": 29444736,
      "step": 139525
    },
    {
      "epoch": 15.34983498349835,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004675203660774559,
      "loss": 0.2293,
      "num_input_tokens_seen": 29445760,
      "step": 139530
    },
    {
      "epoch": 15.35038503850385,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.004674159092203698,
      "loss": 0.2319,
      "num_input_tokens_seen": 29446848,
      "step": 139535
    },
    {
      "epoch": 15.350935093509351,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.004673114618801157,
      "loss": 0.2314,
      "num_input_tokens_seen": 29447872,
      "step": 139540
    },
    {
      "epoch": 15.351485148514852,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0046720702405765565,
      "loss": 0.2314,
      "num_input_tokens_seen": 29448960,
      "step": 139545
    },
    {
      "epoch": 15.352035203520352,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.004671025957539529,
      "loss": 0.2313,
      "num_input_tokens_seen": 29450048,
      "step": 139550
    },
    {
      "epoch": 15.352585258525853,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.004669981769699701,
      "loss": 0.2309,
      "num_input_tokens_seen": 29451104,
      "step": 139555
    },
    {
      "epoch": 15.353135313531354,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004668937677066698,
      "loss": 0.2314,
      "num_input_tokens_seen": 29452128,
      "step": 139560
    },
    {
      "epoch": 15.353685368536853,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.004667893679650139,
      "loss": 0.2329,
      "num_input_tokens_seen": 29453216,
      "step": 139565
    },
    {
      "epoch": 15.354235423542354,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.004666849777459642,
      "loss": 0.2303,
      "num_input_tokens_seen": 29454272,
      "step": 139570
    },
    {
      "epoch": 15.354785478547855,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.004665805970504831,
      "loss": 0.2303,
      "num_input_tokens_seen": 29455296,
      "step": 139575
    },
    {
      "epoch": 15.355335533553355,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.004664762258795334,
      "loss": 0.2319,
      "num_input_tokens_seen": 29456352,
      "step": 139580
    },
    {
      "epoch": 15.355885588558856,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.0046637186423407586,
      "loss": 0.2314,
      "num_input_tokens_seen": 29457376,
      "step": 139585
    },
    {
      "epoch": 15.356435643564357,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.004662675121150733,
      "loss": 0.2303,
      "num_input_tokens_seen": 29458368,
      "step": 139590
    },
    {
      "epoch": 15.356985698569858,
      "grad_norm": 0.000797271728515625,
      "learning_rate": 0.004661631695234865,
      "loss": 0.2298,
      "num_input_tokens_seen": 29459424,
      "step": 139595
    },
    {
      "epoch": 15.357535753575357,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.004660588364602783,
      "loss": 0.2293,
      "num_input_tokens_seen": 29460512,
      "step": 139600
    },
    {
      "epoch": 15.358085808580858,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.00465954512926409,
      "loss": 0.2319,
      "num_input_tokens_seen": 29461632,
      "step": 139605
    },
    {
      "epoch": 15.35863586358636,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.00465850198922841,
      "loss": 0.2319,
      "num_input_tokens_seen": 29462720,
      "step": 139610
    },
    {
      "epoch": 15.359185918591859,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004657458944505359,
      "loss": 0.2314,
      "num_input_tokens_seen": 29463808,
      "step": 139615
    },
    {
      "epoch": 15.35973597359736,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00465641599510454,
      "loss": 0.2308,
      "num_input_tokens_seen": 29464864,
      "step": 139620
    },
    {
      "epoch": 15.36028602860286,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.004655373141035576,
      "loss": 0.2309,
      "num_input_tokens_seen": 29465920,
      "step": 139625
    },
    {
      "epoch": 15.36083608360836,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004654330382308071,
      "loss": 0.234,
      "num_input_tokens_seen": 29467008,
      "step": 139630
    },
    {
      "epoch": 15.361386138613861,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.004653287718931638,
      "loss": 0.2329,
      "num_input_tokens_seen": 29468032,
      "step": 139635
    },
    {
      "epoch": 15.361936193619362,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.004652245150915893,
      "loss": 0.2324,
      "num_input_tokens_seen": 29469120,
      "step": 139640
    },
    {
      "epoch": 15.362486248624862,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004651202678270432,
      "loss": 0.2303,
      "num_input_tokens_seen": 29470176,
      "step": 139645
    },
    {
      "epoch": 15.363036303630363,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004650160301004878,
      "loss": 0.2303,
      "num_input_tokens_seen": 29471232,
      "step": 139650
    },
    {
      "epoch": 15.363586358635864,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.004649118019128827,
      "loss": 0.2329,
      "num_input_tokens_seen": 29472320,
      "step": 139655
    },
    {
      "epoch": 15.364136413641365,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.004648075832651886,
      "loss": 0.2314,
      "num_input_tokens_seen": 29473472,
      "step": 139660
    },
    {
      "epoch": 15.364686468646864,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00464703374158366,
      "loss": 0.2324,
      "num_input_tokens_seen": 29474560,
      "step": 139665
    },
    {
      "epoch": 15.365236523652365,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.004645991745933757,
      "loss": 0.2314,
      "num_input_tokens_seen": 29475584,
      "step": 139670
    },
    {
      "epoch": 15.365786578657866,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.004644949845711785,
      "loss": 0.2304,
      "num_input_tokens_seen": 29476640,
      "step": 139675
    },
    {
      "epoch": 15.366336633663366,
      "grad_norm": 0.001953125,
      "learning_rate": 0.004643908040927343,
      "loss": 0.2314,
      "num_input_tokens_seen": 29477696,
      "step": 139680
    },
    {
      "epoch": 15.366886688668867,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.004642866331590024,
      "loss": 0.2298,
      "num_input_tokens_seen": 29478720,
      "step": 139685
    },
    {
      "epoch": 15.367436743674368,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.004641824717709438,
      "loss": 0.2324,
      "num_input_tokens_seen": 29479776,
      "step": 139690
    },
    {
      "epoch": 15.367986798679867,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.004640783199295183,
      "loss": 0.2303,
      "num_input_tokens_seen": 29480832,
      "step": 139695
    },
    {
      "epoch": 15.368536853685368,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.004639741776356862,
      "loss": 0.2324,
      "num_input_tokens_seen": 29481856,
      "step": 139700
    },
    {
      "epoch": 15.36908690869087,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00463870044890407,
      "loss": 0.2293,
      "num_input_tokens_seen": 29482880,
      "step": 139705
    },
    {
      "epoch": 15.369636963696369,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.004637659216946401,
      "loss": 0.2303,
      "num_input_tokens_seen": 29483872,
      "step": 139710
    },
    {
      "epoch": 15.37018701870187,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.004636618080493459,
      "loss": 0.2309,
      "num_input_tokens_seen": 29484960,
      "step": 139715
    },
    {
      "epoch": 15.370737073707371,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004635577039554829,
      "loss": 0.2314,
      "num_input_tokens_seen": 29485952,
      "step": 139720
    },
    {
      "epoch": 15.371287128712872,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.004634536094140112,
      "loss": 0.2298,
      "num_input_tokens_seen": 29486944,
      "step": 139725
    },
    {
      "epoch": 15.371837183718371,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004633495244258908,
      "loss": 0.2324,
      "num_input_tokens_seen": 29487968,
      "step": 139730
    },
    {
      "epoch": 15.372387238723872,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004632454489920799,
      "loss": 0.2314,
      "num_input_tokens_seen": 29488960,
      "step": 139735
    },
    {
      "epoch": 15.372937293729374,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.004631413831135387,
      "loss": 0.2324,
      "num_input_tokens_seen": 29489984,
      "step": 139740
    },
    {
      "epoch": 15.373487348734873,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.004630373267912254,
      "loss": 0.2314,
      "num_input_tokens_seen": 29491008,
      "step": 139745
    },
    {
      "epoch": 15.374037403740374,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.004629332800260994,
      "loss": 0.2308,
      "num_input_tokens_seen": 29492032,
      "step": 139750
    },
    {
      "epoch": 15.374587458745875,
      "grad_norm": 0.005859375,
      "learning_rate": 0.004628292428191202,
      "loss": 0.2309,
      "num_input_tokens_seen": 29492992,
      "step": 139755
    },
    {
      "epoch": 15.375137513751374,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.004627252151712456,
      "loss": 0.2314,
      "num_input_tokens_seen": 29494048,
      "step": 139760
    },
    {
      "epoch": 15.375687568756875,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.004626211970834356,
      "loss": 0.2308,
      "num_input_tokens_seen": 29495104,
      "step": 139765
    },
    {
      "epoch": 15.376237623762377,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.004625171885566482,
      "loss": 0.2298,
      "num_input_tokens_seen": 29496128,
      "step": 139770
    },
    {
      "epoch": 15.376787678767876,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.004624131895918416,
      "loss": 0.2314,
      "num_input_tokens_seen": 29497152,
      "step": 139775
    },
    {
      "epoch": 15.377337733773377,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004623092001899747,
      "loss": 0.2293,
      "num_input_tokens_seen": 29498208,
      "step": 139780
    },
    {
      "epoch": 15.377887788778878,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004622052203520061,
      "loss": 0.2319,
      "num_input_tokens_seen": 29499296,
      "step": 139785
    },
    {
      "epoch": 15.37843784378438,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.004621012500788943,
      "loss": 0.2329,
      "num_input_tokens_seen": 29500320,
      "step": 139790
    },
    {
      "epoch": 15.378987898789878,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.004619972893715974,
      "loss": 0.2308,
      "num_input_tokens_seen": 29501408,
      "step": 139795
    },
    {
      "epoch": 15.37953795379538,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.004618933382310728,
      "loss": 0.2324,
      "num_input_tokens_seen": 29502464,
      "step": 139800
    },
    {
      "epoch": 15.38008800880088,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004617893966582791,
      "loss": 0.2309,
      "num_input_tokens_seen": 29503616,
      "step": 139805
    },
    {
      "epoch": 15.38063806380638,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.004616854646541745,
      "loss": 0.2303,
      "num_input_tokens_seen": 29504640,
      "step": 139810
    },
    {
      "epoch": 15.381188118811881,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.004615815422197171,
      "loss": 0.2288,
      "num_input_tokens_seen": 29505696,
      "step": 139815
    },
    {
      "epoch": 15.381738173817382,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.0046147762935586445,
      "loss": 0.2298,
      "num_input_tokens_seen": 29506752,
      "step": 139820
    },
    {
      "epoch": 15.382288228822881,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004613737260635736,
      "loss": 0.2309,
      "num_input_tokens_seen": 29507872,
      "step": 139825
    },
    {
      "epoch": 15.382838283828383,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.0046126983234380334,
      "loss": 0.2314,
      "num_input_tokens_seen": 29508896,
      "step": 139830
    },
    {
      "epoch": 15.383388338833884,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.004611659481975101,
      "loss": 0.2309,
      "num_input_tokens_seen": 29509920,
      "step": 139835
    },
    {
      "epoch": 15.383938393839385,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.004610620736256518,
      "loss": 0.2324,
      "num_input_tokens_seen": 29511008,
      "step": 139840
    },
    {
      "epoch": 15.384488448844884,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004609582086291863,
      "loss": 0.2324,
      "num_input_tokens_seen": 29512000,
      "step": 139845
    },
    {
      "epoch": 15.385038503850385,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.004608543532090699,
      "loss": 0.233,
      "num_input_tokens_seen": 29513024,
      "step": 139850
    },
    {
      "epoch": 15.385588558855886,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004607505073662607,
      "loss": 0.2319,
      "num_input_tokens_seen": 29514112,
      "step": 139855
    },
    {
      "epoch": 15.386138613861386,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.004606466711017151,
      "loss": 0.2319,
      "num_input_tokens_seen": 29515200,
      "step": 139860
    },
    {
      "epoch": 15.386688668866887,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.004605428444163903,
      "loss": 0.2335,
      "num_input_tokens_seen": 29516192,
      "step": 139865
    },
    {
      "epoch": 15.387238723872388,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.004604390273112438,
      "loss": 0.2293,
      "num_input_tokens_seen": 29517344,
      "step": 139870
    },
    {
      "epoch": 15.387788778877887,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0046033521978723155,
      "loss": 0.2319,
      "num_input_tokens_seen": 29518368,
      "step": 139875
    },
    {
      "epoch": 15.388338833883388,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.004602314218453111,
      "loss": 0.2308,
      "num_input_tokens_seen": 29519456,
      "step": 139880
    },
    {
      "epoch": 15.38888888888889,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0046012763348643815,
      "loss": 0.2304,
      "num_input_tokens_seen": 29520576,
      "step": 139885
    },
    {
      "epoch": 15.389438943894389,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004600238547115706,
      "loss": 0.2319,
      "num_input_tokens_seen": 29521632,
      "step": 139890
    },
    {
      "epoch": 15.38998899889989,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.004599200855216633,
      "loss": 0.2308,
      "num_input_tokens_seen": 29522784,
      "step": 139895
    },
    {
      "epoch": 15.39053905390539,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.004598163259176737,
      "loss": 0.2304,
      "num_input_tokens_seen": 29523808,
      "step": 139900
    },
    {
      "epoch": 15.391089108910892,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.004597125759005584,
      "loss": 0.2298,
      "num_input_tokens_seen": 29524864,
      "step": 139905
    },
    {
      "epoch": 15.391639163916391,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.004596088354712726,
      "loss": 0.2314,
      "num_input_tokens_seen": 29525952,
      "step": 139910
    },
    {
      "epoch": 15.392189218921892,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.004595051046307732,
      "loss": 0.2314,
      "num_input_tokens_seen": 29527072,
      "step": 139915
    },
    {
      "epoch": 15.392739273927393,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.004594013833800162,
      "loss": 0.2298,
      "num_input_tokens_seen": 29528064,
      "step": 139920
    },
    {
      "epoch": 15.393289328932893,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.004592976717199564,
      "loss": 0.2329,
      "num_input_tokens_seen": 29529088,
      "step": 139925
    },
    {
      "epoch": 15.393839383938394,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004591939696515514,
      "loss": 0.2329,
      "num_input_tokens_seen": 29530176,
      "step": 139930
    },
    {
      "epoch": 15.394389438943895,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004590902771757558,
      "loss": 0.2303,
      "num_input_tokens_seen": 29531232,
      "step": 139935
    },
    {
      "epoch": 15.394939493949394,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.00458986594293526,
      "loss": 0.2293,
      "num_input_tokens_seen": 29532320,
      "step": 139940
    },
    {
      "epoch": 15.395489548954895,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004588829210058173,
      "loss": 0.2309,
      "num_input_tokens_seen": 29533344,
      "step": 139945
    },
    {
      "epoch": 15.396039603960396,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004587792573135847,
      "loss": 0.2308,
      "num_input_tokens_seen": 29534336,
      "step": 139950
    },
    {
      "epoch": 15.396589658965897,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.00458675603217784,
      "loss": 0.2298,
      "num_input_tokens_seen": 29535360,
      "step": 139955
    },
    {
      "epoch": 15.397139713971397,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0045857195871937114,
      "loss": 0.2335,
      "num_input_tokens_seen": 29536480,
      "step": 139960
    },
    {
      "epoch": 15.397689768976898,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.004584683238193002,
      "loss": 0.2309,
      "num_input_tokens_seen": 29537504,
      "step": 139965
    },
    {
      "epoch": 15.398239823982399,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0045836469851852765,
      "loss": 0.2356,
      "num_input_tokens_seen": 29538560,
      "step": 139970
    },
    {
      "epoch": 15.398789878987898,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.004582610828180071,
      "loss": 0.2319,
      "num_input_tokens_seen": 29539616,
      "step": 139975
    },
    {
      "epoch": 15.3993399339934,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004581574767186943,
      "loss": 0.2335,
      "num_input_tokens_seen": 29540640,
      "step": 139980
    },
    {
      "epoch": 15.3998899889989,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.004580538802215447,
      "loss": 0.2335,
      "num_input_tokens_seen": 29541664,
      "step": 139985
    },
    {
      "epoch": 15.4004400440044,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004579502933275121,
      "loss": 0.2319,
      "num_input_tokens_seen": 29542752,
      "step": 139990
    },
    {
      "epoch": 15.400990099009901,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00457846716037552,
      "loss": 0.2309,
      "num_input_tokens_seen": 29543808,
      "step": 139995
    },
    {
      "epoch": 15.401540154015402,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0045774314835261825,
      "loss": 0.2308,
      "num_input_tokens_seen": 29544896,
      "step": 140000
    },
    {
      "epoch": 15.402090209020901,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004576395902736661,
      "loss": 0.2319,
      "num_input_tokens_seen": 29545984,
      "step": 140005
    },
    {
      "epoch": 15.402640264026402,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.004575360418016493,
      "loss": 0.2314,
      "num_input_tokens_seen": 29547072,
      "step": 140010
    },
    {
      "epoch": 15.403190319031903,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.004574325029375225,
      "loss": 0.2314,
      "num_input_tokens_seen": 29548192,
      "step": 140015
    },
    {
      "epoch": 15.403740374037405,
      "grad_norm": 0.00054931640625,
      "learning_rate": 0.004573289736822406,
      "loss": 0.2308,
      "num_input_tokens_seen": 29549216,
      "step": 140020
    },
    {
      "epoch": 15.404290429042904,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004572254540367568,
      "loss": 0.2293,
      "num_input_tokens_seen": 29550208,
      "step": 140025
    },
    {
      "epoch": 15.404840484048405,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0045712194400202594,
      "loss": 0.2309,
      "num_input_tokens_seen": 29551264,
      "step": 140030
    },
    {
      "epoch": 15.405390539053906,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004570184435790018,
      "loss": 0.2309,
      "num_input_tokens_seen": 29552352,
      "step": 140035
    },
    {
      "epoch": 15.405940594059405,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004569149527686373,
      "loss": 0.2309,
      "num_input_tokens_seen": 29553408,
      "step": 140040
    },
    {
      "epoch": 15.406490649064907,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.004568114715718881,
      "loss": 0.2324,
      "num_input_tokens_seen": 29554432,
      "step": 140045
    },
    {
      "epoch": 15.407040704070408,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.0045670799998970635,
      "loss": 0.2308,
      "num_input_tokens_seen": 29555456,
      "step": 140050
    },
    {
      "epoch": 15.407590759075907,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.00456604538023047,
      "loss": 0.2329,
      "num_input_tokens_seen": 29556512,
      "step": 140055
    },
    {
      "epoch": 15.408140814081408,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004565010856728629,
      "loss": 0.2314,
      "num_input_tokens_seen": 29557472,
      "step": 140060
    },
    {
      "epoch": 15.408690869086909,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.004563976429401071,
      "loss": 0.2324,
      "num_input_tokens_seen": 29558496,
      "step": 140065
    },
    {
      "epoch": 15.409240924092408,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004562942098257335,
      "loss": 0.2314,
      "num_input_tokens_seen": 29559552,
      "step": 140070
    },
    {
      "epoch": 15.40979097909791,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.0045619078633069516,
      "loss": 0.2329,
      "num_input_tokens_seen": 29560544,
      "step": 140075
    },
    {
      "epoch": 15.41034103410341,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.004560873724559461,
      "loss": 0.2314,
      "num_input_tokens_seen": 29561568,
      "step": 140080
    },
    {
      "epoch": 15.410891089108912,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.004559839682024387,
      "loss": 0.2303,
      "num_input_tokens_seen": 29562688,
      "step": 140085
    },
    {
      "epoch": 15.411441144114411,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.004558805735711258,
      "loss": 0.2304,
      "num_input_tokens_seen": 29563712,
      "step": 140090
    },
    {
      "epoch": 15.411991199119912,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.004557771885629611,
      "loss": 0.2319,
      "num_input_tokens_seen": 29564736,
      "step": 140095
    },
    {
      "epoch": 15.412541254125413,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0045567381317889585,
      "loss": 0.2324,
      "num_input_tokens_seen": 29565824,
      "step": 140100
    },
    {
      "epoch": 15.413091309130913,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.00455570447419885,
      "loss": 0.2319,
      "num_input_tokens_seen": 29566880,
      "step": 140105
    },
    {
      "epoch": 15.413641364136414,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0045546709128688025,
      "loss": 0.2288,
      "num_input_tokens_seen": 29567872,
      "step": 140110
    },
    {
      "epoch": 15.414191419141915,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0045536374478083366,
      "loss": 0.2308,
      "num_input_tokens_seen": 29568928,
      "step": 140115
    },
    {
      "epoch": 15.414741474147414,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004552604079026985,
      "loss": 0.2329,
      "num_input_tokens_seen": 29570016,
      "step": 140120
    },
    {
      "epoch": 15.415291529152915,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.004551570806534264,
      "loss": 0.2288,
      "num_input_tokens_seen": 29571072,
      "step": 140125
    },
    {
      "epoch": 15.415841584158416,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.004550537630339701,
      "loss": 0.2308,
      "num_input_tokens_seen": 29572224,
      "step": 140130
    },
    {
      "epoch": 15.416391639163916,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004549504550452823,
      "loss": 0.2303,
      "num_input_tokens_seen": 29573312,
      "step": 140135
    },
    {
      "epoch": 15.416941694169417,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.004548471566883142,
      "loss": 0.2303,
      "num_input_tokens_seen": 29574368,
      "step": 140140
    },
    {
      "epoch": 15.417491749174918,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0045474386796401905,
      "loss": 0.2304,
      "num_input_tokens_seen": 29575392,
      "step": 140145
    },
    {
      "epoch": 15.418041804180419,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004546405888733479,
      "loss": 0.2314,
      "num_input_tokens_seen": 29576384,
      "step": 140150
    },
    {
      "epoch": 15.418591859185918,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004545373194172524,
      "loss": 0.2319,
      "num_input_tokens_seen": 29577568,
      "step": 140155
    },
    {
      "epoch": 15.41914191419142,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.004544340595966847,
      "loss": 0.2324,
      "num_input_tokens_seen": 29578560,
      "step": 140160
    },
    {
      "epoch": 15.41969196919692,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.004543308094125967,
      "loss": 0.2309,
      "num_input_tokens_seen": 29579648,
      "step": 140165
    },
    {
      "epoch": 15.42024202420242,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.004542275688659404,
      "loss": 0.2329,
      "num_input_tokens_seen": 29580768,
      "step": 140170
    },
    {
      "epoch": 15.42079207920792,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0045412433795766665,
      "loss": 0.2314,
      "num_input_tokens_seen": 29581888,
      "step": 140175
    },
    {
      "epoch": 15.421342134213422,
      "grad_norm": 0.000705718994140625,
      "learning_rate": 0.004540211166887266,
      "loss": 0.2309,
      "num_input_tokens_seen": 29582912,
      "step": 140180
    },
    {
      "epoch": 15.421892189218921,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004539179050600719,
      "loss": 0.2299,
      "num_input_tokens_seen": 29584000,
      "step": 140185
    },
    {
      "epoch": 15.422442244224422,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.00453814703072654,
      "loss": 0.2309,
      "num_input_tokens_seen": 29585024,
      "step": 140190
    },
    {
      "epoch": 15.422992299229923,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004537115107274244,
      "loss": 0.2303,
      "num_input_tokens_seen": 29586112,
      "step": 140195
    },
    {
      "epoch": 15.423542354235423,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0045360832802533385,
      "loss": 0.2319,
      "num_input_tokens_seen": 29587296,
      "step": 140200
    },
    {
      "epoch": 15.424092409240924,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004535051549673327,
      "loss": 0.2319,
      "num_input_tokens_seen": 29588384,
      "step": 140205
    },
    {
      "epoch": 15.424642464246425,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.004534019915543728,
      "loss": 0.2324,
      "num_input_tokens_seen": 29589408,
      "step": 140210
    },
    {
      "epoch": 15.425192519251926,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00453298837787404,
      "loss": 0.2314,
      "num_input_tokens_seen": 29590432,
      "step": 140215
    },
    {
      "epoch": 15.425742574257425,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004531956936673777,
      "loss": 0.2324,
      "num_input_tokens_seen": 29591488,
      "step": 140220
    },
    {
      "epoch": 15.426292629262926,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004530925591952449,
      "loss": 0.2324,
      "num_input_tokens_seen": 29592544,
      "step": 140225
    },
    {
      "epoch": 15.426842684268427,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.004529894343719549,
      "loss": 0.2304,
      "num_input_tokens_seen": 29593600,
      "step": 140230
    },
    {
      "epoch": 15.427392739273927,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.004528863191984596,
      "loss": 0.2319,
      "num_input_tokens_seen": 29594624,
      "step": 140235
    },
    {
      "epoch": 15.427942794279428,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.004527832136757081,
      "loss": 0.2319,
      "num_input_tokens_seen": 29595744,
      "step": 140240
    },
    {
      "epoch": 15.428492849284929,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.004526801178046511,
      "loss": 0.2314,
      "num_input_tokens_seen": 29596800,
      "step": 140245
    },
    {
      "epoch": 15.429042904290428,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.004525770315862395,
      "loss": 0.2319,
      "num_input_tokens_seen": 29597888,
      "step": 140250
    },
    {
      "epoch": 15.42959295929593,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.004524739550214225,
      "loss": 0.2309,
      "num_input_tokens_seen": 29598944,
      "step": 140255
    },
    {
      "epoch": 15.43014301430143,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.004523708881111505,
      "loss": 0.2313,
      "num_input_tokens_seen": 29599936,
      "step": 140260
    },
    {
      "epoch": 15.430693069306932,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.004522678308563731,
      "loss": 0.2309,
      "num_input_tokens_seen": 29601056,
      "step": 140265
    },
    {
      "epoch": 15.43124312431243,
      "grad_norm": 0.000514984130859375,
      "learning_rate": 0.004521647832580407,
      "loss": 0.2324,
      "num_input_tokens_seen": 29602112,
      "step": 140270
    },
    {
      "epoch": 15.431793179317932,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004520617453171024,
      "loss": 0.2324,
      "num_input_tokens_seen": 29603232,
      "step": 140275
    },
    {
      "epoch": 15.432343234323433,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.004519587170345079,
      "loss": 0.2309,
      "num_input_tokens_seen": 29604288,
      "step": 140280
    },
    {
      "epoch": 15.432893289328932,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.004518556984112076,
      "loss": 0.2319,
      "num_input_tokens_seen": 29605344,
      "step": 140285
    },
    {
      "epoch": 15.433443344334433,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004517526894481498,
      "loss": 0.2309,
      "num_input_tokens_seen": 29606432,
      "step": 140290
    },
    {
      "epoch": 15.433993399339935,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.0045164969014628515,
      "loss": 0.2319,
      "num_input_tokens_seen": 29607488,
      "step": 140295
    },
    {
      "epoch": 15.434543454345434,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.004515467005065618,
      "loss": 0.2303,
      "num_input_tokens_seen": 29608480,
      "step": 140300
    },
    {
      "epoch": 15.435093509350935,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004514437205299291,
      "loss": 0.2319,
      "num_input_tokens_seen": 29609504,
      "step": 140305
    },
    {
      "epoch": 15.435643564356436,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004513407502173373,
      "loss": 0.2314,
      "num_input_tokens_seen": 29610624,
      "step": 140310
    },
    {
      "epoch": 15.436193619361935,
      "grad_norm": 0.000606536865234375,
      "learning_rate": 0.00451237789569734,
      "loss": 0.2304,
      "num_input_tokens_seen": 29611616,
      "step": 140315
    },
    {
      "epoch": 15.436743674367436,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.004511348385880692,
      "loss": 0.2309,
      "num_input_tokens_seen": 29612640,
      "step": 140320
    },
    {
      "epoch": 15.437293729372938,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004510318972732913,
      "loss": 0.233,
      "num_input_tokens_seen": 29613728,
      "step": 140325
    },
    {
      "epoch": 15.437843784378439,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.004509289656263486,
      "loss": 0.2314,
      "num_input_tokens_seen": 29614784,
      "step": 140330
    },
    {
      "epoch": 15.438393839383938,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.004508260436481902,
      "loss": 0.2288,
      "num_input_tokens_seen": 29615808,
      "step": 140335
    },
    {
      "epoch": 15.438943894389439,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.004507231313397651,
      "loss": 0.2314,
      "num_input_tokens_seen": 29616928,
      "step": 140340
    },
    {
      "epoch": 15.43949394939494,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004506202287020211,
      "loss": 0.2314,
      "num_input_tokens_seen": 29618080,
      "step": 140345
    },
    {
      "epoch": 15.44004400440044,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.004505173357359073,
      "loss": 0.2303,
      "num_input_tokens_seen": 29619168,
      "step": 140350
    },
    {
      "epoch": 15.44059405940594,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004504144524423712,
      "loss": 0.2324,
      "num_input_tokens_seen": 29620160,
      "step": 140355
    },
    {
      "epoch": 15.441144114411442,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.004503115788223614,
      "loss": 0.2308,
      "num_input_tokens_seen": 29621184,
      "step": 140360
    },
    {
      "epoch": 15.441694169416941,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.004502087148768265,
      "loss": 0.2329,
      "num_input_tokens_seen": 29622208,
      "step": 140365
    },
    {
      "epoch": 15.442244224422442,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004501058606067138,
      "loss": 0.2299,
      "num_input_tokens_seen": 29623232,
      "step": 140370
    },
    {
      "epoch": 15.442794279427943,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004500030160129721,
      "loss": 0.2308,
      "num_input_tokens_seen": 29624320,
      "step": 140375
    },
    {
      "epoch": 15.443344334433444,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.004499001810965481,
      "loss": 0.2309,
      "num_input_tokens_seen": 29625376,
      "step": 140380
    },
    {
      "epoch": 15.443894389438944,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004497973558583909,
      "loss": 0.2314,
      "num_input_tokens_seen": 29626496,
      "step": 140385
    },
    {
      "epoch": 15.444444444444445,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004496945402994471,
      "loss": 0.233,
      "num_input_tokens_seen": 29627520,
      "step": 140390
    },
    {
      "epoch": 15.444994499449946,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.004495917344206646,
      "loss": 0.2304,
      "num_input_tokens_seen": 29628576,
      "step": 140395
    },
    {
      "epoch": 15.445544554455445,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004494889382229915,
      "loss": 0.2304,
      "num_input_tokens_seen": 29629632,
      "step": 140400
    },
    {
      "epoch": 15.446094609460946,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004493861517073743,
      "loss": 0.2308,
      "num_input_tokens_seen": 29630688,
      "step": 140405
    },
    {
      "epoch": 15.446644664466447,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.004492833748747613,
      "loss": 0.2319,
      "num_input_tokens_seen": 29631712,
      "step": 140410
    },
    {
      "epoch": 15.447194719471947,
      "grad_norm": 0.005859375,
      "learning_rate": 0.004491806077260989,
      "loss": 0.2304,
      "num_input_tokens_seen": 29632768,
      "step": 140415
    },
    {
      "epoch": 15.447744774477448,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004490778502623345,
      "loss": 0.2324,
      "num_input_tokens_seen": 29633824,
      "step": 140420
    },
    {
      "epoch": 15.448294829482949,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004489751024844158,
      "loss": 0.2314,
      "num_input_tokens_seen": 29634848,
      "step": 140425
    },
    {
      "epoch": 15.448844884488448,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.004488723643932886,
      "loss": 0.2324,
      "num_input_tokens_seen": 29635808,
      "step": 140430
    },
    {
      "epoch": 15.44939493949395,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.004487696359899011,
      "loss": 0.2309,
      "num_input_tokens_seen": 29636832,
      "step": 140435
    },
    {
      "epoch": 15.44994499449945,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.004486669172751994,
      "loss": 0.2314,
      "num_input_tokens_seen": 29637888,
      "step": 140440
    },
    {
      "epoch": 15.450495049504951,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.004485642082501298,
      "loss": 0.2324,
      "num_input_tokens_seen": 29638944,
      "step": 140445
    },
    {
      "epoch": 15.45104510451045,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.004484615089156392,
      "loss": 0.2314,
      "num_input_tokens_seen": 29640128,
      "step": 140450
    },
    {
      "epoch": 15.451595159515952,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.004483588192726744,
      "loss": 0.2324,
      "num_input_tokens_seen": 29641216,
      "step": 140455
    },
    {
      "epoch": 15.452145214521453,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.004482561393221822,
      "loss": 0.2324,
      "num_input_tokens_seen": 29642304,
      "step": 140460
    },
    {
      "epoch": 15.452695269526952,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.004481534690651086,
      "loss": 0.2314,
      "num_input_tokens_seen": 29643360,
      "step": 140465
    },
    {
      "epoch": 15.453245324532453,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0044805080850239916,
      "loss": 0.2293,
      "num_input_tokens_seen": 29644416,
      "step": 140470
    },
    {
      "epoch": 15.453795379537954,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.004479481576350007,
      "loss": 0.2324,
      "num_input_tokens_seen": 29645440,
      "step": 140475
    },
    {
      "epoch": 15.454345434543454,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004478455164638593,
      "loss": 0.2319,
      "num_input_tokens_seen": 29646496,
      "step": 140480
    },
    {
      "epoch": 15.454895489548955,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.004477428849899212,
      "loss": 0.2329,
      "num_input_tokens_seen": 29647488,
      "step": 140485
    },
    {
      "epoch": 15.455445544554456,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.004476402632141321,
      "loss": 0.2308,
      "num_input_tokens_seen": 29648480,
      "step": 140490
    },
    {
      "epoch": 15.455995599559955,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004475376511374373,
      "loss": 0.2293,
      "num_input_tokens_seen": 29649536,
      "step": 140495
    },
    {
      "epoch": 15.456545654565456,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004474350487607833,
      "loss": 0.2324,
      "num_input_tokens_seen": 29650624,
      "step": 140500
    },
    {
      "epoch": 15.457095709570957,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00447332456085115,
      "loss": 0.2303,
      "num_input_tokens_seen": 29651616,
      "step": 140505
    },
    {
      "epoch": 15.457645764576458,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.004472298731113783,
      "loss": 0.2298,
      "num_input_tokens_seen": 29652672,
      "step": 140510
    },
    {
      "epoch": 15.458195819581958,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.004471272998405192,
      "loss": 0.2298,
      "num_input_tokens_seen": 29653696,
      "step": 140515
    },
    {
      "epoch": 15.458745874587459,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.004470247362734821,
      "loss": 0.2314,
      "num_input_tokens_seen": 29654688,
      "step": 140520
    },
    {
      "epoch": 15.45929592959296,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004469221824112131,
      "loss": 0.2324,
      "num_input_tokens_seen": 29655840,
      "step": 140525
    },
    {
      "epoch": 15.45984598459846,
      "grad_norm": 0.005859375,
      "learning_rate": 0.004468196382546567,
      "loss": 0.2329,
      "num_input_tokens_seen": 29656896,
      "step": 140530
    },
    {
      "epoch": 15.46039603960396,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004467171038047582,
      "loss": 0.2309,
      "num_input_tokens_seen": 29657952,
      "step": 140535
    },
    {
      "epoch": 15.460946094609461,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004466145790624634,
      "loss": 0.2314,
      "num_input_tokens_seen": 29659008,
      "step": 140540
    },
    {
      "epoch": 15.46149614961496,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00446512064028716,
      "loss": 0.2308,
      "num_input_tokens_seen": 29660032,
      "step": 140545
    },
    {
      "epoch": 15.462046204620462,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.004464095587044617,
      "loss": 0.2308,
      "num_input_tokens_seen": 29661184,
      "step": 140550
    },
    {
      "epoch": 15.462596259625963,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.004463070630906452,
      "loss": 0.2314,
      "num_input_tokens_seen": 29662240,
      "step": 140555
    },
    {
      "epoch": 15.463146314631462,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004462045771882101,
      "loss": 0.2299,
      "num_input_tokens_seen": 29663264,
      "step": 140560
    },
    {
      "epoch": 15.463696369636963,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.004461021009981019,
      "loss": 0.2335,
      "num_input_tokens_seen": 29664256,
      "step": 140565
    },
    {
      "epoch": 15.464246424642464,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00445999634521265,
      "loss": 0.2319,
      "num_input_tokens_seen": 29665344,
      "step": 140570
    },
    {
      "epoch": 15.464796479647966,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004458971777586442,
      "loss": 0.2314,
      "num_input_tokens_seen": 29666368,
      "step": 140575
    },
    {
      "epoch": 15.465346534653465,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.004457947307111832,
      "loss": 0.2309,
      "num_input_tokens_seen": 29667392,
      "step": 140580
    },
    {
      "epoch": 15.465896589658966,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.00445692293379826,
      "loss": 0.2314,
      "num_input_tokens_seen": 29668416,
      "step": 140585
    },
    {
      "epoch": 15.466446644664467,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.004455898657655168,
      "loss": 0.2304,
      "num_input_tokens_seen": 29669472,
      "step": 140590
    },
    {
      "epoch": 15.466996699669966,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.004454874478691999,
      "loss": 0.2329,
      "num_input_tokens_seen": 29670528,
      "step": 140595
    },
    {
      "epoch": 15.467546754675467,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004453850396918198,
      "loss": 0.2314,
      "num_input_tokens_seen": 29671520,
      "step": 140600
    },
    {
      "epoch": 15.468096809680969,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004452826412343197,
      "loss": 0.2329,
      "num_input_tokens_seen": 29672608,
      "step": 140605
    },
    {
      "epoch": 15.468646864686468,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.004451802524976429,
      "loss": 0.2298,
      "num_input_tokens_seen": 29673696,
      "step": 140610
    },
    {
      "epoch": 15.469196919691969,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.00445077873482734,
      "loss": 0.2309,
      "num_input_tokens_seen": 29674688,
      "step": 140615
    },
    {
      "epoch": 15.46974697469747,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004449755041905356,
      "loss": 0.2303,
      "num_input_tokens_seen": 29675808,
      "step": 140620
    },
    {
      "epoch": 15.47029702970297,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.004448731446219919,
      "loss": 0.2298,
      "num_input_tokens_seen": 29676896,
      "step": 140625
    },
    {
      "epoch": 15.47084708470847,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.004447707947780465,
      "loss": 0.2303,
      "num_input_tokens_seen": 29677952,
      "step": 140630
    },
    {
      "epoch": 15.471397139713972,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004446684546596419,
      "loss": 0.2308,
      "num_input_tokens_seen": 29678976,
      "step": 140635
    },
    {
      "epoch": 15.471947194719473,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.004445661242677221,
      "loss": 0.2309,
      "num_input_tokens_seen": 29680000,
      "step": 140640
    },
    {
      "epoch": 15.472497249724972,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004444638036032296,
      "loss": 0.2308,
      "num_input_tokens_seen": 29681120,
      "step": 140645
    },
    {
      "epoch": 15.473047304730473,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004443614926671077,
      "loss": 0.2324,
      "num_input_tokens_seen": 29682208,
      "step": 140650
    },
    {
      "epoch": 15.473597359735974,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.004442591914602998,
      "loss": 0.2272,
      "num_input_tokens_seen": 29683232,
      "step": 140655
    },
    {
      "epoch": 15.474147414741473,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0044415689998374775,
      "loss": 0.2319,
      "num_input_tokens_seen": 29684320,
      "step": 140660
    },
    {
      "epoch": 15.474697469746975,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.004440546182383956,
      "loss": 0.2319,
      "num_input_tokens_seen": 29685344,
      "step": 140665
    },
    {
      "epoch": 15.475247524752476,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004439523462251847,
      "loss": 0.2329,
      "num_input_tokens_seen": 29686336,
      "step": 140670
    },
    {
      "epoch": 15.475797579757975,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004438500839450588,
      "loss": 0.2313,
      "num_input_tokens_seen": 29687424,
      "step": 140675
    },
    {
      "epoch": 15.476347634763476,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004437478313989593,
      "loss": 0.2314,
      "num_input_tokens_seen": 29688416,
      "step": 140680
    },
    {
      "epoch": 15.476897689768977,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.004436455885878293,
      "loss": 0.2329,
      "num_input_tokens_seen": 29689440,
      "step": 140685
    },
    {
      "epoch": 15.477447744774478,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004435433555126116,
      "loss": 0.2314,
      "num_input_tokens_seen": 29690432,
      "step": 140690
    },
    {
      "epoch": 15.477997799779978,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004434411321742474,
      "loss": 0.2324,
      "num_input_tokens_seen": 29691456,
      "step": 140695
    },
    {
      "epoch": 15.478547854785479,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.004433389185736796,
      "loss": 0.2319,
      "num_input_tokens_seen": 29692544,
      "step": 140700
    },
    {
      "epoch": 15.47909790979098,
      "grad_norm": 0.000537872314453125,
      "learning_rate": 0.0044323671471185,
      "loss": 0.2335,
      "num_input_tokens_seen": 29693536,
      "step": 140705
    },
    {
      "epoch": 15.479647964796479,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.004431345205896996,
      "loss": 0.2329,
      "num_input_tokens_seen": 29694560,
      "step": 140710
    },
    {
      "epoch": 15.48019801980198,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004430323362081723,
      "loss": 0.2329,
      "num_input_tokens_seen": 29695520,
      "step": 140715
    },
    {
      "epoch": 15.480748074807481,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00442930161568208,
      "loss": 0.2308,
      "num_input_tokens_seen": 29696512,
      "step": 140720
    },
    {
      "epoch": 15.48129812981298,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004428279966707499,
      "loss": 0.2309,
      "num_input_tokens_seen": 29697504,
      "step": 140725
    },
    {
      "epoch": 15.481848184818482,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0044272584151673865,
      "loss": 0.2335,
      "num_input_tokens_seen": 29698528,
      "step": 140730
    },
    {
      "epoch": 15.482398239823983,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004426236961071157,
      "loss": 0.2324,
      "num_input_tokens_seen": 29699584,
      "step": 140735
    },
    {
      "epoch": 15.482948294829482,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.004425215604428226,
      "loss": 0.2304,
      "num_input_tokens_seen": 29700640,
      "step": 140740
    },
    {
      "epoch": 15.483498349834983,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0044241943452480136,
      "loss": 0.2329,
      "num_input_tokens_seen": 29701696,
      "step": 140745
    },
    {
      "epoch": 15.484048404840484,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.004423173183539923,
      "loss": 0.2314,
      "num_input_tokens_seen": 29702720,
      "step": 140750
    },
    {
      "epoch": 15.484598459845985,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004422152119313375,
      "loss": 0.2298,
      "num_input_tokens_seen": 29703776,
      "step": 140755
    },
    {
      "epoch": 15.485148514851485,
      "grad_norm": 0.0004482269287109375,
      "learning_rate": 0.004421131152577771,
      "loss": 0.235,
      "num_input_tokens_seen": 29704800,
      "step": 140760
    },
    {
      "epoch": 15.485698569856986,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004420110283342523,
      "loss": 0.2308,
      "num_input_tokens_seen": 29705984,
      "step": 140765
    },
    {
      "epoch": 15.486248624862487,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0044190895116170475,
      "loss": 0.2309,
      "num_input_tokens_seen": 29707072,
      "step": 140770
    },
    {
      "epoch": 15.486798679867986,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004418068837410741,
      "loss": 0.2304,
      "num_input_tokens_seen": 29708096,
      "step": 140775
    },
    {
      "epoch": 15.487348734873487,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004417048260733023,
      "loss": 0.2319,
      "num_input_tokens_seen": 29709120,
      "step": 140780
    },
    {
      "epoch": 15.487898789878988,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004416027781593287,
      "loss": 0.2309,
      "num_input_tokens_seen": 29710208,
      "step": 140785
    },
    {
      "epoch": 15.488448844884488,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.0044150074000009485,
      "loss": 0.2293,
      "num_input_tokens_seen": 29711328,
      "step": 140790
    },
    {
      "epoch": 15.488998899889989,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.004413987115965404,
      "loss": 0.2303,
      "num_input_tokens_seen": 29712352,
      "step": 140795
    },
    {
      "epoch": 15.48954895489549,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004412966929496062,
      "loss": 0.2324,
      "num_input_tokens_seen": 29713440,
      "step": 140800
    },
    {
      "epoch": 15.490099009900991,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.004411946840602325,
      "loss": 0.2303,
      "num_input_tokens_seen": 29714496,
      "step": 140805
    },
    {
      "epoch": 15.49064906490649,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004410926849293591,
      "loss": 0.2329,
      "num_input_tokens_seen": 29715584,
      "step": 140810
    },
    {
      "epoch": 15.491199119911991,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0044099069555792675,
      "loss": 0.2309,
      "num_input_tokens_seen": 29716736,
      "step": 140815
    },
    {
      "epoch": 15.491749174917492,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.004408887159468748,
      "loss": 0.2319,
      "num_input_tokens_seen": 29717792,
      "step": 140820
    },
    {
      "epoch": 15.492299229922992,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.004407867460971427,
      "loss": 0.2308,
      "num_input_tokens_seen": 29718880,
      "step": 140825
    },
    {
      "epoch": 15.492849284928493,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.004406847860096717,
      "loss": 0.2309,
      "num_input_tokens_seen": 29719904,
      "step": 140830
    },
    {
      "epoch": 15.493399339933994,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004405828356854001,
      "loss": 0.2308,
      "num_input_tokens_seen": 29720928,
      "step": 140835
    },
    {
      "epoch": 15.493949394939493,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004404808951252689,
      "loss": 0.2314,
      "num_input_tokens_seen": 29722016,
      "step": 140840
    },
    {
      "epoch": 15.494499449944994,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004403789643302167,
      "loss": 0.2309,
      "num_input_tokens_seen": 29723040,
      "step": 140845
    },
    {
      "epoch": 15.495049504950495,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.004402770433011827,
      "loss": 0.2314,
      "num_input_tokens_seen": 29724160,
      "step": 140850
    },
    {
      "epoch": 15.495599559955995,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004401751320391066,
      "loss": 0.2309,
      "num_input_tokens_seen": 29725248,
      "step": 140855
    },
    {
      "epoch": 15.496149614961496,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.004400732305449279,
      "loss": 0.2319,
      "num_input_tokens_seen": 29726368,
      "step": 140860
    },
    {
      "epoch": 15.496699669966997,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0043997133881958595,
      "loss": 0.2324,
      "num_input_tokens_seen": 29727392,
      "step": 140865
    },
    {
      "epoch": 15.497249724972498,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.004398694568640198,
      "loss": 0.2329,
      "num_input_tokens_seen": 29728480,
      "step": 140870
    },
    {
      "epoch": 15.497799779977997,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004397675846791674,
      "loss": 0.2308,
      "num_input_tokens_seen": 29729536,
      "step": 140875
    },
    {
      "epoch": 15.498349834983498,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00439665722265969,
      "loss": 0.2329,
      "num_input_tokens_seen": 29730560,
      "step": 140880
    },
    {
      "epoch": 15.498899889989,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004395638696253619,
      "loss": 0.2324,
      "num_input_tokens_seen": 29731648,
      "step": 140885
    },
    {
      "epoch": 15.499449944994499,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004394620267582868,
      "loss": 0.2308,
      "num_input_tokens_seen": 29732704,
      "step": 140890
    },
    {
      "epoch": 15.5,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004393601936656811,
      "loss": 0.2308,
      "num_input_tokens_seen": 29733792,
      "step": 140895
    },
    {
      "epoch": 15.500550055005501,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004392583703484832,
      "loss": 0.2324,
      "num_input_tokens_seen": 29734784,
      "step": 140900
    },
    {
      "epoch": 15.501100110011,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.004391565568076324,
      "loss": 0.2324,
      "num_input_tokens_seen": 29735808,
      "step": 140905
    },
    {
      "epoch": 15.501650165016502,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.004390547530440661,
      "loss": 0.2319,
      "num_input_tokens_seen": 29736832,
      "step": 140910
    },
    {
      "epoch": 15.502200220022003,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.00438952959058723,
      "loss": 0.2324,
      "num_input_tokens_seen": 29737824,
      "step": 140915
    },
    {
      "epoch": 15.502750275027502,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004388511748525419,
      "loss": 0.2293,
      "num_input_tokens_seen": 29738912,
      "step": 140920
    },
    {
      "epoch": 15.503300330033003,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004387494004264598,
      "loss": 0.2298,
      "num_input_tokens_seen": 29740000,
      "step": 140925
    },
    {
      "epoch": 15.503850385038504,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004386476357814157,
      "loss": 0.2293,
      "num_input_tokens_seen": 29741024,
      "step": 140930
    },
    {
      "epoch": 15.504400440044005,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00438545880918347,
      "loss": 0.2324,
      "num_input_tokens_seen": 29742080,
      "step": 140935
    },
    {
      "epoch": 15.504950495049505,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.004384441358381912,
      "loss": 0.2319,
      "num_input_tokens_seen": 29743168,
      "step": 140940
    },
    {
      "epoch": 15.505500550055006,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004383424005418864,
      "loss": 0.2319,
      "num_input_tokens_seen": 29744192,
      "step": 140945
    },
    {
      "epoch": 15.506050605060507,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004382406750303702,
      "loss": 0.2319,
      "num_input_tokens_seen": 29745312,
      "step": 140950
    },
    {
      "epoch": 15.506600660066006,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.004381389593045808,
      "loss": 0.2314,
      "num_input_tokens_seen": 29746336,
      "step": 140955
    },
    {
      "epoch": 15.507150715071507,
      "grad_norm": 0.000713348388671875,
      "learning_rate": 0.00438037253365455,
      "loss": 0.2303,
      "num_input_tokens_seen": 29747424,
      "step": 140960
    },
    {
      "epoch": 15.507700770077008,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.0043793555721392975,
      "loss": 0.2314,
      "num_input_tokens_seen": 29748448,
      "step": 140965
    },
    {
      "epoch": 15.508250825082508,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.004378338708509429,
      "loss": 0.2329,
      "num_input_tokens_seen": 29749536,
      "step": 140970
    },
    {
      "epoch": 15.508800880088009,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004377321942774315,
      "loss": 0.2303,
      "num_input_tokens_seen": 29750624,
      "step": 140975
    },
    {
      "epoch": 15.50935093509351,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004376305274943333,
      "loss": 0.2324,
      "num_input_tokens_seen": 29751648,
      "step": 140980
    },
    {
      "epoch": 15.509900990099009,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.004375288705025848,
      "loss": 0.2303,
      "num_input_tokens_seen": 29752672,
      "step": 140985
    },
    {
      "epoch": 15.51045104510451,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004374272233031224,
      "loss": 0.2313,
      "num_input_tokens_seen": 29753760,
      "step": 140990
    },
    {
      "epoch": 15.511001100110011,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.004373255858968838,
      "loss": 0.2303,
      "num_input_tokens_seen": 29754784,
      "step": 140995
    },
    {
      "epoch": 15.511551155115512,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0043722395828480485,
      "loss": 0.2314,
      "num_input_tokens_seen": 29755840,
      "step": 141000
    },
    {
      "epoch": 15.512101210121012,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004371223404678228,
      "loss": 0.2324,
      "num_input_tokens_seen": 29756896,
      "step": 141005
    },
    {
      "epoch": 15.512651265126513,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.004370207324468746,
      "loss": 0.2309,
      "num_input_tokens_seen": 29757984,
      "step": 141010
    },
    {
      "epoch": 15.513201320132014,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0043691913422289575,
      "loss": 0.2319,
      "num_input_tokens_seen": 29759040,
      "step": 141015
    },
    {
      "epoch": 15.513751375137513,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004368175457968236,
      "loss": 0.2303,
      "num_input_tokens_seen": 29760064,
      "step": 141020
    },
    {
      "epoch": 15.514301430143014,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.004367159671695934,
      "loss": 0.2303,
      "num_input_tokens_seen": 29761120,
      "step": 141025
    },
    {
      "epoch": 15.514851485148515,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.004366143983421419,
      "loss": 0.2303,
      "num_input_tokens_seen": 29762240,
      "step": 141030
    },
    {
      "epoch": 15.515401540154015,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.004365128393154059,
      "loss": 0.2298,
      "num_input_tokens_seen": 29763296,
      "step": 141035
    },
    {
      "epoch": 15.515951595159516,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.004364112900903202,
      "loss": 0.2308,
      "num_input_tokens_seen": 29764320,
      "step": 141040
    },
    {
      "epoch": 15.516501650165017,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004363097506678217,
      "loss": 0.2314,
      "num_input_tokens_seen": 29765408,
      "step": 141045
    },
    {
      "epoch": 15.517051705170516,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0043620822104884546,
      "loss": 0.2309,
      "num_input_tokens_seen": 29766432,
      "step": 141050
    },
    {
      "epoch": 15.517601760176017,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.004361067012343279,
      "loss": 0.2314,
      "num_input_tokens_seen": 29767456,
      "step": 141055
    },
    {
      "epoch": 15.518151815181518,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00436005191225204,
      "loss": 0.2308,
      "num_input_tokens_seen": 29768544,
      "step": 141060
    },
    {
      "epoch": 15.51870187018702,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.004359036910224097,
      "loss": 0.2303,
      "num_input_tokens_seen": 29769600,
      "step": 141065
    },
    {
      "epoch": 15.519251925192519,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.004358022006268808,
      "loss": 0.2329,
      "num_input_tokens_seen": 29770688,
      "step": 141070
    },
    {
      "epoch": 15.51980198019802,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.004357007200395519,
      "loss": 0.2298,
      "num_input_tokens_seen": 29771712,
      "step": 141075
    },
    {
      "epoch": 15.520352035203521,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0043559924926135935,
      "loss": 0.2335,
      "num_input_tokens_seen": 29772704,
      "step": 141080
    },
    {
      "epoch": 15.52090209020902,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004354977882932374,
      "loss": 0.2308,
      "num_input_tokens_seen": 29773760,
      "step": 141085
    },
    {
      "epoch": 15.521452145214521,
      "grad_norm": 0.0033111572265625,
      "learning_rate": 0.004353963371361214,
      "loss": 0.2314,
      "num_input_tokens_seen": 29774784,
      "step": 141090
    },
    {
      "epoch": 15.522002200220022,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.004352948957909471,
      "loss": 0.2309,
      "num_input_tokens_seen": 29775872,
      "step": 141095
    },
    {
      "epoch": 15.522552255225522,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.004351934642586482,
      "loss": 0.2293,
      "num_input_tokens_seen": 29776896,
      "step": 141100
    },
    {
      "epoch": 15.523102310231023,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004350920425401608,
      "loss": 0.2319,
      "num_input_tokens_seen": 29777984,
      "step": 141105
    },
    {
      "epoch": 15.523652365236524,
      "grad_norm": 0.000621795654296875,
      "learning_rate": 0.004349906306364189,
      "loss": 0.2314,
      "num_input_tokens_seen": 29779040,
      "step": 141110
    },
    {
      "epoch": 15.524202420242025,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.004348892285483571,
      "loss": 0.2303,
      "num_input_tokens_seen": 29780096,
      "step": 141115
    },
    {
      "epoch": 15.524752475247524,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0043478783627691,
      "loss": 0.2293,
      "num_input_tokens_seen": 29781248,
      "step": 141120
    },
    {
      "epoch": 15.525302530253025,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004346864538230128,
      "loss": 0.2329,
      "num_input_tokens_seen": 29782272,
      "step": 141125
    },
    {
      "epoch": 15.525852585258527,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0043458508118759875,
      "loss": 0.2319,
      "num_input_tokens_seen": 29783232,
      "step": 141130
    },
    {
      "epoch": 15.526402640264026,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.004344837183716034,
      "loss": 0.2293,
      "num_input_tokens_seen": 29784320,
      "step": 141135
    },
    {
      "epoch": 15.526952695269527,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004343823653759599,
      "loss": 0.2299,
      "num_input_tokens_seen": 29785408,
      "step": 141140
    },
    {
      "epoch": 15.527502750275028,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.004342810222016027,
      "loss": 0.2314,
      "num_input_tokens_seen": 29786496,
      "step": 141145
    },
    {
      "epoch": 15.528052805280527,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.004341796888494665,
      "loss": 0.2314,
      "num_input_tokens_seen": 29787520,
      "step": 141150
    },
    {
      "epoch": 15.528602860286028,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.004340783653204841,
      "loss": 0.2303,
      "num_input_tokens_seen": 29788608,
      "step": 141155
    },
    {
      "epoch": 15.52915291529153,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.004339770516155906,
      "loss": 0.2308,
      "num_input_tokens_seen": 29789664,
      "step": 141160
    },
    {
      "epoch": 15.52970297029703,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.004338757477357184,
      "loss": 0.2308,
      "num_input_tokens_seen": 29790720,
      "step": 141165
    },
    {
      "epoch": 15.53025302530253,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004337744536818023,
      "loss": 0.2314,
      "num_input_tokens_seen": 29791776,
      "step": 141170
    },
    {
      "epoch": 15.530803080308031,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.004336731694547751,
      "loss": 0.2324,
      "num_input_tokens_seen": 29792832,
      "step": 141175
    },
    {
      "epoch": 15.531353135313532,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.004335718950555706,
      "loss": 0.2298,
      "num_input_tokens_seen": 29793888,
      "step": 141180
    },
    {
      "epoch": 15.531903190319031,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.004334706304851227,
      "loss": 0.2309,
      "num_input_tokens_seen": 29794912,
      "step": 141185
    },
    {
      "epoch": 15.532453245324533,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.004333693757443639,
      "loss": 0.2324,
      "num_input_tokens_seen": 29796000,
      "step": 141190
    },
    {
      "epoch": 15.533003300330034,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0043326813083422785,
      "loss": 0.2329,
      "num_input_tokens_seen": 29796992,
      "step": 141195
    },
    {
      "epoch": 15.533553355335533,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0043316689575564745,
      "loss": 0.2324,
      "num_input_tokens_seen": 29798144,
      "step": 141200
    },
    {
      "epoch": 15.534103410341034,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.004330656705095557,
      "loss": 0.2314,
      "num_input_tokens_seen": 29799168,
      "step": 141205
    },
    {
      "epoch": 15.534653465346535,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0043296445509688605,
      "loss": 0.2314,
      "num_input_tokens_seen": 29800160,
      "step": 141210
    },
    {
      "epoch": 15.535203520352034,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004328632495185707,
      "loss": 0.2313,
      "num_input_tokens_seen": 29801216,
      "step": 141215
    },
    {
      "epoch": 15.535753575357536,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.004327620537755431,
      "loss": 0.2324,
      "num_input_tokens_seen": 29802240,
      "step": 141220
    },
    {
      "epoch": 15.536303630363037,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.004326608678687357,
      "loss": 0.2319,
      "num_input_tokens_seen": 29803264,
      "step": 141225
    },
    {
      "epoch": 15.536853685368538,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.004325596917990804,
      "loss": 0.2314,
      "num_input_tokens_seen": 29804352,
      "step": 141230
    },
    {
      "epoch": 15.537403740374037,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0043245852556751015,
      "loss": 0.2308,
      "num_input_tokens_seen": 29805408,
      "step": 141235
    },
    {
      "epoch": 15.537953795379538,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.004323573691749574,
      "loss": 0.2314,
      "num_input_tokens_seen": 29806432,
      "step": 141240
    },
    {
      "epoch": 15.53850385038504,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004322562226223549,
      "loss": 0.2319,
      "num_input_tokens_seen": 29807488,
      "step": 141245
    },
    {
      "epoch": 15.539053905390539,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.004321550859106345,
      "loss": 0.2324,
      "num_input_tokens_seen": 29808544,
      "step": 141250
    },
    {
      "epoch": 15.53960396039604,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.0043205395904072795,
      "loss": 0.2309,
      "num_input_tokens_seen": 29809696,
      "step": 141255
    },
    {
      "epoch": 15.54015401540154,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.004319528420135673,
      "loss": 0.2324,
      "num_input_tokens_seen": 29810688,
      "step": 141260
    },
    {
      "epoch": 15.54070407040704,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.004318517348300852,
      "loss": 0.2329,
      "num_input_tokens_seen": 29811680,
      "step": 141265
    },
    {
      "epoch": 15.541254125412541,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.004317506374912132,
      "loss": 0.2303,
      "num_input_tokens_seen": 29812768,
      "step": 141270
    },
    {
      "epoch": 15.541804180418042,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.004316495499978832,
      "loss": 0.2314,
      "num_input_tokens_seen": 29813792,
      "step": 141275
    },
    {
      "epoch": 15.542354235423542,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004315484723510261,
      "loss": 0.2288,
      "num_input_tokens_seen": 29814816,
      "step": 141280
    },
    {
      "epoch": 15.542904290429043,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.004314474045515746,
      "loss": 0.2314,
      "num_input_tokens_seen": 29815904,
      "step": 141285
    },
    {
      "epoch": 15.543454345434544,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004313463466004591,
      "loss": 0.2309,
      "num_input_tokens_seen": 29816928,
      "step": 141290
    },
    {
      "epoch": 15.544004400440045,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.004312452984986114,
      "loss": 0.2324,
      "num_input_tokens_seen": 29817888,
      "step": 141295
    },
    {
      "epoch": 15.544554455445544,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004311442602469636,
      "loss": 0.2324,
      "num_input_tokens_seen": 29818912,
      "step": 141300
    },
    {
      "epoch": 15.545104510451045,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.004310432318464457,
      "loss": 0.2298,
      "num_input_tokens_seen": 29819936,
      "step": 141305
    },
    {
      "epoch": 15.545654565456546,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.004309422132979898,
      "loss": 0.2319,
      "num_input_tokens_seen": 29820992,
      "step": 141310
    },
    {
      "epoch": 15.546204620462046,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004308412046025264,
      "loss": 0.2319,
      "num_input_tokens_seen": 29822016,
      "step": 141315
    },
    {
      "epoch": 15.546754675467547,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.004307402057609862,
      "loss": 0.2314,
      "num_input_tokens_seen": 29823072,
      "step": 141320
    },
    {
      "epoch": 15.547304730473048,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.00430639216774301,
      "loss": 0.2308,
      "num_input_tokens_seen": 29824128,
      "step": 141325
    },
    {
      "epoch": 15.547854785478547,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.004305382376434006,
      "loss": 0.2309,
      "num_input_tokens_seen": 29825248,
      "step": 141330
    },
    {
      "epoch": 15.548404840484048,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.004304372683692165,
      "loss": 0.2335,
      "num_input_tokens_seen": 29826336,
      "step": 141335
    },
    {
      "epoch": 15.54895489548955,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004303363089526791,
      "loss": 0.2308,
      "num_input_tokens_seen": 29827360,
      "step": 141340
    },
    {
      "epoch": 15.549504950495049,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.00430235359394718,
      "loss": 0.2298,
      "num_input_tokens_seen": 29828448,
      "step": 141345
    },
    {
      "epoch": 15.55005500550055,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.004301344196962643,
      "loss": 0.2319,
      "num_input_tokens_seen": 29829536,
      "step": 141350
    },
    {
      "epoch": 15.55060506050605,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.004300334898582483,
      "loss": 0.2303,
      "num_input_tokens_seen": 29830592,
      "step": 141355
    },
    {
      "epoch": 15.551155115511552,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.0042993256988160065,
      "loss": 0.2319,
      "num_input_tokens_seen": 29831680,
      "step": 141360
    },
    {
      "epoch": 15.551705170517051,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0042983165976725095,
      "loss": 0.2303,
      "num_input_tokens_seen": 29832704,
      "step": 141365
    },
    {
      "epoch": 15.552255225522552,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004297307595161289,
      "loss": 0.2308,
      "num_input_tokens_seen": 29833760,
      "step": 141370
    },
    {
      "epoch": 15.552805280528053,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.00429629869129165,
      "loss": 0.2324,
      "num_input_tokens_seen": 29834848,
      "step": 141375
    },
    {
      "epoch": 15.553355335533553,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004295289886072889,
      "loss": 0.2303,
      "num_input_tokens_seen": 29835904,
      "step": 141380
    },
    {
      "epoch": 15.553905390539054,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004294281179514308,
      "loss": 0.2303,
      "num_input_tokens_seen": 29836960,
      "step": 141385
    },
    {
      "epoch": 15.554455445544555,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004293272571625202,
      "loss": 0.2298,
      "num_input_tokens_seen": 29838016,
      "step": 141390
    },
    {
      "epoch": 15.555005500550054,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.004292264062414861,
      "loss": 0.2298,
      "num_input_tokens_seen": 29839072,
      "step": 141395
    },
    {
      "epoch": 15.555555555555555,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.0042912556518925875,
      "loss": 0.2324,
      "num_input_tokens_seen": 29840160,
      "step": 141400
    },
    {
      "epoch": 15.556105610561056,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004290247340067667,
      "loss": 0.2329,
      "num_input_tokens_seen": 29841216,
      "step": 141405
    },
    {
      "epoch": 15.556655665566556,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004289239126949399,
      "loss": 0.2314,
      "num_input_tokens_seen": 29842176,
      "step": 141410
    },
    {
      "epoch": 15.557205720572057,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004288231012547077,
      "loss": 0.2319,
      "num_input_tokens_seen": 29843328,
      "step": 141415
    },
    {
      "epoch": 15.557755775577558,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.004287222996869987,
      "loss": 0.2298,
      "num_input_tokens_seen": 29844320,
      "step": 141420
    },
    {
      "epoch": 15.558305830583059,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004286215079927427,
      "loss": 0.2324,
      "num_input_tokens_seen": 29845408,
      "step": 141425
    },
    {
      "epoch": 15.558855885588558,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.004285207261728675,
      "loss": 0.2319,
      "num_input_tokens_seen": 29846496,
      "step": 141430
    },
    {
      "epoch": 15.55940594059406,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.004284199542283028,
      "loss": 0.234,
      "num_input_tokens_seen": 29847520,
      "step": 141435
    },
    {
      "epoch": 15.55995599559956,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004283191921599777,
      "loss": 0.2345,
      "num_input_tokens_seen": 29848512,
      "step": 141440
    },
    {
      "epoch": 15.56050605060506,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004282184399688197,
      "loss": 0.2308,
      "num_input_tokens_seen": 29849536,
      "step": 141445
    },
    {
      "epoch": 15.561056105610561,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.004281176976557587,
      "loss": 0.2335,
      "num_input_tokens_seen": 29850592,
      "step": 141450
    },
    {
      "epoch": 15.561606160616062,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004280169652217219,
      "loss": 0.2335,
      "num_input_tokens_seen": 29851648,
      "step": 141455
    },
    {
      "epoch": 15.562156215621561,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.004279162426676391,
      "loss": 0.2314,
      "num_input_tokens_seen": 29852640,
      "step": 141460
    },
    {
      "epoch": 15.562706270627062,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004278155299944371,
      "loss": 0.2303,
      "num_input_tokens_seen": 29853728,
      "step": 141465
    },
    {
      "epoch": 15.563256325632564,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.004277148272030448,
      "loss": 0.2308,
      "num_input_tokens_seen": 29854816,
      "step": 141470
    },
    {
      "epoch": 15.563806380638063,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004276141342943911,
      "loss": 0.2303,
      "num_input_tokens_seen": 29855904,
      "step": 141475
    },
    {
      "epoch": 15.564356435643564,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.004275134512694029,
      "loss": 0.2309,
      "num_input_tokens_seen": 29856960,
      "step": 141480
    },
    {
      "epoch": 15.564906490649065,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.00427412778129009,
      "loss": 0.2314,
      "num_input_tokens_seen": 29857984,
      "step": 141485
    },
    {
      "epoch": 15.565456545654566,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.004273121148741368,
      "loss": 0.2329,
      "num_input_tokens_seen": 29859008,
      "step": 141490
    },
    {
      "epoch": 15.566006600660065,
      "grad_norm": 0.005859375,
      "learning_rate": 0.004272114615057131,
      "loss": 0.2319,
      "num_input_tokens_seen": 29860032,
      "step": 141495
    },
    {
      "epoch": 15.566556655665567,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.004271108180246676,
      "loss": 0.2283,
      "num_input_tokens_seen": 29861088,
      "step": 141500
    },
    {
      "epoch": 15.567106710671068,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.004270101844319269,
      "loss": 0.2298,
      "num_input_tokens_seen": 29862176,
      "step": 141505
    },
    {
      "epoch": 15.567656765676567,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.004269095607284181,
      "loss": 0.2309,
      "num_input_tokens_seen": 29863296,
      "step": 141510
    },
    {
      "epoch": 15.568206820682068,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.004268089469150692,
      "loss": 0.2314,
      "num_input_tokens_seen": 29864288,
      "step": 141515
    },
    {
      "epoch": 15.56875687568757,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004267083429928069,
      "loss": 0.2308,
      "num_input_tokens_seen": 29865440,
      "step": 141520
    },
    {
      "epoch": 15.569306930693068,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0042660774896255875,
      "loss": 0.2309,
      "num_input_tokens_seen": 29866464,
      "step": 141525
    },
    {
      "epoch": 15.56985698569857,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.004265071648252524,
      "loss": 0.2308,
      "num_input_tokens_seen": 29867520,
      "step": 141530
    },
    {
      "epoch": 15.57040704070407,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004264065905818141,
      "loss": 0.2314,
      "num_input_tokens_seen": 29868608,
      "step": 141535
    },
    {
      "epoch": 15.570957095709572,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004263060262331713,
      "loss": 0.2324,
      "num_input_tokens_seen": 29869632,
      "step": 141540
    },
    {
      "epoch": 15.571507150715071,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.004262054717802504,
      "loss": 0.2303,
      "num_input_tokens_seen": 29870624,
      "step": 141545
    },
    {
      "epoch": 15.572057205720572,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004261049272239781,
      "loss": 0.2309,
      "num_input_tokens_seen": 29871712,
      "step": 141550
    },
    {
      "epoch": 15.572607260726073,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.0042600439256528224,
      "loss": 0.2298,
      "num_input_tokens_seen": 29872768,
      "step": 141555
    },
    {
      "epoch": 15.573157315731573,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.004259038678050879,
      "loss": 0.2324,
      "num_input_tokens_seen": 29873824,
      "step": 141560
    },
    {
      "epoch": 15.573707370737074,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.004258033529443226,
      "loss": 0.2314,
      "num_input_tokens_seen": 29874880,
      "step": 141565
    },
    {
      "epoch": 15.574257425742575,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004257028479839119,
      "loss": 0.2314,
      "num_input_tokens_seen": 29875904,
      "step": 141570
    },
    {
      "epoch": 15.574807480748074,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004256023529247831,
      "loss": 0.2324,
      "num_input_tokens_seen": 29877024,
      "step": 141575
    },
    {
      "epoch": 15.575357535753575,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004255018677678615,
      "loss": 0.2298,
      "num_input_tokens_seen": 29878048,
      "step": 141580
    },
    {
      "epoch": 15.575907590759076,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.004254013925140733,
      "loss": 0.2298,
      "num_input_tokens_seen": 29879072,
      "step": 141585
    },
    {
      "epoch": 15.576457645764577,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.004253009271643454,
      "loss": 0.2319,
      "num_input_tokens_seen": 29880160,
      "step": 141590
    },
    {
      "epoch": 15.577007700770077,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0042520047171960285,
      "loss": 0.2303,
      "num_input_tokens_seen": 29881216,
      "step": 141595
    },
    {
      "epoch": 15.577557755775578,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.004251000261807722,
      "loss": 0.2314,
      "num_input_tokens_seen": 29882208,
      "step": 141600
    },
    {
      "epoch": 15.578107810781079,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0042499959054877875,
      "loss": 0.2308,
      "num_input_tokens_seen": 29883232,
      "step": 141605
    },
    {
      "epoch": 15.578657865786578,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0042489916482454735,
      "loss": 0.2303,
      "num_input_tokens_seen": 29884256,
      "step": 141610
    },
    {
      "epoch": 15.57920792079208,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0042479874900900536,
      "loss": 0.2329,
      "num_input_tokens_seen": 29885376,
      "step": 141615
    },
    {
      "epoch": 15.57975797579758,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00424698343103077,
      "loss": 0.2283,
      "num_input_tokens_seen": 29886432,
      "step": 141620
    },
    {
      "epoch": 15.58030803080308,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.004245979471076884,
      "loss": 0.2314,
      "num_input_tokens_seen": 29887520,
      "step": 141625
    },
    {
      "epoch": 15.58085808580858,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004244975610237646,
      "loss": 0.2314,
      "num_input_tokens_seen": 29888576,
      "step": 141630
    },
    {
      "epoch": 15.581408140814082,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004243971848522301,
      "loss": 0.2308,
      "num_input_tokens_seen": 29889664,
      "step": 141635
    },
    {
      "epoch": 15.581958195819581,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.004242968185940108,
      "loss": 0.2314,
      "num_input_tokens_seen": 29890720,
      "step": 141640
    },
    {
      "epoch": 15.582508250825082,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.0042419646225003126,
      "loss": 0.2309,
      "num_input_tokens_seen": 29891808,
      "step": 141645
    },
    {
      "epoch": 15.583058305830583,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.004240961158212175,
      "loss": 0.2303,
      "num_input_tokens_seen": 29892864,
      "step": 141650
    },
    {
      "epoch": 15.583608360836084,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.004239957793084933,
      "loss": 0.2319,
      "num_input_tokens_seen": 29893888,
      "step": 141655
    },
    {
      "epoch": 15.584158415841584,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.004238954527127833,
      "loss": 0.2308,
      "num_input_tokens_seen": 29894976,
      "step": 141660
    },
    {
      "epoch": 15.584708470847085,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004237951360350132,
      "loss": 0.2324,
      "num_input_tokens_seen": 29896032,
      "step": 141665
    },
    {
      "epoch": 15.585258525852586,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0042369482927610575,
      "loss": 0.2314,
      "num_input_tokens_seen": 29897152,
      "step": 141670
    },
    {
      "epoch": 15.585808580858085,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.004235945324369876,
      "loss": 0.2303,
      "num_input_tokens_seen": 29898176,
      "step": 141675
    },
    {
      "epoch": 15.586358635863586,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.004234942455185821,
      "loss": 0.2319,
      "num_input_tokens_seen": 29899200,
      "step": 141680
    },
    {
      "epoch": 15.586908690869087,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.004233939685218132,
      "loss": 0.2303,
      "num_input_tokens_seen": 29900320,
      "step": 141685
    },
    {
      "epoch": 15.587458745874587,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00423293701447606,
      "loss": 0.2308,
      "num_input_tokens_seen": 29901344,
      "step": 141690
    },
    {
      "epoch": 15.588008800880088,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004231934442968835,
      "loss": 0.2329,
      "num_input_tokens_seen": 29902400,
      "step": 141695
    },
    {
      "epoch": 15.588558855885589,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.004230931970705704,
      "loss": 0.2308,
      "num_input_tokens_seen": 29903456,
      "step": 141700
    },
    {
      "epoch": 15.589108910891088,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0042299295976959105,
      "loss": 0.2308,
      "num_input_tokens_seen": 29904512,
      "step": 141705
    },
    {
      "epoch": 15.58965896589659,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.004228927323948682,
      "loss": 0.2309,
      "num_input_tokens_seen": 29905568,
      "step": 141710
    },
    {
      "epoch": 15.59020902090209,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.004227925149473267,
      "loss": 0.2319,
      "num_input_tokens_seen": 29906624,
      "step": 141715
    },
    {
      "epoch": 15.590759075907592,
      "grad_norm": 0.0008392333984375,
      "learning_rate": 0.004226923074278898,
      "loss": 0.2324,
      "num_input_tokens_seen": 29907616,
      "step": 141720
    },
    {
      "epoch": 15.591309130913091,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004225921098374804,
      "loss": 0.2319,
      "num_input_tokens_seen": 29908704,
      "step": 141725
    },
    {
      "epoch": 15.591859185918592,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.004224919221770226,
      "loss": 0.2314,
      "num_input_tokens_seen": 29909792,
      "step": 141730
    },
    {
      "epoch": 15.592409240924093,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004223917444474395,
      "loss": 0.2298,
      "num_input_tokens_seen": 29910816,
      "step": 141735
    },
    {
      "epoch": 15.592959295929592,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0042229157664965536,
      "loss": 0.2298,
      "num_input_tokens_seen": 29911904,
      "step": 141740
    },
    {
      "epoch": 15.593509350935093,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004221914187845926,
      "loss": 0.2309,
      "num_input_tokens_seen": 29913024,
      "step": 141745
    },
    {
      "epoch": 15.594059405940595,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004220912708531736,
      "loss": 0.2313,
      "num_input_tokens_seen": 29914016,
      "step": 141750
    },
    {
      "epoch": 15.594609460946094,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004219911328563223,
      "loss": 0.2324,
      "num_input_tokens_seen": 29915040,
      "step": 141755
    },
    {
      "epoch": 15.595159515951595,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004218910047949615,
      "loss": 0.2309,
      "num_input_tokens_seen": 29916192,
      "step": 141760
    },
    {
      "epoch": 15.595709570957096,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.004217908866700143,
      "loss": 0.2308,
      "num_input_tokens_seen": 29917280,
      "step": 141765
    },
    {
      "epoch": 15.596259625962595,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.004216907784824033,
      "loss": 0.2314,
      "num_input_tokens_seen": 29918336,
      "step": 141770
    },
    {
      "epoch": 15.596809680968097,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.004215906802330505,
      "loss": 0.2298,
      "num_input_tokens_seen": 29919392,
      "step": 141775
    },
    {
      "epoch": 15.597359735973598,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0042149059192287935,
      "loss": 0.2324,
      "num_input_tokens_seen": 29920480,
      "step": 141780
    },
    {
      "epoch": 15.597909790979099,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004213905135528114,
      "loss": 0.2319,
      "num_input_tokens_seen": 29921536,
      "step": 141785
    },
    {
      "epoch": 15.598459845984598,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.004212904451237695,
      "loss": 0.2314,
      "num_input_tokens_seen": 29922688,
      "step": 141790
    },
    {
      "epoch": 15.599009900990099,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.004211903866366765,
      "loss": 0.2308,
      "num_input_tokens_seen": 29923680,
      "step": 141795
    },
    {
      "epoch": 15.5995599559956,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.0042109033809245335,
      "loss": 0.2309,
      "num_input_tokens_seen": 29924768,
      "step": 141800
    },
    {
      "epoch": 15.6001100110011,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.004209902994920235,
      "loss": 0.2314,
      "num_input_tokens_seen": 29925792,
      "step": 141805
    },
    {
      "epoch": 15.6006600660066,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.004208902708363079,
      "loss": 0.2309,
      "num_input_tokens_seen": 29926880,
      "step": 141810
    },
    {
      "epoch": 15.601210121012102,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.004207902521262288,
      "loss": 0.2314,
      "num_input_tokens_seen": 29927872,
      "step": 141815
    },
    {
      "epoch": 15.601760176017601,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004206902433627085,
      "loss": 0.2324,
      "num_input_tokens_seen": 29928960,
      "step": 141820
    },
    {
      "epoch": 15.602310231023102,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.00420590244546668,
      "loss": 0.2309,
      "num_input_tokens_seen": 29930048,
      "step": 141825
    },
    {
      "epoch": 15.602860286028603,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.004204902556790295,
      "loss": 0.2304,
      "num_input_tokens_seen": 29931136,
      "step": 141830
    },
    {
      "epoch": 15.603410341034103,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.004203902767607141,
      "loss": 0.2314,
      "num_input_tokens_seen": 29932192,
      "step": 141835
    },
    {
      "epoch": 15.603960396039604,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.004202903077926438,
      "loss": 0.2309,
      "num_input_tokens_seen": 29933312,
      "step": 141840
    },
    {
      "epoch": 15.604510451045105,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.004201903487757391,
      "loss": 0.2303,
      "num_input_tokens_seen": 29934304,
      "step": 141845
    },
    {
      "epoch": 15.605060506050606,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.00420090399710922,
      "loss": 0.2319,
      "num_input_tokens_seen": 29935424,
      "step": 141850
    },
    {
      "epoch": 15.605610561056105,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.004199904605991138,
      "loss": 0.2303,
      "num_input_tokens_seen": 29936384,
      "step": 141855
    },
    {
      "epoch": 15.606160616061606,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004198905314412348,
      "loss": 0.2319,
      "num_input_tokens_seen": 29937408,
      "step": 141860
    },
    {
      "epoch": 15.606710671067107,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004197906122382071,
      "loss": 0.2335,
      "num_input_tokens_seen": 29938464,
      "step": 141865
    },
    {
      "epoch": 15.607260726072607,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.004196907029909504,
      "loss": 0.2309,
      "num_input_tokens_seen": 29939552,
      "step": 141870
    },
    {
      "epoch": 15.607810781078108,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004195908037003863,
      "loss": 0.2308,
      "num_input_tokens_seen": 29940608,
      "step": 141875
    },
    {
      "epoch": 15.608360836083609,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.004194909143674356,
      "loss": 0.2324,
      "num_input_tokens_seen": 29941632,
      "step": 141880
    },
    {
      "epoch": 15.608910891089108,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.004193910349930187,
      "loss": 0.2309,
      "num_input_tokens_seen": 29942688,
      "step": 141885
    },
    {
      "epoch": 15.60946094609461,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004192911655780556,
      "loss": 0.2293,
      "num_input_tokens_seen": 29943712,
      "step": 141890
    },
    {
      "epoch": 15.61001100110011,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004191913061234677,
      "loss": 0.2303,
      "num_input_tokens_seen": 29944768,
      "step": 141895
    },
    {
      "epoch": 15.61056105610561,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004190914566301747,
      "loss": 0.2298,
      "num_input_tokens_seen": 29945856,
      "step": 141900
    },
    {
      "epoch": 15.61111111111111,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.004189916170990969,
      "loss": 0.2303,
      "num_input_tokens_seen": 29946912,
      "step": 141905
    },
    {
      "epoch": 15.611661166116612,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00418891787531155,
      "loss": 0.2319,
      "num_input_tokens_seen": 29947968,
      "step": 141910
    },
    {
      "epoch": 15.612211221122113,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.004187919679272684,
      "loss": 0.2303,
      "num_input_tokens_seen": 29949056,
      "step": 141915
    },
    {
      "epoch": 15.612761276127612,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004186921582883578,
      "loss": 0.2293,
      "num_input_tokens_seen": 29950080,
      "step": 141920
    },
    {
      "epoch": 15.613311331133113,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004185923586153423,
      "loss": 0.2314,
      "num_input_tokens_seen": 29951136,
      "step": 141925
    },
    {
      "epoch": 15.613861386138614,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.004184925689091422,
      "loss": 0.2324,
      "num_input_tokens_seen": 29952224,
      "step": 141930
    },
    {
      "epoch": 15.614411441144114,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.004183927891706776,
      "loss": 0.2314,
      "num_input_tokens_seen": 29953280,
      "step": 141935
    },
    {
      "epoch": 15.614961496149615,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.00418293019400867,
      "loss": 0.2319,
      "num_input_tokens_seen": 29954272,
      "step": 141940
    },
    {
      "epoch": 15.615511551155116,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004181932596006312,
      "loss": 0.2314,
      "num_input_tokens_seen": 29955296,
      "step": 141945
    },
    {
      "epoch": 15.616061606160617,
      "grad_norm": 0.000560760498046875,
      "learning_rate": 0.004180935097708886,
      "loss": 0.2319,
      "num_input_tokens_seen": 29956256,
      "step": 141950
    },
    {
      "epoch": 15.616611661166116,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004179937699125593,
      "loss": 0.2293,
      "num_input_tokens_seen": 29957344,
      "step": 141955
    },
    {
      "epoch": 15.617161716171617,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.004178940400265618,
      "loss": 0.2298,
      "num_input_tokens_seen": 29958400,
      "step": 141960
    },
    {
      "epoch": 15.617711771177119,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.004177943201138157,
      "loss": 0.2303,
      "num_input_tokens_seen": 29959520,
      "step": 141965
    },
    {
      "epoch": 15.618261826182618,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.004176946101752406,
      "loss": 0.2319,
      "num_input_tokens_seen": 29960544,
      "step": 141970
    },
    {
      "epoch": 15.618811881188119,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0041759491021175444,
      "loss": 0.2298,
      "num_input_tokens_seen": 29961600,
      "step": 141975
    },
    {
      "epoch": 15.61936193619362,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.004174952202242769,
      "loss": 0.2303,
      "num_input_tokens_seen": 29962624,
      "step": 141980
    },
    {
      "epoch": 15.61991199119912,
      "grad_norm": 0.000812530517578125,
      "learning_rate": 0.004173955402137262,
      "loss": 0.2298,
      "num_input_tokens_seen": 29963584,
      "step": 141985
    },
    {
      "epoch": 15.62046204620462,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004172958701810213,
      "loss": 0.2309,
      "num_input_tokens_seen": 29964640,
      "step": 141990
    },
    {
      "epoch": 15.621012101210122,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.004171962101270813,
      "loss": 0.2324,
      "num_input_tokens_seen": 29965728,
      "step": 141995
    },
    {
      "epoch": 15.62156215621562,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.004170965600528237,
      "loss": 0.2304,
      "num_input_tokens_seen": 29966816,
      "step": 142000
    },
    {
      "epoch": 15.622112211221122,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00416996919959168,
      "loss": 0.2309,
      "num_input_tokens_seen": 29967872,
      "step": 142005
    },
    {
      "epoch": 15.622662266226623,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.00416897289847032,
      "loss": 0.2314,
      "num_input_tokens_seen": 29968960,
      "step": 142010
    },
    {
      "epoch": 15.623212321232124,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.004167976697173336,
      "loss": 0.2324,
      "num_input_tokens_seen": 29970016,
      "step": 142015
    },
    {
      "epoch": 15.623762376237623,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.004166980595709913,
      "loss": 0.2314,
      "num_input_tokens_seen": 29971040,
      "step": 142020
    },
    {
      "epoch": 15.624312431243125,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00416598459408923,
      "loss": 0.2324,
      "num_input_tokens_seen": 29972128,
      "step": 142025
    },
    {
      "epoch": 15.624862486248626,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.004164988692320473,
      "loss": 0.233,
      "num_input_tokens_seen": 29973152,
      "step": 142030
    },
    {
      "epoch": 15.625412541254125,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.004163992890412817,
      "loss": 0.2308,
      "num_input_tokens_seen": 29974208,
      "step": 142035
    },
    {
      "epoch": 15.625962596259626,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004162997188375435,
      "loss": 0.2309,
      "num_input_tokens_seen": 29975296,
      "step": 142040
    },
    {
      "epoch": 15.626512651265127,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.004162001586217507,
      "loss": 0.2308,
      "num_input_tokens_seen": 29976288,
      "step": 142045
    },
    {
      "epoch": 15.627062706270626,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.00416100608394821,
      "loss": 0.2324,
      "num_input_tokens_seen": 29977312,
      "step": 142050
    },
    {
      "epoch": 15.627612761276128,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.004160010681576723,
      "loss": 0.2314,
      "num_input_tokens_seen": 29978400,
      "step": 142055
    },
    {
      "epoch": 15.628162816281629,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.004159015379112215,
      "loss": 0.2314,
      "num_input_tokens_seen": 29979456,
      "step": 142060
    },
    {
      "epoch": 15.628712871287128,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004158020176563858,
      "loss": 0.2308,
      "num_input_tokens_seen": 29980512,
      "step": 142065
    },
    {
      "epoch": 15.629262926292629,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00415702507394083,
      "loss": 0.2293,
      "num_input_tokens_seen": 29981568,
      "step": 142070
    },
    {
      "epoch": 15.62981298129813,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.004156030071252294,
      "loss": 0.2319,
      "num_input_tokens_seen": 29982688,
      "step": 142075
    },
    {
      "epoch": 15.630363036303631,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004155035168507425,
      "loss": 0.2319,
      "num_input_tokens_seen": 29983680,
      "step": 142080
    },
    {
      "epoch": 15.63091309130913,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.004154040365715399,
      "loss": 0.2324,
      "num_input_tokens_seen": 29984832,
      "step": 142085
    },
    {
      "epoch": 15.631463146314632,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0041530456628853735,
      "loss": 0.2319,
      "num_input_tokens_seen": 29985888,
      "step": 142090
    },
    {
      "epoch": 15.632013201320133,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004152051060026526,
      "loss": 0.2319,
      "num_input_tokens_seen": 29986976,
      "step": 142095
    },
    {
      "epoch": 15.632563256325632,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004151056557148012,
      "loss": 0.234,
      "num_input_tokens_seen": 29988032,
      "step": 142100
    },
    {
      "epoch": 15.633113311331133,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004150062154259005,
      "loss": 0.2298,
      "num_input_tokens_seen": 29989120,
      "step": 142105
    },
    {
      "epoch": 15.633663366336634,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0041490678513686746,
      "loss": 0.2298,
      "num_input_tokens_seen": 29990240,
      "step": 142110
    },
    {
      "epoch": 15.634213421342134,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004148073648486174,
      "loss": 0.2319,
      "num_input_tokens_seen": 29991296,
      "step": 142115
    },
    {
      "epoch": 15.634763476347635,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004147079545620675,
      "loss": 0.2298,
      "num_input_tokens_seen": 29992352,
      "step": 142120
    },
    {
      "epoch": 15.635313531353136,
      "grad_norm": 0.005859375,
      "learning_rate": 0.004146085542781336,
      "loss": 0.2309,
      "num_input_tokens_seen": 29993408,
      "step": 142125
    },
    {
      "epoch": 15.635863586358635,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.004145091639977316,
      "loss": 0.2308,
      "num_input_tokens_seen": 29994432,
      "step": 142130
    },
    {
      "epoch": 15.636413641364136,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004144097837217775,
      "loss": 0.2319,
      "num_input_tokens_seen": 29995520,
      "step": 142135
    },
    {
      "epoch": 15.636963696369637,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.004143104134511876,
      "loss": 0.2314,
      "num_input_tokens_seen": 29996544,
      "step": 142140
    },
    {
      "epoch": 15.637513751375138,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.004142110531868781,
      "loss": 0.2319,
      "num_input_tokens_seen": 29997632,
      "step": 142145
    },
    {
      "epoch": 15.638063806380638,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004141117029297645,
      "loss": 0.2319,
      "num_input_tokens_seen": 29998688,
      "step": 142150
    },
    {
      "epoch": 15.638613861386139,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.004140123626807616,
      "loss": 0.2324,
      "num_input_tokens_seen": 29999712,
      "step": 142155
    },
    {
      "epoch": 15.63916391639164,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.004139130324407858,
      "loss": 0.2288,
      "num_input_tokens_seen": 30000704,
      "step": 142160
    },
    {
      "epoch": 15.63971397139714,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004138137122107522,
      "loss": 0.2303,
      "num_input_tokens_seen": 30001792,
      "step": 142165
    },
    {
      "epoch": 15.64026402640264,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00413714401991577,
      "loss": 0.2314,
      "num_input_tokens_seen": 30002816,
      "step": 142170
    },
    {
      "epoch": 15.640814081408141,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.004136151017841749,
      "loss": 0.2314,
      "num_input_tokens_seen": 30003872,
      "step": 142175
    },
    {
      "epoch": 15.64136413641364,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.004135158115894606,
      "loss": 0.2309,
      "num_input_tokens_seen": 30004960,
      "step": 142180
    },
    {
      "epoch": 15.641914191419142,
      "grad_norm": 0.000713348388671875,
      "learning_rate": 0.0041341653140835035,
      "loss": 0.2303,
      "num_input_tokens_seen": 30006016,
      "step": 142185
    },
    {
      "epoch": 15.642464246424643,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.0041331726124175785,
      "loss": 0.2293,
      "num_input_tokens_seen": 30007072,
      "step": 142190
    },
    {
      "epoch": 15.643014301430142,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.004132180010905989,
      "loss": 0.2324,
      "num_input_tokens_seen": 30008160,
      "step": 142195
    },
    {
      "epoch": 15.643564356435643,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004131187509557885,
      "loss": 0.2304,
      "num_input_tokens_seen": 30009184,
      "step": 142200
    },
    {
      "epoch": 15.644114411441144,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.004130195108382405,
      "loss": 0.2298,
      "num_input_tokens_seen": 30010144,
      "step": 142205
    },
    {
      "epoch": 15.644664466446645,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.004129202807388707,
      "loss": 0.2335,
      "num_input_tokens_seen": 30011264,
      "step": 142210
    },
    {
      "epoch": 15.645214521452145,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.004128210606585926,
      "loss": 0.2309,
      "num_input_tokens_seen": 30012288,
      "step": 142215
    },
    {
      "epoch": 15.645764576457646,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004127218505983209,
      "loss": 0.2308,
      "num_input_tokens_seen": 30013312,
      "step": 142220
    },
    {
      "epoch": 15.646314631463147,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0041262265055897075,
      "loss": 0.2329,
      "num_input_tokens_seen": 30014368,
      "step": 142225
    },
    {
      "epoch": 15.646864686468646,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.004125234605414554,
      "loss": 0.2324,
      "num_input_tokens_seen": 30015392,
      "step": 142230
    },
    {
      "epoch": 15.647414741474147,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.004124242805466899,
      "loss": 0.2309,
      "num_input_tokens_seen": 30016512,
      "step": 142235
    },
    {
      "epoch": 15.647964796479648,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004123251105755878,
      "loss": 0.2314,
      "num_input_tokens_seen": 30017632,
      "step": 142240
    },
    {
      "epoch": 15.648514851485148,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004122259506290633,
      "loss": 0.2309,
      "num_input_tokens_seen": 30018688,
      "step": 142245
    },
    {
      "epoch": 15.649064906490649,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004121268007080299,
      "loss": 0.2314,
      "num_input_tokens_seen": 30019744,
      "step": 142250
    },
    {
      "epoch": 15.64961496149615,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004120276608134019,
      "loss": 0.2319,
      "num_input_tokens_seen": 30020704,
      "step": 142255
    },
    {
      "epoch": 15.65016501650165,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004119285309460931,
      "loss": 0.2303,
      "num_input_tokens_seen": 30021728,
      "step": 142260
    },
    {
      "epoch": 15.65071507150715,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004118294111070166,
      "loss": 0.2309,
      "num_input_tokens_seen": 30022752,
      "step": 142265
    },
    {
      "epoch": 15.651265126512651,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0041173030129708675,
      "loss": 0.2309,
      "num_input_tokens_seen": 30023840,
      "step": 142270
    },
    {
      "epoch": 15.651815181518153,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.004116312015172165,
      "loss": 0.2314,
      "num_input_tokens_seen": 30024864,
      "step": 142275
    },
    {
      "epoch": 15.652365236523652,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.004115321117683182,
      "loss": 0.2335,
      "num_input_tokens_seen": 30025952,
      "step": 142280
    },
    {
      "epoch": 15.652915291529153,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.004114330320513072,
      "loss": 0.2309,
      "num_input_tokens_seen": 30027040,
      "step": 142285
    },
    {
      "epoch": 15.653465346534654,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.004113339623670956,
      "loss": 0.2324,
      "num_input_tokens_seen": 30028128,
      "step": 142290
    },
    {
      "epoch": 15.654015401540153,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.004112349027165959,
      "loss": 0.2308,
      "num_input_tokens_seen": 30029184,
      "step": 142295
    },
    {
      "epoch": 15.654565456545654,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.0041113585310072216,
      "loss": 0.2329,
      "num_input_tokens_seen": 30030208,
      "step": 142300
    },
    {
      "epoch": 15.655115511551156,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0041103681352038625,
      "loss": 0.2314,
      "num_input_tokens_seen": 30031232,
      "step": 142305
    },
    {
      "epoch": 15.655665566556655,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.004109377839765016,
      "loss": 0.2308,
      "num_input_tokens_seen": 30032256,
      "step": 142310
    },
    {
      "epoch": 15.656215621562156,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004108387644699813,
      "loss": 0.2314,
      "num_input_tokens_seen": 30033344,
      "step": 142315
    },
    {
      "epoch": 15.656765676567657,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00410739755001737,
      "loss": 0.2314,
      "num_input_tokens_seen": 30034400,
      "step": 142320
    },
    {
      "epoch": 15.657315731573158,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.004106407555726821,
      "loss": 0.2314,
      "num_input_tokens_seen": 30035520,
      "step": 142325
    },
    {
      "epoch": 15.657865786578657,
      "grad_norm": 0.001953125,
      "learning_rate": 0.004105417661837281,
      "loss": 0.2303,
      "num_input_tokens_seen": 30036608,
      "step": 142330
    },
    {
      "epoch": 15.658415841584159,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.004104427868357879,
      "loss": 0.2288,
      "num_input_tokens_seen": 30037664,
      "step": 142335
    },
    {
      "epoch": 15.65896589658966,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.004103438175297743,
      "loss": 0.2309,
      "num_input_tokens_seen": 30038720,
      "step": 142340
    },
    {
      "epoch": 15.659515951595159,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004102448582665984,
      "loss": 0.2293,
      "num_input_tokens_seen": 30039712,
      "step": 142345
    },
    {
      "epoch": 15.66006600660066,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.004101459090471731,
      "loss": 0.2319,
      "num_input_tokens_seen": 30040736,
      "step": 142350
    },
    {
      "epoch": 15.660616061606161,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004100469698724095,
      "loss": 0.2314,
      "num_input_tokens_seen": 30041792,
      "step": 142355
    },
    {
      "epoch": 15.66116611661166,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.004099480407432205,
      "loss": 0.2345,
      "num_input_tokens_seen": 30042880,
      "step": 142360
    },
    {
      "epoch": 15.661716171617162,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004098491216605168,
      "loss": 0.2298,
      "num_input_tokens_seen": 30043936,
      "step": 142365
    },
    {
      "epoch": 15.662266226622663,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004097502126252108,
      "loss": 0.2309,
      "num_input_tokens_seen": 30045024,
      "step": 142370
    },
    {
      "epoch": 15.662816281628164,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004096513136382142,
      "loss": 0.2303,
      "num_input_tokens_seen": 30046080,
      "step": 142375
    },
    {
      "epoch": 15.663366336633663,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.004095524247004379,
      "loss": 0.2293,
      "num_input_tokens_seen": 30047072,
      "step": 142380
    },
    {
      "epoch": 15.663916391639164,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0040945354581279425,
      "loss": 0.2309,
      "num_input_tokens_seen": 30048160,
      "step": 142385
    },
    {
      "epoch": 15.664466446644665,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.004093546769761937,
      "loss": 0.2319,
      "num_input_tokens_seen": 30049184,
      "step": 142390
    },
    {
      "epoch": 15.665016501650165,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004092558181915469,
      "loss": 0.2314,
      "num_input_tokens_seen": 30050272,
      "step": 142395
    },
    {
      "epoch": 15.665566556655666,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.004091569694597668,
      "loss": 0.2303,
      "num_input_tokens_seen": 30051296,
      "step": 142400
    },
    {
      "epoch": 15.666116611661167,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.004090581307817631,
      "loss": 0.2324,
      "num_input_tokens_seen": 30052320,
      "step": 142405
    },
    {
      "epoch": 15.666666666666666,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.004089593021584476,
      "loss": 0.2314,
      "num_input_tokens_seen": 30053440,
      "step": 142410
    },
    {
      "epoch": 15.667216721672167,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004088604835907304,
      "loss": 0.2303,
      "num_input_tokens_seen": 30054528,
      "step": 142415
    },
    {
      "epoch": 15.667766776677668,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004087616750795224,
      "loss": 0.2314,
      "num_input_tokens_seen": 30055520,
      "step": 142420
    },
    {
      "epoch": 15.668316831683168,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.004086628766257343,
      "loss": 0.2324,
      "num_input_tokens_seen": 30056640,
      "step": 142425
    },
    {
      "epoch": 15.668866886688669,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004085640882302768,
      "loss": 0.2314,
      "num_input_tokens_seen": 30057728,
      "step": 142430
    },
    {
      "epoch": 15.66941694169417,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.004084653098940608,
      "loss": 0.2303,
      "num_input_tokens_seen": 30058720,
      "step": 142435
    },
    {
      "epoch": 15.66996699669967,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.004083665416179963,
      "loss": 0.2324,
      "num_input_tokens_seen": 30059808,
      "step": 142440
    },
    {
      "epoch": 15.67051705170517,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.00408267783402993,
      "loss": 0.2303,
      "num_input_tokens_seen": 30060832,
      "step": 142445
    },
    {
      "epoch": 15.671067106710671,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004081690352499623,
      "loss": 0.2329,
      "num_input_tokens_seen": 30061888,
      "step": 142450
    },
    {
      "epoch": 15.671617161716172,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.004080702971598127,
      "loss": 0.2319,
      "num_input_tokens_seen": 30062944,
      "step": 142455
    },
    {
      "epoch": 15.672167216721672,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.004079715691334562,
      "loss": 0.2303,
      "num_input_tokens_seen": 30063968,
      "step": 142460
    },
    {
      "epoch": 15.672717271727173,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.004078728511718016,
      "loss": 0.2314,
      "num_input_tokens_seen": 30065056,
      "step": 142465
    },
    {
      "epoch": 15.673267326732674,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.004077741432757585,
      "loss": 0.2324,
      "num_input_tokens_seen": 30066112,
      "step": 142470
    },
    {
      "epoch": 15.673817381738173,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004076754454462374,
      "loss": 0.2298,
      "num_input_tokens_seen": 30067168,
      "step": 142475
    },
    {
      "epoch": 15.674367436743674,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.004075767576841473,
      "loss": 0.2324,
      "num_input_tokens_seen": 30068192,
      "step": 142480
    },
    {
      "epoch": 15.674917491749175,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004074780799903978,
      "loss": 0.2329,
      "num_input_tokens_seen": 30069248,
      "step": 142485
    },
    {
      "epoch": 15.675467546754675,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.004073794123658992,
      "loss": 0.2309,
      "num_input_tokens_seen": 30070272,
      "step": 142490
    },
    {
      "epoch": 15.676017601760176,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.0040728075481155975,
      "loss": 0.2309,
      "num_input_tokens_seen": 30071264,
      "step": 142495
    },
    {
      "epoch": 15.676567656765677,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.004071821073282899,
      "loss": 0.2308,
      "num_input_tokens_seen": 30072256,
      "step": 142500
    },
    {
      "epoch": 15.677117711771178,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004070834699169979,
      "loss": 0.233,
      "num_input_tokens_seen": 30073280,
      "step": 142505
    },
    {
      "epoch": 15.677667766776677,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.004069848425785928,
      "loss": 0.2314,
      "num_input_tokens_seen": 30074336,
      "step": 142510
    },
    {
      "epoch": 15.678217821782178,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004068862253139838,
      "loss": 0.2309,
      "num_input_tokens_seen": 30075424,
      "step": 142515
    },
    {
      "epoch": 15.67876787678768,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004067876181240801,
      "loss": 0.2309,
      "num_input_tokens_seen": 30076544,
      "step": 142520
    },
    {
      "epoch": 15.679317931793179,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.004066890210097907,
      "loss": 0.2319,
      "num_input_tokens_seen": 30077696,
      "step": 142525
    },
    {
      "epoch": 15.67986798679868,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004065904339720241,
      "loss": 0.2308,
      "num_input_tokens_seen": 30078720,
      "step": 142530
    },
    {
      "epoch": 15.680418041804181,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.004064918570116882,
      "loss": 0.2314,
      "num_input_tokens_seen": 30079776,
      "step": 142535
    },
    {
      "epoch": 15.68096809680968,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.004063932901296922,
      "loss": 0.2314,
      "num_input_tokens_seen": 30080864,
      "step": 142540
    },
    {
      "epoch": 15.681518151815181,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.004062947333269445,
      "loss": 0.2308,
      "num_input_tokens_seen": 30081952,
      "step": 142545
    },
    {
      "epoch": 15.682068206820682,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004061961866043538,
      "loss": 0.2309,
      "num_input_tokens_seen": 30083008,
      "step": 142550
    },
    {
      "epoch": 15.682618261826182,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.004060976499628281,
      "loss": 0.2314,
      "num_input_tokens_seen": 30084032,
      "step": 142555
    },
    {
      "epoch": 15.683168316831683,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004059991234032749,
      "loss": 0.2293,
      "num_input_tokens_seen": 30085024,
      "step": 142560
    },
    {
      "epoch": 15.683718371837184,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004059006069266034,
      "loss": 0.2314,
      "num_input_tokens_seen": 30086112,
      "step": 142565
    },
    {
      "epoch": 15.684268426842685,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.004058021005337204,
      "loss": 0.2308,
      "num_input_tokens_seen": 30087200,
      "step": 142570
    },
    {
      "epoch": 15.684818481848184,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004057036042255345,
      "loss": 0.2314,
      "num_input_tokens_seen": 30088256,
      "step": 142575
    },
    {
      "epoch": 15.685368536853685,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.004056051180029539,
      "loss": 0.2309,
      "num_input_tokens_seen": 30089376,
      "step": 142580
    },
    {
      "epoch": 15.685918591859187,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.004055066418668853,
      "loss": 0.2314,
      "num_input_tokens_seen": 30090432,
      "step": 142585
    },
    {
      "epoch": 15.686468646864686,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00405408175818237,
      "loss": 0.2309,
      "num_input_tokens_seen": 30091488,
      "step": 142590
    },
    {
      "epoch": 15.687018701870187,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.004053097198579162,
      "loss": 0.2308,
      "num_input_tokens_seen": 30092640,
      "step": 142595
    },
    {
      "epoch": 15.687568756875688,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.004052112739868302,
      "loss": 0.2319,
      "num_input_tokens_seen": 30093632,
      "step": 142600
    },
    {
      "epoch": 15.688118811881187,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004051128382058872,
      "loss": 0.2319,
      "num_input_tokens_seen": 30094752,
      "step": 142605
    },
    {
      "epoch": 15.688668866886688,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004050144125159931,
      "loss": 0.2308,
      "num_input_tokens_seen": 30095776,
      "step": 142610
    },
    {
      "epoch": 15.68921892189219,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.004049159969180564,
      "loss": 0.2303,
      "num_input_tokens_seen": 30096768,
      "step": 142615
    },
    {
      "epoch": 15.689768976897689,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00404817591412983,
      "loss": 0.2308,
      "num_input_tokens_seen": 30097824,
      "step": 142620
    },
    {
      "epoch": 15.69031903190319,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004047191960016806,
      "loss": 0.2319,
      "num_input_tokens_seen": 30098848,
      "step": 142625
    },
    {
      "epoch": 15.690869086908691,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.004046208106850556,
      "loss": 0.2303,
      "num_input_tokens_seen": 30099904,
      "step": 142630
    },
    {
      "epoch": 15.691419141914192,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004045224354640148,
      "loss": 0.2314,
      "num_input_tokens_seen": 30101024,
      "step": 142635
    },
    {
      "epoch": 15.691969196919691,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004044240703394655,
      "loss": 0.2324,
      "num_input_tokens_seen": 30102048,
      "step": 142640
    },
    {
      "epoch": 15.692519251925193,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004043257153123135,
      "loss": 0.2293,
      "num_input_tokens_seen": 30103104,
      "step": 142645
    },
    {
      "epoch": 15.693069306930694,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.004042273703834661,
      "loss": 0.2303,
      "num_input_tokens_seen": 30104224,
      "step": 142650
    },
    {
      "epoch": 15.693619361936193,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0040412903555382875,
      "loss": 0.2314,
      "num_input_tokens_seen": 30105280,
      "step": 142655
    },
    {
      "epoch": 15.694169416941694,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.004040307108243081,
      "loss": 0.2308,
      "num_input_tokens_seen": 30106400,
      "step": 142660
    },
    {
      "epoch": 15.694719471947195,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.004039323961958111,
      "loss": 0.2324,
      "num_input_tokens_seen": 30107392,
      "step": 142665
    },
    {
      "epoch": 15.695269526952695,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.004038340916692431,
      "loss": 0.2309,
      "num_input_tokens_seen": 30108448,
      "step": 142670
    },
    {
      "epoch": 15.695819581958196,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0040373579724551,
      "loss": 0.2298,
      "num_input_tokens_seen": 30109472,
      "step": 142675
    },
    {
      "epoch": 15.696369636963697,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004036375129255184,
      "loss": 0.2288,
      "num_input_tokens_seen": 30110496,
      "step": 142680
    },
    {
      "epoch": 15.696919691969196,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004035392387101732,
      "loss": 0.2314,
      "num_input_tokens_seen": 30111520,
      "step": 142685
    },
    {
      "epoch": 15.697469746974697,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004034409746003808,
      "loss": 0.2324,
      "num_input_tokens_seen": 30112576,
      "step": 142690
    },
    {
      "epoch": 15.698019801980198,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.0040334272059704725,
      "loss": 0.2303,
      "num_input_tokens_seen": 30113568,
      "step": 142695
    },
    {
      "epoch": 15.6985698569857,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004032444767010771,
      "loss": 0.2356,
      "num_input_tokens_seen": 30114560,
      "step": 142700
    },
    {
      "epoch": 15.699119911991199,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004031462429133767,
      "loss": 0.2298,
      "num_input_tokens_seen": 30115520,
      "step": 142705
    },
    {
      "epoch": 15.6996699669967,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.004030480192348508,
      "loss": 0.233,
      "num_input_tokens_seen": 30116576,
      "step": 142710
    },
    {
      "epoch": 15.7002200220022,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.004029498056664049,
      "loss": 0.2314,
      "num_input_tokens_seen": 30117632,
      "step": 142715
    },
    {
      "epoch": 15.7007700770077,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.004028516022089446,
      "loss": 0.2319,
      "num_input_tokens_seen": 30118656,
      "step": 142720
    },
    {
      "epoch": 15.701320132013201,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.004027534088633742,
      "loss": 0.2303,
      "num_input_tokens_seen": 30119648,
      "step": 142725
    },
    {
      "epoch": 15.701870187018702,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.004026552256305995,
      "loss": 0.2313,
      "num_input_tokens_seen": 30120704,
      "step": 142730
    },
    {
      "epoch": 15.702420242024202,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004025570525115247,
      "loss": 0.2314,
      "num_input_tokens_seen": 30121728,
      "step": 142735
    },
    {
      "epoch": 15.702970297029703,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004024588895070554,
      "loss": 0.2309,
      "num_input_tokens_seen": 30122784,
      "step": 142740
    },
    {
      "epoch": 15.703520352035204,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.004023607366180954,
      "loss": 0.2319,
      "num_input_tokens_seen": 30123872,
      "step": 142745
    },
    {
      "epoch": 15.704070407040705,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0040226259384554975,
      "loss": 0.2308,
      "num_input_tokens_seen": 30124864,
      "step": 142750
    },
    {
      "epoch": 15.704620462046204,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.004021644611903234,
      "loss": 0.2293,
      "num_input_tokens_seen": 30125888,
      "step": 142755
    },
    {
      "epoch": 15.705170517051705,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.004020663386533199,
      "loss": 0.2314,
      "num_input_tokens_seen": 30126976,
      "step": 142760
    },
    {
      "epoch": 15.705720572057206,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004019682262354447,
      "loss": 0.2324,
      "num_input_tokens_seen": 30127968,
      "step": 142765
    },
    {
      "epoch": 15.706270627062706,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.004018701239376012,
      "loss": 0.2319,
      "num_input_tokens_seen": 30128992,
      "step": 142770
    },
    {
      "epoch": 15.706820682068207,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.004017720317606937,
      "loss": 0.2314,
      "num_input_tokens_seen": 30130112,
      "step": 142775
    },
    {
      "epoch": 15.707370737073708,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.004016739497056268,
      "loss": 0.2288,
      "num_input_tokens_seen": 30131200,
      "step": 142780
    },
    {
      "epoch": 15.707920792079207,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.004015758777733036,
      "loss": 0.2309,
      "num_input_tokens_seen": 30132256,
      "step": 142785
    },
    {
      "epoch": 15.708470847084708,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0040147781596462885,
      "loss": 0.2314,
      "num_input_tokens_seen": 30133248,
      "step": 142790
    },
    {
      "epoch": 15.70902090209021,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.004013797642805062,
      "loss": 0.2335,
      "num_input_tokens_seen": 30134368,
      "step": 142795
    },
    {
      "epoch": 15.70957095709571,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.004012817227218385,
      "loss": 0.2308,
      "num_input_tokens_seen": 30135424,
      "step": 142800
    },
    {
      "epoch": 15.71012101210121,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.004011836912895299,
      "loss": 0.2324,
      "num_input_tokens_seen": 30136416,
      "step": 142805
    },
    {
      "epoch": 15.710671067106711,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.004010856699844839,
      "loss": 0.2303,
      "num_input_tokens_seen": 30137472,
      "step": 142810
    },
    {
      "epoch": 15.711221122112212,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004009876588076046,
      "loss": 0.2335,
      "num_input_tokens_seen": 30138496,
      "step": 142815
    },
    {
      "epoch": 15.711771177117711,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004008896577597946,
      "loss": 0.2324,
      "num_input_tokens_seen": 30139584,
      "step": 142820
    },
    {
      "epoch": 15.712321232123212,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.004007916668419567,
      "loss": 0.2308,
      "num_input_tokens_seen": 30140704,
      "step": 142825
    },
    {
      "epoch": 15.712871287128714,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.004006936860549946,
      "loss": 0.2314,
      "num_input_tokens_seen": 30141760,
      "step": 142830
    },
    {
      "epoch": 15.713421342134213,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.004005957153998114,
      "loss": 0.2314,
      "num_input_tokens_seen": 30142816,
      "step": 142835
    },
    {
      "epoch": 15.713971397139714,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.004004977548773102,
      "loss": 0.2298,
      "num_input_tokens_seen": 30143808,
      "step": 142840
    },
    {
      "epoch": 15.714521452145215,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.004003998044883938,
      "loss": 0.2319,
      "num_input_tokens_seen": 30144864,
      "step": 142845
    },
    {
      "epoch": 15.715071507150714,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.004003018642339642,
      "loss": 0.2329,
      "num_input_tokens_seen": 30145888,
      "step": 142850
    },
    {
      "epoch": 15.715621562156215,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.004002039341149252,
      "loss": 0.2314,
      "num_input_tokens_seen": 30146944,
      "step": 142855
    },
    {
      "epoch": 15.716171617161717,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.004001060141321783,
      "loss": 0.2329,
      "num_input_tokens_seen": 30147968,
      "step": 142860
    },
    {
      "epoch": 15.716721672167218,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0040000810428662634,
      "loss": 0.2309,
      "num_input_tokens_seen": 30149056,
      "step": 142865
    },
    {
      "epoch": 15.717271727172717,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.003999102045791724,
      "loss": 0.2319,
      "num_input_tokens_seen": 30150112,
      "step": 142870
    },
    {
      "epoch": 15.717821782178218,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.003998123150107179,
      "loss": 0.2309,
      "num_input_tokens_seen": 30151200,
      "step": 142875
    },
    {
      "epoch": 15.718371837183719,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003997144355821657,
      "loss": 0.2309,
      "num_input_tokens_seen": 30152224,
      "step": 142880
    },
    {
      "epoch": 15.718921892189218,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003996165662944172,
      "loss": 0.2319,
      "num_input_tokens_seen": 30153248,
      "step": 142885
    },
    {
      "epoch": 15.71947194719472,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.003995187071483747,
      "loss": 0.2303,
      "num_input_tokens_seen": 30154272,
      "step": 142890
    },
    {
      "epoch": 15.72002200220022,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.003994208581449407,
      "loss": 0.2314,
      "num_input_tokens_seen": 30155296,
      "step": 142895
    },
    {
      "epoch": 15.72057205720572,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.003993230192850161,
      "loss": 0.2303,
      "num_input_tokens_seen": 30156352,
      "step": 142900
    },
    {
      "epoch": 15.721122112211221,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.003992251905695035,
      "loss": 0.2293,
      "num_input_tokens_seen": 30157440,
      "step": 142905
    },
    {
      "epoch": 15.721672167216722,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.003991273719993041,
      "loss": 0.2303,
      "num_input_tokens_seen": 30158528,
      "step": 142910
    },
    {
      "epoch": 15.722222222222221,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.003990295635753191,
      "loss": 0.2303,
      "num_input_tokens_seen": 30159616,
      "step": 142915
    },
    {
      "epoch": 15.722772277227723,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.003989317652984503,
      "loss": 0.2319,
      "num_input_tokens_seen": 30160672,
      "step": 142920
    },
    {
      "epoch": 15.723322332233224,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0039883397716959906,
      "loss": 0.2319,
      "num_input_tokens_seen": 30161728,
      "step": 142925
    },
    {
      "epoch": 15.723872387238725,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.003987361991896672,
      "loss": 0.2309,
      "num_input_tokens_seen": 30162752,
      "step": 142930
    },
    {
      "epoch": 15.724422442244224,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.003986384313595551,
      "loss": 0.2308,
      "num_input_tokens_seen": 30163808,
      "step": 142935
    },
    {
      "epoch": 15.724972497249725,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003985406736801638,
      "loss": 0.2298,
      "num_input_tokens_seen": 30164864,
      "step": 142940
    },
    {
      "epoch": 15.725522552255226,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.003984429261523946,
      "loss": 0.2345,
      "num_input_tokens_seen": 30165920,
      "step": 142945
    },
    {
      "epoch": 15.726072607260726,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.003983451887771483,
      "loss": 0.2309,
      "num_input_tokens_seen": 30166944,
      "step": 142950
    },
    {
      "epoch": 15.726622662266227,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.003982474615553263,
      "loss": 0.2303,
      "num_input_tokens_seen": 30168000,
      "step": 142955
    },
    {
      "epoch": 15.727172717271728,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003981497444878288,
      "loss": 0.2293,
      "num_input_tokens_seen": 30169056,
      "step": 142960
    },
    {
      "epoch": 15.727722772277227,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003980520375755559,
      "loss": 0.2319,
      "num_input_tokens_seen": 30170144,
      "step": 142965
    },
    {
      "epoch": 15.728272827282728,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.003979543408194091,
      "loss": 0.2329,
      "num_input_tokens_seen": 30171168,
      "step": 142970
    },
    {
      "epoch": 15.72882288228823,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0039785665422028774,
      "loss": 0.2324,
      "num_input_tokens_seen": 30172256,
      "step": 142975
    },
    {
      "epoch": 15.729372937293729,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.003977589777790928,
      "loss": 0.234,
      "num_input_tokens_seen": 30173376,
      "step": 142980
    },
    {
      "epoch": 15.72992299229923,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.003976613114967248,
      "loss": 0.2329,
      "num_input_tokens_seen": 30174432,
      "step": 142985
    },
    {
      "epoch": 15.73047304730473,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.003975636553740833,
      "loss": 0.2319,
      "num_input_tokens_seen": 30175456,
      "step": 142990
    },
    {
      "epoch": 15.731023102310232,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.003974660094120688,
      "loss": 0.2308,
      "num_input_tokens_seen": 30176544,
      "step": 142995
    },
    {
      "epoch": 15.731573157315731,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.0039736837361158085,
      "loss": 0.2324,
      "num_input_tokens_seen": 30177600,
      "step": 143000
    },
    {
      "epoch": 15.732123212321232,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.003972707479735193,
      "loss": 0.2309,
      "num_input_tokens_seen": 30178688,
      "step": 143005
    },
    {
      "epoch": 15.732673267326733,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0039717313249878474,
      "loss": 0.2314,
      "num_input_tokens_seen": 30179744,
      "step": 143010
    },
    {
      "epoch": 15.733223322332233,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003970755271882758,
      "loss": 0.2324,
      "num_input_tokens_seen": 30180832,
      "step": 143015
    },
    {
      "epoch": 15.733773377337734,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.003969779320428927,
      "loss": 0.2309,
      "num_input_tokens_seen": 30181888,
      "step": 143020
    },
    {
      "epoch": 15.734323432343235,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.003968803470635343,
      "loss": 0.2308,
      "num_input_tokens_seen": 30182944,
      "step": 143025
    },
    {
      "epoch": 15.734873487348734,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003967827722511011,
      "loss": 0.2314,
      "num_input_tokens_seen": 30184064,
      "step": 143030
    },
    {
      "epoch": 15.735423542354235,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.003966852076064911,
      "loss": 0.2319,
      "num_input_tokens_seen": 30185088,
      "step": 143035
    },
    {
      "epoch": 15.735973597359736,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.003965876531306042,
      "loss": 0.2309,
      "num_input_tokens_seen": 30186176,
      "step": 143040
    },
    {
      "epoch": 15.736523652365236,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.003964901088243399,
      "loss": 0.2319,
      "num_input_tokens_seen": 30187232,
      "step": 143045
    },
    {
      "epoch": 15.737073707370737,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0039639257468859655,
      "loss": 0.2324,
      "num_input_tokens_seen": 30188320,
      "step": 143050
    },
    {
      "epoch": 15.737623762376238,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00396295050724273,
      "loss": 0.2314,
      "num_input_tokens_seen": 30189376,
      "step": 143055
    },
    {
      "epoch": 15.738173817381739,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.003961975369322689,
      "loss": 0.2319,
      "num_input_tokens_seen": 30190432,
      "step": 143060
    },
    {
      "epoch": 15.738723872387238,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.003961000333134813,
      "loss": 0.2308,
      "num_input_tokens_seen": 30191520,
      "step": 143065
    },
    {
      "epoch": 15.73927392739274,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003960025398688111,
      "loss": 0.2293,
      "num_input_tokens_seen": 30192576,
      "step": 143070
    },
    {
      "epoch": 15.73982398239824,
      "grad_norm": 0.000476837158203125,
      "learning_rate": 0.003959050565991556,
      "loss": 0.2324,
      "num_input_tokens_seen": 30193664,
      "step": 143075
    },
    {
      "epoch": 15.74037403740374,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.00395807583505413,
      "loss": 0.2314,
      "num_input_tokens_seen": 30194720,
      "step": 143080
    },
    {
      "epoch": 15.74092409240924,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.003957101205884825,
      "loss": 0.233,
      "num_input_tokens_seen": 30195776,
      "step": 143085
    },
    {
      "epoch": 15.741474147414742,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0039561266784926144,
      "loss": 0.2314,
      "num_input_tokens_seen": 30196864,
      "step": 143090
    },
    {
      "epoch": 15.742024202420241,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.003955152252886486,
      "loss": 0.2329,
      "num_input_tokens_seen": 30197920,
      "step": 143095
    },
    {
      "epoch": 15.742574257425742,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.003954177929075424,
      "loss": 0.2293,
      "num_input_tokens_seen": 30199040,
      "step": 143100
    },
    {
      "epoch": 15.743124312431243,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0039532037070683975,
      "loss": 0.2329,
      "num_input_tokens_seen": 30200032,
      "step": 143105
    },
    {
      "epoch": 15.743674367436743,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.003952229586874398,
      "loss": 0.2314,
      "num_input_tokens_seen": 30201056,
      "step": 143110
    },
    {
      "epoch": 15.744224422442244,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.003951255568502392,
      "loss": 0.2309,
      "num_input_tokens_seen": 30202016,
      "step": 143115
    },
    {
      "epoch": 15.744774477447745,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0039502816519613616,
      "loss": 0.2324,
      "num_input_tokens_seen": 30203072,
      "step": 143120
    },
    {
      "epoch": 15.745324532453246,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.003949307837260287,
      "loss": 0.2308,
      "num_input_tokens_seen": 30204160,
      "step": 143125
    },
    {
      "epoch": 15.745874587458745,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.003948334124408136,
      "loss": 0.2345,
      "num_input_tokens_seen": 30205216,
      "step": 143130
    },
    {
      "epoch": 15.746424642464246,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00394736051341389,
      "loss": 0.2314,
      "num_input_tokens_seen": 30206240,
      "step": 143135
    },
    {
      "epoch": 15.746974697469748,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.003946387004286514,
      "loss": 0.2314,
      "num_input_tokens_seen": 30207232,
      "step": 143140
    },
    {
      "epoch": 15.747524752475247,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.003945413597034989,
      "loss": 0.2308,
      "num_input_tokens_seen": 30208256,
      "step": 143145
    },
    {
      "epoch": 15.748074807480748,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003944440291668279,
      "loss": 0.2309,
      "num_input_tokens_seen": 30209280,
      "step": 143150
    },
    {
      "epoch": 15.748624862486249,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003943467088195357,
      "loss": 0.2303,
      "num_input_tokens_seen": 30210400,
      "step": 143155
    },
    {
      "epoch": 15.749174917491748,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.003942493986625199,
      "loss": 0.2335,
      "num_input_tokens_seen": 30211392,
      "step": 143160
    },
    {
      "epoch": 15.74972497249725,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003941520986966763,
      "loss": 0.2314,
      "num_input_tokens_seen": 30212544,
      "step": 143165
    },
    {
      "epoch": 15.75027502750275,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003940548089229026,
      "loss": 0.2303,
      "num_input_tokens_seen": 30213664,
      "step": 143170
    },
    {
      "epoch": 15.750825082508252,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.003939575293420951,
      "loss": 0.233,
      "num_input_tokens_seen": 30214688,
      "step": 143175
    },
    {
      "epoch": 15.751375137513751,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003938602599551492,
      "loss": 0.2325,
      "num_input_tokens_seen": 30215744,
      "step": 143180
    },
    {
      "epoch": 15.751925192519252,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.003937630007629636,
      "loss": 0.2319,
      "num_input_tokens_seen": 30216800,
      "step": 143185
    },
    {
      "epoch": 15.752475247524753,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003936657517664331,
      "loss": 0.2324,
      "num_input_tokens_seen": 30217888,
      "step": 143190
    },
    {
      "epoch": 15.753025302530252,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003935685129664551,
      "loss": 0.2288,
      "num_input_tokens_seen": 30218912,
      "step": 143195
    },
    {
      "epoch": 15.753575357535754,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.0039347128436392495,
      "loss": 0.2303,
      "num_input_tokens_seen": 30219936,
      "step": 143200
    },
    {
      "epoch": 15.754125412541255,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.003933740659597387,
      "loss": 0.2319,
      "num_input_tokens_seen": 30221024,
      "step": 143205
    },
    {
      "epoch": 15.754675467546754,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.003932768577547927,
      "loss": 0.2298,
      "num_input_tokens_seen": 30222144,
      "step": 143210
    },
    {
      "epoch": 15.755225522552255,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.003931796597499829,
      "loss": 0.2314,
      "num_input_tokens_seen": 30223232,
      "step": 143215
    },
    {
      "epoch": 15.755775577557756,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.003930824719462053,
      "loss": 0.2303,
      "num_input_tokens_seen": 30224256,
      "step": 143220
    },
    {
      "epoch": 15.756325632563257,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.003929852943443554,
      "loss": 0.2303,
      "num_input_tokens_seen": 30225312,
      "step": 143225
    },
    {
      "epoch": 15.756875687568757,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.003928881269453286,
      "loss": 0.2304,
      "num_input_tokens_seen": 30226432,
      "step": 143230
    },
    {
      "epoch": 15.757425742574258,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.003927909697500211,
      "loss": 0.2309,
      "num_input_tokens_seen": 30227488,
      "step": 143235
    },
    {
      "epoch": 15.757975797579759,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.003926938227593274,
      "loss": 0.2298,
      "num_input_tokens_seen": 30228544,
      "step": 143240
    },
    {
      "epoch": 15.758525852585258,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.003925966859741435,
      "loss": 0.2298,
      "num_input_tokens_seen": 30229632,
      "step": 143245
    },
    {
      "epoch": 15.75907590759076,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00392499559395365,
      "loss": 0.2319,
      "num_input_tokens_seen": 30230688,
      "step": 143250
    },
    {
      "epoch": 15.75962596259626,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.003924024430238861,
      "loss": 0.2324,
      "num_input_tokens_seen": 30231808,
      "step": 143255
    },
    {
      "epoch": 15.76017601760176,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.003923053368606029,
      "loss": 0.2308,
      "num_input_tokens_seen": 30232928,
      "step": 143260
    },
    {
      "epoch": 15.76072607260726,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.003922082409064094,
      "loss": 0.2329,
      "num_input_tokens_seen": 30233984,
      "step": 143265
    },
    {
      "epoch": 15.761276127612762,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003921111551622011,
      "loss": 0.2324,
      "num_input_tokens_seen": 30235072,
      "step": 143270
    },
    {
      "epoch": 15.761826182618261,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.003920140796288729,
      "loss": 0.2293,
      "num_input_tokens_seen": 30236192,
      "step": 143275
    },
    {
      "epoch": 15.762376237623762,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.00391917014307319,
      "loss": 0.2298,
      "num_input_tokens_seen": 30237152,
      "step": 143280
    },
    {
      "epoch": 15.762926292629263,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.003918199591984347,
      "loss": 0.2314,
      "num_input_tokens_seen": 30238176,
      "step": 143285
    },
    {
      "epoch": 15.763476347634764,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.003917229143031139,
      "loss": 0.2298,
      "num_input_tokens_seen": 30239200,
      "step": 143290
    },
    {
      "epoch": 15.764026402640264,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003916258796222509,
      "loss": 0.2309,
      "num_input_tokens_seen": 30240256,
      "step": 143295
    },
    {
      "epoch": 15.764576457645765,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.003915288551567403,
      "loss": 0.2303,
      "num_input_tokens_seen": 30241280,
      "step": 143300
    },
    {
      "epoch": 15.765126512651266,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.003914318409074764,
      "loss": 0.2335,
      "num_input_tokens_seen": 30242400,
      "step": 143305
    },
    {
      "epoch": 15.765676567656765,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003913348368753536,
      "loss": 0.2319,
      "num_input_tokens_seen": 30243488,
      "step": 143310
    },
    {
      "epoch": 15.766226622662266,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.003912378430612657,
      "loss": 0.2319,
      "num_input_tokens_seen": 30244512,
      "step": 143315
    },
    {
      "epoch": 15.766776677667767,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.003911408594661061,
      "loss": 0.2319,
      "num_input_tokens_seen": 30245568,
      "step": 143320
    },
    {
      "epoch": 15.767326732673267,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.003910438860907692,
      "loss": 0.2314,
      "num_input_tokens_seen": 30246624,
      "step": 143325
    },
    {
      "epoch": 15.767876787678768,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.003909469229361485,
      "loss": 0.2319,
      "num_input_tokens_seen": 30247680,
      "step": 143330
    },
    {
      "epoch": 15.768426842684269,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.003908499700031384,
      "loss": 0.2319,
      "num_input_tokens_seen": 30248768,
      "step": 143335
    },
    {
      "epoch": 15.768976897689768,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.003907530272926318,
      "loss": 0.2309,
      "num_input_tokens_seen": 30249856,
      "step": 143340
    },
    {
      "epoch": 15.76952695269527,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00390656094805522,
      "loss": 0.2309,
      "num_input_tokens_seen": 30250944,
      "step": 143345
    },
    {
      "epoch": 15.77007700770077,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.00390559172542703,
      "loss": 0.2308,
      "num_input_tokens_seen": 30252064,
      "step": 143350
    },
    {
      "epoch": 15.770627062706271,
      "grad_norm": 0.002593994140625,
      "learning_rate": 0.003904622605050673,
      "loss": 0.233,
      "num_input_tokens_seen": 30253088,
      "step": 143355
    },
    {
      "epoch": 15.77117711771177,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003903653586935086,
      "loss": 0.2324,
      "num_input_tokens_seen": 30254144,
      "step": 143360
    },
    {
      "epoch": 15.771727172717272,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.0039026846710892038,
      "loss": 0.2308,
      "num_input_tokens_seen": 30255264,
      "step": 143365
    },
    {
      "epoch": 15.772277227722773,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0039017158575219467,
      "loss": 0.2298,
      "num_input_tokens_seen": 30256320,
      "step": 143370
    },
    {
      "epoch": 15.772827282728272,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.003900747146242253,
      "loss": 0.2319,
      "num_input_tokens_seen": 30257344,
      "step": 143375
    },
    {
      "epoch": 15.773377337733773,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.003899778537259044,
      "loss": 0.234,
      "num_input_tokens_seen": 30258400,
      "step": 143380
    },
    {
      "epoch": 15.773927392739274,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.0038988100305812493,
      "loss": 0.2309,
      "num_input_tokens_seen": 30259424,
      "step": 143385
    },
    {
      "epoch": 15.774477447744774,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0038978416262178,
      "loss": 0.2298,
      "num_input_tokens_seen": 30260480,
      "step": 143390
    },
    {
      "epoch": 15.775027502750275,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003896873324177612,
      "loss": 0.2324,
      "num_input_tokens_seen": 30261568,
      "step": 143395
    },
    {
      "epoch": 15.775577557755776,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0038959051244696202,
      "loss": 0.2303,
      "num_input_tokens_seen": 30262592,
      "step": 143400
    },
    {
      "epoch": 15.776127612761275,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0038949370271027387,
      "loss": 0.2309,
      "num_input_tokens_seen": 30263616,
      "step": 143405
    },
    {
      "epoch": 15.776677667766776,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0038939690320858975,
      "loss": 0.2319,
      "num_input_tokens_seen": 30264704,
      "step": 143410
    },
    {
      "epoch": 15.777227722772277,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.003893001139428009,
      "loss": 0.2319,
      "num_input_tokens_seen": 30265824,
      "step": 143415
    },
    {
      "epoch": 15.777777777777779,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0038920333491380002,
      "loss": 0.2303,
      "num_input_tokens_seen": 30266880,
      "step": 143420
    },
    {
      "epoch": 15.778327832783278,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.003891065661224795,
      "loss": 0.2298,
      "num_input_tokens_seen": 30267968,
      "step": 143425
    },
    {
      "epoch": 15.778877887788779,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.0038900980756973063,
      "loss": 0.2298,
      "num_input_tokens_seen": 30269024,
      "step": 143430
    },
    {
      "epoch": 15.77942794279428,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.0038891305925644487,
      "loss": 0.234,
      "num_input_tokens_seen": 30270112,
      "step": 143435
    },
    {
      "epoch": 15.77997799779978,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.003888163211835143,
      "loss": 0.2314,
      "num_input_tokens_seen": 30271232,
      "step": 143440
    },
    {
      "epoch": 15.78052805280528,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0038871959335183047,
      "loss": 0.2303,
      "num_input_tokens_seen": 30272320,
      "step": 143445
    },
    {
      "epoch": 15.781078107810782,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.0038862287576228527,
      "loss": 0.2314,
      "num_input_tokens_seen": 30273376,
      "step": 143450
    },
    {
      "epoch": 15.781628162816281,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.003885261684157698,
      "loss": 0.2298,
      "num_input_tokens_seen": 30274496,
      "step": 143455
    },
    {
      "epoch": 15.782178217821782,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.003884294713131749,
      "loss": 0.2298,
      "num_input_tokens_seen": 30275520,
      "step": 143460
    },
    {
      "epoch": 15.782728272827283,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0038833278445539266,
      "loss": 0.2303,
      "num_input_tokens_seen": 30276512,
      "step": 143465
    },
    {
      "epoch": 15.783278327832782,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.003882361078433132,
      "loss": 0.2309,
      "num_input_tokens_seen": 30277504,
      "step": 143470
    },
    {
      "epoch": 15.783828382838283,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.0038813944147782797,
      "loss": 0.2298,
      "num_input_tokens_seen": 30278528,
      "step": 143475
    },
    {
      "epoch": 15.784378437843785,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.003880427853598286,
      "loss": 0.2309,
      "num_input_tokens_seen": 30279584,
      "step": 143480
    },
    {
      "epoch": 15.784928492849286,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003879461394902047,
      "loss": 0.2309,
      "num_input_tokens_seen": 30280640,
      "step": 143485
    },
    {
      "epoch": 15.785478547854785,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.003878495038698481,
      "loss": 0.2335,
      "num_input_tokens_seen": 30281664,
      "step": 143490
    },
    {
      "epoch": 15.786028602860286,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0038775287849964857,
      "loss": 0.2325,
      "num_input_tokens_seen": 30282752,
      "step": 143495
    },
    {
      "epoch": 15.786578657865787,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.00387656263380497,
      "loss": 0.2325,
      "num_input_tokens_seen": 30283776,
      "step": 143500
    },
    {
      "epoch": 15.787128712871286,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.0038755965851328443,
      "loss": 0.2314,
      "num_input_tokens_seen": 30284800,
      "step": 143505
    },
    {
      "epoch": 15.787678767876788,
      "grad_norm": 0.001953125,
      "learning_rate": 0.0038746306389890028,
      "loss": 0.2314,
      "num_input_tokens_seen": 30285888,
      "step": 143510
    },
    {
      "epoch": 15.788228822882289,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003873664795382355,
      "loss": 0.2298,
      "num_input_tokens_seen": 30286944,
      "step": 143515
    },
    {
      "epoch": 15.788778877887788,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.003872699054321795,
      "loss": 0.2303,
      "num_input_tokens_seen": 30287936,
      "step": 143520
    },
    {
      "epoch": 15.789328932893289,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.003871733415816235,
      "loss": 0.2325,
      "num_input_tokens_seen": 30288992,
      "step": 143525
    },
    {
      "epoch": 15.78987898789879,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0038707678798745623,
      "loss": 0.2298,
      "num_input_tokens_seen": 30290048,
      "step": 143530
    },
    {
      "epoch": 15.79042904290429,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0038698024465056834,
      "loss": 0.2298,
      "num_input_tokens_seen": 30291040,
      "step": 143535
    },
    {
      "epoch": 15.79097909790979,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0038688371157184973,
      "loss": 0.2283,
      "num_input_tokens_seen": 30292064,
      "step": 143540
    },
    {
      "epoch": 15.791529152915292,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0038678718875218942,
      "loss": 0.2309,
      "num_input_tokens_seen": 30293152,
      "step": 143545
    },
    {
      "epoch": 15.792079207920793,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0038669067619247796,
      "loss": 0.2309,
      "num_input_tokens_seen": 30294240,
      "step": 143550
    },
    {
      "epoch": 15.792629262926292,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0038659417389360373,
      "loss": 0.2309,
      "num_input_tokens_seen": 30295264,
      "step": 143555
    },
    {
      "epoch": 15.793179317931793,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0038649768185645677,
      "loss": 0.2319,
      "num_input_tokens_seen": 30296256,
      "step": 143560
    },
    {
      "epoch": 15.793729372937294,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003864012000819269,
      "loss": 0.2304,
      "num_input_tokens_seen": 30297312,
      "step": 143565
    },
    {
      "epoch": 15.794279427942794,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.0038630472857090225,
      "loss": 0.2335,
      "num_input_tokens_seen": 30298400,
      "step": 143570
    },
    {
      "epoch": 15.794829482948295,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003862082673242731,
      "loss": 0.2329,
      "num_input_tokens_seen": 30299456,
      "step": 143575
    },
    {
      "epoch": 15.795379537953796,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0038611181634292772,
      "loss": 0.2324,
      "num_input_tokens_seen": 30300480,
      "step": 143580
    },
    {
      "epoch": 15.795929592959295,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0038601537562775495,
      "loss": 0.2329,
      "num_input_tokens_seen": 30301600,
      "step": 143585
    },
    {
      "epoch": 15.796479647964796,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0038591894517964387,
      "loss": 0.2314,
      "num_input_tokens_seen": 30302656,
      "step": 143590
    },
    {
      "epoch": 15.797029702970297,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.0038582252499948324,
      "loss": 0.2304,
      "num_input_tokens_seen": 30303712,
      "step": 143595
    },
    {
      "epoch": 15.797579757975798,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.0038572611508816226,
      "loss": 0.2288,
      "num_input_tokens_seen": 30304800,
      "step": 143600
    },
    {
      "epoch": 15.798129812981298,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00385629715446569,
      "loss": 0.2304,
      "num_input_tokens_seen": 30305824,
      "step": 143605
    },
    {
      "epoch": 15.798679867986799,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.0038553332607559145,
      "loss": 0.2314,
      "num_input_tokens_seen": 30306848,
      "step": 143610
    },
    {
      "epoch": 15.7992299229923,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.003854369469761184,
      "loss": 0.2345,
      "num_input_tokens_seen": 30307936,
      "step": 143615
    },
    {
      "epoch": 15.7997799779978,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0038534057814903874,
      "loss": 0.2288,
      "num_input_tokens_seen": 30308992,
      "step": 143620
    },
    {
      "epoch": 15.8003300330033,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003852442195952397,
      "loss": 0.2303,
      "num_input_tokens_seen": 30310048,
      "step": 143625
    },
    {
      "epoch": 15.800880088008801,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.003851478713156101,
      "loss": 0.2309,
      "num_input_tokens_seen": 30311040,
      "step": 143630
    },
    {
      "epoch": 15.8014301430143,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.003850515333110372,
      "loss": 0.2319,
      "num_input_tokens_seen": 30312096,
      "step": 143635
    },
    {
      "epoch": 15.801980198019802,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0038495520558240967,
      "loss": 0.2314,
      "num_input_tokens_seen": 30313120,
      "step": 143640
    },
    {
      "epoch": 15.802530253025303,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.0038485888813061474,
      "loss": 0.2298,
      "num_input_tokens_seen": 30314240,
      "step": 143645
    },
    {
      "epoch": 15.803080308030804,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.003847625809565401,
      "loss": 0.2319,
      "num_input_tokens_seen": 30315328,
      "step": 143650
    },
    {
      "epoch": 15.803630363036303,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0038466628406107414,
      "loss": 0.2345,
      "num_input_tokens_seen": 30316384,
      "step": 143655
    },
    {
      "epoch": 15.804180418041804,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.003845699974451036,
      "loss": 0.2319,
      "num_input_tokens_seen": 30317472,
      "step": 143660
    },
    {
      "epoch": 15.804730473047305,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0038447372110951643,
      "loss": 0.2324,
      "num_input_tokens_seen": 30318496,
      "step": 143665
    },
    {
      "epoch": 15.805280528052805,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0038437745505519923,
      "loss": 0.2314,
      "num_input_tokens_seen": 30319584,
      "step": 143670
    },
    {
      "epoch": 15.805830583058306,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.003842811992830397,
      "loss": 0.2319,
      "num_input_tokens_seen": 30320672,
      "step": 143675
    },
    {
      "epoch": 15.806380638063807,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003841849537939254,
      "loss": 0.2324,
      "num_input_tokens_seen": 30321760,
      "step": 143680
    },
    {
      "epoch": 15.806930693069306,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0038408871858874264,
      "loss": 0.2324,
      "num_input_tokens_seen": 30322784,
      "step": 143685
    },
    {
      "epoch": 15.807480748074807,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0038399249366837906,
      "loss": 0.2309,
      "num_input_tokens_seen": 30323808,
      "step": 143690
    },
    {
      "epoch": 15.808030803080309,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0038389627903372106,
      "loss": 0.2314,
      "num_input_tokens_seen": 30324864,
      "step": 143695
    },
    {
      "epoch": 15.808580858085808,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0038380007468565517,
      "loss": 0.2319,
      "num_input_tokens_seen": 30325920,
      "step": 143700
    },
    {
      "epoch": 15.809130913091309,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003837038806250685,
      "loss": 0.2324,
      "num_input_tokens_seen": 30327040,
      "step": 143705
    },
    {
      "epoch": 15.80968096809681,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.0038360769685284726,
      "loss": 0.2324,
      "num_input_tokens_seen": 30328096,
      "step": 143710
    },
    {
      "epoch": 15.810231023102311,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.003835115233698787,
      "loss": 0.2308,
      "num_input_tokens_seen": 30329120,
      "step": 143715
    },
    {
      "epoch": 15.81078107810781,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0038341536017704883,
      "loss": 0.2319,
      "num_input_tokens_seen": 30330176,
      "step": 143720
    },
    {
      "epoch": 15.811331133113312,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.003833192072752432,
      "loss": 0.2319,
      "num_input_tokens_seen": 30331168,
      "step": 143725
    },
    {
      "epoch": 15.811881188118813,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.003832230646653487,
      "loss": 0.2324,
      "num_input_tokens_seen": 30332224,
      "step": 143730
    },
    {
      "epoch": 15.812431243124312,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0038312693234825133,
      "loss": 0.2324,
      "num_input_tokens_seen": 30333280,
      "step": 143735
    },
    {
      "epoch": 15.812981298129813,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0038303081032483735,
      "loss": 0.2324,
      "num_input_tokens_seen": 30334336,
      "step": 143740
    },
    {
      "epoch": 15.813531353135314,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003829346985959926,
      "loss": 0.2314,
      "num_input_tokens_seen": 30335392,
      "step": 143745
    },
    {
      "epoch": 15.814081408140813,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0038283859716260232,
      "loss": 0.2319,
      "num_input_tokens_seen": 30336448,
      "step": 143750
    },
    {
      "epoch": 15.814631463146315,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.00382742506025553,
      "loss": 0.2298,
      "num_input_tokens_seen": 30337472,
      "step": 143755
    },
    {
      "epoch": 15.815181518151816,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0038264642518572936,
      "loss": 0.2314,
      "num_input_tokens_seen": 30338560,
      "step": 143760
    },
    {
      "epoch": 15.815731573157315,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.003825503546440175,
      "loss": 0.2324,
      "num_input_tokens_seen": 30339584,
      "step": 143765
    },
    {
      "epoch": 15.816281628162816,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.0038245429440130325,
      "loss": 0.2314,
      "num_input_tokens_seen": 30340640,
      "step": 143770
    },
    {
      "epoch": 15.816831683168317,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.003823582444584711,
      "loss": 0.2314,
      "num_input_tokens_seen": 30341696,
      "step": 143775
    },
    {
      "epoch": 15.817381738173818,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003822622048164072,
      "loss": 0.2319,
      "num_input_tokens_seen": 30342752,
      "step": 143780
    },
    {
      "epoch": 15.817931793179318,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003821661754759958,
      "loss": 0.2324,
      "num_input_tokens_seen": 30343744,
      "step": 143785
    },
    {
      "epoch": 15.818481848184819,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.003820701564381223,
      "loss": 0.2314,
      "num_input_tokens_seen": 30344768,
      "step": 143790
    },
    {
      "epoch": 15.81903190319032,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.003819741477036722,
      "loss": 0.2308,
      "num_input_tokens_seen": 30345856,
      "step": 143795
    },
    {
      "epoch": 15.819581958195819,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0038187814927352934,
      "loss": 0.2298,
      "num_input_tokens_seen": 30346848,
      "step": 143800
    },
    {
      "epoch": 15.82013201320132,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0038178216114857968,
      "loss": 0.2314,
      "num_input_tokens_seen": 30347872,
      "step": 143805
    },
    {
      "epoch": 15.820682068206821,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.003816861833297072,
      "loss": 0.2299,
      "num_input_tokens_seen": 30348992,
      "step": 143810
    },
    {
      "epoch": 15.82123212321232,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00381590215817796,
      "loss": 0.2325,
      "num_input_tokens_seen": 30350048,
      "step": 143815
    },
    {
      "epoch": 15.821782178217822,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.0038149425861373136,
      "loss": 0.2314,
      "num_input_tokens_seen": 30351072,
      "step": 143820
    },
    {
      "epoch": 15.822332233223323,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.0038139831171839726,
      "loss": 0.2293,
      "num_input_tokens_seen": 30352096,
      "step": 143825
    },
    {
      "epoch": 15.822882288228822,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.003813023751326785,
      "loss": 0.2298,
      "num_input_tokens_seen": 30353152,
      "step": 143830
    },
    {
      "epoch": 15.823432343234323,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003812064488574592,
      "loss": 0.2308,
      "num_input_tokens_seen": 30354240,
      "step": 143835
    },
    {
      "epoch": 15.823982398239824,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003811105328936226,
      "loss": 0.2298,
      "num_input_tokens_seen": 30355360,
      "step": 143840
    },
    {
      "epoch": 15.824532453245325,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0038101462724205374,
      "loss": 0.2314,
      "num_input_tokens_seen": 30356416,
      "step": 143845
    },
    {
      "epoch": 15.825082508250825,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0038091873190363537,
      "loss": 0.2314,
      "num_input_tokens_seen": 30357472,
      "step": 143850
    },
    {
      "epoch": 15.825632563256326,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.003808228468792528,
      "loss": 0.2304,
      "num_input_tokens_seen": 30358496,
      "step": 143855
    },
    {
      "epoch": 15.826182618261827,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003807269721697892,
      "loss": 0.2303,
      "num_input_tokens_seen": 30359552,
      "step": 143860
    },
    {
      "epoch": 15.826732673267326,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.003806311077761274,
      "loss": 0.234,
      "num_input_tokens_seen": 30360640,
      "step": 143865
    },
    {
      "epoch": 15.827282728272827,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0038053525369915195,
      "loss": 0.2304,
      "num_input_tokens_seen": 30361728,
      "step": 143870
    },
    {
      "epoch": 15.827832783278328,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0038043940993974564,
      "loss": 0.2314,
      "num_input_tokens_seen": 30362816,
      "step": 143875
    },
    {
      "epoch": 15.828382838283828,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.0038034357649879193,
      "loss": 0.233,
      "num_input_tokens_seen": 30363904,
      "step": 143880
    },
    {
      "epoch": 15.828932893289329,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003802477533771748,
      "loss": 0.2314,
      "num_input_tokens_seen": 30364960,
      "step": 143885
    },
    {
      "epoch": 15.82948294829483,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.003801519405757763,
      "loss": 0.2288,
      "num_input_tokens_seen": 30366016,
      "step": 143890
    },
    {
      "epoch": 15.83003300330033,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.003800561380954805,
      "loss": 0.2309,
      "num_input_tokens_seen": 30367040,
      "step": 143895
    },
    {
      "epoch": 15.83058305830583,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.0037996034593716925,
      "loss": 0.2329,
      "num_input_tokens_seen": 30368064,
      "step": 143900
    },
    {
      "epoch": 15.831133113311331,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.0037986456410172625,
      "loss": 0.2324,
      "num_input_tokens_seen": 30369088,
      "step": 143905
    },
    {
      "epoch": 15.831683168316832,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0037976879259003444,
      "loss": 0.2298,
      "num_input_tokens_seen": 30370176,
      "step": 143910
    },
    {
      "epoch": 15.832233223322332,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0037967303140297557,
      "loss": 0.2325,
      "num_input_tokens_seen": 30371232,
      "step": 143915
    },
    {
      "epoch": 15.832783278327833,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.0037957728054143335,
      "loss": 0.2329,
      "num_input_tokens_seen": 30372288,
      "step": 143920
    },
    {
      "epoch": 15.833333333333334,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0037948154000628934,
      "loss": 0.2329,
      "num_input_tokens_seen": 30373312,
      "step": 143925
    },
    {
      "epoch": 15.833883388338833,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.0037938580979842654,
      "loss": 0.2314,
      "num_input_tokens_seen": 30374272,
      "step": 143930
    },
    {
      "epoch": 15.834433443344334,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0037929008991872673,
      "loss": 0.2314,
      "num_input_tokens_seen": 30375392,
      "step": 143935
    },
    {
      "epoch": 15.834983498349835,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003791943803680724,
      "loss": 0.2319,
      "num_input_tokens_seen": 30376448,
      "step": 143940
    },
    {
      "epoch": 15.835533553355335,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0037909868114734593,
      "loss": 0.2309,
      "num_input_tokens_seen": 30377504,
      "step": 143945
    },
    {
      "epoch": 15.836083608360836,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0037900299225742872,
      "loss": 0.2314,
      "num_input_tokens_seen": 30378528,
      "step": 143950
    },
    {
      "epoch": 15.836633663366337,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.003789073136992035,
      "loss": 0.2304,
      "num_input_tokens_seen": 30379616,
      "step": 143955
    },
    {
      "epoch": 15.837183718371836,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.0037881164547355155,
      "loss": 0.233,
      "num_input_tokens_seen": 30380640,
      "step": 143960
    },
    {
      "epoch": 15.837733773377337,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.003787159875813539,
      "loss": 0.2304,
      "num_input_tokens_seen": 30381664,
      "step": 143965
    },
    {
      "epoch": 15.838283828382838,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0037862034002349373,
      "loss": 0.2319,
      "num_input_tokens_seen": 30382848,
      "step": 143970
    },
    {
      "epoch": 15.83883388338834,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003785247028008513,
      "loss": 0.2319,
      "num_input_tokens_seen": 30383936,
      "step": 143975
    },
    {
      "epoch": 15.839383938393839,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.00378429075914309,
      "loss": 0.2309,
      "num_input_tokens_seen": 30384992,
      "step": 143980
    },
    {
      "epoch": 15.83993399339934,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.003783334593647478,
      "loss": 0.234,
      "num_input_tokens_seen": 30386112,
      "step": 143985
    },
    {
      "epoch": 15.840484048404841,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0037823785315304853,
      "loss": 0.2319,
      "num_input_tokens_seen": 30387200,
      "step": 143990
    },
    {
      "epoch": 15.84103410341034,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.003781422572800926,
      "loss": 0.2314,
      "num_input_tokens_seen": 30388256,
      "step": 143995
    },
    {
      "epoch": 15.841584158415841,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0037804667174676115,
      "loss": 0.2319,
      "num_input_tokens_seen": 30389280,
      "step": 144000
    },
    {
      "epoch": 15.842134213421343,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.003779510965539355,
      "loss": 0.2319,
      "num_input_tokens_seen": 30390272,
      "step": 144005
    },
    {
      "epoch": 15.842684268426842,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0037785553170249626,
      "loss": 0.2304,
      "num_input_tokens_seen": 30391328,
      "step": 144010
    },
    {
      "epoch": 15.843234323432343,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.003777599771933237,
      "loss": 0.2319,
      "num_input_tokens_seen": 30392384,
      "step": 144015
    },
    {
      "epoch": 15.843784378437844,
      "grad_norm": 0.00055694580078125,
      "learning_rate": 0.0037766443302729934,
      "loss": 0.2329,
      "num_input_tokens_seen": 30393408,
      "step": 144020
    },
    {
      "epoch": 15.844334433443345,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.0037756889920530293,
      "loss": 0.2298,
      "num_input_tokens_seen": 30394464,
      "step": 144025
    },
    {
      "epoch": 15.844884488448844,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0037747337572821547,
      "loss": 0.2308,
      "num_input_tokens_seen": 30395520,
      "step": 144030
    },
    {
      "epoch": 15.845434543454346,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0037737786259691756,
      "loss": 0.233,
      "num_input_tokens_seen": 30396512,
      "step": 144035
    },
    {
      "epoch": 15.845984598459847,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.0037728235981228883,
      "loss": 0.2298,
      "num_input_tokens_seen": 30397536,
      "step": 144040
    },
    {
      "epoch": 15.846534653465346,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0037718686737521043,
      "loss": 0.2319,
      "num_input_tokens_seen": 30398528,
      "step": 144045
    },
    {
      "epoch": 15.847084708470847,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.003770913852865613,
      "loss": 0.2314,
      "num_input_tokens_seen": 30399552,
      "step": 144050
    },
    {
      "epoch": 15.847634763476348,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.003769959135472221,
      "loss": 0.2314,
      "num_input_tokens_seen": 30400608,
      "step": 144055
    },
    {
      "epoch": 15.848184818481847,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.003769004521580731,
      "loss": 0.2319,
      "num_input_tokens_seen": 30401664,
      "step": 144060
    },
    {
      "epoch": 15.848734873487349,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0037680500111999327,
      "loss": 0.2324,
      "num_input_tokens_seen": 30402752,
      "step": 144065
    },
    {
      "epoch": 15.84928492849285,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.003767095604338632,
      "loss": 0.2298,
      "num_input_tokens_seen": 30403808,
      "step": 144070
    },
    {
      "epoch": 15.84983498349835,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.003766141301005621,
      "loss": 0.2309,
      "num_input_tokens_seen": 30404832,
      "step": 144075
    },
    {
      "epoch": 15.85038503850385,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.003765187101209692,
      "loss": 0.2303,
      "num_input_tokens_seen": 30405824,
      "step": 144080
    },
    {
      "epoch": 15.850935093509351,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0037642330049596417,
      "loss": 0.2314,
      "num_input_tokens_seen": 30406816,
      "step": 144085
    },
    {
      "epoch": 15.851485148514852,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0037632790122642643,
      "loss": 0.2298,
      "num_input_tokens_seen": 30407936,
      "step": 144090
    },
    {
      "epoch": 15.852035203520352,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.0037623251231323567,
      "loss": 0.2319,
      "num_input_tokens_seen": 30408960,
      "step": 144095
    },
    {
      "epoch": 15.852585258525853,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.0037613713375727045,
      "loss": 0.2319,
      "num_input_tokens_seen": 30409952,
      "step": 144100
    },
    {
      "epoch": 15.853135313531354,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.003760417655594098,
      "loss": 0.2299,
      "num_input_tokens_seen": 30410944,
      "step": 144105
    },
    {
      "epoch": 15.853685368536853,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.003759464077205328,
      "loss": 0.2319,
      "num_input_tokens_seen": 30412064,
      "step": 144110
    },
    {
      "epoch": 15.854235423542354,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003758510602415183,
      "loss": 0.2304,
      "num_input_tokens_seen": 30413120,
      "step": 144115
    },
    {
      "epoch": 15.854785478547855,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.003757557231232457,
      "loss": 0.2324,
      "num_input_tokens_seen": 30414240,
      "step": 144120
    },
    {
      "epoch": 15.855335533553355,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.0037566039636659303,
      "loss": 0.2303,
      "num_input_tokens_seen": 30415264,
      "step": 144125
    },
    {
      "epoch": 15.855885588558856,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.003755650799724387,
      "loss": 0.2329,
      "num_input_tokens_seen": 30416256,
      "step": 144130
    },
    {
      "epoch": 15.856435643564357,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0037546977394166186,
      "loss": 0.2314,
      "num_input_tokens_seen": 30417248,
      "step": 144135
    },
    {
      "epoch": 15.856985698569858,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0037537447827513996,
      "loss": 0.233,
      "num_input_tokens_seen": 30418304,
      "step": 144140
    },
    {
      "epoch": 15.857535753575357,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.0037527919297375184,
      "loss": 0.2293,
      "num_input_tokens_seen": 30419328,
      "step": 144145
    },
    {
      "epoch": 15.858085808580858,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0037518391803837625,
      "loss": 0.2303,
      "num_input_tokens_seen": 30420416,
      "step": 144150
    },
    {
      "epoch": 15.85863586358636,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0037508865346989034,
      "loss": 0.2314,
      "num_input_tokens_seen": 30421504,
      "step": 144155
    },
    {
      "epoch": 15.859185918591859,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003749933992691729,
      "loss": 0.2303,
      "num_input_tokens_seen": 30422528,
      "step": 144160
    },
    {
      "epoch": 15.85973597359736,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.003748981554371009,
      "loss": 0.2314,
      "num_input_tokens_seen": 30423552,
      "step": 144165
    },
    {
      "epoch": 15.86028602860286,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00374802921974553,
      "loss": 0.2309,
      "num_input_tokens_seen": 30424576,
      "step": 144170
    },
    {
      "epoch": 15.86083608360836,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0037470769888240693,
      "loss": 0.2303,
      "num_input_tokens_seen": 30425664,
      "step": 144175
    },
    {
      "epoch": 15.861386138613861,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0037461248616153958,
      "loss": 0.2314,
      "num_input_tokens_seen": 30426752,
      "step": 144180
    },
    {
      "epoch": 15.861936193619362,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.003745172838128293,
      "loss": 0.2303,
      "num_input_tokens_seen": 30427808,
      "step": 144185
    },
    {
      "epoch": 15.862486248624862,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0037442209183715285,
      "loss": 0.2314,
      "num_input_tokens_seen": 30428864,
      "step": 144190
    },
    {
      "epoch": 15.863036303630363,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0037432691023538828,
      "loss": 0.2319,
      "num_input_tokens_seen": 30429888,
      "step": 144195
    },
    {
      "epoch": 15.863586358635864,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0037423173900841206,
      "loss": 0.2314,
      "num_input_tokens_seen": 30431008,
      "step": 144200
    },
    {
      "epoch": 15.864136413641365,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003741365781571015,
      "loss": 0.2319,
      "num_input_tokens_seen": 30432128,
      "step": 144205
    },
    {
      "epoch": 15.864686468646864,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.003740414276823344,
      "loss": 0.2314,
      "num_input_tokens_seen": 30433184,
      "step": 144210
    },
    {
      "epoch": 15.865236523652365,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003739462875849872,
      "loss": 0.2329,
      "num_input_tokens_seen": 30434272,
      "step": 144215
    },
    {
      "epoch": 15.865786578657866,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.0037385115786593643,
      "loss": 0.2308,
      "num_input_tokens_seen": 30435392,
      "step": 144220
    },
    {
      "epoch": 15.866336633663366,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00373756038526059,
      "loss": 0.2298,
      "num_input_tokens_seen": 30436448,
      "step": 144225
    },
    {
      "epoch": 15.866886688668867,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.003736609295662317,
      "loss": 0.2335,
      "num_input_tokens_seen": 30437472,
      "step": 144230
    },
    {
      "epoch": 15.867436743674368,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003735658309873317,
      "loss": 0.2298,
      "num_input_tokens_seen": 30438560,
      "step": 144235
    },
    {
      "epoch": 15.867986798679867,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0037347074279023475,
      "loss": 0.2309,
      "num_input_tokens_seen": 30439584,
      "step": 144240
    },
    {
      "epoch": 15.868536853685368,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.003733756649758171,
      "loss": 0.2314,
      "num_input_tokens_seen": 30440672,
      "step": 144245
    },
    {
      "epoch": 15.86908690869087,
      "grad_norm": 0.000492095947265625,
      "learning_rate": 0.003732805975449557,
      "loss": 0.2324,
      "num_input_tokens_seen": 30441664,
      "step": 144250
    },
    {
      "epoch": 15.869636963696369,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0037318554049852602,
      "loss": 0.2303,
      "num_input_tokens_seen": 30442720,
      "step": 144255
    },
    {
      "epoch": 15.87018701870187,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.003730904938374046,
      "loss": 0.2308,
      "num_input_tokens_seen": 30443744,
      "step": 144260
    },
    {
      "epoch": 15.870737073707371,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.0037299545756246757,
      "loss": 0.2314,
      "num_input_tokens_seen": 30444864,
      "step": 144265
    },
    {
      "epoch": 15.871287128712872,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.003729004316745903,
      "loss": 0.2298,
      "num_input_tokens_seen": 30445952,
      "step": 144270
    },
    {
      "epoch": 15.871837183718371,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0037280541617464935,
      "loss": 0.2314,
      "num_input_tokens_seen": 30447072,
      "step": 144275
    },
    {
      "epoch": 15.872387238723872,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.003727104110635195,
      "loss": 0.233,
      "num_input_tokens_seen": 30448096,
      "step": 144280
    },
    {
      "epoch": 15.872937293729374,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.00372615416342077,
      "loss": 0.2329,
      "num_input_tokens_seen": 30449120,
      "step": 144285
    },
    {
      "epoch": 15.873487348734873,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0037252043201119753,
      "loss": 0.2304,
      "num_input_tokens_seen": 30450208,
      "step": 144290
    },
    {
      "epoch": 15.874037403740374,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0037242545807175597,
      "loss": 0.2309,
      "num_input_tokens_seen": 30451296,
      "step": 144295
    },
    {
      "epoch": 15.874587458745875,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0037233049452462825,
      "loss": 0.2314,
      "num_input_tokens_seen": 30452352,
      "step": 144300
    },
    {
      "epoch": 15.875137513751374,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.0037223554137068866,
      "loss": 0.2319,
      "num_input_tokens_seen": 30453408,
      "step": 144305
    },
    {
      "epoch": 15.875687568756875,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.003721405986108135,
      "loss": 0.2319,
      "num_input_tokens_seen": 30454368,
      "step": 144310
    },
    {
      "epoch": 15.876237623762377,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0037204566624587676,
      "loss": 0.2309,
      "num_input_tokens_seen": 30455456,
      "step": 144315
    },
    {
      "epoch": 15.876787678767876,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0037195074427675387,
      "loss": 0.2335,
      "num_input_tokens_seen": 30456480,
      "step": 144320
    },
    {
      "epoch": 15.877337733773377,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.003718558327043202,
      "loss": 0.2324,
      "num_input_tokens_seen": 30457536,
      "step": 144325
    },
    {
      "epoch": 15.877887788778878,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0037176093152944947,
      "loss": 0.2298,
      "num_input_tokens_seen": 30458624,
      "step": 144330
    },
    {
      "epoch": 15.87843784378438,
      "grad_norm": 0.00067901611328125,
      "learning_rate": 0.0037166604075301738,
      "loss": 0.2324,
      "num_input_tokens_seen": 30459680,
      "step": 144335
    },
    {
      "epoch": 15.878987898789878,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.003715711603758975,
      "loss": 0.2319,
      "num_input_tokens_seen": 30460704,
      "step": 144340
    },
    {
      "epoch": 15.87953795379538,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0037147629039896474,
      "loss": 0.2314,
      "num_input_tokens_seen": 30461728,
      "step": 144345
    },
    {
      "epoch": 15.88008800880088,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.003713814308230939,
      "loss": 0.2319,
      "num_input_tokens_seen": 30462816,
      "step": 144350
    },
    {
      "epoch": 15.88063806380638,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0037128658164915845,
      "loss": 0.2314,
      "num_input_tokens_seen": 30463936,
      "step": 144355
    },
    {
      "epoch": 15.881188118811881,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.003711917428780334,
      "loss": 0.2303,
      "num_input_tokens_seen": 30465024,
      "step": 144360
    },
    {
      "epoch": 15.881738173817382,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0037109691451059227,
      "loss": 0.2308,
      "num_input_tokens_seen": 30466176,
      "step": 144365
    },
    {
      "epoch": 15.882288228822881,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.0037100209654770885,
      "loss": 0.2314,
      "num_input_tokens_seen": 30467296,
      "step": 144370
    },
    {
      "epoch": 15.882838283828383,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.0037090728899025747,
      "loss": 0.2288,
      "num_input_tokens_seen": 30468320,
      "step": 144375
    },
    {
      "epoch": 15.883388338833884,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.003708124918391117,
      "loss": 0.2324,
      "num_input_tokens_seen": 30469408,
      "step": 144380
    },
    {
      "epoch": 15.883938393839383,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.0037071770509514578,
      "loss": 0.2319,
      "num_input_tokens_seen": 30470464,
      "step": 144385
    },
    {
      "epoch": 15.884488448844884,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003706229287592329,
      "loss": 0.2314,
      "num_input_tokens_seen": 30471488,
      "step": 144390
    },
    {
      "epoch": 15.885038503850385,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.003705281628322461,
      "loss": 0.2293,
      "num_input_tokens_seen": 30472512,
      "step": 144395
    },
    {
      "epoch": 15.885588558855886,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003704334073150593,
      "loss": 0.2335,
      "num_input_tokens_seen": 30473504,
      "step": 144400
    },
    {
      "epoch": 15.886138613861386,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0037033866220854616,
      "loss": 0.2303,
      "num_input_tokens_seen": 30474624,
      "step": 144405
    },
    {
      "epoch": 15.886688668866887,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.0037024392751357914,
      "loss": 0.2314,
      "num_input_tokens_seen": 30475648,
      "step": 144410
    },
    {
      "epoch": 15.887238723872388,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.003701492032310322,
      "loss": 0.2308,
      "num_input_tokens_seen": 30476672,
      "step": 144415
    },
    {
      "epoch": 15.887788778877887,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.0037005448936177736,
      "loss": 0.2309,
      "num_input_tokens_seen": 30477728,
      "step": 144420
    },
    {
      "epoch": 15.888338833883388,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.003699597859066885,
      "loss": 0.2319,
      "num_input_tokens_seen": 30478848,
      "step": 144425
    },
    {
      "epoch": 15.88888888888889,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003698650928666377,
      "loss": 0.2319,
      "num_input_tokens_seen": 30479936,
      "step": 144430
    },
    {
      "epoch": 15.88943894389439,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00369770410242498,
      "loss": 0.2309,
      "num_input_tokens_seen": 30480960,
      "step": 144435
    },
    {
      "epoch": 15.88998899889989,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.003696757380351426,
      "loss": 0.2319,
      "num_input_tokens_seen": 30482016,
      "step": 144440
    },
    {
      "epoch": 15.89053905390539,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0036958107624544306,
      "loss": 0.2319,
      "num_input_tokens_seen": 30483072,
      "step": 144445
    },
    {
      "epoch": 15.891089108910892,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0036948642487427273,
      "loss": 0.2308,
      "num_input_tokens_seen": 30484128,
      "step": 144450
    },
    {
      "epoch": 15.891639163916391,
      "grad_norm": 0.00078582763671875,
      "learning_rate": 0.0036939178392250307,
      "loss": 0.2314,
      "num_input_tokens_seen": 30485152,
      "step": 144455
    },
    {
      "epoch": 15.892189218921892,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0036929715339100674,
      "loss": 0.2309,
      "num_input_tokens_seen": 30486144,
      "step": 144460
    },
    {
      "epoch": 15.892739273927393,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003692025332806567,
      "loss": 0.2314,
      "num_input_tokens_seen": 30487200,
      "step": 144465
    },
    {
      "epoch": 15.893289328932893,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0036910792359232374,
      "loss": 0.2329,
      "num_input_tokens_seen": 30488224,
      "step": 144470
    },
    {
      "epoch": 15.893839383938394,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.0036901332432688065,
      "loss": 0.2314,
      "num_input_tokens_seen": 30489312,
      "step": 144475
    },
    {
      "epoch": 15.894389438943895,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.0036891873548519924,
      "loss": 0.2303,
      "num_input_tokens_seen": 30490336,
      "step": 144480
    },
    {
      "epoch": 15.894939493949394,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003688241570681507,
      "loss": 0.2324,
      "num_input_tokens_seen": 30491456,
      "step": 144485
    },
    {
      "epoch": 15.895489548954895,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.003687295890766069,
      "loss": 0.2298,
      "num_input_tokens_seen": 30492416,
      "step": 144490
    },
    {
      "epoch": 15.896039603960396,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0036863503151143987,
      "loss": 0.2314,
      "num_input_tokens_seen": 30493472,
      "step": 144495
    },
    {
      "epoch": 15.896589658965897,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0036854048437352125,
      "loss": 0.2329,
      "num_input_tokens_seen": 30494528,
      "step": 144500
    },
    {
      "epoch": 15.897139713971397,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00368445947663722,
      "loss": 0.2314,
      "num_input_tokens_seen": 30495584,
      "step": 144505
    },
    {
      "epoch": 15.897689768976898,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0036835142138291308,
      "loss": 0.2314,
      "num_input_tokens_seen": 30496704,
      "step": 144510
    },
    {
      "epoch": 15.898239823982399,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00368256905531966,
      "loss": 0.2314,
      "num_input_tokens_seen": 30497664,
      "step": 144515
    },
    {
      "epoch": 15.898789878987898,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0036816240011175215,
      "loss": 0.2319,
      "num_input_tokens_seen": 30498752,
      "step": 144520
    },
    {
      "epoch": 15.8993399339934,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.003680679051231426,
      "loss": 0.2314,
      "num_input_tokens_seen": 30499872,
      "step": 144525
    },
    {
      "epoch": 15.8998899889989,
      "grad_norm": 0.000797271728515625,
      "learning_rate": 0.0036797342056700804,
      "loss": 0.2329,
      "num_input_tokens_seen": 30500896,
      "step": 144530
    },
    {
      "epoch": 15.9004400440044,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0036787894644421887,
      "loss": 0.2324,
      "num_input_tokens_seen": 30501952,
      "step": 144535
    },
    {
      "epoch": 15.900990099009901,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0036778448275564665,
      "loss": 0.2314,
      "num_input_tokens_seen": 30503072,
      "step": 144540
    },
    {
      "epoch": 15.901540154015402,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003676900295021611,
      "loss": 0.2324,
      "num_input_tokens_seen": 30504192,
      "step": 144545
    },
    {
      "epoch": 15.902090209020901,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0036759558668463333,
      "loss": 0.2314,
      "num_input_tokens_seen": 30505312,
      "step": 144550
    },
    {
      "epoch": 15.902640264026402,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0036750115430393394,
      "loss": 0.233,
      "num_input_tokens_seen": 30506336,
      "step": 144555
    },
    {
      "epoch": 15.903190319031903,
      "grad_norm": 0.00077056884765625,
      "learning_rate": 0.0036740673236093256,
      "loss": 0.2329,
      "num_input_tokens_seen": 30507456,
      "step": 144560
    },
    {
      "epoch": 15.903740374037405,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0036731232085650037,
      "loss": 0.2324,
      "num_input_tokens_seen": 30508512,
      "step": 144565
    },
    {
      "epoch": 15.904290429042904,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0036721791979150652,
      "loss": 0.2303,
      "num_input_tokens_seen": 30509600,
      "step": 144570
    },
    {
      "epoch": 15.904840484048405,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.003671235291668216,
      "loss": 0.2335,
      "num_input_tokens_seen": 30510592,
      "step": 144575
    },
    {
      "epoch": 15.905390539053906,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0036702914898331593,
      "loss": 0.2314,
      "num_input_tokens_seen": 30511680,
      "step": 144580
    },
    {
      "epoch": 15.905940594059405,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.003669347792418585,
      "loss": 0.2324,
      "num_input_tokens_seen": 30512704,
      "step": 144585
    },
    {
      "epoch": 15.906490649064907,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0036684041994331992,
      "loss": 0.2308,
      "num_input_tokens_seen": 30513728,
      "step": 144590
    },
    {
      "epoch": 15.907040704070408,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0036674607108856936,
      "loss": 0.2319,
      "num_input_tokens_seen": 30514816,
      "step": 144595
    },
    {
      "epoch": 15.907590759075907,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.003666517326784761,
      "loss": 0.2314,
      "num_input_tokens_seen": 30515904,
      "step": 144600
    },
    {
      "epoch": 15.908140814081408,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0036655740471391012,
      "loss": 0.2314,
      "num_input_tokens_seen": 30516960,
      "step": 144605
    },
    {
      "epoch": 15.908690869086909,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0036646308719574055,
      "loss": 0.2324,
      "num_input_tokens_seen": 30518016,
      "step": 144610
    },
    {
      "epoch": 15.909240924092408,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.003663687801248372,
      "loss": 0.2319,
      "num_input_tokens_seen": 30519104,
      "step": 144615
    },
    {
      "epoch": 15.90979097909791,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.0036627448350206888,
      "loss": 0.2309,
      "num_input_tokens_seen": 30520192,
      "step": 144620
    },
    {
      "epoch": 15.91034103410341,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.003661801973283042,
      "loss": 0.2308,
      "num_input_tokens_seen": 30521248,
      "step": 144625
    },
    {
      "epoch": 15.910891089108912,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00366085921604413,
      "loss": 0.2314,
      "num_input_tokens_seen": 30522368,
      "step": 144630
    },
    {
      "epoch": 15.911441144114411,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.003659916563312629,
      "loss": 0.2324,
      "num_input_tokens_seen": 30523456,
      "step": 144635
    },
    {
      "epoch": 15.911991199119912,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.0036589740150972435,
      "loss": 0.2319,
      "num_input_tokens_seen": 30524576,
      "step": 144640
    },
    {
      "epoch": 15.912541254125413,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0036580315714066524,
      "loss": 0.2309,
      "num_input_tokens_seen": 30525632,
      "step": 144645
    },
    {
      "epoch": 15.913091309130913,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.003657089232249538,
      "loss": 0.2319,
      "num_input_tokens_seen": 30526688,
      "step": 144650
    },
    {
      "epoch": 15.913641364136414,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.003656146997634595,
      "loss": 0.2314,
      "num_input_tokens_seen": 30527744,
      "step": 144655
    },
    {
      "epoch": 15.914191419141915,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.003655204867570495,
      "loss": 0.2314,
      "num_input_tokens_seen": 30528800,
      "step": 144660
    },
    {
      "epoch": 15.914741474147414,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0036542628420659293,
      "loss": 0.2303,
      "num_input_tokens_seen": 30529856,
      "step": 144665
    },
    {
      "epoch": 15.915291529152915,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.003653320921129583,
      "loss": 0.2335,
      "num_input_tokens_seen": 30530848,
      "step": 144670
    },
    {
      "epoch": 15.915841584158416,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0036523791047701304,
      "loss": 0.2309,
      "num_input_tokens_seen": 30531936,
      "step": 144675
    },
    {
      "epoch": 15.916391639163916,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.003651437392996256,
      "loss": 0.2319,
      "num_input_tokens_seen": 30533024,
      "step": 144680
    },
    {
      "epoch": 15.916941694169417,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0036504957858166354,
      "loss": 0.2319,
      "num_input_tokens_seen": 30534048,
      "step": 144685
    },
    {
      "epoch": 15.917491749174918,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.0036495542832399525,
      "loss": 0.2303,
      "num_input_tokens_seen": 30535104,
      "step": 144690
    },
    {
      "epoch": 15.918041804180419,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003648612885274877,
      "loss": 0.2324,
      "num_input_tokens_seen": 30536128,
      "step": 144695
    },
    {
      "epoch": 15.918591859185918,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0036476715919300895,
      "loss": 0.2298,
      "num_input_tokens_seen": 30537216,
      "step": 144700
    },
    {
      "epoch": 15.91914191419142,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0036467304032142705,
      "loss": 0.2303,
      "num_input_tokens_seen": 30538304,
      "step": 144705
    },
    {
      "epoch": 15.91969196919692,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0036457893191360834,
      "loss": 0.2303,
      "num_input_tokens_seen": 30539392,
      "step": 144710
    },
    {
      "epoch": 15.92024202420242,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0036448483397042144,
      "loss": 0.2303,
      "num_input_tokens_seen": 30540480,
      "step": 144715
    },
    {
      "epoch": 15.92079207920792,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0036439074649273237,
      "loss": 0.2303,
      "num_input_tokens_seen": 30541536,
      "step": 144720
    },
    {
      "epoch": 15.921342134213422,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.003642966694814089,
      "loss": 0.2293,
      "num_input_tokens_seen": 30542592,
      "step": 144725
    },
    {
      "epoch": 15.921892189218921,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0036420260293731853,
      "loss": 0.2293,
      "num_input_tokens_seen": 30543616,
      "step": 144730
    },
    {
      "epoch": 15.922442244224422,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.0036410854686132737,
      "loss": 0.2304,
      "num_input_tokens_seen": 30544704,
      "step": 144735
    },
    {
      "epoch": 15.922992299229923,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0036401450125430304,
      "loss": 0.2314,
      "num_input_tokens_seen": 30545792,
      "step": 144740
    },
    {
      "epoch": 15.923542354235423,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003639204661171121,
      "loss": 0.2324,
      "num_input_tokens_seen": 30546880,
      "step": 144745
    },
    {
      "epoch": 15.924092409240924,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.0036382644145062006,
      "loss": 0.2283,
      "num_input_tokens_seen": 30547904,
      "step": 144750
    },
    {
      "epoch": 15.924642464246425,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.0036373242725569535,
      "loss": 0.2319,
      "num_input_tokens_seen": 30548928,
      "step": 144755
    },
    {
      "epoch": 15.925192519251926,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0036363842353320324,
      "loss": 0.2319,
      "num_input_tokens_seen": 30550016,
      "step": 144760
    },
    {
      "epoch": 15.925742574257425,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00363544430284011,
      "loss": 0.2298,
      "num_input_tokens_seen": 30551008,
      "step": 144765
    },
    {
      "epoch": 15.926292629262926,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0036345044750898444,
      "loss": 0.2314,
      "num_input_tokens_seen": 30552032,
      "step": 144770
    },
    {
      "epoch": 15.926842684268427,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0036335647520898927,
      "loss": 0.2308,
      "num_input_tokens_seen": 30553024,
      "step": 144775
    },
    {
      "epoch": 15.927392739273927,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003632625133848919,
      "loss": 0.2335,
      "num_input_tokens_seen": 30554048,
      "step": 144780
    },
    {
      "epoch": 15.927942794279428,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0036316856203755916,
      "loss": 0.2319,
      "num_input_tokens_seen": 30555168,
      "step": 144785
    },
    {
      "epoch": 15.928492849284929,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0036307462116785564,
      "loss": 0.2303,
      "num_input_tokens_seen": 30556224,
      "step": 144790
    },
    {
      "epoch": 15.929042904290428,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.003629806907766484,
      "loss": 0.2324,
      "num_input_tokens_seen": 30557280,
      "step": 144795
    },
    {
      "epoch": 15.92959295929593,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00362886770864802,
      "loss": 0.2319,
      "num_input_tokens_seen": 30558304,
      "step": 144800
    },
    {
      "epoch": 15.93014301430143,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.00362792861433183,
      "loss": 0.2319,
      "num_input_tokens_seen": 30559296,
      "step": 144805
    },
    {
      "epoch": 15.930693069306932,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0036269896248265613,
      "loss": 0.2309,
      "num_input_tokens_seen": 30560320,
      "step": 144810
    },
    {
      "epoch": 15.93124312431243,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.0036260507401408727,
      "loss": 0.2309,
      "num_input_tokens_seen": 30561344,
      "step": 144815
    },
    {
      "epoch": 15.931793179317932,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0036251119602834195,
      "loss": 0.2303,
      "num_input_tokens_seen": 30562400,
      "step": 144820
    },
    {
      "epoch": 15.932343234323433,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0036241732852628477,
      "loss": 0.2314,
      "num_input_tokens_seen": 30563456,
      "step": 144825
    },
    {
      "epoch": 15.932893289328932,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.003623234715087816,
      "loss": 0.2324,
      "num_input_tokens_seen": 30564544,
      "step": 144830
    },
    {
      "epoch": 15.933443344334433,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0036222962497669668,
      "loss": 0.2309,
      "num_input_tokens_seen": 30565600,
      "step": 144835
    },
    {
      "epoch": 15.933993399339935,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0036213578893089524,
      "loss": 0.2298,
      "num_input_tokens_seen": 30566688,
      "step": 144840
    },
    {
      "epoch": 15.934543454345434,
      "grad_norm": 0.000553131103515625,
      "learning_rate": 0.003620419633722428,
      "loss": 0.2308,
      "num_input_tokens_seen": 30567744,
      "step": 144845
    },
    {
      "epoch": 15.935093509350935,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.0036194814830160316,
      "loss": 0.2303,
      "num_input_tokens_seen": 30568768,
      "step": 144850
    },
    {
      "epoch": 15.935643564356436,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.003618543437198416,
      "loss": 0.2303,
      "num_input_tokens_seen": 30569856,
      "step": 144855
    },
    {
      "epoch": 15.936193619361937,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.0036176054962782245,
      "loss": 0.2314,
      "num_input_tokens_seen": 30570976,
      "step": 144860
    },
    {
      "epoch": 15.936743674367436,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0036166676602640982,
      "loss": 0.2304,
      "num_input_tokens_seen": 30572064,
      "step": 144865
    },
    {
      "epoch": 15.937293729372938,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.0036157299291646825,
      "loss": 0.2308,
      "num_input_tokens_seen": 30573152,
      "step": 144870
    },
    {
      "epoch": 15.937843784378439,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.003614792302988622,
      "loss": 0.2329,
      "num_input_tokens_seen": 30574240,
      "step": 144875
    },
    {
      "epoch": 15.938393839383938,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0036138547817445624,
      "loss": 0.2324,
      "num_input_tokens_seen": 30575328,
      "step": 144880
    },
    {
      "epoch": 15.938943894389439,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0036129173654411394,
      "loss": 0.2303,
      "num_input_tokens_seen": 30576352,
      "step": 144885
    },
    {
      "epoch": 15.93949394939494,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0036119800540869887,
      "loss": 0.2324,
      "num_input_tokens_seen": 30577472,
      "step": 144890
    },
    {
      "epoch": 15.94004400440044,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.003611042847690754,
      "loss": 0.2309,
      "num_input_tokens_seen": 30578560,
      "step": 144895
    },
    {
      "epoch": 15.94059405940594,
      "grad_norm": 0.002777099609375,
      "learning_rate": 0.0036101057462610713,
      "loss": 0.2308,
      "num_input_tokens_seen": 30579712,
      "step": 144900
    },
    {
      "epoch": 15.941144114411442,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.003609168749806582,
      "loss": 0.2319,
      "num_input_tokens_seen": 30580736,
      "step": 144905
    },
    {
      "epoch": 15.941694169416941,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.0036082318583359214,
      "loss": 0.2303,
      "num_input_tokens_seen": 30581792,
      "step": 144910
    },
    {
      "epoch": 15.942244224422442,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003607295071857715,
      "loss": 0.2314,
      "num_input_tokens_seen": 30582816,
      "step": 144915
    },
    {
      "epoch": 15.942794279427943,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.0036063583903806085,
      "loss": 0.2308,
      "num_input_tokens_seen": 30583872,
      "step": 144920
    },
    {
      "epoch": 15.943344334433444,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0036054218139132243,
      "loss": 0.2319,
      "num_input_tokens_seen": 30584928,
      "step": 144925
    },
    {
      "epoch": 15.943894389438944,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.003604485342464199,
      "loss": 0.2309,
      "num_input_tokens_seen": 30585952,
      "step": 144930
    },
    {
      "epoch": 15.944444444444445,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0036035489760421697,
      "loss": 0.2324,
      "num_input_tokens_seen": 30586944,
      "step": 144935
    },
    {
      "epoch": 15.944994499449946,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0036026127146557567,
      "loss": 0.2309,
      "num_input_tokens_seen": 30587904,
      "step": 144940
    },
    {
      "epoch": 15.945544554455445,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.003601676558313597,
      "loss": 0.2314,
      "num_input_tokens_seen": 30588960,
      "step": 144945
    },
    {
      "epoch": 15.946094609460946,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.003600740507024311,
      "loss": 0.2314,
      "num_input_tokens_seen": 30590016,
      "step": 144950
    },
    {
      "epoch": 15.946644664466447,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003599804560796529,
      "loss": 0.2324,
      "num_input_tokens_seen": 30591008,
      "step": 144955
    },
    {
      "epoch": 15.947194719471947,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.0035988687196388844,
      "loss": 0.2308,
      "num_input_tokens_seen": 30592128,
      "step": 144960
    },
    {
      "epoch": 15.947744774477448,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.0035979329835599894,
      "loss": 0.233,
      "num_input_tokens_seen": 30593184,
      "step": 144965
    },
    {
      "epoch": 15.948294829482949,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.003596997352568481,
      "loss": 0.2303,
      "num_input_tokens_seen": 30594208,
      "step": 144970
    },
    {
      "epoch": 15.948844884488448,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.0035960618266729747,
      "loss": 0.2324,
      "num_input_tokens_seen": 30595232,
      "step": 144975
    },
    {
      "epoch": 15.94939493949395,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003595126405882092,
      "loss": 0.2324,
      "num_input_tokens_seen": 30596224,
      "step": 144980
    },
    {
      "epoch": 15.94994499449945,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.0035941910902044555,
      "loss": 0.2329,
      "num_input_tokens_seen": 30597312,
      "step": 144985
    },
    {
      "epoch": 15.950495049504951,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.003593255879648687,
      "loss": 0.2314,
      "num_input_tokens_seen": 30598368,
      "step": 144990
    },
    {
      "epoch": 15.95104510451045,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.003592320774223409,
      "loss": 0.2319,
      "num_input_tokens_seen": 30599392,
      "step": 144995
    },
    {
      "epoch": 15.951595159515952,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.003591385773937237,
      "loss": 0.2309,
      "num_input_tokens_seen": 30600544,
      "step": 145000
    },
    {
      "epoch": 15.952145214521453,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.003590450878798784,
      "loss": 0.2308,
      "num_input_tokens_seen": 30601632,
      "step": 145005
    },
    {
      "epoch": 15.952695269526952,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00358951608881667,
      "loss": 0.2319,
      "num_input_tokens_seen": 30602688,
      "step": 145010
    },
    {
      "epoch": 15.953245324532453,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0035885814039995105,
      "loss": 0.2319,
      "num_input_tokens_seen": 30603712,
      "step": 145015
    },
    {
      "epoch": 15.953795379537954,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.003587646824355926,
      "loss": 0.2319,
      "num_input_tokens_seen": 30604768,
      "step": 145020
    },
    {
      "epoch": 15.954345434543454,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.003586712349894523,
      "loss": 0.2314,
      "num_input_tokens_seen": 30605792,
      "step": 145025
    },
    {
      "epoch": 15.954895489548955,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0035857779806239124,
      "loss": 0.2309,
      "num_input_tokens_seen": 30606816,
      "step": 145030
    },
    {
      "epoch": 15.955445544554456,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003584843716552713,
      "loss": 0.2319,
      "num_input_tokens_seen": 30607840,
      "step": 145035
    },
    {
      "epoch": 15.955995599559955,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0035839095576895275,
      "loss": 0.2314,
      "num_input_tokens_seen": 30608832,
      "step": 145040
    },
    {
      "epoch": 15.956545654565456,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.0035829755040429705,
      "loss": 0.2319,
      "num_input_tokens_seen": 30609888,
      "step": 145045
    },
    {
      "epoch": 15.957095709570957,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.003582041555621652,
      "loss": 0.2314,
      "num_input_tokens_seen": 30610944,
      "step": 145050
    },
    {
      "epoch": 15.957645764576458,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0035811077124341755,
      "loss": 0.2319,
      "num_input_tokens_seen": 30612032,
      "step": 145055
    },
    {
      "epoch": 15.958195819581958,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.003580173974489152,
      "loss": 0.2308,
      "num_input_tokens_seen": 30613088,
      "step": 145060
    },
    {
      "epoch": 15.958745874587459,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003579240341795183,
      "loss": 0.2324,
      "num_input_tokens_seen": 30614112,
      "step": 145065
    },
    {
      "epoch": 15.95929592959296,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0035783068143608746,
      "loss": 0.2324,
      "num_input_tokens_seen": 30615200,
      "step": 145070
    },
    {
      "epoch": 15.95984598459846,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0035773733921948356,
      "loss": 0.2314,
      "num_input_tokens_seen": 30616256,
      "step": 145075
    },
    {
      "epoch": 15.96039603960396,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0035764400753056608,
      "loss": 0.2324,
      "num_input_tokens_seen": 30617344,
      "step": 145080
    },
    {
      "epoch": 15.960946094609461,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.003575506863701961,
      "loss": 0.2303,
      "num_input_tokens_seen": 30618400,
      "step": 145085
    },
    {
      "epoch": 15.96149614961496,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003574573757392328,
      "loss": 0.2309,
      "num_input_tokens_seen": 30619424,
      "step": 145090
    },
    {
      "epoch": 15.962046204620462,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0035736407563853718,
      "loss": 0.2298,
      "num_input_tokens_seen": 30620480,
      "step": 145095
    },
    {
      "epoch": 15.962596259625963,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.0035727078606896797,
      "loss": 0.2314,
      "num_input_tokens_seen": 30621632,
      "step": 145100
    },
    {
      "epoch": 15.963146314631462,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.003571775070313856,
      "loss": 0.2319,
      "num_input_tokens_seen": 30622656,
      "step": 145105
    },
    {
      "epoch": 15.963696369636963,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.003570842385266503,
      "loss": 0.2319,
      "num_input_tokens_seen": 30623680,
      "step": 145110
    },
    {
      "epoch": 15.964246424642464,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.003569909805556206,
      "loss": 0.2319,
      "num_input_tokens_seen": 30624768,
      "step": 145115
    },
    {
      "epoch": 15.964796479647966,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.003568977331191571,
      "loss": 0.2329,
      "num_input_tokens_seen": 30625760,
      "step": 145120
    },
    {
      "epoch": 15.965346534653465,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0035680449621811813,
      "loss": 0.2324,
      "num_input_tokens_seen": 30626784,
      "step": 145125
    },
    {
      "epoch": 15.965896589658966,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.003567112698533637,
      "loss": 0.233,
      "num_input_tokens_seen": 30627808,
      "step": 145130
    },
    {
      "epoch": 15.966446644664467,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0035661805402575322,
      "loss": 0.2303,
      "num_input_tokens_seen": 30628864,
      "step": 145135
    },
    {
      "epoch": 15.966996699669966,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0035652484873614507,
      "loss": 0.2304,
      "num_input_tokens_seen": 30629856,
      "step": 145140
    },
    {
      "epoch": 15.967546754675467,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003564316539853992,
      "loss": 0.2319,
      "num_input_tokens_seen": 30631008,
      "step": 145145
    },
    {
      "epoch": 15.968096809680969,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.0035633846977437393,
      "loss": 0.2314,
      "num_input_tokens_seen": 30632032,
      "step": 145150
    },
    {
      "epoch": 15.968646864686468,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0035624529610392788,
      "loss": 0.2293,
      "num_input_tokens_seen": 30633088,
      "step": 145155
    },
    {
      "epoch": 15.969196919691969,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.003561521329749201,
      "loss": 0.2319,
      "num_input_tokens_seen": 30634112,
      "step": 145160
    },
    {
      "epoch": 15.96974697469747,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.0035605898038820966,
      "loss": 0.2309,
      "num_input_tokens_seen": 30635168,
      "step": 145165
    },
    {
      "epoch": 15.97029702970297,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0035596583834465404,
      "loss": 0.2309,
      "num_input_tokens_seen": 30636192,
      "step": 145170
    },
    {
      "epoch": 15.97084708470847,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.0035587270684511296,
      "loss": 0.2314,
      "num_input_tokens_seen": 30637216,
      "step": 145175
    },
    {
      "epoch": 15.971397139713972,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0035577958589044366,
      "loss": 0.2314,
      "num_input_tokens_seen": 30638336,
      "step": 145180
    },
    {
      "epoch": 15.971947194719473,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0035568647548150497,
      "loss": 0.2314,
      "num_input_tokens_seen": 30639456,
      "step": 145185
    },
    {
      "epoch": 15.972497249724972,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.003555933756191553,
      "loss": 0.2293,
      "num_input_tokens_seen": 30640512,
      "step": 145190
    },
    {
      "epoch": 15.973047304730473,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.003555002863042519,
      "loss": 0.2319,
      "num_input_tokens_seen": 30641536,
      "step": 145195
    },
    {
      "epoch": 15.973597359735974,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.003554072075376535,
      "loss": 0.2309,
      "num_input_tokens_seen": 30642592,
      "step": 145200
    },
    {
      "epoch": 15.974147414741473,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003553141393202175,
      "loss": 0.2308,
      "num_input_tokens_seen": 30643584,
      "step": 145205
    },
    {
      "epoch": 15.974697469746975,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0035522108165280198,
      "loss": 0.2314,
      "num_input_tokens_seen": 30644672,
      "step": 145210
    },
    {
      "epoch": 15.975247524752476,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.0035512803453626412,
      "loss": 0.2298,
      "num_input_tokens_seen": 30645696,
      "step": 145215
    },
    {
      "epoch": 15.975797579757975,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.0035503499797146175,
      "loss": 0.2293,
      "num_input_tokens_seen": 30646784,
      "step": 145220
    },
    {
      "epoch": 15.976347634763476,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0035494197195925286,
      "loss": 0.2293,
      "num_input_tokens_seen": 30647808,
      "step": 145225
    },
    {
      "epoch": 15.976897689768977,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.003548489565004939,
      "loss": 0.2325,
      "num_input_tokens_seen": 30648864,
      "step": 145230
    },
    {
      "epoch": 15.977447744774478,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0035475595159604307,
      "loss": 0.2314,
      "num_input_tokens_seen": 30649952,
      "step": 145235
    },
    {
      "epoch": 15.977997799779978,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.003546629572467566,
      "loss": 0.2319,
      "num_input_tokens_seen": 30651008,
      "step": 145240
    },
    {
      "epoch": 15.978547854785479,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0035456997345349205,
      "loss": 0.2329,
      "num_input_tokens_seen": 30652064,
      "step": 145245
    },
    {
      "epoch": 15.97909790979098,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.0035447700021710702,
      "loss": 0.2319,
      "num_input_tokens_seen": 30653152,
      "step": 145250
    },
    {
      "epoch": 15.979647964796479,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.003543840375384572,
      "loss": 0.2309,
      "num_input_tokens_seen": 30654208,
      "step": 145255
    },
    {
      "epoch": 15.98019801980198,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0035429108541840055,
      "loss": 0.2308,
      "num_input_tokens_seen": 30655264,
      "step": 145260
    },
    {
      "epoch": 15.980748074807481,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.00354198143857793,
      "loss": 0.2303,
      "num_input_tokens_seen": 30656352,
      "step": 145265
    },
    {
      "epoch": 15.98129812981298,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0035410521285749114,
      "loss": 0.2309,
      "num_input_tokens_seen": 30657472,
      "step": 145270
    },
    {
      "epoch": 15.981848184818482,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003540122924183515,
      "loss": 0.2324,
      "num_input_tokens_seen": 30658496,
      "step": 145275
    },
    {
      "epoch": 15.982398239823983,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.0035391938254123083,
      "loss": 0.2319,
      "num_input_tokens_seen": 30659552,
      "step": 145280
    },
    {
      "epoch": 15.982948294829484,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0035382648322698556,
      "loss": 0.2314,
      "num_input_tokens_seen": 30660608,
      "step": 145285
    },
    {
      "epoch": 15.983498349834983,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0035373359447647165,
      "loss": 0.2308,
      "num_input_tokens_seen": 30661696,
      "step": 145290
    },
    {
      "epoch": 15.984048404840484,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0035364071629054483,
      "loss": 0.2308,
      "num_input_tokens_seen": 30662688,
      "step": 145295
    },
    {
      "epoch": 15.984598459845985,
      "grad_norm": 0.0026702880859375,
      "learning_rate": 0.0035354784867006133,
      "loss": 0.2309,
      "num_input_tokens_seen": 30663712,
      "step": 145300
    },
    {
      "epoch": 15.985148514851485,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0035345499161587735,
      "loss": 0.234,
      "num_input_tokens_seen": 30664768,
      "step": 145305
    },
    {
      "epoch": 15.985698569856986,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.003533621451288488,
      "loss": 0.2329,
      "num_input_tokens_seen": 30665856,
      "step": 145310
    },
    {
      "epoch": 15.986248624862487,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.003532693092098313,
      "loss": 0.2314,
      "num_input_tokens_seen": 30666912,
      "step": 145315
    },
    {
      "epoch": 15.986798679867986,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.0035317648385967968,
      "loss": 0.2329,
      "num_input_tokens_seen": 30667968,
      "step": 145320
    },
    {
      "epoch": 15.987348734873487,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.003530836690792507,
      "loss": 0.2324,
      "num_input_tokens_seen": 30668960,
      "step": 145325
    },
    {
      "epoch": 15.987898789878988,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.003529908648693988,
      "loss": 0.2303,
      "num_input_tokens_seen": 30670016,
      "step": 145330
    },
    {
      "epoch": 15.988448844884488,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.003528980712309796,
      "loss": 0.2314,
      "num_input_tokens_seen": 30671072,
      "step": 145335
    },
    {
      "epoch": 15.988998899889989,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.003528052881648488,
      "loss": 0.2293,
      "num_input_tokens_seen": 30672128,
      "step": 145340
    },
    {
      "epoch": 15.98954895489549,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00352712515671861,
      "loss": 0.2314,
      "num_input_tokens_seen": 30673248,
      "step": 145345
    },
    {
      "epoch": 15.990099009900991,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0035261975375287156,
      "loss": 0.2314,
      "num_input_tokens_seen": 30674336,
      "step": 145350
    },
    {
      "epoch": 15.99064906490649,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0035252700240873546,
      "loss": 0.2329,
      "num_input_tokens_seen": 30675360,
      "step": 145355
    },
    {
      "epoch": 15.991199119911991,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.003524342616403065,
      "loss": 0.2309,
      "num_input_tokens_seen": 30676384,
      "step": 145360
    },
    {
      "epoch": 15.991749174917492,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0035234153144844114,
      "loss": 0.2308,
      "num_input_tokens_seen": 30677408,
      "step": 145365
    },
    {
      "epoch": 15.992299229922992,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0035224881183399275,
      "loss": 0.2325,
      "num_input_tokens_seen": 30678432,
      "step": 145370
    },
    {
      "epoch": 15.992849284928493,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003521561027978168,
      "loss": 0.2319,
      "num_input_tokens_seen": 30679456,
      "step": 145375
    },
    {
      "epoch": 15.993399339933994,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.003520634043407672,
      "loss": 0.2314,
      "num_input_tokens_seen": 30680544,
      "step": 145380
    },
    {
      "epoch": 15.993949394939493,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0035197071646369802,
      "loss": 0.2293,
      "num_input_tokens_seen": 30681568,
      "step": 145385
    },
    {
      "epoch": 15.994499449944994,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003518780391674637,
      "loss": 0.2314,
      "num_input_tokens_seen": 30682592,
      "step": 145390
    },
    {
      "epoch": 15.995049504950495,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003517853724529188,
      "loss": 0.2314,
      "num_input_tokens_seen": 30683616,
      "step": 145395
    },
    {
      "epoch": 15.995599559955995,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.0035169271632091736,
      "loss": 0.2303,
      "num_input_tokens_seen": 30684704,
      "step": 145400
    },
    {
      "epoch": 15.996149614961496,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0035160007077231323,
      "loss": 0.2288,
      "num_input_tokens_seen": 30685792,
      "step": 145405
    },
    {
      "epoch": 15.996699669966997,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0035150743580795976,
      "loss": 0.2298,
      "num_input_tokens_seen": 30686880,
      "step": 145410
    },
    {
      "epoch": 15.997249724972498,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003514148114287116,
      "loss": 0.2314,
      "num_input_tokens_seen": 30687936,
      "step": 145415
    },
    {
      "epoch": 15.997799779977997,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.003513221976354212,
      "loss": 0.2309,
      "num_input_tokens_seen": 30689024,
      "step": 145420
    },
    {
      "epoch": 15.998349834983498,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.0035122959442894356,
      "loss": 0.2329,
      "num_input_tokens_seen": 30690112,
      "step": 145425
    },
    {
      "epoch": 15.998899889989,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0035113700181013167,
      "loss": 0.2298,
      "num_input_tokens_seen": 30691200,
      "step": 145430
    },
    {
      "epoch": 15.999449944994499,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.0035104441977983833,
      "loss": 0.2314,
      "num_input_tokens_seen": 30692192,
      "step": 145435
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.0035095184833891773,
      "loss": 0.2324,
      "num_input_tokens_seen": 30693088,
      "step": 145440
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.23126927018165588,
      "eval_runtime": 60.6251,
      "eval_samples_per_second": 66.639,
      "eval_steps_per_second": 16.66,
      "num_input_tokens_seen": 30693088,
      "step": 145440
    },
    {
      "epoch": 16.0005500550055,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0035085928748822197,
      "loss": 0.2314,
      "num_input_tokens_seen": 30694208,
      "step": 145445
    },
    {
      "epoch": 16.001100110011002,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.003507667372286049,
      "loss": 0.2314,
      "num_input_tokens_seen": 30695232,
      "step": 145450
    },
    {
      "epoch": 16.001650165016503,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0035067419756091966,
      "loss": 0.2293,
      "num_input_tokens_seen": 30696256,
      "step": 145455
    },
    {
      "epoch": 16.002200220022,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0035058166848601856,
      "loss": 0.2309,
      "num_input_tokens_seen": 30697344,
      "step": 145460
    },
    {
      "epoch": 16.002750275027502,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0035048915000475494,
      "loss": 0.2335,
      "num_input_tokens_seen": 30698304,
      "step": 145465
    },
    {
      "epoch": 16.003300330033003,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.003503966421179809,
      "loss": 0.2298,
      "num_input_tokens_seen": 30699488,
      "step": 145470
    },
    {
      "epoch": 16.003850385038504,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0035030414482654986,
      "loss": 0.2304,
      "num_input_tokens_seen": 30700544,
      "step": 145475
    },
    {
      "epoch": 16.004400440044005,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.003502116581313134,
      "loss": 0.2303,
      "num_input_tokens_seen": 30701632,
      "step": 145480
    },
    {
      "epoch": 16.004950495049506,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.003501191820331243,
      "loss": 0.2303,
      "num_input_tokens_seen": 30702688,
      "step": 145485
    },
    {
      "epoch": 16.005500550055004,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.0035002671653283526,
      "loss": 0.2309,
      "num_input_tokens_seen": 30703744,
      "step": 145490
    },
    {
      "epoch": 16.006050605060505,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0034993426163129797,
      "loss": 0.2329,
      "num_input_tokens_seen": 30704832,
      "step": 145495
    },
    {
      "epoch": 16.006600660066006,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0034984181732936504,
      "loss": 0.2319,
      "num_input_tokens_seen": 30705856,
      "step": 145500
    },
    {
      "epoch": 16.007150715071507,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0034974938362788777,
      "loss": 0.2314,
      "num_input_tokens_seen": 30706880,
      "step": 145505
    },
    {
      "epoch": 16.007700770077008,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.003496569605277185,
      "loss": 0.2314,
      "num_input_tokens_seen": 30708000,
      "step": 145510
    },
    {
      "epoch": 16.00825082508251,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0034956454802970937,
      "loss": 0.2293,
      "num_input_tokens_seen": 30708960,
      "step": 145515
    },
    {
      "epoch": 16.00880088008801,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.003494721461347114,
      "loss": 0.2309,
      "num_input_tokens_seen": 30709984,
      "step": 145520
    },
    {
      "epoch": 16.009350935093508,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.00349379754843577,
      "loss": 0.2314,
      "num_input_tokens_seen": 30711040,
      "step": 145525
    },
    {
      "epoch": 16.00990099009901,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0034928737415715724,
      "loss": 0.2309,
      "num_input_tokens_seen": 30712128,
      "step": 145530
    },
    {
      "epoch": 16.01045104510451,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0034919500407630294,
      "loss": 0.2324,
      "num_input_tokens_seen": 30713152,
      "step": 145535
    },
    {
      "epoch": 16.01100110011001,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.003491026446018667,
      "loss": 0.2329,
      "num_input_tokens_seen": 30714272,
      "step": 145540
    },
    {
      "epoch": 16.011551155115512,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0034901029573469883,
      "loss": 0.2319,
      "num_input_tokens_seen": 30715360,
      "step": 145545
    },
    {
      "epoch": 16.012101210121013,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00348917957475651,
      "loss": 0.2324,
      "num_input_tokens_seen": 30716416,
      "step": 145550
    },
    {
      "epoch": 16.01265126512651,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0034882562982557425,
      "loss": 0.2324,
      "num_input_tokens_seen": 30717568,
      "step": 145555
    },
    {
      "epoch": 16.013201320132012,
      "grad_norm": 0.00058746337890625,
      "learning_rate": 0.003487333127853188,
      "loss": 0.2319,
      "num_input_tokens_seen": 30718592,
      "step": 145560
    },
    {
      "epoch": 16.013751375137513,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.003486410063557359,
      "loss": 0.2303,
      "num_input_tokens_seen": 30719712,
      "step": 145565
    },
    {
      "epoch": 16.014301430143014,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.0034854871053767683,
      "loss": 0.2309,
      "num_input_tokens_seen": 30720736,
      "step": 145570
    },
    {
      "epoch": 16.014851485148515,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.003484564253319914,
      "loss": 0.2308,
      "num_input_tokens_seen": 30721792,
      "step": 145575
    },
    {
      "epoch": 16.015401540154016,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003483641507395309,
      "loss": 0.2319,
      "num_input_tokens_seen": 30722784,
      "step": 145580
    },
    {
      "epoch": 16.015951595159517,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.003482718867611449,
      "loss": 0.2298,
      "num_input_tokens_seen": 30723808,
      "step": 145585
    },
    {
      "epoch": 16.016501650165015,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.003481796333976848,
      "loss": 0.2304,
      "num_input_tokens_seen": 30724864,
      "step": 145590
    },
    {
      "epoch": 16.017051705170516,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0034808739065,
      "loss": 0.2304,
      "num_input_tokens_seen": 30725952,
      "step": 145595
    },
    {
      "epoch": 16.017601760176017,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.0034799515851894076,
      "loss": 0.2319,
      "num_input_tokens_seen": 30727008,
      "step": 145600
    },
    {
      "epoch": 16.01815181518152,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0034790293700535787,
      "loss": 0.2314,
      "num_input_tokens_seen": 30728032,
      "step": 145605
    },
    {
      "epoch": 16.01870187018702,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.0034781072611010043,
      "loss": 0.2299,
      "num_input_tokens_seen": 30729152,
      "step": 145610
    },
    {
      "epoch": 16.01925192519252,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.00347718525834019,
      "loss": 0.2329,
      "num_input_tokens_seen": 30730208,
      "step": 145615
    },
    {
      "epoch": 16.019801980198018,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.003476263361779626,
      "loss": 0.2319,
      "num_input_tokens_seen": 30731200,
      "step": 145620
    },
    {
      "epoch": 16.02035203520352,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.0034753415714278138,
      "loss": 0.2314,
      "num_input_tokens_seen": 30732288,
      "step": 145625
    },
    {
      "epoch": 16.02090209020902,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0034744198872932518,
      "loss": 0.2319,
      "num_input_tokens_seen": 30733312,
      "step": 145630
    },
    {
      "epoch": 16.02145214521452,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0034734983093844263,
      "loss": 0.2319,
      "num_input_tokens_seen": 30734368,
      "step": 145635
    },
    {
      "epoch": 16.022002200220022,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.00347257683770984,
      "loss": 0.2319,
      "num_input_tokens_seen": 30735424,
      "step": 145640
    },
    {
      "epoch": 16.022552255225524,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0034716554722779825,
      "loss": 0.2304,
      "num_input_tokens_seen": 30736448,
      "step": 145645
    },
    {
      "epoch": 16.023102310231025,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0034707342130973406,
      "loss": 0.2309,
      "num_input_tokens_seen": 30737408,
      "step": 145650
    },
    {
      "epoch": 16.023652365236522,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003469813060176408,
      "loss": 0.2308,
      "num_input_tokens_seen": 30738464,
      "step": 145655
    },
    {
      "epoch": 16.024202420242023,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0034688920135236764,
      "loss": 0.2319,
      "num_input_tokens_seen": 30739488,
      "step": 145660
    },
    {
      "epoch": 16.024752475247524,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0034679710731476392,
      "loss": 0.2288,
      "num_input_tokens_seen": 30740576,
      "step": 145665
    },
    {
      "epoch": 16.025302530253025,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0034670502390567776,
      "loss": 0.2314,
      "num_input_tokens_seen": 30741696,
      "step": 145670
    },
    {
      "epoch": 16.025852585258527,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0034661295112595764,
      "loss": 0.2303,
      "num_input_tokens_seen": 30742784,
      "step": 145675
    },
    {
      "epoch": 16.026402640264028,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0034652088897645255,
      "loss": 0.2298,
      "num_input_tokens_seen": 30743840,
      "step": 145680
    },
    {
      "epoch": 16.02695269526953,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.003464288374580108,
      "loss": 0.233,
      "num_input_tokens_seen": 30744864,
      "step": 145685
    },
    {
      "epoch": 16.027502750275026,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0034633679657148147,
      "loss": 0.2309,
      "num_input_tokens_seen": 30745920,
      "step": 145690
    },
    {
      "epoch": 16.028052805280527,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.003462447663177121,
      "loss": 0.2329,
      "num_input_tokens_seen": 30747008,
      "step": 145695
    },
    {
      "epoch": 16.02860286028603,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.003461527466975509,
      "loss": 0.2345,
      "num_input_tokens_seen": 30748128,
      "step": 145700
    },
    {
      "epoch": 16.02915291529153,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.003460607377118465,
      "loss": 0.2308,
      "num_input_tokens_seen": 30749184,
      "step": 145705
    },
    {
      "epoch": 16.02970297029703,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.0034596873936144613,
      "loss": 0.2303,
      "num_input_tokens_seen": 30750240,
      "step": 145710
    },
    {
      "epoch": 16.03025302530253,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0034587675164719817,
      "loss": 0.2329,
      "num_input_tokens_seen": 30751328,
      "step": 145715
    },
    {
      "epoch": 16.03080308030803,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0034578477456995065,
      "loss": 0.2314,
      "num_input_tokens_seen": 30752416,
      "step": 145720
    },
    {
      "epoch": 16.03135313531353,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0034569280813055076,
      "loss": 0.2309,
      "num_input_tokens_seen": 30753504,
      "step": 145725
    },
    {
      "epoch": 16.03190319031903,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.0034560085232984675,
      "loss": 0.2309,
      "num_input_tokens_seen": 30754592,
      "step": 145730
    },
    {
      "epoch": 16.032453245324533,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.0034550890716868537,
      "loss": 0.2298,
      "num_input_tokens_seen": 30755584,
      "step": 145735
    },
    {
      "epoch": 16.033003300330034,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0034541697264791438,
      "loss": 0.2298,
      "num_input_tokens_seen": 30756672,
      "step": 145740
    },
    {
      "epoch": 16.033553355335535,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.003453250487683816,
      "loss": 0.2335,
      "num_input_tokens_seen": 30757728,
      "step": 145745
    },
    {
      "epoch": 16.034103410341036,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.003452331355309332,
      "loss": 0.2303,
      "num_input_tokens_seen": 30758784,
      "step": 145750
    },
    {
      "epoch": 16.034653465346533,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0034514123293641736,
      "loss": 0.2319,
      "num_input_tokens_seen": 30759872,
      "step": 145755
    },
    {
      "epoch": 16.035203520352034,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0034504934098568067,
      "loss": 0.2314,
      "num_input_tokens_seen": 30760896,
      "step": 145760
    },
    {
      "epoch": 16.035753575357536,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.003449574596795697,
      "loss": 0.2329,
      "num_input_tokens_seen": 30761856,
      "step": 145765
    },
    {
      "epoch": 16.036303630363037,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003448655890189314,
      "loss": 0.2304,
      "num_input_tokens_seen": 30762976,
      "step": 145770
    },
    {
      "epoch": 16.036853685368538,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003447737290046127,
      "loss": 0.2319,
      "num_input_tokens_seen": 30764096,
      "step": 145775
    },
    {
      "epoch": 16.03740374037404,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0034468187963746072,
      "loss": 0.2319,
      "num_input_tokens_seen": 30765152,
      "step": 145780
    },
    {
      "epoch": 16.037953795379536,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0034459004091832134,
      "loss": 0.2314,
      "num_input_tokens_seen": 30766208,
      "step": 145785
    },
    {
      "epoch": 16.038503850385037,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.003444982128480408,
      "loss": 0.2298,
      "num_input_tokens_seen": 30767296,
      "step": 145790
    },
    {
      "epoch": 16.03905390539054,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.0034440639542746582,
      "loss": 0.2293,
      "num_input_tokens_seen": 30768320,
      "step": 145795
    },
    {
      "epoch": 16.03960396039604,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.0034431458865744244,
      "loss": 0.2324,
      "num_input_tokens_seen": 30769376,
      "step": 145800
    },
    {
      "epoch": 16.04015401540154,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0034422279253881753,
      "loss": 0.2314,
      "num_input_tokens_seen": 30770432,
      "step": 145805
    },
    {
      "epoch": 16.040704070407042,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0034413100707243655,
      "loss": 0.2319,
      "num_input_tokens_seen": 30771456,
      "step": 145810
    },
    {
      "epoch": 16.041254125412543,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0034403923225914492,
      "loss": 0.2308,
      "num_input_tokens_seen": 30772448,
      "step": 145815
    },
    {
      "epoch": 16.04180418041804,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.003439474680997894,
      "loss": 0.2283,
      "num_input_tokens_seen": 30773504,
      "step": 145820
    },
    {
      "epoch": 16.04235423542354,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0034385571459521497,
      "loss": 0.2303,
      "num_input_tokens_seen": 30774496,
      "step": 145825
    },
    {
      "epoch": 16.042904290429043,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0034376397174626754,
      "loss": 0.2314,
      "num_input_tokens_seen": 30775520,
      "step": 145830
    },
    {
      "epoch": 16.043454345434544,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0034367223955379333,
      "loss": 0.2308,
      "num_input_tokens_seen": 30776576,
      "step": 145835
    },
    {
      "epoch": 16.044004400440045,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.003435805180186368,
      "loss": 0.2303,
      "num_input_tokens_seen": 30777632,
      "step": 145840
    },
    {
      "epoch": 16.044554455445546,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.0034348880714164414,
      "loss": 0.2314,
      "num_input_tokens_seen": 30778688,
      "step": 145845
    },
    {
      "epoch": 16.045104510451043,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0034339710692365966,
      "loss": 0.2309,
      "num_input_tokens_seen": 30779712,
      "step": 145850
    },
    {
      "epoch": 16.045654565456545,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0034330541736552916,
      "loss": 0.2314,
      "num_input_tokens_seen": 30780832,
      "step": 145855
    },
    {
      "epoch": 16.046204620462046,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.00343213738468098,
      "loss": 0.2309,
      "num_input_tokens_seen": 30781920,
      "step": 145860
    },
    {
      "epoch": 16.046754675467547,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.003431220702322103,
      "loss": 0.2298,
      "num_input_tokens_seen": 30782976,
      "step": 145865
    },
    {
      "epoch": 16.047304730473048,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0034303041265871174,
      "loss": 0.2303,
      "num_input_tokens_seen": 30784032,
      "step": 145870
    },
    {
      "epoch": 16.04785478547855,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0034293876574844635,
      "loss": 0.2314,
      "num_input_tokens_seen": 30785088,
      "step": 145875
    },
    {
      "epoch": 16.04840484048405,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.003428471295022595,
      "loss": 0.2303,
      "num_input_tokens_seen": 30786144,
      "step": 145880
    },
    {
      "epoch": 16.048954895489548,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.00342755503920995,
      "loss": 0.2314,
      "num_input_tokens_seen": 30787232,
      "step": 145885
    },
    {
      "epoch": 16.04950495049505,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0034266388900549777,
      "loss": 0.2314,
      "num_input_tokens_seen": 30788256,
      "step": 145890
    },
    {
      "epoch": 16.05005500550055,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.003425722847566126,
      "loss": 0.2309,
      "num_input_tokens_seen": 30789248,
      "step": 145895
    },
    {
      "epoch": 16.05060506050605,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003424806911751827,
      "loss": 0.2293,
      "num_input_tokens_seen": 30790336,
      "step": 145900
    },
    {
      "epoch": 16.051155115511552,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0034238910826205338,
      "loss": 0.2335,
      "num_input_tokens_seen": 30791392,
      "step": 145905
    },
    {
      "epoch": 16.051705170517053,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.003422975360180678,
      "loss": 0.2314,
      "num_input_tokens_seen": 30792416,
      "step": 145910
    },
    {
      "epoch": 16.05225522552255,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0034220597444407025,
      "loss": 0.2335,
      "num_input_tokens_seen": 30793440,
      "step": 145915
    },
    {
      "epoch": 16.05280528052805,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.003421144235409051,
      "loss": 0.2298,
      "num_input_tokens_seen": 30794560,
      "step": 145920
    },
    {
      "epoch": 16.053355335533553,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.0034202288330941514,
      "loss": 0.2319,
      "num_input_tokens_seen": 30795616,
      "step": 145925
    },
    {
      "epoch": 16.053905390539054,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0034193135375044526,
      "loss": 0.2324,
      "num_input_tokens_seen": 30796736,
      "step": 145930
    },
    {
      "epoch": 16.054455445544555,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.003418398348648382,
      "loss": 0.2319,
      "num_input_tokens_seen": 30797696,
      "step": 145935
    },
    {
      "epoch": 16.055005500550056,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0034174832665343734,
      "loss": 0.2293,
      "num_input_tokens_seen": 30798752,
      "step": 145940
    },
    {
      "epoch": 16.055555555555557,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.003416568291170863,
      "loss": 0.2303,
      "num_input_tokens_seen": 30799776,
      "step": 145945
    },
    {
      "epoch": 16.056105610561055,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.003415653422566288,
      "loss": 0.2298,
      "num_input_tokens_seen": 30800800,
      "step": 145950
    },
    {
      "epoch": 16.056655665566556,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003414738660729075,
      "loss": 0.2319,
      "num_input_tokens_seen": 30801888,
      "step": 145955
    },
    {
      "epoch": 16.057205720572057,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0034138240056676592,
      "loss": 0.2324,
      "num_input_tokens_seen": 30802944,
      "step": 145960
    },
    {
      "epoch": 16.057755775577558,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0034129094573904636,
      "loss": 0.2335,
      "num_input_tokens_seen": 30803968,
      "step": 145965
    },
    {
      "epoch": 16.05830583058306,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0034119950159059212,
      "loss": 0.2319,
      "num_input_tokens_seen": 30805024,
      "step": 145970
    },
    {
      "epoch": 16.05885588558856,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.003411080681222465,
      "loss": 0.2309,
      "num_input_tokens_seen": 30806112,
      "step": 145975
    },
    {
      "epoch": 16.059405940594058,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.0034101664533485137,
      "loss": 0.2329,
      "num_input_tokens_seen": 30807168,
      "step": 145980
    },
    {
      "epoch": 16.05995599559956,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0034092523322924997,
      "loss": 0.2319,
      "num_input_tokens_seen": 30808224,
      "step": 145985
    },
    {
      "epoch": 16.06050605060506,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0034083383180628425,
      "loss": 0.2303,
      "num_input_tokens_seen": 30809312,
      "step": 145990
    },
    {
      "epoch": 16.06105610561056,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.0034074244106679723,
      "loss": 0.2319,
      "num_input_tokens_seen": 30810368,
      "step": 145995
    },
    {
      "epoch": 16.061606160616062,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.003406510610116303,
      "loss": 0.2293,
      "num_input_tokens_seen": 30811392,
      "step": 146000
    },
    {
      "epoch": 16.062156215621563,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.0034055969164162635,
      "loss": 0.2308,
      "num_input_tokens_seen": 30812448,
      "step": 146005
    },
    {
      "epoch": 16.062706270627064,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0034046833295762777,
      "loss": 0.2314,
      "num_input_tokens_seen": 30813472,
      "step": 146010
    },
    {
      "epoch": 16.063256325632562,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0034037698496047577,
      "loss": 0.2319,
      "num_input_tokens_seen": 30814464,
      "step": 146015
    },
    {
      "epoch": 16.063806380638063,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.003402856476510132,
      "loss": 0.2308,
      "num_input_tokens_seen": 30815488,
      "step": 146020
    },
    {
      "epoch": 16.064356435643564,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.003401943210300808,
      "loss": 0.2303,
      "num_input_tokens_seen": 30816608,
      "step": 146025
    },
    {
      "epoch": 16.064906490649065,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0034010300509852084,
      "loss": 0.2314,
      "num_input_tokens_seen": 30817664,
      "step": 146030
    },
    {
      "epoch": 16.065456545654566,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0034001169985717525,
      "loss": 0.2314,
      "num_input_tokens_seen": 30818720,
      "step": 146035
    },
    {
      "epoch": 16.066006600660067,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0033992040530688486,
      "loss": 0.2314,
      "num_input_tokens_seen": 30819840,
      "step": 146040
    },
    {
      "epoch": 16.066556655665565,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0033982912144849184,
      "loss": 0.2303,
      "num_input_tokens_seen": 30820928,
      "step": 146045
    },
    {
      "epoch": 16.067106710671066,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00339737848282837,
      "loss": 0.2298,
      "num_input_tokens_seen": 30822016,
      "step": 146050
    },
    {
      "epoch": 16.067656765676567,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.003396465858107614,
      "loss": 0.2319,
      "num_input_tokens_seen": 30823072,
      "step": 146055
    },
    {
      "epoch": 16.068206820682068,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0033955533403310645,
      "loss": 0.2283,
      "num_input_tokens_seen": 30824128,
      "step": 146060
    },
    {
      "epoch": 16.06875687568757,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.00339464092950713,
      "loss": 0.2319,
      "num_input_tokens_seen": 30825152,
      "step": 146065
    },
    {
      "epoch": 16.06930693069307,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.0033937286256442256,
      "loss": 0.2319,
      "num_input_tokens_seen": 30826272,
      "step": 146070
    },
    {
      "epoch": 16.06985698569857,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003392816428750756,
      "loss": 0.234,
      "num_input_tokens_seen": 30827392,
      "step": 146075
    },
    {
      "epoch": 16.07040704070407,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.003391904338835123,
      "loss": 0.2309,
      "num_input_tokens_seen": 30828416,
      "step": 146080
    },
    {
      "epoch": 16.07095709570957,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0033909923559057387,
      "loss": 0.2309,
      "num_input_tokens_seen": 30829504,
      "step": 146085
    },
    {
      "epoch": 16.07150715071507,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.003390080479971007,
      "loss": 0.2288,
      "num_input_tokens_seen": 30830592,
      "step": 146090
    },
    {
      "epoch": 16.072057205720572,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.003389168711039335,
      "loss": 0.2298,
      "num_input_tokens_seen": 30831648,
      "step": 146095
    },
    {
      "epoch": 16.072607260726073,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.003388257049119125,
      "loss": 0.2308,
      "num_input_tokens_seen": 30832640,
      "step": 146100
    },
    {
      "epoch": 16.073157315731574,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.003387345494218774,
      "loss": 0.2329,
      "num_input_tokens_seen": 30833696,
      "step": 146105
    },
    {
      "epoch": 16.073707370737075,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.003386434046346691,
      "loss": 0.2314,
      "num_input_tokens_seen": 30834720,
      "step": 146110
    },
    {
      "epoch": 16.074257425742573,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.003385522705511268,
      "loss": 0.2329,
      "num_input_tokens_seen": 30835744,
      "step": 146115
    },
    {
      "epoch": 16.074807480748074,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.003384611471720909,
      "loss": 0.2314,
      "num_input_tokens_seen": 30836800,
      "step": 146120
    },
    {
      "epoch": 16.075357535753575,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.003383700344984018,
      "loss": 0.2309,
      "num_input_tokens_seen": 30837952,
      "step": 146125
    },
    {
      "epoch": 16.075907590759076,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0033827893253089813,
      "loss": 0.2314,
      "num_input_tokens_seen": 30838976,
      "step": 146130
    },
    {
      "epoch": 16.076457645764577,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.0033818784127042055,
      "loss": 0.2345,
      "num_input_tokens_seen": 30840000,
      "step": 146135
    },
    {
      "epoch": 16.07700770077008,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0033809676071780825,
      "loss": 0.2309,
      "num_input_tokens_seen": 30841088,
      "step": 146140
    },
    {
      "epoch": 16.077557755775576,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.003380056908738997,
      "loss": 0.2324,
      "num_input_tokens_seen": 30842176,
      "step": 146145
    },
    {
      "epoch": 16.078107810781077,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0033791463173953587,
      "loss": 0.2304,
      "num_input_tokens_seen": 30843296,
      "step": 146150
    },
    {
      "epoch": 16.078657865786578,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0033782358331555483,
      "loss": 0.2298,
      "num_input_tokens_seen": 30844352,
      "step": 146155
    },
    {
      "epoch": 16.07920792079208,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.003377325456027965,
      "loss": 0.2314,
      "num_input_tokens_seen": 30845344,
      "step": 146160
    },
    {
      "epoch": 16.07975797579758,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.003376415186020998,
      "loss": 0.2319,
      "num_input_tokens_seen": 30846336,
      "step": 146165
    },
    {
      "epoch": 16.08030803080308,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.003375505023143029,
      "loss": 0.2319,
      "num_input_tokens_seen": 30847328,
      "step": 146170
    },
    {
      "epoch": 16.080858085808583,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.003374594967402453,
      "loss": 0.2309,
      "num_input_tokens_seen": 30848384,
      "step": 146175
    },
    {
      "epoch": 16.08140814081408,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.003373685018807656,
      "loss": 0.2314,
      "num_input_tokens_seen": 30849504,
      "step": 146180
    },
    {
      "epoch": 16.08195819581958,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003372775177367031,
      "loss": 0.2303,
      "num_input_tokens_seen": 30850560,
      "step": 146185
    },
    {
      "epoch": 16.082508250825082,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.0033718654430889565,
      "loss": 0.2303,
      "num_input_tokens_seen": 30851616,
      "step": 146190
    },
    {
      "epoch": 16.083058305830583,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003370955815981814,
      "loss": 0.2313,
      "num_input_tokens_seen": 30852640,
      "step": 146195
    },
    {
      "epoch": 16.083608360836084,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.003370046296053997,
      "loss": 0.2303,
      "num_input_tokens_seen": 30853696,
      "step": 146200
    },
    {
      "epoch": 16.084158415841586,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003369136883313874,
      "loss": 0.2324,
      "num_input_tokens_seen": 30854688,
      "step": 146205
    },
    {
      "epoch": 16.084708470847083,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.0033682275777698442,
      "loss": 0.2309,
      "num_input_tokens_seen": 30855744,
      "step": 146210
    },
    {
      "epoch": 16.085258525852584,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0033673183794302786,
      "loss": 0.2298,
      "num_input_tokens_seen": 30856736,
      "step": 146215
    },
    {
      "epoch": 16.085808580858085,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.003366409288303553,
      "loss": 0.2314,
      "num_input_tokens_seen": 30857856,
      "step": 146220
    },
    {
      "epoch": 16.086358635863586,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.003365500304398056,
      "loss": 0.2319,
      "num_input_tokens_seen": 30858976,
      "step": 146225
    },
    {
      "epoch": 16.086908690869087,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003364591427722154,
      "loss": 0.2303,
      "num_input_tokens_seen": 30860032,
      "step": 146230
    },
    {
      "epoch": 16.08745874587459,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0033636826582842303,
      "loss": 0.2335,
      "num_input_tokens_seen": 30861088,
      "step": 146235
    },
    {
      "epoch": 16.08800880088009,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0033627739960926653,
      "loss": 0.2324,
      "num_input_tokens_seen": 30862144,
      "step": 146240
    },
    {
      "epoch": 16.088558855885587,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0033618654411558223,
      "loss": 0.2298,
      "num_input_tokens_seen": 30863200,
      "step": 146245
    },
    {
      "epoch": 16.08910891089109,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.0033609569934820874,
      "loss": 0.2298,
      "num_input_tokens_seen": 30864288,
      "step": 146250
    },
    {
      "epoch": 16.08965896589659,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.003360048653079821,
      "loss": 0.2303,
      "num_input_tokens_seen": 30865440,
      "step": 146255
    },
    {
      "epoch": 16.09020902090209,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.003359140419957405,
      "loss": 0.2324,
      "num_input_tokens_seen": 30866528,
      "step": 146260
    },
    {
      "epoch": 16.09075907590759,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0033582322941232033,
      "loss": 0.2314,
      "num_input_tokens_seen": 30867488,
      "step": 146265
    },
    {
      "epoch": 16.091309130913093,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.003357324275585587,
      "loss": 0.2314,
      "num_input_tokens_seen": 30868608,
      "step": 146270
    },
    {
      "epoch": 16.09185918591859,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0033564163643529303,
      "loss": 0.2303,
      "num_input_tokens_seen": 30869600,
      "step": 146275
    },
    {
      "epoch": 16.09240924092409,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.003355508560433592,
      "loss": 0.2298,
      "num_input_tokens_seen": 30870624,
      "step": 146280
    },
    {
      "epoch": 16.092959295929592,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0033546008638359486,
      "loss": 0.2324,
      "num_input_tokens_seen": 30871776,
      "step": 146285
    },
    {
      "epoch": 16.093509350935093,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.003353693274568358,
      "loss": 0.2313,
      "num_input_tokens_seen": 30872864,
      "step": 146290
    },
    {
      "epoch": 16.094059405940595,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.003352785792639186,
      "loss": 0.2319,
      "num_input_tokens_seen": 30873888,
      "step": 146295
    },
    {
      "epoch": 16.094609460946096,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.003351878418056804,
      "loss": 0.2324,
      "num_input_tokens_seen": 30874880,
      "step": 146300
    },
    {
      "epoch": 16.095159515951597,
      "grad_norm": 0.000514984130859375,
      "learning_rate": 0.0033509711508295644,
      "loss": 0.2314,
      "num_input_tokens_seen": 30875904,
      "step": 146305
    },
    {
      "epoch": 16.095709570957094,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.0033500639909658384,
      "loss": 0.2309,
      "num_input_tokens_seen": 30876960,
      "step": 146310
    },
    {
      "epoch": 16.096259625962595,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.0033491569384739837,
      "loss": 0.2303,
      "num_input_tokens_seen": 30878016,
      "step": 146315
    },
    {
      "epoch": 16.096809680968097,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.0033482499933623487,
      "loss": 0.2314,
      "num_input_tokens_seen": 30879072,
      "step": 146320
    },
    {
      "epoch": 16.097359735973598,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003347343155639312,
      "loss": 0.2314,
      "num_input_tokens_seen": 30880128,
      "step": 146325
    },
    {
      "epoch": 16.0979097909791,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.003346436425313221,
      "loss": 0.2314,
      "num_input_tokens_seen": 30881216,
      "step": 146330
    },
    {
      "epoch": 16.0984598459846,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.00334552980239243,
      "loss": 0.2298,
      "num_input_tokens_seen": 30882272,
      "step": 146335
    },
    {
      "epoch": 16.099009900990097,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.003344623286885302,
      "loss": 0.2314,
      "num_input_tokens_seen": 30883296,
      "step": 146340
    },
    {
      "epoch": 16.0995599559956,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003343716878800185,
      "loss": 0.2319,
      "num_input_tokens_seen": 30884384,
      "step": 146345
    },
    {
      "epoch": 16.1001100110011,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0033428105781454364,
      "loss": 0.2309,
      "num_input_tokens_seen": 30885472,
      "step": 146350
    },
    {
      "epoch": 16.1006600660066,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0033419043849294123,
      "loss": 0.2293,
      "num_input_tokens_seen": 30886432,
      "step": 146355
    },
    {
      "epoch": 16.1012101210121,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0033409982991604572,
      "loss": 0.2335,
      "num_input_tokens_seen": 30887456,
      "step": 146360
    },
    {
      "epoch": 16.101760176017603,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.003340092320846932,
      "loss": 0.2319,
      "num_input_tokens_seen": 30888512,
      "step": 146365
    },
    {
      "epoch": 16.102310231023104,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.003339186449997175,
      "loss": 0.2319,
      "num_input_tokens_seen": 30889600,
      "step": 146370
    },
    {
      "epoch": 16.1028602860286,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.003338280686619546,
      "loss": 0.2319,
      "num_input_tokens_seen": 30890656,
      "step": 146375
    },
    {
      "epoch": 16.103410341034103,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0033373750307223843,
      "loss": 0.2309,
      "num_input_tokens_seen": 30891680,
      "step": 146380
    },
    {
      "epoch": 16.103960396039604,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0033364694823140406,
      "loss": 0.2303,
      "num_input_tokens_seen": 30892736,
      "step": 146385
    },
    {
      "epoch": 16.104510451045105,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.0033355640414028647,
      "loss": 0.2314,
      "num_input_tokens_seen": 30893760,
      "step": 146390
    },
    {
      "epoch": 16.105060506050606,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0033346587079971933,
      "loss": 0.2309,
      "num_input_tokens_seen": 30894752,
      "step": 146395
    },
    {
      "epoch": 16.105610561056107,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.00333375348210538,
      "loss": 0.2309,
      "num_input_tokens_seen": 30895840,
      "step": 146400
    },
    {
      "epoch": 16.106160616061604,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.0033328483637357575,
      "loss": 0.2314,
      "num_input_tokens_seen": 30896896,
      "step": 146405
    },
    {
      "epoch": 16.106710671067106,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.0033319433528966745,
      "loss": 0.2324,
      "num_input_tokens_seen": 30897920,
      "step": 146410
    },
    {
      "epoch": 16.107260726072607,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.0033310384495964743,
      "loss": 0.2324,
      "num_input_tokens_seen": 30898944,
      "step": 146415
    },
    {
      "epoch": 16.107810781078108,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.0033301336538434888,
      "loss": 0.2324,
      "num_input_tokens_seen": 30900032,
      "step": 146420
    },
    {
      "epoch": 16.10836083608361,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0033292289656460658,
      "loss": 0.2314,
      "num_input_tokens_seen": 30901056,
      "step": 146425
    },
    {
      "epoch": 16.10891089108911,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.003328324385012541,
      "loss": 0.2319,
      "num_input_tokens_seen": 30902144,
      "step": 146430
    },
    {
      "epoch": 16.10946094609461,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0033274199119512436,
      "loss": 0.2303,
      "num_input_tokens_seen": 30903200,
      "step": 146435
    },
    {
      "epoch": 16.11001100110011,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.003326515546470516,
      "loss": 0.2324,
      "num_input_tokens_seen": 30904288,
      "step": 146440
    },
    {
      "epoch": 16.11056105610561,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.003325611288578694,
      "loss": 0.2319,
      "num_input_tokens_seen": 30905312,
      "step": 146445
    },
    {
      "epoch": 16.11111111111111,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003324707138284115,
      "loss": 0.2319,
      "num_input_tokens_seen": 30906368,
      "step": 146450
    },
    {
      "epoch": 16.111661166116612,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0033238030955951075,
      "loss": 0.2314,
      "num_input_tokens_seen": 30907424,
      "step": 146455
    },
    {
      "epoch": 16.112211221122113,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0033228991605200008,
      "loss": 0.2335,
      "num_input_tokens_seen": 30908416,
      "step": 146460
    },
    {
      "epoch": 16.112761276127614,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0033219953330671285,
      "loss": 0.2298,
      "num_input_tokens_seen": 30909472,
      "step": 146465
    },
    {
      "epoch": 16.11331133113311,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.0033210916132448208,
      "loss": 0.2309,
      "num_input_tokens_seen": 30910528,
      "step": 146470
    },
    {
      "epoch": 16.113861386138613,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0033201880010614126,
      "loss": 0.2309,
      "num_input_tokens_seen": 30911584,
      "step": 146475
    },
    {
      "epoch": 16.114411441144114,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0033192844965252272,
      "loss": 0.2324,
      "num_input_tokens_seen": 30912704,
      "step": 146480
    },
    {
      "epoch": 16.114961496149615,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0033183810996445877,
      "loss": 0.2298,
      "num_input_tokens_seen": 30913728,
      "step": 146485
    },
    {
      "epoch": 16.115511551155116,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.003317477810427828,
      "loss": 0.2303,
      "num_input_tokens_seen": 30914816,
      "step": 146490
    },
    {
      "epoch": 16.116061606160617,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003316574628883266,
      "loss": 0.2314,
      "num_input_tokens_seen": 30915904,
      "step": 146495
    },
    {
      "epoch": 16.116611661166118,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.003315671555019228,
      "loss": 0.2314,
      "num_input_tokens_seen": 30916960,
      "step": 146500
    },
    {
      "epoch": 16.117161716171616,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0033147685888440435,
      "loss": 0.2314,
      "num_input_tokens_seen": 30918080,
      "step": 146505
    },
    {
      "epoch": 16.117711771177117,
      "grad_norm": 0.005859375,
      "learning_rate": 0.003313865730366027,
      "loss": 0.2293,
      "num_input_tokens_seen": 30919136,
      "step": 146510
    },
    {
      "epoch": 16.118261826182618,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.0033129629795935044,
      "loss": 0.2319,
      "num_input_tokens_seen": 30920192,
      "step": 146515
    },
    {
      "epoch": 16.11881188118812,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00331206033653479,
      "loss": 0.2298,
      "num_input_tokens_seen": 30921184,
      "step": 146520
    },
    {
      "epoch": 16.11936193619362,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0033111578011982084,
      "loss": 0.2324,
      "num_input_tokens_seen": 30922304,
      "step": 146525
    },
    {
      "epoch": 16.11991199119912,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.00331025537359208,
      "loss": 0.2314,
      "num_input_tokens_seen": 30923424,
      "step": 146530
    },
    {
      "epoch": 16.120462046204622,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0033093530537247153,
      "loss": 0.2303,
      "num_input_tokens_seen": 30924416,
      "step": 146535
    },
    {
      "epoch": 16.12101210121012,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.003308450841604436,
      "loss": 0.2319,
      "num_input_tokens_seen": 30925440,
      "step": 146540
    },
    {
      "epoch": 16.12156215621562,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0033075487372395576,
      "loss": 0.2319,
      "num_input_tokens_seen": 30926496,
      "step": 146545
    },
    {
      "epoch": 16.122112211221122,
      "grad_norm": 0.005859375,
      "learning_rate": 0.003306646740638386,
      "loss": 0.2319,
      "num_input_tokens_seen": 30927552,
      "step": 146550
    },
    {
      "epoch": 16.122662266226623,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00330574485180924,
      "loss": 0.2319,
      "num_input_tokens_seen": 30928608,
      "step": 146555
    },
    {
      "epoch": 16.123212321232124,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003304843070760434,
      "loss": 0.2314,
      "num_input_tokens_seen": 30929696,
      "step": 146560
    },
    {
      "epoch": 16.123762376237625,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.0033039413975002813,
      "loss": 0.2319,
      "num_input_tokens_seen": 30930752,
      "step": 146565
    },
    {
      "epoch": 16.124312431243123,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.003303039832037088,
      "loss": 0.2298,
      "num_input_tokens_seen": 30931776,
      "step": 146570
    },
    {
      "epoch": 16.124862486248624,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.0033021383743791592,
      "loss": 0.2308,
      "num_input_tokens_seen": 30932832,
      "step": 146575
    },
    {
      "epoch": 16.125412541254125,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.003301237024534807,
      "loss": 0.2308,
      "num_input_tokens_seen": 30933824,
      "step": 146580
    },
    {
      "epoch": 16.125962596259626,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.003300335782512341,
      "loss": 0.2314,
      "num_input_tokens_seen": 30934880,
      "step": 146585
    },
    {
      "epoch": 16.126512651265127,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0032994346483200703,
      "loss": 0.2303,
      "num_input_tokens_seen": 30935936,
      "step": 146590
    },
    {
      "epoch": 16.127062706270628,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.003298533621966296,
      "loss": 0.2303,
      "num_input_tokens_seen": 30936960,
      "step": 146595
    },
    {
      "epoch": 16.12761276127613,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0032976327034593177,
      "loss": 0.2303,
      "num_input_tokens_seen": 30938112,
      "step": 146600
    },
    {
      "epoch": 16.128162816281627,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.003296731892807448,
      "loss": 0.2329,
      "num_input_tokens_seen": 30939168,
      "step": 146605
    },
    {
      "epoch": 16.128712871287128,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.0032958311900189797,
      "loss": 0.2303,
      "num_input_tokens_seen": 30940224,
      "step": 146610
    },
    {
      "epoch": 16.12926292629263,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.0032949305951022194,
      "loss": 0.2324,
      "num_input_tokens_seen": 30941312,
      "step": 146615
    },
    {
      "epoch": 16.12981298129813,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0032940301080654727,
      "loss": 0.2309,
      "num_input_tokens_seen": 30942336,
      "step": 146620
    },
    {
      "epoch": 16.13036303630363,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0032931297289170275,
      "loss": 0.2298,
      "num_input_tokens_seen": 30943424,
      "step": 146625
    },
    {
      "epoch": 16.130913091309132,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.003292229457665192,
      "loss": 0.2314,
      "num_input_tokens_seen": 30944512,
      "step": 146630
    },
    {
      "epoch": 16.13146314631463,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.003291329294318257,
      "loss": 0.233,
      "num_input_tokens_seen": 30945568,
      "step": 146635
    },
    {
      "epoch": 16.13201320132013,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.0032904292388845216,
      "loss": 0.2324,
      "num_input_tokens_seen": 30946592,
      "step": 146640
    },
    {
      "epoch": 16.132563256325632,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0032895292913722833,
      "loss": 0.2314,
      "num_input_tokens_seen": 30947584,
      "step": 146645
    },
    {
      "epoch": 16.133113311331133,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0032886294517898316,
      "loss": 0.2314,
      "num_input_tokens_seen": 30948608,
      "step": 146650
    },
    {
      "epoch": 16.133663366336634,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0032877297201454668,
      "loss": 0.2319,
      "num_input_tokens_seen": 30949728,
      "step": 146655
    },
    {
      "epoch": 16.134213421342135,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.003286830096447472,
      "loss": 0.2308,
      "num_input_tokens_seen": 30950752,
      "step": 146660
    },
    {
      "epoch": 16.134763476347636,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0032859305807041495,
      "loss": 0.2314,
      "num_input_tokens_seen": 30951744,
      "step": 146665
    },
    {
      "epoch": 16.135313531353134,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0032850311729237785,
      "loss": 0.2319,
      "num_input_tokens_seen": 30952800,
      "step": 146670
    },
    {
      "epoch": 16.135863586358635,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0032841318731146544,
      "loss": 0.2309,
      "num_input_tokens_seen": 30953856,
      "step": 146675
    },
    {
      "epoch": 16.136413641364136,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0032832326812850692,
      "loss": 0.2309,
      "num_input_tokens_seen": 30954912,
      "step": 146680
    },
    {
      "epoch": 16.136963696369637,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.0032823335974433007,
      "loss": 0.2319,
      "num_input_tokens_seen": 30956000,
      "step": 146685
    },
    {
      "epoch": 16.13751375137514,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0032814346215976464,
      "loss": 0.2319,
      "num_input_tokens_seen": 30957088,
      "step": 146690
    },
    {
      "epoch": 16.13806380638064,
      "grad_norm": 0.000507354736328125,
      "learning_rate": 0.0032805357537563805,
      "loss": 0.2303,
      "num_input_tokens_seen": 30958176,
      "step": 146695
    },
    {
      "epoch": 16.138613861386137,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0032796369939277936,
      "loss": 0.2319,
      "num_input_tokens_seen": 30959200,
      "step": 146700
    },
    {
      "epoch": 16.139163916391638,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0032787383421201736,
      "loss": 0.2288,
      "num_input_tokens_seen": 30960256,
      "step": 146705
    },
    {
      "epoch": 16.13971397139714,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.003277839798341796,
      "loss": 0.2308,
      "num_input_tokens_seen": 30961248,
      "step": 146710
    },
    {
      "epoch": 16.14026402640264,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0032769413626009426,
      "loss": 0.2309,
      "num_input_tokens_seen": 30962336,
      "step": 146715
    },
    {
      "epoch": 16.14081408140814,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.003276043034905898,
      "loss": 0.2324,
      "num_input_tokens_seen": 30963392,
      "step": 146720
    },
    {
      "epoch": 16.141364136413642,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0032751448152649335,
      "loss": 0.233,
      "num_input_tokens_seen": 30964480,
      "step": 146725
    },
    {
      "epoch": 16.141914191419144,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.003274246703686335,
      "loss": 0.2303,
      "num_input_tokens_seen": 30965504,
      "step": 146730
    },
    {
      "epoch": 16.14246424642464,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.003273348700178383,
      "loss": 0.2319,
      "num_input_tokens_seen": 30966560,
      "step": 146735
    },
    {
      "epoch": 16.143014301430142,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.003272450804749343,
      "loss": 0.2324,
      "num_input_tokens_seen": 30967584,
      "step": 146740
    },
    {
      "epoch": 16.143564356435643,
      "grad_norm": 0.002593994140625,
      "learning_rate": 0.0032715530174075016,
      "loss": 0.2314,
      "num_input_tokens_seen": 30968704,
      "step": 146745
    },
    {
      "epoch": 16.144114411441144,
      "grad_norm": 0.00081634521484375,
      "learning_rate": 0.0032706553381611236,
      "loss": 0.2335,
      "num_input_tokens_seen": 30969760,
      "step": 146750
    },
    {
      "epoch": 16.144664466446645,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0032697577670184877,
      "loss": 0.2319,
      "num_input_tokens_seen": 30970752,
      "step": 146755
    },
    {
      "epoch": 16.145214521452147,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.003268860303987868,
      "loss": 0.2319,
      "num_input_tokens_seen": 30971776,
      "step": 146760
    },
    {
      "epoch": 16.145764576457644,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.003267962949077531,
      "loss": 0.2303,
      "num_input_tokens_seen": 30972736,
      "step": 146765
    },
    {
      "epoch": 16.146314631463145,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0032670657022957554,
      "loss": 0.2314,
      "num_input_tokens_seen": 30973824,
      "step": 146770
    },
    {
      "epoch": 16.146864686468646,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.003266168563650799,
      "loss": 0.2319,
      "num_input_tokens_seen": 30974912,
      "step": 146775
    },
    {
      "epoch": 16.147414741474147,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.0032652715331509424,
      "loss": 0.2329,
      "num_input_tokens_seen": 30975936,
      "step": 146780
    },
    {
      "epoch": 16.14796479647965,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0032643746108044418,
      "loss": 0.2308,
      "num_input_tokens_seen": 30976960,
      "step": 146785
    },
    {
      "epoch": 16.14851485148515,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.0032634777966195692,
      "loss": 0.2309,
      "num_input_tokens_seen": 30977984,
      "step": 146790
    },
    {
      "epoch": 16.14906490649065,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.0032625810906045946,
      "loss": 0.2309,
      "num_input_tokens_seen": 30979072,
      "step": 146795
    },
    {
      "epoch": 16.149614961496148,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0032616844927677736,
      "loss": 0.233,
      "num_input_tokens_seen": 30980128,
      "step": 146800
    },
    {
      "epoch": 16.15016501650165,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0032607880031173785,
      "loss": 0.2309,
      "num_input_tokens_seen": 30981248,
      "step": 146805
    },
    {
      "epoch": 16.15071507150715,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.003259891621661662,
      "loss": 0.2309,
      "num_input_tokens_seen": 30982304,
      "step": 146810
    },
    {
      "epoch": 16.15126512651265,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.003258995348408892,
      "loss": 0.2319,
      "num_input_tokens_seen": 30983392,
      "step": 146815
    },
    {
      "epoch": 16.151815181518153,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.0032580991833673317,
      "loss": 0.2314,
      "num_input_tokens_seen": 30984480,
      "step": 146820
    },
    {
      "epoch": 16.152365236523654,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0032572031265452332,
      "loss": 0.2304,
      "num_input_tokens_seen": 30985504,
      "step": 146825
    },
    {
      "epoch": 16.15291529152915,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.003256307177950863,
      "loss": 0.2324,
      "num_input_tokens_seen": 30986560,
      "step": 146830
    },
    {
      "epoch": 16.153465346534652,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0032554113375924737,
      "loss": 0.2303,
      "num_input_tokens_seen": 30987584,
      "step": 146835
    },
    {
      "epoch": 16.154015401540153,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003254515605478319,
      "loss": 0.2298,
      "num_input_tokens_seen": 30988640,
      "step": 146840
    },
    {
      "epoch": 16.154565456545654,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0032536199816166566,
      "loss": 0.2314,
      "num_input_tokens_seen": 30989664,
      "step": 146845
    },
    {
      "epoch": 16.155115511551156,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.0032527244660157445,
      "loss": 0.2303,
      "num_input_tokens_seen": 30990688,
      "step": 146850
    },
    {
      "epoch": 16.155665566556657,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.0032518290586838377,
      "loss": 0.2319,
      "num_input_tokens_seen": 30991808,
      "step": 146855
    },
    {
      "epoch": 16.156215621562158,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.003250933759629183,
      "loss": 0.2329,
      "num_input_tokens_seen": 30992832,
      "step": 146860
    },
    {
      "epoch": 16.156765676567655,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0032500385688600322,
      "loss": 0.2314,
      "num_input_tokens_seen": 30993792,
      "step": 146865
    },
    {
      "epoch": 16.157315731573156,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0032491434863846373,
      "loss": 0.2298,
      "num_input_tokens_seen": 30994816,
      "step": 146870
    },
    {
      "epoch": 16.157865786578657,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.003248248512211248,
      "loss": 0.2329,
      "num_input_tokens_seen": 30995872,
      "step": 146875
    },
    {
      "epoch": 16.15841584158416,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0032473536463481175,
      "loss": 0.2303,
      "num_input_tokens_seen": 30996896,
      "step": 146880
    },
    {
      "epoch": 16.15896589658966,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0032464588888034885,
      "loss": 0.2329,
      "num_input_tokens_seen": 30997984,
      "step": 146885
    },
    {
      "epoch": 16.15951595159516,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0032455642395856047,
      "loss": 0.2324,
      "num_input_tokens_seen": 30999072,
      "step": 146890
    },
    {
      "epoch": 16.16006600660066,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003244669698702719,
      "loss": 0.2314,
      "num_input_tokens_seen": 31000160,
      "step": 146895
    },
    {
      "epoch": 16.16061606160616,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003243775266163066,
      "loss": 0.2298,
      "num_input_tokens_seen": 31001216,
      "step": 146900
    },
    {
      "epoch": 16.16116611661166,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.003242880941974897,
      "loss": 0.2329,
      "num_input_tokens_seen": 31002240,
      "step": 146905
    },
    {
      "epoch": 16.16171617161716,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0032419867261464564,
      "loss": 0.2298,
      "num_input_tokens_seen": 31003296,
      "step": 146910
    },
    {
      "epoch": 16.162266226622663,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0032410926186859784,
      "loss": 0.2319,
      "num_input_tokens_seen": 31004448,
      "step": 146915
    },
    {
      "epoch": 16.162816281628164,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.00324019861960171,
      "loss": 0.2329,
      "num_input_tokens_seen": 31005440,
      "step": 146920
    },
    {
      "epoch": 16.163366336633665,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00323930472890189,
      "loss": 0.2324,
      "num_input_tokens_seen": 31006496,
      "step": 146925
    },
    {
      "epoch": 16.163916391639162,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.0032384109465947463,
      "loss": 0.2319,
      "num_input_tokens_seen": 31007520,
      "step": 146930
    },
    {
      "epoch": 16.164466446644663,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.0032375172726885324,
      "loss": 0.2303,
      "num_input_tokens_seen": 31008576,
      "step": 146935
    },
    {
      "epoch": 16.165016501650165,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.003236623707191474,
      "loss": 0.2308,
      "num_input_tokens_seen": 31009664,
      "step": 146940
    },
    {
      "epoch": 16.165566556655666,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0032357302501118132,
      "loss": 0.2314,
      "num_input_tokens_seen": 31010720,
      "step": 146945
    },
    {
      "epoch": 16.166116611661167,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.003234836901457783,
      "loss": 0.2314,
      "num_input_tokens_seen": 31011776,
      "step": 146950
    },
    {
      "epoch": 16.166666666666668,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0032339436612376115,
      "loss": 0.2308,
      "num_input_tokens_seen": 31012800,
      "step": 146955
    },
    {
      "epoch": 16.16721672167217,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.003233050529459535,
      "loss": 0.2298,
      "num_input_tokens_seen": 31013920,
      "step": 146960
    },
    {
      "epoch": 16.167766776677666,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0032321575061317856,
      "loss": 0.2298,
      "num_input_tokens_seen": 31014944,
      "step": 146965
    },
    {
      "epoch": 16.168316831683168,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.0032312645912625985,
      "loss": 0.2324,
      "num_input_tokens_seen": 31016032,
      "step": 146970
    },
    {
      "epoch": 16.16886688668867,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.003230371784860197,
      "loss": 0.2319,
      "num_input_tokens_seen": 31017120,
      "step": 146975
    },
    {
      "epoch": 16.16941694169417,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0032294790869328094,
      "loss": 0.2309,
      "num_input_tokens_seen": 31018176,
      "step": 146980
    },
    {
      "epoch": 16.16996699669967,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.0032285864974886673,
      "loss": 0.233,
      "num_input_tokens_seen": 31019232,
      "step": 146985
    },
    {
      "epoch": 16.170517051705172,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0032276940165359863,
      "loss": 0.2308,
      "num_input_tokens_seen": 31020320,
      "step": 146990
    },
    {
      "epoch": 16.17106710671067,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.003226801644083009,
      "loss": 0.2314,
      "num_input_tokens_seen": 31021408,
      "step": 146995
    },
    {
      "epoch": 16.17161716171617,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0032259093801379536,
      "loss": 0.2325,
      "num_input_tokens_seen": 31022432,
      "step": 147000
    },
    {
      "epoch": 16.17216721672167,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.003225017224709035,
      "loss": 0.2324,
      "num_input_tokens_seen": 31023520,
      "step": 147005
    },
    {
      "epoch": 16.172717271727173,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.003224125177804488,
      "loss": 0.2309,
      "num_input_tokens_seen": 31024576,
      "step": 147010
    },
    {
      "epoch": 16.173267326732674,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.003223233239432524,
      "loss": 0.2324,
      "num_input_tokens_seen": 31025632,
      "step": 147015
    },
    {
      "epoch": 16.173817381738175,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003222341409601368,
      "loss": 0.2319,
      "num_input_tokens_seen": 31026656,
      "step": 147020
    },
    {
      "epoch": 16.174367436743676,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003221449688319245,
      "loss": 0.2319,
      "num_input_tokens_seen": 31027680,
      "step": 147025
    },
    {
      "epoch": 16.174917491749174,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0032205580755943633,
      "loss": 0.2329,
      "num_input_tokens_seen": 31028800,
      "step": 147030
    },
    {
      "epoch": 16.175467546754675,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.00321966657143495,
      "loss": 0.2298,
      "num_input_tokens_seen": 31029824,
      "step": 147035
    },
    {
      "epoch": 16.176017601760176,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.003218775175849212,
      "loss": 0.2319,
      "num_input_tokens_seen": 31030848,
      "step": 147040
    },
    {
      "epoch": 16.176567656765677,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0032178838888453757,
      "loss": 0.2298,
      "num_input_tokens_seen": 31031936,
      "step": 147045
    },
    {
      "epoch": 16.177117711771178,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0032169927104316446,
      "loss": 0.2293,
      "num_input_tokens_seen": 31032992,
      "step": 147050
    },
    {
      "epoch": 16.17766776677668,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.0032161016406162387,
      "loss": 0.2303,
      "num_input_tokens_seen": 31034080,
      "step": 147055
    },
    {
      "epoch": 16.178217821782177,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.0032152106794073737,
      "loss": 0.2335,
      "num_input_tokens_seen": 31035136,
      "step": 147060
    },
    {
      "epoch": 16.178767876787678,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.003214319826813252,
      "loss": 0.2314,
      "num_input_tokens_seen": 31036160,
      "step": 147065
    },
    {
      "epoch": 16.17931793179318,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.003213429082842093,
      "loss": 0.2303,
      "num_input_tokens_seen": 31037184,
      "step": 147070
    },
    {
      "epoch": 16.17986798679868,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0032125384475020996,
      "loss": 0.2314,
      "num_input_tokens_seen": 31038240,
      "step": 147075
    },
    {
      "epoch": 16.18041804180418,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.0032116479208014817,
      "loss": 0.2299,
      "num_input_tokens_seen": 31039232,
      "step": 147080
    },
    {
      "epoch": 16.180968096809682,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.0032107575027484534,
      "loss": 0.2324,
      "num_input_tokens_seen": 31040288,
      "step": 147085
    },
    {
      "epoch": 16.181518151815183,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.003209867193351216,
      "loss": 0.2314,
      "num_input_tokens_seen": 31041344,
      "step": 147090
    },
    {
      "epoch": 16.18206820682068,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0032089769926179714,
      "loss": 0.2335,
      "num_input_tokens_seen": 31042432,
      "step": 147095
    },
    {
      "epoch": 16.182618261826182,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.0032080869005569317,
      "loss": 0.235,
      "num_input_tokens_seen": 31043456,
      "step": 147100
    },
    {
      "epoch": 16.183168316831683,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003207196917176288,
      "loss": 0.2314,
      "num_input_tokens_seen": 31044416,
      "step": 147105
    },
    {
      "epoch": 16.183718371837184,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.0032063070424842607,
      "loss": 0.2319,
      "num_input_tokens_seen": 31045472,
      "step": 147110
    },
    {
      "epoch": 16.184268426842685,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.003205417276489042,
      "loss": 0.2319,
      "num_input_tokens_seen": 31046560,
      "step": 147115
    },
    {
      "epoch": 16.184818481848186,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.003204527619198828,
      "loss": 0.2319,
      "num_input_tokens_seen": 31047616,
      "step": 147120
    },
    {
      "epoch": 16.185368536853684,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0032036380706218274,
      "loss": 0.2293,
      "num_input_tokens_seen": 31048640,
      "step": 147125
    },
    {
      "epoch": 16.185918591859185,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0032027486307662305,
      "loss": 0.2314,
      "num_input_tokens_seen": 31049728,
      "step": 147130
    },
    {
      "epoch": 16.186468646864686,
      "grad_norm": 0.005859375,
      "learning_rate": 0.003201859299640237,
      "loss": 0.2298,
      "num_input_tokens_seen": 31050784,
      "step": 147135
    },
    {
      "epoch": 16.187018701870187,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.003200970077252051,
      "loss": 0.2303,
      "num_input_tokens_seen": 31051872,
      "step": 147140
    },
    {
      "epoch": 16.187568756875688,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.003200080963609857,
      "loss": 0.2303,
      "num_input_tokens_seen": 31052960,
      "step": 147145
    },
    {
      "epoch": 16.18811881188119,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0031991919587218602,
      "loss": 0.2308,
      "num_input_tokens_seen": 31054016,
      "step": 147150
    },
    {
      "epoch": 16.18866886688669,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0031983030625962424,
      "loss": 0.2319,
      "num_input_tokens_seen": 31055104,
      "step": 147155
    },
    {
      "epoch": 16.189218921892188,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0031974142752412074,
      "loss": 0.2319,
      "num_input_tokens_seen": 31056160,
      "step": 147160
    },
    {
      "epoch": 16.18976897689769,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.003196525596664937,
      "loss": 0.2314,
      "num_input_tokens_seen": 31057248,
      "step": 147165
    },
    {
      "epoch": 16.19031903190319,
      "grad_norm": 0.005859375,
      "learning_rate": 0.003195637026875627,
      "loss": 0.2314,
      "num_input_tokens_seen": 31058272,
      "step": 147170
    },
    {
      "epoch": 16.19086908690869,
      "grad_norm": 0.000537872314453125,
      "learning_rate": 0.0031947485658814704,
      "loss": 0.233,
      "num_input_tokens_seen": 31059296,
      "step": 147175
    },
    {
      "epoch": 16.191419141914192,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.003193860213690647,
      "loss": 0.2314,
      "num_input_tokens_seen": 31060288,
      "step": 147180
    },
    {
      "epoch": 16.191969196919693,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.0031929719703113527,
      "loss": 0.2303,
      "num_input_tokens_seen": 31061376,
      "step": 147185
    },
    {
      "epoch": 16.19251925192519,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.003192083835751767,
      "loss": 0.2293,
      "num_input_tokens_seen": 31062400,
      "step": 147190
    },
    {
      "epoch": 16.193069306930692,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.0031911958100200783,
      "loss": 0.2319,
      "num_input_tokens_seen": 31063488,
      "step": 147195
    },
    {
      "epoch": 16.193619361936193,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.0031903078931244744,
      "loss": 0.2308,
      "num_input_tokens_seen": 31064608,
      "step": 147200
    },
    {
      "epoch": 16.194169416941694,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0031894200850731322,
      "loss": 0.2324,
      "num_input_tokens_seen": 31065728,
      "step": 147205
    },
    {
      "epoch": 16.194719471947195,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.0031885323858742423,
      "loss": 0.2319,
      "num_input_tokens_seen": 31066816,
      "step": 147210
    },
    {
      "epoch": 16.195269526952696,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0031876447955359816,
      "loss": 0.2303,
      "num_input_tokens_seen": 31067872,
      "step": 147215
    },
    {
      "epoch": 16.195819581958197,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.0031867573140665277,
      "loss": 0.2319,
      "num_input_tokens_seen": 31068928,
      "step": 147220
    },
    {
      "epoch": 16.196369636963695,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0031858699414740615,
      "loss": 0.2324,
      "num_input_tokens_seen": 31069984,
      "step": 147225
    },
    {
      "epoch": 16.196919691969196,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.0031849826777667644,
      "loss": 0.2324,
      "num_input_tokens_seen": 31071104,
      "step": 147230
    },
    {
      "epoch": 16.197469746974697,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.003184095522952815,
      "loss": 0.2314,
      "num_input_tokens_seen": 31072224,
      "step": 147235
    },
    {
      "epoch": 16.198019801980198,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.0031832084770403884,
      "loss": 0.2319,
      "num_input_tokens_seen": 31073216,
      "step": 147240
    },
    {
      "epoch": 16.1985698569857,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0031823215400376545,
      "loss": 0.2319,
      "num_input_tokens_seen": 31074272,
      "step": 147245
    },
    {
      "epoch": 16.1991199119912,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.003181434711952792,
      "loss": 0.2324,
      "num_input_tokens_seen": 31075296,
      "step": 147250
    },
    {
      "epoch": 16.199669966996698,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.0031805479927939765,
      "loss": 0.2293,
      "num_input_tokens_seen": 31076352,
      "step": 147255
    },
    {
      "epoch": 16.2002200220022,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003179661382569381,
      "loss": 0.2304,
      "num_input_tokens_seen": 31077408,
      "step": 147260
    },
    {
      "epoch": 16.2007700770077,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0031787748812871753,
      "loss": 0.2314,
      "num_input_tokens_seen": 31078528,
      "step": 147265
    },
    {
      "epoch": 16.2013201320132,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.0031778884889555247,
      "loss": 0.2288,
      "num_input_tokens_seen": 31079520,
      "step": 147270
    },
    {
      "epoch": 16.201870187018702,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0031770022055826064,
      "loss": 0.2319,
      "num_input_tokens_seen": 31080576,
      "step": 147275
    },
    {
      "epoch": 16.202420242024203,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0031761160311765835,
      "loss": 0.2335,
      "num_input_tokens_seen": 31081600,
      "step": 147280
    },
    {
      "epoch": 16.202970297029704,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0031752299657456244,
      "loss": 0.2319,
      "num_input_tokens_seen": 31082720,
      "step": 147285
    },
    {
      "epoch": 16.203520352035202,
      "grad_norm": 0.002716064453125,
      "learning_rate": 0.0031743440092978996,
      "loss": 0.2298,
      "num_input_tokens_seen": 31083776,
      "step": 147290
    },
    {
      "epoch": 16.204070407040703,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.003173458161841569,
      "loss": 0.2324,
      "num_input_tokens_seen": 31084832,
      "step": 147295
    },
    {
      "epoch": 16.204620462046204,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0031725724233848027,
      "loss": 0.2319,
      "num_input_tokens_seen": 31085888,
      "step": 147300
    },
    {
      "epoch": 16.205170517051705,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.003171686793935758,
      "loss": 0.2288,
      "num_input_tokens_seen": 31086976,
      "step": 147305
    },
    {
      "epoch": 16.205720572057206,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.003170801273502599,
      "loss": 0.2303,
      "num_input_tokens_seen": 31088064,
      "step": 147310
    },
    {
      "epoch": 16.206270627062707,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.0031699158620934935,
      "loss": 0.2309,
      "num_input_tokens_seen": 31089120,
      "step": 147315
    },
    {
      "epoch": 16.206820682068205,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0031690305597165927,
      "loss": 0.2319,
      "num_input_tokens_seen": 31090208,
      "step": 147320
    },
    {
      "epoch": 16.207370737073706,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.003168145366380064,
      "loss": 0.2298,
      "num_input_tokens_seen": 31091296,
      "step": 147325
    },
    {
      "epoch": 16.207920792079207,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.003167260282092062,
      "loss": 0.2314,
      "num_input_tokens_seen": 31092384,
      "step": 147330
    },
    {
      "epoch": 16.20847084708471,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0031663753068607403,
      "loss": 0.2304,
      "num_input_tokens_seen": 31093440,
      "step": 147335
    },
    {
      "epoch": 16.20902090209021,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.003165490440694259,
      "loss": 0.2309,
      "num_input_tokens_seen": 31094464,
      "step": 147340
    },
    {
      "epoch": 16.20957095709571,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.0031646056836007732,
      "loss": 0.2308,
      "num_input_tokens_seen": 31095520,
      "step": 147345
    },
    {
      "epoch": 16.21012101210121,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.003163721035588442,
      "loss": 0.2324,
      "num_input_tokens_seen": 31096512,
      "step": 147350
    },
    {
      "epoch": 16.21067106710671,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.0031628364966654146,
      "loss": 0.2309,
      "num_input_tokens_seen": 31097600,
      "step": 147355
    },
    {
      "epoch": 16.21122112211221,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.0031619520668398388,
      "loss": 0.2308,
      "num_input_tokens_seen": 31098624,
      "step": 147360
    },
    {
      "epoch": 16.21177117711771,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.0031610677461198704,
      "loss": 0.2335,
      "num_input_tokens_seen": 31099584,
      "step": 147365
    },
    {
      "epoch": 16.212321232123212,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0031601835345136603,
      "loss": 0.2314,
      "num_input_tokens_seen": 31100704,
      "step": 147370
    },
    {
      "epoch": 16.212871287128714,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0031592994320293597,
      "loss": 0.2314,
      "num_input_tokens_seen": 31101792,
      "step": 147375
    },
    {
      "epoch": 16.213421342134215,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.003158415438675116,
      "loss": 0.2304,
      "num_input_tokens_seen": 31102880,
      "step": 147380
    },
    {
      "epoch": 16.213971397139716,
      "grad_norm": 0.0026702880859375,
      "learning_rate": 0.003157531554459071,
      "loss": 0.2314,
      "num_input_tokens_seen": 31103904,
      "step": 147385
    },
    {
      "epoch": 16.214521452145213,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0031566477793893793,
      "loss": 0.2324,
      "num_input_tokens_seen": 31104992,
      "step": 147390
    },
    {
      "epoch": 16.215071507150714,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.0031557641134741764,
      "loss": 0.2314,
      "num_input_tokens_seen": 31106048,
      "step": 147395
    },
    {
      "epoch": 16.215621562156215,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0031548805567216142,
      "loss": 0.2314,
      "num_input_tokens_seen": 31107040,
      "step": 147400
    },
    {
      "epoch": 16.216171617161717,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0031539971091398378,
      "loss": 0.2304,
      "num_input_tokens_seen": 31108160,
      "step": 147405
    },
    {
      "epoch": 16.216721672167218,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.0031531137707369802,
      "loss": 0.2298,
      "num_input_tokens_seen": 31109184,
      "step": 147410
    },
    {
      "epoch": 16.21727172717272,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0031522305415211927,
      "loss": 0.2298,
      "num_input_tokens_seen": 31110272,
      "step": 147415
    },
    {
      "epoch": 16.217821782178216,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.0031513474215006075,
      "loss": 0.2314,
      "num_input_tokens_seen": 31111296,
      "step": 147420
    },
    {
      "epoch": 16.218371837183717,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.0031504644106833688,
      "loss": 0.2319,
      "num_input_tokens_seen": 31112384,
      "step": 147425
    },
    {
      "epoch": 16.21892189218922,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.0031495815090776167,
      "loss": 0.2309,
      "num_input_tokens_seen": 31113440,
      "step": 147430
    },
    {
      "epoch": 16.21947194719472,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0031486987166914815,
      "loss": 0.2319,
      "num_input_tokens_seen": 31114464,
      "step": 147435
    },
    {
      "epoch": 16.22002200220022,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0031478160335331076,
      "loss": 0.2324,
      "num_input_tokens_seen": 31115488,
      "step": 147440
    },
    {
      "epoch": 16.22057205720572,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.003146933459610621,
      "loss": 0.2309,
      "num_input_tokens_seen": 31116576,
      "step": 147445
    },
    {
      "epoch": 16.221122112211223,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0031460509949321667,
      "loss": 0.2314,
      "num_input_tokens_seen": 31117632,
      "step": 147450
    },
    {
      "epoch": 16.22167216721672,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0031451686395058665,
      "loss": 0.2314,
      "num_input_tokens_seen": 31118656,
      "step": 147455
    },
    {
      "epoch": 16.22222222222222,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.00314428639333986,
      "loss": 0.2314,
      "num_input_tokens_seen": 31119744,
      "step": 147460
    },
    {
      "epoch": 16.222772277227723,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0031434042564422796,
      "loss": 0.2309,
      "num_input_tokens_seen": 31120800,
      "step": 147465
    },
    {
      "epoch": 16.223322332233224,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0031425222288212506,
      "loss": 0.2314,
      "num_input_tokens_seen": 31121824,
      "step": 147470
    },
    {
      "epoch": 16.223872387238725,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0031416403104849075,
      "loss": 0.2293,
      "num_input_tokens_seen": 31122912,
      "step": 147475
    },
    {
      "epoch": 16.224422442244226,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.003140758501441371,
      "loss": 0.2293,
      "num_input_tokens_seen": 31123936,
      "step": 147480
    },
    {
      "epoch": 16.224972497249723,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.003139876801698773,
      "loss": 0.2309,
      "num_input_tokens_seen": 31124960,
      "step": 147485
    },
    {
      "epoch": 16.225522552255224,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0031389952112652435,
      "loss": 0.2314,
      "num_input_tokens_seen": 31126048,
      "step": 147490
    },
    {
      "epoch": 16.226072607260726,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003138113730148904,
      "loss": 0.2314,
      "num_input_tokens_seen": 31127104,
      "step": 147495
    },
    {
      "epoch": 16.226622662266227,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0031372323583578735,
      "loss": 0.2314,
      "num_input_tokens_seen": 31128192,
      "step": 147500
    },
    {
      "epoch": 16.227172717271728,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0031363510959002854,
      "loss": 0.2303,
      "num_input_tokens_seen": 31129248,
      "step": 147505
    },
    {
      "epoch": 16.22772277227723,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0031354699427842514,
      "loss": 0.233,
      "num_input_tokens_seen": 31130336,
      "step": 147510
    },
    {
      "epoch": 16.22827282728273,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.003134588899017898,
      "loss": 0.2335,
      "num_input_tokens_seen": 31131328,
      "step": 147515
    },
    {
      "epoch": 16.228822882288227,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0031337079646093493,
      "loss": 0.2304,
      "num_input_tokens_seen": 31132384,
      "step": 147520
    },
    {
      "epoch": 16.22937293729373,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.003132827139566717,
      "loss": 0.2314,
      "num_input_tokens_seen": 31133472,
      "step": 147525
    },
    {
      "epoch": 16.22992299229923,
      "grad_norm": 0.000667572021484375,
      "learning_rate": 0.0031319464238981245,
      "loss": 0.2319,
      "num_input_tokens_seen": 31134464,
      "step": 147530
    },
    {
      "epoch": 16.23047304730473,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0031310658176116855,
      "loss": 0.2325,
      "num_input_tokens_seen": 31135488,
      "step": 147535
    },
    {
      "epoch": 16.231023102310232,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0031301853207155164,
      "loss": 0.2309,
      "num_input_tokens_seen": 31136544,
      "step": 147540
    },
    {
      "epoch": 16.231573157315733,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0031293049332177377,
      "loss": 0.2335,
      "num_input_tokens_seen": 31137632,
      "step": 147545
    },
    {
      "epoch": 16.23212321232123,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0031284246551264547,
      "loss": 0.2314,
      "num_input_tokens_seen": 31138624,
      "step": 147550
    },
    {
      "epoch": 16.23267326732673,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.003127544486449789,
      "loss": 0.2329,
      "num_input_tokens_seen": 31139648,
      "step": 147555
    },
    {
      "epoch": 16.233223322332233,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.003126664427195846,
      "loss": 0.2309,
      "num_input_tokens_seen": 31140704,
      "step": 147560
    },
    {
      "epoch": 16.233773377337734,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.003125784477372742,
      "loss": 0.2324,
      "num_input_tokens_seen": 31141696,
      "step": 147565
    },
    {
      "epoch": 16.234323432343235,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.003124904636988583,
      "loss": 0.2319,
      "num_input_tokens_seen": 31142880,
      "step": 147570
    },
    {
      "epoch": 16.234873487348736,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.003124024906051477,
      "loss": 0.2351,
      "num_input_tokens_seen": 31143936,
      "step": 147575
    },
    {
      "epoch": 16.235423542354237,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0031231452845695406,
      "loss": 0.2314,
      "num_input_tokens_seen": 31145024,
      "step": 147580
    },
    {
      "epoch": 16.235973597359735,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0031222657725508713,
      "loss": 0.2324,
      "num_input_tokens_seen": 31146080,
      "step": 147585
    },
    {
      "epoch": 16.236523652365236,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.003121386370003582,
      "loss": 0.2314,
      "num_input_tokens_seen": 31147168,
      "step": 147590
    },
    {
      "epoch": 16.237073707370737,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00312050707693577,
      "loss": 0.2329,
      "num_input_tokens_seen": 31148192,
      "step": 147595
    },
    {
      "epoch": 16.237623762376238,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003119627893355545,
      "loss": 0.2314,
      "num_input_tokens_seen": 31149248,
      "step": 147600
    },
    {
      "epoch": 16.23817381738174,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.003118748819271013,
      "loss": 0.2309,
      "num_input_tokens_seen": 31150336,
      "step": 147605
    },
    {
      "epoch": 16.23872387238724,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003117869854690267,
      "loss": 0.2324,
      "num_input_tokens_seen": 31151424,
      "step": 147610
    },
    {
      "epoch": 16.239273927392738,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0031169909996214174,
      "loss": 0.2298,
      "num_input_tokens_seen": 31152448,
      "step": 147615
    },
    {
      "epoch": 16.23982398239824,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.0031161122540725606,
      "loss": 0.2329,
      "num_input_tokens_seen": 31153472,
      "step": 147620
    },
    {
      "epoch": 16.24037403740374,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00311523361805179,
      "loss": 0.2313,
      "num_input_tokens_seen": 31154464,
      "step": 147625
    },
    {
      "epoch": 16.24092409240924,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0031143550915672083,
      "loss": 0.2314,
      "num_input_tokens_seen": 31155520,
      "step": 147630
    },
    {
      "epoch": 16.241474147414742,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.003113476674626912,
      "loss": 0.2309,
      "num_input_tokens_seen": 31156608,
      "step": 147635
    },
    {
      "epoch": 16.242024202420243,
      "grad_norm": 0.005859375,
      "learning_rate": 0.003112598367239003,
      "loss": 0.2324,
      "num_input_tokens_seen": 31157632,
      "step": 147640
    },
    {
      "epoch": 16.242574257425744,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.003111720169411569,
      "loss": 0.2298,
      "num_input_tokens_seen": 31158624,
      "step": 147645
    },
    {
      "epoch": 16.24312431243124,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0031108420811527013,
      "loss": 0.2324,
      "num_input_tokens_seen": 31159616,
      "step": 147650
    },
    {
      "epoch": 16.243674367436743,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003109964102470498,
      "loss": 0.2314,
      "num_input_tokens_seen": 31160672,
      "step": 147655
    },
    {
      "epoch": 16.244224422442244,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.003109086233373051,
      "loss": 0.2319,
      "num_input_tokens_seen": 31161664,
      "step": 147660
    },
    {
      "epoch": 16.244774477447745,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.003108208473868454,
      "loss": 0.2314,
      "num_input_tokens_seen": 31162784,
      "step": 147665
    },
    {
      "epoch": 16.245324532453246,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0031073308239647933,
      "loss": 0.2309,
      "num_input_tokens_seen": 31163872,
      "step": 147670
    },
    {
      "epoch": 16.245874587458747,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.003106453283670153,
      "loss": 0.2325,
      "num_input_tokens_seen": 31164928,
      "step": 147675
    },
    {
      "epoch": 16.246424642464245,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0031055758529926294,
      "loss": 0.2304,
      "num_input_tokens_seen": 31165984,
      "step": 147680
    },
    {
      "epoch": 16.246974697469746,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0031046985319403034,
      "loss": 0.2309,
      "num_input_tokens_seen": 31167008,
      "step": 147685
    },
    {
      "epoch": 16.247524752475247,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0031038213205212615,
      "loss": 0.2283,
      "num_input_tokens_seen": 31168032,
      "step": 147690
    },
    {
      "epoch": 16.248074807480748,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0031029442187435956,
      "loss": 0.2298,
      "num_input_tokens_seen": 31169088,
      "step": 147695
    },
    {
      "epoch": 16.24862486248625,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.0031020672266153793,
      "loss": 0.2303,
      "num_input_tokens_seen": 31170080,
      "step": 147700
    },
    {
      "epoch": 16.24917491749175,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.0031011903441447043,
      "loss": 0.2319,
      "num_input_tokens_seen": 31171168,
      "step": 147705
    },
    {
      "epoch": 16.24972497249725,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.0031003135713396474,
      "loss": 0.2324,
      "num_input_tokens_seen": 31172288,
      "step": 147710
    },
    {
      "epoch": 16.25027502750275,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003099436908208282,
      "loss": 0.2319,
      "num_input_tokens_seen": 31173312,
      "step": 147715
    },
    {
      "epoch": 16.25082508250825,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.0030985603547587057,
      "loss": 0.2314,
      "num_input_tokens_seen": 31174368,
      "step": 147720
    },
    {
      "epoch": 16.25137513751375,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0030976839109989832,
      "loss": 0.2319,
      "num_input_tokens_seen": 31175392,
      "step": 147725
    },
    {
      "epoch": 16.251925192519252,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0030968075769372016,
      "loss": 0.2319,
      "num_input_tokens_seen": 31176448,
      "step": 147730
    },
    {
      "epoch": 16.252475247524753,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0030959313525814307,
      "loss": 0.2308,
      "num_input_tokens_seen": 31177536,
      "step": 147735
    },
    {
      "epoch": 16.253025302530254,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.003095055237939746,
      "loss": 0.2335,
      "num_input_tokens_seen": 31178624,
      "step": 147740
    },
    {
      "epoch": 16.253575357535752,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.0030941792330202245,
      "loss": 0.2308,
      "num_input_tokens_seen": 31179712,
      "step": 147745
    },
    {
      "epoch": 16.254125412541253,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.0030933033378309394,
      "loss": 0.2309,
      "num_input_tokens_seen": 31180864,
      "step": 147750
    },
    {
      "epoch": 16.254675467546754,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003092427552379969,
      "loss": 0.2314,
      "num_input_tokens_seen": 31181920,
      "step": 147755
    },
    {
      "epoch": 16.255225522552255,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.0030915518766753795,
      "loss": 0.2319,
      "num_input_tokens_seen": 31182944,
      "step": 147760
    },
    {
      "epoch": 16.255775577557756,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0030906763107252386,
      "loss": 0.2324,
      "num_input_tokens_seen": 31184000,
      "step": 147765
    },
    {
      "epoch": 16.256325632563257,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0030898008545376236,
      "loss": 0.234,
      "num_input_tokens_seen": 31185088,
      "step": 147770
    },
    {
      "epoch": 16.25687568756876,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0030889255081205896,
      "loss": 0.2308,
      "num_input_tokens_seen": 31186144,
      "step": 147775
    },
    {
      "epoch": 16.257425742574256,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.003088050271482223,
      "loss": 0.2293,
      "num_input_tokens_seen": 31187136,
      "step": 147780
    },
    {
      "epoch": 16.257975797579757,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.00308717514463058,
      "loss": 0.2303,
      "num_input_tokens_seen": 31188224,
      "step": 147785
    },
    {
      "epoch": 16.258525852585258,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0030863001275737247,
      "loss": 0.2303,
      "num_input_tokens_seen": 31189184,
      "step": 147790
    },
    {
      "epoch": 16.25907590759076,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0030854252203197283,
      "loss": 0.2324,
      "num_input_tokens_seen": 31190304,
      "step": 147795
    },
    {
      "epoch": 16.25962596259626,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.0030845504228766457,
      "loss": 0.2293,
      "num_input_tokens_seen": 31191296,
      "step": 147800
    },
    {
      "epoch": 16.26017601760176,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.003083675735252545,
      "loss": 0.2298,
      "num_input_tokens_seen": 31192352,
      "step": 147805
    },
    {
      "epoch": 16.260726072607262,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.00308280115745549,
      "loss": 0.2314,
      "num_input_tokens_seen": 31193408,
      "step": 147810
    },
    {
      "epoch": 16.26127612761276,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0030819266894935348,
      "loss": 0.2314,
      "num_input_tokens_seen": 31194464,
      "step": 147815
    },
    {
      "epoch": 16.26182618261826,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.003081052331374747,
      "loss": 0.2303,
      "num_input_tokens_seen": 31195488,
      "step": 147820
    },
    {
      "epoch": 16.262376237623762,
      "grad_norm": 0.000751495361328125,
      "learning_rate": 0.003080178083107176,
      "loss": 0.2319,
      "num_input_tokens_seen": 31196512,
      "step": 147825
    },
    {
      "epoch": 16.262926292629263,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0030793039446988884,
      "loss": 0.2293,
      "num_input_tokens_seen": 31197600,
      "step": 147830
    },
    {
      "epoch": 16.263476347634764,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.003078429916157931,
      "loss": 0.2335,
      "num_input_tokens_seen": 31198688,
      "step": 147835
    },
    {
      "epoch": 16.264026402640265,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0030775559974923652,
      "loss": 0.2283,
      "num_input_tokens_seen": 31199776,
      "step": 147840
    },
    {
      "epoch": 16.264576457645763,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.003076682188710251,
      "loss": 0.2303,
      "num_input_tokens_seen": 31200800,
      "step": 147845
    },
    {
      "epoch": 16.265126512651264,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0030758084898196286,
      "loss": 0.2319,
      "num_input_tokens_seen": 31201824,
      "step": 147850
    },
    {
      "epoch": 16.265676567656765,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.0030749349008285624,
      "loss": 0.2324,
      "num_input_tokens_seen": 31202880,
      "step": 147855
    },
    {
      "epoch": 16.266226622662266,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.003074061421745096,
      "loss": 0.233,
      "num_input_tokens_seen": 31203936,
      "step": 147860
    },
    {
      "epoch": 16.266776677667767,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.003073188052577281,
      "loss": 0.2308,
      "num_input_tokens_seen": 31204960,
      "step": 147865
    },
    {
      "epoch": 16.26732673267327,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0030723147933331752,
      "loss": 0.2304,
      "num_input_tokens_seen": 31205984,
      "step": 147870
    },
    {
      "epoch": 16.26787678767877,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.003071441644020819,
      "loss": 0.2329,
      "num_input_tokens_seen": 31207040,
      "step": 147875
    },
    {
      "epoch": 16.268426842684267,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.003070568604648257,
      "loss": 0.2324,
      "num_input_tokens_seen": 31208128,
      "step": 147880
    },
    {
      "epoch": 16.268976897689768,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.003069695675223544,
      "loss": 0.2304,
      "num_input_tokens_seen": 31209152,
      "step": 147885
    },
    {
      "epoch": 16.26952695269527,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.003068822855754714,
      "loss": 0.2324,
      "num_input_tokens_seen": 31210240,
      "step": 147890
    },
    {
      "epoch": 16.27007700770077,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.0030679501462498268,
      "loss": 0.2299,
      "num_input_tokens_seen": 31211360,
      "step": 147895
    },
    {
      "epoch": 16.27062706270627,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.003067077546716916,
      "loss": 0.2346,
      "num_input_tokens_seen": 31212384,
      "step": 147900
    },
    {
      "epoch": 16.271177117711773,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0030662050571640224,
      "loss": 0.2309,
      "num_input_tokens_seen": 31213408,
      "step": 147905
    },
    {
      "epoch": 16.27172717271727,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0030653326775991955,
      "loss": 0.2314,
      "num_input_tokens_seen": 31214496,
      "step": 147910
    },
    {
      "epoch": 16.27227722772277,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.003064460408030464,
      "loss": 0.2319,
      "num_input_tokens_seen": 31215616,
      "step": 147915
    },
    {
      "epoch": 16.272827282728272,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.0030635882484658754,
      "loss": 0.233,
      "num_input_tokens_seen": 31216672,
      "step": 147920
    },
    {
      "epoch": 16.273377337733773,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.003062716198913471,
      "loss": 0.2324,
      "num_input_tokens_seen": 31217760,
      "step": 147925
    },
    {
      "epoch": 16.273927392739274,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0030618442593812775,
      "loss": 0.2288,
      "num_input_tokens_seen": 31218816,
      "step": 147930
    },
    {
      "epoch": 16.274477447744776,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.003060972429877343,
      "loss": 0.2298,
      "num_input_tokens_seen": 31219936,
      "step": 147935
    },
    {
      "epoch": 16.275027502750277,
      "grad_norm": 0.000698089599609375,
      "learning_rate": 0.003060100710409691,
      "loss": 0.2325,
      "num_input_tokens_seen": 31220928,
      "step": 147940
    },
    {
      "epoch": 16.275577557755774,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0030592291009863665,
      "loss": 0.2314,
      "num_input_tokens_seen": 31221984,
      "step": 147945
    },
    {
      "epoch": 16.276127612761275,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.0030583576016153923,
      "loss": 0.2329,
      "num_input_tokens_seen": 31223008,
      "step": 147950
    },
    {
      "epoch": 16.276677667766776,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0030574862123048073,
      "loss": 0.2314,
      "num_input_tokens_seen": 31224000,
      "step": 147955
    },
    {
      "epoch": 16.277227722772277,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.0030566149330626432,
      "loss": 0.2303,
      "num_input_tokens_seen": 31225120,
      "step": 147960
    },
    {
      "epoch": 16.27777777777778,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.003055743763896926,
      "loss": 0.2309,
      "num_input_tokens_seen": 31226144,
      "step": 147965
    },
    {
      "epoch": 16.27832783278328,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003054872704815691,
      "loss": 0.2298,
      "num_input_tokens_seen": 31227200,
      "step": 147970
    },
    {
      "epoch": 16.278877887788777,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.003054001755826959,
      "loss": 0.2298,
      "num_input_tokens_seen": 31228256,
      "step": 147975
    },
    {
      "epoch": 16.27942794279428,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00305313091693876,
      "loss": 0.2324,
      "num_input_tokens_seen": 31229280,
      "step": 147980
    },
    {
      "epoch": 16.27997799779978,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0030522601881591248,
      "loss": 0.2298,
      "num_input_tokens_seen": 31230272,
      "step": 147985
    },
    {
      "epoch": 16.28052805280528,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003051389569496069,
      "loss": 0.2309,
      "num_input_tokens_seen": 31231264,
      "step": 147990
    },
    {
      "epoch": 16.28107810781078,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0030505190609576276,
      "loss": 0.2304,
      "num_input_tokens_seen": 31232352,
      "step": 147995
    },
    {
      "epoch": 16.281628162816283,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.0030496486625518175,
      "loss": 0.2319,
      "num_input_tokens_seen": 31233408,
      "step": 148000
    },
    {
      "epoch": 16.282178217821784,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.0030487783742866575,
      "loss": 0.2319,
      "num_input_tokens_seen": 31234464,
      "step": 148005
    },
    {
      "epoch": 16.28272827282728,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.0030479081961701723,
      "loss": 0.2324,
      "num_input_tokens_seen": 31235520,
      "step": 148010
    },
    {
      "epoch": 16.283278327832782,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.003047038128210383,
      "loss": 0.2319,
      "num_input_tokens_seen": 31236576,
      "step": 148015
    },
    {
      "epoch": 16.283828382838283,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.003046168170415311,
      "loss": 0.2309,
      "num_input_tokens_seen": 31237632,
      "step": 148020
    },
    {
      "epoch": 16.284378437843785,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0030452983227929713,
      "loss": 0.2293,
      "num_input_tokens_seen": 31238688,
      "step": 148025
    },
    {
      "epoch": 16.284928492849286,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0030444285853513764,
      "loss": 0.2319,
      "num_input_tokens_seen": 31239712,
      "step": 148030
    },
    {
      "epoch": 16.285478547854787,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.003043558958098546,
      "loss": 0.2324,
      "num_input_tokens_seen": 31240736,
      "step": 148035
    },
    {
      "epoch": 16.286028602860284,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.003042689441042494,
      "loss": 0.2324,
      "num_input_tokens_seen": 31241728,
      "step": 148040
    },
    {
      "epoch": 16.286578657865785,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0030418200341912425,
      "loss": 0.2304,
      "num_input_tokens_seen": 31242752,
      "step": 148045
    },
    {
      "epoch": 16.287128712871286,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0030409507375527963,
      "loss": 0.2319,
      "num_input_tokens_seen": 31243808,
      "step": 148050
    },
    {
      "epoch": 16.287678767876788,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0030400815511351644,
      "loss": 0.2319,
      "num_input_tokens_seen": 31244896,
      "step": 148055
    },
    {
      "epoch": 16.28822882288229,
      "grad_norm": 0.0006866455078125,
      "learning_rate": 0.0030392124749463673,
      "loss": 0.2298,
      "num_input_tokens_seen": 31245920,
      "step": 148060
    },
    {
      "epoch": 16.28877887788779,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.003038343508994404,
      "loss": 0.2303,
      "num_input_tokens_seen": 31246944,
      "step": 148065
    },
    {
      "epoch": 16.28932893289329,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00303747465328729,
      "loss": 0.2324,
      "num_input_tokens_seen": 31248032,
      "step": 148070
    },
    {
      "epoch": 16.28987898789879,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0030366059078330335,
      "loss": 0.2314,
      "num_input_tokens_seen": 31249120,
      "step": 148075
    },
    {
      "epoch": 16.29042904290429,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.003035737272639638,
      "loss": 0.2319,
      "num_input_tokens_seen": 31250176,
      "step": 148080
    },
    {
      "epoch": 16.29097909790979,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0030348687477151147,
      "loss": 0.2324,
      "num_input_tokens_seen": 31251264,
      "step": 148085
    },
    {
      "epoch": 16.29152915291529,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0030340003330674606,
      "loss": 0.2314,
      "num_input_tokens_seen": 31252352,
      "step": 148090
    },
    {
      "epoch": 16.292079207920793,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.003033132028704683,
      "loss": 0.2303,
      "num_input_tokens_seen": 31253408,
      "step": 148095
    },
    {
      "epoch": 16.292629262926294,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00303226383463479,
      "loss": 0.2314,
      "num_input_tokens_seen": 31254464,
      "step": 148100
    },
    {
      "epoch": 16.293179317931795,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.0030313957508657755,
      "loss": 0.2324,
      "num_input_tokens_seen": 31255456,
      "step": 148105
    },
    {
      "epoch": 16.293729372937293,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.003030527777405645,
      "loss": 0.2319,
      "num_input_tokens_seen": 31256480,
      "step": 148110
    },
    {
      "epoch": 16.294279427942794,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.003029659914262398,
      "loss": 0.2319,
      "num_input_tokens_seen": 31257536,
      "step": 148115
    },
    {
      "epoch": 16.294829482948295,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.003028792161444027,
      "loss": 0.2303,
      "num_input_tokens_seen": 31258592,
      "step": 148120
    },
    {
      "epoch": 16.295379537953796,
      "grad_norm": 0.005859375,
      "learning_rate": 0.003027924518958536,
      "loss": 0.2319,
      "num_input_tokens_seen": 31259648,
      "step": 148125
    },
    {
      "epoch": 16.295929592959297,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.0030270569868139172,
      "loss": 0.2303,
      "num_input_tokens_seen": 31260704,
      "step": 148130
    },
    {
      "epoch": 16.296479647964798,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.003026189565018174,
      "loss": 0.2324,
      "num_input_tokens_seen": 31261760,
      "step": 148135
    },
    {
      "epoch": 16.297029702970296,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.003025322253579296,
      "loss": 0.2329,
      "num_input_tokens_seen": 31262816,
      "step": 148140
    },
    {
      "epoch": 16.297579757975797,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.0030244550525052737,
      "loss": 0.2319,
      "num_input_tokens_seen": 31263808,
      "step": 148145
    },
    {
      "epoch": 16.298129812981298,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.003023587961804102,
      "loss": 0.2319,
      "num_input_tokens_seen": 31264896,
      "step": 148150
    },
    {
      "epoch": 16.2986798679868,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0030227209814837736,
      "loss": 0.2309,
      "num_input_tokens_seen": 31265984,
      "step": 148155
    },
    {
      "epoch": 16.2992299229923,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.0030218541115522826,
      "loss": 0.2309,
      "num_input_tokens_seen": 31267040,
      "step": 148160
    },
    {
      "epoch": 16.2997799779978,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.003020987352017614,
      "loss": 0.2314,
      "num_input_tokens_seen": 31268064,
      "step": 148165
    },
    {
      "epoch": 16.300330033003302,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0030201207028877535,
      "loss": 0.2313,
      "num_input_tokens_seen": 31269120,
      "step": 148170
    },
    {
      "epoch": 16.3008800880088,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0030192541641706956,
      "loss": 0.2324,
      "num_input_tokens_seen": 31270080,
      "step": 148175
    },
    {
      "epoch": 16.3014301430143,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.0030183877358744182,
      "loss": 0.2319,
      "num_input_tokens_seen": 31271168,
      "step": 148180
    },
    {
      "epoch": 16.301980198019802,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.003017521418006914,
      "loss": 0.2314,
      "num_input_tokens_seen": 31272192,
      "step": 148185
    },
    {
      "epoch": 16.302530253025303,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.003016655210576167,
      "loss": 0.2314,
      "num_input_tokens_seen": 31273248,
      "step": 148190
    },
    {
      "epoch": 16.303080308030804,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.003015789113590156,
      "loss": 0.2324,
      "num_input_tokens_seen": 31274304,
      "step": 148195
    },
    {
      "epoch": 16.303630363036305,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.00301492312705687,
      "loss": 0.2314,
      "num_input_tokens_seen": 31275328,
      "step": 148200
    },
    {
      "epoch": 16.304180418041803,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0030140572509842825,
      "loss": 0.2319,
      "num_input_tokens_seen": 31276352,
      "step": 148205
    },
    {
      "epoch": 16.304730473047304,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.003013191485380376,
      "loss": 0.2309,
      "num_input_tokens_seen": 31277440,
      "step": 148210
    },
    {
      "epoch": 16.305280528052805,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.003012325830253138,
      "loss": 0.2298,
      "num_input_tokens_seen": 31278464,
      "step": 148215
    },
    {
      "epoch": 16.305830583058306,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0030114602856105375,
      "loss": 0.233,
      "num_input_tokens_seen": 31279488,
      "step": 148220
    },
    {
      "epoch": 16.306380638063807,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.003010594851460557,
      "loss": 0.2319,
      "num_input_tokens_seen": 31280576,
      "step": 148225
    },
    {
      "epoch": 16.306930693069308,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0030097295278111672,
      "loss": 0.2303,
      "num_input_tokens_seen": 31281664,
      "step": 148230
    },
    {
      "epoch": 16.30748074807481,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0030088643146703497,
      "loss": 0.2298,
      "num_input_tokens_seen": 31282816,
      "step": 148235
    },
    {
      "epoch": 16.308030803080307,
      "grad_norm": 0.000652313232421875,
      "learning_rate": 0.003007999212046073,
      "loss": 0.2309,
      "num_input_tokens_seen": 31283872,
      "step": 148240
    },
    {
      "epoch": 16.308580858085808,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.003007134219946314,
      "loss": 0.2329,
      "num_input_tokens_seen": 31284896,
      "step": 148245
    },
    {
      "epoch": 16.30913091309131,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.003006269338379047,
      "loss": 0.2293,
      "num_input_tokens_seen": 31285984,
      "step": 148250
    },
    {
      "epoch": 16.30968096809681,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.003005404567352241,
      "loss": 0.233,
      "num_input_tokens_seen": 31287008,
      "step": 148255
    },
    {
      "epoch": 16.31023102310231,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.003004539906873861,
      "loss": 0.2298,
      "num_input_tokens_seen": 31288032,
      "step": 148260
    },
    {
      "epoch": 16.310781078107812,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.003003675356951881,
      "loss": 0.2309,
      "num_input_tokens_seen": 31289088,
      "step": 148265
    },
    {
      "epoch": 16.31133113311331,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0030028109175942695,
      "loss": 0.2314,
      "num_input_tokens_seen": 31290208,
      "step": 148270
    },
    {
      "epoch": 16.31188118811881,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0030019465888089956,
      "loss": 0.2324,
      "num_input_tokens_seen": 31291168,
      "step": 148275
    },
    {
      "epoch": 16.312431243124312,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0030010823706040236,
      "loss": 0.2309,
      "num_input_tokens_seen": 31292256,
      "step": 148280
    },
    {
      "epoch": 16.312981298129813,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.003000218262987313,
      "loss": 0.2324,
      "num_input_tokens_seen": 31293312,
      "step": 148285
    },
    {
      "epoch": 16.313531353135314,
      "grad_norm": 0.005859375,
      "learning_rate": 0.002999354265966838,
      "loss": 0.233,
      "num_input_tokens_seen": 31294432,
      "step": 148290
    },
    {
      "epoch": 16.314081408140815,
      "grad_norm": 0.00518798828125,
      "learning_rate": 0.0029984903795505506,
      "loss": 0.2268,
      "num_input_tokens_seen": 31295456,
      "step": 148295
    },
    {
      "epoch": 16.314631463146316,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0029976266037464194,
      "loss": 0.2293,
      "num_input_tokens_seen": 31296512,
      "step": 148300
    },
    {
      "epoch": 16.315181518151814,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0029967629385624093,
      "loss": 0.2329,
      "num_input_tokens_seen": 31297600,
      "step": 148305
    },
    {
      "epoch": 16.315731573157315,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0029958993840064714,
      "loss": 0.2298,
      "num_input_tokens_seen": 31298624,
      "step": 148310
    },
    {
      "epoch": 16.316281628162816,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0029950359400865716,
      "loss": 0.2309,
      "num_input_tokens_seen": 31299744,
      "step": 148315
    },
    {
      "epoch": 16.316831683168317,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.002994172606810661,
      "loss": 0.2298,
      "num_input_tokens_seen": 31300832,
      "step": 148320
    },
    {
      "epoch": 16.317381738173818,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.0029933093841867015,
      "loss": 0.2293,
      "num_input_tokens_seen": 31301856,
      "step": 148325
    },
    {
      "epoch": 16.31793179317932,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.0029924462722226513,
      "loss": 0.2324,
      "num_input_tokens_seen": 31302976,
      "step": 148330
    },
    {
      "epoch": 16.318481848184817,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002991583270926458,
      "loss": 0.2314,
      "num_input_tokens_seen": 31303968,
      "step": 148335
    },
    {
      "epoch": 16.319031903190318,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0029907203803060827,
      "loss": 0.2324,
      "num_input_tokens_seen": 31304960,
      "step": 148340
    },
    {
      "epoch": 16.31958195819582,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.002989857600369473,
      "loss": 0.2324,
      "num_input_tokens_seen": 31306016,
      "step": 148345
    },
    {
      "epoch": 16.32013201320132,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.002988994931124586,
      "loss": 0.2298,
      "num_input_tokens_seen": 31307072,
      "step": 148350
    },
    {
      "epoch": 16.32068206820682,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002988132372579365,
      "loss": 0.2308,
      "num_input_tokens_seen": 31308160,
      "step": 148355
    },
    {
      "epoch": 16.321232123212322,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.002987269924741764,
      "loss": 0.2293,
      "num_input_tokens_seen": 31309216,
      "step": 148360
    },
    {
      "epoch": 16.321782178217823,
      "grad_norm": 0.0026702880859375,
      "learning_rate": 0.002986407587619736,
      "loss": 0.2314,
      "num_input_tokens_seen": 31310304,
      "step": 148365
    },
    {
      "epoch": 16.32233223322332,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00298554536122122,
      "loss": 0.2319,
      "num_input_tokens_seen": 31311424,
      "step": 148370
    },
    {
      "epoch": 16.322882288228822,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0029846832455541717,
      "loss": 0.2335,
      "num_input_tokens_seen": 31312448,
      "step": 148375
    },
    {
      "epoch": 16.323432343234323,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.002983821240626529,
      "loss": 0.2319,
      "num_input_tokens_seen": 31313440,
      "step": 148380
    },
    {
      "epoch": 16.323982398239824,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.002982959346446239,
      "loss": 0.2314,
      "num_input_tokens_seen": 31314464,
      "step": 148385
    },
    {
      "epoch": 16.324532453245325,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0029820975630212515,
      "loss": 0.2319,
      "num_input_tokens_seen": 31315488,
      "step": 148390
    },
    {
      "epoch": 16.325082508250826,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.002981235890359499,
      "loss": 0.2324,
      "num_input_tokens_seen": 31316512,
      "step": 148395
    },
    {
      "epoch": 16.325632563256324,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.002980374328468933,
      "loss": 0.2319,
      "num_input_tokens_seen": 31317536,
      "step": 148400
    },
    {
      "epoch": 16.326182618261825,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.0029795128773574884,
      "loss": 0.2329,
      "num_input_tokens_seen": 31318560,
      "step": 148405
    },
    {
      "epoch": 16.326732673267326,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0029786515370331017,
      "loss": 0.2288,
      "num_input_tokens_seen": 31319552,
      "step": 148410
    },
    {
      "epoch": 16.327282728272827,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.0029777903075037166,
      "loss": 0.2314,
      "num_input_tokens_seen": 31320576,
      "step": 148415
    },
    {
      "epoch": 16.32783278327833,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.002976929188777268,
      "loss": 0.2356,
      "num_input_tokens_seen": 31321728,
      "step": 148420
    },
    {
      "epoch": 16.32838283828383,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.002976068180861698,
      "loss": 0.2324,
      "num_input_tokens_seen": 31322784,
      "step": 148425
    },
    {
      "epoch": 16.32893289328933,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002975207283764938,
      "loss": 0.2308,
      "num_input_tokens_seen": 31323808,
      "step": 148430
    },
    {
      "epoch": 16.329482948294828,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0029743464974949186,
      "loss": 0.2319,
      "num_input_tokens_seen": 31324928,
      "step": 148435
    },
    {
      "epoch": 16.33003300330033,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002973485822059577,
      "loss": 0.2303,
      "num_input_tokens_seen": 31325920,
      "step": 148440
    },
    {
      "epoch": 16.33058305830583,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002972625257466851,
      "loss": 0.233,
      "num_input_tokens_seen": 31327040,
      "step": 148445
    },
    {
      "epoch": 16.33113311331133,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.002971764803724662,
      "loss": 0.2308,
      "num_input_tokens_seen": 31328096,
      "step": 148450
    },
    {
      "epoch": 16.331683168316832,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.002970904460840948,
      "loss": 0.2319,
      "num_input_tokens_seen": 31329152,
      "step": 148455
    },
    {
      "epoch": 16.332233223322334,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0029700442288236326,
      "loss": 0.2304,
      "num_input_tokens_seen": 31330208,
      "step": 148460
    },
    {
      "epoch": 16.33278327832783,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.00296918410768065,
      "loss": 0.2314,
      "num_input_tokens_seen": 31331232,
      "step": 148465
    },
    {
      "epoch": 16.333333333333332,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0029683240974199207,
      "loss": 0.2298,
      "num_input_tokens_seen": 31332352,
      "step": 148470
    },
    {
      "epoch": 16.333883388338833,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.0029674641980493755,
      "loss": 0.2324,
      "num_input_tokens_seen": 31333440,
      "step": 148475
    },
    {
      "epoch": 16.334433443344334,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.002966604409576942,
      "loss": 0.2319,
      "num_input_tokens_seen": 31334496,
      "step": 148480
    },
    {
      "epoch": 16.334983498349835,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0029657447320105383,
      "loss": 0.2288,
      "num_input_tokens_seen": 31335520,
      "step": 148485
    },
    {
      "epoch": 16.335533553355337,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0029648851653580957,
      "loss": 0.2324,
      "num_input_tokens_seen": 31336608,
      "step": 148490
    },
    {
      "epoch": 16.336083608360838,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0029640257096275303,
      "loss": 0.2335,
      "num_input_tokens_seen": 31337664,
      "step": 148495
    },
    {
      "epoch": 16.336633663366335,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.002963166364826756,
      "loss": 0.2324,
      "num_input_tokens_seen": 31338720,
      "step": 148500
    },
    {
      "epoch": 16.337183718371836,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.0029623071309637116,
      "loss": 0.2314,
      "num_input_tokens_seen": 31339776,
      "step": 148505
    },
    {
      "epoch": 16.337733773377337,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0029614480080462995,
      "loss": 0.2314,
      "num_input_tokens_seen": 31340896,
      "step": 148510
    },
    {
      "epoch": 16.33828382838284,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0029605889960824506,
      "loss": 0.2319,
      "num_input_tokens_seen": 31342016,
      "step": 148515
    },
    {
      "epoch": 16.33883388338834,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0029597300950800747,
      "loss": 0.2329,
      "num_input_tokens_seen": 31343104,
      "step": 148520
    },
    {
      "epoch": 16.33938393839384,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0029588713050470845,
      "loss": 0.2304,
      "num_input_tokens_seen": 31344160,
      "step": 148525
    },
    {
      "epoch": 16.33993399339934,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.0029580126259914014,
      "loss": 0.2319,
      "num_input_tokens_seen": 31345184,
      "step": 148530
    },
    {
      "epoch": 16.34048404840484,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.002957154057920936,
      "loss": 0.2324,
      "num_input_tokens_seen": 31346304,
      "step": 148535
    },
    {
      "epoch": 16.34103410341034,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002956295600843607,
      "loss": 0.2303,
      "num_input_tokens_seen": 31347360,
      "step": 148540
    },
    {
      "epoch": 16.34158415841584,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0029554372547673227,
      "loss": 0.2298,
      "num_input_tokens_seen": 31348384,
      "step": 148545
    },
    {
      "epoch": 16.342134213421343,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.002954579019699991,
      "loss": 0.2309,
      "num_input_tokens_seen": 31349440,
      "step": 148550
    },
    {
      "epoch": 16.342684268426844,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0029537208956495263,
      "loss": 0.2314,
      "num_input_tokens_seen": 31350496,
      "step": 148555
    },
    {
      "epoch": 16.343234323432345,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.0029528628826238296,
      "loss": 0.2303,
      "num_input_tokens_seen": 31351552,
      "step": 148560
    },
    {
      "epoch": 16.343784378437842,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0029520049806308225,
      "loss": 0.2303,
      "num_input_tokens_seen": 31352640,
      "step": 148565
    },
    {
      "epoch": 16.344334433443343,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0029511471896784035,
      "loss": 0.233,
      "num_input_tokens_seen": 31353696,
      "step": 148570
    },
    {
      "epoch": 16.344884488448844,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0029502895097744753,
      "loss": 0.2324,
      "num_input_tokens_seen": 31354688,
      "step": 148575
    },
    {
      "epoch": 16.345434543454346,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0029494319409269507,
      "loss": 0.2319,
      "num_input_tokens_seen": 31355776,
      "step": 148580
    },
    {
      "epoch": 16.345984598459847,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.002948574483143726,
      "loss": 0.2319,
      "num_input_tokens_seen": 31356800,
      "step": 148585
    },
    {
      "epoch": 16.346534653465348,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0029477171364327075,
      "loss": 0.2304,
      "num_input_tokens_seen": 31357824,
      "step": 148590
    },
    {
      "epoch": 16.34708470847085,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002946859900801801,
      "loss": 0.2309,
      "num_input_tokens_seen": 31358912,
      "step": 148595
    },
    {
      "epoch": 16.347634763476346,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.002946002776258898,
      "loss": 0.2309,
      "num_input_tokens_seen": 31359968,
      "step": 148600
    },
    {
      "epoch": 16.348184818481847,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0029451457628119075,
      "loss": 0.2329,
      "num_input_tokens_seen": 31361056,
      "step": 148605
    },
    {
      "epoch": 16.34873487348735,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0029442888604687216,
      "loss": 0.2309,
      "num_input_tokens_seen": 31362144,
      "step": 148610
    },
    {
      "epoch": 16.34928492849285,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0029434320692372437,
      "loss": 0.2303,
      "num_input_tokens_seen": 31363168,
      "step": 148615
    },
    {
      "epoch": 16.34983498349835,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.002942575389125363,
      "loss": 0.2309,
      "num_input_tokens_seen": 31364224,
      "step": 148620
    },
    {
      "epoch": 16.350385038503852,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0029417188201409793,
      "loss": 0.233,
      "num_input_tokens_seen": 31365216,
      "step": 148625
    },
    {
      "epoch": 16.35093509350935,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0029408623622919896,
      "loss": 0.2309,
      "num_input_tokens_seen": 31366240,
      "step": 148630
    },
    {
      "epoch": 16.35148514851485,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.002940006015586287,
      "loss": 0.2293,
      "num_input_tokens_seen": 31367328,
      "step": 148635
    },
    {
      "epoch": 16.35203520352035,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0029391497800317565,
      "loss": 0.2309,
      "num_input_tokens_seen": 31368352,
      "step": 148640
    },
    {
      "epoch": 16.352585258525853,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.002938293655636297,
      "loss": 0.233,
      "num_input_tokens_seen": 31369408,
      "step": 148645
    },
    {
      "epoch": 16.353135313531354,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0029374376424077947,
      "loss": 0.2304,
      "num_input_tokens_seen": 31370528,
      "step": 148650
    },
    {
      "epoch": 16.353685368536855,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0029365817403541484,
      "loss": 0.2304,
      "num_input_tokens_seen": 31371648,
      "step": 148655
    },
    {
      "epoch": 16.354235423542356,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0029357259494832365,
      "loss": 0.2298,
      "num_input_tokens_seen": 31372672,
      "step": 148660
    },
    {
      "epoch": 16.354785478547853,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0029348702698029482,
      "loss": 0.2298,
      "num_input_tokens_seen": 31373728,
      "step": 148665
    },
    {
      "epoch": 16.355335533553355,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0029340147013211734,
      "loss": 0.2319,
      "num_input_tokens_seen": 31374816,
      "step": 148670
    },
    {
      "epoch": 16.355885588558856,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0029331592440457912,
      "loss": 0.2324,
      "num_input_tokens_seen": 31375872,
      "step": 148675
    },
    {
      "epoch": 16.356435643564357,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.002932303897984692,
      "loss": 0.2314,
      "num_input_tokens_seen": 31376960,
      "step": 148680
    },
    {
      "epoch": 16.356985698569858,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0029314486631457594,
      "loss": 0.2303,
      "num_input_tokens_seen": 31378048,
      "step": 148685
    },
    {
      "epoch": 16.35753575357536,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0029305935395368696,
      "loss": 0.2324,
      "num_input_tokens_seen": 31379168,
      "step": 148690
    },
    {
      "epoch": 16.358085808580856,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.0029297385271659124,
      "loss": 0.2319,
      "num_input_tokens_seen": 31380192,
      "step": 148695
    },
    {
      "epoch": 16.358635863586358,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.002928883626040758,
      "loss": 0.2309,
      "num_input_tokens_seen": 31381312,
      "step": 148700
    },
    {
      "epoch": 16.35918591859186,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0029280288361692914,
      "loss": 0.2293,
      "num_input_tokens_seen": 31382368,
      "step": 148705
    },
    {
      "epoch": 16.35973597359736,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0029271741575593946,
      "loss": 0.2314,
      "num_input_tokens_seen": 31383424,
      "step": 148710
    },
    {
      "epoch": 16.36028602860286,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0029263195902189364,
      "loss": 0.2314,
      "num_input_tokens_seen": 31384512,
      "step": 148715
    },
    {
      "epoch": 16.360836083608362,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0029254651341558023,
      "loss": 0.2309,
      "num_input_tokens_seen": 31385536,
      "step": 148720
    },
    {
      "epoch": 16.361386138613863,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.002924610789377856,
      "loss": 0.2298,
      "num_input_tokens_seen": 31386656,
      "step": 148725
    },
    {
      "epoch": 16.36193619361936,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0029237565558929827,
      "loss": 0.2324,
      "num_input_tokens_seen": 31387744,
      "step": 148730
    },
    {
      "epoch": 16.36248624862486,
      "grad_norm": 0.0003452301025390625,
      "learning_rate": 0.002922902433709048,
      "loss": 0.2309,
      "num_input_tokens_seen": 31388736,
      "step": 148735
    },
    {
      "epoch": 16.363036303630363,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.0029220484228339237,
      "loss": 0.2314,
      "num_input_tokens_seen": 31389760,
      "step": 148740
    },
    {
      "epoch": 16.363586358635864,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.0029211945232754887,
      "loss": 0.2314,
      "num_input_tokens_seen": 31390816,
      "step": 148745
    },
    {
      "epoch": 16.364136413641365,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.002920340735041605,
      "loss": 0.2303,
      "num_input_tokens_seen": 31391872,
      "step": 148750
    },
    {
      "epoch": 16.364686468646866,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002919487058140146,
      "loss": 0.2308,
      "num_input_tokens_seen": 31392928,
      "step": 148755
    },
    {
      "epoch": 16.365236523652364,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0029186334925789758,
      "loss": 0.2304,
      "num_input_tokens_seen": 31393952,
      "step": 148760
    },
    {
      "epoch": 16.365786578657865,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0029177800383659628,
      "loss": 0.2314,
      "num_input_tokens_seen": 31395008,
      "step": 148765
    },
    {
      "epoch": 16.366336633663366,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.002916926695508978,
      "loss": 0.2319,
      "num_input_tokens_seen": 31396064,
      "step": 148770
    },
    {
      "epoch": 16.366886688668867,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0029160734640158773,
      "loss": 0.2293,
      "num_input_tokens_seen": 31397152,
      "step": 148775
    },
    {
      "epoch": 16.367436743674368,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.002915220343894534,
      "loss": 0.2329,
      "num_input_tokens_seen": 31398208,
      "step": 148780
    },
    {
      "epoch": 16.36798679867987,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.002914367335152804,
      "loss": 0.2314,
      "num_input_tokens_seen": 31399296,
      "step": 148785
    },
    {
      "epoch": 16.36853685368537,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.002913514437798549,
      "loss": 0.2298,
      "num_input_tokens_seen": 31400352,
      "step": 148790
    },
    {
      "epoch": 16.369086908690868,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0029126616518396313,
      "loss": 0.2308,
      "num_input_tokens_seen": 31401408,
      "step": 148795
    },
    {
      "epoch": 16.36963696369637,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.0029118089772839106,
      "loss": 0.2303,
      "num_input_tokens_seen": 31402592,
      "step": 148800
    },
    {
      "epoch": 16.37018701870187,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00291095641413925,
      "loss": 0.2324,
      "num_input_tokens_seen": 31403552,
      "step": 148805
    },
    {
      "epoch": 16.37073707370737,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0029101039624135037,
      "loss": 0.2314,
      "num_input_tokens_seen": 31404640,
      "step": 148810
    },
    {
      "epoch": 16.371287128712872,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002909251622114524,
      "loss": 0.2314,
      "num_input_tokens_seen": 31405632,
      "step": 148815
    },
    {
      "epoch": 16.371837183718373,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00290839939325017,
      "loss": 0.2324,
      "num_input_tokens_seen": 31406688,
      "step": 148820
    },
    {
      "epoch": 16.37238723872387,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0029075472758282967,
      "loss": 0.2319,
      "num_input_tokens_seen": 31407712,
      "step": 148825
    },
    {
      "epoch": 16.372937293729372,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.0029066952698567606,
      "loss": 0.2303,
      "num_input_tokens_seen": 31408832,
      "step": 148830
    },
    {
      "epoch": 16.373487348734873,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.0029058433753434134,
      "loss": 0.2314,
      "num_input_tokens_seen": 31409888,
      "step": 148835
    },
    {
      "epoch": 16.374037403740374,
      "grad_norm": 0.000812530517578125,
      "learning_rate": 0.0029049915922960984,
      "loss": 0.2314,
      "num_input_tokens_seen": 31410944,
      "step": 148840
    },
    {
      "epoch": 16.374587458745875,
      "grad_norm": 0.0006561279296875,
      "learning_rate": 0.002904139920722678,
      "loss": 0.2314,
      "num_input_tokens_seen": 31412000,
      "step": 148845
    },
    {
      "epoch": 16.375137513751376,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00290328836063099,
      "loss": 0.2314,
      "num_input_tokens_seen": 31413056,
      "step": 148850
    },
    {
      "epoch": 16.375687568756877,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002902436912028891,
      "loss": 0.2319,
      "num_input_tokens_seen": 31414144,
      "step": 148855
    },
    {
      "epoch": 16.376237623762375,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0029015855749242284,
      "loss": 0.2298,
      "num_input_tokens_seen": 31415168,
      "step": 148860
    },
    {
      "epoch": 16.376787678767876,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.002900734349324843,
      "loss": 0.2304,
      "num_input_tokens_seen": 31416224,
      "step": 148865
    },
    {
      "epoch": 16.377337733773377,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0028998832352385854,
      "loss": 0.2299,
      "num_input_tokens_seen": 31417248,
      "step": 148870
    },
    {
      "epoch": 16.377887788778878,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0028990322326732957,
      "loss": 0.2308,
      "num_input_tokens_seen": 31418272,
      "step": 148875
    },
    {
      "epoch": 16.37843784378438,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0028981813416368188,
      "loss": 0.2314,
      "num_input_tokens_seen": 31419296,
      "step": 148880
    },
    {
      "epoch": 16.37898789878988,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.002897330562137,
      "loss": 0.2335,
      "num_input_tokens_seen": 31420320,
      "step": 148885
    },
    {
      "epoch": 16.379537953795378,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0028964798941816753,
      "loss": 0.2351,
      "num_input_tokens_seen": 31421312,
      "step": 148890
    },
    {
      "epoch": 16.38008800880088,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.002895629337778693,
      "loss": 0.2298,
      "num_input_tokens_seen": 31422368,
      "step": 148895
    },
    {
      "epoch": 16.38063806380638,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.0028947788929358847,
      "loss": 0.2319,
      "num_input_tokens_seen": 31423456,
      "step": 148900
    },
    {
      "epoch": 16.38118811881188,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.002893928559661088,
      "loss": 0.2303,
      "num_input_tokens_seen": 31424576,
      "step": 148905
    },
    {
      "epoch": 16.381738173817382,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.0028930783379621417,
      "loss": 0.2293,
      "num_input_tokens_seen": 31425632,
      "step": 148910
    },
    {
      "epoch": 16.382288228822883,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002892228227846884,
      "loss": 0.2319,
      "num_input_tokens_seen": 31426720,
      "step": 148915
    },
    {
      "epoch": 16.382838283828384,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0028913782293231508,
      "loss": 0.2308,
      "num_input_tokens_seen": 31427744,
      "step": 148920
    },
    {
      "epoch": 16.383388338833882,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.002890528342398776,
      "loss": 0.2324,
      "num_input_tokens_seen": 31428832,
      "step": 148925
    },
    {
      "epoch": 16.383938393839383,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.002889678567081586,
      "loss": 0.2314,
      "num_input_tokens_seen": 31429920,
      "step": 148930
    },
    {
      "epoch": 16.384488448844884,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.0028888289033794185,
      "loss": 0.2335,
      "num_input_tokens_seen": 31431008,
      "step": 148935
    },
    {
      "epoch": 16.385038503850385,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0028879793513001033,
      "loss": 0.2314,
      "num_input_tokens_seen": 31432064,
      "step": 148940
    },
    {
      "epoch": 16.385588558855886,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.0028871299108514745,
      "loss": 0.2319,
      "num_input_tokens_seen": 31433152,
      "step": 148945
    },
    {
      "epoch": 16.386138613861387,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.0028862805820413563,
      "loss": 0.2314,
      "num_input_tokens_seen": 31434208,
      "step": 148950
    },
    {
      "epoch": 16.38668866886689,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.002885431364877573,
      "loss": 0.2308,
      "num_input_tokens_seen": 31435296,
      "step": 148955
    },
    {
      "epoch": 16.387238723872386,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.002884582259367961,
      "loss": 0.2309,
      "num_input_tokens_seen": 31436288,
      "step": 148960
    },
    {
      "epoch": 16.387788778877887,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.002883733265520335,
      "loss": 0.2288,
      "num_input_tokens_seen": 31437376,
      "step": 148965
    },
    {
      "epoch": 16.388338833883388,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0028828843833425275,
      "loss": 0.2309,
      "num_input_tokens_seen": 31438400,
      "step": 148970
    },
    {
      "epoch": 16.38888888888889,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002882035612842362,
      "loss": 0.2319,
      "num_input_tokens_seen": 31439392,
      "step": 148975
    },
    {
      "epoch": 16.38943894389439,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.002881186954027658,
      "loss": 0.2324,
      "num_input_tokens_seen": 31440416,
      "step": 148980
    },
    {
      "epoch": 16.38998899889989,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0028803384069062414,
      "loss": 0.2298,
      "num_input_tokens_seen": 31441408,
      "step": 148985
    },
    {
      "epoch": 16.39053905390539,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.0028794899714859267,
      "loss": 0.2329,
      "num_input_tokens_seen": 31442496,
      "step": 148990
    },
    {
      "epoch": 16.39108910891089,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0028786416477745358,
      "loss": 0.2303,
      "num_input_tokens_seen": 31443520,
      "step": 148995
    },
    {
      "epoch": 16.39163916391639,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0028777934357798922,
      "loss": 0.2309,
      "num_input_tokens_seen": 31444576,
      "step": 149000
    },
    {
      "epoch": 16.392189218921892,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0028769453355098067,
      "loss": 0.2335,
      "num_input_tokens_seen": 31445600,
      "step": 149005
    },
    {
      "epoch": 16.392739273927393,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.0028760973469721027,
      "loss": 0.2319,
      "num_input_tokens_seen": 31446688,
      "step": 149010
    },
    {
      "epoch": 16.393289328932894,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0028752494701745892,
      "loss": 0.2329,
      "num_input_tokens_seen": 31447808,
      "step": 149015
    },
    {
      "epoch": 16.393839383938396,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.002874401705125085,
      "loss": 0.2304,
      "num_input_tokens_seen": 31448832,
      "step": 149020
    },
    {
      "epoch": 16.394389438943893,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002873554051831399,
      "loss": 0.2308,
      "num_input_tokens_seen": 31449824,
      "step": 149025
    },
    {
      "epoch": 16.394939493949394,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.002872706510301346,
      "loss": 0.2319,
      "num_input_tokens_seen": 31450848,
      "step": 149030
    },
    {
      "epoch": 16.395489548954895,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0028718590805427435,
      "loss": 0.2319,
      "num_input_tokens_seen": 31451936,
      "step": 149035
    },
    {
      "epoch": 16.396039603960396,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.002871011762563393,
      "loss": 0.2314,
      "num_input_tokens_seen": 31453024,
      "step": 149040
    },
    {
      "epoch": 16.396589658965897,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0028701645563711064,
      "loss": 0.2309,
      "num_input_tokens_seen": 31454048,
      "step": 149045
    },
    {
      "epoch": 16.3971397139714,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.0028693174619736908,
      "loss": 0.2298,
      "num_input_tokens_seen": 31455168,
      "step": 149050
    },
    {
      "epoch": 16.397689768976896,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002868470479378956,
      "loss": 0.2308,
      "num_input_tokens_seen": 31456224,
      "step": 149055
    },
    {
      "epoch": 16.398239823982397,
      "grad_norm": 0.0007476806640625,
      "learning_rate": 0.0028676236085947104,
      "loss": 0.2324,
      "num_input_tokens_seen": 31457248,
      "step": 149060
    },
    {
      "epoch": 16.3987898789879,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.0028667768496287577,
      "loss": 0.2304,
      "num_input_tokens_seen": 31458304,
      "step": 149065
    },
    {
      "epoch": 16.3993399339934,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.002865930202488895,
      "loss": 0.2308,
      "num_input_tokens_seen": 31459360,
      "step": 149070
    },
    {
      "epoch": 16.3998899889989,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002865083667182936,
      "loss": 0.2324,
      "num_input_tokens_seen": 31460416,
      "step": 149075
    },
    {
      "epoch": 16.4004400440044,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0028642372437186745,
      "loss": 0.2293,
      "num_input_tokens_seen": 31461472,
      "step": 149080
    },
    {
      "epoch": 16.400990099009903,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0028633909321039144,
      "loss": 0.2319,
      "num_input_tokens_seen": 31462496,
      "step": 149085
    },
    {
      "epoch": 16.4015401540154,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.0028625447323464615,
      "loss": 0.2324,
      "num_input_tokens_seen": 31463552,
      "step": 149090
    },
    {
      "epoch": 16.4020902090209,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0028616986444541053,
      "loss": 0.2319,
      "num_input_tokens_seen": 31464672,
      "step": 149095
    },
    {
      "epoch": 16.402640264026402,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.00286085266843465,
      "loss": 0.2308,
      "num_input_tokens_seen": 31465696,
      "step": 149100
    },
    {
      "epoch": 16.403190319031903,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.00286000680429589,
      "loss": 0.2324,
      "num_input_tokens_seen": 31466720,
      "step": 149105
    },
    {
      "epoch": 16.403740374037405,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0028591610520456194,
      "loss": 0.2324,
      "num_input_tokens_seen": 31467712,
      "step": 149110
    },
    {
      "epoch": 16.404290429042906,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.0028583154116916414,
      "loss": 0.2293,
      "num_input_tokens_seen": 31468736,
      "step": 149115
    },
    {
      "epoch": 16.404840484048403,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0028574698832417394,
      "loss": 0.2304,
      "num_input_tokens_seen": 31469824,
      "step": 149120
    },
    {
      "epoch": 16.405390539053904,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0028566244667037138,
      "loss": 0.2319,
      "num_input_tokens_seen": 31470912,
      "step": 149125
    },
    {
      "epoch": 16.405940594059405,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.00285577916208535,
      "loss": 0.2293,
      "num_input_tokens_seen": 31472000,
      "step": 149130
    },
    {
      "epoch": 16.406490649064907,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0028549339693944467,
      "loss": 0.2319,
      "num_input_tokens_seen": 31473056,
      "step": 149135
    },
    {
      "epoch": 16.407040704070408,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0028540888886387865,
      "loss": 0.2293,
      "num_input_tokens_seen": 31474208,
      "step": 149140
    },
    {
      "epoch": 16.40759075907591,
      "grad_norm": 0.001953125,
      "learning_rate": 0.0028532439198261593,
      "loss": 0.2314,
      "num_input_tokens_seen": 31475264,
      "step": 149145
    },
    {
      "epoch": 16.40814081408141,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0028523990629643592,
      "loss": 0.2304,
      "num_input_tokens_seen": 31476352,
      "step": 149150
    },
    {
      "epoch": 16.408690869086907,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002851554318061163,
      "loss": 0.234,
      "num_input_tokens_seen": 31477408,
      "step": 149155
    },
    {
      "epoch": 16.40924092409241,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002850709685124366,
      "loss": 0.2314,
      "num_input_tokens_seen": 31478464,
      "step": 149160
    },
    {
      "epoch": 16.40979097909791,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0028498651641617446,
      "loss": 0.2283,
      "num_input_tokens_seen": 31479456,
      "step": 149165
    },
    {
      "epoch": 16.41034103410341,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0028490207551810846,
      "loss": 0.2303,
      "num_input_tokens_seen": 31480576,
      "step": 149170
    },
    {
      "epoch": 16.41089108910891,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.002848176458190174,
      "loss": 0.2298,
      "num_input_tokens_seen": 31481600,
      "step": 149175
    },
    {
      "epoch": 16.411441144114413,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.002847332273196785,
      "loss": 0.2308,
      "num_input_tokens_seen": 31482624,
      "step": 149180
    },
    {
      "epoch": 16.41199119911991,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.002846488200208709,
      "loss": 0.2314,
      "num_input_tokens_seen": 31483744,
      "step": 149185
    },
    {
      "epoch": 16.41254125412541,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.0028456442392337184,
      "loss": 0.2324,
      "num_input_tokens_seen": 31484800,
      "step": 149190
    },
    {
      "epoch": 16.413091309130913,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002844800390279589,
      "loss": 0.2298,
      "num_input_tokens_seen": 31485888,
      "step": 149195
    },
    {
      "epoch": 16.413641364136414,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0028439566533541003,
      "loss": 0.2329,
      "num_input_tokens_seen": 31486912,
      "step": 149200
    },
    {
      "epoch": 16.414191419141915,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0028431130284650324,
      "loss": 0.2308,
      "num_input_tokens_seen": 31487904,
      "step": 149205
    },
    {
      "epoch": 16.414741474147416,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.00284226951562016,
      "loss": 0.2319,
      "num_input_tokens_seen": 31488928,
      "step": 149210
    },
    {
      "epoch": 16.415291529152917,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.002841426114827257,
      "loss": 0.2303,
      "num_input_tokens_seen": 31489984,
      "step": 149215
    },
    {
      "epoch": 16.415841584158414,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.002840582826094092,
      "loss": 0.2304,
      "num_input_tokens_seen": 31491008,
      "step": 149220
    },
    {
      "epoch": 16.416391639163916,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0028397396494284385,
      "loss": 0.233,
      "num_input_tokens_seen": 31492064,
      "step": 149225
    },
    {
      "epoch": 16.416941694169417,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.0028388965848380747,
      "loss": 0.2335,
      "num_input_tokens_seen": 31493056,
      "step": 149230
    },
    {
      "epoch": 16.417491749174918,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.002838053632330762,
      "loss": 0.2324,
      "num_input_tokens_seen": 31494080,
      "step": 149235
    },
    {
      "epoch": 16.41804180418042,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.0028372107919142756,
      "loss": 0.2324,
      "num_input_tokens_seen": 31495168,
      "step": 149240
    },
    {
      "epoch": 16.41859185918592,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0028363680635963783,
      "loss": 0.2303,
      "num_input_tokens_seen": 31496192,
      "step": 149245
    },
    {
      "epoch": 16.419141914191417,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0028355254473848433,
      "loss": 0.2324,
      "num_input_tokens_seen": 31497248,
      "step": 149250
    },
    {
      "epoch": 16.41969196919692,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.0028346829432874298,
      "loss": 0.2303,
      "num_input_tokens_seen": 31498272,
      "step": 149255
    },
    {
      "epoch": 16.42024202420242,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.002833840551311906,
      "loss": 0.2324,
      "num_input_tokens_seen": 31499328,
      "step": 149260
    },
    {
      "epoch": 16.42079207920792,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00283299827146604,
      "loss": 0.2314,
      "num_input_tokens_seen": 31500384,
      "step": 149265
    },
    {
      "epoch": 16.421342134213422,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.0028321561037575865,
      "loss": 0.2303,
      "num_input_tokens_seen": 31501504,
      "step": 149270
    },
    {
      "epoch": 16.421892189218923,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.002831314048194314,
      "loss": 0.2324,
      "num_input_tokens_seen": 31502496,
      "step": 149275
    },
    {
      "epoch": 16.422442244224424,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.0028304721047839824,
      "loss": 0.2335,
      "num_input_tokens_seen": 31503616,
      "step": 149280
    },
    {
      "epoch": 16.42299229922992,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0028296302735343415,
      "loss": 0.2298,
      "num_input_tokens_seen": 31504704,
      "step": 149285
    },
    {
      "epoch": 16.423542354235423,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.0028287885544531666,
      "loss": 0.2303,
      "num_input_tokens_seen": 31505760,
      "step": 149290
    },
    {
      "epoch": 16.424092409240924,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.002827946947548203,
      "loss": 0.2325,
      "num_input_tokens_seen": 31506784,
      "step": 149295
    },
    {
      "epoch": 16.424642464246425,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.002827105452827217,
      "loss": 0.2298,
      "num_input_tokens_seen": 31507904,
      "step": 149300
    },
    {
      "epoch": 16.425192519251926,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.002826264070297957,
      "loss": 0.2324,
      "num_input_tokens_seen": 31508960,
      "step": 149305
    },
    {
      "epoch": 16.425742574257427,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.002825422799968176,
      "loss": 0.2309,
      "num_input_tokens_seen": 31510016,
      "step": 149310
    },
    {
      "epoch": 16.426292629262925,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0028245816418456327,
      "loss": 0.2308,
      "num_input_tokens_seen": 31511008,
      "step": 149315
    },
    {
      "epoch": 16.426842684268426,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.002823740595938076,
      "loss": 0.2319,
      "num_input_tokens_seen": 31512032,
      "step": 149320
    },
    {
      "epoch": 16.427392739273927,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.002822899662253266,
      "loss": 0.2308,
      "num_input_tokens_seen": 31513088,
      "step": 149325
    },
    {
      "epoch": 16.427942794279428,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0028220588407989436,
      "loss": 0.2335,
      "num_input_tokens_seen": 31514144,
      "step": 149330
    },
    {
      "epoch": 16.42849284928493,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0028212181315828587,
      "loss": 0.2319,
      "num_input_tokens_seen": 31515200,
      "step": 149335
    },
    {
      "epoch": 16.42904290429043,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.002820377534612768,
      "loss": 0.2314,
      "num_input_tokens_seen": 31516256,
      "step": 149340
    },
    {
      "epoch": 16.42959295929593,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.0028195370498964034,
      "loss": 0.2303,
      "num_input_tokens_seen": 31517280,
      "step": 149345
    },
    {
      "epoch": 16.43014301430143,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002818696677441529,
      "loss": 0.2335,
      "num_input_tokens_seen": 31518368,
      "step": 149350
    },
    {
      "epoch": 16.43069306930693,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.002817856417255882,
      "loss": 0.2308,
      "num_input_tokens_seen": 31519360,
      "step": 149355
    },
    {
      "epoch": 16.43124312431243,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0028170162693472037,
      "loss": 0.2309,
      "num_input_tokens_seen": 31520480,
      "step": 149360
    },
    {
      "epoch": 16.431793179317932,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0028161762337232444,
      "loss": 0.2324,
      "num_input_tokens_seen": 31521536,
      "step": 149365
    },
    {
      "epoch": 16.432343234323433,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0028153363103917382,
      "loss": 0.2308,
      "num_input_tokens_seen": 31522624,
      "step": 149370
    },
    {
      "epoch": 16.432893289328934,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.00281449649936043,
      "loss": 0.2314,
      "num_input_tokens_seen": 31523616,
      "step": 149375
    },
    {
      "epoch": 16.433443344334435,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.0028136568006370643,
      "loss": 0.2319,
      "num_input_tokens_seen": 31524640,
      "step": 149380
    },
    {
      "epoch": 16.433993399339933,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.002812817214229372,
      "loss": 0.2319,
      "num_input_tokens_seen": 31525696,
      "step": 149385
    },
    {
      "epoch": 16.434543454345434,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.002811977740145101,
      "loss": 0.2309,
      "num_input_tokens_seen": 31526720,
      "step": 149390
    },
    {
      "epoch": 16.435093509350935,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0028111383783919783,
      "loss": 0.2314,
      "num_input_tokens_seen": 31527744,
      "step": 149395
    },
    {
      "epoch": 16.435643564356436,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.002810299128977748,
      "loss": 0.2298,
      "num_input_tokens_seen": 31528864,
      "step": 149400
    },
    {
      "epoch": 16.436193619361937,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0028094599919101376,
      "loss": 0.2324,
      "num_input_tokens_seen": 31529952,
      "step": 149405
    },
    {
      "epoch": 16.436743674367438,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0028086209671968846,
      "loss": 0.2319,
      "num_input_tokens_seen": 31531040,
      "step": 149410
    },
    {
      "epoch": 16.437293729372936,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0028077820548457265,
      "loss": 0.2319,
      "num_input_tokens_seen": 31532064,
      "step": 149415
    },
    {
      "epoch": 16.437843784378437,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00280694325486439,
      "loss": 0.2319,
      "num_input_tokens_seen": 31533120,
      "step": 149420
    },
    {
      "epoch": 16.438393839383938,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0028061045672606024,
      "loss": 0.2308,
      "num_input_tokens_seen": 31534240,
      "step": 149425
    },
    {
      "epoch": 16.43894389438944,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0028052659920420994,
      "loss": 0.234,
      "num_input_tokens_seen": 31535360,
      "step": 149430
    },
    {
      "epoch": 16.43949394939494,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002804427529216608,
      "loss": 0.2314,
      "num_input_tokens_seen": 31536352,
      "step": 149435
    },
    {
      "epoch": 16.44004400440044,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0028035891787918585,
      "loss": 0.2329,
      "num_input_tokens_seen": 31537376,
      "step": 149440
    },
    {
      "epoch": 16.440594059405942,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.002802750940775577,
      "loss": 0.233,
      "num_input_tokens_seen": 31538496,
      "step": 149445
    },
    {
      "epoch": 16.44114411441144,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.002801912815175483,
      "loss": 0.2325,
      "num_input_tokens_seen": 31539584,
      "step": 149450
    },
    {
      "epoch": 16.44169416941694,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.0028010748019993074,
      "loss": 0.2329,
      "num_input_tokens_seen": 31540640,
      "step": 149455
    },
    {
      "epoch": 16.442244224422442,
      "grad_norm": 0.000690460205078125,
      "learning_rate": 0.0028002369012547694,
      "loss": 0.2314,
      "num_input_tokens_seen": 31541728,
      "step": 149460
    },
    {
      "epoch": 16.442794279427943,
      "grad_norm": 0.00067901611328125,
      "learning_rate": 0.0027993991129495943,
      "loss": 0.2303,
      "num_input_tokens_seen": 31542784,
      "step": 149465
    },
    {
      "epoch": 16.443344334433444,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0027985614370915055,
      "loss": 0.2309,
      "num_input_tokens_seen": 31543808,
      "step": 149470
    },
    {
      "epoch": 16.443894389438945,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.0027977238736882192,
      "loss": 0.2303,
      "num_input_tokens_seen": 31544896,
      "step": 149475
    },
    {
      "epoch": 16.444444444444443,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0027968864227474587,
      "loss": 0.2314,
      "num_input_tokens_seen": 31545952,
      "step": 149480
    },
    {
      "epoch": 16.444994499449944,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.002796049084276938,
      "loss": 0.2314,
      "num_input_tokens_seen": 31547008,
      "step": 149485
    },
    {
      "epoch": 16.445544554455445,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002795211858284376,
      "loss": 0.2324,
      "num_input_tokens_seen": 31548064,
      "step": 149490
    },
    {
      "epoch": 16.446094609460946,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.0027943747447774945,
      "loss": 0.2319,
      "num_input_tokens_seen": 31549056,
      "step": 149495
    },
    {
      "epoch": 16.446644664466447,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.0027935377437639996,
      "loss": 0.2314,
      "num_input_tokens_seen": 31550176,
      "step": 149500
    },
    {
      "epoch": 16.44719471947195,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0027927008552516147,
      "loss": 0.2308,
      "num_input_tokens_seen": 31551264,
      "step": 149505
    },
    {
      "epoch": 16.44774477447745,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.0027918640792480437,
      "loss": 0.2314,
      "num_input_tokens_seen": 31552288,
      "step": 149510
    },
    {
      "epoch": 16.448294829482947,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002791027415761007,
      "loss": 0.2319,
      "num_input_tokens_seen": 31553344,
      "step": 149515
    },
    {
      "epoch": 16.448844884488448,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.002790190864798209,
      "loss": 0.2319,
      "num_input_tokens_seen": 31554432,
      "step": 149520
    },
    {
      "epoch": 16.44939493949395,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0027893544263673624,
      "loss": 0.2309,
      "num_input_tokens_seen": 31555488,
      "step": 149525
    },
    {
      "epoch": 16.44994499449945,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0027885181004761823,
      "loss": 0.2293,
      "num_input_tokens_seen": 31556544,
      "step": 149530
    },
    {
      "epoch": 16.45049504950495,
      "grad_norm": 0.000614166259765625,
      "learning_rate": 0.0027876818871323665,
      "loss": 0.2309,
      "num_input_tokens_seen": 31557664,
      "step": 149535
    },
    {
      "epoch": 16.451045104510452,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0027868457863436296,
      "loss": 0.2298,
      "num_input_tokens_seen": 31558752,
      "step": 149540
    },
    {
      "epoch": 16.45159515951595,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.0027860097981176717,
      "loss": 0.2314,
      "num_input_tokens_seen": 31559808,
      "step": 149545
    },
    {
      "epoch": 16.45214521452145,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0027851739224622,
      "loss": 0.2283,
      "num_input_tokens_seen": 31560832,
      "step": 149550
    },
    {
      "epoch": 16.452695269526952,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0027843381593849236,
      "loss": 0.2324,
      "num_input_tokens_seen": 31561792,
      "step": 149555
    },
    {
      "epoch": 16.453245324532453,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.002783502508893536,
      "loss": 0.2309,
      "num_input_tokens_seen": 31562816,
      "step": 149560
    },
    {
      "epoch": 16.453795379537954,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.002782666970995748,
      "loss": 0.2329,
      "num_input_tokens_seen": 31563872,
      "step": 149565
    },
    {
      "epoch": 16.454345434543455,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0027818315456992564,
      "loss": 0.2324,
      "num_input_tokens_seen": 31564960,
      "step": 149570
    },
    {
      "epoch": 16.454895489548957,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.002780996233011757,
      "loss": 0.2324,
      "num_input_tokens_seen": 31565952,
      "step": 149575
    },
    {
      "epoch": 16.455445544554454,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002780161032940952,
      "loss": 0.2324,
      "num_input_tokens_seen": 31566944,
      "step": 149580
    },
    {
      "epoch": 16.455995599559955,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.0027793259454945386,
      "loss": 0.2319,
      "num_input_tokens_seen": 31568000,
      "step": 149585
    },
    {
      "epoch": 16.456545654565456,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002778490970680217,
      "loss": 0.2303,
      "num_input_tokens_seen": 31569088,
      "step": 149590
    },
    {
      "epoch": 16.457095709570957,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.002777656108505681,
      "loss": 0.2314,
      "num_input_tokens_seen": 31570144,
      "step": 149595
    },
    {
      "epoch": 16.45764576457646,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00277682135897862,
      "loss": 0.2304,
      "num_input_tokens_seen": 31571264,
      "step": 149600
    },
    {
      "epoch": 16.45819581958196,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.0027759867221067306,
      "loss": 0.2308,
      "num_input_tokens_seen": 31572288,
      "step": 149605
    },
    {
      "epoch": 16.458745874587457,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0027751521978977107,
      "loss": 0.2319,
      "num_input_tokens_seen": 31573344,
      "step": 149610
    },
    {
      "epoch": 16.459295929592958,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0027743177863592417,
      "loss": 0.2309,
      "num_input_tokens_seen": 31574464,
      "step": 149615
    },
    {
      "epoch": 16.45984598459846,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.0027734834874990227,
      "loss": 0.2314,
      "num_input_tokens_seen": 31575488,
      "step": 149620
    },
    {
      "epoch": 16.46039603960396,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.0027726493013247376,
      "loss": 0.2319,
      "num_input_tokens_seen": 31576544,
      "step": 149625
    },
    {
      "epoch": 16.46094609460946,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0027718152278440786,
      "loss": 0.2309,
      "num_input_tokens_seen": 31577568,
      "step": 149630
    },
    {
      "epoch": 16.461496149614963,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.0027709812670647298,
      "loss": 0.2324,
      "num_input_tokens_seen": 31578656,
      "step": 149635
    },
    {
      "epoch": 16.462046204620464,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0027701474189943762,
      "loss": 0.2293,
      "num_input_tokens_seen": 31579712,
      "step": 149640
    },
    {
      "epoch": 16.46259625962596,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0027693136836407107,
      "loss": 0.2314,
      "num_input_tokens_seen": 31580736,
      "step": 149645
    },
    {
      "epoch": 16.463146314631462,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0027684800610114062,
      "loss": 0.2298,
      "num_input_tokens_seen": 31581856,
      "step": 149650
    },
    {
      "epoch": 16.463696369636963,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0027676465511141573,
      "loss": 0.2319,
      "num_input_tokens_seen": 31582880,
      "step": 149655
    },
    {
      "epoch": 16.464246424642464,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.002766813153956636,
      "loss": 0.2298,
      "num_input_tokens_seen": 31584000,
      "step": 149660
    },
    {
      "epoch": 16.464796479647966,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0027659798695465294,
      "loss": 0.2303,
      "num_input_tokens_seen": 31585024,
      "step": 149665
    },
    {
      "epoch": 16.465346534653467,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0027651466978915176,
      "loss": 0.2324,
      "num_input_tokens_seen": 31586080,
      "step": 149670
    },
    {
      "epoch": 16.465896589658964,
      "grad_norm": 0.000751495361328125,
      "learning_rate": 0.0027643136389992754,
      "loss": 0.2324,
      "num_input_tokens_seen": 31587136,
      "step": 149675
    },
    {
      "epoch": 16.466446644664465,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0027634806928774863,
      "loss": 0.2314,
      "num_input_tokens_seen": 31588160,
      "step": 149680
    },
    {
      "epoch": 16.466996699669966,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.002762647859533824,
      "loss": 0.2335,
      "num_input_tokens_seen": 31589152,
      "step": 149685
    },
    {
      "epoch": 16.467546754675467,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0027618151389759604,
      "loss": 0.2324,
      "num_input_tokens_seen": 31590240,
      "step": 149690
    },
    {
      "epoch": 16.46809680968097,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0027609825312115747,
      "loss": 0.2293,
      "num_input_tokens_seen": 31591360,
      "step": 149695
    },
    {
      "epoch": 16.46864686468647,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0027601500362483392,
      "loss": 0.2319,
      "num_input_tokens_seen": 31592384,
      "step": 149700
    },
    {
      "epoch": 16.46919691969197,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002759317654093933,
      "loss": 0.2329,
      "num_input_tokens_seen": 31593408,
      "step": 149705
    },
    {
      "epoch": 16.46974697469747,
      "grad_norm": 0.0003757476806640625,
      "learning_rate": 0.0027584853847560206,
      "loss": 0.2293,
      "num_input_tokens_seen": 31594496,
      "step": 149710
    },
    {
      "epoch": 16.47029702970297,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.0027576532282422702,
      "loss": 0.2324,
      "num_input_tokens_seen": 31595584,
      "step": 149715
    },
    {
      "epoch": 16.47084708470847,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0027568211845603575,
      "loss": 0.2304,
      "num_input_tokens_seen": 31596736,
      "step": 149720
    },
    {
      "epoch": 16.47139713971397,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0027559892537179485,
      "loss": 0.2303,
      "num_input_tokens_seen": 31597824,
      "step": 149725
    },
    {
      "epoch": 16.471947194719473,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0027551574357227147,
      "loss": 0.2288,
      "num_input_tokens_seen": 31598912,
      "step": 149730
    },
    {
      "epoch": 16.472497249724974,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0027543257305823196,
      "loss": 0.2319,
      "num_input_tokens_seen": 31600000,
      "step": 149735
    },
    {
      "epoch": 16.47304730473047,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.0027534941383044243,
      "loss": 0.2329,
      "num_input_tokens_seen": 31601120,
      "step": 149740
    },
    {
      "epoch": 16.473597359735972,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0027526626588967,
      "loss": 0.2329,
      "num_input_tokens_seen": 31602144,
      "step": 149745
    },
    {
      "epoch": 16.474147414741473,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.0027518312923668045,
      "loss": 0.2329,
      "num_input_tokens_seen": 31603200,
      "step": 149750
    },
    {
      "epoch": 16.474697469746975,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.002751000038722402,
      "loss": 0.2278,
      "num_input_tokens_seen": 31604288,
      "step": 149755
    },
    {
      "epoch": 16.475247524752476,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.002750168897971159,
      "loss": 0.2303,
      "num_input_tokens_seen": 31605280,
      "step": 149760
    },
    {
      "epoch": 16.475797579757977,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0027493378701207265,
      "loss": 0.2288,
      "num_input_tokens_seen": 31606304,
      "step": 149765
    },
    {
      "epoch": 16.476347634763478,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0027485069551787733,
      "loss": 0.2298,
      "num_input_tokens_seen": 31607328,
      "step": 149770
    },
    {
      "epoch": 16.476897689768975,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.002747676153152948,
      "loss": 0.2293,
      "num_input_tokens_seen": 31608384,
      "step": 149775
    },
    {
      "epoch": 16.477447744774476,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002746845464050913,
      "loss": 0.2314,
      "num_input_tokens_seen": 31609408,
      "step": 149780
    },
    {
      "epoch": 16.477997799779978,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0027460148878803276,
      "loss": 0.2329,
      "num_input_tokens_seen": 31610496,
      "step": 149785
    },
    {
      "epoch": 16.47854785478548,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002745184424648838,
      "loss": 0.2283,
      "num_input_tokens_seen": 31611552,
      "step": 149790
    },
    {
      "epoch": 16.47909790979098,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0027443540743641074,
      "loss": 0.2335,
      "num_input_tokens_seen": 31612608,
      "step": 149795
    },
    {
      "epoch": 16.47964796479648,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.002743523837033784,
      "loss": 0.235,
      "num_input_tokens_seen": 31613696,
      "step": 149800
    },
    {
      "epoch": 16.480198019801982,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.0027426937126655174,
      "loss": 0.2303,
      "num_input_tokens_seen": 31614784,
      "step": 149805
    },
    {
      "epoch": 16.48074807480748,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.002741863701266961,
      "loss": 0.2309,
      "num_input_tokens_seen": 31615872,
      "step": 149810
    },
    {
      "epoch": 16.48129812981298,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.002741033802845765,
      "loss": 0.2329,
      "num_input_tokens_seen": 31616896,
      "step": 149815
    },
    {
      "epoch": 16.48184818481848,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0027402040174095807,
      "loss": 0.2319,
      "num_input_tokens_seen": 31618016,
      "step": 149820
    },
    {
      "epoch": 16.482398239823983,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.002739374344966055,
      "loss": 0.2309,
      "num_input_tokens_seen": 31619072,
      "step": 149825
    },
    {
      "epoch": 16.482948294829484,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0027385447855228285,
      "loss": 0.2329,
      "num_input_tokens_seen": 31620192,
      "step": 149830
    },
    {
      "epoch": 16.483498349834985,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0027377153390875496,
      "loss": 0.2309,
      "num_input_tokens_seen": 31621184,
      "step": 149835
    },
    {
      "epoch": 16.484048404840483,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0027368860056678646,
      "loss": 0.2324,
      "num_input_tokens_seen": 31622272,
      "step": 149840
    },
    {
      "epoch": 16.484598459845984,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.002736056785271422,
      "loss": 0.2319,
      "num_input_tokens_seen": 31623328,
      "step": 149845
    },
    {
      "epoch": 16.485148514851485,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.002735227677905859,
      "loss": 0.2324,
      "num_input_tokens_seen": 31624448,
      "step": 149850
    },
    {
      "epoch": 16.485698569856986,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0027343986835788123,
      "loss": 0.2319,
      "num_input_tokens_seen": 31625472,
      "step": 149855
    },
    {
      "epoch": 16.486248624862487,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.002733569802297932,
      "loss": 0.2329,
      "num_input_tokens_seen": 31626464,
      "step": 149860
    },
    {
      "epoch": 16.486798679867988,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.0027327410340708498,
      "loss": 0.2324,
      "num_input_tokens_seen": 31627552,
      "step": 149865
    },
    {
      "epoch": 16.48734873487349,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.002731912378905206,
      "loss": 0.2314,
      "num_input_tokens_seen": 31628608,
      "step": 149870
    },
    {
      "epoch": 16.487898789878987,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0027310838368086424,
      "loss": 0.2314,
      "num_input_tokens_seen": 31629632,
      "step": 149875
    },
    {
      "epoch": 16.488448844884488,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.002730255407788789,
      "loss": 0.2324,
      "num_input_tokens_seen": 31630720,
      "step": 149880
    },
    {
      "epoch": 16.48899889988999,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0027294270918532875,
      "loss": 0.2309,
      "num_input_tokens_seen": 31631776,
      "step": 149885
    },
    {
      "epoch": 16.48954895489549,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.002728598889009765,
      "loss": 0.2298,
      "num_input_tokens_seen": 31632768,
      "step": 149890
    },
    {
      "epoch": 16.49009900990099,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.002727770799265857,
      "loss": 0.2308,
      "num_input_tokens_seen": 31633920,
      "step": 149895
    },
    {
      "epoch": 16.490649064906492,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0027269428226292013,
      "loss": 0.2319,
      "num_input_tokens_seen": 31634976,
      "step": 149900
    },
    {
      "epoch": 16.49119911991199,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.002726114959107421,
      "loss": 0.2308,
      "num_input_tokens_seen": 31636032,
      "step": 149905
    },
    {
      "epoch": 16.49174917491749,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.0027252872087081526,
      "loss": 0.2314,
      "num_input_tokens_seen": 31637152,
      "step": 149910
    },
    {
      "epoch": 16.492299229922992,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.0027244595714390188,
      "loss": 0.2324,
      "num_input_tokens_seen": 31638176,
      "step": 149915
    },
    {
      "epoch": 16.492849284928493,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0027236320473076545,
      "loss": 0.2314,
      "num_input_tokens_seen": 31639264,
      "step": 149920
    },
    {
      "epoch": 16.493399339933994,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002722804636321678,
      "loss": 0.2345,
      "num_input_tokens_seen": 31640352,
      "step": 149925
    },
    {
      "epoch": 16.493949394939495,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.0027219773384887205,
      "loss": 0.2308,
      "num_input_tokens_seen": 31641440,
      "step": 149930
    },
    {
      "epoch": 16.494499449944996,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0027211501538164104,
      "loss": 0.2329,
      "num_input_tokens_seen": 31642528,
      "step": 149935
    },
    {
      "epoch": 16.495049504950494,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.002720323082312365,
      "loss": 0.2308,
      "num_input_tokens_seen": 31643552,
      "step": 149940
    },
    {
      "epoch": 16.495599559955995,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.002719496123984212,
      "loss": 0.2314,
      "num_input_tokens_seen": 31644608,
      "step": 149945
    },
    {
      "epoch": 16.496149614961496,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.002718669278839566,
      "loss": 0.2314,
      "num_input_tokens_seen": 31645568,
      "step": 149950
    },
    {
      "epoch": 16.496699669966997,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.0027178425468860515,
      "loss": 0.2324,
      "num_input_tokens_seen": 31646624,
      "step": 149955
    },
    {
      "epoch": 16.497249724972498,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.002717015928131294,
      "loss": 0.2329,
      "num_input_tokens_seen": 31647616,
      "step": 149960
    },
    {
      "epoch": 16.497799779978,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.002716189422582903,
      "loss": 0.2319,
      "num_input_tokens_seen": 31648672,
      "step": 149965
    },
    {
      "epoch": 16.498349834983497,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.002715363030248503,
      "loss": 0.2308,
      "num_input_tokens_seen": 31649728,
      "step": 149970
    },
    {
      "epoch": 16.498899889988998,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002714536751135707,
      "loss": 0.2319,
      "num_input_tokens_seen": 31650784,
      "step": 149975
    },
    {
      "epoch": 16.4994499449945,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.002713710585252127,
      "loss": 0.2308,
      "num_input_tokens_seen": 31651840,
      "step": 149980
    },
    {
      "epoch": 16.5,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0027128845326053817,
      "loss": 0.2319,
      "num_input_tokens_seen": 31652960,
      "step": 149985
    },
    {
      "epoch": 16.5005500550055,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002712058593203087,
      "loss": 0.2308,
      "num_input_tokens_seen": 31654048,
      "step": 149990
    },
    {
      "epoch": 16.501100110011002,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0027112327670528484,
      "loss": 0.2324,
      "num_input_tokens_seen": 31655104,
      "step": 149995
    },
    {
      "epoch": 16.501650165016503,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.002710407054162283,
      "loss": 0.2329,
      "num_input_tokens_seen": 31656224,
      "step": 150000
    },
    {
      "epoch": 16.502200220022,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0027095814545389967,
      "loss": 0.2335,
      "num_input_tokens_seen": 31657216,
      "step": 150005
    },
    {
      "epoch": 16.502750275027502,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0027087559681906,
      "loss": 0.2324,
      "num_input_tokens_seen": 31658336,
      "step": 150010
    },
    {
      "epoch": 16.503300330033003,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0027079305951247046,
      "loss": 0.2324,
      "num_input_tokens_seen": 31659392,
      "step": 150015
    },
    {
      "epoch": 16.503850385038504,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0027071053353489114,
      "loss": 0.2293,
      "num_input_tokens_seen": 31660448,
      "step": 150020
    },
    {
      "epoch": 16.504400440044005,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.002706280188870835,
      "loss": 0.2298,
      "num_input_tokens_seen": 31661504,
      "step": 150025
    },
    {
      "epoch": 16.504950495049506,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.002705455155698068,
      "loss": 0.2303,
      "num_input_tokens_seen": 31662624,
      "step": 150030
    },
    {
      "epoch": 16.505500550055004,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0027046302358382274,
      "loss": 0.2309,
      "num_input_tokens_seen": 31663680,
      "step": 150035
    },
    {
      "epoch": 16.506050605060505,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.002703805429298906,
      "loss": 0.2329,
      "num_input_tokens_seen": 31664672,
      "step": 150040
    },
    {
      "epoch": 16.506600660066006,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00270298073608771,
      "loss": 0.2324,
      "num_input_tokens_seen": 31665728,
      "step": 150045
    },
    {
      "epoch": 16.507150715071507,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002702156156212243,
      "loss": 0.2324,
      "num_input_tokens_seen": 31666816,
      "step": 150050
    },
    {
      "epoch": 16.507700770077008,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0027013316896800977,
      "loss": 0.2314,
      "num_input_tokens_seen": 31667904,
      "step": 150055
    },
    {
      "epoch": 16.50825082508251,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.002700507336498881,
      "loss": 0.2319,
      "num_input_tokens_seen": 31668992,
      "step": 150060
    },
    {
      "epoch": 16.50880088008801,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.002699683096676185,
      "loss": 0.2303,
      "num_input_tokens_seen": 31670080,
      "step": 150065
    },
    {
      "epoch": 16.509350935093508,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.002698858970219602,
      "loss": 0.2303,
      "num_input_tokens_seen": 31671168,
      "step": 150070
    },
    {
      "epoch": 16.50990099009901,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0026980349571367394,
      "loss": 0.2303,
      "num_input_tokens_seen": 31672160,
      "step": 150075
    },
    {
      "epoch": 16.51045104510451,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0026972110574351813,
      "loss": 0.2324,
      "num_input_tokens_seen": 31673184,
      "step": 150080
    },
    {
      "epoch": 16.51100110011001,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0026963872711225317,
      "loss": 0.2309,
      "num_input_tokens_seen": 31674304,
      "step": 150085
    },
    {
      "epoch": 16.511551155115512,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.002695563598206374,
      "loss": 0.2303,
      "num_input_tokens_seen": 31675328,
      "step": 150090
    },
    {
      "epoch": 16.512101210121013,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.002694740038694301,
      "loss": 0.2303,
      "num_input_tokens_seen": 31676352,
      "step": 150095
    },
    {
      "epoch": 16.51265126512651,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.002693916592593903,
      "loss": 0.2298,
      "num_input_tokens_seen": 31677408,
      "step": 150100
    },
    {
      "epoch": 16.513201320132012,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.0026930932599127703,
      "loss": 0.2298,
      "num_input_tokens_seen": 31678432,
      "step": 150105
    },
    {
      "epoch": 16.513751375137513,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0026922700406584957,
      "loss": 0.2314,
      "num_input_tokens_seen": 31679392,
      "step": 150110
    },
    {
      "epoch": 16.514301430143014,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.002691446934838663,
      "loss": 0.2319,
      "num_input_tokens_seen": 31680512,
      "step": 150115
    },
    {
      "epoch": 16.514851485148515,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002690623942460853,
      "loss": 0.2324,
      "num_input_tokens_seen": 31681504,
      "step": 150120
    },
    {
      "epoch": 16.515401540154016,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.002689801063532661,
      "loss": 0.2319,
      "num_input_tokens_seen": 31682624,
      "step": 150125
    },
    {
      "epoch": 16.515951595159517,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.0026889782980616572,
      "loss": 0.2288,
      "num_input_tokens_seen": 31683680,
      "step": 150130
    },
    {
      "epoch": 16.516501650165015,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0026881556460554394,
      "loss": 0.2314,
      "num_input_tokens_seen": 31684672,
      "step": 150135
    },
    {
      "epoch": 16.517051705170516,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.0026873331075215833,
      "loss": 0.2324,
      "num_input_tokens_seen": 31685696,
      "step": 150140
    },
    {
      "epoch": 16.517601760176017,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 0.0026865106824676675,
      "loss": 0.2324,
      "num_input_tokens_seen": 31686848,
      "step": 150145
    },
    {
      "epoch": 16.51815181518152,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.002685688370901278,
      "loss": 0.2298,
      "num_input_tokens_seen": 31687904,
      "step": 150150
    },
    {
      "epoch": 16.51870187018702,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.002684866172829985,
      "loss": 0.2308,
      "num_input_tokens_seen": 31688960,
      "step": 150155
    },
    {
      "epoch": 16.51925192519252,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002684044088261371,
      "loss": 0.2324,
      "num_input_tokens_seen": 31690016,
      "step": 150160
    },
    {
      "epoch": 16.519801980198018,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0026832221172030163,
      "loss": 0.2314,
      "num_input_tokens_seen": 31691136,
      "step": 150165
    },
    {
      "epoch": 16.52035203520352,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00268240025966249,
      "loss": 0.2319,
      "num_input_tokens_seen": 31692160,
      "step": 150170
    },
    {
      "epoch": 16.52090209020902,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0026815785156473746,
      "loss": 0.2314,
      "num_input_tokens_seen": 31693216,
      "step": 150175
    },
    {
      "epoch": 16.52145214521452,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0026807568851652374,
      "loss": 0.2319,
      "num_input_tokens_seen": 31694240,
      "step": 150180
    },
    {
      "epoch": 16.522002200220022,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0026799353682236504,
      "loss": 0.2314,
      "num_input_tokens_seen": 31695232,
      "step": 150185
    },
    {
      "epoch": 16.522552255225524,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0026791139648301857,
      "loss": 0.2319,
      "num_input_tokens_seen": 31696288,
      "step": 150190
    },
    {
      "epoch": 16.523102310231025,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.002678292674992418,
      "loss": 0.2319,
      "num_input_tokens_seen": 31697344,
      "step": 150195
    },
    {
      "epoch": 16.523652365236522,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0026774714987179153,
      "loss": 0.2335,
      "num_input_tokens_seen": 31698368,
      "step": 150200
    },
    {
      "epoch": 16.524202420242023,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0026766504360142456,
      "loss": 0.2335,
      "num_input_tokens_seen": 31699392,
      "step": 150205
    },
    {
      "epoch": 16.524752475247524,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0026758294868889725,
      "loss": 0.2298,
      "num_input_tokens_seen": 31700480,
      "step": 150210
    },
    {
      "epoch": 16.525302530253025,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.002675008651349665,
      "loss": 0.2308,
      "num_input_tokens_seen": 31701568,
      "step": 150215
    },
    {
      "epoch": 16.525852585258527,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.0026741879294038887,
      "loss": 0.2303,
      "num_input_tokens_seen": 31702592,
      "step": 150220
    },
    {
      "epoch": 16.526402640264028,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002673367321059211,
      "loss": 0.2319,
      "num_input_tokens_seen": 31703616,
      "step": 150225
    },
    {
      "epoch": 16.52695269526953,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0026725468263231926,
      "loss": 0.2314,
      "num_input_tokens_seen": 31704672,
      "step": 150230
    },
    {
      "epoch": 16.527502750275026,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0026717264452033903,
      "loss": 0.2308,
      "num_input_tokens_seen": 31705728,
      "step": 150235
    },
    {
      "epoch": 16.528052805280527,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002670906177707374,
      "loss": 0.2308,
      "num_input_tokens_seen": 31706816,
      "step": 150240
    },
    {
      "epoch": 16.52860286028603,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.0026700860238426946,
      "loss": 0.2309,
      "num_input_tokens_seen": 31707936,
      "step": 150245
    },
    {
      "epoch": 16.52915291529153,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.002669265983616917,
      "loss": 0.2309,
      "num_input_tokens_seen": 31708960,
      "step": 150250
    },
    {
      "epoch": 16.52970297029703,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.002668446057037602,
      "loss": 0.2309,
      "num_input_tokens_seen": 31710016,
      "step": 150255
    },
    {
      "epoch": 16.53025302530253,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.0026676262441122983,
      "loss": 0.2309,
      "num_input_tokens_seen": 31711072,
      "step": 150260
    },
    {
      "epoch": 16.53080308030803,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0026668065448485693,
      "loss": 0.2319,
      "num_input_tokens_seen": 31712128,
      "step": 150265
    },
    {
      "epoch": 16.53135313531353,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.002665986959253963,
      "loss": 0.2335,
      "num_input_tokens_seen": 31713184,
      "step": 150270
    },
    {
      "epoch": 16.53190319031903,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.0026651674873360348,
      "loss": 0.2319,
      "num_input_tokens_seen": 31714272,
      "step": 150275
    },
    {
      "epoch": 16.532453245324533,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.002664348129102345,
      "loss": 0.2303,
      "num_input_tokens_seen": 31715328,
      "step": 150280
    },
    {
      "epoch": 16.533003300330034,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.002663528884560435,
      "loss": 0.2329,
      "num_input_tokens_seen": 31716384,
      "step": 150285
    },
    {
      "epoch": 16.533553355335535,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002662709753717863,
      "loss": 0.2283,
      "num_input_tokens_seen": 31717504,
      "step": 150290
    },
    {
      "epoch": 16.534103410341036,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.002661890736582171,
      "loss": 0.2298,
      "num_input_tokens_seen": 31718560,
      "step": 150295
    },
    {
      "epoch": 16.534653465346533,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0026610718331609162,
      "loss": 0.2329,
      "num_input_tokens_seen": 31719648,
      "step": 150300
    },
    {
      "epoch": 16.535203520352034,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0026602530434616375,
      "loss": 0.2314,
      "num_input_tokens_seen": 31720736,
      "step": 150305
    },
    {
      "epoch": 16.535753575357536,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.002659434367491885,
      "loss": 0.2308,
      "num_input_tokens_seen": 31721888,
      "step": 150310
    },
    {
      "epoch": 16.536303630363037,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0026586158052592087,
      "loss": 0.2324,
      "num_input_tokens_seen": 31722944,
      "step": 150315
    },
    {
      "epoch": 16.536853685368538,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0026577973567711447,
      "loss": 0.2299,
      "num_input_tokens_seen": 31724000,
      "step": 150320
    },
    {
      "epoch": 16.53740374037404,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0026569790220352433,
      "loss": 0.2314,
      "num_input_tokens_seen": 31724992,
      "step": 150325
    },
    {
      "epoch": 16.537953795379536,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.00265616080105904,
      "loss": 0.2314,
      "num_input_tokens_seen": 31726016,
      "step": 150330
    },
    {
      "epoch": 16.538503850385037,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0026553426938500796,
      "loss": 0.2325,
      "num_input_tokens_seen": 31727040,
      "step": 150335
    },
    {
      "epoch": 16.53905390539054,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.002654524700415905,
      "loss": 0.2304,
      "num_input_tokens_seen": 31728128,
      "step": 150340
    },
    {
      "epoch": 16.53960396039604,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0026537068207640488,
      "loss": 0.234,
      "num_input_tokens_seen": 31729152,
      "step": 150345
    },
    {
      "epoch": 16.54015401540154,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.002652889054902057,
      "loss": 0.2303,
      "num_input_tokens_seen": 31730176,
      "step": 150350
    },
    {
      "epoch": 16.540704070407042,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.002652071402837462,
      "loss": 0.2308,
      "num_input_tokens_seen": 31731264,
      "step": 150355
    },
    {
      "epoch": 16.541254125412543,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002651253864577797,
      "loss": 0.2319,
      "num_input_tokens_seen": 31732416,
      "step": 150360
    },
    {
      "epoch": 16.54180418041804,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.0026504364401305986,
      "loss": 0.2314,
      "num_input_tokens_seen": 31733440,
      "step": 150365
    },
    {
      "epoch": 16.54235423542354,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0026496191295034047,
      "loss": 0.2309,
      "num_input_tokens_seen": 31734560,
      "step": 150370
    },
    {
      "epoch": 16.542904290429043,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.002648801932703742,
      "loss": 0.2309,
      "num_input_tokens_seen": 31735584,
      "step": 150375
    },
    {
      "epoch": 16.543454345434544,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0026479848497391488,
      "loss": 0.2319,
      "num_input_tokens_seen": 31736640,
      "step": 150380
    },
    {
      "epoch": 16.544004400440045,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0026471678806171475,
      "loss": 0.2329,
      "num_input_tokens_seen": 31737696,
      "step": 150385
    },
    {
      "epoch": 16.544554455445546,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0026463510253452744,
      "loss": 0.2308,
      "num_input_tokens_seen": 31738720,
      "step": 150390
    },
    {
      "epoch": 16.545104510451043,
      "grad_norm": 0.00072479248046875,
      "learning_rate": 0.0026455342839310572,
      "loss": 0.2324,
      "num_input_tokens_seen": 31739776,
      "step": 150395
    },
    {
      "epoch": 16.545654565456545,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.002644717656382021,
      "loss": 0.2304,
      "num_input_tokens_seen": 31740864,
      "step": 150400
    },
    {
      "epoch": 16.546204620462046,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0026439011427056964,
      "loss": 0.2303,
      "num_input_tokens_seen": 31741856,
      "step": 150405
    },
    {
      "epoch": 16.546754675467547,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.0026430847429096014,
      "loss": 0.2309,
      "num_input_tokens_seen": 31742944,
      "step": 150410
    },
    {
      "epoch": 16.547304730473048,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0026422684570012694,
      "loss": 0.2308,
      "num_input_tokens_seen": 31744000,
      "step": 150415
    },
    {
      "epoch": 16.54785478547855,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.002641452284988215,
      "loss": 0.2324,
      "num_input_tokens_seen": 31745088,
      "step": 150420
    },
    {
      "epoch": 16.54840484048405,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0026406362268779652,
      "loss": 0.2303,
      "num_input_tokens_seen": 31746144,
      "step": 150425
    },
    {
      "epoch": 16.548954895489548,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.002639820282678045,
      "loss": 0.2288,
      "num_input_tokens_seen": 31747200,
      "step": 150430
    },
    {
      "epoch": 16.54950495049505,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0026390044523959644,
      "loss": 0.2319,
      "num_input_tokens_seen": 31748288,
      "step": 150435
    },
    {
      "epoch": 16.55005500550055,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.002638188736039253,
      "loss": 0.2319,
      "num_input_tokens_seen": 31749280,
      "step": 150440
    },
    {
      "epoch": 16.55060506050605,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0026373731336154214,
      "loss": 0.2309,
      "num_input_tokens_seen": 31750272,
      "step": 150445
    },
    {
      "epoch": 16.551155115511552,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.002636557645131989,
      "loss": 0.2319,
      "num_input_tokens_seen": 31751328,
      "step": 150450
    },
    {
      "epoch": 16.551705170517053,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.002635742270596477,
      "loss": 0.2308,
      "num_input_tokens_seen": 31752352,
      "step": 150455
    },
    {
      "epoch": 16.55225522552255,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0026349270100163905,
      "loss": 0.2319,
      "num_input_tokens_seen": 31753344,
      "step": 150460
    },
    {
      "epoch": 16.55280528052805,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0026341118633992516,
      "loss": 0.2309,
      "num_input_tokens_seen": 31754400,
      "step": 150465
    },
    {
      "epoch": 16.553355335533553,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.002633296830752572,
      "loss": 0.2314,
      "num_input_tokens_seen": 31755424,
      "step": 150470
    },
    {
      "epoch": 16.553905390539054,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.0026324819120838567,
      "loss": 0.2324,
      "num_input_tokens_seen": 31756512,
      "step": 150475
    },
    {
      "epoch": 16.554455445544555,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.0026316671074006204,
      "loss": 0.2303,
      "num_input_tokens_seen": 31757568,
      "step": 150480
    },
    {
      "epoch": 16.555005500550056,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.002630852416710374,
      "loss": 0.2309,
      "num_input_tokens_seen": 31758592,
      "step": 150485
    },
    {
      "epoch": 16.555555555555557,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0026300378400206286,
      "loss": 0.2308,
      "num_input_tokens_seen": 31759680,
      "step": 150490
    },
    {
      "epoch": 16.556105610561055,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0026292233773388893,
      "loss": 0.2314,
      "num_input_tokens_seen": 31760704,
      "step": 150495
    },
    {
      "epoch": 16.556655665566556,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.002628409028672659,
      "loss": 0.2298,
      "num_input_tokens_seen": 31761728,
      "step": 150500
    },
    {
      "epoch": 16.557205720572057,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.0026275947940294446,
      "loss": 0.2329,
      "num_input_tokens_seen": 31762784,
      "step": 150505
    },
    {
      "epoch": 16.557755775577558,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002626780673416754,
      "loss": 0.2309,
      "num_input_tokens_seen": 31763808,
      "step": 150510
    },
    {
      "epoch": 16.55830583058306,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0026259666668420915,
      "loss": 0.2319,
      "num_input_tokens_seen": 31764832,
      "step": 150515
    },
    {
      "epoch": 16.55885588558856,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0026251527743129557,
      "loss": 0.2314,
      "num_input_tokens_seen": 31765888,
      "step": 150520
    },
    {
      "epoch": 16.55940594059406,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0026243389958368474,
      "loss": 0.2314,
      "num_input_tokens_seen": 31766944,
      "step": 150525
    },
    {
      "epoch": 16.55995599559956,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0026235253314212696,
      "loss": 0.2319,
      "num_input_tokens_seen": 31767968,
      "step": 150530
    },
    {
      "epoch": 16.56050605060506,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.002622711781073717,
      "loss": 0.2314,
      "num_input_tokens_seen": 31768992,
      "step": 150535
    },
    {
      "epoch": 16.56105610561056,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.0026218983448016903,
      "loss": 0.2303,
      "num_input_tokens_seen": 31769984,
      "step": 150540
    },
    {
      "epoch": 16.561606160616062,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.0026210850226126912,
      "loss": 0.2324,
      "num_input_tokens_seen": 31771040,
      "step": 150545
    },
    {
      "epoch": 16.562156215621563,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.002620271814514208,
      "loss": 0.2303,
      "num_input_tokens_seen": 31772064,
      "step": 150550
    },
    {
      "epoch": 16.562706270627064,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0026194587205137436,
      "loss": 0.2319,
      "num_input_tokens_seen": 31773056,
      "step": 150555
    },
    {
      "epoch": 16.563256325632562,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.002618645740618783,
      "loss": 0.2303,
      "num_input_tokens_seen": 31774112,
      "step": 150560
    },
    {
      "epoch": 16.563806380638063,
      "grad_norm": 0.000732421875,
      "learning_rate": 0.002617832874836823,
      "loss": 0.2324,
      "num_input_tokens_seen": 31775168,
      "step": 150565
    },
    {
      "epoch": 16.564356435643564,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.002617020123175359,
      "loss": 0.2324,
      "num_input_tokens_seen": 31776256,
      "step": 150570
    },
    {
      "epoch": 16.564906490649065,
      "grad_norm": 0.00051116943359375,
      "learning_rate": 0.002616207485641876,
      "loss": 0.2324,
      "num_input_tokens_seen": 31777312,
      "step": 150575
    },
    {
      "epoch": 16.565456545654566,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.002615394962243869,
      "loss": 0.2309,
      "num_input_tokens_seen": 31778464,
      "step": 150580
    },
    {
      "epoch": 16.566006600660067,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0026145825529888247,
      "loss": 0.2298,
      "num_input_tokens_seen": 31779616,
      "step": 150585
    },
    {
      "epoch": 16.566556655665565,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0026137702578842253,
      "loss": 0.2319,
      "num_input_tokens_seen": 31780640,
      "step": 150590
    },
    {
      "epoch": 16.567106710671066,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.002612958076937564,
      "loss": 0.2324,
      "num_input_tokens_seen": 31781760,
      "step": 150595
    },
    {
      "epoch": 16.567656765676567,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0026121460101563225,
      "loss": 0.2303,
      "num_input_tokens_seen": 31782848,
      "step": 150600
    },
    {
      "epoch": 16.568206820682068,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.002611334057547991,
      "loss": 0.2319,
      "num_input_tokens_seen": 31783872,
      "step": 150605
    },
    {
      "epoch": 16.56875687568757,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0026105222191200495,
      "loss": 0.2329,
      "num_input_tokens_seen": 31784992,
      "step": 150610
    },
    {
      "epoch": 16.56930693069307,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.0026097104948799737,
      "loss": 0.2324,
      "num_input_tokens_seen": 31785984,
      "step": 150615
    },
    {
      "epoch": 16.56985698569857,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.002608898884835253,
      "loss": 0.2308,
      "num_input_tokens_seen": 31787072,
      "step": 150620
    },
    {
      "epoch": 16.57040704070407,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002608087388993366,
      "loss": 0.2308,
      "num_input_tokens_seen": 31788128,
      "step": 150625
    },
    {
      "epoch": 16.57095709570957,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.0026072760073617934,
      "loss": 0.2329,
      "num_input_tokens_seen": 31789248,
      "step": 150630
    },
    {
      "epoch": 16.57150715071507,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.002606464739948011,
      "loss": 0.2303,
      "num_input_tokens_seen": 31790368,
      "step": 150635
    },
    {
      "epoch": 16.572057205720572,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0026056535867594933,
      "loss": 0.2314,
      "num_input_tokens_seen": 31791488,
      "step": 150640
    },
    {
      "epoch": 16.572607260726073,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0026048425478037234,
      "loss": 0.2324,
      "num_input_tokens_seen": 31792512,
      "step": 150645
    },
    {
      "epoch": 16.573157315731574,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0026040316230881676,
      "loss": 0.2303,
      "num_input_tokens_seen": 31793536,
      "step": 150650
    },
    {
      "epoch": 16.573707370737075,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0026032208126203034,
      "loss": 0.2314,
      "num_input_tokens_seen": 31794496,
      "step": 150655
    },
    {
      "epoch": 16.574257425742573,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.002602410116407609,
      "loss": 0.2298,
      "num_input_tokens_seen": 31795616,
      "step": 150660
    },
    {
      "epoch": 16.574807480748074,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.0026015995344575477,
      "loss": 0.2314,
      "num_input_tokens_seen": 31796672,
      "step": 150665
    },
    {
      "epoch": 16.575357535753575,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002600789066777598,
      "loss": 0.2319,
      "num_input_tokens_seen": 31797792,
      "step": 150670
    },
    {
      "epoch": 16.575907590759076,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.002599978713375222,
      "loss": 0.2298,
      "num_input_tokens_seen": 31798848,
      "step": 150675
    },
    {
      "epoch": 16.576457645764577,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0025991684742578914,
      "loss": 0.2314,
      "num_input_tokens_seen": 31799872,
      "step": 150680
    },
    {
      "epoch": 16.57700770077008,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.002598358349433079,
      "loss": 0.2335,
      "num_input_tokens_seen": 31800992,
      "step": 150685
    },
    {
      "epoch": 16.577557755775576,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.002597548338908243,
      "loss": 0.2324,
      "num_input_tokens_seen": 31802048,
      "step": 150690
    },
    {
      "epoch": 16.578107810781077,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0025967384426908557,
      "loss": 0.2319,
      "num_input_tokens_seen": 31803040,
      "step": 150695
    },
    {
      "epoch": 16.578657865786578,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.002595928660788376,
      "loss": 0.2314,
      "num_input_tokens_seen": 31804128,
      "step": 150700
    },
    {
      "epoch": 16.57920792079208,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.002595118993208272,
      "loss": 0.2314,
      "num_input_tokens_seen": 31805152,
      "step": 150705
    },
    {
      "epoch": 16.57975797579758,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.0025943094399580014,
      "loss": 0.234,
      "num_input_tokens_seen": 31806208,
      "step": 150710
    },
    {
      "epoch": 16.58030803080308,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.002593500001045028,
      "loss": 0.2314,
      "num_input_tokens_seen": 31807264,
      "step": 150715
    },
    {
      "epoch": 16.580858085808583,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.0025926906764768155,
      "loss": 0.2308,
      "num_input_tokens_seen": 31808320,
      "step": 150720
    },
    {
      "epoch": 16.58140814081408,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.002591881466260815,
      "loss": 0.2314,
      "num_input_tokens_seen": 31809408,
      "step": 150725
    },
    {
      "epoch": 16.58195819581958,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0025910723704044916,
      "loss": 0.2303,
      "num_input_tokens_seen": 31810496,
      "step": 150730
    },
    {
      "epoch": 16.582508250825082,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0025902633889153025,
      "loss": 0.2324,
      "num_input_tokens_seen": 31811584,
      "step": 150735
    },
    {
      "epoch": 16.583058305830583,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0025894545218006913,
      "loss": 0.2319,
      "num_input_tokens_seen": 31812704,
      "step": 150740
    },
    {
      "epoch": 16.583608360836084,
      "grad_norm": 0.005859375,
      "learning_rate": 0.002588645769068131,
      "loss": 0.2324,
      "num_input_tokens_seen": 31813696,
      "step": 150745
    },
    {
      "epoch": 16.584158415841586,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.0025878371307250644,
      "loss": 0.2314,
      "num_input_tokens_seen": 31814752,
      "step": 150750
    },
    {
      "epoch": 16.584708470847083,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0025870286067789486,
      "loss": 0.2319,
      "num_input_tokens_seen": 31815840,
      "step": 150755
    },
    {
      "epoch": 16.585258525852584,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0025862201972372354,
      "loss": 0.2324,
      "num_input_tokens_seen": 31816928,
      "step": 150760
    },
    {
      "epoch": 16.585808580858085,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.002585411902107369,
      "loss": 0.2308,
      "num_input_tokens_seen": 31818048,
      "step": 150765
    },
    {
      "epoch": 16.586358635863586,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0025846037213968047,
      "loss": 0.2308,
      "num_input_tokens_seen": 31819072,
      "step": 150770
    },
    {
      "epoch": 16.586908690869087,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0025837956551129954,
      "loss": 0.2303,
      "num_input_tokens_seen": 31820096,
      "step": 150775
    },
    {
      "epoch": 16.58745874587459,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.002582987703263378,
      "loss": 0.2314,
      "num_input_tokens_seen": 31821184,
      "step": 150780
    },
    {
      "epoch": 16.58800880088009,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.002582179865855408,
      "loss": 0.2314,
      "num_input_tokens_seen": 31822144,
      "step": 150785
    },
    {
      "epoch": 16.588558855885587,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0025813721428965255,
      "loss": 0.2293,
      "num_input_tokens_seen": 31823200,
      "step": 150790
    },
    {
      "epoch": 16.58910891089109,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0025805645343941774,
      "loss": 0.234,
      "num_input_tokens_seen": 31824256,
      "step": 150795
    },
    {
      "epoch": 16.58965896589659,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0025797570403558088,
      "loss": 0.2314,
      "num_input_tokens_seen": 31825312,
      "step": 150800
    },
    {
      "epoch": 16.59020902090209,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.002578949660788856,
      "loss": 0.2324,
      "num_input_tokens_seen": 31826304,
      "step": 150805
    },
    {
      "epoch": 16.59075907590759,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002578142395700768,
      "loss": 0.2288,
      "num_input_tokens_seen": 31827360,
      "step": 150810
    },
    {
      "epoch": 16.591309130913093,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.002577335245098977,
      "loss": 0.2319,
      "num_input_tokens_seen": 31828448,
      "step": 150815
    },
    {
      "epoch": 16.59185918591859,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0025765282089909307,
      "loss": 0.2308,
      "num_input_tokens_seen": 31829536,
      "step": 150820
    },
    {
      "epoch": 16.59240924092409,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002575721287384058,
      "loss": 0.2303,
      "num_input_tokens_seen": 31830624,
      "step": 150825
    },
    {
      "epoch": 16.592959295929592,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.002574914480285801,
      "loss": 0.2298,
      "num_input_tokens_seen": 31831648,
      "step": 150830
    },
    {
      "epoch": 16.593509350935093,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.002574107787703599,
      "loss": 0.2329,
      "num_input_tokens_seen": 31832704,
      "step": 150835
    },
    {
      "epoch": 16.594059405940595,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.002573301209644879,
      "loss": 0.2303,
      "num_input_tokens_seen": 31833760,
      "step": 150840
    },
    {
      "epoch": 16.594609460946096,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.0025724947461170824,
      "loss": 0.2324,
      "num_input_tokens_seen": 31834816,
      "step": 150845
    },
    {
      "epoch": 16.595159515951597,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0025716883971276384,
      "loss": 0.2304,
      "num_input_tokens_seen": 31835840,
      "step": 150850
    },
    {
      "epoch": 16.595709570957094,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.00257088216268397,
      "loss": 0.2309,
      "num_input_tokens_seen": 31836896,
      "step": 150855
    },
    {
      "epoch": 16.596259625962595,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.0025700760427935254,
      "loss": 0.2324,
      "num_input_tokens_seen": 31837920,
      "step": 150860
    },
    {
      "epoch": 16.596809680968097,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0025692700374637216,
      "loss": 0.2308,
      "num_input_tokens_seen": 31839008,
      "step": 150865
    },
    {
      "epoch": 16.597359735973598,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0025684641467019964,
      "loss": 0.2324,
      "num_input_tokens_seen": 31840064,
      "step": 150870
    },
    {
      "epoch": 16.5979097909791,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00256765837051577,
      "loss": 0.2324,
      "num_input_tokens_seen": 31841088,
      "step": 150875
    },
    {
      "epoch": 16.5984598459846,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.002566852708912467,
      "loss": 0.2298,
      "num_input_tokens_seen": 31842112,
      "step": 150880
    },
    {
      "epoch": 16.599009900990097,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.002566047161899518,
      "loss": 0.2319,
      "num_input_tokens_seen": 31843200,
      "step": 150885
    },
    {
      "epoch": 16.5995599559956,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0025652417294843447,
      "loss": 0.2324,
      "num_input_tokens_seen": 31844320,
      "step": 150890
    },
    {
      "epoch": 16.6001100110011,
      "grad_norm": 0.002655029296875,
      "learning_rate": 0.0025644364116743754,
      "loss": 0.2298,
      "num_input_tokens_seen": 31845344,
      "step": 150895
    },
    {
      "epoch": 16.6006600660066,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.002563631208477029,
      "loss": 0.2314,
      "num_input_tokens_seen": 31846336,
      "step": 150900
    },
    {
      "epoch": 16.6012101210121,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002562826119899721,
      "loss": 0.2319,
      "num_input_tokens_seen": 31847392,
      "step": 150905
    },
    {
      "epoch": 16.601760176017603,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.002562021145949882,
      "loss": 0.2303,
      "num_input_tokens_seen": 31848448,
      "step": 150910
    },
    {
      "epoch": 16.602310231023104,
      "grad_norm": 0.00066375732421875,
      "learning_rate": 0.002561216286634918,
      "loss": 0.2309,
      "num_input_tokens_seen": 31849504,
      "step": 150915
    },
    {
      "epoch": 16.6028602860286,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.002560411541962263,
      "loss": 0.2293,
      "num_input_tokens_seen": 31850592,
      "step": 150920
    },
    {
      "epoch": 16.603410341034103,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.0025596069119393243,
      "loss": 0.2319,
      "num_input_tokens_seen": 31851648,
      "step": 150925
    },
    {
      "epoch": 16.603960396039604,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.002558802396573516,
      "loss": 0.2308,
      "num_input_tokens_seen": 31852672,
      "step": 150930
    },
    {
      "epoch": 16.604510451045105,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.002557997995872259,
      "loss": 0.2303,
      "num_input_tokens_seen": 31853728,
      "step": 150935
    },
    {
      "epoch": 16.605060506050606,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0025571937098429623,
      "loss": 0.2324,
      "num_input_tokens_seen": 31854720,
      "step": 150940
    },
    {
      "epoch": 16.605610561056107,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00255638953849304,
      "loss": 0.2335,
      "num_input_tokens_seen": 31855808,
      "step": 150945
    },
    {
      "epoch": 16.606160616061608,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002555585481829908,
      "loss": 0.2309,
      "num_input_tokens_seen": 31856832,
      "step": 150950
    },
    {
      "epoch": 16.606710671067106,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.002554781539860968,
      "loss": 0.2319,
      "num_input_tokens_seen": 31857824,
      "step": 150955
    },
    {
      "epoch": 16.607260726072607,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0025539777125936397,
      "loss": 0.2314,
      "num_input_tokens_seen": 31858944,
      "step": 150960
    },
    {
      "epoch": 16.607810781078108,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.002553174000035327,
      "loss": 0.2303,
      "num_input_tokens_seen": 31860000,
      "step": 150965
    },
    {
      "epoch": 16.60836083608361,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.002552370402193432,
      "loss": 0.2298,
      "num_input_tokens_seen": 31861024,
      "step": 150970
    },
    {
      "epoch": 16.60891089108911,
      "grad_norm": 0.006561279296875,
      "learning_rate": 0.0025515669190753682,
      "loss": 0.2309,
      "num_input_tokens_seen": 31862080,
      "step": 150975
    },
    {
      "epoch": 16.60946094609461,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0025507635506885366,
      "loss": 0.2314,
      "num_input_tokens_seen": 31863136,
      "step": 150980
    },
    {
      "epoch": 16.61001100110011,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.002549960297040349,
      "loss": 0.2277,
      "num_input_tokens_seen": 31864256,
      "step": 150985
    },
    {
      "epoch": 16.61056105610561,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.002549157158138202,
      "loss": 0.2303,
      "num_input_tokens_seen": 31865280,
      "step": 150990
    },
    {
      "epoch": 16.61111111111111,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0025483541339894954,
      "loss": 0.2314,
      "num_input_tokens_seen": 31866400,
      "step": 150995
    },
    {
      "epoch": 16.611661166116612,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0025475512246016345,
      "loss": 0.2299,
      "num_input_tokens_seen": 31867424,
      "step": 151000
    },
    {
      "epoch": 16.612211221122113,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0025467484299820185,
      "loss": 0.2293,
      "num_input_tokens_seen": 31868480,
      "step": 151005
    },
    {
      "epoch": 16.612761276127614,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0025459457501380497,
      "loss": 0.2314,
      "num_input_tokens_seen": 31869632,
      "step": 151010
    },
    {
      "epoch": 16.61331133113311,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0025451431850771243,
      "loss": 0.2319,
      "num_input_tokens_seen": 31870688,
      "step": 151015
    },
    {
      "epoch": 16.613861386138613,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0025443407348066326,
      "loss": 0.2319,
      "num_input_tokens_seen": 31871776,
      "step": 151020
    },
    {
      "epoch": 16.614411441144114,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0025435383993339805,
      "loss": 0.2324,
      "num_input_tokens_seen": 31872800,
      "step": 151025
    },
    {
      "epoch": 16.614961496149615,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002542736178666552,
      "loss": 0.2298,
      "num_input_tokens_seen": 31873824,
      "step": 151030
    },
    {
      "epoch": 16.615511551155116,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002541934072811748,
      "loss": 0.2303,
      "num_input_tokens_seen": 31874880,
      "step": 151035
    },
    {
      "epoch": 16.616061606160617,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002541132081776964,
      "loss": 0.2314,
      "num_input_tokens_seen": 31875936,
      "step": 151040
    },
    {
      "epoch": 16.616611661166118,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0025403302055695826,
      "loss": 0.2319,
      "num_input_tokens_seen": 31876928,
      "step": 151045
    },
    {
      "epoch": 16.617161716171616,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.0025395284441970026,
      "loss": 0.2329,
      "num_input_tokens_seen": 31877984,
      "step": 151050
    },
    {
      "epoch": 16.617711771177117,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.002538726797666607,
      "loss": 0.2298,
      "num_input_tokens_seen": 31879040,
      "step": 151055
    },
    {
      "epoch": 16.618261826182618,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0025379252659857863,
      "loss": 0.2308,
      "num_input_tokens_seen": 31880096,
      "step": 151060
    },
    {
      "epoch": 16.61881188118812,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0025371238491619314,
      "loss": 0.2308,
      "num_input_tokens_seen": 31881152,
      "step": 151065
    },
    {
      "epoch": 16.61936193619362,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.002536322547202424,
      "loss": 0.2309,
      "num_input_tokens_seen": 31882176,
      "step": 151070
    },
    {
      "epoch": 16.61991199119912,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0025355213601146523,
      "loss": 0.2329,
      "num_input_tokens_seen": 31883232,
      "step": 151075
    },
    {
      "epoch": 16.620462046204622,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.002534720287905997,
      "loss": 0.2329,
      "num_input_tokens_seen": 31884224,
      "step": 151080
    },
    {
      "epoch": 16.62101210121012,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.002533919330583847,
      "loss": 0.2329,
      "num_input_tokens_seen": 31885248,
      "step": 151085
    },
    {
      "epoch": 16.62156215621562,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.002533118488155578,
      "loss": 0.2314,
      "num_input_tokens_seen": 31886304,
      "step": 151090
    },
    {
      "epoch": 16.622112211221122,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0025323177606285724,
      "loss": 0.2329,
      "num_input_tokens_seen": 31887360,
      "step": 151095
    },
    {
      "epoch": 16.622662266226623,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.0025315171480102144,
      "loss": 0.2319,
      "num_input_tokens_seen": 31888384,
      "step": 151100
    },
    {
      "epoch": 16.623212321232124,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002530716650307878,
      "loss": 0.2319,
      "num_input_tokens_seen": 31889408,
      "step": 151105
    },
    {
      "epoch": 16.623762376237625,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.002529916267528946,
      "loss": 0.2324,
      "num_input_tokens_seen": 31890496,
      "step": 151110
    },
    {
      "epoch": 16.624312431243123,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.002529115999680787,
      "loss": 0.2314,
      "num_input_tokens_seen": 31891584,
      "step": 151115
    },
    {
      "epoch": 16.624862486248624,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.002528315846770784,
      "loss": 0.2319,
      "num_input_tokens_seen": 31892672,
      "step": 151120
    },
    {
      "epoch": 16.625412541254125,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.002527515808806312,
      "loss": 0.2324,
      "num_input_tokens_seen": 31893728,
      "step": 151125
    },
    {
      "epoch": 16.625962596259626,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0025267158857947376,
      "loss": 0.2298,
      "num_input_tokens_seen": 31894752,
      "step": 151130
    },
    {
      "epoch": 16.626512651265127,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0025259160777434423,
      "loss": 0.2324,
      "num_input_tokens_seen": 31895776,
      "step": 151135
    },
    {
      "epoch": 16.627062706270628,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.002525116384659791,
      "loss": 0.2314,
      "num_input_tokens_seen": 31896832,
      "step": 151140
    },
    {
      "epoch": 16.62761276127613,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0025243168065511544,
      "loss": 0.2298,
      "num_input_tokens_seen": 31897888,
      "step": 151145
    },
    {
      "epoch": 16.628162816281627,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.002523517343424903,
      "loss": 0.2303,
      "num_input_tokens_seen": 31898944,
      "step": 151150
    },
    {
      "epoch": 16.628712871287128,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.0025227179952884075,
      "loss": 0.2319,
      "num_input_tokens_seen": 31900032,
      "step": 151155
    },
    {
      "epoch": 16.62926292629263,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0025219187621490307,
      "loss": 0.2303,
      "num_input_tokens_seen": 31901184,
      "step": 151160
    },
    {
      "epoch": 16.62981298129813,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0025211196440141446,
      "loss": 0.2314,
      "num_input_tokens_seen": 31902272,
      "step": 151165
    },
    {
      "epoch": 16.63036303630363,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.002520320640891107,
      "loss": 0.2314,
      "num_input_tokens_seen": 31903328,
      "step": 151170
    },
    {
      "epoch": 16.630913091309132,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002519521752787287,
      "loss": 0.2308,
      "num_input_tokens_seen": 31904416,
      "step": 151175
    },
    {
      "epoch": 16.63146314631463,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.0025187229797100485,
      "loss": 0.2298,
      "num_input_tokens_seen": 31905440,
      "step": 151180
    },
    {
      "epoch": 16.63201320132013,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0025179243216667474,
      "loss": 0.2319,
      "num_input_tokens_seen": 31906496,
      "step": 151185
    },
    {
      "epoch": 16.632563256325632,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0025171257786647527,
      "loss": 0.2319,
      "num_input_tokens_seen": 31907552,
      "step": 151190
    },
    {
      "epoch": 16.633113311331133,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002516327350711417,
      "loss": 0.2309,
      "num_input_tokens_seen": 31908576,
      "step": 151195
    },
    {
      "epoch": 16.633663366336634,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002515529037814104,
      "loss": 0.2314,
      "num_input_tokens_seen": 31909664,
      "step": 151200
    },
    {
      "epoch": 16.634213421342135,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0025147308399801655,
      "loss": 0.2319,
      "num_input_tokens_seen": 31910720,
      "step": 151205
    },
    {
      "epoch": 16.634763476347636,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.0025139327572169627,
      "loss": 0.2303,
      "num_input_tokens_seen": 31911776,
      "step": 151210
    },
    {
      "epoch": 16.635313531353134,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0025131347895318537,
      "loss": 0.2319,
      "num_input_tokens_seen": 31912864,
      "step": 151215
    },
    {
      "epoch": 16.635863586358635,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0025123369369321863,
      "loss": 0.2303,
      "num_input_tokens_seen": 31913952,
      "step": 151220
    },
    {
      "epoch": 16.636413641364136,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0025115391994253206,
      "loss": 0.2329,
      "num_input_tokens_seen": 31914976,
      "step": 151225
    },
    {
      "epoch": 16.636963696369637,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.002510741577018601,
      "loss": 0.2293,
      "num_input_tokens_seen": 31916032,
      "step": 151230
    },
    {
      "epoch": 16.63751375137514,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002509944069719385,
      "loss": 0.2319,
      "num_input_tokens_seen": 31917088,
      "step": 151235
    },
    {
      "epoch": 16.63806380638064,
      "grad_norm": 0.0028839111328125,
      "learning_rate": 0.0025091466775350227,
      "loss": 0.2303,
      "num_input_tokens_seen": 31918208,
      "step": 151240
    },
    {
      "epoch": 16.638613861386137,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0025083494004728593,
      "loss": 0.2329,
      "num_input_tokens_seen": 31919264,
      "step": 151245
    },
    {
      "epoch": 16.639163916391638,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.00250755223854025,
      "loss": 0.2324,
      "num_input_tokens_seen": 31920320,
      "step": 151250
    },
    {
      "epoch": 16.63971397139714,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0025067551917445363,
      "loss": 0.2303,
      "num_input_tokens_seen": 31921376,
      "step": 151255
    },
    {
      "epoch": 16.64026402640264,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.002505958260093061,
      "loss": 0.2329,
      "num_input_tokens_seen": 31922528,
      "step": 151260
    },
    {
      "epoch": 16.64081408140814,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0025051614435931745,
      "loss": 0.2314,
      "num_input_tokens_seen": 31923552,
      "step": 151265
    },
    {
      "epoch": 16.641364136413642,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0025043647422522187,
      "loss": 0.2298,
      "num_input_tokens_seen": 31924640,
      "step": 151270
    },
    {
      "epoch": 16.641914191419144,
      "grad_norm": 0.005859375,
      "learning_rate": 0.002503568156077539,
      "loss": 0.2304,
      "num_input_tokens_seen": 31925728,
      "step": 151275
    },
    {
      "epoch": 16.64246424642464,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0025027716850764786,
      "loss": 0.2313,
      "num_input_tokens_seen": 31926720,
      "step": 151280
    },
    {
      "epoch": 16.643014301430142,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.002501975329256368,
      "loss": 0.2303,
      "num_input_tokens_seen": 31927776,
      "step": 151285
    },
    {
      "epoch": 16.643564356435643,
      "grad_norm": 0.00079345703125,
      "learning_rate": 0.002501179088624556,
      "loss": 0.2309,
      "num_input_tokens_seen": 31928832,
      "step": 151290
    },
    {
      "epoch": 16.644114411441144,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0025003829631883774,
      "loss": 0.2335,
      "num_input_tokens_seen": 31929920,
      "step": 151295
    },
    {
      "epoch": 16.644664466446645,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0024995869529551755,
      "loss": 0.2319,
      "num_input_tokens_seen": 31930912,
      "step": 151300
    },
    {
      "epoch": 16.645214521452147,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.002498791057932282,
      "loss": 0.2335,
      "num_input_tokens_seen": 31931968,
      "step": 151305
    },
    {
      "epoch": 16.645764576457644,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.0024979952781270286,
      "loss": 0.2303,
      "num_input_tokens_seen": 31933088,
      "step": 151310
    },
    {
      "epoch": 16.646314631463145,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0024971996135467584,
      "loss": 0.2319,
      "num_input_tokens_seen": 31934080,
      "step": 151315
    },
    {
      "epoch": 16.646864686468646,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0024964040641987967,
      "loss": 0.2303,
      "num_input_tokens_seen": 31935136,
      "step": 151320
    },
    {
      "epoch": 16.647414741474147,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.002495608630090478,
      "loss": 0.2308,
      "num_input_tokens_seen": 31936192,
      "step": 151325
    },
    {
      "epoch": 16.64796479647965,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.0024948133112291378,
      "loss": 0.2324,
      "num_input_tokens_seen": 31937216,
      "step": 151330
    },
    {
      "epoch": 16.64851485148515,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0024940181076220985,
      "loss": 0.2324,
      "num_input_tokens_seen": 31938272,
      "step": 151335
    },
    {
      "epoch": 16.64906490649065,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.002493223019276698,
      "loss": 0.2314,
      "num_input_tokens_seen": 31939264,
      "step": 151340
    },
    {
      "epoch": 16.649614961496148,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.002492428046200257,
      "loss": 0.2303,
      "num_input_tokens_seen": 31940288,
      "step": 151345
    },
    {
      "epoch": 16.65016501650165,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0024916331884001024,
      "loss": 0.2324,
      "num_input_tokens_seen": 31941376,
      "step": 151350
    },
    {
      "epoch": 16.65071507150715,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0024908384458835675,
      "loss": 0.2298,
      "num_input_tokens_seen": 31942400,
      "step": 151355
    },
    {
      "epoch": 16.65126512651265,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.002490043818657969,
      "loss": 0.2298,
      "num_input_tokens_seen": 31943424,
      "step": 151360
    },
    {
      "epoch": 16.651815181518153,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.002489249306730637,
      "loss": 0.2298,
      "num_input_tokens_seen": 31944448,
      "step": 151365
    },
    {
      "epoch": 16.652365236523654,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0024884549101088916,
      "loss": 0.2314,
      "num_input_tokens_seen": 31945472,
      "step": 151370
    },
    {
      "epoch": 16.652915291529155,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.0024876606288000485,
      "loss": 0.2303,
      "num_input_tokens_seen": 31946560,
      "step": 151375
    },
    {
      "epoch": 16.653465346534652,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.002486866462811434,
      "loss": 0.2319,
      "num_input_tokens_seen": 31947616,
      "step": 151380
    },
    {
      "epoch": 16.654015401540153,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002486072412150367,
      "loss": 0.2319,
      "num_input_tokens_seen": 31948672,
      "step": 151385
    },
    {
      "epoch": 16.654565456545654,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0024852784768241696,
      "loss": 0.2309,
      "num_input_tokens_seen": 31949792,
      "step": 151390
    },
    {
      "epoch": 16.655115511551156,
      "grad_norm": 0.000881195068359375,
      "learning_rate": 0.002484484656840155,
      "loss": 0.2319,
      "num_input_tokens_seen": 31950816,
      "step": 151395
    },
    {
      "epoch": 16.655665566556657,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002483690952205637,
      "loss": 0.2298,
      "num_input_tokens_seen": 31951936,
      "step": 151400
    },
    {
      "epoch": 16.656215621562158,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0024828973629279333,
      "loss": 0.2314,
      "num_input_tokens_seen": 31952992,
      "step": 151405
    },
    {
      "epoch": 16.656765676567655,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0024821038890143576,
      "loss": 0.2314,
      "num_input_tokens_seen": 31954016,
      "step": 151410
    },
    {
      "epoch": 16.657315731573156,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.002481310530472228,
      "loss": 0.2314,
      "num_input_tokens_seen": 31955040,
      "step": 151415
    },
    {
      "epoch": 16.657865786578657,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0024805172873088527,
      "loss": 0.2324,
      "num_input_tokens_seen": 31956128,
      "step": 151420
    },
    {
      "epoch": 16.65841584158416,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002479724159531538,
      "loss": 0.2308,
      "num_input_tokens_seen": 31957216,
      "step": 151425
    },
    {
      "epoch": 16.65896589658966,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0024789311471476003,
      "loss": 0.2329,
      "num_input_tokens_seen": 31958208,
      "step": 151430
    },
    {
      "epoch": 16.65951595159516,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.002478138250164343,
      "loss": 0.2324,
      "num_input_tokens_seen": 31959232,
      "step": 151435
    },
    {
      "epoch": 16.66006600660066,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.002477345468589078,
      "loss": 0.234,
      "num_input_tokens_seen": 31960352,
      "step": 151440
    },
    {
      "epoch": 16.66061606160616,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0024765528024291144,
      "loss": 0.2293,
      "num_input_tokens_seen": 31961408,
      "step": 151445
    },
    {
      "epoch": 16.66116611661166,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.002475760251691749,
      "loss": 0.2314,
      "num_input_tokens_seen": 31962464,
      "step": 151450
    },
    {
      "epoch": 16.66171617161716,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.0024749678163842963,
      "loss": 0.2324,
      "num_input_tokens_seen": 31963584,
      "step": 151455
    },
    {
      "epoch": 16.662266226622663,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0024741754965140516,
      "loss": 0.2298,
      "num_input_tokens_seen": 31964640,
      "step": 151460
    },
    {
      "epoch": 16.662816281628164,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.002473383292088321,
      "loss": 0.2324,
      "num_input_tokens_seen": 31965632,
      "step": 151465
    },
    {
      "epoch": 16.663366336633665,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00247259120311441,
      "loss": 0.2319,
      "num_input_tokens_seen": 31966720,
      "step": 151470
    },
    {
      "epoch": 16.663916391639162,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0024717992295996095,
      "loss": 0.2308,
      "num_input_tokens_seen": 31967808,
      "step": 151475
    },
    {
      "epoch": 16.664466446644663,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0024710073715512285,
      "loss": 0.2319,
      "num_input_tokens_seen": 31968864,
      "step": 151480
    },
    {
      "epoch": 16.665016501650165,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.002470215628976557,
      "loss": 0.2309,
      "num_input_tokens_seen": 31969920,
      "step": 151485
    },
    {
      "epoch": 16.665566556655666,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0024694240018828984,
      "loss": 0.2298,
      "num_input_tokens_seen": 31970944,
      "step": 151490
    },
    {
      "epoch": 16.666116611661167,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0024686324902775435,
      "loss": 0.2303,
      "num_input_tokens_seen": 31972000,
      "step": 151495
    },
    {
      "epoch": 16.666666666666668,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.0024678410941677887,
      "loss": 0.2324,
      "num_input_tokens_seen": 31973056,
      "step": 151500
    },
    {
      "epoch": 16.66721672167217,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.002467049813560934,
      "loss": 0.2293,
      "num_input_tokens_seen": 31974112,
      "step": 151505
    },
    {
      "epoch": 16.667766776677666,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.002466258648464263,
      "loss": 0.2324,
      "num_input_tokens_seen": 31975168,
      "step": 151510
    },
    {
      "epoch": 16.668316831683168,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.0024654675988850755,
      "loss": 0.2309,
      "num_input_tokens_seen": 31976224,
      "step": 151515
    },
    {
      "epoch": 16.66886688668867,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.002464676664830658,
      "loss": 0.2319,
      "num_input_tokens_seen": 31977280,
      "step": 151520
    },
    {
      "epoch": 16.66941694169417,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.0024638858463082924,
      "loss": 0.2298,
      "num_input_tokens_seen": 31978304,
      "step": 151525
    },
    {
      "epoch": 16.66996699669967,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002463095143325285,
      "loss": 0.2324,
      "num_input_tokens_seen": 31979360,
      "step": 151530
    },
    {
      "epoch": 16.670517051705172,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0024623045558889115,
      "loss": 0.2309,
      "num_input_tokens_seen": 31980448,
      "step": 151535
    },
    {
      "epoch": 16.67106710671067,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0024615140840064595,
      "loss": 0.2314,
      "num_input_tokens_seen": 31981472,
      "step": 151540
    },
    {
      "epoch": 16.67161716171617,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0024607237276852182,
      "loss": 0.2314,
      "num_input_tokens_seen": 31982592,
      "step": 151545
    },
    {
      "epoch": 16.67216721672167,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.002459933486932465,
      "loss": 0.2314,
      "num_input_tokens_seen": 31983680,
      "step": 151550
    },
    {
      "epoch": 16.672717271727173,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0024591433617554893,
      "loss": 0.2324,
      "num_input_tokens_seen": 31984704,
      "step": 151555
    },
    {
      "epoch": 16.673267326732674,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0024583533521615746,
      "loss": 0.2303,
      "num_input_tokens_seen": 31985760,
      "step": 151560
    },
    {
      "epoch": 16.673817381738175,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.002457563458157994,
      "loss": 0.2324,
      "num_input_tokens_seen": 31986880,
      "step": 151565
    },
    {
      "epoch": 16.674367436743676,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.0024567736797520365,
      "loss": 0.2324,
      "num_input_tokens_seen": 31987936,
      "step": 151570
    },
    {
      "epoch": 16.674917491749174,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.002455984016950975,
      "loss": 0.2309,
      "num_input_tokens_seen": 31988992,
      "step": 151575
    },
    {
      "epoch": 16.675467546754675,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.00245519446976209,
      "loss": 0.2319,
      "num_input_tokens_seen": 31989984,
      "step": 151580
    },
    {
      "epoch": 16.676017601760176,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00245440503819266,
      "loss": 0.2319,
      "num_input_tokens_seen": 31991008,
      "step": 151585
    },
    {
      "epoch": 16.676567656765677,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0024536157222499553,
      "loss": 0.2308,
      "num_input_tokens_seen": 31992064,
      "step": 151590
    },
    {
      "epoch": 16.677117711771178,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002452826521941258,
      "loss": 0.2335,
      "num_input_tokens_seen": 31993088,
      "step": 151595
    },
    {
      "epoch": 16.67766776677668,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002452037437273834,
      "loss": 0.2329,
      "num_input_tokens_seen": 31994112,
      "step": 151600
    },
    {
      "epoch": 16.678217821782177,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.002451248468254964,
      "loss": 0.2298,
      "num_input_tokens_seen": 31995200,
      "step": 151605
    },
    {
      "epoch": 16.678767876787678,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.002450459614891912,
      "loss": 0.2314,
      "num_input_tokens_seen": 31996256,
      "step": 151610
    },
    {
      "epoch": 16.67931793179318,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.002449670877191951,
      "loss": 0.233,
      "num_input_tokens_seen": 31997280,
      "step": 151615
    },
    {
      "epoch": 16.67986798679868,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.002448882255162355,
      "loss": 0.2288,
      "num_input_tokens_seen": 31998304,
      "step": 151620
    },
    {
      "epoch": 16.68041804180418,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.0024480937488103867,
      "loss": 0.2324,
      "num_input_tokens_seen": 31999392,
      "step": 151625
    },
    {
      "epoch": 16.680968096809682,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.002447305358143316,
      "loss": 0.234,
      "num_input_tokens_seen": 32000480,
      "step": 151630
    },
    {
      "epoch": 16.681518151815183,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.002446517083168412,
      "loss": 0.2303,
      "num_input_tokens_seen": 32001504,
      "step": 151635
    },
    {
      "epoch": 16.68206820682068,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0024457289238929256,
      "loss": 0.2324,
      "num_input_tokens_seen": 32002624,
      "step": 151640
    },
    {
      "epoch": 16.682618261826182,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0024449408803241405,
      "loss": 0.2314,
      "num_input_tokens_seen": 32003616,
      "step": 151645
    },
    {
      "epoch": 16.683168316831683,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.002444152952469307,
      "loss": 0.2329,
      "num_input_tokens_seen": 32004640,
      "step": 151650
    },
    {
      "epoch": 16.683718371837184,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0024433651403356947,
      "loss": 0.2314,
      "num_input_tokens_seen": 32005760,
      "step": 151655
    },
    {
      "epoch": 16.684268426842685,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0024425774439305607,
      "loss": 0.2314,
      "num_input_tokens_seen": 32006816,
      "step": 151660
    },
    {
      "epoch": 16.684818481848186,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002441789863261161,
      "loss": 0.2308,
      "num_input_tokens_seen": 32007904,
      "step": 151665
    },
    {
      "epoch": 16.685368536853684,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.0024410023983347582,
      "loss": 0.2309,
      "num_input_tokens_seen": 32008928,
      "step": 151670
    },
    {
      "epoch": 16.685918591859185,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.00244021504915861,
      "loss": 0.2319,
      "num_input_tokens_seen": 32010016,
      "step": 151675
    },
    {
      "epoch": 16.686468646864686,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0024394278157399765,
      "loss": 0.2313,
      "num_input_tokens_seen": 32011104,
      "step": 151680
    },
    {
      "epoch": 16.687018701870187,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.0024386406980861095,
      "loss": 0.2329,
      "num_input_tokens_seen": 32012160,
      "step": 151685
    },
    {
      "epoch": 16.687568756875688,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0024378536962042594,
      "loss": 0.2335,
      "num_input_tokens_seen": 32013152,
      "step": 151690
    },
    {
      "epoch": 16.68811881188119,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.0024370668101016887,
      "loss": 0.2324,
      "num_input_tokens_seen": 32014272,
      "step": 151695
    },
    {
      "epoch": 16.68866886688669,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.002436280039785637,
      "loss": 0.2319,
      "num_input_tokens_seen": 32015296,
      "step": 151700
    },
    {
      "epoch": 16.689218921892188,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0024354933852633698,
      "loss": 0.2303,
      "num_input_tokens_seen": 32016384,
      "step": 151705
    },
    {
      "epoch": 16.68976897689769,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.002434706846542133,
      "loss": 0.2304,
      "num_input_tokens_seen": 32017376,
      "step": 151710
    },
    {
      "epoch": 16.69031903190319,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0024339204236291686,
      "loss": 0.2314,
      "num_input_tokens_seen": 32018432,
      "step": 151715
    },
    {
      "epoch": 16.69086908690869,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.002433134116531734,
      "loss": 0.2319,
      "num_input_tokens_seen": 32019488,
      "step": 151720
    },
    {
      "epoch": 16.691419141914192,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0024323479252570685,
      "loss": 0.2314,
      "num_input_tokens_seen": 32020576,
      "step": 151725
    },
    {
      "epoch": 16.691969196919693,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.002431561849812422,
      "loss": 0.2319,
      "num_input_tokens_seen": 32021696,
      "step": 151730
    },
    {
      "epoch": 16.69251925192519,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0024307758902050417,
      "loss": 0.2319,
      "num_input_tokens_seen": 32022784,
      "step": 151735
    },
    {
      "epoch": 16.693069306930692,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0024299900464421673,
      "loss": 0.2319,
      "num_input_tokens_seen": 32023808,
      "step": 151740
    },
    {
      "epoch": 16.693619361936193,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.0024292043185310448,
      "loss": 0.2304,
      "num_input_tokens_seen": 32024832,
      "step": 151745
    },
    {
      "epoch": 16.694169416941694,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0024284187064789143,
      "loss": 0.2308,
      "num_input_tokens_seen": 32025856,
      "step": 151750
    },
    {
      "epoch": 16.694719471947195,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002427633210293012,
      "loss": 0.2324,
      "num_input_tokens_seen": 32026880,
      "step": 151755
    },
    {
      "epoch": 16.695269526952696,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0024268478299805823,
      "loss": 0.2314,
      "num_input_tokens_seen": 32027968,
      "step": 151760
    },
    {
      "epoch": 16.695819581958197,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0024260625655488626,
      "loss": 0.2324,
      "num_input_tokens_seen": 32029056,
      "step": 151765
    },
    {
      "epoch": 16.696369636963695,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0024252774170050936,
      "loss": 0.2293,
      "num_input_tokens_seen": 32030080,
      "step": 151770
    },
    {
      "epoch": 16.696919691969196,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.00242449238435651,
      "loss": 0.2324,
      "num_input_tokens_seen": 32031200,
      "step": 151775
    },
    {
      "epoch": 16.697469746974697,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.0024237074676103426,
      "loss": 0.2314,
      "num_input_tokens_seen": 32032320,
      "step": 151780
    },
    {
      "epoch": 16.698019801980198,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.002422922666773827,
      "loss": 0.2314,
      "num_input_tokens_seen": 32033408,
      "step": 151785
    },
    {
      "epoch": 16.6985698569857,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0024221379818541986,
      "loss": 0.2314,
      "num_input_tokens_seen": 32034400,
      "step": 151790
    },
    {
      "epoch": 16.6991199119912,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0024213534128586945,
      "loss": 0.2308,
      "num_input_tokens_seen": 32035520,
      "step": 151795
    },
    {
      "epoch": 16.6996699669967,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002420568959794538,
      "loss": 0.2314,
      "num_input_tokens_seen": 32036608,
      "step": 151800
    },
    {
      "epoch": 16.7002200220022,
      "grad_norm": 0.001953125,
      "learning_rate": 0.0024197846226689584,
      "loss": 0.2335,
      "num_input_tokens_seen": 32037696,
      "step": 151805
    },
    {
      "epoch": 16.7007700770077,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.00241900040148919,
      "loss": 0.2309,
      "num_input_tokens_seen": 32038720,
      "step": 151810
    },
    {
      "epoch": 16.7013201320132,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.002418216296262456,
      "loss": 0.233,
      "num_input_tokens_seen": 32039776,
      "step": 151815
    },
    {
      "epoch": 16.701870187018702,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0024174323069959844,
      "loss": 0.2309,
      "num_input_tokens_seen": 32040864,
      "step": 151820
    },
    {
      "epoch": 16.702420242024203,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.002416648433697004,
      "loss": 0.234,
      "num_input_tokens_seen": 32041952,
      "step": 151825
    },
    {
      "epoch": 16.702970297029704,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.002415864676372735,
      "loss": 0.2319,
      "num_input_tokens_seen": 32043072,
      "step": 151830
    },
    {
      "epoch": 16.703520352035202,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0024150810350304064,
      "loss": 0.2314,
      "num_input_tokens_seen": 32044128,
      "step": 151835
    },
    {
      "epoch": 16.704070407040703,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.002414297509677231,
      "loss": 0.2304,
      "num_input_tokens_seen": 32045184,
      "step": 151840
    },
    {
      "epoch": 16.704620462046204,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0024135141003204384,
      "loss": 0.2309,
      "num_input_tokens_seen": 32046336,
      "step": 151845
    },
    {
      "epoch": 16.705170517051705,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.002412730806967251,
      "loss": 0.2314,
      "num_input_tokens_seen": 32047392,
      "step": 151850
    },
    {
      "epoch": 16.705720572057206,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.0024119476296248794,
      "loss": 0.2308,
      "num_input_tokens_seen": 32048448,
      "step": 151855
    },
    {
      "epoch": 16.706270627062707,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.0024111645683005493,
      "loss": 0.2314,
      "num_input_tokens_seen": 32049504,
      "step": 151860
    },
    {
      "epoch": 16.706820682068205,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.002410381623001471,
      "loss": 0.2314,
      "num_input_tokens_seen": 32050624,
      "step": 151865
    },
    {
      "epoch": 16.707370737073706,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.002409598793734867,
      "loss": 0.2319,
      "num_input_tokens_seen": 32051680,
      "step": 151870
    },
    {
      "epoch": 16.707920792079207,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0024088160805079482,
      "loss": 0.2319,
      "num_input_tokens_seen": 32052736,
      "step": 151875
    },
    {
      "epoch": 16.70847084708471,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.002408033483327928,
      "loss": 0.2314,
      "num_input_tokens_seen": 32053792,
      "step": 151880
    },
    {
      "epoch": 16.70902090209021,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.002407251002202024,
      "loss": 0.2309,
      "num_input_tokens_seen": 32054848,
      "step": 151885
    },
    {
      "epoch": 16.70957095709571,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0024064686371374428,
      "loss": 0.2324,
      "num_input_tokens_seen": 32055872,
      "step": 151890
    },
    {
      "epoch": 16.71012101210121,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.0024056863881414007,
      "loss": 0.2298,
      "num_input_tokens_seen": 32056928,
      "step": 151895
    },
    {
      "epoch": 16.71067106710671,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.0024049042552210997,
      "loss": 0.2303,
      "num_input_tokens_seen": 32057952,
      "step": 151900
    },
    {
      "epoch": 16.71122112211221,
      "grad_norm": 0.002899169921875,
      "learning_rate": 0.0024041222383837536,
      "loss": 0.2314,
      "num_input_tokens_seen": 32059040,
      "step": 151905
    },
    {
      "epoch": 16.71177117711771,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0024033403376365714,
      "loss": 0.2314,
      "num_input_tokens_seen": 32060096,
      "step": 151910
    },
    {
      "epoch": 16.712321232123212,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0024025585529867557,
      "loss": 0.2319,
      "num_input_tokens_seen": 32061184,
      "step": 151915
    },
    {
      "epoch": 16.712871287128714,
      "grad_norm": 0.000659942626953125,
      "learning_rate": 0.0024017768844415113,
      "loss": 0.2314,
      "num_input_tokens_seen": 32062336,
      "step": 151920
    },
    {
      "epoch": 16.713421342134215,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0024009953320080473,
      "loss": 0.2314,
      "num_input_tokens_seen": 32063456,
      "step": 151925
    },
    {
      "epoch": 16.713971397139716,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0024002138956935593,
      "loss": 0.2329,
      "num_input_tokens_seen": 32064544,
      "step": 151930
    },
    {
      "epoch": 16.714521452145213,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.0023994325755052545,
      "loss": 0.2308,
      "num_input_tokens_seen": 32065632,
      "step": 151935
    },
    {
      "epoch": 16.715071507150714,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0023986513714503365,
      "loss": 0.2324,
      "num_input_tokens_seen": 32066688,
      "step": 151940
    },
    {
      "epoch": 16.715621562156215,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.002397870283535999,
      "loss": 0.2303,
      "num_input_tokens_seen": 32067776,
      "step": 151945
    },
    {
      "epoch": 16.716171617161717,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0023970893117694473,
      "loss": 0.2309,
      "num_input_tokens_seen": 32068800,
      "step": 151950
    },
    {
      "epoch": 16.716721672167218,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0023963084561578717,
      "loss": 0.2298,
      "num_input_tokens_seen": 32069888,
      "step": 151955
    },
    {
      "epoch": 16.71727172717272,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0023955277167084744,
      "loss": 0.2314,
      "num_input_tokens_seen": 32070944,
      "step": 151960
    },
    {
      "epoch": 16.717821782178216,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0023947470934284542,
      "loss": 0.2324,
      "num_input_tokens_seen": 32072032,
      "step": 151965
    },
    {
      "epoch": 16.718371837183717,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.002393966586324996,
      "loss": 0.233,
      "num_input_tokens_seen": 32073088,
      "step": 151970
    },
    {
      "epoch": 16.71892189218922,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.0023931861954053033,
      "loss": 0.2309,
      "num_input_tokens_seen": 32074144,
      "step": 151975
    },
    {
      "epoch": 16.71947194719472,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0023924059206765595,
      "loss": 0.2303,
      "num_input_tokens_seen": 32075200,
      "step": 151980
    },
    {
      "epoch": 16.72002200220022,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.0023916257621459653,
      "loss": 0.2335,
      "num_input_tokens_seen": 32076256,
      "step": 151985
    },
    {
      "epoch": 16.72057205720572,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0023908457198207026,
      "loss": 0.233,
      "num_input_tokens_seen": 32077312,
      "step": 151990
    },
    {
      "epoch": 16.721122112211223,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.002390065793707965,
      "loss": 0.2313,
      "num_input_tokens_seen": 32078368,
      "step": 151995
    },
    {
      "epoch": 16.72167216721672,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.002389285983814943,
      "loss": 0.2324,
      "num_input_tokens_seen": 32079392,
      "step": 152000
    },
    {
      "epoch": 16.72222222222222,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0023885062901488167,
      "loss": 0.2314,
      "num_input_tokens_seen": 32080512,
      "step": 152005
    },
    {
      "epoch": 16.722772277227723,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0023877267127167816,
      "loss": 0.2309,
      "num_input_tokens_seen": 32081600,
      "step": 152010
    },
    {
      "epoch": 16.723322332233224,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.0023869472515260113,
      "loss": 0.2314,
      "num_input_tokens_seen": 32082656,
      "step": 152015
    },
    {
      "epoch": 16.723872387238725,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.0023861679065836985,
      "loss": 0.2303,
      "num_input_tokens_seen": 32083648,
      "step": 152020
    },
    {
      "epoch": 16.724422442244226,
      "grad_norm": 0.0005645751953125,
      "learning_rate": 0.0023853886778970263,
      "loss": 0.2314,
      "num_input_tokens_seen": 32084736,
      "step": 152025
    },
    {
      "epoch": 16.724972497249723,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00238460956547317,
      "loss": 0.2324,
      "num_input_tokens_seen": 32085760,
      "step": 152030
    },
    {
      "epoch": 16.725522552255224,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.002383830569319315,
      "loss": 0.2314,
      "num_input_tokens_seen": 32086784,
      "step": 152035
    },
    {
      "epoch": 16.726072607260726,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0023830516894426433,
      "loss": 0.2293,
      "num_input_tokens_seen": 32087776,
      "step": 152040
    },
    {
      "epoch": 16.726622662266227,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.002382272925850324,
      "loss": 0.2324,
      "num_input_tokens_seen": 32088832,
      "step": 152045
    },
    {
      "epoch": 16.727172717271728,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0023814942785495424,
      "loss": 0.2319,
      "num_input_tokens_seen": 32089920,
      "step": 152050
    },
    {
      "epoch": 16.72772277227723,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.0023807157475474717,
      "loss": 0.2324,
      "num_input_tokens_seen": 32090944,
      "step": 152055
    },
    {
      "epoch": 16.72827282728273,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.0023799373328512925,
      "loss": 0.2335,
      "num_input_tokens_seen": 32092032,
      "step": 152060
    },
    {
      "epoch": 16.728822882288227,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0023791590344681758,
      "loss": 0.2303,
      "num_input_tokens_seen": 32093088,
      "step": 152065
    },
    {
      "epoch": 16.72937293729373,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.0023783808524052907,
      "loss": 0.2319,
      "num_input_tokens_seen": 32094112,
      "step": 152070
    },
    {
      "epoch": 16.72992299229923,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.0023776027866698117,
      "loss": 0.2319,
      "num_input_tokens_seen": 32095168,
      "step": 152075
    },
    {
      "epoch": 16.73047304730473,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0023768248372689114,
      "loss": 0.2314,
      "num_input_tokens_seen": 32096192,
      "step": 152080
    },
    {
      "epoch": 16.731023102310232,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.002376047004209764,
      "loss": 0.2329,
      "num_input_tokens_seen": 32097280,
      "step": 152085
    },
    {
      "epoch": 16.731573157315733,
      "grad_norm": 0.00058746337890625,
      "learning_rate": 0.002375269287499532,
      "loss": 0.2319,
      "num_input_tokens_seen": 32098304,
      "step": 152090
    },
    {
      "epoch": 16.73212321232123,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0023744916871453834,
      "loss": 0.2309,
      "num_input_tokens_seen": 32099296,
      "step": 152095
    },
    {
      "epoch": 16.73267326732673,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.0023737142031544887,
      "loss": 0.2314,
      "num_input_tokens_seen": 32100288,
      "step": 152100
    },
    {
      "epoch": 16.733223322332233,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.002372936835534009,
      "loss": 0.2314,
      "num_input_tokens_seen": 32101376,
      "step": 152105
    },
    {
      "epoch": 16.733773377337734,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0023721595842911103,
      "loss": 0.2303,
      "num_input_tokens_seen": 32102400,
      "step": 152110
    },
    {
      "epoch": 16.734323432343235,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0023713824494329613,
      "loss": 0.2298,
      "num_input_tokens_seen": 32103392,
      "step": 152115
    },
    {
      "epoch": 16.734873487348736,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002370605430966717,
      "loss": 0.2314,
      "num_input_tokens_seen": 32104416,
      "step": 152120
    },
    {
      "epoch": 16.735423542354237,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0023698285288995466,
      "loss": 0.2314,
      "num_input_tokens_seen": 32105472,
      "step": 152125
    },
    {
      "epoch": 16.735973597359735,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.0023690517432386027,
      "loss": 0.2304,
      "num_input_tokens_seen": 32106496,
      "step": 152130
    },
    {
      "epoch": 16.736523652365236,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.002368275073991047,
      "loss": 0.2303,
      "num_input_tokens_seen": 32107616,
      "step": 152135
    },
    {
      "epoch": 16.737073707370737,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.002367498521164042,
      "loss": 0.2308,
      "num_input_tokens_seen": 32108640,
      "step": 152140
    },
    {
      "epoch": 16.737623762376238,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.002366722084764739,
      "loss": 0.2309,
      "num_input_tokens_seen": 32109664,
      "step": 152145
    },
    {
      "epoch": 16.73817381738174,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0023659457648002994,
      "loss": 0.2319,
      "num_input_tokens_seen": 32110720,
      "step": 152150
    },
    {
      "epoch": 16.73872387238724,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0023651695612778743,
      "loss": 0.2314,
      "num_input_tokens_seen": 32111808,
      "step": 152155
    },
    {
      "epoch": 16.739273927392738,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0023643934742046163,
      "loss": 0.2308,
      "num_input_tokens_seen": 32112768,
      "step": 152160
    },
    {
      "epoch": 16.73982398239824,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.002363617503587681,
      "loss": 0.2324,
      "num_input_tokens_seen": 32113792,
      "step": 152165
    },
    {
      "epoch": 16.74037403740374,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.002362841649434218,
      "loss": 0.2319,
      "num_input_tokens_seen": 32114880,
      "step": 152170
    },
    {
      "epoch": 16.74092409240924,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.002362065911751385,
      "loss": 0.2319,
      "num_input_tokens_seen": 32115968,
      "step": 152175
    },
    {
      "epoch": 16.741474147414742,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.002361290290546324,
      "loss": 0.2319,
      "num_input_tokens_seen": 32117056,
      "step": 152180
    },
    {
      "epoch": 16.742024202420243,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.002360514785826183,
      "loss": 0.2298,
      "num_input_tokens_seen": 32118112,
      "step": 152185
    },
    {
      "epoch": 16.742574257425744,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0023597393975981133,
      "loss": 0.2308,
      "num_input_tokens_seen": 32119168,
      "step": 152190
    },
    {
      "epoch": 16.74312431243124,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.002358964125869259,
      "loss": 0.2324,
      "num_input_tokens_seen": 32120256,
      "step": 152195
    },
    {
      "epoch": 16.743674367436743,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0023581889706467706,
      "loss": 0.2319,
      "num_input_tokens_seen": 32121312,
      "step": 152200
    },
    {
      "epoch": 16.744224422442244,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002357413931937788,
      "loss": 0.2298,
      "num_input_tokens_seen": 32122432,
      "step": 152205
    },
    {
      "epoch": 16.744774477447745,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.0023566390097494515,
      "loss": 0.2329,
      "num_input_tokens_seen": 32123520,
      "step": 152210
    },
    {
      "epoch": 16.745324532453246,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.002355864204088911,
      "loss": 0.2314,
      "num_input_tokens_seen": 32124512,
      "step": 152215
    },
    {
      "epoch": 16.745874587458747,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.002355089514963297,
      "loss": 0.2308,
      "num_input_tokens_seen": 32125600,
      "step": 152220
    },
    {
      "epoch": 16.746424642464248,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.0023543149423797577,
      "loss": 0.2293,
      "num_input_tokens_seen": 32126656,
      "step": 152225
    },
    {
      "epoch": 16.746974697469746,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0023535404863454317,
      "loss": 0.2308,
      "num_input_tokens_seen": 32127744,
      "step": 152230
    },
    {
      "epoch": 16.747524752475247,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.002352766146867452,
      "loss": 0.2309,
      "num_input_tokens_seen": 32128864,
      "step": 152235
    },
    {
      "epoch": 16.748074807480748,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0023519919239529596,
      "loss": 0.2314,
      "num_input_tokens_seen": 32129856,
      "step": 152240
    },
    {
      "epoch": 16.74862486248625,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.002351217817609087,
      "loss": 0.2303,
      "num_input_tokens_seen": 32130848,
      "step": 152245
    },
    {
      "epoch": 16.74917491749175,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.00235044382784297,
      "loss": 0.2329,
      "num_input_tokens_seen": 32131904,
      "step": 152250
    },
    {
      "epoch": 16.74972497249725,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002349669954661745,
      "loss": 0.2324,
      "num_input_tokens_seen": 32132992,
      "step": 152255
    },
    {
      "epoch": 16.75027502750275,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0023488961980725393,
      "loss": 0.233,
      "num_input_tokens_seen": 32133952,
      "step": 152260
    },
    {
      "epoch": 16.75082508250825,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.0023481225580824906,
      "loss": 0.2319,
      "num_input_tokens_seen": 32135072,
      "step": 152265
    },
    {
      "epoch": 16.75137513751375,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0023473490346987195,
      "loss": 0.2319,
      "num_input_tokens_seen": 32136096,
      "step": 152270
    },
    {
      "epoch": 16.751925192519252,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.002346575627928366,
      "loss": 0.2303,
      "num_input_tokens_seen": 32137184,
      "step": 152275
    },
    {
      "epoch": 16.752475247524753,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.002345802337778551,
      "loss": 0.2298,
      "num_input_tokens_seen": 32138240,
      "step": 152280
    },
    {
      "epoch": 16.753025302530254,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.0023450291642564012,
      "loss": 0.2303,
      "num_input_tokens_seen": 32139328,
      "step": 152285
    },
    {
      "epoch": 16.753575357535752,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00234425610736905,
      "loss": 0.2308,
      "num_input_tokens_seen": 32140320,
      "step": 152290
    },
    {
      "epoch": 16.754125412541253,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.0023434831671236157,
      "loss": 0.2278,
      "num_input_tokens_seen": 32141344,
      "step": 152295
    },
    {
      "epoch": 16.754675467546754,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.0023427103435272245,
      "loss": 0.2319,
      "num_input_tokens_seen": 32142368,
      "step": 152300
    },
    {
      "epoch": 16.755225522552255,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.002341937636587,
      "loss": 0.2324,
      "num_input_tokens_seen": 32143392,
      "step": 152305
    },
    {
      "epoch": 16.755775577557756,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.002341165046310054,
      "loss": 0.2283,
      "num_input_tokens_seen": 32144416,
      "step": 152310
    },
    {
      "epoch": 16.756325632563257,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0023403925727035232,
      "loss": 0.2298,
      "num_input_tokens_seen": 32145408,
      "step": 152315
    },
    {
      "epoch": 16.75687568756876,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.0023396202157745204,
      "loss": 0.2314,
      "num_input_tokens_seen": 32146464,
      "step": 152320
    },
    {
      "epoch": 16.757425742574256,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0023388479755301595,
      "loss": 0.2309,
      "num_input_tokens_seen": 32147520,
      "step": 152325
    },
    {
      "epoch": 16.757975797579757,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0023380758519775647,
      "loss": 0.2319,
      "num_input_tokens_seen": 32148576,
      "step": 152330
    },
    {
      "epoch": 16.758525852585258,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0023373038451238435,
      "loss": 0.2319,
      "num_input_tokens_seen": 32149664,
      "step": 152335
    },
    {
      "epoch": 16.75907590759076,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.002336531954976119,
      "loss": 0.2303,
      "num_input_tokens_seen": 32150720,
      "step": 152340
    },
    {
      "epoch": 16.75962596259626,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.002335760181541506,
      "loss": 0.2324,
      "num_input_tokens_seen": 32151776,
      "step": 152345
    },
    {
      "epoch": 16.76017601760176,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.002334988524827109,
      "loss": 0.2314,
      "num_input_tokens_seen": 32152800,
      "step": 152350
    },
    {
      "epoch": 16.760726072607262,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.0023342169848400516,
      "loss": 0.234,
      "num_input_tokens_seen": 32153856,
      "step": 152355
    },
    {
      "epoch": 16.76127612761276,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.002333445561587433,
      "loss": 0.2324,
      "num_input_tokens_seen": 32154880,
      "step": 152360
    },
    {
      "epoch": 16.76182618261826,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0023326742550763713,
      "loss": 0.2319,
      "num_input_tokens_seen": 32155904,
      "step": 152365
    },
    {
      "epoch": 16.762376237623762,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.002331903065313973,
      "loss": 0.2308,
      "num_input_tokens_seen": 32156992,
      "step": 152370
    },
    {
      "epoch": 16.762926292629263,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0023311319923073437,
      "loss": 0.2319,
      "num_input_tokens_seen": 32158112,
      "step": 152375
    },
    {
      "epoch": 16.763476347634764,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0023303610360635952,
      "loss": 0.2319,
      "num_input_tokens_seen": 32159168,
      "step": 152380
    },
    {
      "epoch": 16.764026402640265,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.002329590196589825,
      "loss": 0.2303,
      "num_input_tokens_seen": 32160224,
      "step": 152385
    },
    {
      "epoch": 16.764576457645763,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0023288194738931467,
      "loss": 0.2293,
      "num_input_tokens_seen": 32161248,
      "step": 152390
    },
    {
      "epoch": 16.765126512651264,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0023280488679806553,
      "loss": 0.2314,
      "num_input_tokens_seen": 32162368,
      "step": 152395
    },
    {
      "epoch": 16.765676567656765,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0023272783788594576,
      "loss": 0.233,
      "num_input_tokens_seen": 32163392,
      "step": 152400
    },
    {
      "epoch": 16.766226622662266,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002326508006536657,
      "loss": 0.2314,
      "num_input_tokens_seen": 32164448,
      "step": 152405
    },
    {
      "epoch": 16.766776677667767,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.002325737751019347,
      "loss": 0.2319,
      "num_input_tokens_seen": 32165440,
      "step": 152410
    },
    {
      "epoch": 16.76732673267327,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0023249676123146333,
      "loss": 0.2303,
      "num_input_tokens_seen": 32166496,
      "step": 152415
    },
    {
      "epoch": 16.76787678767877,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.002324197590429612,
      "loss": 0.2329,
      "num_input_tokens_seen": 32167488,
      "step": 152420
    },
    {
      "epoch": 16.768426842684267,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.002323427685371372,
      "loss": 0.2309,
      "num_input_tokens_seen": 32168544,
      "step": 152425
    },
    {
      "epoch": 16.768976897689768,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.002322657897147023,
      "loss": 0.2314,
      "num_input_tokens_seen": 32169600,
      "step": 152430
    },
    {
      "epoch": 16.76952695269527,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0023218882257636506,
      "loss": 0.2319,
      "num_input_tokens_seen": 32170656,
      "step": 152435
    },
    {
      "epoch": 16.77007700770077,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0023211186712283533,
      "loss": 0.2303,
      "num_input_tokens_seen": 32171680,
      "step": 152440
    },
    {
      "epoch": 16.77062706270627,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0023203492335482213,
      "loss": 0.2319,
      "num_input_tokens_seen": 32172768,
      "step": 152445
    },
    {
      "epoch": 16.771177117711773,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0023195799127303423,
      "loss": 0.2309,
      "num_input_tokens_seen": 32173792,
      "step": 152450
    },
    {
      "epoch": 16.77172717271727,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002318810708781811,
      "loss": 0.2303,
      "num_input_tokens_seen": 32174816,
      "step": 152455
    },
    {
      "epoch": 16.77227722772277,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.0023180416217097165,
      "loss": 0.2303,
      "num_input_tokens_seen": 32175872,
      "step": 152460
    },
    {
      "epoch": 16.772827282728272,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00231727265152115,
      "loss": 0.2314,
      "num_input_tokens_seen": 32176960,
      "step": 152465
    },
    {
      "epoch": 16.773377337733773,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0023165037982231973,
      "loss": 0.2314,
      "num_input_tokens_seen": 32177984,
      "step": 152470
    },
    {
      "epoch": 16.773927392739274,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0023157350618229376,
      "loss": 0.2324,
      "num_input_tokens_seen": 32179136,
      "step": 152475
    },
    {
      "epoch": 16.774477447744776,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 0.002314966442327466,
      "loss": 0.2313,
      "num_input_tokens_seen": 32180160,
      "step": 152480
    },
    {
      "epoch": 16.775027502750277,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.002314197939743854,
      "loss": 0.2308,
      "num_input_tokens_seen": 32181280,
      "step": 152485
    },
    {
      "epoch": 16.775577557755774,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0023134295540791997,
      "loss": 0.2303,
      "num_input_tokens_seen": 32182368,
      "step": 152490
    },
    {
      "epoch": 16.776127612761275,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.0023126612853405764,
      "loss": 0.2324,
      "num_input_tokens_seen": 32183392,
      "step": 152495
    },
    {
      "epoch": 16.776677667766776,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0023118931335350643,
      "loss": 0.2314,
      "num_input_tokens_seen": 32184384,
      "step": 152500
    },
    {
      "epoch": 16.777227722772277,
      "grad_norm": 0.0030364990234375,
      "learning_rate": 0.0023111250986697484,
      "loss": 0.2303,
      "num_input_tokens_seen": 32185408,
      "step": 152505
    },
    {
      "epoch": 16.77777777777778,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.002310357180751699,
      "loss": 0.2324,
      "num_input_tokens_seen": 32186464,
      "step": 152510
    },
    {
      "epoch": 16.77832783278328,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.002309589379787998,
      "loss": 0.2303,
      "num_input_tokens_seen": 32187488,
      "step": 152515
    },
    {
      "epoch": 16.778877887788777,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.002308821695785727,
      "loss": 0.2308,
      "num_input_tokens_seen": 32188576,
      "step": 152520
    },
    {
      "epoch": 16.77942794279428,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.002308054128751953,
      "loss": 0.2314,
      "num_input_tokens_seen": 32189568,
      "step": 152525
    },
    {
      "epoch": 16.77997799779978,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002307286678693758,
      "loss": 0.2298,
      "num_input_tokens_seen": 32190624,
      "step": 152530
    },
    {
      "epoch": 16.78052805280528,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.0023065193456182098,
      "loss": 0.2313,
      "num_input_tokens_seen": 32191680,
      "step": 152535
    },
    {
      "epoch": 16.78107810781078,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0023057521295323795,
      "loss": 0.2308,
      "num_input_tokens_seen": 32192672,
      "step": 152540
    },
    {
      "epoch": 16.781628162816283,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.00230498503044334,
      "loss": 0.2314,
      "num_input_tokens_seen": 32193760,
      "step": 152545
    },
    {
      "epoch": 16.782178217821784,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.00230421804835816,
      "loss": 0.2314,
      "num_input_tokens_seen": 32194848,
      "step": 152550
    },
    {
      "epoch": 16.78272827282728,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.0023034511832839165,
      "loss": 0.2293,
      "num_input_tokens_seen": 32195936,
      "step": 152555
    },
    {
      "epoch": 16.783278327832782,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0023026844352276696,
      "loss": 0.2309,
      "num_input_tokens_seen": 32196992,
      "step": 152560
    },
    {
      "epoch": 16.783828382838283,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0023019178041964843,
      "loss": 0.2319,
      "num_input_tokens_seen": 32198048,
      "step": 152565
    },
    {
      "epoch": 16.784378437843785,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0023011512901974294,
      "loss": 0.2314,
      "num_input_tokens_seen": 32199136,
      "step": 152570
    },
    {
      "epoch": 16.784928492849286,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.00230038489323757,
      "loss": 0.2329,
      "num_input_tokens_seen": 32200128,
      "step": 152575
    },
    {
      "epoch": 16.785478547854787,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.0022996186133239726,
      "loss": 0.2293,
      "num_input_tokens_seen": 32201280,
      "step": 152580
    },
    {
      "epoch": 16.786028602860284,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.002298852450463696,
      "loss": 0.2314,
      "num_input_tokens_seen": 32202304,
      "step": 152585
    },
    {
      "epoch": 16.786578657865785,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.002298086404663799,
      "loss": 0.2319,
      "num_input_tokens_seen": 32203392,
      "step": 152590
    },
    {
      "epoch": 16.787128712871286,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0022973204759313475,
      "loss": 0.2283,
      "num_input_tokens_seen": 32204448,
      "step": 152595
    },
    {
      "epoch": 16.787678767876788,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.002296554664273393,
      "loss": 0.2324,
      "num_input_tokens_seen": 32205504,
      "step": 152600
    },
    {
      "epoch": 16.78822882288229,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0022957889696969993,
      "loss": 0.2314,
      "num_input_tokens_seen": 32206624,
      "step": 152605
    },
    {
      "epoch": 16.78877887788779,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0022950233922092267,
      "loss": 0.2298,
      "num_input_tokens_seen": 32207648,
      "step": 152610
    },
    {
      "epoch": 16.78932893289329,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0022942579318171235,
      "loss": 0.2298,
      "num_input_tokens_seen": 32208704,
      "step": 152615
    },
    {
      "epoch": 16.78987898789879,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.002293492588527751,
      "loss": 0.2309,
      "num_input_tokens_seen": 32209728,
      "step": 152620
    },
    {
      "epoch": 16.79042904290429,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.0022927273623481576,
      "loss": 0.2314,
      "num_input_tokens_seen": 32210784,
      "step": 152625
    },
    {
      "epoch": 16.79097909790979,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.002291962253285398,
      "loss": 0.2324,
      "num_input_tokens_seen": 32211744,
      "step": 152630
    },
    {
      "epoch": 16.79152915291529,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002291197261346528,
      "loss": 0.2308,
      "num_input_tokens_seen": 32212800,
      "step": 152635
    },
    {
      "epoch": 16.792079207920793,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.002290432386538591,
      "loss": 0.2329,
      "num_input_tokens_seen": 32213856,
      "step": 152640
    },
    {
      "epoch": 16.792629262926294,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0022896676288686434,
      "loss": 0.2314,
      "num_input_tokens_seen": 32214944,
      "step": 152645
    },
    {
      "epoch": 16.793179317931795,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002288902988343728,
      "loss": 0.2309,
      "num_input_tokens_seen": 32215968,
      "step": 152650
    },
    {
      "epoch": 16.793729372937293,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.0022881384649708967,
      "loss": 0.2293,
      "num_input_tokens_seen": 32216992,
      "step": 152655
    },
    {
      "epoch": 16.794279427942794,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00228737405875719,
      "loss": 0.2314,
      "num_input_tokens_seen": 32218080,
      "step": 152660
    },
    {
      "epoch": 16.794829482948295,
      "grad_norm": 0.00064849853515625,
      "learning_rate": 0.0022866097697096565,
      "loss": 0.2303,
      "num_input_tokens_seen": 32219104,
      "step": 152665
    },
    {
      "epoch": 16.795379537953796,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.002285845597835344,
      "loss": 0.2308,
      "num_input_tokens_seen": 32220096,
      "step": 152670
    },
    {
      "epoch": 16.795929592959297,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.002285081543141287,
      "loss": 0.2319,
      "num_input_tokens_seen": 32221120,
      "step": 152675
    },
    {
      "epoch": 16.796479647964798,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.002284317605634537,
      "loss": 0.2319,
      "num_input_tokens_seen": 32222144,
      "step": 152680
    },
    {
      "epoch": 16.797029702970296,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.002283553785322126,
      "loss": 0.2308,
      "num_input_tokens_seen": 32223168,
      "step": 152685
    },
    {
      "epoch": 16.797579757975797,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0022827900822110984,
      "loss": 0.2325,
      "num_input_tokens_seen": 32224288,
      "step": 152690
    },
    {
      "epoch": 16.798129812981298,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0022820264963084955,
      "loss": 0.2314,
      "num_input_tokens_seen": 32225312,
      "step": 152695
    },
    {
      "epoch": 16.7986798679868,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0022812630276213518,
      "loss": 0.2324,
      "num_input_tokens_seen": 32226464,
      "step": 152700
    },
    {
      "epoch": 16.7992299229923,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.0022804996761567014,
      "loss": 0.2298,
      "num_input_tokens_seen": 32227552,
      "step": 152705
    },
    {
      "epoch": 16.7997799779978,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.0022797364419215836,
      "loss": 0.2309,
      "num_input_tokens_seen": 32228608,
      "step": 152710
    },
    {
      "epoch": 16.8003300330033,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0022789733249230283,
      "loss": 0.2309,
      "num_input_tokens_seen": 32229632,
      "step": 152715
    },
    {
      "epoch": 16.8008800880088,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0022782103251680723,
      "loss": 0.2303,
      "num_input_tokens_seen": 32230688,
      "step": 152720
    },
    {
      "epoch": 16.8014301430143,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.002277447442663751,
      "loss": 0.2303,
      "num_input_tokens_seen": 32231712,
      "step": 152725
    },
    {
      "epoch": 16.801980198019802,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.002276684677417086,
      "loss": 0.2324,
      "num_input_tokens_seen": 32232800,
      "step": 152730
    },
    {
      "epoch": 16.802530253025303,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0022759220294351194,
      "loss": 0.2314,
      "num_input_tokens_seen": 32233888,
      "step": 152735
    },
    {
      "epoch": 16.803080308030804,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0022751594987248684,
      "loss": 0.2303,
      "num_input_tokens_seen": 32234912,
      "step": 152740
    },
    {
      "epoch": 16.803630363036305,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0022743970852933675,
      "loss": 0.2298,
      "num_input_tokens_seen": 32235936,
      "step": 152745
    },
    {
      "epoch": 16.804180418041803,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002273634789147646,
      "loss": 0.2288,
      "num_input_tokens_seen": 32236992,
      "step": 152750
    },
    {
      "epoch": 16.804730473047304,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0022728726102947214,
      "loss": 0.2303,
      "num_input_tokens_seen": 32237984,
      "step": 152755
    },
    {
      "epoch": 16.805280528052805,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.002272110548741627,
      "loss": 0.2309,
      "num_input_tokens_seen": 32239040,
      "step": 152760
    },
    {
      "epoch": 16.805830583058306,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0022713486044953785,
      "loss": 0.2314,
      "num_input_tokens_seen": 32240096,
      "step": 152765
    },
    {
      "epoch": 16.806380638063807,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.002270586777563005,
      "loss": 0.2319,
      "num_input_tokens_seen": 32241184,
      "step": 152770
    },
    {
      "epoch": 16.806930693069308,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.002269825067951523,
      "loss": 0.2314,
      "num_input_tokens_seen": 32242176,
      "step": 152775
    },
    {
      "epoch": 16.80748074807481,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.0022690634756679534,
      "loss": 0.2324,
      "num_input_tokens_seen": 32243200,
      "step": 152780
    },
    {
      "epoch": 16.808030803080307,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0022683020007193203,
      "loss": 0.2319,
      "num_input_tokens_seen": 32244288,
      "step": 152785
    },
    {
      "epoch": 16.808580858085808,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0022675406431126354,
      "loss": 0.2319,
      "num_input_tokens_seen": 32245312,
      "step": 152790
    },
    {
      "epoch": 16.80913091309131,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0022667794028549226,
      "loss": 0.2324,
      "num_input_tokens_seen": 32246368,
      "step": 152795
    },
    {
      "epoch": 16.80968096809681,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002266018279953191,
      "loss": 0.2314,
      "num_input_tokens_seen": 32247392,
      "step": 152800
    },
    {
      "epoch": 16.81023102310231,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.002265257274414458,
      "loss": 0.2319,
      "num_input_tokens_seen": 32248480,
      "step": 152805
    },
    {
      "epoch": 16.810781078107812,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.002264496386245741,
      "loss": 0.2319,
      "num_input_tokens_seen": 32249536,
      "step": 152810
    },
    {
      "epoch": 16.81133113311331,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002263735615454046,
      "loss": 0.2324,
      "num_input_tokens_seen": 32250624,
      "step": 152815
    },
    {
      "epoch": 16.81188118811881,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0022629749620463936,
      "loss": 0.2309,
      "num_input_tokens_seen": 32251648,
      "step": 152820
    },
    {
      "epoch": 16.812431243124312,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.0022622144260297876,
      "loss": 0.2329,
      "num_input_tokens_seen": 32252640,
      "step": 152825
    },
    {
      "epoch": 16.812981298129813,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0022614540074112365,
      "loss": 0.2314,
      "num_input_tokens_seen": 32253696,
      "step": 152830
    },
    {
      "epoch": 16.813531353135314,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0022606937061977494,
      "loss": 0.2335,
      "num_input_tokens_seen": 32254752,
      "step": 152835
    },
    {
      "epoch": 16.814081408140815,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.002259933522396337,
      "loss": 0.2303,
      "num_input_tokens_seen": 32255776,
      "step": 152840
    },
    {
      "epoch": 16.814631463146316,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.002259173456014008,
      "loss": 0.2314,
      "num_input_tokens_seen": 32256800,
      "step": 152845
    },
    {
      "epoch": 16.815181518151814,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0022584135070577624,
      "loss": 0.2313,
      "num_input_tokens_seen": 32257824,
      "step": 152850
    },
    {
      "epoch": 16.815731573157315,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0022576536755346036,
      "loss": 0.2298,
      "num_input_tokens_seen": 32258880,
      "step": 152855
    },
    {
      "epoch": 16.816281628162816,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002256893961451534,
      "loss": 0.2319,
      "num_input_tokens_seen": 32259968,
      "step": 152860
    },
    {
      "epoch": 16.816831683168317,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00225613436481556,
      "loss": 0.2303,
      "num_input_tokens_seen": 32260960,
      "step": 152865
    },
    {
      "epoch": 16.817381738173818,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.002255374885633684,
      "loss": 0.2319,
      "num_input_tokens_seen": 32261984,
      "step": 152870
    },
    {
      "epoch": 16.81793179317932,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0022546155239129024,
      "loss": 0.2309,
      "num_input_tokens_seen": 32263040,
      "step": 152875
    },
    {
      "epoch": 16.818481848184817,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.002253856279660209,
      "loss": 0.2303,
      "num_input_tokens_seen": 32264096,
      "step": 152880
    },
    {
      "epoch": 16.819031903190318,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0022530971528826106,
      "loss": 0.2298,
      "num_input_tokens_seen": 32265184,
      "step": 152885
    },
    {
      "epoch": 16.81958195819582,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.002252338143587094,
      "loss": 0.2303,
      "num_input_tokens_seen": 32266176,
      "step": 152890
    },
    {
      "epoch": 16.82013201320132,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0022515792517806615,
      "loss": 0.2314,
      "num_input_tokens_seen": 32267200,
      "step": 152895
    },
    {
      "epoch": 16.82068206820682,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0022508204774703082,
      "loss": 0.2303,
      "num_input_tokens_seen": 32268288,
      "step": 152900
    },
    {
      "epoch": 16.821232123212322,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.002250061820663022,
      "loss": 0.2314,
      "num_input_tokens_seen": 32269344,
      "step": 152905
    },
    {
      "epoch": 16.821782178217823,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0022493032813658016,
      "loss": 0.2303,
      "num_input_tokens_seen": 32270368,
      "step": 152910
    },
    {
      "epoch": 16.82233223322332,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00224854485958563,
      "loss": 0.2319,
      "num_input_tokens_seen": 32271424,
      "step": 152915
    },
    {
      "epoch": 16.822882288228822,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.002247786555329504,
      "loss": 0.2309,
      "num_input_tokens_seen": 32272512,
      "step": 152920
    },
    {
      "epoch": 16.823432343234323,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.002247028368604411,
      "loss": 0.2319,
      "num_input_tokens_seen": 32273568,
      "step": 152925
    },
    {
      "epoch": 16.823982398239824,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.002246270299417336,
      "loss": 0.2324,
      "num_input_tokens_seen": 32274560,
      "step": 152930
    },
    {
      "epoch": 16.824532453245325,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.0022455123477752715,
      "loss": 0.2309,
      "num_input_tokens_seen": 32275648,
      "step": 152935
    },
    {
      "epoch": 16.825082508250826,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.002244754513685198,
      "loss": 0.2319,
      "num_input_tokens_seen": 32276704,
      "step": 152940
    },
    {
      "epoch": 16.825632563256324,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002243996797154099,
      "loss": 0.2314,
      "num_input_tokens_seen": 32277728,
      "step": 152945
    },
    {
      "epoch": 16.826182618261825,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.002243239198188962,
      "loss": 0.2319,
      "num_input_tokens_seen": 32278752,
      "step": 152950
    },
    {
      "epoch": 16.826732673267326,
      "grad_norm": 0.002899169921875,
      "learning_rate": 0.002242481716796766,
      "loss": 0.2308,
      "num_input_tokens_seen": 32279808,
      "step": 152955
    },
    {
      "epoch": 16.827282728272827,
      "grad_norm": 0.001953125,
      "learning_rate": 0.002241724352984498,
      "loss": 0.2335,
      "num_input_tokens_seen": 32280864,
      "step": 152960
    },
    {
      "epoch": 16.82783278327833,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0022409671067591353,
      "loss": 0.2314,
      "num_input_tokens_seen": 32281920,
      "step": 152965
    },
    {
      "epoch": 16.82838283828383,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0022402099781276535,
      "loss": 0.2324,
      "num_input_tokens_seen": 32283008,
      "step": 152970
    },
    {
      "epoch": 16.82893289328933,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0022394529670970325,
      "loss": 0.2314,
      "num_input_tokens_seen": 32284128,
      "step": 152975
    },
    {
      "epoch": 16.829482948294828,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.0022386960736742517,
      "loss": 0.2324,
      "num_input_tokens_seen": 32285152,
      "step": 152980
    },
    {
      "epoch": 16.83003300330033,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.002237939297866287,
      "loss": 0.2319,
      "num_input_tokens_seen": 32286176,
      "step": 152985
    },
    {
      "epoch": 16.83058305830583,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0022371826396801148,
      "loss": 0.234,
      "num_input_tokens_seen": 32287264,
      "step": 152990
    },
    {
      "epoch": 16.83113311331133,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.0022364260991227006,
      "loss": 0.2319,
      "num_input_tokens_seen": 32288320,
      "step": 152995
    },
    {
      "epoch": 16.831683168316832,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.002235669676201026,
      "loss": 0.2324,
      "num_input_tokens_seen": 32289408,
      "step": 153000
    },
    {
      "epoch": 16.832233223322334,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.002234913370922056,
      "loss": 0.2329,
      "num_input_tokens_seen": 32290464,
      "step": 153005
    },
    {
      "epoch": 16.83278327832783,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.0022341571832927634,
      "loss": 0.2314,
      "num_input_tokens_seen": 32291488,
      "step": 153010
    },
    {
      "epoch": 16.833333333333332,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.0022334011133201227,
      "loss": 0.2303,
      "num_input_tokens_seen": 32292544,
      "step": 153015
    },
    {
      "epoch": 16.833883388338833,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002232645161011095,
      "loss": 0.2293,
      "num_input_tokens_seen": 32293600,
      "step": 153020
    },
    {
      "epoch": 16.834433443344334,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.002231889326372653,
      "loss": 0.2303,
      "num_input_tokens_seen": 32294656,
      "step": 153025
    },
    {
      "epoch": 16.834983498349835,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0022311336094117565,
      "loss": 0.2303,
      "num_input_tokens_seen": 32295616,
      "step": 153030
    },
    {
      "epoch": 16.835533553355337,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.002230378010135376,
      "loss": 0.2314,
      "num_input_tokens_seen": 32296672,
      "step": 153035
    },
    {
      "epoch": 16.836083608360838,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.0022296225285504765,
      "loss": 0.2298,
      "num_input_tokens_seen": 32297760,
      "step": 153040
    },
    {
      "epoch": 16.836633663366335,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.002228867164664015,
      "loss": 0.2293,
      "num_input_tokens_seen": 32298784,
      "step": 153045
    },
    {
      "epoch": 16.837183718371836,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0022281119184829608,
      "loss": 0.2308,
      "num_input_tokens_seen": 32299904,
      "step": 153050
    },
    {
      "epoch": 16.837733773377337,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.002227356790014266,
      "loss": 0.2319,
      "num_input_tokens_seen": 32301024,
      "step": 153055
    },
    {
      "epoch": 16.83828382838284,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0022266017792648973,
      "loss": 0.2324,
      "num_input_tokens_seen": 32302048,
      "step": 153060
    },
    {
      "epoch": 16.83883388338834,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0022258468862418087,
      "loss": 0.2324,
      "num_input_tokens_seen": 32303008,
      "step": 153065
    },
    {
      "epoch": 16.83938393839384,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0022250921109519587,
      "loss": 0.2324,
      "num_input_tokens_seen": 32303968,
      "step": 153070
    },
    {
      "epoch": 16.83993399339934,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0022243374534023095,
      "loss": 0.2314,
      "num_input_tokens_seen": 32305024,
      "step": 153075
    },
    {
      "epoch": 16.84048404840484,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.002223582913599811,
      "loss": 0.2319,
      "num_input_tokens_seen": 32306080,
      "step": 153080
    },
    {
      "epoch": 16.84103410341034,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002222828491551415,
      "loss": 0.2304,
      "num_input_tokens_seen": 32307136,
      "step": 153085
    },
    {
      "epoch": 16.84158415841584,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.00222207418726408,
      "loss": 0.2314,
      "num_input_tokens_seen": 32308224,
      "step": 153090
    },
    {
      "epoch": 16.842134213421343,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0022213200007447508,
      "loss": 0.2324,
      "num_input_tokens_seen": 32309280,
      "step": 153095
    },
    {
      "epoch": 16.842684268426844,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0022205659320003878,
      "loss": 0.2293,
      "num_input_tokens_seen": 32310304,
      "step": 153100
    },
    {
      "epoch": 16.843234323432345,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002219811981037939,
      "loss": 0.2314,
      "num_input_tokens_seen": 32311360,
      "step": 153105
    },
    {
      "epoch": 16.843784378437842,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0022190581478643456,
      "loss": 0.2324,
      "num_input_tokens_seen": 32312448,
      "step": 153110
    },
    {
      "epoch": 16.844334433443343,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0022183044324865634,
      "loss": 0.2303,
      "num_input_tokens_seen": 32313568,
      "step": 153115
    },
    {
      "epoch": 16.844884488448844,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.002217550834911534,
      "loss": 0.2314,
      "num_input_tokens_seen": 32314688,
      "step": 153120
    },
    {
      "epoch": 16.845434543454346,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.002216797355146203,
      "loss": 0.2319,
      "num_input_tokens_seen": 32315712,
      "step": 153125
    },
    {
      "epoch": 16.845984598459847,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.002216043993197522,
      "loss": 0.2314,
      "num_input_tokens_seen": 32316736,
      "step": 153130
    },
    {
      "epoch": 16.846534653465348,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0022152907490724256,
      "loss": 0.2314,
      "num_input_tokens_seen": 32317728,
      "step": 153135
    },
    {
      "epoch": 16.847084708470845,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.002214537622777862,
      "loss": 0.2314,
      "num_input_tokens_seen": 32318784,
      "step": 153140
    },
    {
      "epoch": 16.847634763476346,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.002213784614320769,
      "loss": 0.2293,
      "num_input_tokens_seen": 32319808,
      "step": 153145
    },
    {
      "epoch": 16.848184818481847,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0022130317237080846,
      "loss": 0.2298,
      "num_input_tokens_seen": 32320864,
      "step": 153150
    },
    {
      "epoch": 16.84873487348735,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.002212278950946756,
      "loss": 0.2303,
      "num_input_tokens_seen": 32321888,
      "step": 153155
    },
    {
      "epoch": 16.84928492849285,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.0022115262960437135,
      "loss": 0.2319,
      "num_input_tokens_seen": 32322912,
      "step": 153160
    },
    {
      "epoch": 16.84983498349835,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.002210773759005899,
      "loss": 0.2309,
      "num_input_tokens_seen": 32323936,
      "step": 153165
    },
    {
      "epoch": 16.850385038503852,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.0022100213398402427,
      "loss": 0.2309,
      "num_input_tokens_seen": 32325024,
      "step": 153170
    },
    {
      "epoch": 16.85093509350935,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.0022092690385536865,
      "loss": 0.2314,
      "num_input_tokens_seen": 32326048,
      "step": 153175
    },
    {
      "epoch": 16.85148514851485,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0022085168551531564,
      "loss": 0.2319,
      "num_input_tokens_seen": 32327136,
      "step": 153180
    },
    {
      "epoch": 16.85203520352035,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.0022077647896455864,
      "loss": 0.2293,
      "num_input_tokens_seen": 32328128,
      "step": 153185
    },
    {
      "epoch": 16.852585258525853,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0022070128420379163,
      "loss": 0.2314,
      "num_input_tokens_seen": 32329184,
      "step": 153190
    },
    {
      "epoch": 16.853135313531354,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0022062610123370656,
      "loss": 0.2314,
      "num_input_tokens_seen": 32330208,
      "step": 153195
    },
    {
      "epoch": 16.853685368536855,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.002205509300549972,
      "loss": 0.2298,
      "num_input_tokens_seen": 32331168,
      "step": 153200
    },
    {
      "epoch": 16.854235423542356,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0022047577066835584,
      "loss": 0.2324,
      "num_input_tokens_seen": 32332224,
      "step": 153205
    },
    {
      "epoch": 16.854785478547853,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.002204006230744747,
      "loss": 0.2324,
      "num_input_tokens_seen": 32333344,
      "step": 153210
    },
    {
      "epoch": 16.855335533553355,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0022032548727404784,
      "loss": 0.2298,
      "num_input_tokens_seen": 32334528,
      "step": 153215
    },
    {
      "epoch": 16.855885588558856,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.0022025036326776644,
      "loss": 0.2314,
      "num_input_tokens_seen": 32335552,
      "step": 153220
    },
    {
      "epoch": 16.856435643564357,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.002201752510563236,
      "loss": 0.2309,
      "num_input_tokens_seen": 32336544,
      "step": 153225
    },
    {
      "epoch": 16.856985698569858,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.0022010015064041154,
      "loss": 0.2308,
      "num_input_tokens_seen": 32337568,
      "step": 153230
    },
    {
      "epoch": 16.85753575357536,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002200250620207218,
      "loss": 0.2303,
      "num_input_tokens_seen": 32338688,
      "step": 153235
    },
    {
      "epoch": 16.858085808580856,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.0021994998519794685,
      "loss": 0.2319,
      "num_input_tokens_seen": 32339776,
      "step": 153240
    },
    {
      "epoch": 16.858635863586358,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0021987492017277875,
      "loss": 0.2324,
      "num_input_tokens_seen": 32340864,
      "step": 153245
    },
    {
      "epoch": 16.85918591859186,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002197998669459095,
      "loss": 0.2293,
      "num_input_tokens_seen": 32341888,
      "step": 153250
    },
    {
      "epoch": 16.85973597359736,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.002197248255180307,
      "loss": 0.2324,
      "num_input_tokens_seen": 32342912,
      "step": 153255
    },
    {
      "epoch": 16.86028602860286,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0021964979588983335,
      "loss": 0.2298,
      "num_input_tokens_seen": 32343968,
      "step": 153260
    },
    {
      "epoch": 16.860836083608362,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0021957477806201,
      "loss": 0.2314,
      "num_input_tokens_seen": 32344992,
      "step": 153265
    },
    {
      "epoch": 16.861386138613863,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.0021949977203525087,
      "loss": 0.2324,
      "num_input_tokens_seen": 32346048,
      "step": 153270
    },
    {
      "epoch": 16.86193619361936,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.002194247778102482,
      "loss": 0.2314,
      "num_input_tokens_seen": 32347104,
      "step": 153275
    },
    {
      "epoch": 16.86248624862486,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.0021934979538769305,
      "loss": 0.2308,
      "num_input_tokens_seen": 32348192,
      "step": 153280
    },
    {
      "epoch": 16.863036303630363,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.00219274824768276,
      "loss": 0.2319,
      "num_input_tokens_seen": 32349248,
      "step": 153285
    },
    {
      "epoch": 16.863586358635864,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0021919986595268877,
      "loss": 0.234,
      "num_input_tokens_seen": 32350240,
      "step": 153290
    },
    {
      "epoch": 16.864136413641365,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.0021912491894162133,
      "loss": 0.2324,
      "num_input_tokens_seen": 32351264,
      "step": 153295
    },
    {
      "epoch": 16.864686468646866,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0021904998373576488,
      "loss": 0.2314,
      "num_input_tokens_seen": 32352288,
      "step": 153300
    },
    {
      "epoch": 16.865236523652364,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.0021897506033581055,
      "loss": 0.2319,
      "num_input_tokens_seen": 32353376,
      "step": 153305
    },
    {
      "epoch": 16.865786578657865,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.0021890014874244787,
      "loss": 0.2314,
      "num_input_tokens_seen": 32354400,
      "step": 153310
    },
    {
      "epoch": 16.866336633663366,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.0021882524895636825,
      "loss": 0.2309,
      "num_input_tokens_seen": 32355520,
      "step": 153315
    },
    {
      "epoch": 16.866886688668867,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.002187503609782615,
      "loss": 0.2309,
      "num_input_tokens_seen": 32356544,
      "step": 153320
    },
    {
      "epoch": 16.867436743674368,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.0021867548480881765,
      "loss": 0.2319,
      "num_input_tokens_seen": 32357600,
      "step": 153325
    },
    {
      "epoch": 16.86798679867987,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0021860062044872677,
      "loss": 0.2319,
      "num_input_tokens_seen": 32358624,
      "step": 153330
    },
    {
      "epoch": 16.86853685368537,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.002185257678986794,
      "loss": 0.2283,
      "num_input_tokens_seen": 32359680,
      "step": 153335
    },
    {
      "epoch": 16.869086908690868,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.002184509271593652,
      "loss": 0.2314,
      "num_input_tokens_seen": 32360736,
      "step": 153340
    },
    {
      "epoch": 16.86963696369637,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0021837609823147403,
      "loss": 0.2314,
      "num_input_tokens_seen": 32361760,
      "step": 153345
    },
    {
      "epoch": 16.87018701870187,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0021830128111569496,
      "loss": 0.2314,
      "num_input_tokens_seen": 32362752,
      "step": 153350
    },
    {
      "epoch": 16.87073707370737,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0021822647581271797,
      "loss": 0.2308,
      "num_input_tokens_seen": 32363744,
      "step": 153355
    },
    {
      "epoch": 16.871287128712872,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0021815168232323236,
      "loss": 0.2309,
      "num_input_tokens_seen": 32364800,
      "step": 153360
    },
    {
      "epoch": 16.871837183718373,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0021807690064792815,
      "loss": 0.2308,
      "num_input_tokens_seen": 32365856,
      "step": 153365
    },
    {
      "epoch": 16.87238723872387,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.002180021307874939,
      "loss": 0.2288,
      "num_input_tokens_seen": 32366912,
      "step": 153370
    },
    {
      "epoch": 16.872937293729372,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.002179273727426185,
      "loss": 0.2298,
      "num_input_tokens_seen": 32367904,
      "step": 153375
    },
    {
      "epoch": 16.873487348734873,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.002178526265139918,
      "loss": 0.2324,
      "num_input_tokens_seen": 32368960,
      "step": 153380
    },
    {
      "epoch": 16.874037403740374,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.002177778921023018,
      "loss": 0.2319,
      "num_input_tokens_seen": 32370016,
      "step": 153385
    },
    {
      "epoch": 16.874587458745875,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.002177031695082377,
      "loss": 0.2288,
      "num_input_tokens_seen": 32371072,
      "step": 153390
    },
    {
      "epoch": 16.875137513751376,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.002176284587324884,
      "loss": 0.2314,
      "num_input_tokens_seen": 32372160,
      "step": 153395
    },
    {
      "epoch": 16.875687568756877,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0021755375977574208,
      "loss": 0.2319,
      "num_input_tokens_seen": 32373248,
      "step": 153400
    },
    {
      "epoch": 16.876237623762375,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0021747907263868767,
      "loss": 0.2314,
      "num_input_tokens_seen": 32374304,
      "step": 153405
    },
    {
      "epoch": 16.876787678767876,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.002174043973220127,
      "loss": 0.2283,
      "num_input_tokens_seen": 32375424,
      "step": 153410
    },
    {
      "epoch": 16.877337733773377,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.0021732973382640624,
      "loss": 0.2335,
      "num_input_tokens_seen": 32376480,
      "step": 153415
    },
    {
      "epoch": 16.877887788778878,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.0021725508215255634,
      "loss": 0.2324,
      "num_input_tokens_seen": 32377536,
      "step": 153420
    },
    {
      "epoch": 16.87843784378438,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0021718044230115046,
      "loss": 0.2324,
      "num_input_tokens_seen": 32378624,
      "step": 153425
    },
    {
      "epoch": 16.87898789878988,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.0021710581427287724,
      "loss": 0.2314,
      "num_input_tokens_seen": 32379680,
      "step": 153430
    },
    {
      "epoch": 16.879537953795378,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.002170311980684239,
      "loss": 0.2319,
      "num_input_tokens_seen": 32380800,
      "step": 153435
    },
    {
      "epoch": 16.88008800880088,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0021695659368847864,
      "loss": 0.2319,
      "num_input_tokens_seen": 32381856,
      "step": 153440
    },
    {
      "epoch": 16.88063806380638,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.0021688200113372832,
      "loss": 0.233,
      "num_input_tokens_seen": 32382912,
      "step": 153445
    },
    {
      "epoch": 16.88118811881188,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0021680742040486116,
      "loss": 0.2298,
      "num_input_tokens_seen": 32384032,
      "step": 153450
    },
    {
      "epoch": 16.881738173817382,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0021673285150256437,
      "loss": 0.2335,
      "num_input_tokens_seen": 32385088,
      "step": 153455
    },
    {
      "epoch": 16.882288228822883,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.0021665829442752524,
      "loss": 0.234,
      "num_input_tokens_seen": 32386144,
      "step": 153460
    },
    {
      "epoch": 16.882838283828384,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002165837491804304,
      "loss": 0.2308,
      "num_input_tokens_seen": 32387232,
      "step": 153465
    },
    {
      "epoch": 16.883388338833882,
      "grad_norm": 0.001953125,
      "learning_rate": 0.002165092157619675,
      "loss": 0.2303,
      "num_input_tokens_seen": 32388224,
      "step": 153470
    },
    {
      "epoch": 16.883938393839383,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.002164346941728231,
      "loss": 0.2303,
      "num_input_tokens_seen": 32389280,
      "step": 153475
    },
    {
      "epoch": 16.884488448844884,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0021636018441368453,
      "loss": 0.2293,
      "num_input_tokens_seen": 32390368,
      "step": 153480
    },
    {
      "epoch": 16.885038503850385,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.002162856864852382,
      "loss": 0.2324,
      "num_input_tokens_seen": 32391392,
      "step": 153485
    },
    {
      "epoch": 16.885588558855886,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0021621120038817043,
      "loss": 0.2308,
      "num_input_tokens_seen": 32392416,
      "step": 153490
    },
    {
      "epoch": 16.886138613861387,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0021613672612316847,
      "loss": 0.2309,
      "num_input_tokens_seen": 32393504,
      "step": 153495
    },
    {
      "epoch": 16.88668866886689,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.002160622636909176,
      "loss": 0.2303,
      "num_input_tokens_seen": 32394624,
      "step": 153500
    },
    {
      "epoch": 16.887238723872386,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.00215987813092105,
      "loss": 0.2319,
      "num_input_tokens_seen": 32395680,
      "step": 153505
    },
    {
      "epoch": 16.887788778877887,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.002159133743274169,
      "loss": 0.2309,
      "num_input_tokens_seen": 32396800,
      "step": 153510
    },
    {
      "epoch": 16.888338833883388,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.002158389473975386,
      "loss": 0.233,
      "num_input_tokens_seen": 32397920,
      "step": 153515
    },
    {
      "epoch": 16.88888888888889,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0021576453230315704,
      "loss": 0.2319,
      "num_input_tokens_seen": 32398944,
      "step": 153520
    },
    {
      "epoch": 16.88943894389439,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0021569012904495703,
      "loss": 0.2314,
      "num_input_tokens_seen": 32400000,
      "step": 153525
    },
    {
      "epoch": 16.88998899889989,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.00215615737623625,
      "loss": 0.2303,
      "num_input_tokens_seen": 32401024,
      "step": 153530
    },
    {
      "epoch": 16.89053905390539,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002155413580398466,
      "loss": 0.2314,
      "num_input_tokens_seen": 32402112,
      "step": 153535
    },
    {
      "epoch": 16.89108910891089,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0021546699029430685,
      "loss": 0.2308,
      "num_input_tokens_seen": 32403200,
      "step": 153540
    },
    {
      "epoch": 16.89163916391639,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0021539263438769183,
      "loss": 0.2293,
      "num_input_tokens_seen": 32404288,
      "step": 153545
    },
    {
      "epoch": 16.892189218921892,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.002153182903206861,
      "loss": 0.2314,
      "num_input_tokens_seen": 32405376,
      "step": 153550
    },
    {
      "epoch": 16.892739273927393,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0021524395809397576,
      "loss": 0.2309,
      "num_input_tokens_seen": 32406432,
      "step": 153555
    },
    {
      "epoch": 16.893289328932894,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002151696377082448,
      "loss": 0.2313,
      "num_input_tokens_seen": 32407424,
      "step": 153560
    },
    {
      "epoch": 16.893839383938392,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.00215095329164179,
      "loss": 0.2319,
      "num_input_tokens_seen": 32408480,
      "step": 153565
    },
    {
      "epoch": 16.894389438943893,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.002150210324624631,
      "loss": 0.2335,
      "num_input_tokens_seen": 32409536,
      "step": 153570
    },
    {
      "epoch": 16.894939493949394,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.0021494674760378146,
      "loss": 0.2308,
      "num_input_tokens_seen": 32410592,
      "step": 153575
    },
    {
      "epoch": 16.895489548954895,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.002148724745888195,
      "loss": 0.234,
      "num_input_tokens_seen": 32411616,
      "step": 153580
    },
    {
      "epoch": 16.896039603960396,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002147982134182608,
      "loss": 0.2303,
      "num_input_tokens_seen": 32412704,
      "step": 153585
    },
    {
      "epoch": 16.896589658965897,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.002147239640927903,
      "loss": 0.2293,
      "num_input_tokens_seen": 32413728,
      "step": 153590
    },
    {
      "epoch": 16.8971397139714,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.002146497266130927,
      "loss": 0.2298,
      "num_input_tokens_seen": 32414848,
      "step": 153595
    },
    {
      "epoch": 16.897689768976896,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0021457550097985145,
      "loss": 0.2329,
      "num_input_tokens_seen": 32416000,
      "step": 153600
    },
    {
      "epoch": 16.898239823982397,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0021450128719375126,
      "loss": 0.2319,
      "num_input_tokens_seen": 32417120,
      "step": 153605
    },
    {
      "epoch": 16.8987898789879,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.0021442708525547596,
      "loss": 0.2314,
      "num_input_tokens_seen": 32418080,
      "step": 153610
    },
    {
      "epoch": 16.8993399339934,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0021435289516570886,
      "loss": 0.2309,
      "num_input_tokens_seen": 32419168,
      "step": 153615
    },
    {
      "epoch": 16.8998899889989,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.002142787169251344,
      "loss": 0.2288,
      "num_input_tokens_seen": 32420224,
      "step": 153620
    },
    {
      "epoch": 16.9004400440044,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0021420455053443583,
      "loss": 0.2319,
      "num_input_tokens_seen": 32421312,
      "step": 153625
    },
    {
      "epoch": 16.900990099009903,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.0021413039599429743,
      "loss": 0.2335,
      "num_input_tokens_seen": 32422400,
      "step": 153630
    },
    {
      "epoch": 16.9015401540154,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.0021405625330540206,
      "loss": 0.2309,
      "num_input_tokens_seen": 32423488,
      "step": 153635
    },
    {
      "epoch": 16.9020902090209,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0021398212246843294,
      "loss": 0.2314,
      "num_input_tokens_seen": 32424544,
      "step": 153640
    },
    {
      "epoch": 16.902640264026402,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.002139080034840734,
      "loss": 0.2303,
      "num_input_tokens_seen": 32425568,
      "step": 153645
    },
    {
      "epoch": 16.903190319031903,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0021383389635300704,
      "loss": 0.2319,
      "num_input_tokens_seen": 32426592,
      "step": 153650
    },
    {
      "epoch": 16.903740374037405,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0021375980107591615,
      "loss": 0.2324,
      "num_input_tokens_seen": 32427584,
      "step": 153655
    },
    {
      "epoch": 16.904290429042906,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.002136857176534843,
      "loss": 0.2324,
      "num_input_tokens_seen": 32428544,
      "step": 153660
    },
    {
      "epoch": 16.904840484048403,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0021361164608639355,
      "loss": 0.2309,
      "num_input_tokens_seen": 32429632,
      "step": 153665
    },
    {
      "epoch": 16.905390539053904,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.0021353758637532743,
      "loss": 0.2314,
      "num_input_tokens_seen": 32430688,
      "step": 153670
    },
    {
      "epoch": 16.905940594059405,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0021346353852096777,
      "loss": 0.2303,
      "num_input_tokens_seen": 32431680,
      "step": 153675
    },
    {
      "epoch": 16.906490649064907,
      "grad_norm": 0.00081634521484375,
      "learning_rate": 0.002133895025239972,
      "loss": 0.2324,
      "num_input_tokens_seen": 32432736,
      "step": 153680
    },
    {
      "epoch": 16.907040704070408,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0021331547838509862,
      "loss": 0.2288,
      "num_input_tokens_seen": 32433792,
      "step": 153685
    },
    {
      "epoch": 16.90759075907591,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.0021324146610495354,
      "loss": 0.2309,
      "num_input_tokens_seen": 32434848,
      "step": 153690
    },
    {
      "epoch": 16.90814081408141,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0021316746568424473,
      "loss": 0.2308,
      "num_input_tokens_seen": 32435904,
      "step": 153695
    },
    {
      "epoch": 16.908690869086907,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.002130934771236536,
      "loss": 0.2319,
      "num_input_tokens_seen": 32436992,
      "step": 153700
    },
    {
      "epoch": 16.90924092409241,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0021301950042386234,
      "loss": 0.2314,
      "num_input_tokens_seen": 32437984,
      "step": 153705
    },
    {
      "epoch": 16.90979097909791,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.0021294553558555323,
      "loss": 0.2304,
      "num_input_tokens_seen": 32439008,
      "step": 153710
    },
    {
      "epoch": 16.91034103410341,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00212871582609407,
      "loss": 0.2329,
      "num_input_tokens_seen": 32440064,
      "step": 153715
    },
    {
      "epoch": 16.91089108910891,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0021279764149610618,
      "loss": 0.2303,
      "num_input_tokens_seen": 32441088,
      "step": 153720
    },
    {
      "epoch": 16.911441144114413,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.00212723712246332,
      "loss": 0.2303,
      "num_input_tokens_seen": 32442080,
      "step": 153725
    },
    {
      "epoch": 16.91199119911991,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.0021264979486076514,
      "loss": 0.2303,
      "num_input_tokens_seen": 32443136,
      "step": 153730
    },
    {
      "epoch": 16.91254125412541,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.0021257588934008747,
      "loss": 0.2329,
      "num_input_tokens_seen": 32444192,
      "step": 153735
    },
    {
      "epoch": 16.913091309130913,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.0021250199568498,
      "loss": 0.2313,
      "num_input_tokens_seen": 32445280,
      "step": 153740
    },
    {
      "epoch": 16.913641364136414,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.002124281138961242,
      "loss": 0.2303,
      "num_input_tokens_seen": 32446272,
      "step": 153745
    },
    {
      "epoch": 16.914191419141915,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0021235424397420067,
      "loss": 0.2319,
      "num_input_tokens_seen": 32447392,
      "step": 153750
    },
    {
      "epoch": 16.914741474147416,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0021228038591988986,
      "loss": 0.2314,
      "num_input_tokens_seen": 32448416,
      "step": 153755
    },
    {
      "epoch": 16.915291529152917,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0021220653973387286,
      "loss": 0.2303,
      "num_input_tokens_seen": 32449472,
      "step": 153760
    },
    {
      "epoch": 16.915841584158414,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.002121327054168302,
      "loss": 0.234,
      "num_input_tokens_seen": 32450592,
      "step": 153765
    },
    {
      "epoch": 16.916391639163916,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.0021205888296944265,
      "loss": 0.2303,
      "num_input_tokens_seen": 32451744,
      "step": 153770
    },
    {
      "epoch": 16.916941694169417,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0021198507239239043,
      "loss": 0.2298,
      "num_input_tokens_seen": 32452768,
      "step": 153775
    },
    {
      "epoch": 16.917491749174918,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0021191127368635344,
      "loss": 0.2319,
      "num_input_tokens_seen": 32453856,
      "step": 153780
    },
    {
      "epoch": 16.91804180418042,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0021183748685201253,
      "loss": 0.2309,
      "num_input_tokens_seen": 32454944,
      "step": 153785
    },
    {
      "epoch": 16.91859185918592,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.0021176371189004705,
      "loss": 0.2319,
      "num_input_tokens_seen": 32456000,
      "step": 153790
    },
    {
      "epoch": 16.919141914191417,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.002116899488011373,
      "loss": 0.2309,
      "num_input_tokens_seen": 32457056,
      "step": 153795
    },
    {
      "epoch": 16.91969196919692,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.002116161975859633,
      "loss": 0.2309,
      "num_input_tokens_seen": 32458144,
      "step": 153800
    },
    {
      "epoch": 16.92024202420242,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0021154245824520435,
      "loss": 0.2324,
      "num_input_tokens_seen": 32459168,
      "step": 153805
    },
    {
      "epoch": 16.92079207920792,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002114687307795406,
      "loss": 0.2324,
      "num_input_tokens_seen": 32460224,
      "step": 153810
    },
    {
      "epoch": 16.921342134213422,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0021139501518965087,
      "loss": 0.2277,
      "num_input_tokens_seen": 32461248,
      "step": 153815
    },
    {
      "epoch": 16.921892189218923,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0021132131147621495,
      "loss": 0.2319,
      "num_input_tokens_seen": 32462336,
      "step": 153820
    },
    {
      "epoch": 16.922442244224424,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.002112476196399127,
      "loss": 0.2308,
      "num_input_tokens_seen": 32463424,
      "step": 153825
    },
    {
      "epoch": 16.92299229922992,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.0021117393968142213,
      "loss": 0.2335,
      "num_input_tokens_seen": 32464512,
      "step": 153830
    },
    {
      "epoch": 16.923542354235423,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0021110027160142343,
      "loss": 0.2293,
      "num_input_tokens_seen": 32465664,
      "step": 153835
    },
    {
      "epoch": 16.924092409240924,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.002110266154005945,
      "loss": 0.2303,
      "num_input_tokens_seen": 32466720,
      "step": 153840
    },
    {
      "epoch": 16.924642464246425,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0021095297107961526,
      "loss": 0.2288,
      "num_input_tokens_seen": 32467776,
      "step": 153845
    },
    {
      "epoch": 16.925192519251926,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.002108793386391637,
      "loss": 0.2314,
      "num_input_tokens_seen": 32468768,
      "step": 153850
    },
    {
      "epoch": 16.925742574257427,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0021080571807991846,
      "loss": 0.2314,
      "num_input_tokens_seen": 32469792,
      "step": 153855
    },
    {
      "epoch": 16.926292629262925,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.0021073210940255878,
      "loss": 0.2324,
      "num_input_tokens_seen": 32470784,
      "step": 153860
    },
    {
      "epoch": 16.926842684268426,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.0021065851260776264,
      "loss": 0.2314,
      "num_input_tokens_seen": 32471840,
      "step": 153865
    },
    {
      "epoch": 16.927392739273927,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0021058492769620773,
      "loss": 0.2314,
      "num_input_tokens_seen": 32472896,
      "step": 153870
    },
    {
      "epoch": 16.927942794279428,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0021051135466857336,
      "loss": 0.2308,
      "num_input_tokens_seen": 32473952,
      "step": 153875
    },
    {
      "epoch": 16.92849284928493,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.002104377935255364,
      "loss": 0.2319,
      "num_input_tokens_seen": 32475008,
      "step": 153880
    },
    {
      "epoch": 16.92904290429043,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.002103642442677761,
      "loss": 0.2314,
      "num_input_tokens_seen": 32476096,
      "step": 153885
    },
    {
      "epoch": 16.92959295929593,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0021029070689596984,
      "loss": 0.2314,
      "num_input_tokens_seen": 32477120,
      "step": 153890
    },
    {
      "epoch": 16.93014301430143,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0021021718141079476,
      "loss": 0.2314,
      "num_input_tokens_seen": 32478112,
      "step": 153895
    },
    {
      "epoch": 16.93069306930693,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.002101436678129295,
      "loss": 0.2308,
      "num_input_tokens_seen": 32479168,
      "step": 153900
    },
    {
      "epoch": 16.93124312431243,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002100701661030508,
      "loss": 0.2314,
      "num_input_tokens_seen": 32480288,
      "step": 153905
    },
    {
      "epoch": 16.931793179317932,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.002099966762818363,
      "loss": 0.2298,
      "num_input_tokens_seen": 32481376,
      "step": 153910
    },
    {
      "epoch": 16.932343234323433,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0020992319834996385,
      "loss": 0.2329,
      "num_input_tokens_seen": 32482432,
      "step": 153915
    },
    {
      "epoch": 16.932893289328934,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.0020984973230810987,
      "loss": 0.2329,
      "num_input_tokens_seen": 32483456,
      "step": 153920
    },
    {
      "epoch": 16.933443344334435,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.0020977627815695213,
      "loss": 0.2304,
      "num_input_tokens_seen": 32484448,
      "step": 153925
    },
    {
      "epoch": 16.933993399339933,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0020970283589716696,
      "loss": 0.2314,
      "num_input_tokens_seen": 32485568,
      "step": 153930
    },
    {
      "epoch": 16.934543454345434,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002096294055294318,
      "loss": 0.2314,
      "num_input_tokens_seen": 32486624,
      "step": 153935
    },
    {
      "epoch": 16.935093509350935,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0020955598705442347,
      "loss": 0.2309,
      "num_input_tokens_seen": 32487680,
      "step": 153940
    },
    {
      "epoch": 16.935643564356436,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.00209482580472818,
      "loss": 0.2303,
      "num_input_tokens_seen": 32488768,
      "step": 153945
    },
    {
      "epoch": 16.936193619361937,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0020940918578529265,
      "loss": 0.2319,
      "num_input_tokens_seen": 32489760,
      "step": 153950
    },
    {
      "epoch": 16.936743674367438,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.002093358029925233,
      "loss": 0.2308,
      "num_input_tokens_seen": 32490784,
      "step": 153955
    },
    {
      "epoch": 16.937293729372936,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.002092624320951869,
      "loss": 0.2319,
      "num_input_tokens_seen": 32491840,
      "step": 153960
    },
    {
      "epoch": 16.937843784378437,
      "grad_norm": 0.006866455078125,
      "learning_rate": 0.002091890730939588,
      "loss": 0.2303,
      "num_input_tokens_seen": 32492832,
      "step": 153965
    },
    {
      "epoch": 16.938393839383938,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.002091157259895157,
      "loss": 0.2324,
      "num_input_tokens_seen": 32493920,
      "step": 153970
    },
    {
      "epoch": 16.93894389438944,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00209042390782534,
      "loss": 0.2329,
      "num_input_tokens_seen": 32494976,
      "step": 153975
    },
    {
      "epoch": 16.93949394939494,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.002089690674736886,
      "loss": 0.2314,
      "num_input_tokens_seen": 32496064,
      "step": 153980
    },
    {
      "epoch": 16.94004400440044,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.002088957560636563,
      "loss": 0.2319,
      "num_input_tokens_seen": 32497120,
      "step": 153985
    },
    {
      "epoch": 16.94059405940594,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002088224565531122,
      "loss": 0.2303,
      "num_input_tokens_seen": 32498240,
      "step": 153990
    },
    {
      "epoch": 16.94114411441144,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0020874916894273133,
      "loss": 0.2298,
      "num_input_tokens_seen": 32499232,
      "step": 153995
    },
    {
      "epoch": 16.94169416941694,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.002086758932331904,
      "loss": 0.2298,
      "num_input_tokens_seen": 32500256,
      "step": 154000
    },
    {
      "epoch": 16.942244224422442,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0020860262942516378,
      "loss": 0.2329,
      "num_input_tokens_seen": 32501312,
      "step": 154005
    },
    {
      "epoch": 16.942794279427943,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.0020852937751932735,
      "loss": 0.2314,
      "num_input_tokens_seen": 32502368,
      "step": 154010
    },
    {
      "epoch": 16.943344334433444,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00208456137516356,
      "loss": 0.2319,
      "num_input_tokens_seen": 32503456,
      "step": 154015
    },
    {
      "epoch": 16.943894389438945,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.002083829094169243,
      "loss": 0.2335,
      "num_input_tokens_seen": 32504576,
      "step": 154020
    },
    {
      "epoch": 16.944444444444443,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0020830969322170765,
      "loss": 0.2313,
      "num_input_tokens_seen": 32505600,
      "step": 154025
    },
    {
      "epoch": 16.944994499449944,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0020823648893138064,
      "loss": 0.2319,
      "num_input_tokens_seen": 32506656,
      "step": 154030
    },
    {
      "epoch": 16.945544554455445,
      "grad_norm": 0.0003948211669921875,
      "learning_rate": 0.002081632965466185,
      "loss": 0.2303,
      "num_input_tokens_seen": 32507712,
      "step": 154035
    },
    {
      "epoch": 16.946094609460946,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.002080901160680953,
      "loss": 0.2324,
      "num_input_tokens_seen": 32508736,
      "step": 154040
    },
    {
      "epoch": 16.946644664466447,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0020801694749648524,
      "loss": 0.2278,
      "num_input_tokens_seen": 32509760,
      "step": 154045
    },
    {
      "epoch": 16.94719471947195,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.002079437908324633,
      "loss": 0.2298,
      "num_input_tokens_seen": 32510784,
      "step": 154050
    },
    {
      "epoch": 16.94774477447745,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.0020787064607670318,
      "loss": 0.2319,
      "num_input_tokens_seen": 32511872,
      "step": 154055
    },
    {
      "epoch": 16.948294829482947,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.0020779751322987927,
      "loss": 0.234,
      "num_input_tokens_seen": 32512960,
      "step": 154060
    },
    {
      "epoch": 16.948844884488448,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.00207724392292666,
      "loss": 0.2324,
      "num_input_tokens_seen": 32513984,
      "step": 154065
    },
    {
      "epoch": 16.94939493949395,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002076512832657366,
      "loss": 0.2324,
      "num_input_tokens_seen": 32514976,
      "step": 154070
    },
    {
      "epoch": 16.94994499449945,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.0020757818614976545,
      "loss": 0.2324,
      "num_input_tokens_seen": 32516032,
      "step": 154075
    },
    {
      "epoch": 16.95049504950495,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.002075051009454257,
      "loss": 0.2293,
      "num_input_tokens_seen": 32517184,
      "step": 154080
    },
    {
      "epoch": 16.951045104510452,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.002074320276533912,
      "loss": 0.2303,
      "num_input_tokens_seen": 32518304,
      "step": 154085
    },
    {
      "epoch": 16.95159515951595,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002073589662743358,
      "loss": 0.2319,
      "num_input_tokens_seen": 32519360,
      "step": 154090
    },
    {
      "epoch": 16.95214521452145,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.002072859168089323,
      "loss": 0.2308,
      "num_input_tokens_seen": 32520416,
      "step": 154095
    },
    {
      "epoch": 16.952695269526952,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.0020721287925785452,
      "loss": 0.2308,
      "num_input_tokens_seen": 32521440,
      "step": 154100
    },
    {
      "epoch": 16.953245324532453,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0020713985362177524,
      "loss": 0.2309,
      "num_input_tokens_seen": 32522592,
      "step": 154105
    },
    {
      "epoch": 16.953795379537954,
      "grad_norm": 0.005859375,
      "learning_rate": 0.002070668399013672,
      "loss": 0.2324,
      "num_input_tokens_seen": 32523648,
      "step": 154110
    },
    {
      "epoch": 16.954345434543455,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.002069938380973036,
      "loss": 0.2324,
      "num_input_tokens_seen": 32524704,
      "step": 154115
    },
    {
      "epoch": 16.954895489548957,
      "grad_norm": 0.005859375,
      "learning_rate": 0.002069208482102575,
      "loss": 0.2283,
      "num_input_tokens_seen": 32525760,
      "step": 154120
    },
    {
      "epoch": 16.955445544554454,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.002068478702409018,
      "loss": 0.2314,
      "num_input_tokens_seen": 32526816,
      "step": 154125
    },
    {
      "epoch": 16.955995599559955,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.002067749041899087,
      "loss": 0.2324,
      "num_input_tokens_seen": 32527904,
      "step": 154130
    },
    {
      "epoch": 16.956545654565456,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.0020670195005795047,
      "loss": 0.2319,
      "num_input_tokens_seen": 32529024,
      "step": 154135
    },
    {
      "epoch": 16.957095709570957,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.002066290078456997,
      "loss": 0.2314,
      "num_input_tokens_seen": 32530048,
      "step": 154140
    },
    {
      "epoch": 16.95764576457646,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0020655607755382874,
      "loss": 0.2314,
      "num_input_tokens_seen": 32531104,
      "step": 154145
    },
    {
      "epoch": 16.95819581958196,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.0020648315918301028,
      "loss": 0.2319,
      "num_input_tokens_seen": 32532288,
      "step": 154150
    },
    {
      "epoch": 16.958745874587457,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0020641025273391575,
      "loss": 0.2314,
      "num_input_tokens_seen": 32533344,
      "step": 154155
    },
    {
      "epoch": 16.959295929592958,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.002063373582072168,
      "loss": 0.2308,
      "num_input_tokens_seen": 32534336,
      "step": 154160
    },
    {
      "epoch": 16.95984598459846,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.002062644756035861,
      "loss": 0.2329,
      "num_input_tokens_seen": 32535456,
      "step": 154165
    },
    {
      "epoch": 16.96039603960396,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.002061916049236945,
      "loss": 0.2324,
      "num_input_tokens_seen": 32536512,
      "step": 154170
    },
    {
      "epoch": 16.96094609460946,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00206118746168214,
      "loss": 0.2314,
      "num_input_tokens_seen": 32537568,
      "step": 154175
    },
    {
      "epoch": 16.961496149614963,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0020604589933781664,
      "loss": 0.2314,
      "num_input_tokens_seen": 32538624,
      "step": 154180
    },
    {
      "epoch": 16.962046204620464,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0020597306443317282,
      "loss": 0.2335,
      "num_input_tokens_seen": 32539680,
      "step": 154185
    },
    {
      "epoch": 16.96259625962596,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002059002414549548,
      "loss": 0.2319,
      "num_input_tokens_seen": 32540672,
      "step": 154190
    },
    {
      "epoch": 16.963146314631462,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.00205827430403833,
      "loss": 0.2309,
      "num_input_tokens_seen": 32541728,
      "step": 154195
    },
    {
      "epoch": 16.963696369636963,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.002057546312804786,
      "loss": 0.2324,
      "num_input_tokens_seen": 32542784,
      "step": 154200
    },
    {
      "epoch": 16.964246424642464,
      "grad_norm": 0.00057220458984375,
      "learning_rate": 0.0020568184408556313,
      "loss": 0.2303,
      "num_input_tokens_seen": 32543808,
      "step": 154205
    },
    {
      "epoch": 16.964796479647966,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.0020560906881975666,
      "loss": 0.2308,
      "num_input_tokens_seen": 32544832,
      "step": 154210
    },
    {
      "epoch": 16.965346534653467,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.0020553630548373067,
      "loss": 0.2324,
      "num_input_tokens_seen": 32545888,
      "step": 154215
    },
    {
      "epoch": 16.965896589658964,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.002054635540781552,
      "loss": 0.2329,
      "num_input_tokens_seen": 32546976,
      "step": 154220
    },
    {
      "epoch": 16.966446644664465,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.002053908146037011,
      "loss": 0.2324,
      "num_input_tokens_seen": 32547936,
      "step": 154225
    },
    {
      "epoch": 16.966996699669966,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002053180870610384,
      "loss": 0.2314,
      "num_input_tokens_seen": 32548960,
      "step": 154230
    },
    {
      "epoch": 16.967546754675467,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0020524537145083757,
      "loss": 0.2319,
      "num_input_tokens_seen": 32550048,
      "step": 154235
    },
    {
      "epoch": 16.96809680968097,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.002051726677737694,
      "loss": 0.2314,
      "num_input_tokens_seen": 32551104,
      "step": 154240
    },
    {
      "epoch": 16.96864686468647,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0020509997603050333,
      "loss": 0.2319,
      "num_input_tokens_seen": 32552224,
      "step": 154245
    },
    {
      "epoch": 16.96919691969197,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00205027296221709,
      "loss": 0.2314,
      "num_input_tokens_seen": 32553312,
      "step": 154250
    },
    {
      "epoch": 16.96974697469747,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0020495462834805673,
      "loss": 0.2324,
      "num_input_tokens_seen": 32554336,
      "step": 154255
    },
    {
      "epoch": 16.97029702970297,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0020488197241021627,
      "loss": 0.2324,
      "num_input_tokens_seen": 32555456,
      "step": 154260
    },
    {
      "epoch": 16.97084708470847,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.0020480932840885764,
      "loss": 0.2319,
      "num_input_tokens_seen": 32556480,
      "step": 154265
    },
    {
      "epoch": 16.97139713971397,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0020473669634464976,
      "loss": 0.2303,
      "num_input_tokens_seen": 32557504,
      "step": 154270
    },
    {
      "epoch": 16.971947194719473,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0020466407621826193,
      "loss": 0.2314,
      "num_input_tokens_seen": 32558528,
      "step": 154275
    },
    {
      "epoch": 16.972497249724974,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.0020459146803036414,
      "loss": 0.2308,
      "num_input_tokens_seen": 32559552,
      "step": 154280
    },
    {
      "epoch": 16.97304730473047,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0020451887178162486,
      "loss": 0.2314,
      "num_input_tokens_seen": 32560608,
      "step": 154285
    },
    {
      "epoch": 16.973597359735972,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0020444628747271335,
      "loss": 0.2314,
      "num_input_tokens_seen": 32561664,
      "step": 154290
    },
    {
      "epoch": 16.974147414741473,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002043737151042993,
      "loss": 0.2303,
      "num_input_tokens_seen": 32562752,
      "step": 154295
    },
    {
      "epoch": 16.974697469746975,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002043011546770506,
      "loss": 0.2335,
      "num_input_tokens_seen": 32563744,
      "step": 154300
    },
    {
      "epoch": 16.975247524752476,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.002042286061916366,
      "loss": 0.2314,
      "num_input_tokens_seen": 32564704,
      "step": 154305
    },
    {
      "epoch": 16.975797579757977,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.0020415606964872563,
      "loss": 0.2308,
      "num_input_tokens_seen": 32565792,
      "step": 154310
    },
    {
      "epoch": 16.976347634763478,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.002040835450489864,
      "loss": 0.2335,
      "num_input_tokens_seen": 32566848,
      "step": 154315
    },
    {
      "epoch": 16.976897689768975,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0020401103239308742,
      "loss": 0.2303,
      "num_input_tokens_seen": 32567904,
      "step": 154320
    },
    {
      "epoch": 16.977447744774476,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0020393853168169667,
      "loss": 0.2314,
      "num_input_tokens_seen": 32568960,
      "step": 154325
    },
    {
      "epoch": 16.977997799779978,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.002038660429154829,
      "loss": 0.2314,
      "num_input_tokens_seen": 32569952,
      "step": 154330
    },
    {
      "epoch": 16.97854785478548,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0020379356609511352,
      "loss": 0.2303,
      "num_input_tokens_seen": 32571008,
      "step": 154335
    },
    {
      "epoch": 16.97909790979098,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002037211012212572,
      "loss": 0.2324,
      "num_input_tokens_seen": 32572160,
      "step": 154340
    },
    {
      "epoch": 16.97964796479648,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002036486482945811,
      "loss": 0.2319,
      "num_input_tokens_seen": 32573248,
      "step": 154345
    },
    {
      "epoch": 16.980198019801982,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0020357620731575343,
      "loss": 0.2319,
      "num_input_tokens_seen": 32574304,
      "step": 154350
    },
    {
      "epoch": 16.98074807480748,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0020350377828544206,
      "loss": 0.2303,
      "num_input_tokens_seen": 32575392,
      "step": 154355
    },
    {
      "epoch": 16.98129812981298,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0020343136120431387,
      "loss": 0.2329,
      "num_input_tokens_seen": 32576448,
      "step": 154360
    },
    {
      "epoch": 16.98184818481848,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.002033589560730371,
      "loss": 0.2303,
      "num_input_tokens_seen": 32577472,
      "step": 154365
    },
    {
      "epoch": 16.982398239823983,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.0020328656289227824,
      "loss": 0.2309,
      "num_input_tokens_seen": 32578560,
      "step": 154370
    },
    {
      "epoch": 16.982948294829484,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0020321418166270493,
      "loss": 0.2329,
      "num_input_tokens_seen": 32579616,
      "step": 154375
    },
    {
      "epoch": 16.983498349834985,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0020314181238498446,
      "loss": 0.2309,
      "num_input_tokens_seen": 32580640,
      "step": 154380
    },
    {
      "epoch": 16.984048404840483,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0020306945505978347,
      "loss": 0.2314,
      "num_input_tokens_seen": 32581728,
      "step": 154385
    },
    {
      "epoch": 16.984598459845984,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0020299710968776923,
      "loss": 0.2309,
      "num_input_tokens_seen": 32582720,
      "step": 154390
    },
    {
      "epoch": 16.985148514851485,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0020292477626960826,
      "loss": 0.2308,
      "num_input_tokens_seen": 32583744,
      "step": 154395
    },
    {
      "epoch": 16.985698569856986,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.0020285245480596696,
      "loss": 0.2334,
      "num_input_tokens_seen": 32584832,
      "step": 154400
    },
    {
      "epoch": 16.986248624862487,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.00202780145297512,
      "loss": 0.2309,
      "num_input_tokens_seen": 32585888,
      "step": 154405
    },
    {
      "epoch": 16.986798679867988,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0020270784774491023,
      "loss": 0.2319,
      "num_input_tokens_seen": 32586912,
      "step": 154410
    },
    {
      "epoch": 16.98734873487349,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.002026355621488279,
      "loss": 0.2303,
      "num_input_tokens_seen": 32587936,
      "step": 154415
    },
    {
      "epoch": 16.987898789878987,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.002025632885099311,
      "loss": 0.2309,
      "num_input_tokens_seen": 32588960,
      "step": 154420
    },
    {
      "epoch": 16.988448844884488,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002024910268288855,
      "loss": 0.2314,
      "num_input_tokens_seen": 32589984,
      "step": 154425
    },
    {
      "epoch": 16.98899889988999,
      "grad_norm": 0.00070953369140625,
      "learning_rate": 0.0020241877710635747,
      "loss": 0.2335,
      "num_input_tokens_seen": 32591072,
      "step": 154430
    },
    {
      "epoch": 16.98954895489549,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.002023465393430134,
      "loss": 0.2319,
      "num_input_tokens_seen": 32592160,
      "step": 154435
    },
    {
      "epoch": 16.99009900990099,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0020227431353951808,
      "loss": 0.2308,
      "num_input_tokens_seen": 32593312,
      "step": 154440
    },
    {
      "epoch": 16.990649064906492,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0020220209969653816,
      "loss": 0.2298,
      "num_input_tokens_seen": 32594368,
      "step": 154445
    },
    {
      "epoch": 16.99119911991199,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.002021298978147385,
      "loss": 0.2319,
      "num_input_tokens_seen": 32595392,
      "step": 154450
    },
    {
      "epoch": 16.99174917491749,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.00202057707894785,
      "loss": 0.2314,
      "num_input_tokens_seen": 32596416,
      "step": 154455
    },
    {
      "epoch": 16.992299229922992,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.0020198552993734246,
      "loss": 0.2324,
      "num_input_tokens_seen": 32597504,
      "step": 154460
    },
    {
      "epoch": 16.992849284928493,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.0020191336394307655,
      "loss": 0.234,
      "num_input_tokens_seen": 32598560,
      "step": 154465
    },
    {
      "epoch": 16.993399339933994,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.002018412099126524,
      "loss": 0.2309,
      "num_input_tokens_seen": 32599584,
      "step": 154470
    },
    {
      "epoch": 16.993949394939495,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.002017690678467348,
      "loss": 0.2303,
      "num_input_tokens_seen": 32600672,
      "step": 154475
    },
    {
      "epoch": 16.994499449944996,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0020169693774598906,
      "loss": 0.2298,
      "num_input_tokens_seen": 32601792,
      "step": 154480
    },
    {
      "epoch": 16.995049504950494,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0020162481961107912,
      "loss": 0.2283,
      "num_input_tokens_seen": 32602848,
      "step": 154485
    },
    {
      "epoch": 16.995599559955995,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0020155271344267033,
      "loss": 0.2324,
      "num_input_tokens_seen": 32603872,
      "step": 154490
    },
    {
      "epoch": 16.996149614961496,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0020148061924142747,
      "loss": 0.2329,
      "num_input_tokens_seen": 32604896,
      "step": 154495
    },
    {
      "epoch": 16.996699669966997,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.002014085370080143,
      "loss": 0.2298,
      "num_input_tokens_seen": 32606048,
      "step": 154500
    },
    {
      "epoch": 16.997249724972498,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0020133646674309583,
      "loss": 0.2314,
      "num_input_tokens_seen": 32607104,
      "step": 154505
    },
    {
      "epoch": 16.997799779978,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.0020126440844733607,
      "loss": 0.2309,
      "num_input_tokens_seen": 32608160,
      "step": 154510
    },
    {
      "epoch": 16.998349834983497,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0020119236212139875,
      "loss": 0.2298,
      "num_input_tokens_seen": 32609280,
      "step": 154515
    },
    {
      "epoch": 16.998899889988998,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.002011203277659482,
      "loss": 0.2314,
      "num_input_tokens_seen": 32610336,
      "step": 154520
    },
    {
      "epoch": 16.9994499449945,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.002010483053816482,
      "loss": 0.2303,
      "num_input_tokens_seen": 32611456,
      "step": 154525
    },
    {
      "epoch": 17.0,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.002009762949691631,
      "loss": 0.2298,
      "num_input_tokens_seen": 32612480,
      "step": 154530
    },
    {
      "epoch": 17.0,
      "eval_loss": 0.2313462793827057,
      "eval_runtime": 60.6475,
      "eval_samples_per_second": 66.614,
      "eval_steps_per_second": 16.654,
      "num_input_tokens_seen": 32612480,
      "step": 154530
    },
    {
      "epoch": 17.0005500550055,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0020090429652915597,
      "loss": 0.2298,
      "num_input_tokens_seen": 32613504,
      "step": 154535
    },
    {
      "epoch": 17.001100110011002,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.002008323100622905,
      "loss": 0.2309,
      "num_input_tokens_seen": 32614592,
      "step": 154540
    },
    {
      "epoch": 17.001650165016503,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0020076033556923,
      "loss": 0.2319,
      "num_input_tokens_seen": 32615680,
      "step": 154545
    },
    {
      "epoch": 17.002200220022,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0020068837305063825,
      "loss": 0.2314,
      "num_input_tokens_seen": 32616736,
      "step": 154550
    },
    {
      "epoch": 17.002750275027502,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0020061642250717854,
      "loss": 0.2324,
      "num_input_tokens_seen": 32617792,
      "step": 154555
    },
    {
      "epoch": 17.003300330033003,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0020054448393951366,
      "loss": 0.2303,
      "num_input_tokens_seen": 32618848,
      "step": 154560
    },
    {
      "epoch": 17.003850385038504,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.002004725573483065,
      "loss": 0.2303,
      "num_input_tokens_seen": 32619968,
      "step": 154565
    },
    {
      "epoch": 17.004400440044005,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.002004006427342206,
      "loss": 0.2314,
      "num_input_tokens_seen": 32621056,
      "step": 154570
    },
    {
      "epoch": 17.004950495049506,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0020032874009791794,
      "loss": 0.2329,
      "num_input_tokens_seen": 32622112,
      "step": 154575
    },
    {
      "epoch": 17.005500550055004,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0020025684944006174,
      "loss": 0.2298,
      "num_input_tokens_seen": 32623136,
      "step": 154580
    },
    {
      "epoch": 17.006050605060505,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.002001849707613147,
      "loss": 0.2335,
      "num_input_tokens_seen": 32624192,
      "step": 154585
    },
    {
      "epoch": 17.006600660066006,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0020011310406233885,
      "loss": 0.2314,
      "num_input_tokens_seen": 32625216,
      "step": 154590
    },
    {
      "epoch": 17.007150715071507,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.0020004124934379698,
      "loss": 0.2335,
      "num_input_tokens_seen": 32626304,
      "step": 154595
    },
    {
      "epoch": 17.007700770077008,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.0019996940660635086,
      "loss": 0.2324,
      "num_input_tokens_seen": 32627328,
      "step": 154600
    },
    {
      "epoch": 17.00825082508251,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0019989757585066287,
      "loss": 0.2309,
      "num_input_tokens_seen": 32628416,
      "step": 154605
    },
    {
      "epoch": 17.00880088008801,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.001998257570773954,
      "loss": 0.2314,
      "num_input_tokens_seen": 32629472,
      "step": 154610
    },
    {
      "epoch": 17.009350935093508,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.001997539502872098,
      "loss": 0.2314,
      "num_input_tokens_seen": 32630496,
      "step": 154615
    },
    {
      "epoch": 17.00990099009901,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.001996821554807684,
      "loss": 0.2309,
      "num_input_tokens_seen": 32631552,
      "step": 154620
    },
    {
      "epoch": 17.01045104510451,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.001996103726587326,
      "loss": 0.234,
      "num_input_tokens_seen": 32632544,
      "step": 154625
    },
    {
      "epoch": 17.01100110011001,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0019953860182176375,
      "loss": 0.2329,
      "num_input_tokens_seen": 32633568,
      "step": 154630
    },
    {
      "epoch": 17.011551155115512,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.0019946684297052365,
      "loss": 0.2319,
      "num_input_tokens_seen": 32634688,
      "step": 154635
    },
    {
      "epoch": 17.012101210121013,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.001993950961056734,
      "loss": 0.2314,
      "num_input_tokens_seen": 32635712,
      "step": 154640
    },
    {
      "epoch": 17.01265126512651,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0019932336122787486,
      "loss": 0.2303,
      "num_input_tokens_seen": 32636736,
      "step": 154645
    },
    {
      "epoch": 17.013201320132012,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.001992516383377888,
      "loss": 0.2308,
      "num_input_tokens_seen": 32637792,
      "step": 154650
    },
    {
      "epoch": 17.013751375137513,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0019917992743607588,
      "loss": 0.2319,
      "num_input_tokens_seen": 32638816,
      "step": 154655
    },
    {
      "epoch": 17.014301430143014,
      "grad_norm": 0.00069427490234375,
      "learning_rate": 0.001991082285233977,
      "loss": 0.2319,
      "num_input_tokens_seen": 32639904,
      "step": 154660
    },
    {
      "epoch": 17.014851485148515,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0019903654160041413,
      "loss": 0.2319,
      "num_input_tokens_seen": 32640992,
      "step": 154665
    },
    {
      "epoch": 17.015401540154016,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0019896486666778723,
      "loss": 0.2319,
      "num_input_tokens_seen": 32642048,
      "step": 154670
    },
    {
      "epoch": 17.015951595159517,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0019889320372617667,
      "loss": 0.2314,
      "num_input_tokens_seen": 32643104,
      "step": 154675
    },
    {
      "epoch": 17.016501650165015,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.001988215527762428,
      "loss": 0.2324,
      "num_input_tokens_seen": 32644128,
      "step": 154680
    },
    {
      "epoch": 17.017051705170516,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.0019874991381864686,
      "loss": 0.2308,
      "num_input_tokens_seen": 32645184,
      "step": 154685
    },
    {
      "epoch": 17.017601760176017,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0019867828685404803,
      "loss": 0.2288,
      "num_input_tokens_seen": 32646336,
      "step": 154690
    },
    {
      "epoch": 17.01815181518152,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.001986066718831071,
      "loss": 0.2314,
      "num_input_tokens_seen": 32647456,
      "step": 154695
    },
    {
      "epoch": 17.01870187018702,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.001985350689064842,
      "loss": 0.2324,
      "num_input_tokens_seen": 32648544,
      "step": 154700
    },
    {
      "epoch": 17.01925192519252,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.001984634779248388,
      "loss": 0.2303,
      "num_input_tokens_seen": 32649568,
      "step": 154705
    },
    {
      "epoch": 17.019801980198018,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.001983918989388314,
      "loss": 0.2329,
      "num_input_tokens_seen": 32650656,
      "step": 154710
    },
    {
      "epoch": 17.02035203520352,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.0019832033194912098,
      "loss": 0.2288,
      "num_input_tokens_seen": 32651744,
      "step": 154715
    },
    {
      "epoch": 17.02090209020902,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001982487769563676,
      "loss": 0.2329,
      "num_input_tokens_seen": 32652736,
      "step": 154720
    },
    {
      "epoch": 17.02145214521452,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.001981772339612305,
      "loss": 0.2319,
      "num_input_tokens_seen": 32653728,
      "step": 154725
    },
    {
      "epoch": 17.022002200220022,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.001981057029643689,
      "loss": 0.2319,
      "num_input_tokens_seen": 32654752,
      "step": 154730
    },
    {
      "epoch": 17.022552255225524,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0019803418396644283,
      "loss": 0.2335,
      "num_input_tokens_seen": 32655744,
      "step": 154735
    },
    {
      "epoch": 17.023102310231025,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.0019796267696811057,
      "loss": 0.2319,
      "num_input_tokens_seen": 32656768,
      "step": 154740
    },
    {
      "epoch": 17.023652365236522,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0019789118197003196,
      "loss": 0.2319,
      "num_input_tokens_seen": 32657856,
      "step": 154745
    },
    {
      "epoch": 17.024202420242023,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0019781969897286512,
      "loss": 0.2329,
      "num_input_tokens_seen": 32658944,
      "step": 154750
    },
    {
      "epoch": 17.024752475247524,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0019774822797726916,
      "loss": 0.2309,
      "num_input_tokens_seen": 32660000,
      "step": 154755
    },
    {
      "epoch": 17.025302530253025,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.001976767689839034,
      "loss": 0.2309,
      "num_input_tokens_seen": 32661056,
      "step": 154760
    },
    {
      "epoch": 17.025852585258527,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.001976053219934255,
      "loss": 0.2324,
      "num_input_tokens_seen": 32662144,
      "step": 154765
    },
    {
      "epoch": 17.026402640264028,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0019753388700649477,
      "loss": 0.2324,
      "num_input_tokens_seen": 32663200,
      "step": 154770
    },
    {
      "epoch": 17.02695269526953,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.001974624640237693,
      "loss": 0.2314,
      "num_input_tokens_seen": 32664224,
      "step": 154775
    },
    {
      "epoch": 17.027502750275026,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.001973910530459065,
      "loss": 0.2314,
      "num_input_tokens_seen": 32665312,
      "step": 154780
    },
    {
      "epoch": 17.028052805280527,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.0019731965407356607,
      "loss": 0.2319,
      "num_input_tokens_seen": 32666368,
      "step": 154785
    },
    {
      "epoch": 17.02860286028603,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.001972482671074048,
      "loss": 0.2309,
      "num_input_tokens_seen": 32667392,
      "step": 154790
    },
    {
      "epoch": 17.02915291529153,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0019717689214808166,
      "loss": 0.2324,
      "num_input_tokens_seen": 32668416,
      "step": 154795
    },
    {
      "epoch": 17.02970297029703,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0019710552919625377,
      "loss": 0.2298,
      "num_input_tokens_seen": 32669472,
      "step": 154800
    },
    {
      "epoch": 17.03025302530253,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.0019703417825257872,
      "loss": 0.2329,
      "num_input_tokens_seen": 32670464,
      "step": 154805
    },
    {
      "epoch": 17.03080308030803,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0019696283931771464,
      "loss": 0.2308,
      "num_input_tokens_seen": 32671520,
      "step": 154810
    },
    {
      "epoch": 17.03135313531353,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0019689151239231893,
      "loss": 0.2314,
      "num_input_tokens_seen": 32672512,
      "step": 154815
    },
    {
      "epoch": 17.03190319031903,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0019682019747704863,
      "loss": 0.2324,
      "num_input_tokens_seen": 32673536,
      "step": 154820
    },
    {
      "epoch": 17.032453245324533,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0019674889457256158,
      "loss": 0.2303,
      "num_input_tokens_seen": 32674560,
      "step": 154825
    },
    {
      "epoch": 17.033003300330034,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0019667760367951415,
      "loss": 0.2324,
      "num_input_tokens_seen": 32675648,
      "step": 154830
    },
    {
      "epoch": 17.033553355335535,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.001966063247985642,
      "loss": 0.2324,
      "num_input_tokens_seen": 32676672,
      "step": 154835
    },
    {
      "epoch": 17.034103410341036,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001965350579303681,
      "loss": 0.2335,
      "num_input_tokens_seen": 32677792,
      "step": 154840
    },
    {
      "epoch": 17.034653465346533,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0019646380307558288,
      "loss": 0.2298,
      "num_input_tokens_seen": 32678912,
      "step": 154845
    },
    {
      "epoch": 17.035203520352034,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0019639256023486554,
      "loss": 0.2319,
      "num_input_tokens_seen": 32680000,
      "step": 154850
    },
    {
      "epoch": 17.035753575357536,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.001963213294088723,
      "loss": 0.2314,
      "num_input_tokens_seen": 32681088,
      "step": 154855
    },
    {
      "epoch": 17.036303630363037,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.001962501105982599,
      "loss": 0.2308,
      "num_input_tokens_seen": 32682112,
      "step": 154860
    },
    {
      "epoch": 17.036853685368538,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.001961789038036845,
      "loss": 0.2303,
      "num_input_tokens_seen": 32683072,
      "step": 154865
    },
    {
      "epoch": 17.03740374037404,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0019610770902580245,
      "loss": 0.2293,
      "num_input_tokens_seen": 32684128,
      "step": 154870
    },
    {
      "epoch": 17.037953795379536,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.001960365262652703,
      "loss": 0.2324,
      "num_input_tokens_seen": 32685120,
      "step": 154875
    },
    {
      "epoch": 17.038503850385037,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001959653555227434,
      "loss": 0.2335,
      "num_input_tokens_seen": 32686208,
      "step": 154880
    },
    {
      "epoch": 17.03905390539054,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0019589419679887842,
      "loss": 0.2324,
      "num_input_tokens_seen": 32687392,
      "step": 154885
    },
    {
      "epoch": 17.03960396039604,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.001958230500943308,
      "loss": 0.2303,
      "num_input_tokens_seen": 32688416,
      "step": 154890
    },
    {
      "epoch": 17.04015401540154,
      "grad_norm": 0.0006866455078125,
      "learning_rate": 0.0019575191540975624,
      "loss": 0.2329,
      "num_input_tokens_seen": 32689440,
      "step": 154895
    },
    {
      "epoch": 17.040704070407042,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.0019568079274581013,
      "loss": 0.2324,
      "num_input_tokens_seen": 32690496,
      "step": 154900
    },
    {
      "epoch": 17.041254125412543,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.001956096821031485,
      "loss": 0.2314,
      "num_input_tokens_seen": 32691584,
      "step": 154905
    },
    {
      "epoch": 17.04180418041804,
      "grad_norm": 0.005126953125,
      "learning_rate": 0.001955385834824268,
      "loss": 0.2309,
      "num_input_tokens_seen": 32692640,
      "step": 154910
    },
    {
      "epoch": 17.04235423542354,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.001954674968843,
      "loss": 0.2309,
      "num_input_tokens_seen": 32693696,
      "step": 154915
    },
    {
      "epoch": 17.042904290429043,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00195396422309423,
      "loss": 0.2314,
      "num_input_tokens_seen": 32694784,
      "step": 154920
    },
    {
      "epoch": 17.043454345434544,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0019532535975845117,
      "loss": 0.2319,
      "num_input_tokens_seen": 32695776,
      "step": 154925
    },
    {
      "epoch": 17.044004400440045,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.0019525430923203935,
      "loss": 0.2314,
      "num_input_tokens_seen": 32696864,
      "step": 154930
    },
    {
      "epoch": 17.044554455445546,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0019518327073084285,
      "loss": 0.2303,
      "num_input_tokens_seen": 32697856,
      "step": 154935
    },
    {
      "epoch": 17.045104510451043,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0019511224425551614,
      "loss": 0.2293,
      "num_input_tokens_seen": 32698944,
      "step": 154940
    },
    {
      "epoch": 17.045654565456545,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.001950412298067134,
      "loss": 0.2329,
      "num_input_tokens_seen": 32700000,
      "step": 154945
    },
    {
      "epoch": 17.046204620462046,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0019497022738508972,
      "loss": 0.2309,
      "num_input_tokens_seen": 32701024,
      "step": 154950
    },
    {
      "epoch": 17.046754675467547,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.001948992369912988,
      "loss": 0.2308,
      "num_input_tokens_seen": 32702048,
      "step": 154955
    },
    {
      "epoch": 17.047304730473048,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.0019482825862599555,
      "loss": 0.233,
      "num_input_tokens_seen": 32703200,
      "step": 154960
    },
    {
      "epoch": 17.04785478547855,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0019475729228983418,
      "loss": 0.2314,
      "num_input_tokens_seen": 32704224,
      "step": 154965
    },
    {
      "epoch": 17.04840484048405,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.0019468633798346829,
      "loss": 0.2324,
      "num_input_tokens_seen": 32705312,
      "step": 154970
    },
    {
      "epoch": 17.048954895489548,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0019461539570755236,
      "loss": 0.2314,
      "num_input_tokens_seen": 32706336,
      "step": 154975
    },
    {
      "epoch": 17.04950495049505,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0019454446546273956,
      "loss": 0.2324,
      "num_input_tokens_seen": 32707392,
      "step": 154980
    },
    {
      "epoch": 17.05005500550055,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0019447354724968417,
      "loss": 0.2308,
      "num_input_tokens_seen": 32708448,
      "step": 154985
    },
    {
      "epoch": 17.05060506050605,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0019440264106903987,
      "loss": 0.2314,
      "num_input_tokens_seen": 32709472,
      "step": 154990
    },
    {
      "epoch": 17.051155115511552,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.001943317469214596,
      "loss": 0.2308,
      "num_input_tokens_seen": 32710592,
      "step": 154995
    },
    {
      "epoch": 17.051705170517053,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0019426086480759735,
      "loss": 0.2324,
      "num_input_tokens_seen": 32711616,
      "step": 155000
    },
    {
      "epoch": 17.05225522552255,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0019418999472810626,
      "loss": 0.2303,
      "num_input_tokens_seen": 32712576,
      "step": 155005
    },
    {
      "epoch": 17.05280528052805,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.00194119136683639,
      "loss": 0.2309,
      "num_input_tokens_seen": 32713664,
      "step": 155010
    },
    {
      "epoch": 17.053355335533553,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0019404829067484917,
      "loss": 0.2324,
      "num_input_tokens_seen": 32714752,
      "step": 155015
    },
    {
      "epoch": 17.053905390539054,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.0019397745670238946,
      "loss": 0.2314,
      "num_input_tokens_seen": 32715808,
      "step": 155020
    },
    {
      "epoch": 17.054455445544555,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.0019390663476691317,
      "loss": 0.2324,
      "num_input_tokens_seen": 32716864,
      "step": 155025
    },
    {
      "epoch": 17.055005500550056,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0019383582486907274,
      "loss": 0.2314,
      "num_input_tokens_seen": 32717920,
      "step": 155030
    },
    {
      "epoch": 17.055555555555557,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0019376502700952035,
      "loss": 0.2303,
      "num_input_tokens_seen": 32719008,
      "step": 155035
    },
    {
      "epoch": 17.056105610561055,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0019369424118890898,
      "loss": 0.2308,
      "num_input_tokens_seen": 32720128,
      "step": 155040
    },
    {
      "epoch": 17.056655665566556,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.001936234674078911,
      "loss": 0.2298,
      "num_input_tokens_seen": 32721216,
      "step": 155045
    },
    {
      "epoch": 17.057205720572057,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0019355270566711901,
      "loss": 0.2324,
      "num_input_tokens_seen": 32722240,
      "step": 155050
    },
    {
      "epoch": 17.057755775577558,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.001934819559672447,
      "loss": 0.2314,
      "num_input_tokens_seen": 32723296,
      "step": 155055
    },
    {
      "epoch": 17.05830583058306,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.0019341121830892016,
      "loss": 0.2314,
      "num_input_tokens_seen": 32724352,
      "step": 155060
    },
    {
      "epoch": 17.05885588558856,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0019334049269279768,
      "loss": 0.2314,
      "num_input_tokens_seen": 32725376,
      "step": 155065
    },
    {
      "epoch": 17.059405940594058,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.0019326977911952841,
      "loss": 0.2319,
      "num_input_tokens_seen": 32726496,
      "step": 155070
    },
    {
      "epoch": 17.05995599559956,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0019319907758976467,
      "loss": 0.2335,
      "num_input_tokens_seen": 32727584,
      "step": 155075
    },
    {
      "epoch": 17.06050605060506,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0019312838810415843,
      "loss": 0.2324,
      "num_input_tokens_seen": 32728608,
      "step": 155080
    },
    {
      "epoch": 17.06105610561056,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0019305771066336018,
      "loss": 0.2298,
      "num_input_tokens_seen": 32729632,
      "step": 155085
    },
    {
      "epoch": 17.061606160616062,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.001929870452680224,
      "loss": 0.2324,
      "num_input_tokens_seen": 32730720,
      "step": 155090
    },
    {
      "epoch": 17.062156215621563,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0019291639191879538,
      "loss": 0.2329,
      "num_input_tokens_seen": 32731776,
      "step": 155095
    },
    {
      "epoch": 17.062706270627064,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.001928457506163308,
      "loss": 0.2303,
      "num_input_tokens_seen": 32732864,
      "step": 155100
    },
    {
      "epoch": 17.063256325632562,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.001927751213612801,
      "loss": 0.2303,
      "num_input_tokens_seen": 32733952,
      "step": 155105
    },
    {
      "epoch": 17.063806380638063,
      "grad_norm": 0.001953125,
      "learning_rate": 0.0019270450415429345,
      "loss": 0.2324,
      "num_input_tokens_seen": 32735008,
      "step": 155110
    },
    {
      "epoch": 17.064356435643564,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.001926338989960225,
      "loss": 0.2309,
      "num_input_tokens_seen": 32736064,
      "step": 155115
    },
    {
      "epoch": 17.064906490649065,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0019256330588711722,
      "loss": 0.2309,
      "num_input_tokens_seen": 32737120,
      "step": 155120
    },
    {
      "epoch": 17.065456545654566,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.0019249272482822876,
      "loss": 0.2308,
      "num_input_tokens_seen": 32738208,
      "step": 155125
    },
    {
      "epoch": 17.066006600660067,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0019242215582000726,
      "loss": 0.2298,
      "num_input_tokens_seen": 32739264,
      "step": 155130
    },
    {
      "epoch": 17.066556655665565,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0019235159886310338,
      "loss": 0.2308,
      "num_input_tokens_seen": 32740256,
      "step": 155135
    },
    {
      "epoch": 17.067106710671066,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0019228105395816758,
      "loss": 0.2324,
      "num_input_tokens_seen": 32741344,
      "step": 155140
    },
    {
      "epoch": 17.067656765676567,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.0019221052110584935,
      "loss": 0.2314,
      "num_input_tokens_seen": 32742464,
      "step": 155145
    },
    {
      "epoch": 17.068206820682068,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0019214000030679966,
      "loss": 0.2314,
      "num_input_tokens_seen": 32743488,
      "step": 155150
    },
    {
      "epoch": 17.06875687568757,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0019206949156166768,
      "loss": 0.2309,
      "num_input_tokens_seen": 32744544,
      "step": 155155
    },
    {
      "epoch": 17.06930693069307,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0019199899487110355,
      "loss": 0.2303,
      "num_input_tokens_seen": 32745632,
      "step": 155160
    },
    {
      "epoch": 17.06985698569857,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0019192851023575757,
      "loss": 0.2308,
      "num_input_tokens_seen": 32746656,
      "step": 155165
    },
    {
      "epoch": 17.07040704070407,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0019185803765627823,
      "loss": 0.2324,
      "num_input_tokens_seen": 32747712,
      "step": 155170
    },
    {
      "epoch": 17.07095709570957,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0019178757713331617,
      "loss": 0.2319,
      "num_input_tokens_seen": 32748768,
      "step": 155175
    },
    {
      "epoch": 17.07150715071507,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001917171286675202,
      "loss": 0.2308,
      "num_input_tokens_seen": 32749760,
      "step": 155180
    },
    {
      "epoch": 17.072057205720572,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0019164669225953934,
      "loss": 0.2319,
      "num_input_tokens_seen": 32750848,
      "step": 155185
    },
    {
      "epoch": 17.072607260726073,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0019157626791002319,
      "loss": 0.2298,
      "num_input_tokens_seen": 32751904,
      "step": 155190
    },
    {
      "epoch": 17.073157315731574,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.001915058556196209,
      "loss": 0.2298,
      "num_input_tokens_seen": 32752896,
      "step": 155195
    },
    {
      "epoch": 17.073707370737075,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.00191435455388981,
      "loss": 0.2324,
      "num_input_tokens_seen": 32753984,
      "step": 155200
    },
    {
      "epoch": 17.074257425742573,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0019136506721875307,
      "loss": 0.2324,
      "num_input_tokens_seen": 32755040,
      "step": 155205
    },
    {
      "epoch": 17.074807480748074,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0019129469110958479,
      "loss": 0.2319,
      "num_input_tokens_seen": 32756096,
      "step": 155210
    },
    {
      "epoch": 17.075357535753575,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.0019122432706212566,
      "loss": 0.2304,
      "num_input_tokens_seen": 32757088,
      "step": 155215
    },
    {
      "epoch": 17.075907590759076,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.0019115397507702397,
      "loss": 0.2309,
      "num_input_tokens_seen": 32758144,
      "step": 155220
    },
    {
      "epoch": 17.076457645764577,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0019108363515492788,
      "loss": 0.2298,
      "num_input_tokens_seen": 32759168,
      "step": 155225
    },
    {
      "epoch": 17.07700770077008,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0019101330729648601,
      "loss": 0.2329,
      "num_input_tokens_seen": 32760224,
      "step": 155230
    },
    {
      "epoch": 17.077557755775576,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0019094299150234622,
      "loss": 0.2329,
      "num_input_tokens_seen": 32761248,
      "step": 155235
    },
    {
      "epoch": 17.078107810781077,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0019087268777315696,
      "loss": 0.2303,
      "num_input_tokens_seen": 32762272,
      "step": 155240
    },
    {
      "epoch": 17.078657865786578,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0019080239610956573,
      "loss": 0.2293,
      "num_input_tokens_seen": 32763328,
      "step": 155245
    },
    {
      "epoch": 17.07920792079208,
      "grad_norm": 0.001953125,
      "learning_rate": 0.001907321165122205,
      "loss": 0.2309,
      "num_input_tokens_seen": 32764384,
      "step": 155250
    },
    {
      "epoch": 17.07975797579758,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0019066184898176941,
      "loss": 0.2329,
      "num_input_tokens_seen": 32765440,
      "step": 155255
    },
    {
      "epoch": 17.08030803080308,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0019059159351885945,
      "loss": 0.2314,
      "num_input_tokens_seen": 32766432,
      "step": 155260
    },
    {
      "epoch": 17.080858085808583,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.0019052135012413895,
      "loss": 0.2319,
      "num_input_tokens_seen": 32767488,
      "step": 155265
    },
    {
      "epoch": 17.08140814081408,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0019045111879825438,
      "loss": 0.2314,
      "num_input_tokens_seen": 32768448,
      "step": 155270
    },
    {
      "epoch": 17.08195819581958,
      "grad_norm": 0.0004100799560546875,
      "learning_rate": 0.0019038089954185337,
      "loss": 0.2303,
      "num_input_tokens_seen": 32769472,
      "step": 155275
    },
    {
      "epoch": 17.082508250825082,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.001903106923555836,
      "loss": 0.2314,
      "num_input_tokens_seen": 32770528,
      "step": 155280
    },
    {
      "epoch": 17.083058305830583,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.001902404972400914,
      "loss": 0.2314,
      "num_input_tokens_seen": 32771584,
      "step": 155285
    },
    {
      "epoch": 17.083608360836084,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.0019017031419602437,
      "loss": 0.2324,
      "num_input_tokens_seen": 32772672,
      "step": 155290
    },
    {
      "epoch": 17.084158415841586,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.00190100143224029,
      "loss": 0.2324,
      "num_input_tokens_seen": 32773728,
      "step": 155295
    },
    {
      "epoch": 17.084708470847083,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001900299843247518,
      "loss": 0.2314,
      "num_input_tokens_seen": 32774752,
      "step": 155300
    },
    {
      "epoch": 17.085258525852584,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0018995983749883977,
      "loss": 0.2293,
      "num_input_tokens_seen": 32775840,
      "step": 155305
    },
    {
      "epoch": 17.085808580858085,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0018988970274693916,
      "loss": 0.2319,
      "num_input_tokens_seen": 32776896,
      "step": 155310
    },
    {
      "epoch": 17.086358635863586,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0018981958006969685,
      "loss": 0.2303,
      "num_input_tokens_seen": 32777984,
      "step": 155315
    },
    {
      "epoch": 17.086908690869087,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.001897494694677586,
      "loss": 0.2314,
      "num_input_tokens_seen": 32779072,
      "step": 155320
    },
    {
      "epoch": 17.08745874587459,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0018967937094177079,
      "loss": 0.2319,
      "num_input_tokens_seen": 32780160,
      "step": 155325
    },
    {
      "epoch": 17.08800880088009,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0018960928449237918,
      "loss": 0.2314,
      "num_input_tokens_seen": 32781120,
      "step": 155330
    },
    {
      "epoch": 17.088558855885587,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.0018953921012023012,
      "loss": 0.2319,
      "num_input_tokens_seen": 32782208,
      "step": 155335
    },
    {
      "epoch": 17.08910891089109,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0018946914782596972,
      "loss": 0.2329,
      "num_input_tokens_seen": 32783200,
      "step": 155340
    },
    {
      "epoch": 17.08965896589659,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.001893990976102432,
      "loss": 0.2303,
      "num_input_tokens_seen": 32784288,
      "step": 155345
    },
    {
      "epoch": 17.09020902090209,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.0018932905947369614,
      "loss": 0.2293,
      "num_input_tokens_seen": 32785408,
      "step": 155350
    },
    {
      "epoch": 17.09075907590759,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0018925903341697442,
      "loss": 0.2309,
      "num_input_tokens_seen": 32786528,
      "step": 155355
    },
    {
      "epoch": 17.091309130913093,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0018918901944072297,
      "loss": 0.2314,
      "num_input_tokens_seen": 32787616,
      "step": 155360
    },
    {
      "epoch": 17.09185918591859,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.001891190175455873,
      "loss": 0.2314,
      "num_input_tokens_seen": 32788608,
      "step": 155365
    },
    {
      "epoch": 17.09240924092409,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0018904902773221304,
      "loss": 0.2324,
      "num_input_tokens_seen": 32789632,
      "step": 155370
    },
    {
      "epoch": 17.092959295929592,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.0018897905000124437,
      "loss": 0.2309,
      "num_input_tokens_seen": 32790720,
      "step": 155375
    },
    {
      "epoch": 17.093509350935093,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.001889090843533271,
      "loss": 0.2309,
      "num_input_tokens_seen": 32791808,
      "step": 155380
    },
    {
      "epoch": 17.094059405940595,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0018883913078910535,
      "loss": 0.2303,
      "num_input_tokens_seen": 32792864,
      "step": 155385
    },
    {
      "epoch": 17.094609460946096,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0018876918930922414,
      "loss": 0.2324,
      "num_input_tokens_seen": 32793888,
      "step": 155390
    },
    {
      "epoch": 17.095159515951597,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.001886992599143286,
      "loss": 0.2298,
      "num_input_tokens_seen": 32794880,
      "step": 155395
    },
    {
      "epoch": 17.095709570957094,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0018862934260506226,
      "loss": 0.2298,
      "num_input_tokens_seen": 32795840,
      "step": 155400
    },
    {
      "epoch": 17.096259625962595,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0018855943738207035,
      "loss": 0.2314,
      "num_input_tokens_seen": 32796960,
      "step": 155405
    },
    {
      "epoch": 17.096809680968097,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0018848954424599694,
      "loss": 0.2309,
      "num_input_tokens_seen": 32798080,
      "step": 155410
    },
    {
      "epoch": 17.097359735973598,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0018841966319748564,
      "loss": 0.2329,
      "num_input_tokens_seen": 32799168,
      "step": 155415
    },
    {
      "epoch": 17.0979097909791,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0018834979423718095,
      "loss": 0.2314,
      "num_input_tokens_seen": 32800192,
      "step": 155420
    },
    {
      "epoch": 17.0984598459846,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0018827993736572684,
      "loss": 0.2308,
      "num_input_tokens_seen": 32801184,
      "step": 155425
    },
    {
      "epoch": 17.099009900990097,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.0018821009258376746,
      "loss": 0.2324,
      "num_input_tokens_seen": 32802208,
      "step": 155430
    },
    {
      "epoch": 17.0995599559956,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0018814025989194632,
      "loss": 0.2324,
      "num_input_tokens_seen": 32803232,
      "step": 155435
    },
    {
      "epoch": 17.1001100110011,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0018807043929090638,
      "loss": 0.2319,
      "num_input_tokens_seen": 32804320,
      "step": 155440
    },
    {
      "epoch": 17.1006600660066,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0018800063078129215,
      "loss": 0.2313,
      "num_input_tokens_seen": 32805408,
      "step": 155445
    },
    {
      "epoch": 17.1012101210121,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.0018793083436374574,
      "loss": 0.2319,
      "num_input_tokens_seen": 32806432,
      "step": 155450
    },
    {
      "epoch": 17.101760176017603,
      "grad_norm": 0.0006103515625,
      "learning_rate": 0.00187861050038912,
      "loss": 0.2308,
      "num_input_tokens_seen": 32807424,
      "step": 155455
    },
    {
      "epoch": 17.102310231023104,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0018779127780743342,
      "loss": 0.2303,
      "num_input_tokens_seen": 32808448,
      "step": 155460
    },
    {
      "epoch": 17.1028602860286,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.0018772151766995248,
      "loss": 0.2303,
      "num_input_tokens_seen": 32809536,
      "step": 155465
    },
    {
      "epoch": 17.103410341034103,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0018765176962711298,
      "loss": 0.2319,
      "num_input_tokens_seen": 32810528,
      "step": 155470
    },
    {
      "epoch": 17.103960396039604,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.001875820336795571,
      "loss": 0.2329,
      "num_input_tokens_seen": 32811616,
      "step": 155475
    },
    {
      "epoch": 17.104510451045105,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.0018751230982792782,
      "loss": 0.2319,
      "num_input_tokens_seen": 32812672,
      "step": 155480
    },
    {
      "epoch": 17.105060506050606,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.001874425980728681,
      "loss": 0.2329,
      "num_input_tokens_seen": 32813760,
      "step": 155485
    },
    {
      "epoch": 17.105610561056107,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0018737289841501996,
      "loss": 0.2314,
      "num_input_tokens_seen": 32814784,
      "step": 155490
    },
    {
      "epoch": 17.106160616061604,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001873032108550262,
      "loss": 0.2319,
      "num_input_tokens_seen": 32815968,
      "step": 155495
    },
    {
      "epoch": 17.106710671067106,
      "grad_norm": 0.000823974609375,
      "learning_rate": 0.0018723353539352849,
      "loss": 0.2314,
      "num_input_tokens_seen": 32816992,
      "step": 155500
    },
    {
      "epoch": 17.107260726072607,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.0018716387203116963,
      "loss": 0.2293,
      "num_input_tokens_seen": 32818080,
      "step": 155505
    },
    {
      "epoch": 17.107810781078108,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.001870942207685911,
      "loss": 0.2314,
      "num_input_tokens_seen": 32819136,
      "step": 155510
    },
    {
      "epoch": 17.10836083608361,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0018702458160643526,
      "loss": 0.2308,
      "num_input_tokens_seen": 32820160,
      "step": 155515
    },
    {
      "epoch": 17.10891089108911,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.0018695495454534405,
      "loss": 0.2308,
      "num_input_tokens_seen": 32821216,
      "step": 155520
    },
    {
      "epoch": 17.10946094609461,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0018688533958595864,
      "loss": 0.2314,
      "num_input_tokens_seen": 32822272,
      "step": 155525
    },
    {
      "epoch": 17.11001100110011,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0018681573672892136,
      "loss": 0.2298,
      "num_input_tokens_seen": 32823296,
      "step": 155530
    },
    {
      "epoch": 17.11056105610561,
      "grad_norm": 0.00049591064453125,
      "learning_rate": 0.0018674614597487286,
      "loss": 0.2303,
      "num_input_tokens_seen": 32824384,
      "step": 155535
    },
    {
      "epoch": 17.11111111111111,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0018667656732445498,
      "loss": 0.2314,
      "num_input_tokens_seen": 32825440,
      "step": 155540
    },
    {
      "epoch": 17.111661166116612,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.001866070007783095,
      "loss": 0.2335,
      "num_input_tokens_seen": 32826496,
      "step": 155545
    },
    {
      "epoch": 17.112211221122113,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0018653744633707657,
      "loss": 0.2324,
      "num_input_tokens_seen": 32827584,
      "step": 155550
    },
    {
      "epoch": 17.112761276127614,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0018646790400139806,
      "loss": 0.2324,
      "num_input_tokens_seen": 32828704,
      "step": 155555
    },
    {
      "epoch": 17.11331133113311,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0018639837377191458,
      "loss": 0.2335,
      "num_input_tokens_seen": 32829728,
      "step": 155560
    },
    {
      "epoch": 17.113861386138613,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0018632885564926647,
      "loss": 0.2308,
      "num_input_tokens_seen": 32830752,
      "step": 155565
    },
    {
      "epoch": 17.114411441144114,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.001862593496340954,
      "loss": 0.2329,
      "num_input_tokens_seen": 32831840,
      "step": 155570
    },
    {
      "epoch": 17.114961496149615,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0018618985572704133,
      "loss": 0.2303,
      "num_input_tokens_seen": 32832960,
      "step": 155575
    },
    {
      "epoch": 17.115511551155116,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0018612037392874508,
      "loss": 0.2329,
      "num_input_tokens_seen": 32833952,
      "step": 155580
    },
    {
      "epoch": 17.116061606160617,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00186050904239847,
      "loss": 0.233,
      "num_input_tokens_seen": 32834976,
      "step": 155585
    },
    {
      "epoch": 17.116611661166118,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0018598144666098686,
      "loss": 0.2314,
      "num_input_tokens_seen": 32836064,
      "step": 155590
    },
    {
      "epoch": 17.117161716171616,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.0018591200119280537,
      "loss": 0.2319,
      "num_input_tokens_seen": 32837152,
      "step": 155595
    },
    {
      "epoch": 17.117711771177117,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.001858425678359425,
      "loss": 0.2288,
      "num_input_tokens_seen": 32838272,
      "step": 155600
    },
    {
      "epoch": 17.118261826182618,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0018577314659103787,
      "loss": 0.2303,
      "num_input_tokens_seen": 32839232,
      "step": 155605
    },
    {
      "epoch": 17.11881188118812,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0018570373745873186,
      "loss": 0.2329,
      "num_input_tokens_seen": 32840256,
      "step": 155610
    },
    {
      "epoch": 17.11936193619362,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.0018563434043966359,
      "loss": 0.2319,
      "num_input_tokens_seen": 32841344,
      "step": 155615
    },
    {
      "epoch": 17.11991199119912,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.0018556495553447305,
      "loss": 0.2314,
      "num_input_tokens_seen": 32842368,
      "step": 155620
    },
    {
      "epoch": 17.120462046204622,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.001854955827437994,
      "loss": 0.2319,
      "num_input_tokens_seen": 32843392,
      "step": 155625
    },
    {
      "epoch": 17.12101210121012,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0018542622206828212,
      "loss": 0.2329,
      "num_input_tokens_seen": 32844416,
      "step": 155630
    },
    {
      "epoch": 17.12156215621562,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0018535687350856088,
      "loss": 0.2314,
      "num_input_tokens_seen": 32845536,
      "step": 155635
    },
    {
      "epoch": 17.122112211221122,
      "grad_norm": 0.00080108642578125,
      "learning_rate": 0.0018528753706527434,
      "loss": 0.2314,
      "num_input_tokens_seen": 32846592,
      "step": 155640
    },
    {
      "epoch": 17.122662266226623,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.0018521821273906197,
      "loss": 0.2314,
      "num_input_tokens_seen": 32847648,
      "step": 155645
    },
    {
      "epoch": 17.123212321232124,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0018514890053056192,
      "loss": 0.2319,
      "num_input_tokens_seen": 32848672,
      "step": 155650
    },
    {
      "epoch": 17.123762376237625,
      "grad_norm": 0.000812530517578125,
      "learning_rate": 0.001850796004404137,
      "loss": 0.2314,
      "num_input_tokens_seen": 32849696,
      "step": 155655
    },
    {
      "epoch": 17.124312431243123,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0018501031246925615,
      "loss": 0.2309,
      "num_input_tokens_seen": 32850720,
      "step": 155660
    },
    {
      "epoch": 17.124862486248624,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.001849410366177272,
      "loss": 0.2309,
      "num_input_tokens_seen": 32851744,
      "step": 155665
    },
    {
      "epoch": 17.125412541254125,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.0018487177288646604,
      "loss": 0.2303,
      "num_input_tokens_seen": 32852768,
      "step": 155670
    },
    {
      "epoch": 17.125962596259626,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0018480252127611068,
      "loss": 0.2319,
      "num_input_tokens_seen": 32853760,
      "step": 155675
    },
    {
      "epoch": 17.126512651265127,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.0018473328178729908,
      "loss": 0.2308,
      "num_input_tokens_seen": 32854848,
      "step": 155680
    },
    {
      "epoch": 17.127062706270628,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0018466405442066974,
      "loss": 0.2319,
      "num_input_tokens_seen": 32855936,
      "step": 155685
    },
    {
      "epoch": 17.12761276127613,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0018459483917686063,
      "loss": 0.2314,
      "num_input_tokens_seen": 32857056,
      "step": 155690
    },
    {
      "epoch": 17.128162816281627,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0018452563605650995,
      "loss": 0.2314,
      "num_input_tokens_seen": 32858016,
      "step": 155695
    },
    {
      "epoch": 17.128712871287128,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.0018445644506025532,
      "loss": 0.2319,
      "num_input_tokens_seen": 32859008,
      "step": 155700
    },
    {
      "epoch": 17.12926292629263,
      "grad_norm": 0.00061798095703125,
      "learning_rate": 0.001843872661887339,
      "loss": 0.2324,
      "num_input_tokens_seen": 32860064,
      "step": 155705
    },
    {
      "epoch": 17.12981298129813,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0018431809944258403,
      "loss": 0.2324,
      "num_input_tokens_seen": 32861120,
      "step": 155710
    },
    {
      "epoch": 17.13036303630363,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0018424894482244269,
      "loss": 0.2319,
      "num_input_tokens_seen": 32862112,
      "step": 155715
    },
    {
      "epoch": 17.130913091309132,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.0018417980232894786,
      "loss": 0.2314,
      "num_input_tokens_seen": 32863168,
      "step": 155720
    },
    {
      "epoch": 17.13146314631463,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.001841106719627364,
      "loss": 0.2293,
      "num_input_tokens_seen": 32864224,
      "step": 155725
    },
    {
      "epoch": 17.13201320132013,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0018404155372444507,
      "loss": 0.2324,
      "num_input_tokens_seen": 32865312,
      "step": 155730
    },
    {
      "epoch": 17.132563256325632,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0018397244761471177,
      "loss": 0.2298,
      "num_input_tokens_seen": 32866368,
      "step": 155735
    },
    {
      "epoch": 17.133113311331133,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0018390335363417242,
      "loss": 0.2303,
      "num_input_tokens_seen": 32867456,
      "step": 155740
    },
    {
      "epoch": 17.133663366336634,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.001838342717834644,
      "loss": 0.2329,
      "num_input_tokens_seen": 32868480,
      "step": 155745
    },
    {
      "epoch": 17.134213421342135,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0018376520206322465,
      "loss": 0.2319,
      "num_input_tokens_seen": 32869632,
      "step": 155750
    },
    {
      "epoch": 17.134763476347636,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0018369614447408921,
      "loss": 0.2319,
      "num_input_tokens_seen": 32870720,
      "step": 155755
    },
    {
      "epoch": 17.135313531353134,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0018362709901669505,
      "loss": 0.2293,
      "num_input_tokens_seen": 32871776,
      "step": 155760
    },
    {
      "epoch": 17.135863586358635,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0018355806569167797,
      "loss": 0.2314,
      "num_input_tokens_seen": 32872768,
      "step": 155765
    },
    {
      "epoch": 17.136413641364136,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.001834890444996745,
      "loss": 0.2298,
      "num_input_tokens_seen": 32873856,
      "step": 155770
    },
    {
      "epoch": 17.136963696369637,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.0018342003544132095,
      "loss": 0.2319,
      "num_input_tokens_seen": 32874944,
      "step": 155775
    },
    {
      "epoch": 17.13751375137514,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0018335103851725297,
      "loss": 0.2308,
      "num_input_tokens_seen": 32876000,
      "step": 155780
    },
    {
      "epoch": 17.13806380638064,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0018328205372810707,
      "loss": 0.2308,
      "num_input_tokens_seen": 32876992,
      "step": 155785
    },
    {
      "epoch": 17.138613861386137,
      "grad_norm": 0.00077056884765625,
      "learning_rate": 0.0018321308107451856,
      "loss": 0.2319,
      "num_input_tokens_seen": 32878016,
      "step": 155790
    },
    {
      "epoch": 17.139163916391638,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0018314412055712292,
      "loss": 0.2319,
      "num_input_tokens_seen": 32879008,
      "step": 155795
    },
    {
      "epoch": 17.13971397139714,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0018307517217655616,
      "loss": 0.2314,
      "num_input_tokens_seen": 32880064,
      "step": 155800
    },
    {
      "epoch": 17.14026402640264,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0018300623593345343,
      "loss": 0.2324,
      "num_input_tokens_seen": 32881056,
      "step": 155805
    },
    {
      "epoch": 17.14081408140814,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.0018293731182845074,
      "loss": 0.2314,
      "num_input_tokens_seen": 32882080,
      "step": 155810
    },
    {
      "epoch": 17.141364136413642,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0018286839986218272,
      "loss": 0.2303,
      "num_input_tokens_seen": 32883104,
      "step": 155815
    },
    {
      "epoch": 17.141914191419144,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0018279950003528438,
      "loss": 0.2329,
      "num_input_tokens_seen": 32884192,
      "step": 155820
    },
    {
      "epoch": 17.14246424642464,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0018273061234839105,
      "loss": 0.2314,
      "num_input_tokens_seen": 32885216,
      "step": 155825
    },
    {
      "epoch": 17.143014301430142,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.0018266173680213753,
      "loss": 0.2319,
      "num_input_tokens_seen": 32886336,
      "step": 155830
    },
    {
      "epoch": 17.143564356435643,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0018259287339715902,
      "loss": 0.2329,
      "num_input_tokens_seen": 32887424,
      "step": 155835
    },
    {
      "epoch": 17.144114411441144,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.001825240221340898,
      "loss": 0.2314,
      "num_input_tokens_seen": 32888416,
      "step": 155840
    },
    {
      "epoch": 17.144664466446645,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.0018245518301356406,
      "loss": 0.2319,
      "num_input_tokens_seen": 32889408,
      "step": 155845
    },
    {
      "epoch": 17.145214521452147,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.001823863560362171,
      "loss": 0.2303,
      "num_input_tokens_seen": 32890464,
      "step": 155850
    },
    {
      "epoch": 17.145764576457644,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001823175412026826,
      "loss": 0.2314,
      "num_input_tokens_seen": 32891488,
      "step": 155855
    },
    {
      "epoch": 17.146314631463145,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.001822487385135949,
      "loss": 0.2308,
      "num_input_tokens_seen": 32892544,
      "step": 155860
    },
    {
      "epoch": 17.146864686468646,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0018217994796958863,
      "loss": 0.2303,
      "num_input_tokens_seen": 32893632,
      "step": 155865
    },
    {
      "epoch": 17.147414741474147,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0018211116957129712,
      "loss": 0.233,
      "num_input_tokens_seen": 32894656,
      "step": 155870
    },
    {
      "epoch": 17.14796479647965,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0018204240331935488,
      "loss": 0.2308,
      "num_input_tokens_seen": 32895712,
      "step": 155875
    },
    {
      "epoch": 17.14851485148515,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0018197364921439506,
      "loss": 0.2324,
      "num_input_tokens_seen": 32896736,
      "step": 155880
    },
    {
      "epoch": 17.14906490649065,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0018190490725705183,
      "loss": 0.2314,
      "num_input_tokens_seen": 32897760,
      "step": 155885
    },
    {
      "epoch": 17.149614961496148,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0018183617744795865,
      "loss": 0.2308,
      "num_input_tokens_seen": 32898784,
      "step": 155890
    },
    {
      "epoch": 17.15016501650165,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0018176745978774871,
      "loss": 0.2319,
      "num_input_tokens_seen": 32899808,
      "step": 155895
    },
    {
      "epoch": 17.15071507150715,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0018169875427705583,
      "loss": 0.2314,
      "num_input_tokens_seen": 32900832,
      "step": 155900
    },
    {
      "epoch": 17.15126512651265,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0018163006091651268,
      "loss": 0.2309,
      "num_input_tokens_seen": 32901888,
      "step": 155905
    },
    {
      "epoch": 17.151815181518153,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.001815613797067529,
      "loss": 0.2308,
      "num_input_tokens_seen": 32902912,
      "step": 155910
    },
    {
      "epoch": 17.152365236523654,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.0018149271064840898,
      "loss": 0.2298,
      "num_input_tokens_seen": 32904000,
      "step": 155915
    },
    {
      "epoch": 17.15291529152915,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0018142405374211412,
      "loss": 0.2314,
      "num_input_tokens_seen": 32905056,
      "step": 155920
    },
    {
      "epoch": 17.153465346534652,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0018135540898850126,
      "loss": 0.2298,
      "num_input_tokens_seen": 32906112,
      "step": 155925
    },
    {
      "epoch": 17.154015401540153,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0018128677638820244,
      "loss": 0.2298,
      "num_input_tokens_seen": 32907104,
      "step": 155930
    },
    {
      "epoch": 17.154565456545654,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0018121815594185115,
      "loss": 0.2309,
      "num_input_tokens_seen": 32908192,
      "step": 155935
    },
    {
      "epoch": 17.155115511551156,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0018114954765007885,
      "loss": 0.2303,
      "num_input_tokens_seen": 32909248,
      "step": 155940
    },
    {
      "epoch": 17.155665566556657,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0018108095151351837,
      "loss": 0.2314,
      "num_input_tokens_seen": 32910304,
      "step": 155945
    },
    {
      "epoch": 17.156215621562158,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0018101236753280225,
      "loss": 0.2308,
      "num_input_tokens_seen": 32911360,
      "step": 155950
    },
    {
      "epoch": 17.156765676567655,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0018094379570856177,
      "loss": 0.235,
      "num_input_tokens_seen": 32912416,
      "step": 155955
    },
    {
      "epoch": 17.157315731573156,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0018087523604142979,
      "loss": 0.2309,
      "num_input_tokens_seen": 32913440,
      "step": 155960
    },
    {
      "epoch": 17.157865786578657,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0018080668853203779,
      "loss": 0.2319,
      "num_input_tokens_seen": 32914432,
      "step": 155965
    },
    {
      "epoch": 17.15841584158416,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.001807381531810171,
      "loss": 0.2303,
      "num_input_tokens_seen": 32915488,
      "step": 155970
    },
    {
      "epoch": 17.15896589658966,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.0018066962998899987,
      "loss": 0.2314,
      "num_input_tokens_seen": 32916512,
      "step": 155975
    },
    {
      "epoch": 17.15951595159516,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0018060111895661795,
      "loss": 0.2293,
      "num_input_tokens_seen": 32917568,
      "step": 155980
    },
    {
      "epoch": 17.16006600660066,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0018053262008450215,
      "loss": 0.2303,
      "num_input_tokens_seen": 32918592,
      "step": 155985
    },
    {
      "epoch": 17.16061606160616,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.001804641333732843,
      "loss": 0.2309,
      "num_input_tokens_seen": 32919744,
      "step": 155990
    },
    {
      "epoch": 17.16116611661166,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0018039565882359493,
      "loss": 0.2309,
      "num_input_tokens_seen": 32920832,
      "step": 155995
    },
    {
      "epoch": 17.16171617161716,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0018032719643606564,
      "loss": 0.2324,
      "num_input_tokens_seen": 32921888,
      "step": 156000
    },
    {
      "epoch": 17.162266226622663,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.001802587462113278,
      "loss": 0.2308,
      "num_input_tokens_seen": 32922976,
      "step": 156005
    },
    {
      "epoch": 17.162816281628164,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.001801903081500114,
      "loss": 0.2314,
      "num_input_tokens_seen": 32924064,
      "step": 156010
    },
    {
      "epoch": 17.163366336633665,
      "grad_norm": 0.000579833984375,
      "learning_rate": 0.0018012188225274793,
      "loss": 0.2314,
      "num_input_tokens_seen": 32925120,
      "step": 156015
    },
    {
      "epoch": 17.163916391639162,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.001800534685201674,
      "loss": 0.2303,
      "num_input_tokens_seen": 32926144,
      "step": 156020
    },
    {
      "epoch": 17.164466446644663,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.0017998506695290095,
      "loss": 0.2324,
      "num_input_tokens_seen": 32927200,
      "step": 156025
    },
    {
      "epoch": 17.165016501650165,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.001799166775515784,
      "loss": 0.2324,
      "num_input_tokens_seen": 32928256,
      "step": 156030
    },
    {
      "epoch": 17.165566556655666,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0017984830031683046,
      "loss": 0.2298,
      "num_input_tokens_seen": 32929312,
      "step": 156035
    },
    {
      "epoch": 17.166116611661167,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.001797799352492876,
      "loss": 0.2319,
      "num_input_tokens_seen": 32930304,
      "step": 156040
    },
    {
      "epoch": 17.166666666666668,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0017971158234957928,
      "loss": 0.2324,
      "num_input_tokens_seen": 32931360,
      "step": 156045
    },
    {
      "epoch": 17.16721672167217,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0017964324161833605,
      "loss": 0.2329,
      "num_input_tokens_seen": 32932448,
      "step": 156050
    },
    {
      "epoch": 17.167766776677666,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0017957491305618722,
      "loss": 0.2319,
      "num_input_tokens_seen": 32933536,
      "step": 156055
    },
    {
      "epoch": 17.168316831683168,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0017950659666376278,
      "loss": 0.2314,
      "num_input_tokens_seen": 32934624,
      "step": 156060
    },
    {
      "epoch": 17.16886688668867,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.0017943829244169273,
      "loss": 0.2319,
      "num_input_tokens_seen": 32935712,
      "step": 156065
    },
    {
      "epoch": 17.16941694169417,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0017937000039060591,
      "loss": 0.2303,
      "num_input_tokens_seen": 32936768,
      "step": 156070
    },
    {
      "epoch": 17.16996699669967,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.0017930172051113247,
      "loss": 0.2324,
      "num_input_tokens_seen": 32937824,
      "step": 156075
    },
    {
      "epoch": 17.170517051705172,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.0017923345280390123,
      "loss": 0.2314,
      "num_input_tokens_seen": 32938816,
      "step": 156080
    },
    {
      "epoch": 17.17106710671067,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.001791651972695412,
      "loss": 0.2329,
      "num_input_tokens_seen": 32939904,
      "step": 156085
    },
    {
      "epoch": 17.17161716171617,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0017909695390868191,
      "loss": 0.2298,
      "num_input_tokens_seen": 32940992,
      "step": 156090
    },
    {
      "epoch": 17.17216721672167,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.0017902872272195197,
      "loss": 0.2298,
      "num_input_tokens_seen": 32942112,
      "step": 156095
    },
    {
      "epoch": 17.172717271727173,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.001789605037099809,
      "loss": 0.2314,
      "num_input_tokens_seen": 32943168,
      "step": 156100
    },
    {
      "epoch": 17.173267326732674,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0017889229687339686,
      "loss": 0.2319,
      "num_input_tokens_seen": 32944192,
      "step": 156105
    },
    {
      "epoch": 17.173817381738175,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0017882410221282835,
      "loss": 0.2319,
      "num_input_tokens_seen": 32945216,
      "step": 156110
    },
    {
      "epoch": 17.174367436743676,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0017875591972890403,
      "loss": 0.2324,
      "num_input_tokens_seen": 32946240,
      "step": 156115
    },
    {
      "epoch": 17.174917491749174,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0017868774942225256,
      "loss": 0.2288,
      "num_input_tokens_seen": 32947328,
      "step": 156120
    },
    {
      "epoch": 17.175467546754675,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.001786195912935023,
      "loss": 0.2293,
      "num_input_tokens_seen": 32948384,
      "step": 156125
    },
    {
      "epoch": 17.176017601760176,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0017855144534328121,
      "loss": 0.2293,
      "num_input_tokens_seen": 32949440,
      "step": 156130
    },
    {
      "epoch": 17.176567656765677,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.0017848331157221697,
      "loss": 0.2314,
      "num_input_tokens_seen": 32950432,
      "step": 156135
    },
    {
      "epoch": 17.177117711771178,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.001784151899809384,
      "loss": 0.2324,
      "num_input_tokens_seen": 32951488,
      "step": 156140
    },
    {
      "epoch": 17.17766776677668,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0017834708057007237,
      "loss": 0.2324,
      "num_input_tokens_seen": 32952512,
      "step": 156145
    },
    {
      "epoch": 17.178217821782177,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0017827898334024732,
      "loss": 0.2314,
      "num_input_tokens_seen": 32953504,
      "step": 156150
    },
    {
      "epoch": 17.178767876787678,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.001782108982920908,
      "loss": 0.2325,
      "num_input_tokens_seen": 32954592,
      "step": 156155
    },
    {
      "epoch": 17.17931793179318,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0017814282542622994,
      "loss": 0.233,
      "num_input_tokens_seen": 32955712,
      "step": 156160
    },
    {
      "epoch": 17.17986798679868,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.001780747647432926,
      "loss": 0.2314,
      "num_input_tokens_seen": 32956864,
      "step": 156165
    },
    {
      "epoch": 17.18041804180418,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.001780067162439059,
      "loss": 0.2314,
      "num_input_tokens_seen": 32957920,
      "step": 156170
    },
    {
      "epoch": 17.180968096809682,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.0017793867992869621,
      "loss": 0.2324,
      "num_input_tokens_seen": 32958944,
      "step": 156175
    },
    {
      "epoch": 17.181518151815183,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.0017787065579829202,
      "loss": 0.2314,
      "num_input_tokens_seen": 32960000,
      "step": 156180
    },
    {
      "epoch": 17.18206820682068,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0017780264385331933,
      "loss": 0.2314,
      "num_input_tokens_seen": 32961056,
      "step": 156185
    },
    {
      "epoch": 17.182618261826182,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0017773464409440547,
      "loss": 0.2324,
      "num_input_tokens_seen": 32962112,
      "step": 156190
    },
    {
      "epoch": 17.183168316831683,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.0017766665652217694,
      "loss": 0.2319,
      "num_input_tokens_seen": 32963104,
      "step": 156195
    },
    {
      "epoch": 17.183718371837184,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0017759868113726006,
      "loss": 0.2324,
      "num_input_tokens_seen": 32964160,
      "step": 156200
    },
    {
      "epoch": 17.184268426842685,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0017753071794028151,
      "loss": 0.2324,
      "num_input_tokens_seen": 32965152,
      "step": 156205
    },
    {
      "epoch": 17.184818481848186,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0017746276693186762,
      "loss": 0.2309,
      "num_input_tokens_seen": 32966208,
      "step": 156210
    },
    {
      "epoch": 17.185368536853684,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.0017739482811264522,
      "loss": 0.2314,
      "num_input_tokens_seen": 32967232,
      "step": 156215
    },
    {
      "epoch": 17.185918591859185,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0017732690148324014,
      "loss": 0.2324,
      "num_input_tokens_seen": 32968256,
      "step": 156220
    },
    {
      "epoch": 17.186468646864686,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0017725898704427788,
      "loss": 0.2314,
      "num_input_tokens_seen": 32969248,
      "step": 156225
    },
    {
      "epoch": 17.187018701870187,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0017719108479638528,
      "loss": 0.2319,
      "num_input_tokens_seen": 32970336,
      "step": 156230
    },
    {
      "epoch": 17.187568756875688,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00177123194740187,
      "loss": 0.2308,
      "num_input_tokens_seen": 32971328,
      "step": 156235
    },
    {
      "epoch": 17.18811881188119,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0017705531687631003,
      "loss": 0.2324,
      "num_input_tokens_seen": 32972416,
      "step": 156240
    },
    {
      "epoch": 17.18866886688669,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.001769874512053794,
      "loss": 0.2303,
      "num_input_tokens_seen": 32973472,
      "step": 156245
    },
    {
      "epoch": 17.189218921892188,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.001769195977280204,
      "loss": 0.2314,
      "num_input_tokens_seen": 32974528,
      "step": 156250
    },
    {
      "epoch": 17.18976897689769,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0017685175644485873,
      "loss": 0.2319,
      "num_input_tokens_seen": 32975584,
      "step": 156255
    },
    {
      "epoch": 17.19031903190319,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0017678392735651937,
      "loss": 0.2324,
      "num_input_tokens_seen": 32976608,
      "step": 156260
    },
    {
      "epoch": 17.19086908690869,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.001767161104636275,
      "loss": 0.2319,
      "num_input_tokens_seen": 32977664,
      "step": 156265
    },
    {
      "epoch": 17.191419141914192,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0017664830576680861,
      "loss": 0.2319,
      "num_input_tokens_seen": 32978720,
      "step": 156270
    },
    {
      "epoch": 17.191969196919693,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0017658051326668689,
      "loss": 0.2334,
      "num_input_tokens_seen": 32979808,
      "step": 156275
    },
    {
      "epoch": 17.19251925192519,
      "grad_norm": 0.00061798095703125,
      "learning_rate": 0.0017651273296388796,
      "loss": 0.2324,
      "num_input_tokens_seen": 32980832,
      "step": 156280
    },
    {
      "epoch": 17.193069306930692,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0017644496485903587,
      "loss": 0.2314,
      "num_input_tokens_seen": 32981888,
      "step": 156285
    },
    {
      "epoch": 17.193619361936193,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.001763772089527556,
      "loss": 0.2309,
      "num_input_tokens_seen": 32982944,
      "step": 156290
    },
    {
      "epoch": 17.194169416941694,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0017630946524567115,
      "loss": 0.2314,
      "num_input_tokens_seen": 32984064,
      "step": 156295
    },
    {
      "epoch": 17.194719471947195,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0017624173373840717,
      "loss": 0.2324,
      "num_input_tokens_seen": 32985152,
      "step": 156300
    },
    {
      "epoch": 17.195269526952696,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0017617401443158836,
      "loss": 0.2303,
      "num_input_tokens_seen": 32986176,
      "step": 156305
    },
    {
      "epoch": 17.195819581958197,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.0017610630732583803,
      "loss": 0.2319,
      "num_input_tokens_seen": 32987264,
      "step": 156310
    },
    {
      "epoch": 17.196369636963695,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0017603861242178086,
      "loss": 0.2324,
      "num_input_tokens_seen": 32988320,
      "step": 156315
    },
    {
      "epoch": 17.196919691969196,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0017597092972004035,
      "loss": 0.2303,
      "num_input_tokens_seen": 32989312,
      "step": 156320
    },
    {
      "epoch": 17.197469746974697,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0017590325922124032,
      "loss": 0.2324,
      "num_input_tokens_seen": 32990432,
      "step": 156325
    },
    {
      "epoch": 17.198019801980198,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.0017583560092600497,
      "loss": 0.2329,
      "num_input_tokens_seen": 32991488,
      "step": 156330
    },
    {
      "epoch": 17.1985698569857,
      "grad_norm": 0.000576019287109375,
      "learning_rate": 0.001757679548349571,
      "loss": 0.2309,
      "num_input_tokens_seen": 32992576,
      "step": 156335
    },
    {
      "epoch": 17.1991199119912,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.0017570032094872089,
      "loss": 0.2303,
      "num_input_tokens_seen": 32993600,
      "step": 156340
    },
    {
      "epoch": 17.199669966996698,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0017563269926791935,
      "loss": 0.2314,
      "num_input_tokens_seen": 32994624,
      "step": 156345
    },
    {
      "epoch": 17.2002200220022,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0017556508979317514,
      "loss": 0.2313,
      "num_input_tokens_seen": 32995648,
      "step": 156350
    },
    {
      "epoch": 17.2007700770077,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.0017549749252511275,
      "loss": 0.2314,
      "num_input_tokens_seen": 32996736,
      "step": 156355
    },
    {
      "epoch": 17.2013201320132,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0017542990746435422,
      "loss": 0.2308,
      "num_input_tokens_seen": 32997760,
      "step": 156360
    },
    {
      "epoch": 17.201870187018702,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0017536233461152234,
      "loss": 0.2314,
      "num_input_tokens_seen": 32998816,
      "step": 156365
    },
    {
      "epoch": 17.202420242024203,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0017529477396724046,
      "loss": 0.2319,
      "num_input_tokens_seen": 32999872,
      "step": 156370
    },
    {
      "epoch": 17.202970297029704,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.0017522722553213077,
      "loss": 0.2298,
      "num_input_tokens_seen": 33000864,
      "step": 156375
    },
    {
      "epoch": 17.203520352035202,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0017515968930681607,
      "loss": 0.2298,
      "num_input_tokens_seen": 33001952,
      "step": 156380
    },
    {
      "epoch": 17.204070407040703,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.0017509216529191906,
      "loss": 0.2298,
      "num_input_tokens_seen": 33003008,
      "step": 156385
    },
    {
      "epoch": 17.204620462046204,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0017502465348806155,
      "loss": 0.2319,
      "num_input_tokens_seen": 33004128,
      "step": 156390
    },
    {
      "epoch": 17.205170517051705,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.001749571538958664,
      "loss": 0.2314,
      "num_input_tokens_seen": 33005184,
      "step": 156395
    },
    {
      "epoch": 17.205720572057206,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0017488966651595494,
      "loss": 0.2329,
      "num_input_tokens_seen": 33006272,
      "step": 156400
    },
    {
      "epoch": 17.206270627062707,
      "grad_norm": 0.002716064453125,
      "learning_rate": 0.0017482219134894981,
      "loss": 0.2335,
      "num_input_tokens_seen": 33007296,
      "step": 156405
    },
    {
      "epoch": 17.206820682068205,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.001747547283954724,
      "loss": 0.2303,
      "num_input_tokens_seen": 33008320,
      "step": 156410
    },
    {
      "epoch": 17.207370737073706,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0017468727765614482,
      "loss": 0.2319,
      "num_input_tokens_seen": 33009376,
      "step": 156415
    },
    {
      "epoch": 17.207920792079207,
      "grad_norm": 0.000514984130859375,
      "learning_rate": 0.0017461983913158897,
      "loss": 0.2314,
      "num_input_tokens_seen": 33010400,
      "step": 156420
    },
    {
      "epoch": 17.20847084708471,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0017455241282242561,
      "loss": 0.2324,
      "num_input_tokens_seen": 33011456,
      "step": 156425
    },
    {
      "epoch": 17.20902090209021,
      "grad_norm": 0.003021240234375,
      "learning_rate": 0.0017448499872927713,
      "loss": 0.2308,
      "num_input_tokens_seen": 33012512,
      "step": 156430
    },
    {
      "epoch": 17.20957095709571,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0017441759685276402,
      "loss": 0.2319,
      "num_input_tokens_seen": 33013536,
      "step": 156435
    },
    {
      "epoch": 17.21012101210121,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.001743502071935078,
      "loss": 0.2319,
      "num_input_tokens_seen": 33014528,
      "step": 156440
    },
    {
      "epoch": 17.21067106710671,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.0017428282975212993,
      "loss": 0.2319,
      "num_input_tokens_seen": 33015552,
      "step": 156445
    },
    {
      "epoch": 17.21122112211221,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.001742154645292508,
      "loss": 0.2319,
      "num_input_tokens_seen": 33016544,
      "step": 156450
    },
    {
      "epoch": 17.21177117711771,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0017414811152549187,
      "loss": 0.2335,
      "num_input_tokens_seen": 33017632,
      "step": 156455
    },
    {
      "epoch": 17.212321232123212,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.001740807707414735,
      "loss": 0.2314,
      "num_input_tokens_seen": 33018656,
      "step": 156460
    },
    {
      "epoch": 17.212871287128714,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0017401344217781616,
      "loss": 0.2308,
      "num_input_tokens_seen": 33019744,
      "step": 156465
    },
    {
      "epoch": 17.213421342134215,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0017394612583514074,
      "loss": 0.2308,
      "num_input_tokens_seen": 33020768,
      "step": 156470
    },
    {
      "epoch": 17.213971397139716,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0017387882171406755,
      "loss": 0.2335,
      "num_input_tokens_seen": 33021792,
      "step": 156475
    },
    {
      "epoch": 17.214521452145213,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0017381152981521724,
      "loss": 0.2314,
      "num_input_tokens_seen": 33022848,
      "step": 156480
    },
    {
      "epoch": 17.215071507150714,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0017374425013920952,
      "loss": 0.233,
      "num_input_tokens_seen": 33023936,
      "step": 156485
    },
    {
      "epoch": 17.215621562156215,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0017367698268666452,
      "loss": 0.2308,
      "num_input_tokens_seen": 33024928,
      "step": 156490
    },
    {
      "epoch": 17.216171617161717,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.001736097274582023,
      "loss": 0.2309,
      "num_input_tokens_seen": 33026048,
      "step": 156495
    },
    {
      "epoch": 17.216721672167218,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.001735424844544428,
      "loss": 0.2298,
      "num_input_tokens_seen": 33027072,
      "step": 156500
    },
    {
      "epoch": 17.21727172717272,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0017347525367600591,
      "loss": 0.2324,
      "num_input_tokens_seen": 33028096,
      "step": 156505
    },
    {
      "epoch": 17.217821782178216,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0017340803512351126,
      "loss": 0.2303,
      "num_input_tokens_seen": 33029088,
      "step": 156510
    },
    {
      "epoch": 17.218371837183717,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.0017334082879757772,
      "loss": 0.2329,
      "num_input_tokens_seen": 33030240,
      "step": 156515
    },
    {
      "epoch": 17.21892189218922,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0017327363469882544,
      "loss": 0.2308,
      "num_input_tokens_seen": 33031264,
      "step": 156520
    },
    {
      "epoch": 17.21947194719472,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0017320645282787328,
      "loss": 0.2303,
      "num_input_tokens_seen": 33032288,
      "step": 156525
    },
    {
      "epoch": 17.22002200220022,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0017313928318534054,
      "loss": 0.2293,
      "num_input_tokens_seen": 33033376,
      "step": 156530
    },
    {
      "epoch": 17.22057205720572,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.001730721257718466,
      "loss": 0.2298,
      "num_input_tokens_seen": 33034432,
      "step": 156535
    },
    {
      "epoch": 17.221122112211223,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.0017300498058800994,
      "loss": 0.2314,
      "num_input_tokens_seen": 33035488,
      "step": 156540
    },
    {
      "epoch": 17.22167216721672,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.0017293784763444992,
      "loss": 0.2303,
      "num_input_tokens_seen": 33036544,
      "step": 156545
    },
    {
      "epoch": 17.22222222222222,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.001728707269117845,
      "loss": 0.2319,
      "num_input_tokens_seen": 33037504,
      "step": 156550
    },
    {
      "epoch": 17.222772277227723,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.001728036184206329,
      "loss": 0.2303,
      "num_input_tokens_seen": 33038528,
      "step": 156555
    },
    {
      "epoch": 17.223322332233224,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0017273652216161378,
      "loss": 0.2309,
      "num_input_tokens_seen": 33039552,
      "step": 156560
    },
    {
      "epoch": 17.223872387238725,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0017266943813534497,
      "loss": 0.233,
      "num_input_tokens_seen": 33040640,
      "step": 156565
    },
    {
      "epoch": 17.224422442244226,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.0017260236634244546,
      "loss": 0.2314,
      "num_input_tokens_seen": 33041632,
      "step": 156570
    },
    {
      "epoch": 17.224972497249723,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0017253530678353278,
      "loss": 0.2319,
      "num_input_tokens_seen": 33042720,
      "step": 156575
    },
    {
      "epoch": 17.225522552255224,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.0017246825945922495,
      "loss": 0.2314,
      "num_input_tokens_seen": 33043840,
      "step": 156580
    },
    {
      "epoch": 17.226072607260726,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0017240122437014026,
      "loss": 0.2314,
      "num_input_tokens_seen": 33044928,
      "step": 156585
    },
    {
      "epoch": 17.226622662266227,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0017233420151689625,
      "loss": 0.2309,
      "num_input_tokens_seen": 33045952,
      "step": 156590
    },
    {
      "epoch": 17.227172717271728,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.0017226719090011127,
      "loss": 0.2303,
      "num_input_tokens_seen": 33046976,
      "step": 156595
    },
    {
      "epoch": 17.22772277227723,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0017220019252040246,
      "loss": 0.2314,
      "num_input_tokens_seen": 33048064,
      "step": 156600
    },
    {
      "epoch": 17.22827282728273,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0017213320637838702,
      "loss": 0.2324,
      "num_input_tokens_seen": 33049056,
      "step": 156605
    },
    {
      "epoch": 17.228822882288227,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0017206623247468277,
      "loss": 0.2314,
      "num_input_tokens_seen": 33050112,
      "step": 156610
    },
    {
      "epoch": 17.22937293729373,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.0017199927080990673,
      "loss": 0.2309,
      "num_input_tokens_seen": 33051136,
      "step": 156615
    },
    {
      "epoch": 17.22992299229923,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.001719323213846764,
      "loss": 0.2309,
      "num_input_tokens_seen": 33052160,
      "step": 156620
    },
    {
      "epoch": 17.23047304730473,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0017186538419960878,
      "loss": 0.233,
      "num_input_tokens_seen": 33053184,
      "step": 156625
    },
    {
      "epoch": 17.231023102310232,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0017179845925532021,
      "loss": 0.2314,
      "num_input_tokens_seen": 33054208,
      "step": 156630
    },
    {
      "epoch": 17.231573157315733,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0017173154655242822,
      "loss": 0.2319,
      "num_input_tokens_seen": 33055168,
      "step": 156635
    },
    {
      "epoch": 17.23212321232123,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001716646460915488,
      "loss": 0.2324,
      "num_input_tokens_seen": 33056192,
      "step": 156640
    },
    {
      "epoch": 17.23267326732673,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.001715977578732991,
      "loss": 0.2303,
      "num_input_tokens_seen": 33057248,
      "step": 156645
    },
    {
      "epoch": 17.233223322332233,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0017153088189829568,
      "loss": 0.2309,
      "num_input_tokens_seen": 33058272,
      "step": 156650
    },
    {
      "epoch": 17.233773377337734,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.0017146401816715434,
      "loss": 0.2308,
      "num_input_tokens_seen": 33059328,
      "step": 156655
    },
    {
      "epoch": 17.234323432343235,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0017139716668049193,
      "loss": 0.2298,
      "num_input_tokens_seen": 33060384,
      "step": 156660
    },
    {
      "epoch": 17.234873487348736,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.0017133032743892394,
      "loss": 0.2303,
      "num_input_tokens_seen": 33061408,
      "step": 156665
    },
    {
      "epoch": 17.235423542354237,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.0017126350044306693,
      "loss": 0.2309,
      "num_input_tokens_seen": 33062464,
      "step": 156670
    },
    {
      "epoch": 17.235973597359735,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0017119668569353685,
      "loss": 0.2309,
      "num_input_tokens_seen": 33063456,
      "step": 156675
    },
    {
      "epoch": 17.236523652365236,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.0017112988319094907,
      "loss": 0.2324,
      "num_input_tokens_seen": 33064512,
      "step": 156680
    },
    {
      "epoch": 17.237073707370737,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0017106309293591975,
      "loss": 0.2329,
      "num_input_tokens_seen": 33065568,
      "step": 156685
    },
    {
      "epoch": 17.237623762376238,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.0017099631492906392,
      "loss": 0.2303,
      "num_input_tokens_seen": 33066656,
      "step": 156690
    },
    {
      "epoch": 17.23817381738174,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0017092954917099772,
      "loss": 0.2298,
      "num_input_tokens_seen": 33067712,
      "step": 156695
    },
    {
      "epoch": 17.23872387238724,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.001708627956623357,
      "loss": 0.2324,
      "num_input_tokens_seen": 33068736,
      "step": 156700
    },
    {
      "epoch": 17.239273927392738,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.001707960544036935,
      "loss": 0.2314,
      "num_input_tokens_seen": 33069824,
      "step": 156705
    },
    {
      "epoch": 17.23982398239824,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0017072932539568663,
      "loss": 0.2298,
      "num_input_tokens_seen": 33070944,
      "step": 156710
    },
    {
      "epoch": 17.24037403740374,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0017066260863892928,
      "loss": 0.2303,
      "num_input_tokens_seen": 33071968,
      "step": 156715
    },
    {
      "epoch": 17.24092409240924,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.001705959041340373,
      "loss": 0.2314,
      "num_input_tokens_seen": 33073088,
      "step": 156720
    },
    {
      "epoch": 17.241474147414742,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.001705292118816245,
      "loss": 0.2314,
      "num_input_tokens_seen": 33074080,
      "step": 156725
    },
    {
      "epoch": 17.242024202420243,
      "grad_norm": 0.0006256103515625,
      "learning_rate": 0.0017046253188230608,
      "loss": 0.2293,
      "num_input_tokens_seen": 33075104,
      "step": 156730
    },
    {
      "epoch": 17.242574257425744,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.0017039586413669689,
      "loss": 0.2319,
      "num_input_tokens_seen": 33076128,
      "step": 156735
    },
    {
      "epoch": 17.24312431243124,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.001703292086454109,
      "loss": 0.2309,
      "num_input_tokens_seen": 33077248,
      "step": 156740
    },
    {
      "epoch": 17.243674367436743,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0017026256540906232,
      "loss": 0.2314,
      "num_input_tokens_seen": 33078304,
      "step": 156745
    },
    {
      "epoch": 17.244224422442244,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.001701959344282658,
      "loss": 0.2324,
      "num_input_tokens_seen": 33079328,
      "step": 156750
    },
    {
      "epoch": 17.244774477447745,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.0017012931570363503,
      "loss": 0.2298,
      "num_input_tokens_seen": 33080352,
      "step": 156755
    },
    {
      "epoch": 17.245324532453246,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.0017006270923578437,
      "loss": 0.2314,
      "num_input_tokens_seen": 33081408,
      "step": 156760
    },
    {
      "epoch": 17.245874587458747,
      "grad_norm": 0.000766754150390625,
      "learning_rate": 0.0016999611502532762,
      "loss": 0.2314,
      "num_input_tokens_seen": 33082464,
      "step": 156765
    },
    {
      "epoch": 17.246424642464245,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0016992953307287833,
      "loss": 0.2314,
      "num_input_tokens_seen": 33083456,
      "step": 156770
    },
    {
      "epoch": 17.246974697469746,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.0016986296337905048,
      "loss": 0.2293,
      "num_input_tokens_seen": 33084480,
      "step": 156775
    },
    {
      "epoch": 17.247524752475247,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.0016979640594445727,
      "loss": 0.2298,
      "num_input_tokens_seen": 33085504,
      "step": 156780
    },
    {
      "epoch": 17.248074807480748,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0016972986076971219,
      "loss": 0.2319,
      "num_input_tokens_seen": 33086560,
      "step": 156785
    },
    {
      "epoch": 17.24862486248625,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0016966332785542891,
      "loss": 0.2304,
      "num_input_tokens_seen": 33087616,
      "step": 156790
    },
    {
      "epoch": 17.24917491749175,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.001695968072022203,
      "loss": 0.2319,
      "num_input_tokens_seen": 33088736,
      "step": 156795
    },
    {
      "epoch": 17.24972497249725,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.001695302988106997,
      "loss": 0.2319,
      "num_input_tokens_seen": 33089856,
      "step": 156800
    },
    {
      "epoch": 17.25027502750275,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0016946380268147959,
      "loss": 0.2324,
      "num_input_tokens_seen": 33090880,
      "step": 156805
    },
    {
      "epoch": 17.25082508250825,
      "grad_norm": 0.00079345703125,
      "learning_rate": 0.001693973188151735,
      "loss": 0.2298,
      "num_input_tokens_seen": 33092000,
      "step": 156810
    },
    {
      "epoch": 17.25137513751375,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.0016933084721239343,
      "loss": 0.2324,
      "num_input_tokens_seen": 33093024,
      "step": 156815
    },
    {
      "epoch": 17.251925192519252,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0016926438787375241,
      "loss": 0.2309,
      "num_input_tokens_seen": 33094112,
      "step": 156820
    },
    {
      "epoch": 17.252475247524753,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0016919794079986344,
      "loss": 0.2314,
      "num_input_tokens_seen": 33095136,
      "step": 156825
    },
    {
      "epoch": 17.253025302530254,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0016913150599133819,
      "loss": 0.2308,
      "num_input_tokens_seen": 33096192,
      "step": 156830
    },
    {
      "epoch": 17.253575357535752,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0016906508344878934,
      "loss": 0.2309,
      "num_input_tokens_seen": 33097248,
      "step": 156835
    },
    {
      "epoch": 17.254125412541253,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.0016899867317282874,
      "loss": 0.2314,
      "num_input_tokens_seen": 33098272,
      "step": 156840
    },
    {
      "epoch": 17.254675467546754,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0016893227516406872,
      "loss": 0.2304,
      "num_input_tokens_seen": 33099328,
      "step": 156845
    },
    {
      "epoch": 17.255225522552255,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0016886588942312164,
      "loss": 0.2314,
      "num_input_tokens_seen": 33100384,
      "step": 156850
    },
    {
      "epoch": 17.255775577557756,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0016879951595059849,
      "loss": 0.2314,
      "num_input_tokens_seen": 33101376,
      "step": 156855
    },
    {
      "epoch": 17.256325632563257,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0016873315474711164,
      "loss": 0.2319,
      "num_input_tokens_seen": 33102400,
      "step": 156860
    },
    {
      "epoch": 17.25687568756876,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.001686668058132726,
      "loss": 0.233,
      "num_input_tokens_seen": 33103424,
      "step": 156865
    },
    {
      "epoch": 17.257425742574256,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.001686004691496925,
      "loss": 0.233,
      "num_input_tokens_seen": 33104576,
      "step": 156870
    },
    {
      "epoch": 17.257975797579757,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0016853414475698292,
      "loss": 0.2324,
      "num_input_tokens_seen": 33105600,
      "step": 156875
    },
    {
      "epoch": 17.258525852585258,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0016846783263575532,
      "loss": 0.2319,
      "num_input_tokens_seen": 33106624,
      "step": 156880
    },
    {
      "epoch": 17.25907590759076,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.0016840153278662106,
      "loss": 0.2309,
      "num_input_tokens_seen": 33107648,
      "step": 156885
    },
    {
      "epoch": 17.25962596259626,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0016833524521019099,
      "loss": 0.2314,
      "num_input_tokens_seen": 33108768,
      "step": 156890
    },
    {
      "epoch": 17.26017601760176,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.0016826896990707546,
      "loss": 0.2314,
      "num_input_tokens_seen": 33109792,
      "step": 156895
    },
    {
      "epoch": 17.260726072607262,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.0016820270687788596,
      "loss": 0.2309,
      "num_input_tokens_seen": 33110752,
      "step": 156900
    },
    {
      "epoch": 17.26127612761276,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0016813645612323302,
      "loss": 0.2319,
      "num_input_tokens_seen": 33111808,
      "step": 156905
    },
    {
      "epoch": 17.26182618261826,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0016807021764372748,
      "loss": 0.2324,
      "num_input_tokens_seen": 33112768,
      "step": 156910
    },
    {
      "epoch": 17.262376237623762,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.0016800399143997984,
      "loss": 0.2309,
      "num_input_tokens_seen": 33113792,
      "step": 156915
    },
    {
      "epoch": 17.262926292629263,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001679377775125998,
      "loss": 0.2314,
      "num_input_tokens_seen": 33114816,
      "step": 156920
    },
    {
      "epoch": 17.263476347634764,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.0016787157586219835,
      "loss": 0.2324,
      "num_input_tokens_seen": 33115904,
      "step": 156925
    },
    {
      "epoch": 17.264026402640265,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.001678053864893852,
      "loss": 0.2324,
      "num_input_tokens_seen": 33116928,
      "step": 156930
    },
    {
      "epoch": 17.264576457645763,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.0016773920939477031,
      "loss": 0.2319,
      "num_input_tokens_seen": 33118016,
      "step": 156935
    },
    {
      "epoch": 17.265126512651264,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.001676730445789644,
      "loss": 0.2309,
      "num_input_tokens_seen": 33119040,
      "step": 156940
    },
    {
      "epoch": 17.265676567656765,
      "grad_norm": 0.000659942626953125,
      "learning_rate": 0.0016760689204257633,
      "loss": 0.2314,
      "num_input_tokens_seen": 33120096,
      "step": 156945
    },
    {
      "epoch": 17.266226622662266,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0016754075178621658,
      "loss": 0.2319,
      "num_input_tokens_seen": 33121184,
      "step": 156950
    },
    {
      "epoch": 17.266776677667767,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0016747462381049415,
      "loss": 0.2303,
      "num_input_tokens_seen": 33122208,
      "step": 156955
    },
    {
      "epoch": 17.26732673267327,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0016740850811601826,
      "loss": 0.2319,
      "num_input_tokens_seen": 33123264,
      "step": 156960
    },
    {
      "epoch": 17.26787678767877,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0016734240470339921,
      "loss": 0.2313,
      "num_input_tokens_seen": 33124320,
      "step": 156965
    },
    {
      "epoch": 17.268426842684267,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.0016727631357324557,
      "loss": 0.2319,
      "num_input_tokens_seen": 33125472,
      "step": 156970
    },
    {
      "epoch": 17.268976897689768,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.001672102347261668,
      "loss": 0.2303,
      "num_input_tokens_seen": 33126528,
      "step": 156975
    },
    {
      "epoch": 17.26952695269527,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0016714416816277194,
      "loss": 0.2308,
      "num_input_tokens_seen": 33127616,
      "step": 156980
    },
    {
      "epoch": 17.27007700770077,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0016707811388366933,
      "loss": 0.2309,
      "num_input_tokens_seen": 33128608,
      "step": 156985
    },
    {
      "epoch": 17.27062706270627,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0016701207188946814,
      "loss": 0.2329,
      "num_input_tokens_seen": 33129664,
      "step": 156990
    },
    {
      "epoch": 17.271177117711773,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.001669460421807769,
      "loss": 0.2298,
      "num_input_tokens_seen": 33130720,
      "step": 156995
    },
    {
      "epoch": 17.27172717271727,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0016688002475820478,
      "loss": 0.2298,
      "num_input_tokens_seen": 33131776,
      "step": 157000
    },
    {
      "epoch": 17.27227722772277,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.001668140196223598,
      "loss": 0.2314,
      "num_input_tokens_seen": 33132864,
      "step": 157005
    },
    {
      "epoch": 17.272827282728272,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0016674802677384998,
      "loss": 0.2303,
      "num_input_tokens_seen": 33133856,
      "step": 157010
    },
    {
      "epoch": 17.273377337733773,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0016668204621328414,
      "loss": 0.2319,
      "num_input_tokens_seen": 33134848,
      "step": 157015
    },
    {
      "epoch": 17.273927392739274,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.0016661607794126948,
      "loss": 0.2324,
      "num_input_tokens_seen": 33135904,
      "step": 157020
    },
    {
      "epoch": 17.274477447744776,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0016655012195841516,
      "loss": 0.2303,
      "num_input_tokens_seen": 33136960,
      "step": 157025
    },
    {
      "epoch": 17.275027502750277,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.0016648417826532857,
      "loss": 0.2314,
      "num_input_tokens_seen": 33137952,
      "step": 157030
    },
    {
      "epoch": 17.275577557755774,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0016641824686261701,
      "loss": 0.2304,
      "num_input_tokens_seen": 33139072,
      "step": 157035
    },
    {
      "epoch": 17.276127612761275,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.0016635232775088903,
      "loss": 0.2329,
      "num_input_tokens_seen": 33140096,
      "step": 157040
    },
    {
      "epoch": 17.276677667766776,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.001662864209307513,
      "loss": 0.2299,
      "num_input_tokens_seen": 33141088,
      "step": 157045
    },
    {
      "epoch": 17.277227722772277,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0016622052640281164,
      "loss": 0.2308,
      "num_input_tokens_seen": 33142112,
      "step": 157050
    },
    {
      "epoch": 17.27777777777778,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.0016615464416767777,
      "loss": 0.2319,
      "num_input_tokens_seen": 33143168,
      "step": 157055
    },
    {
      "epoch": 17.27832783278328,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0016608877422595602,
      "loss": 0.2303,
      "num_input_tokens_seen": 33144256,
      "step": 157060
    },
    {
      "epoch": 17.278877887788777,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.001660229165782544,
      "loss": 0.2309,
      "num_input_tokens_seen": 33145280,
      "step": 157065
    },
    {
      "epoch": 17.27942794279428,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.001659570712251791,
      "loss": 0.2329,
      "num_input_tokens_seen": 33146240,
      "step": 157070
    },
    {
      "epoch": 17.27997799779978,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001658912381673378,
      "loss": 0.2314,
      "num_input_tokens_seen": 33147296,
      "step": 157075
    },
    {
      "epoch": 17.28052805280528,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0016582541740533634,
      "loss": 0.2319,
      "num_input_tokens_seen": 33148320,
      "step": 157080
    },
    {
      "epoch": 17.28107810781078,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0016575960893978174,
      "loss": 0.2324,
      "num_input_tokens_seen": 33149344,
      "step": 157085
    },
    {
      "epoch": 17.281628162816283,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0016569381277128103,
      "loss": 0.2308,
      "num_input_tokens_seen": 33150496,
      "step": 157090
    },
    {
      "epoch": 17.282178217821784,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0016562802890043987,
      "loss": 0.2298,
      "num_input_tokens_seen": 33151552,
      "step": 157095
    },
    {
      "epoch": 17.28272827282728,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0016556225732786512,
      "loss": 0.2308,
      "num_input_tokens_seen": 33152576,
      "step": 157100
    },
    {
      "epoch": 17.283278327832782,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0016549649805416243,
      "loss": 0.2309,
      "num_input_tokens_seen": 33153728,
      "step": 157105
    },
    {
      "epoch": 17.283828382838283,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.001654307510799382,
      "loss": 0.235,
      "num_input_tokens_seen": 33154784,
      "step": 157110
    },
    {
      "epoch": 17.284378437843785,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.001653650164057986,
      "loss": 0.2308,
      "num_input_tokens_seen": 33155840,
      "step": 157115
    },
    {
      "epoch": 17.284928492849286,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0016529929403234893,
      "loss": 0.2319,
      "num_input_tokens_seen": 33156832,
      "step": 157120
    },
    {
      "epoch": 17.285478547854787,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.001652335839601956,
      "loss": 0.2298,
      "num_input_tokens_seen": 33157888,
      "step": 157125
    },
    {
      "epoch": 17.286028602860284,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.0016516788618994376,
      "loss": 0.2303,
      "num_input_tokens_seen": 33158912,
      "step": 157130
    },
    {
      "epoch": 17.286578657865785,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0016510220072219843,
      "loss": 0.2324,
      "num_input_tokens_seen": 33159968,
      "step": 157135
    },
    {
      "epoch": 17.287128712871286,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0016503652755756615,
      "loss": 0.233,
      "num_input_tokens_seen": 33160992,
      "step": 157140
    },
    {
      "epoch": 17.287678767876788,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0016497086669665156,
      "loss": 0.2329,
      "num_input_tokens_seen": 33162048,
      "step": 157145
    },
    {
      "epoch": 17.28822882288229,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.001649052181400597,
      "loss": 0.2309,
      "num_input_tokens_seen": 33163104,
      "step": 157150
    },
    {
      "epoch": 17.28877887788779,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001648395818883959,
      "loss": 0.2314,
      "num_input_tokens_seen": 33164192,
      "step": 157155
    },
    {
      "epoch": 17.28932893289329,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.001647739579422649,
      "loss": 0.2324,
      "num_input_tokens_seen": 33165216,
      "step": 157160
    },
    {
      "epoch": 17.28987898789879,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0016470834630227164,
      "loss": 0.2303,
      "num_input_tokens_seen": 33166240,
      "step": 157165
    },
    {
      "epoch": 17.29042904290429,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0016464274696902103,
      "loss": 0.2293,
      "num_input_tokens_seen": 33167296,
      "step": 157170
    },
    {
      "epoch": 17.29097909790979,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.001645771599431172,
      "loss": 0.2308,
      "num_input_tokens_seen": 33168384,
      "step": 157175
    },
    {
      "epoch": 17.29152915291529,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0016451158522516507,
      "loss": 0.2319,
      "num_input_tokens_seen": 33169440,
      "step": 157180
    },
    {
      "epoch": 17.292079207920793,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0016444602281576858,
      "loss": 0.2314,
      "num_input_tokens_seen": 33170560,
      "step": 157185
    },
    {
      "epoch": 17.292629262926294,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0016438047271553262,
      "loss": 0.2324,
      "num_input_tokens_seen": 33171616,
      "step": 157190
    },
    {
      "epoch": 17.293179317931795,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0016431493492506072,
      "loss": 0.2309,
      "num_input_tokens_seen": 33172736,
      "step": 157195
    },
    {
      "epoch": 17.293729372937293,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0016424940944495703,
      "loss": 0.2303,
      "num_input_tokens_seen": 33173728,
      "step": 157200
    },
    {
      "epoch": 17.294279427942794,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0016418389627582575,
      "loss": 0.2303,
      "num_input_tokens_seen": 33174816,
      "step": 157205
    },
    {
      "epoch": 17.294829482948295,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0016411839541827038,
      "loss": 0.2308,
      "num_input_tokens_seen": 33175840,
      "step": 157210
    },
    {
      "epoch": 17.295379537953796,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0016405290687289498,
      "loss": 0.2335,
      "num_input_tokens_seen": 33176864,
      "step": 157215
    },
    {
      "epoch": 17.295929592959297,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.0016398743064030268,
      "loss": 0.2314,
      "num_input_tokens_seen": 33177888,
      "step": 157220
    },
    {
      "epoch": 17.296479647964798,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0016392196672109704,
      "loss": 0.233,
      "num_input_tokens_seen": 33178912,
      "step": 157225
    },
    {
      "epoch": 17.297029702970296,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.001638565151158819,
      "loss": 0.2314,
      "num_input_tokens_seen": 33179968,
      "step": 157230
    },
    {
      "epoch": 17.297579757975797,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0016379107582525974,
      "loss": 0.2314,
      "num_input_tokens_seen": 33181024,
      "step": 157235
    },
    {
      "epoch": 17.298129812981298,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.001637256488498343,
      "loss": 0.2314,
      "num_input_tokens_seen": 33182112,
      "step": 157240
    },
    {
      "epoch": 17.2986798679868,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0016366023419020859,
      "loss": 0.2303,
      "num_input_tokens_seen": 33183136,
      "step": 157245
    },
    {
      "epoch": 17.2992299229923,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.0016359483184698474,
      "loss": 0.2324,
      "num_input_tokens_seen": 33184128,
      "step": 157250
    },
    {
      "epoch": 17.2997799779978,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.0016352944182076617,
      "loss": 0.2324,
      "num_input_tokens_seen": 33185184,
      "step": 157255
    },
    {
      "epoch": 17.300330033003302,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0016346406411215536,
      "loss": 0.2303,
      "num_input_tokens_seen": 33186240,
      "step": 157260
    },
    {
      "epoch": 17.3008800880088,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0016339869872175532,
      "loss": 0.2329,
      "num_input_tokens_seen": 33187296,
      "step": 157265
    },
    {
      "epoch": 17.3014301430143,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0016333334565016794,
      "loss": 0.2335,
      "num_input_tokens_seen": 33188384,
      "step": 157270
    },
    {
      "epoch": 17.301980198019802,
      "grad_norm": 0.0026092529296875,
      "learning_rate": 0.0016326800489799554,
      "loss": 0.2283,
      "num_input_tokens_seen": 33189408,
      "step": 157275
    },
    {
      "epoch": 17.302530253025303,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0016320267646584062,
      "loss": 0.2303,
      "num_input_tokens_seen": 33190432,
      "step": 157280
    },
    {
      "epoch": 17.303080308030804,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.0016313736035430508,
      "loss": 0.2293,
      "num_input_tokens_seen": 33191456,
      "step": 157285
    },
    {
      "epoch": 17.303630363036305,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0016307205656399125,
      "loss": 0.2309,
      "num_input_tokens_seen": 33192512,
      "step": 157290
    },
    {
      "epoch": 17.304180418041803,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0016300676509550082,
      "loss": 0.2329,
      "num_input_tokens_seen": 33193568,
      "step": 157295
    },
    {
      "epoch": 17.304730473047304,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.0016294148594943514,
      "loss": 0.2298,
      "num_input_tokens_seen": 33194560,
      "step": 157300
    },
    {
      "epoch": 17.305280528052805,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0016287621912639653,
      "loss": 0.2319,
      "num_input_tokens_seen": 33195648,
      "step": 157305
    },
    {
      "epoch": 17.305830583058306,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0016281096462698573,
      "loss": 0.2293,
      "num_input_tokens_seen": 33196640,
      "step": 157310
    },
    {
      "epoch": 17.306380638063807,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.001627457224518049,
      "loss": 0.2324,
      "num_input_tokens_seen": 33197760,
      "step": 157315
    },
    {
      "epoch": 17.306930693069308,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0016268049260145523,
      "loss": 0.2335,
      "num_input_tokens_seen": 33198880,
      "step": 157320
    },
    {
      "epoch": 17.30748074807481,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.0016261527507653738,
      "loss": 0.2314,
      "num_input_tokens_seen": 33200000,
      "step": 157325
    },
    {
      "epoch": 17.308030803080307,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0016255006987765325,
      "loss": 0.2303,
      "num_input_tokens_seen": 33201088,
      "step": 157330
    },
    {
      "epoch": 17.308580858085808,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0016248487700540298,
      "loss": 0.2314,
      "num_input_tokens_seen": 33202144,
      "step": 157335
    },
    {
      "epoch": 17.30913091309131,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0016241969646038777,
      "loss": 0.2319,
      "num_input_tokens_seen": 33203200,
      "step": 157340
    },
    {
      "epoch": 17.30968096809681,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0016235452824320867,
      "loss": 0.2324,
      "num_input_tokens_seen": 33204288,
      "step": 157345
    },
    {
      "epoch": 17.31023102310231,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.0016228937235446583,
      "loss": 0.2319,
      "num_input_tokens_seen": 33205344,
      "step": 157350
    },
    {
      "epoch": 17.310781078107812,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0016222422879476027,
      "loss": 0.2319,
      "num_input_tokens_seen": 33206464,
      "step": 157355
    },
    {
      "epoch": 17.31133113311331,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0016215909756469187,
      "loss": 0.2319,
      "num_input_tokens_seen": 33207552,
      "step": 157360
    },
    {
      "epoch": 17.31188118811881,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.0016209397866486096,
      "loss": 0.2324,
      "num_input_tokens_seen": 33208640,
      "step": 157365
    },
    {
      "epoch": 17.312431243124312,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.0016202887209586792,
      "loss": 0.2298,
      "num_input_tokens_seen": 33209664,
      "step": 157370
    },
    {
      "epoch": 17.312981298129813,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0016196377785831256,
      "loss": 0.2324,
      "num_input_tokens_seen": 33210688,
      "step": 157375
    },
    {
      "epoch": 17.313531353135314,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.0016189869595279543,
      "loss": 0.2309,
      "num_input_tokens_seen": 33211744,
      "step": 157380
    },
    {
      "epoch": 17.314081408140815,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0016183362637991587,
      "loss": 0.2298,
      "num_input_tokens_seen": 33212832,
      "step": 157385
    },
    {
      "epoch": 17.314631463146316,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0016176856914027343,
      "loss": 0.2314,
      "num_input_tokens_seen": 33213920,
      "step": 157390
    },
    {
      "epoch": 17.315181518151814,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.0016170352423446792,
      "loss": 0.2303,
      "num_input_tokens_seen": 33214976,
      "step": 157395
    },
    {
      "epoch": 17.315731573157315,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.0016163849166309873,
      "loss": 0.2303,
      "num_input_tokens_seen": 33216000,
      "step": 157400
    },
    {
      "epoch": 17.316281628162816,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0016157347142676587,
      "loss": 0.2314,
      "num_input_tokens_seen": 33217024,
      "step": 157405
    },
    {
      "epoch": 17.316831683168317,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.0016150846352606784,
      "loss": 0.2308,
      "num_input_tokens_seen": 33218080,
      "step": 157410
    },
    {
      "epoch": 17.317381738173818,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.0016144346796160385,
      "loss": 0.2314,
      "num_input_tokens_seen": 33219104,
      "step": 157415
    },
    {
      "epoch": 17.31793179317932,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0016137848473397342,
      "loss": 0.2319,
      "num_input_tokens_seen": 33220192,
      "step": 157420
    },
    {
      "epoch": 17.318481848184817,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.001613135138437749,
      "loss": 0.2314,
      "num_input_tokens_seen": 33221280,
      "step": 157425
    },
    {
      "epoch": 17.319031903190318,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.001612485552916073,
      "loss": 0.2314,
      "num_input_tokens_seen": 33222336,
      "step": 157430
    },
    {
      "epoch": 17.31958195819582,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0016118360907806982,
      "loss": 0.2314,
      "num_input_tokens_seen": 33223328,
      "step": 157435
    },
    {
      "epoch": 17.32013201320132,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0016111867520376016,
      "loss": 0.2319,
      "num_input_tokens_seen": 33224320,
      "step": 157440
    },
    {
      "epoch": 17.32068206820682,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0016105375366927749,
      "loss": 0.2319,
      "num_input_tokens_seen": 33225440,
      "step": 157445
    },
    {
      "epoch": 17.321232123212322,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.0016098884447521965,
      "loss": 0.2293,
      "num_input_tokens_seen": 33226560,
      "step": 157450
    },
    {
      "epoch": 17.321782178217823,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.0016092394762218503,
      "loss": 0.2283,
      "num_input_tokens_seen": 33227680,
      "step": 157455
    },
    {
      "epoch": 17.32233223322332,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0016085906311077212,
      "loss": 0.2298,
      "num_input_tokens_seen": 33228672,
      "step": 157460
    },
    {
      "epoch": 17.322882288228822,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.0016079419094157847,
      "loss": 0.2314,
      "num_input_tokens_seen": 33229728,
      "step": 157465
    },
    {
      "epoch": 17.323432343234323,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0016072933111520226,
      "loss": 0.2319,
      "num_input_tokens_seen": 33230784,
      "step": 157470
    },
    {
      "epoch": 17.323982398239824,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.00160664483632241,
      "loss": 0.2329,
      "num_input_tokens_seen": 33231872,
      "step": 157475
    },
    {
      "epoch": 17.324532453245325,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0016059964849329288,
      "loss": 0.2309,
      "num_input_tokens_seen": 33232992,
      "step": 157480
    },
    {
      "epoch": 17.325082508250826,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0016053482569895461,
      "loss": 0.2308,
      "num_input_tokens_seen": 33234016,
      "step": 157485
    },
    {
      "epoch": 17.325632563256324,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0016047001524982433,
      "loss": 0.2314,
      "num_input_tokens_seen": 33235104,
      "step": 157490
    },
    {
      "epoch": 17.326182618261825,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001604052171464993,
      "loss": 0.2283,
      "num_input_tokens_seen": 33236128,
      "step": 157495
    },
    {
      "epoch": 17.326732673267326,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.0016034043138957632,
      "loss": 0.2319,
      "num_input_tokens_seen": 33237216,
      "step": 157500
    },
    {
      "epoch": 17.327282728272827,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.001602756579796531,
      "loss": 0.2314,
      "num_input_tokens_seen": 33238240,
      "step": 157505
    },
    {
      "epoch": 17.32783278327833,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.001602108969173258,
      "loss": 0.2313,
      "num_input_tokens_seen": 33239296,
      "step": 157510
    },
    {
      "epoch": 17.32838283828383,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.00160146148203192,
      "loss": 0.2309,
      "num_input_tokens_seen": 33240384,
      "step": 157515
    },
    {
      "epoch": 17.32893289328933,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0016008141183784835,
      "loss": 0.2345,
      "num_input_tokens_seen": 33241408,
      "step": 157520
    },
    {
      "epoch": 17.329482948294828,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0016001668782189154,
      "loss": 0.2335,
      "num_input_tokens_seen": 33242400,
      "step": 157525
    },
    {
      "epoch": 17.33003300330033,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0015995197615591744,
      "loss": 0.2308,
      "num_input_tokens_seen": 33243424,
      "step": 157530
    },
    {
      "epoch": 17.33058305830583,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.0015988727684052339,
      "loss": 0.2298,
      "num_input_tokens_seen": 33244544,
      "step": 157535
    },
    {
      "epoch": 17.33113311331133,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0015982258987630492,
      "loss": 0.2324,
      "num_input_tokens_seen": 33245632,
      "step": 157540
    },
    {
      "epoch": 17.331683168316832,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0015975791526385856,
      "loss": 0.2329,
      "num_input_tokens_seen": 33246656,
      "step": 157545
    },
    {
      "epoch": 17.332233223322334,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0015969325300378067,
      "loss": 0.2319,
      "num_input_tokens_seen": 33247744,
      "step": 157550
    },
    {
      "epoch": 17.33278327832783,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0015962860309666675,
      "loss": 0.2324,
      "num_input_tokens_seen": 33248832,
      "step": 157555
    },
    {
      "epoch": 17.333333333333332,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0015956396554311302,
      "loss": 0.2308,
      "num_input_tokens_seen": 33249824,
      "step": 157560
    },
    {
      "epoch": 17.333883388338833,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0015949934034371466,
      "loss": 0.2308,
      "num_input_tokens_seen": 33250944,
      "step": 157565
    },
    {
      "epoch": 17.334433443344334,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0015943472749906766,
      "loss": 0.2319,
      "num_input_tokens_seen": 33251968,
      "step": 157570
    },
    {
      "epoch": 17.334983498349835,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.0015937012700976794,
      "loss": 0.2335,
      "num_input_tokens_seen": 33253120,
      "step": 157575
    },
    {
      "epoch": 17.335533553355337,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0015930553887640996,
      "loss": 0.2319,
      "num_input_tokens_seen": 33254176,
      "step": 157580
    },
    {
      "epoch": 17.336083608360838,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0015924096309958996,
      "loss": 0.2324,
      "num_input_tokens_seen": 33255200,
      "step": 157585
    },
    {
      "epoch": 17.336633663366335,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0015917639967990226,
      "loss": 0.2319,
      "num_input_tokens_seen": 33256256,
      "step": 157590
    },
    {
      "epoch": 17.337183718371836,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0015911184861794275,
      "loss": 0.2303,
      "num_input_tokens_seen": 33257344,
      "step": 157595
    },
    {
      "epoch": 17.337733773377337,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0015904730991430542,
      "loss": 0.2314,
      "num_input_tokens_seen": 33258336,
      "step": 157600
    },
    {
      "epoch": 17.33828382838284,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.0015898278356958567,
      "loss": 0.2303,
      "num_input_tokens_seen": 33259392,
      "step": 157605
    },
    {
      "epoch": 17.33883388338834,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.001589182695843785,
      "loss": 0.2308,
      "num_input_tokens_seen": 33260448,
      "step": 157610
    },
    {
      "epoch": 17.33938393839384,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0015885376795927773,
      "loss": 0.2324,
      "num_input_tokens_seen": 33261472,
      "step": 157615
    },
    {
      "epoch": 17.33993399339934,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.0015878927869487846,
      "loss": 0.2314,
      "num_input_tokens_seen": 33262560,
      "step": 157620
    },
    {
      "epoch": 17.34048404840484,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.0015872480179177466,
      "loss": 0.2324,
      "num_input_tokens_seen": 33263552,
      "step": 157625
    },
    {
      "epoch": 17.34103410341034,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.001586603372505607,
      "loss": 0.2319,
      "num_input_tokens_seen": 33264544,
      "step": 157630
    },
    {
      "epoch": 17.34158415841584,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0015859588507183113,
      "loss": 0.2298,
      "num_input_tokens_seen": 33265536,
      "step": 157635
    },
    {
      "epoch": 17.342134213421343,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.0015853144525617912,
      "loss": 0.233,
      "num_input_tokens_seen": 33266624,
      "step": 157640
    },
    {
      "epoch": 17.342684268426844,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.001584670178041997,
      "loss": 0.2303,
      "num_input_tokens_seen": 33267616,
      "step": 157645
    },
    {
      "epoch": 17.343234323432345,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0015840260271648588,
      "loss": 0.2319,
      "num_input_tokens_seen": 33268672,
      "step": 157650
    },
    {
      "epoch": 17.343784378437842,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.001583381999936312,
      "loss": 0.2303,
      "num_input_tokens_seen": 33269664,
      "step": 157655
    },
    {
      "epoch": 17.344334433443343,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.0015827380963622949,
      "loss": 0.2314,
      "num_input_tokens_seen": 33270624,
      "step": 157660
    },
    {
      "epoch": 17.344884488448844,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0015820943164487433,
      "loss": 0.2298,
      "num_input_tokens_seen": 33271712,
      "step": 157665
    },
    {
      "epoch": 17.345434543454346,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.001581450660201592,
      "loss": 0.2314,
      "num_input_tokens_seen": 33272768,
      "step": 157670
    },
    {
      "epoch": 17.345984598459847,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0015808071276267714,
      "loss": 0.2303,
      "num_input_tokens_seen": 33273824,
      "step": 157675
    },
    {
      "epoch": 17.346534653465348,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0015801637187302102,
      "loss": 0.2303,
      "num_input_tokens_seen": 33274848,
      "step": 157680
    },
    {
      "epoch": 17.34708470847085,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.0015795204335178402,
      "loss": 0.2324,
      "num_input_tokens_seen": 33275840,
      "step": 157685
    },
    {
      "epoch": 17.347634763476346,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0015788772719955901,
      "loss": 0.2314,
      "num_input_tokens_seen": 33276832,
      "step": 157690
    },
    {
      "epoch": 17.348184818481847,
      "grad_norm": 0.0031890869140625,
      "learning_rate": 0.0015782342341693915,
      "loss": 0.2314,
      "num_input_tokens_seen": 33277952,
      "step": 157695
    },
    {
      "epoch": 17.34873487348735,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0015775913200451685,
      "loss": 0.2319,
      "num_input_tokens_seen": 33279040,
      "step": 157700
    },
    {
      "epoch": 17.34928492849285,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.001576948529628841,
      "loss": 0.2298,
      "num_input_tokens_seen": 33280096,
      "step": 157705
    },
    {
      "epoch": 17.34983498349835,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.001576305862926341,
      "loss": 0.2329,
      "num_input_tokens_seen": 33281152,
      "step": 157710
    },
    {
      "epoch": 17.350385038503852,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.001575663319943587,
      "loss": 0.2319,
      "num_input_tokens_seen": 33282240,
      "step": 157715
    },
    {
      "epoch": 17.35093509350935,
      "grad_norm": 0.0004634857177734375,
      "learning_rate": 0.0015750209006865012,
      "loss": 0.2314,
      "num_input_tokens_seen": 33283264,
      "step": 157720
    },
    {
      "epoch": 17.35148514851485,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.0015743786051610087,
      "loss": 0.2303,
      "num_input_tokens_seen": 33284256,
      "step": 157725
    },
    {
      "epoch": 17.35203520352035,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0015737364333730246,
      "loss": 0.2298,
      "num_input_tokens_seen": 33285312,
      "step": 157730
    },
    {
      "epoch": 17.352585258525853,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.001573094385328471,
      "loss": 0.2324,
      "num_input_tokens_seen": 33286368,
      "step": 157735
    },
    {
      "epoch": 17.353135313531354,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0015724524610332652,
      "loss": 0.2298,
      "num_input_tokens_seen": 33287424,
      "step": 157740
    },
    {
      "epoch": 17.353685368536855,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.001571810660493315,
      "loss": 0.2319,
      "num_input_tokens_seen": 33288512,
      "step": 157745
    },
    {
      "epoch": 17.354235423542356,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0015711689837145481,
      "loss": 0.2329,
      "num_input_tokens_seen": 33289600,
      "step": 157750
    },
    {
      "epoch": 17.354785478547853,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0015705274307028694,
      "loss": 0.2303,
      "num_input_tokens_seen": 33290688,
      "step": 157755
    },
    {
      "epoch": 17.355335533553355,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0015698860014641991,
      "loss": 0.2303,
      "num_input_tokens_seen": 33291744,
      "step": 157760
    },
    {
      "epoch": 17.355885588558856,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.001569244696004446,
      "loss": 0.2319,
      "num_input_tokens_seen": 33292704,
      "step": 157765
    },
    {
      "epoch": 17.356435643564357,
      "grad_norm": 0.00078582763671875,
      "learning_rate": 0.0015686035143295152,
      "loss": 0.2314,
      "num_input_tokens_seen": 33293696,
      "step": 157770
    },
    {
      "epoch": 17.356985698569858,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.001567962456445322,
      "loss": 0.2309,
      "num_input_tokens_seen": 33294848,
      "step": 157775
    },
    {
      "epoch": 17.35753575357536,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0015673215223577718,
      "loss": 0.2309,
      "num_input_tokens_seen": 33295904,
      "step": 157780
    },
    {
      "epoch": 17.358085808580856,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.0015666807120727781,
      "loss": 0.2319,
      "num_input_tokens_seen": 33296960,
      "step": 157785
    },
    {
      "epoch": 17.358635863586358,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0015660400255962414,
      "loss": 0.2329,
      "num_input_tokens_seen": 33297984,
      "step": 157790
    },
    {
      "epoch": 17.35918591859186,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.001565399462934065,
      "loss": 0.2314,
      "num_input_tokens_seen": 33299008,
      "step": 157795
    },
    {
      "epoch": 17.35973597359736,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0015647590240921592,
      "loss": 0.2298,
      "num_input_tokens_seen": 33300128,
      "step": 157800
    },
    {
      "epoch": 17.36028602860286,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0015641187090764163,
      "loss": 0.2319,
      "num_input_tokens_seen": 33301152,
      "step": 157805
    },
    {
      "epoch": 17.360836083608362,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0015634785178927496,
      "loss": 0.2324,
      "num_input_tokens_seen": 33302240,
      "step": 157810
    },
    {
      "epoch": 17.361386138613863,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.0015628384505470526,
      "loss": 0.2314,
      "num_input_tokens_seen": 33303328,
      "step": 157815
    },
    {
      "epoch": 17.36193619361936,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0015621985070452243,
      "loss": 0.2319,
      "num_input_tokens_seen": 33304416,
      "step": 157820
    },
    {
      "epoch": 17.36248624862486,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.001561558687393168,
      "loss": 0.2303,
      "num_input_tokens_seen": 33305472,
      "step": 157825
    },
    {
      "epoch": 17.363036303630363,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0015609189915967725,
      "loss": 0.2324,
      "num_input_tokens_seen": 33306496,
      "step": 157830
    },
    {
      "epoch": 17.363586358635864,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.001560279419661938,
      "loss": 0.2324,
      "num_input_tokens_seen": 33307520,
      "step": 157835
    },
    {
      "epoch": 17.364136413641365,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0015596399715945629,
      "loss": 0.2293,
      "num_input_tokens_seen": 33308512,
      "step": 157840
    },
    {
      "epoch": 17.364686468646866,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.001559000647400533,
      "loss": 0.2308,
      "num_input_tokens_seen": 33309568,
      "step": 157845
    },
    {
      "epoch": 17.365236523652364,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.001558361447085748,
      "loss": 0.2308,
      "num_input_tokens_seen": 33310592,
      "step": 157850
    },
    {
      "epoch": 17.365786578657865,
      "grad_norm": 0.005859375,
      "learning_rate": 0.001557722370656092,
      "loss": 0.2324,
      "num_input_tokens_seen": 33311744,
      "step": 157855
    },
    {
      "epoch": 17.366336633663366,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0015570834181174635,
      "loss": 0.2314,
      "num_input_tokens_seen": 33312768,
      "step": 157860
    },
    {
      "epoch": 17.366886688668867,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0015564445894757411,
      "loss": 0.2324,
      "num_input_tokens_seen": 33313824,
      "step": 157865
    },
    {
      "epoch": 17.367436743674368,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.00155580588473682,
      "loss": 0.2303,
      "num_input_tokens_seen": 33314848,
      "step": 157870
    },
    {
      "epoch": 17.36798679867987,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.001555167303906587,
      "loss": 0.233,
      "num_input_tokens_seen": 33315872,
      "step": 157875
    },
    {
      "epoch": 17.36853685368537,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0015545288469909228,
      "loss": 0.2334,
      "num_input_tokens_seen": 33316896,
      "step": 157880
    },
    {
      "epoch": 17.369086908690868,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0015538905139957176,
      "loss": 0.2324,
      "num_input_tokens_seen": 33317952,
      "step": 157885
    },
    {
      "epoch": 17.36963696369637,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.001553252304926848,
      "loss": 0.2308,
      "num_input_tokens_seen": 33319040,
      "step": 157890
    },
    {
      "epoch": 17.37018701870187,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.0015526142197901998,
      "loss": 0.2314,
      "num_input_tokens_seen": 33320064,
      "step": 157895
    },
    {
      "epoch": 17.37073707370737,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0015519762585916562,
      "loss": 0.2319,
      "num_input_tokens_seen": 33321088,
      "step": 157900
    },
    {
      "epoch": 17.371287128712872,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0015513384213370945,
      "loss": 0.2319,
      "num_input_tokens_seen": 33322208,
      "step": 157905
    },
    {
      "epoch": 17.371837183718373,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0015507007080323913,
      "loss": 0.2309,
      "num_input_tokens_seen": 33323296,
      "step": 157910
    },
    {
      "epoch": 17.37238723872387,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.001550063118683429,
      "loss": 0.2303,
      "num_input_tokens_seen": 33324416,
      "step": 157915
    },
    {
      "epoch": 17.372937293729372,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0015494256532960742,
      "loss": 0.2309,
      "num_input_tokens_seen": 33325504,
      "step": 157920
    },
    {
      "epoch": 17.373487348734873,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.001548788311876214,
      "loss": 0.2319,
      "num_input_tokens_seen": 33326496,
      "step": 157925
    },
    {
      "epoch": 17.374037403740374,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0015481510944297188,
      "loss": 0.2324,
      "num_input_tokens_seen": 33327616,
      "step": 157930
    },
    {
      "epoch": 17.374587458745875,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0015475140009624572,
      "loss": 0.2329,
      "num_input_tokens_seen": 33328640,
      "step": 157935
    },
    {
      "epoch": 17.375137513751376,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0015468770314803076,
      "loss": 0.2309,
      "num_input_tokens_seen": 33329664,
      "step": 157940
    },
    {
      "epoch": 17.375687568756877,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.0015462401859891322,
      "loss": 0.2314,
      "num_input_tokens_seen": 33330688,
      "step": 157945
    },
    {
      "epoch": 17.376237623762375,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.0015456034644948063,
      "loss": 0.2313,
      "num_input_tokens_seen": 33331712,
      "step": 157950
    },
    {
      "epoch": 17.376787678767876,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0015449668670032002,
      "loss": 0.2303,
      "num_input_tokens_seen": 33332800,
      "step": 157955
    },
    {
      "epoch": 17.377337733773377,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.0015443303935201741,
      "loss": 0.2324,
      "num_input_tokens_seen": 33333856,
      "step": 157960
    },
    {
      "epoch": 17.377887788778878,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0015436940440516017,
      "loss": 0.2314,
      "num_input_tokens_seen": 33334912,
      "step": 157965
    },
    {
      "epoch": 17.37843784378438,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0015430578186033416,
      "loss": 0.2314,
      "num_input_tokens_seen": 33335936,
      "step": 157970
    },
    {
      "epoch": 17.37898789878988,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0015424217171812642,
      "loss": 0.2324,
      "num_input_tokens_seen": 33336928,
      "step": 157975
    },
    {
      "epoch": 17.379537953795378,
      "grad_norm": 0.00054931640625,
      "learning_rate": 0.0015417857397912248,
      "loss": 0.2308,
      "num_input_tokens_seen": 33337952,
      "step": 157980
    },
    {
      "epoch": 17.38008800880088,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.001541149886439087,
      "loss": 0.2298,
      "num_input_tokens_seen": 33339072,
      "step": 157985
    },
    {
      "epoch": 17.38063806380638,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0015405141571307162,
      "loss": 0.2324,
      "num_input_tokens_seen": 33340128,
      "step": 157990
    },
    {
      "epoch": 17.38118811881188,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0015398785518719658,
      "loss": 0.2303,
      "num_input_tokens_seen": 33341184,
      "step": 157995
    },
    {
      "epoch": 17.381738173817382,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0015392430706686981,
      "loss": 0.2309,
      "num_input_tokens_seen": 33342240,
      "step": 158000
    },
    {
      "epoch": 17.382288228822883,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.001538607713526765,
      "loss": 0.2303,
      "num_input_tokens_seen": 33343296,
      "step": 158005
    },
    {
      "epoch": 17.382838283828384,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001537972480452025,
      "loss": 0.2298,
      "num_input_tokens_seen": 33344352,
      "step": 158010
    },
    {
      "epoch": 17.383388338833882,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0015373373714503368,
      "loss": 0.2303,
      "num_input_tokens_seen": 33345440,
      "step": 158015
    },
    {
      "epoch": 17.383938393839383,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.001536702386527546,
      "loss": 0.2314,
      "num_input_tokens_seen": 33346464,
      "step": 158020
    },
    {
      "epoch": 17.384488448844884,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0015360675256895122,
      "loss": 0.2293,
      "num_input_tokens_seen": 33347520,
      "step": 158025
    },
    {
      "epoch": 17.385038503850385,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.0015354327889420848,
      "loss": 0.2324,
      "num_input_tokens_seen": 33348544,
      "step": 158030
    },
    {
      "epoch": 17.385588558855886,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0015347981762911071,
      "loss": 0.2309,
      "num_input_tokens_seen": 33349568,
      "step": 158035
    },
    {
      "epoch": 17.386138613861387,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.0015341636877424364,
      "loss": 0.2303,
      "num_input_tokens_seen": 33350592,
      "step": 158040
    },
    {
      "epoch": 17.38668866886689,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.0015335293233019142,
      "loss": 0.2314,
      "num_input_tokens_seen": 33351712,
      "step": 158045
    },
    {
      "epoch": 17.387238723872386,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0015328950829753961,
      "loss": 0.2303,
      "num_input_tokens_seen": 33352832,
      "step": 158050
    },
    {
      "epoch": 17.387788778877887,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.0015322609667687192,
      "loss": 0.2288,
      "num_input_tokens_seen": 33353856,
      "step": 158055
    },
    {
      "epoch": 17.388338833883388,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0015316269746877304,
      "loss": 0.2314,
      "num_input_tokens_seen": 33354912,
      "step": 158060
    },
    {
      "epoch": 17.38888888888889,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.0015309931067382715,
      "loss": 0.2314,
      "num_input_tokens_seen": 33355904,
      "step": 158065
    },
    {
      "epoch": 17.38943894389439,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.0015303593629261863,
      "loss": 0.2309,
      "num_input_tokens_seen": 33356960,
      "step": 158070
    },
    {
      "epoch": 17.38998899889989,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0015297257432573168,
      "loss": 0.2324,
      "num_input_tokens_seen": 33358016,
      "step": 158075
    },
    {
      "epoch": 17.39053905390539,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.0015290922477375034,
      "loss": 0.2298,
      "num_input_tokens_seen": 33359040,
      "step": 158080
    },
    {
      "epoch": 17.39108910891089,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0015284588763725797,
      "loss": 0.2319,
      "num_input_tokens_seen": 33360096,
      "step": 158085
    },
    {
      "epoch": 17.39163916391639,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0015278256291683895,
      "loss": 0.2319,
      "num_input_tokens_seen": 33361184,
      "step": 158090
    },
    {
      "epoch": 17.392189218921892,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0015271925061307611,
      "loss": 0.2324,
      "num_input_tokens_seen": 33362240,
      "step": 158095
    },
    {
      "epoch": 17.392739273927393,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.001526559507265537,
      "loss": 0.234,
      "num_input_tokens_seen": 33363296,
      "step": 158100
    },
    {
      "epoch": 17.393289328932894,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0015259266325785502,
      "loss": 0.2308,
      "num_input_tokens_seen": 33364352,
      "step": 158105
    },
    {
      "epoch": 17.393839383938396,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00152529388207563,
      "loss": 0.2319,
      "num_input_tokens_seen": 33365440,
      "step": 158110
    },
    {
      "epoch": 17.394389438943893,
      "grad_norm": 0.001953125,
      "learning_rate": 0.0015246612557626115,
      "loss": 0.2314,
      "num_input_tokens_seen": 33366432,
      "step": 158115
    },
    {
      "epoch": 17.394939493949394,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0015240287536453234,
      "loss": 0.2304,
      "num_input_tokens_seen": 33367584,
      "step": 158120
    },
    {
      "epoch": 17.395489548954895,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0015233963757295944,
      "loss": 0.2308,
      "num_input_tokens_seen": 33368672,
      "step": 158125
    },
    {
      "epoch": 17.396039603960396,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.001522764122021258,
      "loss": 0.2324,
      "num_input_tokens_seen": 33369760,
      "step": 158130
    },
    {
      "epoch": 17.396589658965897,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0015221319925261345,
      "loss": 0.2298,
      "num_input_tokens_seen": 33370816,
      "step": 158135
    },
    {
      "epoch": 17.3971397139714,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0015214999872500544,
      "loss": 0.2324,
      "num_input_tokens_seen": 33371872,
      "step": 158140
    },
    {
      "epoch": 17.397689768976896,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.0015208681061988432,
      "loss": 0.2319,
      "num_input_tokens_seen": 33372960,
      "step": 158145
    },
    {
      "epoch": 17.398239823982397,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0015202363493783194,
      "loss": 0.233,
      "num_input_tokens_seen": 33374016,
      "step": 158150
    },
    {
      "epoch": 17.3987898789879,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.00151960471679431,
      "loss": 0.2324,
      "num_input_tokens_seen": 33375136,
      "step": 158155
    },
    {
      "epoch": 17.3993399339934,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 0.0015189732084526335,
      "loss": 0.2314,
      "num_input_tokens_seen": 33376224,
      "step": 158160
    },
    {
      "epoch": 17.3998899889989,
      "grad_norm": 0.00074005126953125,
      "learning_rate": 0.0015183418243591173,
      "loss": 0.2308,
      "num_input_tokens_seen": 33377280,
      "step": 158165
    },
    {
      "epoch": 17.4004400440044,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0015177105645195733,
      "loss": 0.2303,
      "num_input_tokens_seen": 33378304,
      "step": 158170
    },
    {
      "epoch": 17.400990099009903,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.0015170794289398198,
      "loss": 0.2308,
      "num_input_tokens_seen": 33379392,
      "step": 158175
    },
    {
      "epoch": 17.4015401540154,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0015164484176256742,
      "loss": 0.2324,
      "num_input_tokens_seen": 33380480,
      "step": 158180
    },
    {
      "epoch": 17.4020902090209,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0015158175305829552,
      "loss": 0.2324,
      "num_input_tokens_seen": 33381536,
      "step": 158185
    },
    {
      "epoch": 17.402640264026402,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.001515186767817478,
      "loss": 0.2324,
      "num_input_tokens_seen": 33382624,
      "step": 158190
    },
    {
      "epoch": 17.403190319031903,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0015145561293350529,
      "loss": 0.2329,
      "num_input_tokens_seen": 33383648,
      "step": 158195
    },
    {
      "epoch": 17.403740374037405,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0015139256151414904,
      "loss": 0.2314,
      "num_input_tokens_seen": 33384768,
      "step": 158200
    },
    {
      "epoch": 17.404290429042906,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0015132952252426073,
      "loss": 0.2304,
      "num_input_tokens_seen": 33385824,
      "step": 158205
    },
    {
      "epoch": 17.404840484048403,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.0015126649596442076,
      "loss": 0.2314,
      "num_input_tokens_seen": 33386848,
      "step": 158210
    },
    {
      "epoch": 17.405390539053904,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.0015120348183521048,
      "loss": 0.2319,
      "num_input_tokens_seen": 33387872,
      "step": 158215
    },
    {
      "epoch": 17.405940594059405,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0015114048013721059,
      "loss": 0.2314,
      "num_input_tokens_seen": 33388864,
      "step": 158220
    },
    {
      "epoch": 17.406490649064907,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.001510774908710013,
      "loss": 0.2309,
      "num_input_tokens_seen": 33389984,
      "step": 158225
    },
    {
      "epoch": 17.407040704070408,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.0015101451403716397,
      "loss": 0.2314,
      "num_input_tokens_seen": 33391072,
      "step": 158230
    },
    {
      "epoch": 17.40759075907591,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0015095154963627815,
      "loss": 0.2308,
      "num_input_tokens_seen": 33392096,
      "step": 158235
    },
    {
      "epoch": 17.40814081408141,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001508885976689247,
      "loss": 0.2319,
      "num_input_tokens_seen": 33393216,
      "step": 158240
    },
    {
      "epoch": 17.408690869086907,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0015082565813568382,
      "loss": 0.2314,
      "num_input_tokens_seen": 33394304,
      "step": 158245
    },
    {
      "epoch": 17.40924092409241,
      "grad_norm": 0.000644683837890625,
      "learning_rate": 0.0015076273103713522,
      "loss": 0.2308,
      "num_input_tokens_seen": 33395264,
      "step": 158250
    },
    {
      "epoch": 17.40979097909791,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.0015069981637385941,
      "loss": 0.2324,
      "num_input_tokens_seen": 33396384,
      "step": 158255
    },
    {
      "epoch": 17.41034103410341,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0015063691414643565,
      "loss": 0.2319,
      "num_input_tokens_seen": 33397440,
      "step": 158260
    },
    {
      "epoch": 17.41089108910891,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.001505740243554441,
      "loss": 0.2319,
      "num_input_tokens_seen": 33398464,
      "step": 158265
    },
    {
      "epoch": 17.411441144114413,
      "grad_norm": 0.000797271728515625,
      "learning_rate": 0.001505111470014641,
      "loss": 0.2309,
      "num_input_tokens_seen": 33399488,
      "step": 158270
    },
    {
      "epoch": 17.41199119911991,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0015044828208507527,
      "loss": 0.2298,
      "num_input_tokens_seen": 33400576,
      "step": 158275
    },
    {
      "epoch": 17.41254125412541,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.0015038542960685742,
      "loss": 0.2303,
      "num_input_tokens_seen": 33401728,
      "step": 158280
    },
    {
      "epoch": 17.413091309130913,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.0015032258956738942,
      "loss": 0.2314,
      "num_input_tokens_seen": 33402816,
      "step": 158285
    },
    {
      "epoch": 17.413641364136414,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.0015025976196725015,
      "loss": 0.2319,
      "num_input_tokens_seen": 33403840,
      "step": 158290
    },
    {
      "epoch": 17.414191419141915,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00150196946807019,
      "loss": 0.2319,
      "num_input_tokens_seen": 33404864,
      "step": 158295
    },
    {
      "epoch": 17.414741474147416,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0015013414408727483,
      "loss": 0.2308,
      "num_input_tokens_seen": 33405984,
      "step": 158300
    },
    {
      "epoch": 17.415291529152917,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0015007135380859698,
      "loss": 0.2309,
      "num_input_tokens_seen": 33406976,
      "step": 158305
    },
    {
      "epoch": 17.415841584158414,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0015000857597156352,
      "loss": 0.2309,
      "num_input_tokens_seen": 33408032,
      "step": 158310
    },
    {
      "epoch": 17.416391639163916,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0014994581057675299,
      "loss": 0.2298,
      "num_input_tokens_seen": 33409056,
      "step": 158315
    },
    {
      "epoch": 17.416941694169417,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0014988305762474441,
      "loss": 0.2308,
      "num_input_tokens_seen": 33410080,
      "step": 158320
    },
    {
      "epoch": 17.417491749174918,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0014982031711611565,
      "loss": 0.2298,
      "num_input_tokens_seen": 33411072,
      "step": 158325
    },
    {
      "epoch": 17.41804180418042,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.001497575890514451,
      "loss": 0.2324,
      "num_input_tokens_seen": 33412064,
      "step": 158330
    },
    {
      "epoch": 17.41859185918592,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.001496948734313111,
      "loss": 0.2314,
      "num_input_tokens_seen": 33413120,
      "step": 158335
    },
    {
      "epoch": 17.419141914191417,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0014963217025629138,
      "loss": 0.2329,
      "num_input_tokens_seen": 33414208,
      "step": 158340
    },
    {
      "epoch": 17.41969196919692,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.0014956947952696415,
      "loss": 0.2303,
      "num_input_tokens_seen": 33415264,
      "step": 158345
    },
    {
      "epoch": 17.42024202420242,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0014950680124390674,
      "loss": 0.2298,
      "num_input_tokens_seen": 33416352,
      "step": 158350
    },
    {
      "epoch": 17.42079207920792,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.0014944413540769724,
      "loss": 0.2324,
      "num_input_tokens_seen": 33417408,
      "step": 158355
    },
    {
      "epoch": 17.421342134213422,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0014938148201891332,
      "loss": 0.2324,
      "num_input_tokens_seen": 33418464,
      "step": 158360
    },
    {
      "epoch": 17.421892189218923,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0014931884107813186,
      "loss": 0.234,
      "num_input_tokens_seen": 33419488,
      "step": 158365
    },
    {
      "epoch": 17.422442244224424,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0014925621258593074,
      "loss": 0.2314,
      "num_input_tokens_seen": 33420576,
      "step": 158370
    },
    {
      "epoch": 17.42299229922992,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.0014919359654288681,
      "loss": 0.2314,
      "num_input_tokens_seen": 33421664,
      "step": 158375
    },
    {
      "epoch": 17.423542354235423,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.0014913099294957747,
      "loss": 0.2319,
      "num_input_tokens_seen": 33422752,
      "step": 158380
    },
    {
      "epoch": 17.424092409240924,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0014906840180657942,
      "loss": 0.2298,
      "num_input_tokens_seen": 33423776,
      "step": 158385
    },
    {
      "epoch": 17.424642464246425,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001490058231144697,
      "loss": 0.2314,
      "num_input_tokens_seen": 33424896,
      "step": 158390
    },
    {
      "epoch": 17.425192519251926,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.0014894325687382513,
      "loss": 0.2308,
      "num_input_tokens_seen": 33425984,
      "step": 158395
    },
    {
      "epoch": 17.425742574257427,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0014888070308522216,
      "loss": 0.2324,
      "num_input_tokens_seen": 33426976,
      "step": 158400
    },
    {
      "epoch": 17.426292629262925,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.0014881816174923761,
      "loss": 0.2314,
      "num_input_tokens_seen": 33428096,
      "step": 158405
    },
    {
      "epoch": 17.426842684268426,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0014875563286644738,
      "loss": 0.2314,
      "num_input_tokens_seen": 33429216,
      "step": 158410
    },
    {
      "epoch": 17.427392739273927,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0014869311643742815,
      "loss": 0.2329,
      "num_input_tokens_seen": 33430240,
      "step": 158415
    },
    {
      "epoch": 17.427942794279428,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0014863061246275632,
      "loss": 0.2319,
      "num_input_tokens_seen": 33431328,
      "step": 158420
    },
    {
      "epoch": 17.42849284928493,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0014856812094300742,
      "loss": 0.2324,
      "num_input_tokens_seen": 33432352,
      "step": 158425
    },
    {
      "epoch": 17.42904290429043,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0014850564187875797,
      "loss": 0.2298,
      "num_input_tokens_seen": 33433440,
      "step": 158430
    },
    {
      "epoch": 17.42959295929593,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.0014844317527058337,
      "loss": 0.2288,
      "num_input_tokens_seen": 33434464,
      "step": 158435
    },
    {
      "epoch": 17.43014301430143,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.001483807211190593,
      "loss": 0.2329,
      "num_input_tokens_seen": 33435456,
      "step": 158440
    },
    {
      "epoch": 17.43069306930693,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0014831827942476167,
      "loss": 0.2314,
      "num_input_tokens_seen": 33436480,
      "step": 158445
    },
    {
      "epoch": 17.43124312431243,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0014825585018826565,
      "loss": 0.2303,
      "num_input_tokens_seen": 33437568,
      "step": 158450
    },
    {
      "epoch": 17.431793179317932,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.001481934334101473,
      "loss": 0.2303,
      "num_input_tokens_seen": 33438656,
      "step": 158455
    },
    {
      "epoch": 17.432343234323433,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.001481310290909813,
      "loss": 0.2298,
      "num_input_tokens_seen": 33439744,
      "step": 158460
    },
    {
      "epoch": 17.432893289328934,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0014806863723134255,
      "loss": 0.2329,
      "num_input_tokens_seen": 33440768,
      "step": 158465
    },
    {
      "epoch": 17.433443344334435,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.0014800625783180658,
      "loss": 0.2314,
      "num_input_tokens_seen": 33441856,
      "step": 158470
    },
    {
      "epoch": 17.433993399339933,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0014794389089294846,
      "loss": 0.2308,
      "num_input_tokens_seen": 33442912,
      "step": 158475
    },
    {
      "epoch": 17.434543454345434,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0014788153641534252,
      "loss": 0.2309,
      "num_input_tokens_seen": 33444032,
      "step": 158480
    },
    {
      "epoch": 17.435093509350935,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0014781919439956382,
      "loss": 0.2319,
      "num_input_tokens_seen": 33445184,
      "step": 158485
    },
    {
      "epoch": 17.435643564356436,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.001477568648461864,
      "loss": 0.2319,
      "num_input_tokens_seen": 33446208,
      "step": 158490
    },
    {
      "epoch": 17.436193619361937,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.0014769454775578543,
      "loss": 0.2303,
      "num_input_tokens_seen": 33447200,
      "step": 158495
    },
    {
      "epoch": 17.436743674367438,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.0014763224312893469,
      "loss": 0.2319,
      "num_input_tokens_seen": 33448224,
      "step": 158500
    },
    {
      "epoch": 17.437293729372936,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0014756995096620866,
      "loss": 0.2293,
      "num_input_tokens_seen": 33449280,
      "step": 158505
    },
    {
      "epoch": 17.437843784378437,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.001475076712681816,
      "loss": 0.2314,
      "num_input_tokens_seen": 33450368,
      "step": 158510
    },
    {
      "epoch": 17.438393839383938,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0014744540403542716,
      "loss": 0.2303,
      "num_input_tokens_seen": 33451424,
      "step": 158515
    },
    {
      "epoch": 17.43894389438944,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.001473831492685196,
      "loss": 0.2314,
      "num_input_tokens_seen": 33452480,
      "step": 158520
    },
    {
      "epoch": 17.43949394939494,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0014732090696803256,
      "loss": 0.2309,
      "num_input_tokens_seen": 33453568,
      "step": 158525
    },
    {
      "epoch": 17.44004400440044,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.0014725867713453915,
      "loss": 0.2313,
      "num_input_tokens_seen": 33454560,
      "step": 158530
    },
    {
      "epoch": 17.440594059405942,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.001471964597686139,
      "loss": 0.235,
      "num_input_tokens_seen": 33455616,
      "step": 158535
    },
    {
      "epoch": 17.44114411441144,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.0014713425487082948,
      "loss": 0.2314,
      "num_input_tokens_seen": 33456640,
      "step": 158540
    },
    {
      "epoch": 17.44169416941694,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0014707206244175962,
      "loss": 0.2319,
      "num_input_tokens_seen": 33457728,
      "step": 158545
    },
    {
      "epoch": 17.442244224422442,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0014700988248197754,
      "loss": 0.2288,
      "num_input_tokens_seen": 33458752,
      "step": 158550
    },
    {
      "epoch": 17.442794279427943,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0014694771499205578,
      "loss": 0.2319,
      "num_input_tokens_seen": 33459872,
      "step": 158555
    },
    {
      "epoch": 17.443344334433444,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.001468855599725677,
      "loss": 0.2309,
      "num_input_tokens_seen": 33460992,
      "step": 158560
    },
    {
      "epoch": 17.443894389438945,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.0014682341742408617,
      "loss": 0.2324,
      "num_input_tokens_seen": 33462112,
      "step": 158565
    },
    {
      "epoch": 17.444444444444443,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0014676128734718407,
      "loss": 0.2303,
      "num_input_tokens_seen": 33463200,
      "step": 158570
    },
    {
      "epoch": 17.444994499449944,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.0014669916974243379,
      "loss": 0.2303,
      "num_input_tokens_seen": 33464224,
      "step": 158575
    },
    {
      "epoch": 17.445544554455445,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.001466370646104077,
      "loss": 0.2314,
      "num_input_tokens_seen": 33465280,
      "step": 158580
    },
    {
      "epoch": 17.446094609460946,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0014657497195167868,
      "loss": 0.2308,
      "num_input_tokens_seen": 33466304,
      "step": 158585
    },
    {
      "epoch": 17.446644664466447,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.0014651289176681808,
      "loss": 0.2324,
      "num_input_tokens_seen": 33467328,
      "step": 158590
    },
    {
      "epoch": 17.44719471947195,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.001464508240563993,
      "loss": 0.2319,
      "num_input_tokens_seen": 33468416,
      "step": 158595
    },
    {
      "epoch": 17.44774477447745,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0014638876882099388,
      "loss": 0.2303,
      "num_input_tokens_seen": 33469472,
      "step": 158600
    },
    {
      "epoch": 17.448294829482947,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001463267260611732,
      "loss": 0.2314,
      "num_input_tokens_seen": 33470464,
      "step": 158605
    },
    {
      "epoch": 17.448844884488448,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.0014626469577750994,
      "loss": 0.2309,
      "num_input_tokens_seen": 33471552,
      "step": 158610
    },
    {
      "epoch": 17.44939493949395,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.00146202677970575,
      "loss": 0.2314,
      "num_input_tokens_seen": 33472576,
      "step": 158615
    },
    {
      "epoch": 17.44994499449945,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.0014614067264094044,
      "loss": 0.2324,
      "num_input_tokens_seen": 33473600,
      "step": 158620
    },
    {
      "epoch": 17.45049504950495,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.0014607867978917775,
      "loss": 0.2309,
      "num_input_tokens_seen": 33474720,
      "step": 158625
    },
    {
      "epoch": 17.451045104510452,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.001460166994158582,
      "loss": 0.2329,
      "num_input_tokens_seen": 33475776,
      "step": 158630
    },
    {
      "epoch": 17.45159515951595,
      "grad_norm": 0.000637054443359375,
      "learning_rate": 0.0014595473152155313,
      "loss": 0.2293,
      "num_input_tokens_seen": 33476832,
      "step": 158635
    },
    {
      "epoch": 17.45214521452145,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0014589277610683325,
      "loss": 0.2309,
      "num_input_tokens_seen": 33477888,
      "step": 158640
    },
    {
      "epoch": 17.452695269526952,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.0014583083317227029,
      "loss": 0.2303,
      "num_input_tokens_seen": 33478880,
      "step": 158645
    },
    {
      "epoch": 17.453245324532453,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0014576890271843445,
      "loss": 0.2319,
      "num_input_tokens_seen": 33479968,
      "step": 158650
    },
    {
      "epoch": 17.453795379537954,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0014570698474589677,
      "loss": 0.2329,
      "num_input_tokens_seen": 33480992,
      "step": 158655
    },
    {
      "epoch": 17.454345434543455,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.001456450792552283,
      "loss": 0.2319,
      "num_input_tokens_seen": 33482144,
      "step": 158660
    },
    {
      "epoch": 17.454895489548957,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.001455831862469989,
      "loss": 0.2324,
      "num_input_tokens_seen": 33483168,
      "step": 158665
    },
    {
      "epoch": 17.455445544554454,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0014552130572177962,
      "loss": 0.2319,
      "num_input_tokens_seen": 33484224,
      "step": 158670
    },
    {
      "epoch": 17.455995599559955,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.0014545943768014035,
      "loss": 0.2314,
      "num_input_tokens_seen": 33485280,
      "step": 158675
    },
    {
      "epoch": 17.456545654565456,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.001453975821226513,
      "loss": 0.2314,
      "num_input_tokens_seen": 33486304,
      "step": 158680
    },
    {
      "epoch": 17.457095709570957,
      "grad_norm": 0.000652313232421875,
      "learning_rate": 0.0014533573904988316,
      "loss": 0.2314,
      "num_input_tokens_seen": 33487360,
      "step": 158685
    },
    {
      "epoch": 17.45764576457646,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0014527390846240551,
      "loss": 0.2298,
      "num_input_tokens_seen": 33488384,
      "step": 158690
    },
    {
      "epoch": 17.45819581958196,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.0014521209036078785,
      "loss": 0.2324,
      "num_input_tokens_seen": 33489440,
      "step": 158695
    },
    {
      "epoch": 17.458745874587457,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0014515028474560044,
      "loss": 0.2324,
      "num_input_tokens_seen": 33490496,
      "step": 158700
    },
    {
      "epoch": 17.459295929592958,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.001450884916174126,
      "loss": 0.2324,
      "num_input_tokens_seen": 33491520,
      "step": 158705
    },
    {
      "epoch": 17.45984598459846,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0014502671097679392,
      "loss": 0.2308,
      "num_input_tokens_seen": 33492608,
      "step": 158710
    },
    {
      "epoch": 17.46039603960396,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.001449649428243141,
      "loss": 0.2319,
      "num_input_tokens_seen": 33493696,
      "step": 158715
    },
    {
      "epoch": 17.46094609460946,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.0014490318716054201,
      "loss": 0.2319,
      "num_input_tokens_seen": 33494816,
      "step": 158720
    },
    {
      "epoch": 17.461496149614963,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.001448414439860472,
      "loss": 0.2293,
      "num_input_tokens_seen": 33495808,
      "step": 158725
    },
    {
      "epoch": 17.462046204620464,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0014477971330139837,
      "loss": 0.2303,
      "num_input_tokens_seen": 33496800,
      "step": 158730
    },
    {
      "epoch": 17.46259625962596,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.001447179951071646,
      "loss": 0.2308,
      "num_input_tokens_seen": 33497888,
      "step": 158735
    },
    {
      "epoch": 17.463146314631462,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.001446562894039149,
      "loss": 0.2308,
      "num_input_tokens_seen": 33498976,
      "step": 158740
    },
    {
      "epoch": 17.463696369636963,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.0014459459619221764,
      "loss": 0.2319,
      "num_input_tokens_seen": 33499968,
      "step": 158745
    },
    {
      "epoch": 17.464246424642464,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.0014453291547264191,
      "loss": 0.2314,
      "num_input_tokens_seen": 33500992,
      "step": 158750
    },
    {
      "epoch": 17.464796479647966,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0014447124724575539,
      "loss": 0.234,
      "num_input_tokens_seen": 33502144,
      "step": 158755
    },
    {
      "epoch": 17.465346534653467,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0014440959151212746,
      "loss": 0.2319,
      "num_input_tokens_seen": 33503232,
      "step": 158760
    },
    {
      "epoch": 17.465896589658964,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.001443479482723255,
      "loss": 0.2324,
      "num_input_tokens_seen": 33504288,
      "step": 158765
    },
    {
      "epoch": 17.466446644664465,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001442863175269179,
      "loss": 0.2319,
      "num_input_tokens_seen": 33505344,
      "step": 158770
    },
    {
      "epoch": 17.466996699669966,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0014422469927647319,
      "loss": 0.2314,
      "num_input_tokens_seen": 33506400,
      "step": 158775
    },
    {
      "epoch": 17.467546754675467,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.001441630935215586,
      "loss": 0.2303,
      "num_input_tokens_seen": 33507424,
      "step": 158780
    },
    {
      "epoch": 17.46809680968097,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0014410150026274232,
      "loss": 0.2308,
      "num_input_tokens_seen": 33508512,
      "step": 158785
    },
    {
      "epoch": 17.46864686468647,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0014403991950059174,
      "loss": 0.2314,
      "num_input_tokens_seen": 33509600,
      "step": 158790
    },
    {
      "epoch": 17.46919691969197,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.0014397835123567443,
      "loss": 0.2319,
      "num_input_tokens_seen": 33510688,
      "step": 158795
    },
    {
      "epoch": 17.46974697469747,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001439167954685584,
      "loss": 0.2319,
      "num_input_tokens_seen": 33511744,
      "step": 158800
    },
    {
      "epoch": 17.47029702970297,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0014385525219981021,
      "loss": 0.2324,
      "num_input_tokens_seen": 33512800,
      "step": 158805
    },
    {
      "epoch": 17.47084708470847,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0014379372142999758,
      "loss": 0.2324,
      "num_input_tokens_seen": 33513824,
      "step": 158810
    },
    {
      "epoch": 17.47139713971397,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001437322031596877,
      "loss": 0.2319,
      "num_input_tokens_seen": 33514880,
      "step": 158815
    },
    {
      "epoch": 17.471947194719473,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.0014367069738944682,
      "loss": 0.2314,
      "num_input_tokens_seen": 33515904,
      "step": 158820
    },
    {
      "epoch": 17.472497249724974,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0014360920411984229,
      "loss": 0.2303,
      "num_input_tokens_seen": 33517024,
      "step": 158825
    },
    {
      "epoch": 17.47304730473047,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.00143547723351441,
      "loss": 0.2298,
      "num_input_tokens_seen": 33518016,
      "step": 158830
    },
    {
      "epoch": 17.473597359735972,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0014348625508480966,
      "loss": 0.2293,
      "num_input_tokens_seen": 33519008,
      "step": 158835
    },
    {
      "epoch": 17.474147414741473,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.0014342479932051466,
      "loss": 0.2309,
      "num_input_tokens_seen": 33520160,
      "step": 158840
    },
    {
      "epoch": 17.474697469746975,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0014336335605912203,
      "loss": 0.2309,
      "num_input_tokens_seen": 33521184,
      "step": 158845
    },
    {
      "epoch": 17.475247524752476,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0014330192530119833,
      "loss": 0.2324,
      "num_input_tokens_seen": 33522272,
      "step": 158850
    },
    {
      "epoch": 17.475797579757977,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0014324050704730977,
      "loss": 0.2303,
      "num_input_tokens_seen": 33523296,
      "step": 158855
    },
    {
      "epoch": 17.476347634763478,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0014317910129802258,
      "loss": 0.2314,
      "num_input_tokens_seen": 33524320,
      "step": 158860
    },
    {
      "epoch": 17.476897689768975,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0014311770805390278,
      "loss": 0.2309,
      "num_input_tokens_seen": 33525312,
      "step": 158865
    },
    {
      "epoch": 17.477447744774476,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.001430563273155156,
      "loss": 0.2314,
      "num_input_tokens_seen": 33526400,
      "step": 158870
    },
    {
      "epoch": 17.477997799779978,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.001429949590834274,
      "loss": 0.2309,
      "num_input_tokens_seen": 33527488,
      "step": 158875
    },
    {
      "epoch": 17.47854785478548,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0014293360335820326,
      "loss": 0.2309,
      "num_input_tokens_seen": 33528544,
      "step": 158880
    },
    {
      "epoch": 17.47909790979098,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0014287226014040887,
      "loss": 0.2303,
      "num_input_tokens_seen": 33529600,
      "step": 158885
    },
    {
      "epoch": 17.47964796479648,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001428109294306098,
      "loss": 0.2309,
      "num_input_tokens_seen": 33530656,
      "step": 158890
    },
    {
      "epoch": 17.480198019801982,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0014274961122937107,
      "loss": 0.2314,
      "num_input_tokens_seen": 33531744,
      "step": 158895
    },
    {
      "epoch": 17.48074807480748,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0014268830553725808,
      "loss": 0.2319,
      "num_input_tokens_seen": 33532800,
      "step": 158900
    },
    {
      "epoch": 17.48129812981298,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0014262701235483555,
      "loss": 0.2308,
      "num_input_tokens_seen": 33533856,
      "step": 158905
    },
    {
      "epoch": 17.48184818481848,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0014256573168266833,
      "loss": 0.2298,
      "num_input_tokens_seen": 33534912,
      "step": 158910
    },
    {
      "epoch": 17.482398239823983,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0014250446352132168,
      "loss": 0.2319,
      "num_input_tokens_seen": 33535968,
      "step": 158915
    },
    {
      "epoch": 17.482948294829484,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0014244320787135977,
      "loss": 0.2303,
      "num_input_tokens_seen": 33537056,
      "step": 158920
    },
    {
      "epoch": 17.483498349834985,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0014238196473334785,
      "loss": 0.2319,
      "num_input_tokens_seen": 33538048,
      "step": 158925
    },
    {
      "epoch": 17.484048404840483,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001423207341078496,
      "loss": 0.2308,
      "num_input_tokens_seen": 33539104,
      "step": 158930
    },
    {
      "epoch": 17.484598459845984,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0014225951599542964,
      "loss": 0.2309,
      "num_input_tokens_seen": 33540128,
      "step": 158935
    },
    {
      "epoch": 17.485148514851485,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0014219831039665209,
      "loss": 0.2308,
      "num_input_tokens_seen": 33541216,
      "step": 158940
    },
    {
      "epoch": 17.485698569856986,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0014213711731208122,
      "loss": 0.2314,
      "num_input_tokens_seen": 33542208,
      "step": 158945
    },
    {
      "epoch": 17.486248624862487,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0014207593674228124,
      "loss": 0.2309,
      "num_input_tokens_seen": 33543232,
      "step": 158950
    },
    {
      "epoch": 17.486798679867988,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.001420147686878157,
      "loss": 0.2298,
      "num_input_tokens_seen": 33544256,
      "step": 158955
    },
    {
      "epoch": 17.48734873487349,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0014195361314924815,
      "loss": 0.2309,
      "num_input_tokens_seen": 33545312,
      "step": 158960
    },
    {
      "epoch": 17.487898789878987,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.0014189247012714261,
      "loss": 0.2314,
      "num_input_tokens_seen": 33546304,
      "step": 158965
    },
    {
      "epoch": 17.488448844884488,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.0014183133962206234,
      "loss": 0.2314,
      "num_input_tokens_seen": 33547328,
      "step": 158970
    },
    {
      "epoch": 17.48899889988999,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.0014177022163457135,
      "loss": 0.2319,
      "num_input_tokens_seen": 33548352,
      "step": 158975
    },
    {
      "epoch": 17.48954895489549,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.001417091161652324,
      "loss": 0.2324,
      "num_input_tokens_seen": 33549440,
      "step": 158980
    },
    {
      "epoch": 17.49009900990099,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0014164802321460833,
      "loss": 0.2324,
      "num_input_tokens_seen": 33550528,
      "step": 158985
    },
    {
      "epoch": 17.490649064906492,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0014158694278326306,
      "loss": 0.2324,
      "num_input_tokens_seen": 33551616,
      "step": 158990
    },
    {
      "epoch": 17.49119911991199,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.0014152587487175877,
      "loss": 0.2319,
      "num_input_tokens_seen": 33552608,
      "step": 158995
    },
    {
      "epoch": 17.49174917491749,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001414648194806587,
      "loss": 0.2314,
      "num_input_tokens_seen": 33553696,
      "step": 159000
    },
    {
      "epoch": 17.492299229922992,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.001414037766105257,
      "loss": 0.2308,
      "num_input_tokens_seen": 33554784,
      "step": 159005
    },
    {
      "epoch": 17.492849284928493,
      "grad_norm": 0.00067901611328125,
      "learning_rate": 0.0014134274626192206,
      "loss": 0.2314,
      "num_input_tokens_seen": 33555872,
      "step": 159010
    },
    {
      "epoch": 17.493399339933994,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0014128172843541043,
      "loss": 0.2319,
      "num_input_tokens_seen": 33556928,
      "step": 159015
    },
    {
      "epoch": 17.493949394939495,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.0014122072313155303,
      "loss": 0.2314,
      "num_input_tokens_seen": 33557952,
      "step": 159020
    },
    {
      "epoch": 17.494499449944996,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0014115973035091212,
      "loss": 0.2303,
      "num_input_tokens_seen": 33558976,
      "step": 159025
    },
    {
      "epoch": 17.495049504950494,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.001410987500940502,
      "loss": 0.2309,
      "num_input_tokens_seen": 33560064,
      "step": 159030
    },
    {
      "epoch": 17.495599559955995,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0014103778236152886,
      "loss": 0.2324,
      "num_input_tokens_seen": 33561056,
      "step": 159035
    },
    {
      "epoch": 17.496149614961496,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0014097682715391046,
      "loss": 0.2303,
      "num_input_tokens_seen": 33562144,
      "step": 159040
    },
    {
      "epoch": 17.496699669966997,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.0014091588447175624,
      "loss": 0.2314,
      "num_input_tokens_seen": 33563200,
      "step": 159045
    },
    {
      "epoch": 17.497249724972498,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0014085495431562839,
      "loss": 0.2319,
      "num_input_tokens_seen": 33564224,
      "step": 159050
    },
    {
      "epoch": 17.497799779978,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.0014079403668608797,
      "loss": 0.2319,
      "num_input_tokens_seen": 33565248,
      "step": 159055
    },
    {
      "epoch": 17.498349834983497,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.001407331315836967,
      "loss": 0.2319,
      "num_input_tokens_seen": 33566336,
      "step": 159060
    },
    {
      "epoch": 17.498899889988998,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0014067223900901616,
      "loss": 0.2324,
      "num_input_tokens_seen": 33567424,
      "step": 159065
    },
    {
      "epoch": 17.4994499449945,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0014061135896260735,
      "loss": 0.2308,
      "num_input_tokens_seen": 33568448,
      "step": 159070
    },
    {
      "epoch": 17.5,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0014055049144503118,
      "loss": 0.2309,
      "num_input_tokens_seen": 33569504,
      "step": 159075
    },
    {
      "epoch": 17.5005500550055,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.001404896364568487,
      "loss": 0.2329,
      "num_input_tokens_seen": 33570560,
      "step": 159080
    },
    {
      "epoch": 17.501100110011002,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.001404287939986208,
      "loss": 0.2309,
      "num_input_tokens_seen": 33571616,
      "step": 159085
    },
    {
      "epoch": 17.501650165016503,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0014036796407090868,
      "loss": 0.2319,
      "num_input_tokens_seen": 33572672,
      "step": 159090
    },
    {
      "epoch": 17.502200220022,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0014030714667427257,
      "loss": 0.2324,
      "num_input_tokens_seen": 33573728,
      "step": 159095
    },
    {
      "epoch": 17.502750275027502,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.0014024634180927287,
      "loss": 0.2314,
      "num_input_tokens_seen": 33574784,
      "step": 159100
    },
    {
      "epoch": 17.503300330033003,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0014018554947647043,
      "loss": 0.2314,
      "num_input_tokens_seen": 33575840,
      "step": 159105
    },
    {
      "epoch": 17.503850385038504,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00140124769676425,
      "loss": 0.2308,
      "num_input_tokens_seen": 33576896,
      "step": 159110
    },
    {
      "epoch": 17.504400440044005,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0014006400240969695,
      "loss": 0.2293,
      "num_input_tokens_seen": 33577952,
      "step": 159115
    },
    {
      "epoch": 17.504950495049506,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0014000324767684685,
      "loss": 0.2314,
      "num_input_tokens_seen": 33579072,
      "step": 159120
    },
    {
      "epoch": 17.505500550055004,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0013994250547843374,
      "loss": 0.2319,
      "num_input_tokens_seen": 33580160,
      "step": 159125
    },
    {
      "epoch": 17.506050605060505,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0013988177581501832,
      "loss": 0.2324,
      "num_input_tokens_seen": 33581216,
      "step": 159130
    },
    {
      "epoch": 17.506600660066006,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0013982105868715964,
      "loss": 0.2319,
      "num_input_tokens_seen": 33582304,
      "step": 159135
    },
    {
      "epoch": 17.507150715071507,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0013976035409541765,
      "loss": 0.2309,
      "num_input_tokens_seen": 33583328,
      "step": 159140
    },
    {
      "epoch": 17.507700770077008,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00139699662040352,
      "loss": 0.2335,
      "num_input_tokens_seen": 33584320,
      "step": 159145
    },
    {
      "epoch": 17.50825082508251,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.001396389825225216,
      "loss": 0.2314,
      "num_input_tokens_seen": 33585312,
      "step": 159150
    },
    {
      "epoch": 17.50880088008801,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0013957831554248616,
      "loss": 0.234,
      "num_input_tokens_seen": 33586400,
      "step": 159155
    },
    {
      "epoch": 17.509350935093508,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0013951766110080427,
      "loss": 0.2314,
      "num_input_tokens_seen": 33587488,
      "step": 159160
    },
    {
      "epoch": 17.50990099009901,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.001394570191980356,
      "loss": 0.2308,
      "num_input_tokens_seen": 33588544,
      "step": 159165
    },
    {
      "epoch": 17.51045104510451,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.0013939638983473856,
      "loss": 0.2309,
      "num_input_tokens_seen": 33589568,
      "step": 159170
    },
    {
      "epoch": 17.51100110011001,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0013933577301147204,
      "loss": 0.2319,
      "num_input_tokens_seen": 33590688,
      "step": 159175
    },
    {
      "epoch": 17.511551155115512,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0013927516872879492,
      "loss": 0.2308,
      "num_input_tokens_seen": 33591776,
      "step": 159180
    },
    {
      "epoch": 17.512101210121013,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.0013921457698726557,
      "loss": 0.2324,
      "num_input_tokens_seen": 33592832,
      "step": 159185
    },
    {
      "epoch": 17.51265126512651,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0013915399778744258,
      "loss": 0.2298,
      "num_input_tokens_seen": 33593920,
      "step": 159190
    },
    {
      "epoch": 17.513201320132012,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0013909343112988414,
      "loss": 0.2283,
      "num_input_tokens_seen": 33595040,
      "step": 159195
    },
    {
      "epoch": 17.513751375137513,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.0013903287701514833,
      "loss": 0.2303,
      "num_input_tokens_seen": 33596000,
      "step": 159200
    },
    {
      "epoch": 17.514301430143014,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.0013897233544379382,
      "loss": 0.2314,
      "num_input_tokens_seen": 33597088,
      "step": 159205
    },
    {
      "epoch": 17.514851485148515,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.001389118064163779,
      "loss": 0.234,
      "num_input_tokens_seen": 33598144,
      "step": 159210
    },
    {
      "epoch": 17.515401540154016,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001388512899334589,
      "loss": 0.2309,
      "num_input_tokens_seen": 33599296,
      "step": 159215
    },
    {
      "epoch": 17.515951595159517,
      "grad_norm": 0.000774383544921875,
      "learning_rate": 0.0013879078599559457,
      "loss": 0.2314,
      "num_input_tokens_seen": 33600352,
      "step": 159220
    },
    {
      "epoch": 17.516501650165015,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0013873029460334213,
      "loss": 0.2309,
      "num_input_tokens_seen": 33601376,
      "step": 159225
    },
    {
      "epoch": 17.517051705170516,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0013866981575725928,
      "loss": 0.2303,
      "num_input_tokens_seen": 33602432,
      "step": 159230
    },
    {
      "epoch": 17.517601760176017,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0013860934945790358,
      "loss": 0.2298,
      "num_input_tokens_seen": 33603456,
      "step": 159235
    },
    {
      "epoch": 17.51815181518152,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0013854889570583246,
      "loss": 0.2319,
      "num_input_tokens_seen": 33604544,
      "step": 159240
    },
    {
      "epoch": 17.51870187018702,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0013848845450160291,
      "loss": 0.2309,
      "num_input_tokens_seen": 33605664,
      "step": 159245
    },
    {
      "epoch": 17.51925192519252,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0013842802584577167,
      "loss": 0.2319,
      "num_input_tokens_seen": 33606752,
      "step": 159250
    },
    {
      "epoch": 17.519801980198018,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.0013836760973889599,
      "loss": 0.2298,
      "num_input_tokens_seen": 33607808,
      "step": 159255
    },
    {
      "epoch": 17.52035203520352,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.001383072061815329,
      "loss": 0.2314,
      "num_input_tokens_seen": 33608864,
      "step": 159260
    },
    {
      "epoch": 17.52090209020902,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0013824681517423864,
      "loss": 0.2314,
      "num_input_tokens_seen": 33609888,
      "step": 159265
    },
    {
      "epoch": 17.52145214521452,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0013818643671757041,
      "loss": 0.2319,
      "num_input_tokens_seen": 33610944,
      "step": 159270
    },
    {
      "epoch": 17.522002200220022,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0013812607081208393,
      "loss": 0.2303,
      "num_input_tokens_seen": 33612032,
      "step": 159275
    },
    {
      "epoch": 17.522552255225524,
      "grad_norm": 0.00069427490234375,
      "learning_rate": 0.001380657174583363,
      "loss": 0.2319,
      "num_input_tokens_seen": 33613056,
      "step": 159280
    },
    {
      "epoch": 17.523102310231025,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0013800537665688316,
      "loss": 0.2293,
      "num_input_tokens_seen": 33614080,
      "step": 159285
    },
    {
      "epoch": 17.523652365236522,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.001379450484082808,
      "loss": 0.2324,
      "num_input_tokens_seen": 33615104,
      "step": 159290
    },
    {
      "epoch": 17.524202420242023,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0013788473271308577,
      "loss": 0.2319,
      "num_input_tokens_seen": 33616160,
      "step": 159295
    },
    {
      "epoch": 17.524752475247524,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.0013782442957185326,
      "loss": 0.2319,
      "num_input_tokens_seen": 33617216,
      "step": 159300
    },
    {
      "epoch": 17.525302530253025,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0013776413898513966,
      "loss": 0.2303,
      "num_input_tokens_seen": 33618272,
      "step": 159305
    },
    {
      "epoch": 17.525852585258527,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0013770386095350022,
      "loss": 0.2308,
      "num_input_tokens_seen": 33619424,
      "step": 159310
    },
    {
      "epoch": 17.526402640264028,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0013764359547749016,
      "loss": 0.2303,
      "num_input_tokens_seen": 33620480,
      "step": 159315
    },
    {
      "epoch": 17.52695269526953,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0013758334255766585,
      "loss": 0.2329,
      "num_input_tokens_seen": 33621472,
      "step": 159320
    },
    {
      "epoch": 17.527502750275026,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0013752310219458169,
      "loss": 0.2335,
      "num_input_tokens_seen": 33622528,
      "step": 159325
    },
    {
      "epoch": 17.528052805280527,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001374628743887939,
      "loss": 0.2303,
      "num_input_tokens_seen": 33623616,
      "step": 159330
    },
    {
      "epoch": 17.52860286028603,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.0013740265914085668,
      "loss": 0.2303,
      "num_input_tokens_seen": 33624608,
      "step": 159335
    },
    {
      "epoch": 17.52915291529153,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.001373424564513253,
      "loss": 0.2293,
      "num_input_tokens_seen": 33625664,
      "step": 159340
    },
    {
      "epoch": 17.52970297029703,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.0013728226632075445,
      "loss": 0.2335,
      "num_input_tokens_seen": 33626720,
      "step": 159345
    },
    {
      "epoch": 17.53025302530253,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.001372220887496992,
      "loss": 0.2298,
      "num_input_tokens_seen": 33627744,
      "step": 159350
    },
    {
      "epoch": 17.53080308030803,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0013716192373871411,
      "loss": 0.2303,
      "num_input_tokens_seen": 33628800,
      "step": 159355
    },
    {
      "epoch": 17.53135313531353,
      "grad_norm": 0.0029144287109375,
      "learning_rate": 0.0013710177128835386,
      "loss": 0.2298,
      "num_input_tokens_seen": 33629888,
      "step": 159360
    },
    {
      "epoch": 17.53190319031903,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0013704163139917223,
      "loss": 0.2319,
      "num_input_tokens_seen": 33630944,
      "step": 159365
    },
    {
      "epoch": 17.532453245324533,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.001369815040717241,
      "loss": 0.2309,
      "num_input_tokens_seen": 33631968,
      "step": 159370
    },
    {
      "epoch": 17.533003300330034,
      "grad_norm": 0.002899169921875,
      "learning_rate": 0.0013692138930656284,
      "loss": 0.2324,
      "num_input_tokens_seen": 33633152,
      "step": 159375
    },
    {
      "epoch": 17.533553355335535,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.0013686128710424367,
      "loss": 0.2314,
      "num_input_tokens_seen": 33634176,
      "step": 159380
    },
    {
      "epoch": 17.534103410341036,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0013680119746531981,
      "loss": 0.2298,
      "num_input_tokens_seen": 33635168,
      "step": 159385
    },
    {
      "epoch": 17.534653465346533,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.0013674112039034501,
      "loss": 0.2329,
      "num_input_tokens_seen": 33636224,
      "step": 159390
    },
    {
      "epoch": 17.535203520352034,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.001366810558798735,
      "loss": 0.2309,
      "num_input_tokens_seen": 33637312,
      "step": 159395
    },
    {
      "epoch": 17.535753575357536,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0013662100393445813,
      "loss": 0.2319,
      "num_input_tokens_seen": 33638368,
      "step": 159400
    },
    {
      "epoch": 17.536303630363037,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0013656096455465282,
      "loss": 0.2309,
      "num_input_tokens_seen": 33639488,
      "step": 159405
    },
    {
      "epoch": 17.536853685368538,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.0013650093774101113,
      "loss": 0.2335,
      "num_input_tokens_seen": 33640512,
      "step": 159410
    },
    {
      "epoch": 17.53740374037404,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.0013644092349408576,
      "loss": 0.2308,
      "num_input_tokens_seen": 33641568,
      "step": 159415
    },
    {
      "epoch": 17.537953795379536,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.0013638092181443029,
      "loss": 0.2314,
      "num_input_tokens_seen": 33642624,
      "step": 159420
    },
    {
      "epoch": 17.538503850385037,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.0013632093270259725,
      "loss": 0.2303,
      "num_input_tokens_seen": 33643712,
      "step": 159425
    },
    {
      "epoch": 17.53905390539054,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0013626095615914024,
      "loss": 0.2303,
      "num_input_tokens_seen": 33644768,
      "step": 159430
    },
    {
      "epoch": 17.53960396039604,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 0.0013620099218461111,
      "loss": 0.2324,
      "num_input_tokens_seen": 33645792,
      "step": 159435
    },
    {
      "epoch": 17.54015401540154,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.001361410407795633,
      "loss": 0.2304,
      "num_input_tokens_seen": 33646784,
      "step": 159440
    },
    {
      "epoch": 17.540704070407042,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0013608110194454914,
      "loss": 0.2324,
      "num_input_tokens_seen": 33647840,
      "step": 159445
    },
    {
      "epoch": 17.541254125412543,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0013602117568012105,
      "loss": 0.2298,
      "num_input_tokens_seen": 33648960,
      "step": 159450
    },
    {
      "epoch": 17.54180418041804,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.0013596126198683095,
      "loss": 0.2308,
      "num_input_tokens_seen": 33650048,
      "step": 159455
    },
    {
      "epoch": 17.54235423542354,
      "grad_norm": 0.000705718994140625,
      "learning_rate": 0.0013590136086523152,
      "loss": 0.2309,
      "num_input_tokens_seen": 33651104,
      "step": 159460
    },
    {
      "epoch": 17.542904290429043,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.001358414723158745,
      "loss": 0.2303,
      "num_input_tokens_seen": 33652160,
      "step": 159465
    },
    {
      "epoch": 17.543454345434544,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0013578159633931247,
      "loss": 0.2319,
      "num_input_tokens_seen": 33653280,
      "step": 159470
    },
    {
      "epoch": 17.544004400440045,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.001357217329360968,
      "loss": 0.2314,
      "num_input_tokens_seen": 33654304,
      "step": 159475
    },
    {
      "epoch": 17.544554455445546,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0013566188210677903,
      "loss": 0.2319,
      "num_input_tokens_seen": 33655424,
      "step": 159480
    },
    {
      "epoch": 17.545104510451043,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.0013560204385191126,
      "loss": 0.2298,
      "num_input_tokens_seen": 33656480,
      "step": 159485
    },
    {
      "epoch": 17.545654565456545,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0013554221817204436,
      "loss": 0.2308,
      "num_input_tokens_seen": 33657472,
      "step": 159490
    },
    {
      "epoch": 17.546204620462046,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.0013548240506773023,
      "loss": 0.2314,
      "num_input_tokens_seen": 33658528,
      "step": 159495
    },
    {
      "epoch": 17.546754675467547,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.0013542260453952026,
      "loss": 0.2308,
      "num_input_tokens_seen": 33659520,
      "step": 159500
    },
    {
      "epoch": 17.547304730473048,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0013536281658796518,
      "loss": 0.2293,
      "num_input_tokens_seen": 33660576,
      "step": 159505
    },
    {
      "epoch": 17.54785478547855,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0013530304121361636,
      "loss": 0.2314,
      "num_input_tokens_seen": 33661632,
      "step": 159510
    },
    {
      "epoch": 17.54840484048405,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0013524327841702437,
      "loss": 0.2319,
      "num_input_tokens_seen": 33662688,
      "step": 159515
    },
    {
      "epoch": 17.548954895489548,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0013518352819874013,
      "loss": 0.2319,
      "num_input_tokens_seen": 33663776,
      "step": 159520
    },
    {
      "epoch": 17.54950495049505,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0013512379055931466,
      "loss": 0.2329,
      "num_input_tokens_seen": 33664896,
      "step": 159525
    },
    {
      "epoch": 17.55005500550055,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0013506406549929806,
      "loss": 0.2324,
      "num_input_tokens_seen": 33666016,
      "step": 159530
    },
    {
      "epoch": 17.55060506050605,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0013500435301924134,
      "loss": 0.2319,
      "num_input_tokens_seen": 33667072,
      "step": 159535
    },
    {
      "epoch": 17.551155115511552,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0013494465311969428,
      "loss": 0.2319,
      "num_input_tokens_seen": 33668096,
      "step": 159540
    },
    {
      "epoch": 17.551705170517053,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.001348849658012074,
      "loss": 0.2324,
      "num_input_tokens_seen": 33669152,
      "step": 159545
    },
    {
      "epoch": 17.55225522552255,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.001348252910643306,
      "loss": 0.2329,
      "num_input_tokens_seen": 33670176,
      "step": 159550
    },
    {
      "epoch": 17.55280528052805,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0013476562890961396,
      "loss": 0.2303,
      "num_input_tokens_seen": 33671232,
      "step": 159555
    },
    {
      "epoch": 17.553355335533553,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0013470597933760787,
      "loss": 0.2314,
      "num_input_tokens_seen": 33672320,
      "step": 159560
    },
    {
      "epoch": 17.553905390539054,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0013464634234886118,
      "loss": 0.2303,
      "num_input_tokens_seen": 33673408,
      "step": 159565
    },
    {
      "epoch": 17.554455445544555,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0013458671794392418,
      "loss": 0.2303,
      "num_input_tokens_seen": 33674464,
      "step": 159570
    },
    {
      "epoch": 17.555005500550056,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0013452710612334607,
      "loss": 0.2303,
      "num_input_tokens_seen": 33675488,
      "step": 159575
    },
    {
      "epoch": 17.555555555555557,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.001344675068876764,
      "loss": 0.2308,
      "num_input_tokens_seen": 33676608,
      "step": 159580
    },
    {
      "epoch": 17.556105610561055,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.001344079202374649,
      "loss": 0.2303,
      "num_input_tokens_seen": 33677664,
      "step": 159585
    },
    {
      "epoch": 17.556655665566556,
      "grad_norm": 0.0003910064697265625,
      "learning_rate": 0.001343483461732598,
      "loss": 0.2324,
      "num_input_tokens_seen": 33678720,
      "step": 159590
    },
    {
      "epoch": 17.557205720572057,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.00134288784695611,
      "loss": 0.2303,
      "num_input_tokens_seen": 33679840,
      "step": 159595
    },
    {
      "epoch": 17.557755775577558,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0013422923580506724,
      "loss": 0.2303,
      "num_input_tokens_seen": 33680960,
      "step": 159600
    },
    {
      "epoch": 17.55830583058306,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0013416969950217706,
      "loss": 0.2314,
      "num_input_tokens_seen": 33682080,
      "step": 159605
    },
    {
      "epoch": 17.55885588558856,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.0013411017578748917,
      "loss": 0.2314,
      "num_input_tokens_seen": 33683168,
      "step": 159610
    },
    {
      "epoch": 17.55940594059406,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0013405066466155252,
      "loss": 0.2298,
      "num_input_tokens_seen": 33684192,
      "step": 159615
    },
    {
      "epoch": 17.55995599559956,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.0013399116612491562,
      "loss": 0.2314,
      "num_input_tokens_seen": 33685248,
      "step": 159620
    },
    {
      "epoch": 17.56050605060506,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0013393168017812674,
      "loss": 0.2335,
      "num_input_tokens_seen": 33686240,
      "step": 159625
    },
    {
      "epoch": 17.56105610561056,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0013387220682173373,
      "loss": 0.2303,
      "num_input_tokens_seen": 33687296,
      "step": 159630
    },
    {
      "epoch": 17.561606160616062,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0013381274605628502,
      "loss": 0.2303,
      "num_input_tokens_seen": 33688352,
      "step": 159635
    },
    {
      "epoch": 17.562156215621563,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0013375329788232898,
      "loss": 0.235,
      "num_input_tokens_seen": 33689472,
      "step": 159640
    },
    {
      "epoch": 17.562706270627064,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0013369386230041302,
      "loss": 0.2319,
      "num_input_tokens_seen": 33690560,
      "step": 159645
    },
    {
      "epoch": 17.563256325632562,
      "grad_norm": 0.0004444122314453125,
      "learning_rate": 0.001336344393110852,
      "loss": 0.2335,
      "num_input_tokens_seen": 33691616,
      "step": 159650
    },
    {
      "epoch": 17.563806380638063,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.001335750289148929,
      "loss": 0.2298,
      "num_input_tokens_seen": 33692640,
      "step": 159655
    },
    {
      "epoch": 17.564356435643564,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0013351563111238423,
      "loss": 0.2308,
      "num_input_tokens_seen": 33693696,
      "step": 159660
    },
    {
      "epoch": 17.564906490649065,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.0013345624590410604,
      "loss": 0.2293,
      "num_input_tokens_seen": 33694752,
      "step": 159665
    },
    {
      "epoch": 17.565456545654566,
      "grad_norm": 0.000507354736328125,
      "learning_rate": 0.001333968732906059,
      "loss": 0.2324,
      "num_input_tokens_seen": 33695776,
      "step": 159670
    },
    {
      "epoch": 17.566006600660067,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.001333375132724312,
      "loss": 0.2334,
      "num_input_tokens_seen": 33696832,
      "step": 159675
    },
    {
      "epoch": 17.566556655665565,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0013327816585012868,
      "loss": 0.2293,
      "num_input_tokens_seen": 33697920,
      "step": 159680
    },
    {
      "epoch": 17.567106710671066,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.0013321883102424558,
      "loss": 0.2324,
      "num_input_tokens_seen": 33698976,
      "step": 159685
    },
    {
      "epoch": 17.567656765676567,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0013315950879532862,
      "loss": 0.2329,
      "num_input_tokens_seen": 33700000,
      "step": 159690
    },
    {
      "epoch": 17.568206820682068,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0013310019916392452,
      "loss": 0.2324,
      "num_input_tokens_seen": 33701120,
      "step": 159695
    },
    {
      "epoch": 17.56875687568757,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.0013304090213058016,
      "loss": 0.2308,
      "num_input_tokens_seen": 33702176,
      "step": 159700
    },
    {
      "epoch": 17.56930693069307,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.0013298161769584166,
      "loss": 0.2309,
      "num_input_tokens_seen": 33703232,
      "step": 159705
    },
    {
      "epoch": 17.56985698569857,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0013292234586025585,
      "loss": 0.2329,
      "num_input_tokens_seen": 33704320,
      "step": 159710
    },
    {
      "epoch": 17.57040704070407,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.0013286308662436885,
      "loss": 0.2314,
      "num_input_tokens_seen": 33705344,
      "step": 159715
    },
    {
      "epoch": 17.57095709570957,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0013280383998872651,
      "loss": 0.2298,
      "num_input_tokens_seen": 33706368,
      "step": 159720
    },
    {
      "epoch": 17.57150715071507,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.001327446059538751,
      "loss": 0.2319,
      "num_input_tokens_seen": 33707424,
      "step": 159725
    },
    {
      "epoch": 17.572057205720572,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0013268538452036066,
      "loss": 0.2319,
      "num_input_tokens_seen": 33708480,
      "step": 159730
    },
    {
      "epoch": 17.572607260726073,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0013262617568872907,
      "loss": 0.2324,
      "num_input_tokens_seen": 33709536,
      "step": 159735
    },
    {
      "epoch": 17.573157315731574,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0013256697945952611,
      "loss": 0.2309,
      "num_input_tokens_seen": 33710592,
      "step": 159740
    },
    {
      "epoch": 17.573707370737075,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0013250779583329664,
      "loss": 0.2314,
      "num_input_tokens_seen": 33711616,
      "step": 159745
    },
    {
      "epoch": 17.574257425742573,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0013244862481058672,
      "loss": 0.2309,
      "num_input_tokens_seen": 33712672,
      "step": 159750
    },
    {
      "epoch": 17.574807480748074,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.001323894663919416,
      "loss": 0.2314,
      "num_input_tokens_seen": 33713696,
      "step": 159755
    },
    {
      "epoch": 17.575357535753575,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0013233032057790683,
      "loss": 0.2314,
      "num_input_tokens_seen": 33714752,
      "step": 159760
    },
    {
      "epoch": 17.575907590759076,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0013227118736902715,
      "loss": 0.2303,
      "num_input_tokens_seen": 33715840,
      "step": 159765
    },
    {
      "epoch": 17.576457645764577,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0013221206676584745,
      "loss": 0.2309,
      "num_input_tokens_seen": 33716896,
      "step": 159770
    },
    {
      "epoch": 17.57700770077008,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0013215295876891297,
      "loss": 0.2329,
      "num_input_tokens_seen": 33718016,
      "step": 159775
    },
    {
      "epoch": 17.577557755775576,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.001320938633787681,
      "loss": 0.2324,
      "num_input_tokens_seen": 33719040,
      "step": 159780
    },
    {
      "epoch": 17.578107810781077,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.0013203478059595774,
      "loss": 0.2308,
      "num_input_tokens_seen": 33720032,
      "step": 159785
    },
    {
      "epoch": 17.578657865786578,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0013197571042102662,
      "loss": 0.2319,
      "num_input_tokens_seen": 33721120,
      "step": 159790
    },
    {
      "epoch": 17.57920792079208,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0013191665285451864,
      "loss": 0.2308,
      "num_input_tokens_seen": 33722176,
      "step": 159795
    },
    {
      "epoch": 17.57975797579758,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.001318576078969787,
      "loss": 0.2314,
      "num_input_tokens_seen": 33723232,
      "step": 159800
    },
    {
      "epoch": 17.58030803080308,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.0013179857554895052,
      "loss": 0.2329,
      "num_input_tokens_seen": 33724288,
      "step": 159805
    },
    {
      "epoch": 17.580858085808583,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0013173955581097818,
      "loss": 0.2314,
      "num_input_tokens_seen": 33725376,
      "step": 159810
    },
    {
      "epoch": 17.58140814081408,
      "grad_norm": 0.00069427490234375,
      "learning_rate": 0.0013168054868360607,
      "loss": 0.2298,
      "num_input_tokens_seen": 33726400,
      "step": 159815
    },
    {
      "epoch": 17.58195819581958,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0013162155416737758,
      "loss": 0.2319,
      "num_input_tokens_seen": 33727424,
      "step": 159820
    },
    {
      "epoch": 17.582508250825082,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.0013156257226283696,
      "loss": 0.2309,
      "num_input_tokens_seen": 33728448,
      "step": 159825
    },
    {
      "epoch": 17.583058305830583,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.0013150360297052726,
      "loss": 0.2329,
      "num_input_tokens_seen": 33729536,
      "step": 159830
    },
    {
      "epoch": 17.583608360836084,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0013144464629099206,
      "loss": 0.2309,
      "num_input_tokens_seen": 33730656,
      "step": 159835
    },
    {
      "epoch": 17.584158415841586,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0013138570222477474,
      "loss": 0.2314,
      "num_input_tokens_seen": 33731776,
      "step": 159840
    },
    {
      "epoch": 17.584708470847083,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0013132677077241888,
      "loss": 0.2324,
      "num_input_tokens_seen": 33732864,
      "step": 159845
    },
    {
      "epoch": 17.585258525852584,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0013126785193446754,
      "loss": 0.2324,
      "num_input_tokens_seen": 33733952,
      "step": 159850
    },
    {
      "epoch": 17.585808580858085,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0013120894571146363,
      "loss": 0.2314,
      "num_input_tokens_seen": 33734976,
      "step": 159855
    },
    {
      "epoch": 17.586358635863586,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0013115005210394986,
      "loss": 0.2319,
      "num_input_tokens_seen": 33736064,
      "step": 159860
    },
    {
      "epoch": 17.586908690869087,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0013109117111246914,
      "loss": 0.2308,
      "num_input_tokens_seen": 33737088,
      "step": 159865
    },
    {
      "epoch": 17.58745874587459,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.001310323027375642,
      "loss": 0.2314,
      "num_input_tokens_seen": 33738112,
      "step": 159870
    },
    {
      "epoch": 17.58800880088009,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.0013097344697977798,
      "loss": 0.2314,
      "num_input_tokens_seen": 33739200,
      "step": 159875
    },
    {
      "epoch": 17.588558855885587,
      "grad_norm": 0.000762939453125,
      "learning_rate": 0.0013091460383965264,
      "loss": 0.2303,
      "num_input_tokens_seen": 33740224,
      "step": 159880
    },
    {
      "epoch": 17.58910891089109,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.0013085577331773013,
      "loss": 0.2303,
      "num_input_tokens_seen": 33741312,
      "step": 159885
    },
    {
      "epoch": 17.58965896589659,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0013079695541455315,
      "loss": 0.2329,
      "num_input_tokens_seen": 33742496,
      "step": 159890
    },
    {
      "epoch": 17.59020902090209,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.0013073815013066347,
      "loss": 0.2324,
      "num_input_tokens_seen": 33743584,
      "step": 159895
    },
    {
      "epoch": 17.59075907590759,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.0013067935746660314,
      "loss": 0.2298,
      "num_input_tokens_seen": 33744640,
      "step": 159900
    },
    {
      "epoch": 17.591309130913093,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0013062057742291439,
      "loss": 0.2309,
      "num_input_tokens_seen": 33745664,
      "step": 159905
    },
    {
      "epoch": 17.59185918591859,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0013056181000013845,
      "loss": 0.2324,
      "num_input_tokens_seen": 33746752,
      "step": 159910
    },
    {
      "epoch": 17.59240924092409,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.001305030551988174,
      "loss": 0.2303,
      "num_input_tokens_seen": 33747776,
      "step": 159915
    },
    {
      "epoch": 17.592959295929592,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0013044431301949227,
      "loss": 0.2309,
      "num_input_tokens_seen": 33748832,
      "step": 159920
    },
    {
      "epoch": 17.593509350935093,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0013038558346270485,
      "loss": 0.2314,
      "num_input_tokens_seen": 33749888,
      "step": 159925
    },
    {
      "epoch": 17.594059405940595,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0013032686652899648,
      "loss": 0.2314,
      "num_input_tokens_seen": 33750976,
      "step": 159930
    },
    {
      "epoch": 17.594609460946096,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0013026816221890795,
      "loss": 0.2319,
      "num_input_tokens_seen": 33752064,
      "step": 159935
    },
    {
      "epoch": 17.595159515951597,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0013020947053298064,
      "loss": 0.2314,
      "num_input_tokens_seen": 33753088,
      "step": 159940
    },
    {
      "epoch": 17.595709570957094,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.001301507914717553,
      "loss": 0.2324,
      "num_input_tokens_seen": 33754208,
      "step": 159945
    },
    {
      "epoch": 17.596259625962595,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0013009212503577276,
      "loss": 0.2309,
      "num_input_tokens_seen": 33755264,
      "step": 159950
    },
    {
      "epoch": 17.596809680968097,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.0013003347122557368,
      "loss": 0.2304,
      "num_input_tokens_seen": 33756320,
      "step": 159955
    },
    {
      "epoch": 17.597359735973598,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.0012997483004169874,
      "loss": 0.2324,
      "num_input_tokens_seen": 33757312,
      "step": 159960
    },
    {
      "epoch": 17.5979097909791,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.0012991620148468867,
      "loss": 0.2303,
      "num_input_tokens_seen": 33758432,
      "step": 159965
    },
    {
      "epoch": 17.5984598459846,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0012985758555508326,
      "loss": 0.2303,
      "num_input_tokens_seen": 33759520,
      "step": 159970
    },
    {
      "epoch": 17.599009900990097,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0012979898225342317,
      "loss": 0.2308,
      "num_input_tokens_seen": 33760640,
      "step": 159975
    },
    {
      "epoch": 17.5995599559956,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.001297403915802482,
      "loss": 0.2335,
      "num_input_tokens_seen": 33761696,
      "step": 159980
    },
    {
      "epoch": 17.6001100110011,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0012968181353609854,
      "loss": 0.2314,
      "num_input_tokens_seen": 33762752,
      "step": 159985
    },
    {
      "epoch": 17.6006600660066,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0012962324812151426,
      "loss": 0.2309,
      "num_input_tokens_seen": 33763808,
      "step": 159990
    },
    {
      "epoch": 17.6012101210121,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0012956469533703479,
      "loss": 0.2324,
      "num_input_tokens_seen": 33764832,
      "step": 159995
    },
    {
      "epoch": 17.601760176017603,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0012950615518320001,
      "loss": 0.2324,
      "num_input_tokens_seen": 33765856,
      "step": 160000
    },
    {
      "epoch": 17.602310231023104,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0012944762766054952,
      "loss": 0.2319,
      "num_input_tokens_seen": 33766912,
      "step": 160005
    },
    {
      "epoch": 17.6028602860286,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0012938911276962233,
      "loss": 0.2298,
      "num_input_tokens_seen": 33767936,
      "step": 160010
    },
    {
      "epoch": 17.603410341034103,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.001293306105109579,
      "loss": 0.2309,
      "num_input_tokens_seen": 33768960,
      "step": 160015
    },
    {
      "epoch": 17.603960396039604,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0012927212088509592,
      "loss": 0.2298,
      "num_input_tokens_seen": 33770048,
      "step": 160020
    },
    {
      "epoch": 17.604510451045105,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0012921364389257466,
      "loss": 0.2298,
      "num_input_tokens_seen": 33771072,
      "step": 160025
    },
    {
      "epoch": 17.605060506050606,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0012915517953393384,
      "loss": 0.2293,
      "num_input_tokens_seen": 33772128,
      "step": 160030
    },
    {
      "epoch": 17.605610561056107,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.0012909672780971153,
      "loss": 0.2298,
      "num_input_tokens_seen": 33773216,
      "step": 160035
    },
    {
      "epoch": 17.606160616061608,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0012903828872044697,
      "loss": 0.2314,
      "num_input_tokens_seen": 33774304,
      "step": 160040
    },
    {
      "epoch": 17.606710671067106,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.001289798622666789,
      "loss": 0.2319,
      "num_input_tokens_seen": 33775392,
      "step": 160045
    },
    {
      "epoch": 17.607260726072607,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0012892144844894537,
      "loss": 0.2319,
      "num_input_tokens_seen": 33776480,
      "step": 160050
    },
    {
      "epoch": 17.607810781078108,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.0012886304726778513,
      "loss": 0.2314,
      "num_input_tokens_seen": 33777504,
      "step": 160055
    },
    {
      "epoch": 17.60836083608361,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.0012880465872373592,
      "loss": 0.2324,
      "num_input_tokens_seen": 33778560,
      "step": 160060
    },
    {
      "epoch": 17.60891089108911,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0012874628281733663,
      "loss": 0.2314,
      "num_input_tokens_seen": 33779616,
      "step": 160065
    },
    {
      "epoch": 17.60946094609461,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.001286879195491245,
      "loss": 0.2324,
      "num_input_tokens_seen": 33780640,
      "step": 160070
    },
    {
      "epoch": 17.61001100110011,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0012862956891963777,
      "loss": 0.2319,
      "num_input_tokens_seen": 33781696,
      "step": 160075
    },
    {
      "epoch": 17.61056105610561,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0012857123092941453,
      "loss": 0.2308,
      "num_input_tokens_seen": 33782752,
      "step": 160080
    },
    {
      "epoch": 17.61111111111111,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0012851290557899196,
      "loss": 0.2319,
      "num_input_tokens_seen": 33783776,
      "step": 160085
    },
    {
      "epoch": 17.611661166116612,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0012845459286890803,
      "loss": 0.2293,
      "num_input_tokens_seen": 33784864,
      "step": 160090
    },
    {
      "epoch": 17.612211221122113,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0012839629279970011,
      "loss": 0.2308,
      "num_input_tokens_seen": 33785856,
      "step": 160095
    },
    {
      "epoch": 17.612761276127614,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.0012833800537190476,
      "loss": 0.2314,
      "num_input_tokens_seen": 33786912,
      "step": 160100
    },
    {
      "epoch": 17.61331133113311,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001282797305860604,
      "loss": 0.2293,
      "num_input_tokens_seen": 33788000,
      "step": 160105
    },
    {
      "epoch": 17.613861386138613,
      "grad_norm": 0.0005645751953125,
      "learning_rate": 0.0012822146844270327,
      "loss": 0.2314,
      "num_input_tokens_seen": 33788992,
      "step": 160110
    },
    {
      "epoch": 17.614411441144114,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0012816321894237091,
      "loss": 0.2324,
      "num_input_tokens_seen": 33790016,
      "step": 160115
    },
    {
      "epoch": 17.614961496149615,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0012810498208559995,
      "loss": 0.2324,
      "num_input_tokens_seen": 33791040,
      "step": 160120
    },
    {
      "epoch": 17.615511551155116,
      "grad_norm": 0.002410888671875,
      "learning_rate": 0.0012804675787292657,
      "loss": 0.234,
      "num_input_tokens_seen": 33792128,
      "step": 160125
    },
    {
      "epoch": 17.616061606160617,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.0012798854630488804,
      "loss": 0.2298,
      "num_input_tokens_seen": 33793216,
      "step": 160130
    },
    {
      "epoch": 17.616611661166118,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0012793034738202075,
      "loss": 0.233,
      "num_input_tokens_seen": 33794336,
      "step": 160135
    },
    {
      "epoch": 17.617161716171616,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0012787216110486143,
      "loss": 0.2329,
      "num_input_tokens_seen": 33795360,
      "step": 160140
    },
    {
      "epoch": 17.617711771177117,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0012781398747394584,
      "loss": 0.2329,
      "num_input_tokens_seen": 33796448,
      "step": 160145
    },
    {
      "epoch": 17.618261826182618,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0012775582648981005,
      "loss": 0.2309,
      "num_input_tokens_seen": 33797504,
      "step": 160150
    },
    {
      "epoch": 17.61881188118812,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0012769767815299059,
      "loss": 0.2308,
      "num_input_tokens_seen": 33798624,
      "step": 160155
    },
    {
      "epoch": 17.61936193619362,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.0012763954246402258,
      "loss": 0.2303,
      "num_input_tokens_seen": 33799712,
      "step": 160160
    },
    {
      "epoch": 17.61991199119912,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0012758141942344307,
      "loss": 0.2314,
      "num_input_tokens_seen": 33800800,
      "step": 160165
    },
    {
      "epoch": 17.620462046204622,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00127523309031787,
      "loss": 0.2298,
      "num_input_tokens_seen": 33801952,
      "step": 160170
    },
    {
      "epoch": 17.62101210121012,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001274652112895897,
      "loss": 0.2324,
      "num_input_tokens_seen": 33803072,
      "step": 160175
    },
    {
      "epoch": 17.62156215621562,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0012740712619738713,
      "loss": 0.2314,
      "num_input_tokens_seen": 33804160,
      "step": 160180
    },
    {
      "epoch": 17.622112211221122,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.001273490537557142,
      "loss": 0.2298,
      "num_input_tokens_seen": 33805152,
      "step": 160185
    },
    {
      "epoch": 17.622662266226623,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.0012729099396510644,
      "loss": 0.2314,
      "num_input_tokens_seen": 33806208,
      "step": 160190
    },
    {
      "epoch": 17.623212321232124,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0012723294682609931,
      "loss": 0.2308,
      "num_input_tokens_seen": 33807296,
      "step": 160195
    },
    {
      "epoch": 17.623762376237625,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0012717491233922684,
      "loss": 0.2335,
      "num_input_tokens_seen": 33808384,
      "step": 160200
    },
    {
      "epoch": 17.624312431243123,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.0012711689050502494,
      "loss": 0.2329,
      "num_input_tokens_seen": 33809408,
      "step": 160205
    },
    {
      "epoch": 17.624862486248624,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.0012705888132402753,
      "loss": 0.2324,
      "num_input_tokens_seen": 33810560,
      "step": 160210
    },
    {
      "epoch": 17.625412541254125,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0012700088479676984,
      "loss": 0.2308,
      "num_input_tokens_seen": 33811584,
      "step": 160215
    },
    {
      "epoch": 17.625962596259626,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.0012694290092378595,
      "loss": 0.2298,
      "num_input_tokens_seen": 33812608,
      "step": 160220
    },
    {
      "epoch": 17.626512651265127,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.001268849297056106,
      "loss": 0.2314,
      "num_input_tokens_seen": 33813760,
      "step": 160225
    },
    {
      "epoch": 17.627062706270628,
      "grad_norm": 0.000843048095703125,
      "learning_rate": 0.0012682697114277802,
      "loss": 0.2324,
      "num_input_tokens_seen": 33814816,
      "step": 160230
    },
    {
      "epoch": 17.62761276127613,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0012676902523582245,
      "loss": 0.2319,
      "num_input_tokens_seen": 33815936,
      "step": 160235
    },
    {
      "epoch": 17.628162816281627,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0012671109198527763,
      "loss": 0.2303,
      "num_input_tokens_seen": 33817024,
      "step": 160240
    },
    {
      "epoch": 17.628712871287128,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0012665317139167764,
      "loss": 0.2313,
      "num_input_tokens_seen": 33818016,
      "step": 160245
    },
    {
      "epoch": 17.62926292629263,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.0012659526345555638,
      "loss": 0.2319,
      "num_input_tokens_seen": 33819168,
      "step": 160250
    },
    {
      "epoch": 17.62981298129813,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0012653736817744776,
      "loss": 0.2309,
      "num_input_tokens_seen": 33820224,
      "step": 160255
    },
    {
      "epoch": 17.63036303630363,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0012647948555788518,
      "loss": 0.2319,
      "num_input_tokens_seen": 33821248,
      "step": 160260
    },
    {
      "epoch": 17.630913091309132,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.001264216155974019,
      "loss": 0.2319,
      "num_input_tokens_seen": 33822272,
      "step": 160265
    },
    {
      "epoch": 17.63146314631463,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0012636375829653162,
      "loss": 0.2308,
      "num_input_tokens_seen": 33823296,
      "step": 160270
    },
    {
      "epoch": 17.63201320132013,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0012630591365580728,
      "loss": 0.2324,
      "num_input_tokens_seen": 33824416,
      "step": 160275
    },
    {
      "epoch": 17.632563256325632,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.001262480816757623,
      "loss": 0.2303,
      "num_input_tokens_seen": 33825504,
      "step": 160280
    },
    {
      "epoch": 17.633113311331133,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.001261902623569297,
      "loss": 0.2309,
      "num_input_tokens_seen": 33826528,
      "step": 160285
    },
    {
      "epoch": 17.633663366336634,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0012613245569984193,
      "loss": 0.2309,
      "num_input_tokens_seen": 33827584,
      "step": 160290
    },
    {
      "epoch": 17.634213421342135,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.001260746617050324,
      "loss": 0.2319,
      "num_input_tokens_seen": 33828608,
      "step": 160295
    },
    {
      "epoch": 17.634763476347636,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0012601688037303298,
      "loss": 0.2314,
      "num_input_tokens_seen": 33829632,
      "step": 160300
    },
    {
      "epoch": 17.635313531353134,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.0012595911170437694,
      "loss": 0.2303,
      "num_input_tokens_seen": 33830688,
      "step": 160305
    },
    {
      "epoch": 17.635863586358635,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0012590135569959654,
      "loss": 0.2298,
      "num_input_tokens_seen": 33831776,
      "step": 160310
    },
    {
      "epoch": 17.636413641364136,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0012584361235922382,
      "loss": 0.2324,
      "num_input_tokens_seen": 33832864,
      "step": 160315
    },
    {
      "epoch": 17.636963696369637,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0012578588168379134,
      "loss": 0.2314,
      "num_input_tokens_seen": 33833920,
      "step": 160320
    },
    {
      "epoch": 17.63751375137514,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0012572816367383071,
      "loss": 0.2319,
      "num_input_tokens_seen": 33834944,
      "step": 160325
    },
    {
      "epoch": 17.63806380638064,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.001256704583298745,
      "loss": 0.2298,
      "num_input_tokens_seen": 33835968,
      "step": 160330
    },
    {
      "epoch": 17.638613861386137,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0012561276565245393,
      "loss": 0.2314,
      "num_input_tokens_seen": 33836992,
      "step": 160335
    },
    {
      "epoch": 17.639163916391638,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0012555508564210094,
      "loss": 0.2303,
      "num_input_tokens_seen": 33838048,
      "step": 160340
    },
    {
      "epoch": 17.63971397139714,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.001254974182993474,
      "loss": 0.2319,
      "num_input_tokens_seen": 33839072,
      "step": 160345
    },
    {
      "epoch": 17.64026402640264,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.001254397636247244,
      "loss": 0.2314,
      "num_input_tokens_seen": 33840160,
      "step": 160350
    },
    {
      "epoch": 17.64081408140814,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0012538212161876388,
      "loss": 0.2298,
      "num_input_tokens_seen": 33841184,
      "step": 160355
    },
    {
      "epoch": 17.641364136413642,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0012532449228199638,
      "loss": 0.2309,
      "num_input_tokens_seen": 33842240,
      "step": 160360
    },
    {
      "epoch": 17.641914191419144,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0012526687561495347,
      "loss": 0.2324,
      "num_input_tokens_seen": 33843296,
      "step": 160365
    },
    {
      "epoch": 17.64246424642464,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0012520927161816641,
      "loss": 0.2319,
      "num_input_tokens_seen": 33844384,
      "step": 160370
    },
    {
      "epoch": 17.643014301430142,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0012515168029216543,
      "loss": 0.2309,
      "num_input_tokens_seen": 33845440,
      "step": 160375
    },
    {
      "epoch": 17.643564356435643,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0012509410163748197,
      "loss": 0.2303,
      "num_input_tokens_seen": 33846432,
      "step": 160380
    },
    {
      "epoch": 17.644114411441144,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0012503653565464638,
      "loss": 0.2314,
      "num_input_tokens_seen": 33847520,
      "step": 160385
    },
    {
      "epoch": 17.644664466446645,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0012497898234418913,
      "loss": 0.2309,
      "num_input_tokens_seen": 33848576,
      "step": 160390
    },
    {
      "epoch": 17.645214521452147,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0012492144170664077,
      "loss": 0.2319,
      "num_input_tokens_seen": 33849632,
      "step": 160395
    },
    {
      "epoch": 17.645764576457644,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.001248639137425317,
      "loss": 0.2314,
      "num_input_tokens_seen": 33850656,
      "step": 160400
    },
    {
      "epoch": 17.646314631463145,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.0012480639845239216,
      "loss": 0.2308,
      "num_input_tokens_seen": 33851680,
      "step": 160405
    },
    {
      "epoch": 17.646864686468646,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0012474889583675242,
      "loss": 0.2303,
      "num_input_tokens_seen": 33852800,
      "step": 160410
    },
    {
      "epoch": 17.647414741474147,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.001246914058961417,
      "loss": 0.2345,
      "num_input_tokens_seen": 33853792,
      "step": 160415
    },
    {
      "epoch": 17.64796479647965,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.001246339286310904,
      "loss": 0.2319,
      "num_input_tokens_seen": 33854848,
      "step": 160420
    },
    {
      "epoch": 17.64851485148515,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.0012457646404212862,
      "loss": 0.2314,
      "num_input_tokens_seen": 33855904,
      "step": 160425
    },
    {
      "epoch": 17.64906490649065,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0012451901212978528,
      "loss": 0.2308,
      "num_input_tokens_seen": 33856960,
      "step": 160430
    },
    {
      "epoch": 17.649614961496148,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0012446157289459025,
      "loss": 0.2314,
      "num_input_tokens_seen": 33858080,
      "step": 160435
    },
    {
      "epoch": 17.65016501650165,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.001244041463370728,
      "loss": 0.2314,
      "num_input_tokens_seen": 33859136,
      "step": 160440
    },
    {
      "epoch": 17.65071507150715,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.001243467324577625,
      "loss": 0.2298,
      "num_input_tokens_seen": 33860224,
      "step": 160445
    },
    {
      "epoch": 17.65126512651265,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0012428933125718793,
      "loss": 0.2324,
      "num_input_tokens_seen": 33861280,
      "step": 160450
    },
    {
      "epoch": 17.651815181518153,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.001242319427358785,
      "loss": 0.2304,
      "num_input_tokens_seen": 33862304,
      "step": 160455
    },
    {
      "epoch": 17.652365236523654,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.001241745668943633,
      "loss": 0.2324,
      "num_input_tokens_seen": 33863296,
      "step": 160460
    },
    {
      "epoch": 17.652915291529155,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.001241172037331707,
      "loss": 0.2303,
      "num_input_tokens_seen": 33864384,
      "step": 160465
    },
    {
      "epoch": 17.653465346534652,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.0012405985325282998,
      "loss": 0.2319,
      "num_input_tokens_seen": 33865408,
      "step": 160470
    },
    {
      "epoch": 17.654015401540153,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0012400251545386903,
      "loss": 0.2308,
      "num_input_tokens_seen": 33866432,
      "step": 160475
    },
    {
      "epoch": 17.654565456545654,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.001239451903368166,
      "loss": 0.2314,
      "num_input_tokens_seen": 33867520,
      "step": 160480
    },
    {
      "epoch": 17.655115511551156,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0012388787790220128,
      "loss": 0.2298,
      "num_input_tokens_seen": 33868576,
      "step": 160485
    },
    {
      "epoch": 17.655665566556657,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.0012383057815055082,
      "loss": 0.2303,
      "num_input_tokens_seen": 33869664,
      "step": 160490
    },
    {
      "epoch": 17.656215621562158,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.0012377329108239393,
      "loss": 0.2308,
      "num_input_tokens_seen": 33870688,
      "step": 160495
    },
    {
      "epoch": 17.656765676567655,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.001237160166982582,
      "loss": 0.2314,
      "num_input_tokens_seen": 33871776,
      "step": 160500
    },
    {
      "epoch": 17.657315731573156,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0012365875499867124,
      "loss": 0.2319,
      "num_input_tokens_seen": 33872896,
      "step": 160505
    },
    {
      "epoch": 17.657865786578657,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.0012360150598416109,
      "loss": 0.2309,
      "num_input_tokens_seen": 33874016,
      "step": 160510
    },
    {
      "epoch": 17.65841584158416,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.001235442696552555,
      "loss": 0.2319,
      "num_input_tokens_seen": 33875104,
      "step": 160515
    },
    {
      "epoch": 17.65896589658966,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.0012348704601248205,
      "loss": 0.2324,
      "num_input_tokens_seen": 33876160,
      "step": 160520
    },
    {
      "epoch": 17.65951595159516,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00123429835056368,
      "loss": 0.2303,
      "num_input_tokens_seen": 33877184,
      "step": 160525
    },
    {
      "epoch": 17.66006600660066,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0012337263678744025,
      "loss": 0.2309,
      "num_input_tokens_seen": 33878240,
      "step": 160530
    },
    {
      "epoch": 17.66061606160616,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0012331545120622656,
      "loss": 0.2303,
      "num_input_tokens_seen": 33879232,
      "step": 160535
    },
    {
      "epoch": 17.66116611661166,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0012325827831325363,
      "loss": 0.2335,
      "num_input_tokens_seen": 33880320,
      "step": 160540
    },
    {
      "epoch": 17.66171617161716,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0012320111810904877,
      "loss": 0.2314,
      "num_input_tokens_seen": 33881376,
      "step": 160545
    },
    {
      "epoch": 17.662266226622663,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0012314397059413867,
      "loss": 0.2319,
      "num_input_tokens_seen": 33882432,
      "step": 160550
    },
    {
      "epoch": 17.662816281628164,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0012308683576904944,
      "loss": 0.2329,
      "num_input_tokens_seen": 33883456,
      "step": 160555
    },
    {
      "epoch": 17.663366336633665,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0012302971363430863,
      "loss": 0.2303,
      "num_input_tokens_seen": 33884480,
      "step": 160560
    },
    {
      "epoch": 17.663916391639162,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0012297260419044187,
      "loss": 0.2313,
      "num_input_tokens_seen": 33885600,
      "step": 160565
    },
    {
      "epoch": 17.664466446644663,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.001229155074379757,
      "loss": 0.2298,
      "num_input_tokens_seen": 33886656,
      "step": 160570
    },
    {
      "epoch": 17.665016501650165,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.00122858423377437,
      "loss": 0.2319,
      "num_input_tokens_seen": 33887744,
      "step": 160575
    },
    {
      "epoch": 17.665566556655666,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.0012280135200935093,
      "loss": 0.2298,
      "num_input_tokens_seen": 33888736,
      "step": 160580
    },
    {
      "epoch": 17.666116611661167,
      "grad_norm": 0.0027923583984375,
      "learning_rate": 0.0012274429333424435,
      "loss": 0.2324,
      "num_input_tokens_seen": 33889792,
      "step": 160585
    },
    {
      "epoch": 17.666666666666668,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0012268724735264253,
      "loss": 0.2308,
      "num_input_tokens_seen": 33890912,
      "step": 160590
    },
    {
      "epoch": 17.66721672167217,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.0012263021406507134,
      "loss": 0.2304,
      "num_input_tokens_seen": 33892032,
      "step": 160595
    },
    {
      "epoch": 17.667766776677666,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.001225731934720569,
      "loss": 0.2298,
      "num_input_tokens_seen": 33893024,
      "step": 160600
    },
    {
      "epoch": 17.668316831683168,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0012251618557412396,
      "loss": 0.2308,
      "num_input_tokens_seen": 33894016,
      "step": 160605
    },
    {
      "epoch": 17.66886688668867,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0012245919037179875,
      "loss": 0.2314,
      "num_input_tokens_seen": 33895136,
      "step": 160610
    },
    {
      "epoch": 17.66941694169417,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0012240220786560619,
      "loss": 0.2308,
      "num_input_tokens_seen": 33896160,
      "step": 160615
    },
    {
      "epoch": 17.66996699669967,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0012234523805607117,
      "loss": 0.2303,
      "num_input_tokens_seen": 33897184,
      "step": 160620
    },
    {
      "epoch": 17.670517051705172,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.0012228828094371896,
      "loss": 0.2319,
      "num_input_tokens_seen": 33898208,
      "step": 160625
    },
    {
      "epoch": 17.67106710671067,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.0012223133652907463,
      "loss": 0.2308,
      "num_input_tokens_seen": 33899232,
      "step": 160630
    },
    {
      "epoch": 17.67161716171617,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0012217440481266312,
      "loss": 0.2324,
      "num_input_tokens_seen": 33900288,
      "step": 160635
    },
    {
      "epoch": 17.67216721672167,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0012211748579500914,
      "loss": 0.2308,
      "num_input_tokens_seen": 33901312,
      "step": 160640
    },
    {
      "epoch": 17.672717271727173,
      "grad_norm": 0.000652313232421875,
      "learning_rate": 0.0012206057947663678,
      "loss": 0.2324,
      "num_input_tokens_seen": 33902272,
      "step": 160645
    },
    {
      "epoch": 17.673267326732674,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.0012200368585807081,
      "loss": 0.2303,
      "num_input_tokens_seen": 33903328,
      "step": 160650
    },
    {
      "epoch": 17.673817381738175,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.001219468049398356,
      "loss": 0.2319,
      "num_input_tokens_seen": 33904416,
      "step": 160655
    },
    {
      "epoch": 17.674367436743676,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.001218899367224558,
      "loss": 0.2329,
      "num_input_tokens_seen": 33905504,
      "step": 160660
    },
    {
      "epoch": 17.674917491749174,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.0012183308120645524,
      "loss": 0.2293,
      "num_input_tokens_seen": 33906592,
      "step": 160665
    },
    {
      "epoch": 17.675467546754675,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.001217762383923574,
      "loss": 0.2303,
      "num_input_tokens_seen": 33907680,
      "step": 160670
    },
    {
      "epoch": 17.676017601760176,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00121719408280687,
      "loss": 0.2324,
      "num_input_tokens_seen": 33908768,
      "step": 160675
    },
    {
      "epoch": 17.676567656765677,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.001216625908719673,
      "loss": 0.2308,
      "num_input_tokens_seen": 33909824,
      "step": 160680
    },
    {
      "epoch": 17.677117711771178,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.0012160578616672202,
      "loss": 0.2319,
      "num_input_tokens_seen": 33910976,
      "step": 160685
    },
    {
      "epoch": 17.67766776677668,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0012154899416547493,
      "loss": 0.2298,
      "num_input_tokens_seen": 33912064,
      "step": 160690
    },
    {
      "epoch": 17.678217821782177,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.0012149221486874929,
      "loss": 0.2319,
      "num_input_tokens_seen": 33913152,
      "step": 160695
    },
    {
      "epoch": 17.678767876787678,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.0012143544827706847,
      "loss": 0.2314,
      "num_input_tokens_seen": 33914144,
      "step": 160700
    },
    {
      "epoch": 17.67931793179318,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0012137869439095544,
      "loss": 0.2335,
      "num_input_tokens_seen": 33915200,
      "step": 160705
    },
    {
      "epoch": 17.67986798679868,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0012132195321093357,
      "loss": 0.2314,
      "num_input_tokens_seen": 33916352,
      "step": 160710
    },
    {
      "epoch": 17.68041804180418,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0012126522473752581,
      "loss": 0.2319,
      "num_input_tokens_seen": 33917472,
      "step": 160715
    },
    {
      "epoch": 17.680968096809682,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0012120850897125474,
      "loss": 0.2314,
      "num_input_tokens_seen": 33918496,
      "step": 160720
    },
    {
      "epoch": 17.681518151815183,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.001211518059126434,
      "loss": 0.2308,
      "num_input_tokens_seen": 33919552,
      "step": 160725
    },
    {
      "epoch": 17.68206820682068,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.0012109511556221409,
      "loss": 0.2314,
      "num_input_tokens_seen": 33920576,
      "step": 160730
    },
    {
      "epoch": 17.682618261826182,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.001210384379204895,
      "loss": 0.2319,
      "num_input_tokens_seen": 33921600,
      "step": 160735
    },
    {
      "epoch": 17.683168316831683,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0012098177298799178,
      "loss": 0.2319,
      "num_input_tokens_seen": 33922688,
      "step": 160740
    },
    {
      "epoch": 17.683718371837184,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0012092512076524347,
      "loss": 0.2319,
      "num_input_tokens_seen": 33923776,
      "step": 160745
    },
    {
      "epoch": 17.684268426842685,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0012086848125276667,
      "loss": 0.2324,
      "num_input_tokens_seen": 33924768,
      "step": 160750
    },
    {
      "epoch": 17.684818481848186,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.0012081185445108312,
      "loss": 0.2298,
      "num_input_tokens_seen": 33925856,
      "step": 160755
    },
    {
      "epoch": 17.685368536853684,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0012075524036071505,
      "loss": 0.2298,
      "num_input_tokens_seen": 33926848,
      "step": 160760
    },
    {
      "epoch": 17.685918591859185,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.0012069863898218425,
      "loss": 0.2314,
      "num_input_tokens_seen": 33927936,
      "step": 160765
    },
    {
      "epoch": 17.686468646864686,
      "grad_norm": 0.0029754638671875,
      "learning_rate": 0.0012064205031601177,
      "loss": 0.2298,
      "num_input_tokens_seen": 33928992,
      "step": 160770
    },
    {
      "epoch": 17.687018701870187,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0012058547436272005,
      "loss": 0.2288,
      "num_input_tokens_seen": 33930112,
      "step": 160775
    },
    {
      "epoch": 17.687568756875688,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.001205289111228298,
      "loss": 0.2298,
      "num_input_tokens_seen": 33931168,
      "step": 160780
    },
    {
      "epoch": 17.68811881188119,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00120472360596863,
      "loss": 0.233,
      "num_input_tokens_seen": 33932288,
      "step": 160785
    },
    {
      "epoch": 17.68866886688669,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.0012041582278534036,
      "loss": 0.2324,
      "num_input_tokens_seen": 33933408,
      "step": 160790
    },
    {
      "epoch": 17.689218921892188,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0012035929768878294,
      "loss": 0.2303,
      "num_input_tokens_seen": 33934464,
      "step": 160795
    },
    {
      "epoch": 17.68976897689769,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0012030278530771187,
      "loss": 0.2309,
      "num_input_tokens_seen": 33935488,
      "step": 160800
    },
    {
      "epoch": 17.69031903190319,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.001202462856426482,
      "loss": 0.2314,
      "num_input_tokens_seen": 33936576,
      "step": 160805
    },
    {
      "epoch": 17.69086908690869,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0012018979869411223,
      "loss": 0.2309,
      "num_input_tokens_seen": 33937632,
      "step": 160810
    },
    {
      "epoch": 17.691419141914192,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.0012013332446262498,
      "loss": 0.2288,
      "num_input_tokens_seen": 33938656,
      "step": 160815
    },
    {
      "epoch": 17.691969196919693,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0012007686294870656,
      "loss": 0.2303,
      "num_input_tokens_seen": 33939680,
      "step": 160820
    },
    {
      "epoch": 17.69251925192519,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.0012002041415287755,
      "loss": 0.2319,
      "num_input_tokens_seen": 33940800,
      "step": 160825
    },
    {
      "epoch": 17.693069306930692,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.001199639780756584,
      "loss": 0.2319,
      "num_input_tokens_seen": 33941888,
      "step": 160830
    },
    {
      "epoch": 17.693619361936193,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.001199075547175688,
      "loss": 0.2309,
      "num_input_tokens_seen": 33942944,
      "step": 160835
    },
    {
      "epoch": 17.694169416941694,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.0011985114407912921,
      "loss": 0.2303,
      "num_input_tokens_seen": 33944000,
      "step": 160840
    },
    {
      "epoch": 17.694719471947195,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0011979474616085923,
      "loss": 0.2319,
      "num_input_tokens_seen": 33945120,
      "step": 160845
    },
    {
      "epoch": 17.695269526952696,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.0011973836096327904,
      "loss": 0.2314,
      "num_input_tokens_seen": 33946176,
      "step": 160850
    },
    {
      "epoch": 17.695819581958197,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.001196819884869078,
      "loss": 0.2319,
      "num_input_tokens_seen": 33947296,
      "step": 160855
    },
    {
      "epoch": 17.696369636963695,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0011962562873226539,
      "loss": 0.2303,
      "num_input_tokens_seen": 33948384,
      "step": 160860
    },
    {
      "epoch": 17.696919691969196,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.001195692816998714,
      "loss": 0.2314,
      "num_input_tokens_seen": 33949440,
      "step": 160865
    },
    {
      "epoch": 17.697469746974697,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.001195129473902446,
      "loss": 0.2314,
      "num_input_tokens_seen": 33950432,
      "step": 160870
    },
    {
      "epoch": 17.698019801980198,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.0011945662580390487,
      "loss": 0.2293,
      "num_input_tokens_seen": 33951456,
      "step": 160875
    },
    {
      "epoch": 17.6985698569857,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0011940031694137097,
      "loss": 0.2314,
      "num_input_tokens_seen": 33952480,
      "step": 160880
    },
    {
      "epoch": 17.6991199119912,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0011934402080316137,
      "loss": 0.2314,
      "num_input_tokens_seen": 33953568,
      "step": 160885
    },
    {
      "epoch": 17.6996699669967,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.0011928773738979592,
      "loss": 0.2298,
      "num_input_tokens_seen": 33954624,
      "step": 160890
    },
    {
      "epoch": 17.7002200220022,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0011923146670179273,
      "loss": 0.2309,
      "num_input_tokens_seen": 33955712,
      "step": 160895
    },
    {
      "epoch": 17.7007700770077,
      "grad_norm": 0.000446319580078125,
      "learning_rate": 0.0011917520873967057,
      "loss": 0.2324,
      "num_input_tokens_seen": 33956800,
      "step": 160900
    },
    {
      "epoch": 17.7013201320132,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0011911896350394818,
      "loss": 0.2319,
      "num_input_tokens_seen": 33957792,
      "step": 160905
    },
    {
      "epoch": 17.701870187018702,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.001190627309951433,
      "loss": 0.2319,
      "num_input_tokens_seen": 33958848,
      "step": 160910
    },
    {
      "epoch": 17.702420242024203,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0011900651121377454,
      "loss": 0.2314,
      "num_input_tokens_seen": 33959872,
      "step": 160915
    },
    {
      "epoch": 17.702970297029704,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.001189503041603603,
      "loss": 0.2335,
      "num_input_tokens_seen": 33960928,
      "step": 160920
    },
    {
      "epoch": 17.703520352035202,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.0011889410983541848,
      "loss": 0.2319,
      "num_input_tokens_seen": 33961984,
      "step": 160925
    },
    {
      "epoch": 17.704070407040703,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.0011883792823946704,
      "loss": 0.2319,
      "num_input_tokens_seen": 33963008,
      "step": 160930
    },
    {
      "epoch": 17.704620462046204,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0011878175937302322,
      "loss": 0.2324,
      "num_input_tokens_seen": 33964096,
      "step": 160935
    },
    {
      "epoch": 17.705170517051705,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.0011872560323660558,
      "loss": 0.2324,
      "num_input_tokens_seen": 33965120,
      "step": 160940
    },
    {
      "epoch": 17.705720572057206,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0011866945983073074,
      "loss": 0.2308,
      "num_input_tokens_seen": 33966144,
      "step": 160945
    },
    {
      "epoch": 17.706270627062707,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.001186133291559171,
      "loss": 0.2288,
      "num_input_tokens_seen": 33967264,
      "step": 160950
    },
    {
      "epoch": 17.706820682068205,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0011855721121268158,
      "loss": 0.2314,
      "num_input_tokens_seen": 33968352,
      "step": 160955
    },
    {
      "epoch": 17.707370737073706,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0011850110600154112,
      "loss": 0.2288,
      "num_input_tokens_seen": 33969504,
      "step": 160960
    },
    {
      "epoch": 17.707920792079207,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0011844501352301345,
      "loss": 0.2293,
      "num_input_tokens_seen": 33970560,
      "step": 160965
    },
    {
      "epoch": 17.70847084708471,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.0011838893377761467,
      "loss": 0.2303,
      "num_input_tokens_seen": 33971616,
      "step": 160970
    },
    {
      "epoch": 17.70902090209021,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0011833286676586217,
      "loss": 0.2298,
      "num_input_tokens_seen": 33972640,
      "step": 160975
    },
    {
      "epoch": 17.70957095709571,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.001182768124882731,
      "loss": 0.2324,
      "num_input_tokens_seen": 33973728,
      "step": 160980
    },
    {
      "epoch": 17.71012101210121,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0011822077094536315,
      "loss": 0.2329,
      "num_input_tokens_seen": 33974720,
      "step": 160985
    },
    {
      "epoch": 17.71067106710671,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.001181647421376496,
      "loss": 0.2319,
      "num_input_tokens_seen": 33975808,
      "step": 160990
    },
    {
      "epoch": 17.71122112211221,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.001181087260656487,
      "loss": 0.2319,
      "num_input_tokens_seen": 33976864,
      "step": 160995
    },
    {
      "epoch": 17.71177117711771,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001180527227298762,
      "loss": 0.2335,
      "num_input_tokens_seen": 33977888,
      "step": 161000
    },
    {
      "epoch": 17.712321232123212,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.001179967321308487,
      "loss": 0.2293,
      "num_input_tokens_seen": 33978880,
      "step": 161005
    },
    {
      "epoch": 17.712871287128714,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0011794075426908213,
      "loss": 0.2303,
      "num_input_tokens_seen": 33979936,
      "step": 161010
    },
    {
      "epoch": 17.713421342134215,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.001178847891450927,
      "loss": 0.234,
      "num_input_tokens_seen": 33980960,
      "step": 161015
    },
    {
      "epoch": 17.713971397139716,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0011782883675939587,
      "loss": 0.2324,
      "num_input_tokens_seen": 33981920,
      "step": 161020
    },
    {
      "epoch": 17.714521452145213,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0011777289711250737,
      "loss": 0.2329,
      "num_input_tokens_seen": 33982944,
      "step": 161025
    },
    {
      "epoch": 17.715071507150714,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0011771697020494265,
      "loss": 0.2319,
      "num_input_tokens_seen": 33984032,
      "step": 161030
    },
    {
      "epoch": 17.715621562156215,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0011766105603721743,
      "loss": 0.2319,
      "num_input_tokens_seen": 33985120,
      "step": 161035
    },
    {
      "epoch": 17.716171617161717,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.0011760515460984716,
      "loss": 0.2314,
      "num_input_tokens_seen": 33986240,
      "step": 161040
    },
    {
      "epoch": 17.716721672167218,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.001175492659233469,
      "loss": 0.233,
      "num_input_tokens_seen": 33987264,
      "step": 161045
    },
    {
      "epoch": 17.71727172717272,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0011749338997823126,
      "loss": 0.2309,
      "num_input_tokens_seen": 33988320,
      "step": 161050
    },
    {
      "epoch": 17.717821782178216,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0011743752677501617,
      "loss": 0.2314,
      "num_input_tokens_seen": 33989376,
      "step": 161055
    },
    {
      "epoch": 17.718371837183717,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.0011738167631421552,
      "loss": 0.2314,
      "num_input_tokens_seen": 33990400,
      "step": 161060
    },
    {
      "epoch": 17.71892189218922,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.001173258385963446,
      "loss": 0.2303,
      "num_input_tokens_seen": 33991424,
      "step": 161065
    },
    {
      "epoch": 17.71947194719472,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0011727001362191814,
      "loss": 0.2303,
      "num_input_tokens_seen": 33992480,
      "step": 161070
    },
    {
      "epoch": 17.72002200220022,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0011721420139145038,
      "loss": 0.2303,
      "num_input_tokens_seen": 33993568,
      "step": 161075
    },
    {
      "epoch": 17.72057205720572,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0011715840190545596,
      "loss": 0.2314,
      "num_input_tokens_seen": 33994592,
      "step": 161080
    },
    {
      "epoch": 17.721122112211223,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0011710261516444857,
      "loss": 0.2314,
      "num_input_tokens_seen": 33995616,
      "step": 161085
    },
    {
      "epoch": 17.72167216721672,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0011704684116894304,
      "loss": 0.2324,
      "num_input_tokens_seen": 33996704,
      "step": 161090
    },
    {
      "epoch": 17.72222222222222,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.001169910799194534,
      "loss": 0.2298,
      "num_input_tokens_seen": 33997824,
      "step": 161095
    },
    {
      "epoch": 17.722772277227723,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0011693533141649292,
      "loss": 0.2329,
      "num_input_tokens_seen": 33998912,
      "step": 161100
    },
    {
      "epoch": 17.723322332233224,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.001168795956605762,
      "loss": 0.2314,
      "num_input_tokens_seen": 33999904,
      "step": 161105
    },
    {
      "epoch": 17.723872387238725,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0011682387265221632,
      "loss": 0.2319,
      "num_input_tokens_seen": 34000960,
      "step": 161110
    },
    {
      "epoch": 17.724422442244226,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.001167681623919274,
      "loss": 0.2303,
      "num_input_tokens_seen": 34002016,
      "step": 161115
    },
    {
      "epoch": 17.724972497249723,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.0011671246488022212,
      "loss": 0.2319,
      "num_input_tokens_seen": 34003040,
      "step": 161120
    },
    {
      "epoch": 17.725522552255224,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0011665678011761448,
      "loss": 0.2314,
      "num_input_tokens_seen": 34004064,
      "step": 161125
    },
    {
      "epoch": 17.726072607260726,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.0011660110810461771,
      "loss": 0.2308,
      "num_input_tokens_seen": 34005088,
      "step": 161130
    },
    {
      "epoch": 17.726622662266227,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.0011654544884174439,
      "loss": 0.2308,
      "num_input_tokens_seen": 34006144,
      "step": 161135
    },
    {
      "epoch": 17.727172717271728,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.001164898023295081,
      "loss": 0.2324,
      "num_input_tokens_seen": 34007168,
      "step": 161140
    },
    {
      "epoch": 17.72772277227723,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0011643416856842131,
      "loss": 0.2308,
      "num_input_tokens_seen": 34008256,
      "step": 161145
    },
    {
      "epoch": 17.72827282728273,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0011637854755899672,
      "loss": 0.2304,
      "num_input_tokens_seen": 34009248,
      "step": 161150
    },
    {
      "epoch": 17.728822882288227,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0011632293930174747,
      "loss": 0.2309,
      "num_input_tokens_seen": 34010208,
      "step": 161155
    },
    {
      "epoch": 17.72937293729373,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.0011626734379718544,
      "loss": 0.2309,
      "num_input_tokens_seen": 34011200,
      "step": 161160
    },
    {
      "epoch": 17.72992299229923,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.0011621176104582359,
      "loss": 0.2314,
      "num_input_tokens_seen": 34012288,
      "step": 161165
    },
    {
      "epoch": 17.73047304730473,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0011615619104817399,
      "loss": 0.2314,
      "num_input_tokens_seen": 34013312,
      "step": 161170
    },
    {
      "epoch": 17.731023102310232,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001161006338047484,
      "loss": 0.2314,
      "num_input_tokens_seen": 34014304,
      "step": 161175
    },
    {
      "epoch": 17.731573157315733,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0011604508931605938,
      "loss": 0.2319,
      "num_input_tokens_seen": 34015392,
      "step": 161180
    },
    {
      "epoch": 17.73212321232123,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.0011598955758261875,
      "loss": 0.234,
      "num_input_tokens_seen": 34016448,
      "step": 161185
    },
    {
      "epoch": 17.73267326732673,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0011593403860493806,
      "loss": 0.2319,
      "num_input_tokens_seen": 34017472,
      "step": 161190
    },
    {
      "epoch": 17.733223322332233,
      "grad_norm": 0.00049591064453125,
      "learning_rate": 0.0011587853238352942,
      "loss": 0.2314,
      "num_input_tokens_seen": 34018528,
      "step": 161195
    },
    {
      "epoch": 17.733773377337734,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0011582303891890405,
      "loss": 0.2303,
      "num_input_tokens_seen": 34019552,
      "step": 161200
    },
    {
      "epoch": 17.734323432343235,
      "grad_norm": 0.000518798828125,
      "learning_rate": 0.0011576755821157342,
      "loss": 0.2319,
      "num_input_tokens_seen": 34020576,
      "step": 161205
    },
    {
      "epoch": 17.734873487348736,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0011571209026204927,
      "loss": 0.2314,
      "num_input_tokens_seen": 34021632,
      "step": 161210
    },
    {
      "epoch": 17.735423542354237,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0011565663507084217,
      "loss": 0.2309,
      "num_input_tokens_seen": 34022688,
      "step": 161215
    },
    {
      "epoch": 17.735973597359735,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 0.0011560119263846408,
      "loss": 0.2319,
      "num_input_tokens_seen": 34023808,
      "step": 161220
    },
    {
      "epoch": 17.736523652365236,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0011554576296542508,
      "loss": 0.2319,
      "num_input_tokens_seen": 34024800,
      "step": 161225
    },
    {
      "epoch": 17.737073707370737,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0011549034605223673,
      "loss": 0.2324,
      "num_input_tokens_seen": 34025824,
      "step": 161230
    },
    {
      "epoch": 17.737623762376238,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0011543494189940934,
      "loss": 0.2298,
      "num_input_tokens_seen": 34026880,
      "step": 161235
    },
    {
      "epoch": 17.73817381738174,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.0011537955050745362,
      "loss": 0.2314,
      "num_input_tokens_seen": 34027968,
      "step": 161240
    },
    {
      "epoch": 17.73872387238724,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0011532417187688038,
      "loss": 0.2303,
      "num_input_tokens_seen": 34029056,
      "step": 161245
    },
    {
      "epoch": 17.739273927392738,
      "grad_norm": 0.0006561279296875,
      "learning_rate": 0.0011526880600819967,
      "loss": 0.2329,
      "num_input_tokens_seen": 34030048,
      "step": 161250
    },
    {
      "epoch": 17.73982398239824,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0011521345290192208,
      "loss": 0.2329,
      "num_input_tokens_seen": 34031072,
      "step": 161255
    },
    {
      "epoch": 17.74037403740374,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0011515811255855756,
      "loss": 0.2314,
      "num_input_tokens_seen": 34032096,
      "step": 161260
    },
    {
      "epoch": 17.74092409240924,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0011510278497861603,
      "loss": 0.2303,
      "num_input_tokens_seen": 34033184,
      "step": 161265
    },
    {
      "epoch": 17.741474147414742,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.0011504747016260791,
      "loss": 0.2303,
      "num_input_tokens_seen": 34034240,
      "step": 161270
    },
    {
      "epoch": 17.742024202420243,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.001149921681110423,
      "loss": 0.2309,
      "num_input_tokens_seen": 34035296,
      "step": 161275
    },
    {
      "epoch": 17.742574257425744,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.0011493687882442975,
      "loss": 0.2319,
      "num_input_tokens_seen": 34036352,
      "step": 161280
    },
    {
      "epoch": 17.74312431243124,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0011488160230327927,
      "loss": 0.2319,
      "num_input_tokens_seen": 34037344,
      "step": 161285
    },
    {
      "epoch": 17.743674367436743,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0011482633854810003,
      "loss": 0.2319,
      "num_input_tokens_seen": 34038400,
      "step": 161290
    },
    {
      "epoch": 17.744224422442244,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0011477108755940202,
      "loss": 0.2314,
      "num_input_tokens_seen": 34039520,
      "step": 161295
    },
    {
      "epoch": 17.744774477447745,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0011471584933769397,
      "loss": 0.2319,
      "num_input_tokens_seen": 34040544,
      "step": 161300
    },
    {
      "epoch": 17.745324532453246,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0011466062388348563,
      "loss": 0.2329,
      "num_input_tokens_seen": 34041568,
      "step": 161305
    },
    {
      "epoch": 17.745874587458747,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0011460541119728545,
      "loss": 0.2303,
      "num_input_tokens_seen": 34042688,
      "step": 161310
    },
    {
      "epoch": 17.746424642464248,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.001145502112796022,
      "loss": 0.2298,
      "num_input_tokens_seen": 34043776,
      "step": 161315
    },
    {
      "epoch": 17.746974697469746,
      "grad_norm": 0.000682830810546875,
      "learning_rate": 0.0011449502413094492,
      "loss": 0.2324,
      "num_input_tokens_seen": 34044832,
      "step": 161320
    },
    {
      "epoch": 17.747524752475247,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0011443984975182209,
      "loss": 0.2319,
      "num_input_tokens_seen": 34045920,
      "step": 161325
    },
    {
      "epoch": 17.748074807480748,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.001143846881427426,
      "loss": 0.2319,
      "num_input_tokens_seen": 34047072,
      "step": 161330
    },
    {
      "epoch": 17.74862486248625,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0011432953930421458,
      "loss": 0.2298,
      "num_input_tokens_seen": 34048192,
      "step": 161335
    },
    {
      "epoch": 17.74917491749175,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0011427440323674592,
      "loss": 0.2303,
      "num_input_tokens_seen": 34049248,
      "step": 161340
    },
    {
      "epoch": 17.74972497249725,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0011421927994084557,
      "loss": 0.2319,
      "num_input_tokens_seen": 34050304,
      "step": 161345
    },
    {
      "epoch": 17.75027502750275,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0011416416941702077,
      "loss": 0.2298,
      "num_input_tokens_seen": 34051392,
      "step": 161350
    },
    {
      "epoch": 17.75082508250825,
      "grad_norm": 0.000858306884765625,
      "learning_rate": 0.0011410907166577982,
      "loss": 0.2293,
      "num_input_tokens_seen": 34052448,
      "step": 161355
    },
    {
      "epoch": 17.75137513751375,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0011405398668763095,
      "loss": 0.2303,
      "num_input_tokens_seen": 34053504,
      "step": 161360
    },
    {
      "epoch": 17.751925192519252,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.0011399891448308109,
      "loss": 0.2303,
      "num_input_tokens_seen": 34054560,
      "step": 161365
    },
    {
      "epoch": 17.752475247524753,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.0011394385505263832,
      "loss": 0.2319,
      "num_input_tokens_seen": 34055584,
      "step": 161370
    },
    {
      "epoch": 17.753025302530254,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0011388880839680975,
      "loss": 0.2303,
      "num_input_tokens_seen": 34056576,
      "step": 161375
    },
    {
      "epoch": 17.753575357535752,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0011383377451610298,
      "loss": 0.2298,
      "num_input_tokens_seen": 34057600,
      "step": 161380
    },
    {
      "epoch": 17.754125412541253,
      "grad_norm": 0.000705718994140625,
      "learning_rate": 0.0011377875341102544,
      "loss": 0.2319,
      "num_input_tokens_seen": 34058624,
      "step": 161385
    },
    {
      "epoch": 17.754675467546754,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.0011372374508208372,
      "loss": 0.2314,
      "num_input_tokens_seen": 34059744,
      "step": 161390
    },
    {
      "epoch": 17.755225522552255,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0011366874952978506,
      "loss": 0.2345,
      "num_input_tokens_seen": 34060928,
      "step": 161395
    },
    {
      "epoch": 17.755775577557756,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.0011361376675463658,
      "loss": 0.2303,
      "num_input_tokens_seen": 34061952,
      "step": 161400
    },
    {
      "epoch": 17.756325632563257,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0011355879675714435,
      "loss": 0.2303,
      "num_input_tokens_seen": 34062976,
      "step": 161405
    },
    {
      "epoch": 17.75687568756876,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.001135038395378155,
      "loss": 0.2293,
      "num_input_tokens_seen": 34064032,
      "step": 161410
    },
    {
      "epoch": 17.757425742574256,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0011344889509715644,
      "loss": 0.2314,
      "num_input_tokens_seen": 34065088,
      "step": 161415
    },
    {
      "epoch": 17.757975797579757,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.0011339396343567375,
      "loss": 0.2314,
      "num_input_tokens_seen": 34066144,
      "step": 161420
    },
    {
      "epoch": 17.758525852585258,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.0011333904455387356,
      "loss": 0.2308,
      "num_input_tokens_seen": 34067104,
      "step": 161425
    },
    {
      "epoch": 17.75907590759076,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0011328413845226191,
      "loss": 0.2324,
      "num_input_tokens_seen": 34068160,
      "step": 161430
    },
    {
      "epoch": 17.75962596259626,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.001132292451313448,
      "loss": 0.2303,
      "num_input_tokens_seen": 34069248,
      "step": 161435
    },
    {
      "epoch": 17.76017601760176,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0011317436459162826,
      "loss": 0.2324,
      "num_input_tokens_seen": 34070272,
      "step": 161440
    },
    {
      "epoch": 17.760726072607262,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.001131194968336186,
      "loss": 0.2293,
      "num_input_tokens_seen": 34071328,
      "step": 161445
    },
    {
      "epoch": 17.76127612761276,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.001130646418578209,
      "loss": 0.2303,
      "num_input_tokens_seen": 34072416,
      "step": 161450
    },
    {
      "epoch": 17.76182618261826,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0011300979966474055,
      "loss": 0.2303,
      "num_input_tokens_seen": 34073440,
      "step": 161455
    },
    {
      "epoch": 17.762376237623762,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.0011295497025488371,
      "loss": 0.2329,
      "num_input_tokens_seen": 34074400,
      "step": 161460
    },
    {
      "epoch": 17.762926292629263,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.001129001536287551,
      "loss": 0.2309,
      "num_input_tokens_seen": 34075488,
      "step": 161465
    },
    {
      "epoch": 17.763476347634764,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0011284534978686018,
      "loss": 0.2298,
      "num_input_tokens_seen": 34076544,
      "step": 161470
    },
    {
      "epoch": 17.764026402640265,
      "grad_norm": 0.000888824462890625,
      "learning_rate": 0.0011279055872970417,
      "loss": 0.2324,
      "num_input_tokens_seen": 34077600,
      "step": 161475
    },
    {
      "epoch": 17.764576457645763,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.001127357804577917,
      "loss": 0.2329,
      "num_input_tokens_seen": 34078688,
      "step": 161480
    },
    {
      "epoch": 17.765126512651264,
      "grad_norm": 0.000568389892578125,
      "learning_rate": 0.001126810149716282,
      "loss": 0.2329,
      "num_input_tokens_seen": 34079776,
      "step": 161485
    },
    {
      "epoch": 17.765676567656765,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.001126262622717179,
      "loss": 0.2314,
      "num_input_tokens_seen": 34080800,
      "step": 161490
    },
    {
      "epoch": 17.766226622662266,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001125715223585656,
      "loss": 0.2314,
      "num_input_tokens_seen": 34081792,
      "step": 161495
    },
    {
      "epoch": 17.766776677667767,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0011251679523267587,
      "loss": 0.2329,
      "num_input_tokens_seen": 34082816,
      "step": 161500
    },
    {
      "epoch": 17.76732673267327,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.00112462080894553,
      "loss": 0.2308,
      "num_input_tokens_seen": 34083904,
      "step": 161505
    },
    {
      "epoch": 17.76787678767877,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0011240737934470157,
      "loss": 0.2314,
      "num_input_tokens_seen": 34084960,
      "step": 161510
    },
    {
      "epoch": 17.768426842684267,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0011235269058362518,
      "loss": 0.2329,
      "num_input_tokens_seen": 34086048,
      "step": 161515
    },
    {
      "epoch": 17.768976897689768,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0011229801461182841,
      "loss": 0.2324,
      "num_input_tokens_seen": 34087104,
      "step": 161520
    },
    {
      "epoch": 17.76952695269527,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0011224335142981472,
      "loss": 0.2314,
      "num_input_tokens_seen": 34088160,
      "step": 161525
    },
    {
      "epoch": 17.77007700770077,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.001121887010380882,
      "loss": 0.2319,
      "num_input_tokens_seen": 34089248,
      "step": 161530
    },
    {
      "epoch": 17.77062706270627,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.0011213406343715275,
      "loss": 0.2303,
      "num_input_tokens_seen": 34090304,
      "step": 161535
    },
    {
      "epoch": 17.771177117711773,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0011207943862751135,
      "loss": 0.2298,
      "num_input_tokens_seen": 34091360,
      "step": 161540
    },
    {
      "epoch": 17.77172717271727,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.0011202482660966806,
      "loss": 0.2314,
      "num_input_tokens_seen": 34092384,
      "step": 161545
    },
    {
      "epoch": 17.77227722772277,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0011197022738412598,
      "loss": 0.2309,
      "num_input_tokens_seen": 34093440,
      "step": 161550
    },
    {
      "epoch": 17.772827282728272,
      "grad_norm": 0.001953125,
      "learning_rate": 0.0011191564095138773,
      "loss": 0.2308,
      "num_input_tokens_seen": 34094496,
      "step": 161555
    },
    {
      "epoch": 17.773377337733773,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.0011186106731195739,
      "loss": 0.2303,
      "num_input_tokens_seen": 34095552,
      "step": 161560
    },
    {
      "epoch": 17.773927392739274,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.0011180650646633755,
      "loss": 0.2319,
      "num_input_tokens_seen": 34096576,
      "step": 161565
    },
    {
      "epoch": 17.774477447744776,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0011175195841503082,
      "loss": 0.2329,
      "num_input_tokens_seen": 34097568,
      "step": 161570
    },
    {
      "epoch": 17.775027502750277,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0011169742315854032,
      "loss": 0.2319,
      "num_input_tokens_seen": 34098656,
      "step": 161575
    },
    {
      "epoch": 17.775577557755774,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.0011164290069736826,
      "loss": 0.2309,
      "num_input_tokens_seen": 34099744,
      "step": 161580
    },
    {
      "epoch": 17.776127612761275,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0011158839103201745,
      "loss": 0.2324,
      "num_input_tokens_seen": 34100832,
      "step": 161585
    },
    {
      "epoch": 17.776677667766776,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001115338941629903,
      "loss": 0.2308,
      "num_input_tokens_seen": 34101856,
      "step": 161590
    },
    {
      "epoch": 17.777227722772277,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0011147941009078894,
      "loss": 0.2314,
      "num_input_tokens_seen": 34102880,
      "step": 161595
    },
    {
      "epoch": 17.77777777777778,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0011142493881591575,
      "loss": 0.2314,
      "num_input_tokens_seen": 34103904,
      "step": 161600
    },
    {
      "epoch": 17.77832783278328,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.0011137048033887237,
      "loss": 0.2335,
      "num_input_tokens_seen": 34104960,
      "step": 161605
    },
    {
      "epoch": 17.778877887788777,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0011131603466016087,
      "loss": 0.2308,
      "num_input_tokens_seen": 34106048,
      "step": 161610
    },
    {
      "epoch": 17.77942794279428,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.0011126160178028337,
      "loss": 0.2324,
      "num_input_tokens_seen": 34107104,
      "step": 161615
    },
    {
      "epoch": 17.77997799779978,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.0011120718169974114,
      "loss": 0.2309,
      "num_input_tokens_seen": 34108160,
      "step": 161620
    },
    {
      "epoch": 17.78052805280528,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.001111527744190361,
      "loss": 0.2314,
      "num_input_tokens_seen": 34109312,
      "step": 161625
    },
    {
      "epoch": 17.78107810781078,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.0011109837993866932,
      "loss": 0.2324,
      "num_input_tokens_seen": 34110464,
      "step": 161630
    },
    {
      "epoch": 17.781628162816283,
      "grad_norm": 0.001953125,
      "learning_rate": 0.0011104399825914246,
      "loss": 0.2303,
      "num_input_tokens_seen": 34111552,
      "step": 161635
    },
    {
      "epoch": 17.782178217821784,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.001109896293809564,
      "loss": 0.2319,
      "num_input_tokens_seen": 34112544,
      "step": 161640
    },
    {
      "epoch": 17.78272827282728,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0011093527330461245,
      "loss": 0.2314,
      "num_input_tokens_seen": 34113600,
      "step": 161645
    },
    {
      "epoch": 17.783278327832782,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0011088093003061167,
      "loss": 0.2303,
      "num_input_tokens_seen": 34114624,
      "step": 161650
    },
    {
      "epoch": 17.783828382838283,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0011082659955945467,
      "loss": 0.2319,
      "num_input_tokens_seen": 34115648,
      "step": 161655
    },
    {
      "epoch": 17.784378437843785,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0011077228189164256,
      "loss": 0.2314,
      "num_input_tokens_seen": 34116736,
      "step": 161660
    },
    {
      "epoch": 17.784928492849286,
      "grad_norm": 0.005096435546875,
      "learning_rate": 0.001107179770276756,
      "loss": 0.2288,
      "num_input_tokens_seen": 34117824,
      "step": 161665
    },
    {
      "epoch": 17.785478547854787,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0011066368496805406,
      "loss": 0.2319,
      "num_input_tokens_seen": 34118848,
      "step": 161670
    },
    {
      "epoch": 17.786028602860284,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0011060940571327904,
      "loss": 0.2319,
      "num_input_tokens_seen": 34119872,
      "step": 161675
    },
    {
      "epoch": 17.786578657865785,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0011055513926385028,
      "loss": 0.2319,
      "num_input_tokens_seen": 34120896,
      "step": 161680
    },
    {
      "epoch": 17.787128712871286,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.0011050088562026827,
      "loss": 0.2335,
      "num_input_tokens_seen": 34121920,
      "step": 161685
    },
    {
      "epoch": 17.787678767876788,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.001104466447830329,
      "loss": 0.2308,
      "num_input_tokens_seen": 34122912,
      "step": 161690
    },
    {
      "epoch": 17.78822882288229,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 0.0011039241675264361,
      "loss": 0.2314,
      "num_input_tokens_seen": 34124000,
      "step": 161695
    },
    {
      "epoch": 17.78877887788779,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0011033820152960084,
      "loss": 0.2293,
      "num_input_tokens_seen": 34125056,
      "step": 161700
    },
    {
      "epoch": 17.78932893289329,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.0011028399911440384,
      "loss": 0.2324,
      "num_input_tokens_seen": 34126144,
      "step": 161705
    },
    {
      "epoch": 17.78987898789879,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.0011022980950755273,
      "loss": 0.2324,
      "num_input_tokens_seen": 34127200,
      "step": 161710
    },
    {
      "epoch": 17.79042904290429,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0011017563270954661,
      "loss": 0.2303,
      "num_input_tokens_seen": 34128288,
      "step": 161715
    },
    {
      "epoch": 17.79097909790979,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0011012146872088457,
      "loss": 0.2314,
      "num_input_tokens_seen": 34129376,
      "step": 161720
    },
    {
      "epoch": 17.79152915291529,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.001100673175420662,
      "loss": 0.2329,
      "num_input_tokens_seen": 34130432,
      "step": 161725
    },
    {
      "epoch": 17.792079207920793,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0011001317917358983,
      "loss": 0.2314,
      "num_input_tokens_seen": 34131520,
      "step": 161730
    },
    {
      "epoch": 17.792629262926294,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0010995905361595565,
      "loss": 0.2324,
      "num_input_tokens_seen": 34132576,
      "step": 161735
    },
    {
      "epoch": 17.793179317931795,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.001099049408696618,
      "loss": 0.2314,
      "num_input_tokens_seen": 34133632,
      "step": 161740
    },
    {
      "epoch": 17.793729372937293,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.0010985084093520685,
      "loss": 0.2314,
      "num_input_tokens_seen": 34134720,
      "step": 161745
    },
    {
      "epoch": 17.794279427942794,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0010979675381308995,
      "loss": 0.2324,
      "num_input_tokens_seen": 34135744,
      "step": 161750
    },
    {
      "epoch": 17.794829482948295,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00109742679503809,
      "loss": 0.2303,
      "num_input_tokens_seen": 34136896,
      "step": 161755
    },
    {
      "epoch": 17.795379537953796,
      "grad_norm": 0.005859375,
      "learning_rate": 0.001096886180078626,
      "loss": 0.2298,
      "num_input_tokens_seen": 34137952,
      "step": 161760
    },
    {
      "epoch": 17.795929592959297,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0010963456932574938,
      "loss": 0.2314,
      "num_input_tokens_seen": 34139008,
      "step": 161765
    },
    {
      "epoch": 17.796479647964798,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0010958053345796676,
      "loss": 0.2313,
      "num_input_tokens_seen": 34140032,
      "step": 161770
    },
    {
      "epoch": 17.797029702970296,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0010952651040501366,
      "loss": 0.2319,
      "num_input_tokens_seen": 34141152,
      "step": 161775
    },
    {
      "epoch": 17.797579757975797,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0010947250016738734,
      "loss": 0.2319,
      "num_input_tokens_seen": 34142176,
      "step": 161780
    },
    {
      "epoch": 17.798129812981298,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.001094185027455856,
      "loss": 0.2314,
      "num_input_tokens_seen": 34143296,
      "step": 161785
    },
    {
      "epoch": 17.7986798679868,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0010936451814010622,
      "loss": 0.2335,
      "num_input_tokens_seen": 34144352,
      "step": 161790
    },
    {
      "epoch": 17.7992299229923,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0010931054635144677,
      "loss": 0.2319,
      "num_input_tokens_seen": 34145408,
      "step": 161795
    },
    {
      "epoch": 17.7997799779978,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0010925658738010484,
      "loss": 0.2298,
      "num_input_tokens_seen": 34146464,
      "step": 161800
    },
    {
      "epoch": 17.8003300330033,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0010920264122657774,
      "loss": 0.2293,
      "num_input_tokens_seen": 34147520,
      "step": 161805
    },
    {
      "epoch": 17.8008800880088,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0010914870789136238,
      "loss": 0.2329,
      "num_input_tokens_seen": 34148544,
      "step": 161810
    },
    {
      "epoch": 17.8014301430143,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.001090947873749557,
      "loss": 0.2319,
      "num_input_tokens_seen": 34149600,
      "step": 161815
    },
    {
      "epoch": 17.801980198019802,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.0010904087967785513,
      "loss": 0.233,
      "num_input_tokens_seen": 34150656,
      "step": 161820
    },
    {
      "epoch": 17.802530253025303,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.0010898698480055762,
      "loss": 0.2319,
      "num_input_tokens_seen": 34151744,
      "step": 161825
    },
    {
      "epoch": 17.803080308030804,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.0010893310274355944,
      "loss": 0.2324,
      "num_input_tokens_seen": 34152768,
      "step": 161830
    },
    {
      "epoch": 17.803630363036305,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.0010887923350735716,
      "loss": 0.2324,
      "num_input_tokens_seen": 34153760,
      "step": 161835
    },
    {
      "epoch": 17.804180418041803,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0010882537709244776,
      "loss": 0.2303,
      "num_input_tokens_seen": 34154816,
      "step": 161840
    },
    {
      "epoch": 17.804730473047304,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00108771533499327,
      "loss": 0.2303,
      "num_input_tokens_seen": 34155840,
      "step": 161845
    },
    {
      "epoch": 17.805280528052805,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0010871770272849145,
      "loss": 0.2308,
      "num_input_tokens_seen": 34156928,
      "step": 161850
    },
    {
      "epoch": 17.805830583058306,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.001086638847804376,
      "loss": 0.2304,
      "num_input_tokens_seen": 34157984,
      "step": 161855
    },
    {
      "epoch": 17.806380638063807,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0010861007965566065,
      "loss": 0.2329,
      "num_input_tokens_seen": 34159072,
      "step": 161860
    },
    {
      "epoch": 17.806930693069308,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0010855628735465745,
      "loss": 0.2324,
      "num_input_tokens_seen": 34160096,
      "step": 161865
    },
    {
      "epoch": 17.80748074807481,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.001085025078779227,
      "loss": 0.2324,
      "num_input_tokens_seen": 34161152,
      "step": 161870
    },
    {
      "epoch": 17.808030803080307,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0010844874122595287,
      "loss": 0.2324,
      "num_input_tokens_seen": 34162176,
      "step": 161875
    },
    {
      "epoch": 17.808580858085808,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.001083949873992434,
      "loss": 0.2314,
      "num_input_tokens_seen": 34163232,
      "step": 161880
    },
    {
      "epoch": 17.80913091309131,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.001083412463982894,
      "loss": 0.2324,
      "num_input_tokens_seen": 34164320,
      "step": 161885
    },
    {
      "epoch": 17.80968096809681,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.0010828751822358645,
      "loss": 0.2314,
      "num_input_tokens_seen": 34165440,
      "step": 161890
    },
    {
      "epoch": 17.81023102310231,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.001082338028756295,
      "loss": 0.2303,
      "num_input_tokens_seen": 34166528,
      "step": 161895
    },
    {
      "epoch": 17.810781078107812,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.0010818010035491398,
      "loss": 0.2303,
      "num_input_tokens_seen": 34167552,
      "step": 161900
    },
    {
      "epoch": 17.81133113311331,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.001081264106619345,
      "loss": 0.2293,
      "num_input_tokens_seen": 34168544,
      "step": 161905
    },
    {
      "epoch": 17.81188118811881,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.0010807273379718585,
      "loss": 0.2288,
      "num_input_tokens_seen": 34169632,
      "step": 161910
    },
    {
      "epoch": 17.812431243124312,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0010801906976116327,
      "loss": 0.2308,
      "num_input_tokens_seen": 34170688,
      "step": 161915
    },
    {
      "epoch": 17.812981298129813,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.0010796541855436054,
      "loss": 0.2319,
      "num_input_tokens_seen": 34171776,
      "step": 161920
    },
    {
      "epoch": 17.813531353135314,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0010791178017727293,
      "loss": 0.2314,
      "num_input_tokens_seen": 34172896,
      "step": 161925
    },
    {
      "epoch": 17.814081408140815,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.001078581546303942,
      "loss": 0.2303,
      "num_input_tokens_seen": 34174048,
      "step": 161930
    },
    {
      "epoch": 17.814631463146316,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.00107804541914219,
      "loss": 0.2308,
      "num_input_tokens_seen": 34175136,
      "step": 161935
    },
    {
      "epoch": 17.815181518151814,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001077509420292414,
      "loss": 0.234,
      "num_input_tokens_seen": 34176128,
      "step": 161940
    },
    {
      "epoch": 17.815731573157315,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.0010769735497595501,
      "loss": 0.2324,
      "num_input_tokens_seen": 34177184,
      "step": 161945
    },
    {
      "epoch": 17.816281628162816,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0010764378075485425,
      "loss": 0.2329,
      "num_input_tokens_seen": 34178272,
      "step": 161950
    },
    {
      "epoch": 17.816831683168317,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.0010759021936643275,
      "loss": 0.2319,
      "num_input_tokens_seen": 34179328,
      "step": 161955
    },
    {
      "epoch": 17.817381738173818,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0010753667081118377,
      "loss": 0.2319,
      "num_input_tokens_seen": 34180384,
      "step": 161960
    },
    {
      "epoch": 17.81793179317932,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.001074831350896011,
      "loss": 0.2293,
      "num_input_tokens_seen": 34181472,
      "step": 161965
    },
    {
      "epoch": 17.818481848184817,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.001074296122021785,
      "loss": 0.2303,
      "num_input_tokens_seen": 34182560,
      "step": 161970
    },
    {
      "epoch": 17.819031903190318,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0010737610214940857,
      "loss": 0.2309,
      "num_input_tokens_seen": 34183680,
      "step": 161975
    },
    {
      "epoch": 17.81958195819582,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0010732260493178524,
      "loss": 0.2303,
      "num_input_tokens_seen": 34184768,
      "step": 161980
    },
    {
      "epoch": 17.82013201320132,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0010726912054980081,
      "loss": 0.2319,
      "num_input_tokens_seen": 34185824,
      "step": 161985
    },
    {
      "epoch": 17.82068206820682,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.001072156490039487,
      "loss": 0.2314,
      "num_input_tokens_seen": 34186912,
      "step": 161990
    },
    {
      "epoch": 17.821232123212322,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.0010716219029472185,
      "loss": 0.2319,
      "num_input_tokens_seen": 34188096,
      "step": 161995
    },
    {
      "epoch": 17.821782178217823,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.0010710874442261254,
      "loss": 0.2314,
      "num_input_tokens_seen": 34189184,
      "step": 162000
    },
    {
      "epoch": 17.82233223322332,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0010705531138811369,
      "loss": 0.2293,
      "num_input_tokens_seen": 34190304,
      "step": 162005
    },
    {
      "epoch": 17.822882288228822,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0010700189119171742,
      "loss": 0.2324,
      "num_input_tokens_seen": 34191360,
      "step": 162010
    },
    {
      "epoch": 17.823432343234323,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0010694848383391669,
      "loss": 0.2324,
      "num_input_tokens_seen": 34192384,
      "step": 162015
    },
    {
      "epoch": 17.823982398239824,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.001068950893152029,
      "loss": 0.2319,
      "num_input_tokens_seen": 34193408,
      "step": 162020
    },
    {
      "epoch": 17.824532453245325,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.001068417076360687,
      "loss": 0.2324,
      "num_input_tokens_seen": 34194464,
      "step": 162025
    },
    {
      "epoch": 17.825082508250826,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0010678833879700616,
      "loss": 0.2309,
      "num_input_tokens_seen": 34195552,
      "step": 162030
    },
    {
      "epoch": 17.825632563256324,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0010673498279850674,
      "loss": 0.233,
      "num_input_tokens_seen": 34196608,
      "step": 162035
    },
    {
      "epoch": 17.826182618261825,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.001066816396410627,
      "loss": 0.2303,
      "num_input_tokens_seen": 34197792,
      "step": 162040
    },
    {
      "epoch": 17.826732673267326,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0010662830932516515,
      "loss": 0.2319,
      "num_input_tokens_seen": 34198816,
      "step": 162045
    },
    {
      "epoch": 17.827282728272827,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0010657499185130585,
      "loss": 0.2303,
      "num_input_tokens_seen": 34199904,
      "step": 162050
    },
    {
      "epoch": 17.82783278327833,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.001065216872199766,
      "loss": 0.2324,
      "num_input_tokens_seen": 34200896,
      "step": 162055
    },
    {
      "epoch": 17.82838283828383,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0010646839543166785,
      "loss": 0.2324,
      "num_input_tokens_seen": 34201984,
      "step": 162060
    },
    {
      "epoch": 17.82893289328933,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0010641511648687167,
      "loss": 0.2329,
      "num_input_tokens_seen": 34203040,
      "step": 162065
    },
    {
      "epoch": 17.829482948294828,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0010636185038607852,
      "loss": 0.2319,
      "num_input_tokens_seen": 34204096,
      "step": 162070
    },
    {
      "epoch": 17.83003300330033,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0010630859712977918,
      "loss": 0.2293,
      "num_input_tokens_seen": 34205216,
      "step": 162075
    },
    {
      "epoch": 17.83058305830583,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.0010625535671846492,
      "loss": 0.2308,
      "num_input_tokens_seen": 34206208,
      "step": 162080
    },
    {
      "epoch": 17.83113311331133,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00106202129152626,
      "loss": 0.2314,
      "num_input_tokens_seen": 34207296,
      "step": 162085
    },
    {
      "epoch": 17.831683168316832,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.001061489144327537,
      "loss": 0.2304,
      "num_input_tokens_seen": 34208320,
      "step": 162090
    },
    {
      "epoch": 17.832233223322334,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0010609571255933796,
      "loss": 0.2298,
      "num_input_tokens_seen": 34209408,
      "step": 162095
    },
    {
      "epoch": 17.83278327832783,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.001060425235328689,
      "loss": 0.2298,
      "num_input_tokens_seen": 34210400,
      "step": 162100
    },
    {
      "epoch": 17.833333333333332,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.0010598934735383697,
      "loss": 0.2319,
      "num_input_tokens_seen": 34211456,
      "step": 162105
    },
    {
      "epoch": 17.833883388338833,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.0010593618402273241,
      "loss": 0.2309,
      "num_input_tokens_seen": 34212480,
      "step": 162110
    },
    {
      "epoch": 17.834433443344334,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0010588303354004535,
      "loss": 0.2335,
      "num_input_tokens_seen": 34213568,
      "step": 162115
    },
    {
      "epoch": 17.834983498349835,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.001058298959062654,
      "loss": 0.2314,
      "num_input_tokens_seen": 34214560,
      "step": 162120
    },
    {
      "epoch": 17.835533553355337,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.00105776771121882,
      "loss": 0.2303,
      "num_input_tokens_seen": 34215616,
      "step": 162125
    },
    {
      "epoch": 17.836083608360838,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.001057236591873854,
      "loss": 0.2324,
      "num_input_tokens_seen": 34216640,
      "step": 162130
    },
    {
      "epoch": 17.836633663366335,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0010567056010326458,
      "loss": 0.2319,
      "num_input_tokens_seen": 34217696,
      "step": 162135
    },
    {
      "epoch": 17.837183718371836,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0010561747387000913,
      "loss": 0.2319,
      "num_input_tokens_seen": 34218752,
      "step": 162140
    },
    {
      "epoch": 17.837733773377337,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.001055644004881085,
      "loss": 0.2324,
      "num_input_tokens_seen": 34219808,
      "step": 162145
    },
    {
      "epoch": 17.83828382838284,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0010551133995805144,
      "loss": 0.2303,
      "num_input_tokens_seen": 34220864,
      "step": 162150
    },
    {
      "epoch": 17.83883388338834,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.0010545829228032744,
      "loss": 0.2324,
      "num_input_tokens_seen": 34221888,
      "step": 162155
    },
    {
      "epoch": 17.83938393839384,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0010540525745542505,
      "loss": 0.233,
      "num_input_tokens_seen": 34223008,
      "step": 162160
    },
    {
      "epoch": 17.83993399339934,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.0010535223548383326,
      "loss": 0.2314,
      "num_input_tokens_seen": 34224000,
      "step": 162165
    },
    {
      "epoch": 17.84048404840484,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0010529922636604084,
      "loss": 0.2313,
      "num_input_tokens_seen": 34225056,
      "step": 162170
    },
    {
      "epoch": 17.84103410341034,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0010524623010253586,
      "loss": 0.2288,
      "num_input_tokens_seen": 34226080,
      "step": 162175
    },
    {
      "epoch": 17.84158415841584,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0010519324669380748,
      "loss": 0.2329,
      "num_input_tokens_seen": 34227168,
      "step": 162180
    },
    {
      "epoch": 17.842134213421343,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.001051402761403436,
      "loss": 0.2319,
      "num_input_tokens_seen": 34228160,
      "step": 162185
    },
    {
      "epoch": 17.842684268426844,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.0010508731844263235,
      "loss": 0.2319,
      "num_input_tokens_seen": 34229216,
      "step": 162190
    },
    {
      "epoch": 17.843234323432345,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.0010503437360116184,
      "loss": 0.2308,
      "num_input_tokens_seen": 34230304,
      "step": 162195
    },
    {
      "epoch": 17.843784378437842,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0010498144161642019,
      "loss": 0.2314,
      "num_input_tokens_seen": 34231360,
      "step": 162200
    },
    {
      "epoch": 17.844334433443343,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.0010492852248889534,
      "loss": 0.2308,
      "num_input_tokens_seen": 34232384,
      "step": 162205
    },
    {
      "epoch": 17.844884488448844,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.0010487561621907503,
      "loss": 0.2314,
      "num_input_tokens_seen": 34233376,
      "step": 162210
    },
    {
      "epoch": 17.845434543454346,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0010482272280744641,
      "loss": 0.2329,
      "num_input_tokens_seen": 34234432,
      "step": 162215
    },
    {
      "epoch": 17.845984598459847,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0010476984225449727,
      "loss": 0.2319,
      "num_input_tokens_seen": 34235456,
      "step": 162220
    },
    {
      "epoch": 17.846534653465348,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00104716974560715,
      "loss": 0.2314,
      "num_input_tokens_seen": 34236512,
      "step": 162225
    },
    {
      "epoch": 17.847084708470845,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0010466411972658706,
      "loss": 0.2314,
      "num_input_tokens_seen": 34237600,
      "step": 162230
    },
    {
      "epoch": 17.847634763476346,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0010461127775260042,
      "loss": 0.2314,
      "num_input_tokens_seen": 34238688,
      "step": 162235
    },
    {
      "epoch": 17.848184818481847,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0010455844863924168,
      "loss": 0.2308,
      "num_input_tokens_seen": 34239744,
      "step": 162240
    },
    {
      "epoch": 17.84873487348735,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.001045056323869986,
      "loss": 0.2319,
      "num_input_tokens_seen": 34240800,
      "step": 162245
    },
    {
      "epoch": 17.84928492849285,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.0010445282899635717,
      "loss": 0.2303,
      "num_input_tokens_seen": 34241856,
      "step": 162250
    },
    {
      "epoch": 17.84983498349835,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0010440003846780427,
      "loss": 0.2324,
      "num_input_tokens_seen": 34242912,
      "step": 162255
    },
    {
      "epoch": 17.850385038503852,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.0010434726080182687,
      "loss": 0.2303,
      "num_input_tokens_seen": 34243904,
      "step": 162260
    },
    {
      "epoch": 17.85093509350935,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.0010429449599891078,
      "loss": 0.2329,
      "num_input_tokens_seen": 34244928,
      "step": 162265
    },
    {
      "epoch": 17.85148514851485,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.001042417440595429,
      "loss": 0.2308,
      "num_input_tokens_seen": 34245984,
      "step": 162270
    },
    {
      "epoch": 17.85203520352035,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0010418900498420868,
      "loss": 0.2319,
      "num_input_tokens_seen": 34246976,
      "step": 162275
    },
    {
      "epoch": 17.852585258525853,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0010413627877339494,
      "loss": 0.2293,
      "num_input_tokens_seen": 34248064,
      "step": 162280
    },
    {
      "epoch": 17.853135313531354,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.0010408356542758739,
      "loss": 0.2303,
      "num_input_tokens_seen": 34249088,
      "step": 162285
    },
    {
      "epoch": 17.853685368536855,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0010403086494727154,
      "loss": 0.2303,
      "num_input_tokens_seen": 34250208,
      "step": 162290
    },
    {
      "epoch": 17.854235423542356,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.0010397817733293364,
      "loss": 0.2308,
      "num_input_tokens_seen": 34251200,
      "step": 162295
    },
    {
      "epoch": 17.854785478547853,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0010392550258505862,
      "loss": 0.2314,
      "num_input_tokens_seen": 34252288,
      "step": 162300
    },
    {
      "epoch": 17.855335533553355,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0010387284070413278,
      "loss": 0.2314,
      "num_input_tokens_seen": 34253312,
      "step": 162305
    },
    {
      "epoch": 17.855885588558856,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.0010382019169064072,
      "loss": 0.2309,
      "num_input_tokens_seen": 34254304,
      "step": 162310
    },
    {
      "epoch": 17.856435643564357,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0010376755554506788,
      "loss": 0.2303,
      "num_input_tokens_seen": 34255328,
      "step": 162315
    },
    {
      "epoch": 17.856985698569858,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.001037149322678999,
      "loss": 0.2324,
      "num_input_tokens_seen": 34256480,
      "step": 162320
    },
    {
      "epoch": 17.85753575357536,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.001036623218596212,
      "loss": 0.2308,
      "num_input_tokens_seen": 34257536,
      "step": 162325
    },
    {
      "epoch": 17.858085808580856,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0010360972432071692,
      "loss": 0.2314,
      "num_input_tokens_seen": 34258560,
      "step": 162330
    },
    {
      "epoch": 17.858635863586358,
      "grad_norm": 0.000682830810546875,
      "learning_rate": 0.0010355713965167178,
      "loss": 0.2325,
      "num_input_tokens_seen": 34259616,
      "step": 162335
    },
    {
      "epoch": 17.85918591859186,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0010350456785297012,
      "loss": 0.2324,
      "num_input_tokens_seen": 34260672,
      "step": 162340
    },
    {
      "epoch": 17.85973597359736,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.0010345200892509703,
      "loss": 0.2319,
      "num_input_tokens_seen": 34261792,
      "step": 162345
    },
    {
      "epoch": 17.86028602860286,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.001033994628685368,
      "loss": 0.2304,
      "num_input_tokens_seen": 34262816,
      "step": 162350
    },
    {
      "epoch": 17.860836083608362,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0010334692968377319,
      "loss": 0.2303,
      "num_input_tokens_seen": 34263840,
      "step": 162355
    },
    {
      "epoch": 17.861386138613863,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0010329440937129097,
      "loss": 0.2319,
      "num_input_tokens_seen": 34264896,
      "step": 162360
    },
    {
      "epoch": 17.86193619361936,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.001032419019315738,
      "loss": 0.2308,
      "num_input_tokens_seen": 34265920,
      "step": 162365
    },
    {
      "epoch": 17.86248624862486,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.001031894073651059,
      "loss": 0.2329,
      "num_input_tokens_seen": 34267008,
      "step": 162370
    },
    {
      "epoch": 17.863036303630363,
      "grad_norm": 0.005859375,
      "learning_rate": 0.001031369256723711,
      "loss": 0.2303,
      "num_input_tokens_seen": 34268096,
      "step": 162375
    },
    {
      "epoch": 17.863586358635864,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0010308445685385265,
      "loss": 0.2319,
      "num_input_tokens_seen": 34269184,
      "step": 162380
    },
    {
      "epoch": 17.864136413641365,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0010303200091003484,
      "loss": 0.2319,
      "num_input_tokens_seen": 34270208,
      "step": 162385
    },
    {
      "epoch": 17.864686468646866,
      "grad_norm": 0.00080108642578125,
      "learning_rate": 0.0010297955784140044,
      "loss": 0.2314,
      "num_input_tokens_seen": 34271168,
      "step": 162390
    },
    {
      "epoch": 17.865236523652364,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0010292712764843309,
      "loss": 0.2335,
      "num_input_tokens_seen": 34272192,
      "step": 162395
    },
    {
      "epoch": 17.865786578657865,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0010287471033161638,
      "loss": 0.2298,
      "num_input_tokens_seen": 34273184,
      "step": 162400
    },
    {
      "epoch": 17.866336633663366,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0010282230589143259,
      "loss": 0.2319,
      "num_input_tokens_seen": 34274272,
      "step": 162405
    },
    {
      "epoch": 17.866886688668867,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0010276991432836568,
      "loss": 0.2319,
      "num_input_tokens_seen": 34275296,
      "step": 162410
    },
    {
      "epoch": 17.867436743674368,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.001027175356428976,
      "loss": 0.2335,
      "num_input_tokens_seen": 34276320,
      "step": 162415
    },
    {
      "epoch": 17.86798679867987,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0010266516983551176,
      "loss": 0.2298,
      "num_input_tokens_seen": 34277344,
      "step": 162420
    },
    {
      "epoch": 17.86853685368537,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.0010261281690669033,
      "loss": 0.2313,
      "num_input_tokens_seen": 34278432,
      "step": 162425
    },
    {
      "epoch": 17.869086908690868,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.0010256047685691604,
      "loss": 0.2309,
      "num_input_tokens_seen": 34279424,
      "step": 162430
    },
    {
      "epoch": 17.86963696369637,
      "grad_norm": 0.0006561279296875,
      "learning_rate": 0.0010250814968667155,
      "loss": 0.2308,
      "num_input_tokens_seen": 34280512,
      "step": 162435
    },
    {
      "epoch": 17.87018701870187,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0010245583539643859,
      "loss": 0.2314,
      "num_input_tokens_seen": 34281568,
      "step": 162440
    },
    {
      "epoch": 17.87073707370737,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.001024035339867,
      "loss": 0.2293,
      "num_input_tokens_seen": 34282656,
      "step": 162445
    },
    {
      "epoch": 17.871287128712872,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0010235124545793716,
      "loss": 0.2308,
      "num_input_tokens_seen": 34283744,
      "step": 162450
    },
    {
      "epoch": 17.871837183718373,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.001022989698106319,
      "loss": 0.2308,
      "num_input_tokens_seen": 34284800,
      "step": 162455
    },
    {
      "epoch": 17.87238723872387,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0010224670704526667,
      "loss": 0.2329,
      "num_input_tokens_seen": 34285792,
      "step": 162460
    },
    {
      "epoch": 17.872937293729372,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0010219445716232272,
      "loss": 0.2309,
      "num_input_tokens_seen": 34286816,
      "step": 162465
    },
    {
      "epoch": 17.873487348734873,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0010214222016228201,
      "loss": 0.2324,
      "num_input_tokens_seen": 34287872,
      "step": 162470
    },
    {
      "epoch": 17.874037403740374,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.0010208999604562551,
      "loss": 0.2303,
      "num_input_tokens_seen": 34288864,
      "step": 162475
    },
    {
      "epoch": 17.874587458745875,
      "grad_norm": 0.002593994140625,
      "learning_rate": 0.0010203778481283465,
      "loss": 0.2314,
      "num_input_tokens_seen": 34289952,
      "step": 162480
    },
    {
      "epoch": 17.875137513751376,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.001019855864643907,
      "loss": 0.2309,
      "num_input_tokens_seen": 34291008,
      "step": 162485
    },
    {
      "epoch": 17.875687568756877,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.001019334010007748,
      "loss": 0.2335,
      "num_input_tokens_seen": 34292096,
      "step": 162490
    },
    {
      "epoch": 17.876237623762375,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.0010188122842246805,
      "loss": 0.2319,
      "num_input_tokens_seen": 34293216,
      "step": 162495
    },
    {
      "epoch": 17.876787678767876,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0010182906872995123,
      "loss": 0.2335,
      "num_input_tokens_seen": 34294272,
      "step": 162500
    },
    {
      "epoch": 17.877337733773377,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0010177692192370462,
      "loss": 0.2313,
      "num_input_tokens_seen": 34295360,
      "step": 162505
    },
    {
      "epoch": 17.877887788778878,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0010172478800420954,
      "loss": 0.2329,
      "num_input_tokens_seen": 34296416,
      "step": 162510
    },
    {
      "epoch": 17.87843784378438,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.0010167266697194554,
      "loss": 0.2299,
      "num_input_tokens_seen": 34297504,
      "step": 162515
    },
    {
      "epoch": 17.87898789878988,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0010162055882739411,
      "loss": 0.234,
      "num_input_tokens_seen": 34298528,
      "step": 162520
    },
    {
      "epoch": 17.879537953795378,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.0010156846357103487,
      "loss": 0.2303,
      "num_input_tokens_seen": 34299552,
      "step": 162525
    },
    {
      "epoch": 17.88008800880088,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0010151638120334794,
      "loss": 0.2319,
      "num_input_tokens_seen": 34300704,
      "step": 162530
    },
    {
      "epoch": 17.88063806380638,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0010146431172481374,
      "loss": 0.2314,
      "num_input_tokens_seen": 34301792,
      "step": 162535
    },
    {
      "epoch": 17.88118811881188,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0010141225513591173,
      "loss": 0.2314,
      "num_input_tokens_seen": 34302816,
      "step": 162540
    },
    {
      "epoch": 17.881738173817382,
      "grad_norm": 0.0008087158203125,
      "learning_rate": 0.001013602114371217,
      "loss": 0.2329,
      "num_input_tokens_seen": 34303840,
      "step": 162545
    },
    {
      "epoch": 17.882288228822883,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.0010130818062892377,
      "loss": 0.2319,
      "num_input_tokens_seen": 34304896,
      "step": 162550
    },
    {
      "epoch": 17.882838283828384,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0010125616271179688,
      "loss": 0.2308,
      "num_input_tokens_seen": 34305952,
      "step": 162555
    },
    {
      "epoch": 17.883388338833882,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0010120415768622114,
      "loss": 0.2329,
      "num_input_tokens_seen": 34306976,
      "step": 162560
    },
    {
      "epoch": 17.883938393839383,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0010115216555267536,
      "loss": 0.2314,
      "num_input_tokens_seen": 34308096,
      "step": 162565
    },
    {
      "epoch": 17.884488448844884,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0010110018631163864,
      "loss": 0.2314,
      "num_input_tokens_seen": 34309120,
      "step": 162570
    },
    {
      "epoch": 17.885038503850385,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.001010482199635901,
      "loss": 0.2314,
      "num_input_tokens_seen": 34310112,
      "step": 162575
    },
    {
      "epoch": 17.885588558855886,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00100996266509009,
      "loss": 0.2319,
      "num_input_tokens_seen": 34311136,
      "step": 162580
    },
    {
      "epoch": 17.886138613861387,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0010094432594837416,
      "loss": 0.2293,
      "num_input_tokens_seen": 34312192,
      "step": 162585
    },
    {
      "epoch": 17.88668866886689,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0010089239828216417,
      "loss": 0.2309,
      "num_input_tokens_seen": 34313216,
      "step": 162590
    },
    {
      "epoch": 17.887238723872386,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.001008404835108575,
      "loss": 0.2303,
      "num_input_tokens_seen": 34314272,
      "step": 162595
    },
    {
      "epoch": 17.887788778877887,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.0010078858163493243,
      "loss": 0.2329,
      "num_input_tokens_seen": 34315296,
      "step": 162600
    },
    {
      "epoch": 17.888338833883388,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.0010073669265486772,
      "loss": 0.2309,
      "num_input_tokens_seen": 34316384,
      "step": 162605
    },
    {
      "epoch": 17.88888888888889,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.0010068481657114186,
      "loss": 0.2314,
      "num_input_tokens_seen": 34317536,
      "step": 162610
    },
    {
      "epoch": 17.88943894389439,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0010063295338423245,
      "loss": 0.2335,
      "num_input_tokens_seen": 34318592,
      "step": 162615
    },
    {
      "epoch": 17.88998899889989,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.0010058110309461743,
      "loss": 0.2319,
      "num_input_tokens_seen": 34319616,
      "step": 162620
    },
    {
      "epoch": 17.89053905390539,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.0010052926570277509,
      "loss": 0.2298,
      "num_input_tokens_seen": 34320768,
      "step": 162625
    },
    {
      "epoch": 17.89108910891089,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0010047744120918272,
      "loss": 0.2303,
      "num_input_tokens_seen": 34321920,
      "step": 162630
    },
    {
      "epoch": 17.89163916391639,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0010042562961431812,
      "loss": 0.234,
      "num_input_tokens_seen": 34322944,
      "step": 162635
    },
    {
      "epoch": 17.892189218921892,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0010037383091865935,
      "loss": 0.233,
      "num_input_tokens_seen": 34324000,
      "step": 162640
    },
    {
      "epoch": 17.892739273927393,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0010032204512268293,
      "loss": 0.2314,
      "num_input_tokens_seen": 34325120,
      "step": 162645
    },
    {
      "epoch": 17.893289328932894,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0010027027222686696,
      "loss": 0.2303,
      "num_input_tokens_seen": 34326240,
      "step": 162650
    },
    {
      "epoch": 17.893839383938392,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0010021851223168804,
      "loss": 0.2324,
      "num_input_tokens_seen": 34327264,
      "step": 162655
    },
    {
      "epoch": 17.894389438943893,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0010016676513762328,
      "loss": 0.2319,
      "num_input_tokens_seen": 34328256,
      "step": 162660
    },
    {
      "epoch": 17.894939493949394,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0010011503094515017,
      "loss": 0.2303,
      "num_input_tokens_seen": 34329344,
      "step": 162665
    },
    {
      "epoch": 17.895489548954895,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.0010006330965474464,
      "loss": 0.2324,
      "num_input_tokens_seen": 34330464,
      "step": 162670
    },
    {
      "epoch": 17.896039603960396,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0010001160126688429,
      "loss": 0.2319,
      "num_input_tokens_seen": 34331616,
      "step": 162675
    },
    {
      "epoch": 17.896589658965897,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0009995990578204478,
      "loss": 0.2314,
      "num_input_tokens_seen": 34332704,
      "step": 162680
    },
    {
      "epoch": 17.8971397139714,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0009990822320070335,
      "loss": 0.2319,
      "num_input_tokens_seen": 34333728,
      "step": 162685
    },
    {
      "epoch": 17.897689768976896,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.0009985655352333584,
      "loss": 0.2313,
      "num_input_tokens_seen": 34334752,
      "step": 162690
    },
    {
      "epoch": 17.898239823982397,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0009980489675041847,
      "loss": 0.2303,
      "num_input_tokens_seen": 34335776,
      "step": 162695
    },
    {
      "epoch": 17.8987898789879,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0009975325288242775,
      "loss": 0.2314,
      "num_input_tokens_seen": 34336832,
      "step": 162700
    },
    {
      "epoch": 17.8993399339934,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0009970162191983928,
      "loss": 0.2319,
      "num_input_tokens_seen": 34337888,
      "step": 162705
    },
    {
      "epoch": 17.8998899889989,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0009965000386312917,
      "loss": 0.2298,
      "num_input_tokens_seen": 34338880,
      "step": 162710
    },
    {
      "epoch": 17.9004400440044,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.000995983987127727,
      "loss": 0.2314,
      "num_input_tokens_seen": 34339936,
      "step": 162715
    },
    {
      "epoch": 17.900990099009903,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0009954680646924602,
      "loss": 0.2335,
      "num_input_tokens_seen": 34340992,
      "step": 162720
    },
    {
      "epoch": 17.9015401540154,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0009949522713302438,
      "loss": 0.2303,
      "num_input_tokens_seen": 34342048,
      "step": 162725
    },
    {
      "epoch": 17.9020902090209,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0009944366070458344,
      "loss": 0.2319,
      "num_input_tokens_seen": 34343104,
      "step": 162730
    },
    {
      "epoch": 17.902640264026402,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0009939210718439794,
      "loss": 0.2308,
      "num_input_tokens_seen": 34344096,
      "step": 162735
    },
    {
      "epoch": 17.903190319031903,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0009934056657294355,
      "loss": 0.234,
      "num_input_tokens_seen": 34345152,
      "step": 162740
    },
    {
      "epoch": 17.903740374037405,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0009928903887069483,
      "loss": 0.2319,
      "num_input_tokens_seen": 34346208,
      "step": 162745
    },
    {
      "epoch": 17.904290429042906,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.000992375240781268,
      "loss": 0.2303,
      "num_input_tokens_seen": 34347296,
      "step": 162750
    },
    {
      "epoch": 17.904840484048403,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.0009918602219571471,
      "loss": 0.2335,
      "num_input_tokens_seen": 34348416,
      "step": 162755
    },
    {
      "epoch": 17.905390539053904,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.000991345332239325,
      "loss": 0.2314,
      "num_input_tokens_seen": 34349408,
      "step": 162760
    },
    {
      "epoch": 17.905940594059405,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.000990830571632555,
      "loss": 0.2319,
      "num_input_tokens_seen": 34350432,
      "step": 162765
    },
    {
      "epoch": 17.906490649064907,
      "grad_norm": 0.000537872314453125,
      "learning_rate": 0.0009903159401415745,
      "loss": 0.2324,
      "num_input_tokens_seen": 34351424,
      "step": 162770
    },
    {
      "epoch": 17.907040704070408,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0009898014377711301,
      "loss": 0.2309,
      "num_input_tokens_seen": 34352480,
      "step": 162775
    },
    {
      "epoch": 17.90759075907591,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0009892870645259644,
      "loss": 0.2309,
      "num_input_tokens_seen": 34353568,
      "step": 162780
    },
    {
      "epoch": 17.90814081408141,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0009887728204108154,
      "loss": 0.2298,
      "num_input_tokens_seen": 34354592,
      "step": 162785
    },
    {
      "epoch": 17.908690869086907,
      "grad_norm": 0.0005035400390625,
      "learning_rate": 0.0009882587054304242,
      "loss": 0.2298,
      "num_input_tokens_seen": 34355648,
      "step": 162790
    },
    {
      "epoch": 17.90924092409241,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0009877447195895288,
      "loss": 0.2324,
      "num_input_tokens_seen": 34356736,
      "step": 162795
    },
    {
      "epoch": 17.90979097909791,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0009872308628928689,
      "loss": 0.2319,
      "num_input_tokens_seen": 34357696,
      "step": 162800
    },
    {
      "epoch": 17.91034103410341,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0009867171353451754,
      "loss": 0.2298,
      "num_input_tokens_seen": 34358752,
      "step": 162805
    },
    {
      "epoch": 17.91089108910891,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.0009862035369511862,
      "loss": 0.2308,
      "num_input_tokens_seen": 34359808,
      "step": 162810
    },
    {
      "epoch": 17.911441144114413,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.000985690067715636,
      "loss": 0.2319,
      "num_input_tokens_seen": 34360864,
      "step": 162815
    },
    {
      "epoch": 17.91199119911991,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0009851767276432544,
      "loss": 0.2329,
      "num_input_tokens_seen": 34361952,
      "step": 162820
    },
    {
      "epoch": 17.91254125412541,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0009846635167387757,
      "loss": 0.2298,
      "num_input_tokens_seen": 34363008,
      "step": 162825
    },
    {
      "epoch": 17.913091309130913,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.0009841504350069246,
      "loss": 0.2329,
      "num_input_tokens_seen": 34364032,
      "step": 162830
    },
    {
      "epoch": 17.913641364136414,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0009836374824524357,
      "loss": 0.2314,
      "num_input_tokens_seen": 34365088,
      "step": 162835
    },
    {
      "epoch": 17.914191419141915,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0009831246590800353,
      "loss": 0.2314,
      "num_input_tokens_seen": 34366208,
      "step": 162840
    },
    {
      "epoch": 17.914741474147416,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0009826119648944475,
      "loss": 0.2293,
      "num_input_tokens_seen": 34367232,
      "step": 162845
    },
    {
      "epoch": 17.915291529152917,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0009820993999004007,
      "loss": 0.2319,
      "num_input_tokens_seen": 34368320,
      "step": 162850
    },
    {
      "epoch": 17.915841584158414,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0009815869641026176,
      "loss": 0.2324,
      "num_input_tokens_seen": 34369344,
      "step": 162855
    },
    {
      "epoch": 17.916391639163916,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0009810746575058176,
      "loss": 0.2314,
      "num_input_tokens_seen": 34370400,
      "step": 162860
    },
    {
      "epoch": 17.916941694169417,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.0009805624801147272,
      "loss": 0.234,
      "num_input_tokens_seen": 34371456,
      "step": 162865
    },
    {
      "epoch": 17.917491749174918,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0009800504319340641,
      "loss": 0.2303,
      "num_input_tokens_seen": 34372512,
      "step": 162870
    },
    {
      "epoch": 17.91804180418042,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0009795385129685514,
      "loss": 0.2314,
      "num_input_tokens_seen": 34373472,
      "step": 162875
    },
    {
      "epoch": 17.91859185918592,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.000979026723222905,
      "loss": 0.2314,
      "num_input_tokens_seen": 34374528,
      "step": 162880
    },
    {
      "epoch": 17.919141914191417,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.000978515062701838,
      "loss": 0.2298,
      "num_input_tokens_seen": 34375552,
      "step": 162885
    },
    {
      "epoch": 17.91969196919692,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.0009780035314100699,
      "loss": 0.2308,
      "num_input_tokens_seen": 34376672,
      "step": 162890
    },
    {
      "epoch": 17.92024202420242,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0009774921293523153,
      "loss": 0.2314,
      "num_input_tokens_seen": 34377760,
      "step": 162895
    },
    {
      "epoch": 17.92079207920792,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0009769808565332889,
      "loss": 0.2308,
      "num_input_tokens_seen": 34378816,
      "step": 162900
    },
    {
      "epoch": 17.921342134213422,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.0009764697129577015,
      "loss": 0.2298,
      "num_input_tokens_seen": 34379840,
      "step": 162905
    },
    {
      "epoch": 17.921892189218923,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0009759586986302615,
      "loss": 0.2309,
      "num_input_tokens_seen": 34380896,
      "step": 162910
    },
    {
      "epoch": 17.922442244224424,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.0009754478135556815,
      "loss": 0.2319,
      "num_input_tokens_seen": 34382016,
      "step": 162915
    },
    {
      "epoch": 17.92299229922992,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0009749370577386678,
      "loss": 0.2308,
      "num_input_tokens_seen": 34383072,
      "step": 162920
    },
    {
      "epoch": 17.923542354235423,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0009744264311839301,
      "loss": 0.2329,
      "num_input_tokens_seen": 34384160,
      "step": 162925
    },
    {
      "epoch": 17.924092409240924,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0009739159338961761,
      "loss": 0.2329,
      "num_input_tokens_seen": 34385216,
      "step": 162930
    },
    {
      "epoch": 17.924642464246425,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0009734055658801055,
      "loss": 0.2319,
      "num_input_tokens_seen": 34386208,
      "step": 162935
    },
    {
      "epoch": 17.925192519251926,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0009728953271404294,
      "loss": 0.2334,
      "num_input_tokens_seen": 34387328,
      "step": 162940
    },
    {
      "epoch": 17.925742574257427,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0009723852176818426,
      "loss": 0.2329,
      "num_input_tokens_seen": 34388416,
      "step": 162945
    },
    {
      "epoch": 17.926292629262925,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0009718752375090494,
      "loss": 0.2309,
      "num_input_tokens_seen": 34389472,
      "step": 162950
    },
    {
      "epoch": 17.926842684268426,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0009713653866267529,
      "loss": 0.2303,
      "num_input_tokens_seen": 34390496,
      "step": 162955
    },
    {
      "epoch": 17.927392739273927,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0009708556650396493,
      "loss": 0.2308,
      "num_input_tokens_seen": 34391488,
      "step": 162960
    },
    {
      "epoch": 17.927942794279428,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0009703460727524382,
      "loss": 0.2308,
      "num_input_tokens_seen": 34392576,
      "step": 162965
    },
    {
      "epoch": 17.92849284928493,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0009698366097698158,
      "loss": 0.2324,
      "num_input_tokens_seen": 34393600,
      "step": 162970
    },
    {
      "epoch": 17.92904290429043,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.000969327276096475,
      "loss": 0.2314,
      "num_input_tokens_seen": 34394624,
      "step": 162975
    },
    {
      "epoch": 17.92959295929593,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.000968818071737112,
      "loss": 0.2298,
      "num_input_tokens_seen": 34395680,
      "step": 162980
    },
    {
      "epoch": 17.93014301430143,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0009683089966964197,
      "loss": 0.2298,
      "num_input_tokens_seen": 34396704,
      "step": 162985
    },
    {
      "epoch": 17.93069306930693,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0009678000509790929,
      "loss": 0.2303,
      "num_input_tokens_seen": 34397760,
      "step": 162990
    },
    {
      "epoch": 17.93124312431243,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0009672912345898194,
      "loss": 0.2314,
      "num_input_tokens_seen": 34398816,
      "step": 162995
    },
    {
      "epoch": 17.931793179317932,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.000966782547533287,
      "loss": 0.2319,
      "num_input_tokens_seen": 34399872,
      "step": 163000
    },
    {
      "epoch": 17.932343234323433,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.0009662739898141869,
      "loss": 0.2319,
      "num_input_tokens_seen": 34400992,
      "step": 163005
    },
    {
      "epoch": 17.932893289328934,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.000965765561437204,
      "loss": 0.2309,
      "num_input_tokens_seen": 34402080,
      "step": 163010
    },
    {
      "epoch": 17.933443344334435,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0009652572624070293,
      "loss": 0.2319,
      "num_input_tokens_seen": 34403168,
      "step": 163015
    },
    {
      "epoch": 17.933993399339933,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0009647490927283425,
      "loss": 0.2314,
      "num_input_tokens_seen": 34404224,
      "step": 163020
    },
    {
      "epoch": 17.934543454345434,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0009642410524058264,
      "loss": 0.2298,
      "num_input_tokens_seen": 34405344,
      "step": 163025
    },
    {
      "epoch": 17.935093509350935,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.000963733141444169,
      "loss": 0.2303,
      "num_input_tokens_seen": 34406432,
      "step": 163030
    },
    {
      "epoch": 17.935643564356436,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.000963225359848045,
      "loss": 0.2319,
      "num_input_tokens_seen": 34407488,
      "step": 163035
    },
    {
      "epoch": 17.936193619361937,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0009627177076221371,
      "loss": 0.2314,
      "num_input_tokens_seen": 34408448,
      "step": 163040
    },
    {
      "epoch": 17.936743674367438,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0009622101847711267,
      "loss": 0.2324,
      "num_input_tokens_seen": 34409440,
      "step": 163045
    },
    {
      "epoch": 17.937293729372936,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0009617027912996867,
      "loss": 0.2319,
      "num_input_tokens_seen": 34410432,
      "step": 163050
    },
    {
      "epoch": 17.937843784378437,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0009611955272124984,
      "loss": 0.2314,
      "num_input_tokens_seen": 34411488,
      "step": 163055
    },
    {
      "epoch": 17.938393839383938,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.0009606883925142312,
      "loss": 0.2314,
      "num_input_tokens_seen": 34412512,
      "step": 163060
    },
    {
      "epoch": 17.93894389438944,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0009601813872095632,
      "loss": 0.2277,
      "num_input_tokens_seen": 34413664,
      "step": 163065
    },
    {
      "epoch": 17.93949394939494,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0009596745113031691,
      "loss": 0.2319,
      "num_input_tokens_seen": 34414784,
      "step": 163070
    },
    {
      "epoch": 17.94004400440044,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0009591677647997148,
      "loss": 0.2303,
      "num_input_tokens_seen": 34415904,
      "step": 163075
    },
    {
      "epoch": 17.94059405940594,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.000958661147703877,
      "loss": 0.2308,
      "num_input_tokens_seen": 34416960,
      "step": 163080
    },
    {
      "epoch": 17.94114411441144,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.0009581546600203184,
      "loss": 0.234,
      "num_input_tokens_seen": 34417952,
      "step": 163085
    },
    {
      "epoch": 17.94169416941694,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0009576483017537135,
      "loss": 0.2324,
      "num_input_tokens_seen": 34418944,
      "step": 163090
    },
    {
      "epoch": 17.942244224422442,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0009571420729087238,
      "loss": 0.2303,
      "num_input_tokens_seen": 34420000,
      "step": 163095
    },
    {
      "epoch": 17.942794279427943,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0009566359734900153,
      "loss": 0.2319,
      "num_input_tokens_seen": 34421088,
      "step": 163100
    },
    {
      "epoch": 17.943344334433444,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.0009561300035022596,
      "loss": 0.2319,
      "num_input_tokens_seen": 34422176,
      "step": 163105
    },
    {
      "epoch": 17.943894389438945,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.0009556241629501127,
      "loss": 0.2304,
      "num_input_tokens_seen": 34423200,
      "step": 163110
    },
    {
      "epoch": 17.944444444444443,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.000955118451838236,
      "loss": 0.2324,
      "num_input_tokens_seen": 34424224,
      "step": 163115
    },
    {
      "epoch": 17.944994499449944,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0009546128701712974,
      "loss": 0.2319,
      "num_input_tokens_seen": 34425280,
      "step": 163120
    },
    {
      "epoch": 17.945544554455445,
      "grad_norm": 0.01055908203125,
      "learning_rate": 0.0009541074179539466,
      "loss": 0.2309,
      "num_input_tokens_seen": 34426304,
      "step": 163125
    },
    {
      "epoch": 17.946094609460946,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.000953602095190853,
      "loss": 0.2319,
      "num_input_tokens_seen": 34427392,
      "step": 163130
    },
    {
      "epoch": 17.946644664466447,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0009530969018866681,
      "loss": 0.2324,
      "num_input_tokens_seen": 34428448,
      "step": 163135
    },
    {
      "epoch": 17.94719471947195,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0009525918380460462,
      "loss": 0.2303,
      "num_input_tokens_seen": 34429472,
      "step": 163140
    },
    {
      "epoch": 17.94774477447745,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0009520869036736473,
      "loss": 0.2298,
      "num_input_tokens_seen": 34430528,
      "step": 163145
    },
    {
      "epoch": 17.948294829482947,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.0009515820987741208,
      "loss": 0.2324,
      "num_input_tokens_seen": 34431616,
      "step": 163150
    },
    {
      "epoch": 17.948844884488448,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0009510774233521213,
      "loss": 0.2314,
      "num_input_tokens_seen": 34432768,
      "step": 163155
    },
    {
      "epoch": 17.94939493949395,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.0009505728774123018,
      "loss": 0.2303,
      "num_input_tokens_seen": 34433728,
      "step": 163160
    },
    {
      "epoch": 17.94994499449945,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0009500684609593068,
      "loss": 0.2314,
      "num_input_tokens_seen": 34434784,
      "step": 163165
    },
    {
      "epoch": 17.95049504950495,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0009495641739977928,
      "loss": 0.2313,
      "num_input_tokens_seen": 34435808,
      "step": 163170
    },
    {
      "epoch": 17.951045104510452,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0009490600165324008,
      "loss": 0.2324,
      "num_input_tokens_seen": 34436832,
      "step": 163175
    },
    {
      "epoch": 17.95159515951595,
      "grad_norm": 0.0030670166015625,
      "learning_rate": 0.0009485559885677807,
      "loss": 0.2319,
      "num_input_tokens_seen": 34437856,
      "step": 163180
    },
    {
      "epoch": 17.95214521452145,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0009480520901085803,
      "loss": 0.2308,
      "num_input_tokens_seen": 34438880,
      "step": 163185
    },
    {
      "epoch": 17.952695269526952,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.0009475483211594376,
      "loss": 0.2303,
      "num_input_tokens_seen": 34439936,
      "step": 163190
    },
    {
      "epoch": 17.953245324532453,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.0009470446817250022,
      "loss": 0.2314,
      "num_input_tokens_seen": 34440992,
      "step": 163195
    },
    {
      "epoch": 17.953795379537954,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0009465411718099103,
      "loss": 0.2329,
      "num_input_tokens_seen": 34442016,
      "step": 163200
    },
    {
      "epoch": 17.954345434543455,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.0009460377914188067,
      "loss": 0.2308,
      "num_input_tokens_seen": 34443136,
      "step": 163205
    },
    {
      "epoch": 17.954895489548957,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0009455345405563275,
      "loss": 0.2303,
      "num_input_tokens_seen": 34444192,
      "step": 163210
    },
    {
      "epoch": 17.955445544554454,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.0009450314192271109,
      "loss": 0.2324,
      "num_input_tokens_seen": 34445248,
      "step": 163215
    },
    {
      "epoch": 17.955995599559955,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0009445284274357996,
      "loss": 0.2319,
      "num_input_tokens_seen": 34446336,
      "step": 163220
    },
    {
      "epoch": 17.956545654565456,
      "grad_norm": 0.000576019287109375,
      "learning_rate": 0.0009440255651870216,
      "loss": 0.2319,
      "num_input_tokens_seen": 34447392,
      "step": 163225
    },
    {
      "epoch": 17.957095709570957,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0009435228324854183,
      "loss": 0.2319,
      "num_input_tokens_seen": 34448416,
      "step": 163230
    },
    {
      "epoch": 17.95764576457646,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0009430202293356193,
      "loss": 0.2329,
      "num_input_tokens_seen": 34449472,
      "step": 163235
    },
    {
      "epoch": 17.95819581958196,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0009425177557422543,
      "loss": 0.2319,
      "num_input_tokens_seen": 34450592,
      "step": 163240
    },
    {
      "epoch": 17.958745874587457,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0009420154117099627,
      "loss": 0.2324,
      "num_input_tokens_seen": 34451552,
      "step": 163245
    },
    {
      "epoch": 17.959295929592958,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 0.000941513197243366,
      "loss": 0.2324,
      "num_input_tokens_seen": 34452576,
      "step": 163250
    },
    {
      "epoch": 17.95984598459846,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0009410111123470987,
      "loss": 0.2283,
      "num_input_tokens_seen": 34453568,
      "step": 163255
    },
    {
      "epoch": 17.96039603960396,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.0009405091570257856,
      "loss": 0.2303,
      "num_input_tokens_seen": 34454560,
      "step": 163260
    },
    {
      "epoch": 17.96094609460946,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.0009400073312840512,
      "loss": 0.2324,
      "num_input_tokens_seen": 34455616,
      "step": 163265
    },
    {
      "epoch": 17.961496149614963,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.0009395056351265218,
      "loss": 0.2298,
      "num_input_tokens_seen": 34456704,
      "step": 163270
    },
    {
      "epoch": 17.962046204620464,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0009390040685578221,
      "loss": 0.2324,
      "num_input_tokens_seen": 34457728,
      "step": 163275
    },
    {
      "epoch": 17.96259625962596,
      "grad_norm": 0.00069427490234375,
      "learning_rate": 0.0009385026315825784,
      "loss": 0.2314,
      "num_input_tokens_seen": 34458816,
      "step": 163280
    },
    {
      "epoch": 17.963146314631462,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.0009380013242054086,
      "loss": 0.2314,
      "num_input_tokens_seen": 34459840,
      "step": 163285
    },
    {
      "epoch": 17.963696369636963,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.000937500146430929,
      "loss": 0.2314,
      "num_input_tokens_seen": 34460864,
      "step": 163290
    },
    {
      "epoch": 17.964246424642464,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.000936999098263766,
      "loss": 0.2293,
      "num_input_tokens_seen": 34461856,
      "step": 163295
    },
    {
      "epoch": 17.964796479647966,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0009364981797085325,
      "loss": 0.2298,
      "num_input_tokens_seen": 34462912,
      "step": 163300
    },
    {
      "epoch": 17.965346534653467,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0009359973907698465,
      "loss": 0.2303,
      "num_input_tokens_seen": 34463936,
      "step": 163305
    },
    {
      "epoch": 17.965896589658964,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.000935496731452326,
      "loss": 0.2319,
      "num_input_tokens_seen": 34464960,
      "step": 163310
    },
    {
      "epoch": 17.966446644664465,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.0009349962017605806,
      "loss": 0.2319,
      "num_input_tokens_seen": 34466016,
      "step": 163315
    },
    {
      "epoch": 17.966996699669966,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0009344958016992299,
      "loss": 0.2319,
      "num_input_tokens_seen": 34467104,
      "step": 163320
    },
    {
      "epoch": 17.967546754675467,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0009339955312728786,
      "loss": 0.2303,
      "num_input_tokens_seen": 34468192,
      "step": 163325
    },
    {
      "epoch": 17.96809680968097,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0009334953904861415,
      "loss": 0.2324,
      "num_input_tokens_seen": 34469248,
      "step": 163330
    },
    {
      "epoch": 17.96864686468647,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.0009329953793436296,
      "loss": 0.2303,
      "num_input_tokens_seen": 34470336,
      "step": 163335
    },
    {
      "epoch": 17.96919691969197,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0009324954978499478,
      "loss": 0.2303,
      "num_input_tokens_seen": 34471392,
      "step": 163340
    },
    {
      "epoch": 17.96974697469747,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.0009319957460097055,
      "loss": 0.2324,
      "num_input_tokens_seen": 34472448,
      "step": 163345
    },
    {
      "epoch": 17.97029702970297,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0009314961238275093,
      "loss": 0.2303,
      "num_input_tokens_seen": 34473504,
      "step": 163350
    },
    {
      "epoch": 17.97084708470847,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0009309966313079587,
      "loss": 0.2308,
      "num_input_tokens_seen": 34474560,
      "step": 163355
    },
    {
      "epoch": 17.97139713971397,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0009304972684556634,
      "loss": 0.2319,
      "num_input_tokens_seen": 34475584,
      "step": 163360
    },
    {
      "epoch": 17.971947194719473,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0009299980352752229,
      "loss": 0.2309,
      "num_input_tokens_seen": 34476672,
      "step": 163365
    },
    {
      "epoch": 17.972497249724974,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.0009294989317712404,
      "loss": 0.2319,
      "num_input_tokens_seen": 34477696,
      "step": 163370
    },
    {
      "epoch": 17.97304730473047,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0009289999579483154,
      "loss": 0.2319,
      "num_input_tokens_seen": 34478720,
      "step": 163375
    },
    {
      "epoch": 17.973597359735972,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.0009285011138110427,
      "loss": 0.2324,
      "num_input_tokens_seen": 34479808,
      "step": 163380
    },
    {
      "epoch": 17.974147414741473,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0009280023993640234,
      "loss": 0.2319,
      "num_input_tokens_seen": 34480864,
      "step": 163385
    },
    {
      "epoch": 17.974697469746975,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0009275038146118541,
      "loss": 0.2308,
      "num_input_tokens_seen": 34481856,
      "step": 163390
    },
    {
      "epoch": 17.975247524752476,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.000927005359559131,
      "loss": 0.2319,
      "num_input_tokens_seen": 34482880,
      "step": 163395
    },
    {
      "epoch": 17.975797579757977,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.000926507034210447,
      "loss": 0.2308,
      "num_input_tokens_seen": 34483936,
      "step": 163400
    },
    {
      "epoch": 17.976347634763478,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0009260088385703934,
      "loss": 0.2308,
      "num_input_tokens_seen": 34484992,
      "step": 163405
    },
    {
      "epoch": 17.976897689768975,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.0009255107726435634,
      "loss": 0.2319,
      "num_input_tokens_seen": 34486016,
      "step": 163410
    },
    {
      "epoch": 17.977447744774476,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0009250128364345466,
      "loss": 0.2324,
      "num_input_tokens_seen": 34487104,
      "step": 163415
    },
    {
      "epoch": 17.977997799779978,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.000924515029947931,
      "loss": 0.2313,
      "num_input_tokens_seen": 34488128,
      "step": 163420
    },
    {
      "epoch": 17.97854785478548,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0009240173531883094,
      "loss": 0.2308,
      "num_input_tokens_seen": 34489152,
      "step": 163425
    },
    {
      "epoch": 17.97909790979098,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.0009235198061602634,
      "loss": 0.2293,
      "num_input_tokens_seen": 34490304,
      "step": 163430
    },
    {
      "epoch": 17.97964796479648,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0009230223888683825,
      "loss": 0.2319,
      "num_input_tokens_seen": 34491392,
      "step": 163435
    },
    {
      "epoch": 17.980198019801982,
      "grad_norm": 0.000579833984375,
      "learning_rate": 0.000922525101317248,
      "loss": 0.2298,
      "num_input_tokens_seen": 34492480,
      "step": 163440
    },
    {
      "epoch": 17.98074807480748,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0009220279435114431,
      "loss": 0.2319,
      "num_input_tokens_seen": 34493568,
      "step": 163445
    },
    {
      "epoch": 17.98129812981298,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.000921530915455554,
      "loss": 0.2304,
      "num_input_tokens_seen": 34494656,
      "step": 163450
    },
    {
      "epoch": 17.98184818481848,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0009210340171541587,
      "loss": 0.2309,
      "num_input_tokens_seen": 34495680,
      "step": 163455
    },
    {
      "epoch": 17.982398239823983,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0009205372486118368,
      "loss": 0.2314,
      "num_input_tokens_seen": 34496736,
      "step": 163460
    },
    {
      "epoch": 17.982948294829484,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0009200406098331665,
      "loss": 0.2329,
      "num_input_tokens_seen": 34497824,
      "step": 163465
    },
    {
      "epoch": 17.983498349834985,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0009195441008227272,
      "loss": 0.2303,
      "num_input_tokens_seen": 34498848,
      "step": 163470
    },
    {
      "epoch": 17.984048404840483,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0009190477215850922,
      "loss": 0.2319,
      "num_input_tokens_seen": 34499904,
      "step": 163475
    },
    {
      "epoch": 17.984598459845984,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0009185514721248361,
      "loss": 0.2313,
      "num_input_tokens_seen": 34500960,
      "step": 163480
    },
    {
      "epoch": 17.985148514851485,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0009180553524465368,
      "loss": 0.2319,
      "num_input_tokens_seen": 34501984,
      "step": 163485
    },
    {
      "epoch": 17.985698569856986,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0009175593625547623,
      "loss": 0.2324,
      "num_input_tokens_seen": 34503040,
      "step": 163490
    },
    {
      "epoch": 17.986248624862487,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.000917063502454089,
      "loss": 0.2303,
      "num_input_tokens_seen": 34504160,
      "step": 163495
    },
    {
      "epoch": 17.986798679867988,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00091656777214908,
      "loss": 0.2314,
      "num_input_tokens_seen": 34505152,
      "step": 163500
    },
    {
      "epoch": 17.98734873487349,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.0009160721716443081,
      "loss": 0.2298,
      "num_input_tokens_seen": 34506176,
      "step": 163505
    },
    {
      "epoch": 17.987898789878987,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0009155767009443432,
      "loss": 0.2319,
      "num_input_tokens_seen": 34507168,
      "step": 163510
    },
    {
      "epoch": 17.988448844884488,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0009150813600537499,
      "loss": 0.2314,
      "num_input_tokens_seen": 34508224,
      "step": 163515
    },
    {
      "epoch": 17.98899889988999,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0009145861489770912,
      "loss": 0.2319,
      "num_input_tokens_seen": 34509312,
      "step": 163520
    },
    {
      "epoch": 17.98954895489549,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.0009140910677189351,
      "loss": 0.2319,
      "num_input_tokens_seen": 34510400,
      "step": 163525
    },
    {
      "epoch": 17.99009900990099,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0009135961162838396,
      "loss": 0.2324,
      "num_input_tokens_seen": 34511392,
      "step": 163530
    },
    {
      "epoch": 17.990649064906492,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0009131012946763694,
      "loss": 0.2288,
      "num_input_tokens_seen": 34512448,
      "step": 163535
    },
    {
      "epoch": 17.99119911991199,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0009126066029010876,
      "loss": 0.2309,
      "num_input_tokens_seen": 34513504,
      "step": 163540
    },
    {
      "epoch": 17.99174917491749,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.0009121120409625488,
      "loss": 0.2319,
      "num_input_tokens_seen": 34514528,
      "step": 163545
    },
    {
      "epoch": 17.992299229922992,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.000911617608865316,
      "loss": 0.2303,
      "num_input_tokens_seen": 34515616,
      "step": 163550
    },
    {
      "epoch": 17.992849284928493,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0009111233066139406,
      "loss": 0.2298,
      "num_input_tokens_seen": 34516672,
      "step": 163555
    },
    {
      "epoch": 17.993399339933994,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0009106291342129807,
      "loss": 0.2324,
      "num_input_tokens_seen": 34517760,
      "step": 163560
    },
    {
      "epoch": 17.993949394939495,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0009101350916669942,
      "loss": 0.2329,
      "num_input_tokens_seen": 34518816,
      "step": 163565
    },
    {
      "epoch": 17.994499449944996,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0009096411789805292,
      "loss": 0.2314,
      "num_input_tokens_seen": 34519904,
      "step": 163570
    },
    {
      "epoch": 17.995049504950494,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0009091473961581436,
      "loss": 0.2319,
      "num_input_tokens_seen": 34520928,
      "step": 163575
    },
    {
      "epoch": 17.995599559955995,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0009086537432043823,
      "loss": 0.2319,
      "num_input_tokens_seen": 34521984,
      "step": 163580
    },
    {
      "epoch": 17.996149614961496,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0009081602201237981,
      "loss": 0.2298,
      "num_input_tokens_seen": 34523040,
      "step": 163585
    },
    {
      "epoch": 17.996699669966997,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0009076668269209392,
      "loss": 0.2319,
      "num_input_tokens_seen": 34524128,
      "step": 163590
    },
    {
      "epoch": 17.997249724972498,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.0009071735636003519,
      "loss": 0.2314,
      "num_input_tokens_seen": 34525184,
      "step": 163595
    },
    {
      "epoch": 17.997799779978,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0009066804301665859,
      "loss": 0.2308,
      "num_input_tokens_seen": 34526176,
      "step": 163600
    },
    {
      "epoch": 17.998349834983497,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0009061874266241809,
      "loss": 0.2319,
      "num_input_tokens_seen": 34527232,
      "step": 163605
    },
    {
      "epoch": 17.998899889988998,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0009056945529776866,
      "loss": 0.2324,
      "num_input_tokens_seen": 34528224,
      "step": 163610
    },
    {
      "epoch": 17.9994499449945,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.000905201809231641,
      "loss": 0.2324,
      "num_input_tokens_seen": 34529184,
      "step": 163615
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0009047091953905855,
      "loss": 0.2309,
      "num_input_tokens_seen": 34530176,
      "step": 163620
    },
    {
      "epoch": 18.0,
      "eval_loss": 0.231306254863739,
      "eval_runtime": 60.5888,
      "eval_samples_per_second": 66.679,
      "eval_steps_per_second": 16.67,
      "num_input_tokens_seen": 34530176,
      "step": 163620
    },
    {
      "epoch": 18.0005500550055,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.0009042167114590648,
      "loss": 0.2309,
      "num_input_tokens_seen": 34531232,
      "step": 163625
    },
    {
      "epoch": 18.001100110011002,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.0009037243574416121,
      "loss": 0.2298,
      "num_input_tokens_seen": 34532256,
      "step": 163630
    },
    {
      "epoch": 18.001650165016503,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.00090323213334277,
      "loss": 0.2309,
      "num_input_tokens_seen": 34533344,
      "step": 163635
    },
    {
      "epoch": 18.002200220022,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.0009027400391670736,
      "loss": 0.2298,
      "num_input_tokens_seen": 34534400,
      "step": 163640
    },
    {
      "epoch": 18.002750275027502,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.0009022480749190559,
      "loss": 0.2314,
      "num_input_tokens_seen": 34535424,
      "step": 163645
    },
    {
      "epoch": 18.003300330033003,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0009017562406032531,
      "loss": 0.2319,
      "num_input_tokens_seen": 34536480,
      "step": 163650
    },
    {
      "epoch": 18.003850385038504,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.0009012645362241966,
      "loss": 0.2308,
      "num_input_tokens_seen": 34537504,
      "step": 163655
    },
    {
      "epoch": 18.004400440044005,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.000900772961786423,
      "loss": 0.2304,
      "num_input_tokens_seen": 34538624,
      "step": 163660
    },
    {
      "epoch": 18.004950495049506,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0009002815172944583,
      "loss": 0.2324,
      "num_input_tokens_seen": 34539616,
      "step": 163665
    },
    {
      "epoch": 18.005500550055004,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0008997902027528326,
      "loss": 0.2314,
      "num_input_tokens_seen": 34540672,
      "step": 163670
    },
    {
      "epoch": 18.006050605060505,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0008992990181660737,
      "loss": 0.2314,
      "num_input_tokens_seen": 34541760,
      "step": 163675
    },
    {
      "epoch": 18.006600660066006,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0008988079635387097,
      "loss": 0.2308,
      "num_input_tokens_seen": 34542816,
      "step": 163680
    },
    {
      "epoch": 18.007150715071507,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.0008983170388752687,
      "loss": 0.2308,
      "num_input_tokens_seen": 34543840,
      "step": 163685
    },
    {
      "epoch": 18.007700770077008,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0008978262441802737,
      "loss": 0.2319,
      "num_input_tokens_seen": 34544896,
      "step": 163690
    },
    {
      "epoch": 18.00825082508251,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0008973355794582426,
      "loss": 0.2308,
      "num_input_tokens_seen": 34546016,
      "step": 163695
    },
    {
      "epoch": 18.00880088008801,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0008968450447137072,
      "loss": 0.2298,
      "num_input_tokens_seen": 34547040,
      "step": 163700
    },
    {
      "epoch": 18.009350935093508,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0008963546399511801,
      "loss": 0.2319,
      "num_input_tokens_seen": 34548064,
      "step": 163705
    },
    {
      "epoch": 18.00990099009901,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0008958643651751846,
      "loss": 0.2314,
      "num_input_tokens_seen": 34549152,
      "step": 163710
    },
    {
      "epoch": 18.01045104510451,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.0008953742203902421,
      "loss": 0.2309,
      "num_input_tokens_seen": 34550240,
      "step": 163715
    },
    {
      "epoch": 18.01100110011001,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0008948842056008654,
      "loss": 0.2277,
      "num_input_tokens_seen": 34551264,
      "step": 163720
    },
    {
      "epoch": 18.011551155115512,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0008943943208115745,
      "loss": 0.2314,
      "num_input_tokens_seen": 34552352,
      "step": 163725
    },
    {
      "epoch": 18.012101210121013,
      "grad_norm": 0.0023193359375,
      "learning_rate": 0.0008939045660268807,
      "loss": 0.2314,
      "num_input_tokens_seen": 34553440,
      "step": 163730
    },
    {
      "epoch": 18.01265126512651,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0008934149412513004,
      "loss": 0.2319,
      "num_input_tokens_seen": 34554528,
      "step": 163735
    },
    {
      "epoch": 18.013201320132012,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0008929254464893465,
      "loss": 0.2319,
      "num_input_tokens_seen": 34555616,
      "step": 163740
    },
    {
      "epoch": 18.013751375137513,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.0008924360817455274,
      "loss": 0.2314,
      "num_input_tokens_seen": 34556672,
      "step": 163745
    },
    {
      "epoch": 18.014301430143014,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0008919468470243575,
      "loss": 0.2303,
      "num_input_tokens_seen": 34557696,
      "step": 163750
    },
    {
      "epoch": 18.014851485148515,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.0008914577423303449,
      "loss": 0.2319,
      "num_input_tokens_seen": 34558752,
      "step": 163755
    },
    {
      "epoch": 18.015401540154016,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0008909687676679928,
      "loss": 0.2303,
      "num_input_tokens_seen": 34559808,
      "step": 163760
    },
    {
      "epoch": 18.015951595159517,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.0008904799230418125,
      "loss": 0.2303,
      "num_input_tokens_seen": 34560768,
      "step": 163765
    },
    {
      "epoch": 18.016501650165015,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0008899912084563088,
      "loss": 0.2293,
      "num_input_tokens_seen": 34561824,
      "step": 163770
    },
    {
      "epoch": 18.017051705170516,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.0008895026239159864,
      "loss": 0.2303,
      "num_input_tokens_seen": 34562976,
      "step": 163775
    },
    {
      "epoch": 18.017601760176017,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0008890141694253484,
      "loss": 0.2335,
      "num_input_tokens_seen": 34564064,
      "step": 163780
    },
    {
      "epoch": 18.01815181518152,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0008885258449888927,
      "loss": 0.2314,
      "num_input_tokens_seen": 34565056,
      "step": 163785
    },
    {
      "epoch": 18.01870187018702,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0008880376506111225,
      "loss": 0.234,
      "num_input_tokens_seen": 34566144,
      "step": 163790
    },
    {
      "epoch": 18.01925192519252,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.0008875495862965393,
      "loss": 0.2335,
      "num_input_tokens_seen": 34567136,
      "step": 163795
    },
    {
      "epoch": 18.019801980198018,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.000887061652049641,
      "loss": 0.2319,
      "num_input_tokens_seen": 34568160,
      "step": 163800
    },
    {
      "epoch": 18.02035203520352,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.0008865738478749241,
      "loss": 0.2314,
      "num_input_tokens_seen": 34569184,
      "step": 163805
    },
    {
      "epoch": 18.02090209020902,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0008860861737768799,
      "loss": 0.2314,
      "num_input_tokens_seen": 34570240,
      "step": 163810
    },
    {
      "epoch": 18.02145214521452,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0008855986297600115,
      "loss": 0.2319,
      "num_input_tokens_seen": 34571360,
      "step": 163815
    },
    {
      "epoch": 18.022002200220022,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0008851112158288038,
      "loss": 0.2319,
      "num_input_tokens_seen": 34572448,
      "step": 163820
    },
    {
      "epoch": 18.022552255225524,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.000884623931987753,
      "loss": 0.2319,
      "num_input_tokens_seen": 34573440,
      "step": 163825
    },
    {
      "epoch": 18.023102310231025,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.000884136778241354,
      "loss": 0.2308,
      "num_input_tokens_seen": 34574496,
      "step": 163830
    },
    {
      "epoch": 18.023652365236522,
      "grad_norm": 0.0030517578125,
      "learning_rate": 0.0008836497545940896,
      "loss": 0.2329,
      "num_input_tokens_seen": 34575520,
      "step": 163835
    },
    {
      "epoch": 18.024202420242023,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.0008831628610504532,
      "loss": 0.2303,
      "num_input_tokens_seen": 34576576,
      "step": 163840
    },
    {
      "epoch": 18.024752475247524,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0008826760976149295,
      "loss": 0.2303,
      "num_input_tokens_seen": 34577632,
      "step": 163845
    },
    {
      "epoch": 18.025302530253025,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0008821894642920063,
      "loss": 0.2345,
      "num_input_tokens_seen": 34578656,
      "step": 163850
    },
    {
      "epoch": 18.025852585258527,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0008817029610861703,
      "loss": 0.233,
      "num_input_tokens_seen": 34579648,
      "step": 163855
    },
    {
      "epoch": 18.026402640264028,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.0008812165880019013,
      "loss": 0.2319,
      "num_input_tokens_seen": 34580672,
      "step": 163860
    },
    {
      "epoch": 18.02695269526953,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0008807303450436854,
      "loss": 0.2308,
      "num_input_tokens_seen": 34581760,
      "step": 163865
    },
    {
      "epoch": 18.027502750275026,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0008802442322160026,
      "loss": 0.2319,
      "num_input_tokens_seen": 34582752,
      "step": 163870
    },
    {
      "epoch": 18.028052805280527,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0008797582495233341,
      "loss": 0.2319,
      "num_input_tokens_seen": 34583872,
      "step": 163875
    },
    {
      "epoch": 18.02860286028603,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0008792723969701566,
      "loss": 0.2319,
      "num_input_tokens_seen": 34584832,
      "step": 163880
    },
    {
      "epoch": 18.02915291529153,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0008787866745609496,
      "loss": 0.2319,
      "num_input_tokens_seen": 34585888,
      "step": 163885
    },
    {
      "epoch": 18.02970297029703,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0008783010823001912,
      "loss": 0.2308,
      "num_input_tokens_seen": 34586944,
      "step": 163890
    },
    {
      "epoch": 18.03025302530253,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0008778156201923565,
      "loss": 0.2329,
      "num_input_tokens_seen": 34587936,
      "step": 163895
    },
    {
      "epoch": 18.03080308030803,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.0008773302882419165,
      "loss": 0.2314,
      "num_input_tokens_seen": 34589024,
      "step": 163900
    },
    {
      "epoch": 18.03135313531353,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0008768450864533477,
      "loss": 0.2309,
      "num_input_tokens_seen": 34590112,
      "step": 163905
    },
    {
      "epoch": 18.03190319031903,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0008763600148311168,
      "loss": 0.2329,
      "num_input_tokens_seen": 34591104,
      "step": 163910
    },
    {
      "epoch": 18.032453245324533,
      "grad_norm": 0.00060272216796875,
      "learning_rate": 0.0008758750733797033,
      "loss": 0.2288,
      "num_input_tokens_seen": 34592160,
      "step": 163915
    },
    {
      "epoch": 18.033003300330034,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 0.0008753902621035719,
      "loss": 0.2324,
      "num_input_tokens_seen": 34593216,
      "step": 163920
    },
    {
      "epoch": 18.033553355335535,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0008749055810071892,
      "loss": 0.2314,
      "num_input_tokens_seen": 34594304,
      "step": 163925
    },
    {
      "epoch": 18.034103410341036,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.000874421030095025,
      "loss": 0.2298,
      "num_input_tokens_seen": 34595360,
      "step": 163930
    },
    {
      "epoch": 18.034653465346533,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.0008739366093715422,
      "loss": 0.2298,
      "num_input_tokens_seen": 34596384,
      "step": 163935
    },
    {
      "epoch": 18.035203520352034,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0008734523188412074,
      "loss": 0.2324,
      "num_input_tokens_seen": 34597472,
      "step": 163940
    },
    {
      "epoch": 18.035753575357536,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0008729681585084853,
      "loss": 0.2324,
      "num_input_tokens_seen": 34598528,
      "step": 163945
    },
    {
      "epoch": 18.036303630363037,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0008724841283778356,
      "loss": 0.2308,
      "num_input_tokens_seen": 34599584,
      "step": 163950
    },
    {
      "epoch": 18.036853685368538,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0008720002284537231,
      "loss": 0.233,
      "num_input_tokens_seen": 34600704,
      "step": 163955
    },
    {
      "epoch": 18.03740374037404,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.0008715164587406026,
      "loss": 0.2335,
      "num_input_tokens_seen": 34601792,
      "step": 163960
    },
    {
      "epoch": 18.037953795379536,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0008710328192429356,
      "loss": 0.2309,
      "num_input_tokens_seen": 34602848,
      "step": 163965
    },
    {
      "epoch": 18.038503850385037,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0008705493099651818,
      "loss": 0.232,
      "num_input_tokens_seen": 34603872,
      "step": 163970
    },
    {
      "epoch": 18.03905390539054,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0008700659309117925,
      "loss": 0.2303,
      "num_input_tokens_seen": 34604864,
      "step": 163975
    },
    {
      "epoch": 18.03960396039604,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0008695826820872259,
      "loss": 0.2314,
      "num_input_tokens_seen": 34605888,
      "step": 163980
    },
    {
      "epoch": 18.04015401540154,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0008690995634959353,
      "loss": 0.2309,
      "num_input_tokens_seen": 34606880,
      "step": 163985
    },
    {
      "epoch": 18.040704070407042,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0008686165751423752,
      "loss": 0.2314,
      "num_input_tokens_seen": 34607968,
      "step": 163990
    },
    {
      "epoch": 18.041254125412543,
      "grad_norm": 0.002655029296875,
      "learning_rate": 0.0008681337170309922,
      "loss": 0.2309,
      "num_input_tokens_seen": 34609024,
      "step": 163995
    },
    {
      "epoch": 18.04180418041804,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.000867650989166241,
      "loss": 0.2324,
      "num_input_tokens_seen": 34610080,
      "step": 164000
    },
    {
      "epoch": 18.04235423542354,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.0008671683915525696,
      "loss": 0.2309,
      "num_input_tokens_seen": 34611168,
      "step": 164005
    },
    {
      "epoch": 18.042904290429043,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0008666859241944246,
      "loss": 0.2324,
      "num_input_tokens_seen": 34612256,
      "step": 164010
    },
    {
      "epoch": 18.043454345434544,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0008662035870962559,
      "loss": 0.2309,
      "num_input_tokens_seen": 34613376,
      "step": 164015
    },
    {
      "epoch": 18.044004400440045,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0008657213802625063,
      "loss": 0.2308,
      "num_input_tokens_seen": 34614464,
      "step": 164020
    },
    {
      "epoch": 18.044554455445546,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0008652393036976157,
      "loss": 0.2298,
      "num_input_tokens_seen": 34615456,
      "step": 164025
    },
    {
      "epoch": 18.045104510451043,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.0008647573574060375,
      "loss": 0.2308,
      "num_input_tokens_seen": 34616448,
      "step": 164030
    },
    {
      "epoch": 18.045654565456545,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0008642755413922042,
      "loss": 0.2324,
      "num_input_tokens_seen": 34617472,
      "step": 164035
    },
    {
      "epoch": 18.046204620462046,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0008637938556605629,
      "loss": 0.2298,
      "num_input_tokens_seen": 34618624,
      "step": 164040
    },
    {
      "epoch": 18.046754675467547,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0008633123002155513,
      "loss": 0.2324,
      "num_input_tokens_seen": 34619680,
      "step": 164045
    },
    {
      "epoch": 18.047304730473048,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.0008628308750616042,
      "loss": 0.2309,
      "num_input_tokens_seen": 34620800,
      "step": 164050
    },
    {
      "epoch": 18.04785478547855,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.0008623495802031617,
      "loss": 0.2298,
      "num_input_tokens_seen": 34621856,
      "step": 164055
    },
    {
      "epoch": 18.04840484048405,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.0008618684156446581,
      "loss": 0.2308,
      "num_input_tokens_seen": 34622944,
      "step": 164060
    },
    {
      "epoch": 18.048954895489548,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.000861387381390532,
      "loss": 0.2314,
      "num_input_tokens_seen": 34624000,
      "step": 164065
    },
    {
      "epoch": 18.04950495049505,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.000860906477445213,
      "loss": 0.2298,
      "num_input_tokens_seen": 34624992,
      "step": 164070
    },
    {
      "epoch": 18.05005500550055,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.000860425703813134,
      "loss": 0.2314,
      "num_input_tokens_seen": 34626016,
      "step": 164075
    },
    {
      "epoch": 18.05060506050605,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0008599450604987269,
      "loss": 0.2314,
      "num_input_tokens_seen": 34627104,
      "step": 164080
    },
    {
      "epoch": 18.051155115511552,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.0008594645475064211,
      "loss": 0.2319,
      "num_input_tokens_seen": 34628160,
      "step": 164085
    },
    {
      "epoch": 18.051705170517053,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0008589841648406432,
      "loss": 0.2314,
      "num_input_tokens_seen": 34629248,
      "step": 164090
    },
    {
      "epoch": 18.05225522552255,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.0008585039125058264,
      "loss": 0.2324,
      "num_input_tokens_seen": 34630368,
      "step": 164095
    },
    {
      "epoch": 18.05280528052805,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0008580237905063903,
      "loss": 0.2293,
      "num_input_tokens_seen": 34631424,
      "step": 164100
    },
    {
      "epoch": 18.053355335533553,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.0008575437988467649,
      "loss": 0.2308,
      "num_input_tokens_seen": 34632512,
      "step": 164105
    },
    {
      "epoch": 18.053905390539054,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0008570639375313698,
      "loss": 0.2319,
      "num_input_tokens_seen": 34633504,
      "step": 164110
    },
    {
      "epoch": 18.054455445544555,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0008565842065646301,
      "loss": 0.234,
      "num_input_tokens_seen": 34634560,
      "step": 164115
    },
    {
      "epoch": 18.055005500550056,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0008561046059509703,
      "loss": 0.2293,
      "num_input_tokens_seen": 34635584,
      "step": 164120
    },
    {
      "epoch": 18.055555555555557,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0008556251356948035,
      "loss": 0.2324,
      "num_input_tokens_seen": 34636640,
      "step": 164125
    },
    {
      "epoch": 18.056105610561055,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0008551457958005565,
      "loss": 0.2319,
      "num_input_tokens_seen": 34637696,
      "step": 164130
    },
    {
      "epoch": 18.056655665566556,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0008546665862726437,
      "loss": 0.2304,
      "num_input_tokens_seen": 34638784,
      "step": 164135
    },
    {
      "epoch": 18.057205720572057,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.000854187507115477,
      "loss": 0.2303,
      "num_input_tokens_seen": 34639840,
      "step": 164140
    },
    {
      "epoch": 18.057755775577558,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0008537085583334774,
      "loss": 0.2293,
      "num_input_tokens_seen": 34640832,
      "step": 164145
    },
    {
      "epoch": 18.05830583058306,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0008532297399310584,
      "loss": 0.2324,
      "num_input_tokens_seen": 34641920,
      "step": 164150
    },
    {
      "epoch": 18.05885588558856,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.0008527510519126347,
      "loss": 0.2298,
      "num_input_tokens_seen": 34642944,
      "step": 164155
    },
    {
      "epoch": 18.059405940594058,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.000852272494282616,
      "loss": 0.2324,
      "num_input_tokens_seen": 34643936,
      "step": 164160
    },
    {
      "epoch": 18.05995599559956,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.0008517940670454105,
      "loss": 0.2303,
      "num_input_tokens_seen": 34644992,
      "step": 164165
    },
    {
      "epoch": 18.06050605060506,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.0008513157702054297,
      "loss": 0.2329,
      "num_input_tokens_seen": 34646048,
      "step": 164170
    },
    {
      "epoch": 18.06105610561056,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0008508376037670833,
      "loss": 0.2324,
      "num_input_tokens_seen": 34647072,
      "step": 164175
    },
    {
      "epoch": 18.061606160616062,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.000850359567734778,
      "loss": 0.2298,
      "num_input_tokens_seen": 34648096,
      "step": 164180
    },
    {
      "epoch": 18.062156215621563,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0008498816621129201,
      "loss": 0.2324,
      "num_input_tokens_seen": 34649056,
      "step": 164185
    },
    {
      "epoch": 18.062706270627064,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0008494038869059111,
      "loss": 0.2309,
      "num_input_tokens_seen": 34650112,
      "step": 164190
    },
    {
      "epoch": 18.063256325632562,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0008489262421181575,
      "loss": 0.2319,
      "num_input_tokens_seen": 34651200,
      "step": 164195
    },
    {
      "epoch": 18.063806380638063,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0008484487277540592,
      "loss": 0.2298,
      "num_input_tokens_seen": 34652192,
      "step": 164200
    },
    {
      "epoch": 18.064356435643564,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0008479713438180174,
      "loss": 0.2298,
      "num_input_tokens_seen": 34653216,
      "step": 164205
    },
    {
      "epoch": 18.064906490649065,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0008474940903144357,
      "loss": 0.2308,
      "num_input_tokens_seen": 34654304,
      "step": 164210
    },
    {
      "epoch": 18.065456545654566,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0008470169672477085,
      "loss": 0.2314,
      "num_input_tokens_seen": 34655360,
      "step": 164215
    },
    {
      "epoch": 18.066006600660067,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.0008465399746222357,
      "loss": 0.2324,
      "num_input_tokens_seen": 34656416,
      "step": 164220
    },
    {
      "epoch": 18.066556655665565,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0008460631124424106,
      "loss": 0.2303,
      "num_input_tokens_seen": 34657472,
      "step": 164225
    },
    {
      "epoch": 18.067106710671066,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.0008455863807126296,
      "loss": 0.2303,
      "num_input_tokens_seen": 34658528,
      "step": 164230
    },
    {
      "epoch": 18.067656765676567,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.0008451097794372892,
      "loss": 0.2303,
      "num_input_tokens_seen": 34659520,
      "step": 164235
    },
    {
      "epoch": 18.068206820682068,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.0008446333086207791,
      "loss": 0.2288,
      "num_input_tokens_seen": 34660512,
      "step": 164240
    },
    {
      "epoch": 18.06875687568757,
      "grad_norm": 0.00262451171875,
      "learning_rate": 0.0008441569682674926,
      "loss": 0.2298,
      "num_input_tokens_seen": 34661664,
      "step": 164245
    },
    {
      "epoch": 18.06930693069307,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.0008436807583818162,
      "loss": 0.2298,
      "num_input_tokens_seen": 34662688,
      "step": 164250
    },
    {
      "epoch": 18.06985698569857,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.000843204678968143,
      "loss": 0.2324,
      "num_input_tokens_seen": 34663744,
      "step": 164255
    },
    {
      "epoch": 18.07040704070407,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0008427287300308578,
      "loss": 0.2304,
      "num_input_tokens_seen": 34664800,
      "step": 164260
    },
    {
      "epoch": 18.07095709570957,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0008422529115743487,
      "loss": 0.2309,
      "num_input_tokens_seen": 34665856,
      "step": 164265
    },
    {
      "epoch": 18.07150715071507,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0008417772236030024,
      "loss": 0.2288,
      "num_input_tokens_seen": 34666944,
      "step": 164270
    },
    {
      "epoch": 18.072057205720572,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 0.0008413016661212036,
      "loss": 0.234,
      "num_input_tokens_seen": 34668032,
      "step": 164275
    },
    {
      "epoch": 18.072607260726073,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0008408262391333287,
      "loss": 0.2308,
      "num_input_tokens_seen": 34669088,
      "step": 164280
    },
    {
      "epoch": 18.073157315731574,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.000840350942643766,
      "loss": 0.2298,
      "num_input_tokens_seen": 34670176,
      "step": 164285
    },
    {
      "epoch": 18.073707370737075,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.0008398757766568937,
      "loss": 0.2319,
      "num_input_tokens_seen": 34671200,
      "step": 164290
    },
    {
      "epoch": 18.074257425742573,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0008394007411770931,
      "loss": 0.2314,
      "num_input_tokens_seen": 34672288,
      "step": 164295
    },
    {
      "epoch": 18.074807480748074,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0008389258362087409,
      "loss": 0.2309,
      "num_input_tokens_seen": 34673376,
      "step": 164300
    },
    {
      "epoch": 18.075357535753575,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.0008384510617562135,
      "loss": 0.2303,
      "num_input_tokens_seen": 34674400,
      "step": 164305
    },
    {
      "epoch": 18.075907590759076,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.0008379764178238874,
      "loss": 0.2319,
      "num_input_tokens_seen": 34675392,
      "step": 164310
    },
    {
      "epoch": 18.076457645764577,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0008375019044161358,
      "loss": 0.2324,
      "num_input_tokens_seen": 34676512,
      "step": 164315
    },
    {
      "epoch": 18.07700770077008,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0008370275215373335,
      "loss": 0.2298,
      "num_input_tokens_seen": 34677568,
      "step": 164320
    },
    {
      "epoch": 18.077557755775576,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0008365532691918553,
      "loss": 0.2314,
      "num_input_tokens_seen": 34678592,
      "step": 164325
    },
    {
      "epoch": 18.078107810781077,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0008360791473840662,
      "loss": 0.2319,
      "num_input_tokens_seen": 34679680,
      "step": 164330
    },
    {
      "epoch": 18.078657865786578,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0008356051561183424,
      "loss": 0.2309,
      "num_input_tokens_seen": 34680672,
      "step": 164335
    },
    {
      "epoch": 18.07920792079208,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0008351312953990459,
      "loss": 0.2314,
      "num_input_tokens_seen": 34681760,
      "step": 164340
    },
    {
      "epoch": 18.07975797579758,
      "grad_norm": 0.000507354736328125,
      "learning_rate": 0.0008346575652305476,
      "loss": 0.2319,
      "num_input_tokens_seen": 34682752,
      "step": 164345
    },
    {
      "epoch": 18.08030803080308,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0008341839656172145,
      "loss": 0.2313,
      "num_input_tokens_seen": 34683840,
      "step": 164350
    },
    {
      "epoch": 18.080858085808583,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0008337104965634095,
      "loss": 0.2308,
      "num_input_tokens_seen": 34684928,
      "step": 164355
    },
    {
      "epoch": 18.08140814081408,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0008332371580734993,
      "loss": 0.2299,
      "num_input_tokens_seen": 34685984,
      "step": 164360
    },
    {
      "epoch": 18.08195819581958,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0008327639501518419,
      "loss": 0.2309,
      "num_input_tokens_seen": 34687040,
      "step": 164365
    },
    {
      "epoch": 18.082508250825082,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.0008322908728028022,
      "loss": 0.2319,
      "num_input_tokens_seen": 34688096,
      "step": 164370
    },
    {
      "epoch": 18.083058305830583,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.0008318179260307384,
      "loss": 0.2309,
      "num_input_tokens_seen": 34689088,
      "step": 164375
    },
    {
      "epoch": 18.083608360836084,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0008313451098400104,
      "loss": 0.2314,
      "num_input_tokens_seen": 34690144,
      "step": 164380
    },
    {
      "epoch": 18.084158415841586,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0008308724242349763,
      "loss": 0.2298,
      "num_input_tokens_seen": 34691136,
      "step": 164385
    },
    {
      "epoch": 18.084708470847083,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0008303998692199909,
      "loss": 0.2308,
      "num_input_tokens_seen": 34692192,
      "step": 164390
    },
    {
      "epoch": 18.085258525852584,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0008299274447994125,
      "loss": 0.2308,
      "num_input_tokens_seen": 34693216,
      "step": 164395
    },
    {
      "epoch": 18.085808580858085,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.0008294551509775927,
      "loss": 0.2308,
      "num_input_tokens_seen": 34694368,
      "step": 164400
    },
    {
      "epoch": 18.086358635863586,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.0008289829877588827,
      "loss": 0.2314,
      "num_input_tokens_seen": 34695392,
      "step": 164405
    },
    {
      "epoch": 18.086908690869087,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0008285109551476411,
      "loss": 0.2324,
      "num_input_tokens_seen": 34696544,
      "step": 164410
    },
    {
      "epoch": 18.08745874587459,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0008280390531482107,
      "loss": 0.2303,
      "num_input_tokens_seen": 34697568,
      "step": 164415
    },
    {
      "epoch": 18.08800880088009,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0008275672817649465,
      "loss": 0.2298,
      "num_input_tokens_seen": 34698624,
      "step": 164420
    },
    {
      "epoch": 18.088558855885587,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0008270956410021934,
      "loss": 0.234,
      "num_input_tokens_seen": 34699680,
      "step": 164425
    },
    {
      "epoch": 18.08910891089109,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0008266241308642963,
      "loss": 0.2309,
      "num_input_tokens_seen": 34700864,
      "step": 164430
    },
    {
      "epoch": 18.08965896589659,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.0008261527513556049,
      "loss": 0.2298,
      "num_input_tokens_seen": 34702016,
      "step": 164435
    },
    {
      "epoch": 18.09020902090209,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0008256815024804609,
      "loss": 0.2329,
      "num_input_tokens_seen": 34703104,
      "step": 164440
    },
    {
      "epoch": 18.09075907590759,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0008252103842432123,
      "loss": 0.2314,
      "num_input_tokens_seen": 34704096,
      "step": 164445
    },
    {
      "epoch": 18.091309130913093,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0008247393966481975,
      "loss": 0.2293,
      "num_input_tokens_seen": 34705152,
      "step": 164450
    },
    {
      "epoch": 18.09185918591859,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0008242685396997562,
      "loss": 0.2319,
      "num_input_tokens_seen": 34706176,
      "step": 164455
    },
    {
      "epoch": 18.09240924092409,
      "grad_norm": 0.000728607177734375,
      "learning_rate": 0.0008237978134022283,
      "loss": 0.2329,
      "num_input_tokens_seen": 34707232,
      "step": 164460
    },
    {
      "epoch": 18.092959295929592,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.0008233272177599571,
      "loss": 0.2303,
      "num_input_tokens_seen": 34708256,
      "step": 164465
    },
    {
      "epoch": 18.093509350935093,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.0008228567527772723,
      "loss": 0.2303,
      "num_input_tokens_seen": 34709312,
      "step": 164470
    },
    {
      "epoch": 18.094059405940595,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0008223864184585172,
      "loss": 0.2309,
      "num_input_tokens_seen": 34710464,
      "step": 164475
    },
    {
      "epoch": 18.094609460946096,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.0008219162148080216,
      "loss": 0.2324,
      "num_input_tokens_seen": 34711488,
      "step": 164480
    },
    {
      "epoch": 18.095159515951597,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0008214461418301222,
      "loss": 0.2314,
      "num_input_tokens_seen": 34712576,
      "step": 164485
    },
    {
      "epoch": 18.095709570957094,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0008209761995291487,
      "loss": 0.2324,
      "num_input_tokens_seen": 34713664,
      "step": 164490
    },
    {
      "epoch": 18.096259625962595,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.0008205063879094343,
      "loss": 0.2329,
      "num_input_tokens_seen": 34714720,
      "step": 164495
    },
    {
      "epoch": 18.096809680968097,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0008200367069753105,
      "loss": 0.2309,
      "num_input_tokens_seen": 34715776,
      "step": 164500
    },
    {
      "epoch": 18.097359735973598,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0008195671567311007,
      "loss": 0.2319,
      "num_input_tokens_seen": 34716896,
      "step": 164505
    },
    {
      "epoch": 18.0979097909791,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0008190977371811398,
      "loss": 0.2314,
      "num_input_tokens_seen": 34717984,
      "step": 164510
    },
    {
      "epoch": 18.0984598459846,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.0008186284483297473,
      "loss": 0.2288,
      "num_input_tokens_seen": 34719040,
      "step": 164515
    },
    {
      "epoch": 18.099009900990097,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0008181592901812517,
      "loss": 0.2335,
      "num_input_tokens_seen": 34720160,
      "step": 164520
    },
    {
      "epoch": 18.0995599559956,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.0008176902627399796,
      "loss": 0.2324,
      "num_input_tokens_seen": 34721312,
      "step": 164525
    },
    {
      "epoch": 18.1001100110011,
      "grad_norm": 0.000583648681640625,
      "learning_rate": 0.0008172213660102473,
      "loss": 0.2335,
      "num_input_tokens_seen": 34722336,
      "step": 164530
    },
    {
      "epoch": 18.1006600660066,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.0008167525999963831,
      "loss": 0.2308,
      "num_input_tokens_seen": 34723392,
      "step": 164535
    },
    {
      "epoch": 18.1012101210121,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0008162839647027053,
      "loss": 0.2314,
      "num_input_tokens_seen": 34724448,
      "step": 164540
    },
    {
      "epoch": 18.101760176017603,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0008158154601335287,
      "loss": 0.2329,
      "num_input_tokens_seen": 34725504,
      "step": 164545
    },
    {
      "epoch": 18.102310231023104,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.000815347086293175,
      "loss": 0.234,
      "num_input_tokens_seen": 34726528,
      "step": 164550
    },
    {
      "epoch": 18.1028602860286,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0008148788431859605,
      "loss": 0.2324,
      "num_input_tokens_seen": 34727584,
      "step": 164555
    },
    {
      "epoch": 18.103410341034103,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0008144107308162052,
      "loss": 0.2308,
      "num_input_tokens_seen": 34728640,
      "step": 164560
    },
    {
      "epoch": 18.103960396039604,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.0008139427491882172,
      "loss": 0.2308,
      "num_input_tokens_seen": 34729696,
      "step": 164565
    },
    {
      "epoch": 18.104510451045105,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00081347489830631,
      "loss": 0.2303,
      "num_input_tokens_seen": 34730816,
      "step": 164570
    },
    {
      "epoch": 18.105060506050606,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0008130071781747983,
      "loss": 0.2324,
      "num_input_tokens_seen": 34731808,
      "step": 164575
    },
    {
      "epoch": 18.105610561056107,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0008125395887979902,
      "loss": 0.2298,
      "num_input_tokens_seen": 34732896,
      "step": 164580
    },
    {
      "epoch": 18.106160616061604,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0008120721301802009,
      "loss": 0.2329,
      "num_input_tokens_seen": 34733952,
      "step": 164585
    },
    {
      "epoch": 18.106710671067106,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0008116048023257333,
      "loss": 0.2329,
      "num_input_tokens_seen": 34735008,
      "step": 164590
    },
    {
      "epoch": 18.107260726072607,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0008111376052388941,
      "loss": 0.2314,
      "num_input_tokens_seen": 34736064,
      "step": 164595
    },
    {
      "epoch": 18.107810781078108,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.0008106705389239932,
      "loss": 0.2324,
      "num_input_tokens_seen": 34737088,
      "step": 164600
    },
    {
      "epoch": 18.10836083608361,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.0008102036033853322,
      "loss": 0.2293,
      "num_input_tokens_seen": 34738144,
      "step": 164605
    },
    {
      "epoch": 18.10891089108911,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 0.0008097367986272141,
      "loss": 0.2283,
      "num_input_tokens_seen": 34739264,
      "step": 164610
    },
    {
      "epoch": 18.10946094609461,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0008092701246539458,
      "loss": 0.2308,
      "num_input_tokens_seen": 34740256,
      "step": 164615
    },
    {
      "epoch": 18.11001100110011,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0008088035814698235,
      "loss": 0.2303,
      "num_input_tokens_seen": 34741376,
      "step": 164620
    },
    {
      "epoch": 18.11056105610561,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0008083371690791508,
      "loss": 0.2314,
      "num_input_tokens_seen": 34742432,
      "step": 164625
    },
    {
      "epoch": 18.11111111111111,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0008078708874862222,
      "loss": 0.2329,
      "num_input_tokens_seen": 34743520,
      "step": 164630
    },
    {
      "epoch": 18.111661166116612,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.0008074047366953363,
      "loss": 0.2314,
      "num_input_tokens_seen": 34744672,
      "step": 164635
    },
    {
      "epoch": 18.112211221122113,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.0008069387167107944,
      "loss": 0.2283,
      "num_input_tokens_seen": 34745728,
      "step": 164640
    },
    {
      "epoch": 18.112761276127614,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0008064728275368865,
      "loss": 0.2298,
      "num_input_tokens_seen": 34746816,
      "step": 164645
    },
    {
      "epoch": 18.11331133113311,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0008060070691779091,
      "loss": 0.2329,
      "num_input_tokens_seen": 34747872,
      "step": 164650
    },
    {
      "epoch": 18.113861386138613,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.000805541441638154,
      "loss": 0.234,
      "num_input_tokens_seen": 34748960,
      "step": 164655
    },
    {
      "epoch": 18.114411441144114,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.0008050759449219091,
      "loss": 0.2298,
      "num_input_tokens_seen": 34750080,
      "step": 164660
    },
    {
      "epoch": 18.114961496149615,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0008046105790334679,
      "loss": 0.2314,
      "num_input_tokens_seen": 34751104,
      "step": 164665
    },
    {
      "epoch": 18.115511551155116,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.00080414534397712,
      "loss": 0.2309,
      "num_input_tokens_seen": 34752128,
      "step": 164670
    },
    {
      "epoch": 18.116061606160617,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0008036802397571541,
      "loss": 0.2309,
      "num_input_tokens_seen": 34753120,
      "step": 164675
    },
    {
      "epoch": 18.116611661166118,
      "grad_norm": 0.000667572021484375,
      "learning_rate": 0.0008032152663778563,
      "loss": 0.2319,
      "num_input_tokens_seen": 34754208,
      "step": 164680
    },
    {
      "epoch": 18.117161716171616,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0008027504238435084,
      "loss": 0.2293,
      "num_input_tokens_seen": 34755264,
      "step": 164685
    },
    {
      "epoch": 18.117711771177117,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0008022857121584003,
      "loss": 0.2324,
      "num_input_tokens_seen": 34756352,
      "step": 164690
    },
    {
      "epoch": 18.118261826182618,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.0008018211313268069,
      "loss": 0.2324,
      "num_input_tokens_seen": 34757504,
      "step": 164695
    },
    {
      "epoch": 18.11881188118812,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0008013566813530198,
      "loss": 0.2283,
      "num_input_tokens_seen": 34758496,
      "step": 164700
    },
    {
      "epoch": 18.11936193619362,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0008008923622413155,
      "loss": 0.2324,
      "num_input_tokens_seen": 34759520,
      "step": 164705
    },
    {
      "epoch": 18.11991199119912,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.000800428173995969,
      "loss": 0.2309,
      "num_input_tokens_seen": 34760640,
      "step": 164710
    },
    {
      "epoch": 18.120462046204622,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0007999641166212667,
      "loss": 0.2314,
      "num_input_tokens_seen": 34761760,
      "step": 164715
    },
    {
      "epoch": 18.12101210121012,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.0007995001901214788,
      "loss": 0.2324,
      "num_input_tokens_seen": 34762752,
      "step": 164720
    },
    {
      "epoch": 18.12156215621562,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.0007990363945008815,
      "loss": 0.2303,
      "num_input_tokens_seen": 34763840,
      "step": 164725
    },
    {
      "epoch": 18.122112211221122,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.0007985727297637552,
      "loss": 0.2298,
      "num_input_tokens_seen": 34764832,
      "step": 164730
    },
    {
      "epoch": 18.122662266226623,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.0007981091959143677,
      "loss": 0.2309,
      "num_input_tokens_seen": 34765824,
      "step": 164735
    },
    {
      "epoch": 18.123212321232124,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.000797645792956994,
      "loss": 0.2314,
      "num_input_tokens_seen": 34766880,
      "step": 164740
    },
    {
      "epoch": 18.123762376237625,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0007971825208959027,
      "loss": 0.2308,
      "num_input_tokens_seen": 34767968,
      "step": 164745
    },
    {
      "epoch": 18.124312431243123,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0007967193797353682,
      "loss": 0.2324,
      "num_input_tokens_seen": 34768992,
      "step": 164750
    },
    {
      "epoch": 18.124862486248624,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.0007962563694796526,
      "loss": 0.2314,
      "num_input_tokens_seen": 34770016,
      "step": 164755
    },
    {
      "epoch": 18.125412541254125,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.0007957934901330255,
      "loss": 0.2309,
      "num_input_tokens_seen": 34771008,
      "step": 164760
    },
    {
      "epoch": 18.125962596259626,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0007953307416997568,
      "loss": 0.2314,
      "num_input_tokens_seen": 34772096,
      "step": 164765
    },
    {
      "epoch": 18.126512651265127,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.0007948681241841066,
      "loss": 0.2298,
      "num_input_tokens_seen": 34773152,
      "step": 164770
    },
    {
      "epoch": 18.127062706270628,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.0007944056375903429,
      "loss": 0.2303,
      "num_input_tokens_seen": 34774272,
      "step": 164775
    },
    {
      "epoch": 18.12761276127613,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.0007939432819227243,
      "loss": 0.2309,
      "num_input_tokens_seen": 34775424,
      "step": 164780
    },
    {
      "epoch": 18.128162816281627,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.0007934810571855121,
      "loss": 0.2303,
      "num_input_tokens_seen": 34776448,
      "step": 164785
    },
    {
      "epoch": 18.128712871287128,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0007930189633829715,
      "loss": 0.2314,
      "num_input_tokens_seen": 34777536,
      "step": 164790
    },
    {
      "epoch": 18.12926292629263,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0007925570005193554,
      "loss": 0.2314,
      "num_input_tokens_seen": 34778656,
      "step": 164795
    },
    {
      "epoch": 18.12981298129813,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0007920951685989258,
      "loss": 0.2303,
      "num_input_tokens_seen": 34779776,
      "step": 164800
    },
    {
      "epoch": 18.13036303630363,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.000791633467625939,
      "loss": 0.2303,
      "num_input_tokens_seen": 34780832,
      "step": 164805
    },
    {
      "epoch": 18.130913091309132,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0007911718976046433,
      "loss": 0.2314,
      "num_input_tokens_seen": 34781856,
      "step": 164810
    },
    {
      "epoch": 18.13146314631463,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.000790710458539302,
      "loss": 0.2324,
      "num_input_tokens_seen": 34782848,
      "step": 164815
    },
    {
      "epoch": 18.13201320132013,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0007902491504341619,
      "loss": 0.2303,
      "num_input_tokens_seen": 34783808,
      "step": 164820
    },
    {
      "epoch": 18.132563256325632,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.0007897879732934793,
      "loss": 0.2303,
      "num_input_tokens_seen": 34784864,
      "step": 164825
    },
    {
      "epoch": 18.133113311331133,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.0007893269271215025,
      "loss": 0.2319,
      "num_input_tokens_seen": 34786016,
      "step": 164830
    },
    {
      "epoch": 18.133663366336634,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0007888660119224782,
      "loss": 0.2314,
      "num_input_tokens_seen": 34787104,
      "step": 164835
    },
    {
      "epoch": 18.134213421342135,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0007884052277006565,
      "loss": 0.2324,
      "num_input_tokens_seen": 34788160,
      "step": 164840
    },
    {
      "epoch": 18.134763476347636,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0007879445744602869,
      "loss": 0.2303,
      "num_input_tokens_seen": 34789216,
      "step": 164845
    },
    {
      "epoch": 18.135313531353134,
      "grad_norm": 0.0027618408203125,
      "learning_rate": 0.0007874840522056098,
      "loss": 0.2319,
      "num_input_tokens_seen": 34790336,
      "step": 164850
    },
    {
      "epoch": 18.135863586358635,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.0007870236609408764,
      "loss": 0.2309,
      "num_input_tokens_seen": 34791392,
      "step": 164855
    },
    {
      "epoch": 18.136413641364136,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.000786563400670322,
      "loss": 0.234,
      "num_input_tokens_seen": 34792448,
      "step": 164860
    },
    {
      "epoch": 18.136963696369637,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0007861032713981946,
      "loss": 0.2303,
      "num_input_tokens_seen": 34793504,
      "step": 164865
    },
    {
      "epoch": 18.13751375137514,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.000785643273128731,
      "loss": 0.2324,
      "num_input_tokens_seen": 34794464,
      "step": 164870
    },
    {
      "epoch": 18.13806380638064,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0007851834058661728,
      "loss": 0.2313,
      "num_input_tokens_seen": 34795520,
      "step": 164875
    },
    {
      "epoch": 18.138613861386137,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0007847236696147597,
      "loss": 0.2329,
      "num_input_tokens_seen": 34796704,
      "step": 164880
    },
    {
      "epoch": 18.139163916391638,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.0007842640643787269,
      "loss": 0.2314,
      "num_input_tokens_seen": 34797760,
      "step": 164885
    },
    {
      "epoch": 18.13971397139714,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0007838045901623109,
      "loss": 0.2293,
      "num_input_tokens_seen": 34798848,
      "step": 164890
    },
    {
      "epoch": 18.14026402640264,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.000783345246969745,
      "loss": 0.2303,
      "num_input_tokens_seen": 34799936,
      "step": 164895
    },
    {
      "epoch": 18.14081408140814,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0007828860348052658,
      "loss": 0.2298,
      "num_input_tokens_seen": 34801024,
      "step": 164900
    },
    {
      "epoch": 18.141364136413642,
      "grad_norm": 0.00078582763671875,
      "learning_rate": 0.0007824269536731032,
      "loss": 0.2293,
      "num_input_tokens_seen": 34802016,
      "step": 164905
    },
    {
      "epoch": 18.141914191419144,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.000781968003577489,
      "loss": 0.2313,
      "num_input_tokens_seen": 34803008,
      "step": 164910
    },
    {
      "epoch": 18.14246424642464,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0007815091845226545,
      "loss": 0.2319,
      "num_input_tokens_seen": 34804032,
      "step": 164915
    },
    {
      "epoch": 18.143014301430142,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0007810504965128284,
      "loss": 0.2324,
      "num_input_tokens_seen": 34805120,
      "step": 164920
    },
    {
      "epoch": 18.143564356435643,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0007805919395522337,
      "loss": 0.2314,
      "num_input_tokens_seen": 34806240,
      "step": 164925
    },
    {
      "epoch": 18.144114411441144,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.0007801335136451004,
      "loss": 0.2324,
      "num_input_tokens_seen": 34807232,
      "step": 164930
    },
    {
      "epoch": 18.144664466446645,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0007796752187956534,
      "loss": 0.2329,
      "num_input_tokens_seen": 34808256,
      "step": 164935
    },
    {
      "epoch": 18.145214521452147,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0007792170550081179,
      "loss": 0.2314,
      "num_input_tokens_seen": 34809248,
      "step": 164940
    },
    {
      "epoch": 18.145764576457644,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.0007787590222867152,
      "loss": 0.2309,
      "num_input_tokens_seen": 34810304,
      "step": 164945
    },
    {
      "epoch": 18.146314631463145,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.0007783011206356638,
      "loss": 0.2319,
      "num_input_tokens_seen": 34811360,
      "step": 164950
    },
    {
      "epoch": 18.146864686468646,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.0007778433500591869,
      "loss": 0.2335,
      "num_input_tokens_seen": 34812416,
      "step": 164955
    },
    {
      "epoch": 18.147414741474147,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0007773857105615045,
      "loss": 0.2329,
      "num_input_tokens_seen": 34813408,
      "step": 164960
    },
    {
      "epoch": 18.14796479647965,
      "grad_norm": 0.000606536865234375,
      "learning_rate": 0.0007769282021468348,
      "loss": 0.2314,
      "num_input_tokens_seen": 34814464,
      "step": 164965
    },
    {
      "epoch": 18.14851485148515,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0007764708248193929,
      "loss": 0.2309,
      "num_input_tokens_seen": 34815456,
      "step": 164970
    },
    {
      "epoch": 18.14906490649065,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0007760135785833921,
      "loss": 0.2345,
      "num_input_tokens_seen": 34816544,
      "step": 164975
    },
    {
      "epoch": 18.149614961496148,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0007755564634430506,
      "loss": 0.2293,
      "num_input_tokens_seen": 34817632,
      "step": 164980
    },
    {
      "epoch": 18.15016501650165,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.0007750994794025784,
      "loss": 0.2303,
      "num_input_tokens_seen": 34818720,
      "step": 164985
    },
    {
      "epoch": 18.15071507150715,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0007746426264661871,
      "loss": 0.2319,
      "num_input_tokens_seen": 34819744,
      "step": 164990
    },
    {
      "epoch": 18.15126512651265,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0007741859046380917,
      "loss": 0.2329,
      "num_input_tokens_seen": 34820800,
      "step": 164995
    },
    {
      "epoch": 18.151815181518153,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0007737293139224971,
      "loss": 0.2298,
      "num_input_tokens_seen": 34821888,
      "step": 165000
    },
    {
      "epoch": 18.152365236523654,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.000773272854323615,
      "loss": 0.2324,
      "num_input_tokens_seen": 34822944,
      "step": 165005
    },
    {
      "epoch": 18.15291529152915,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.000772816525845647,
      "loss": 0.2319,
      "num_input_tokens_seen": 34823968,
      "step": 165010
    },
    {
      "epoch": 18.153465346534652,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0007723603284928032,
      "loss": 0.233,
      "num_input_tokens_seen": 34825024,
      "step": 165015
    },
    {
      "epoch": 18.154015401540153,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0007719042622692901,
      "loss": 0.2293,
      "num_input_tokens_seen": 34826080,
      "step": 165020
    },
    {
      "epoch": 18.154565456545654,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0007714483271793043,
      "loss": 0.2319,
      "num_input_tokens_seen": 34827136,
      "step": 165025
    },
    {
      "epoch": 18.155115511551156,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0007709925232270542,
      "loss": 0.2329,
      "num_input_tokens_seen": 34828288,
      "step": 165030
    },
    {
      "epoch": 18.155665566556657,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0007705368504167398,
      "loss": 0.2324,
      "num_input_tokens_seen": 34829376,
      "step": 165035
    },
    {
      "epoch": 18.156215621562158,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0007700813087525559,
      "loss": 0.2319,
      "num_input_tokens_seen": 34830464,
      "step": 165040
    },
    {
      "epoch": 18.156765676567655,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0007696258982387044,
      "loss": 0.2335,
      "num_input_tokens_seen": 34831552,
      "step": 165045
    },
    {
      "epoch": 18.157315731573156,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0007691706188793818,
      "loss": 0.2309,
      "num_input_tokens_seen": 34832608,
      "step": 165050
    },
    {
      "epoch": 18.157865786578657,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0007687154706787879,
      "loss": 0.2314,
      "num_input_tokens_seen": 34833696,
      "step": 165055
    },
    {
      "epoch": 18.15841584158416,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.0007682604536411147,
      "loss": 0.2314,
      "num_input_tokens_seen": 34834752,
      "step": 165060
    },
    {
      "epoch": 18.15896589658966,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.0007678055677705536,
      "loss": 0.2319,
      "num_input_tokens_seen": 34835840,
      "step": 165065
    },
    {
      "epoch": 18.15951595159516,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0007673508130712997,
      "loss": 0.2309,
      "num_input_tokens_seen": 34836928,
      "step": 165070
    },
    {
      "epoch": 18.16006600660066,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0007668961895475445,
      "loss": 0.2324,
      "num_input_tokens_seen": 34838016,
      "step": 165075
    },
    {
      "epoch": 18.16061606160616,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0007664416972034782,
      "loss": 0.2293,
      "num_input_tokens_seen": 34839040,
      "step": 165080
    },
    {
      "epoch": 18.16116611661166,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0007659873360432889,
      "loss": 0.2314,
      "num_input_tokens_seen": 34840064,
      "step": 165085
    },
    {
      "epoch": 18.16171617161716,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.0007655331060711633,
      "loss": 0.2324,
      "num_input_tokens_seen": 34841152,
      "step": 165090
    },
    {
      "epoch": 18.162266226622663,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.00076507900729129,
      "loss": 0.2319,
      "num_input_tokens_seen": 34842208,
      "step": 165095
    },
    {
      "epoch": 18.162816281628164,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0007646250397078502,
      "loss": 0.2314,
      "num_input_tokens_seen": 34843232,
      "step": 165100
    },
    {
      "epoch": 18.163366336633665,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.0007641712033250325,
      "loss": 0.2314,
      "num_input_tokens_seen": 34844256,
      "step": 165105
    },
    {
      "epoch": 18.163916391639162,
      "grad_norm": 0.0031585693359375,
      "learning_rate": 0.0007637174981470184,
      "loss": 0.2324,
      "num_input_tokens_seen": 34845376,
      "step": 165110
    },
    {
      "epoch": 18.164466446644663,
      "grad_norm": 0.00051116943359375,
      "learning_rate": 0.0007632639241779864,
      "loss": 0.2303,
      "num_input_tokens_seen": 34846368,
      "step": 165115
    },
    {
      "epoch": 18.165016501650165,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0007628104814221231,
      "loss": 0.2319,
      "num_input_tokens_seen": 34847392,
      "step": 165120
    },
    {
      "epoch": 18.165566556655666,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0007623571698836001,
      "loss": 0.2309,
      "num_input_tokens_seen": 34848512,
      "step": 165125
    },
    {
      "epoch": 18.166116611661167,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.0007619039895666007,
      "loss": 0.2319,
      "num_input_tokens_seen": 34849600,
      "step": 165130
    },
    {
      "epoch": 18.166666666666668,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.0007614509404753017,
      "loss": 0.2298,
      "num_input_tokens_seen": 34850656,
      "step": 165135
    },
    {
      "epoch": 18.16721672167217,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0007609980226138746,
      "loss": 0.2329,
      "num_input_tokens_seen": 34851744,
      "step": 165140
    },
    {
      "epoch": 18.167766776677666,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0007605452359864978,
      "loss": 0.2319,
      "num_input_tokens_seen": 34852800,
      "step": 165145
    },
    {
      "epoch": 18.168316831683168,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0007600925805973412,
      "loss": 0.2303,
      "num_input_tokens_seen": 34853856,
      "step": 165150
    },
    {
      "epoch": 18.16886688668867,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00075964005645058,
      "loss": 0.2298,
      "num_input_tokens_seen": 34854912,
      "step": 165155
    },
    {
      "epoch": 18.16941694169417,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0007591876635503824,
      "loss": 0.2308,
      "num_input_tokens_seen": 34855968,
      "step": 165160
    },
    {
      "epoch": 18.16996699669967,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0007587354019009184,
      "loss": 0.2324,
      "num_input_tokens_seen": 34856928,
      "step": 165165
    },
    {
      "epoch": 18.170517051705172,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.0007582832715063581,
      "loss": 0.2303,
      "num_input_tokens_seen": 34857984,
      "step": 165170
    },
    {
      "epoch": 18.17106710671067,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0007578312723708646,
      "loss": 0.2293,
      "num_input_tokens_seen": 34859008,
      "step": 165175
    },
    {
      "epoch": 18.17161716171617,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.0007573794044986082,
      "loss": 0.2324,
      "num_input_tokens_seen": 34860096,
      "step": 165180
    },
    {
      "epoch": 18.17216721672167,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0007569276678937503,
      "loss": 0.234,
      "num_input_tokens_seen": 34861152,
      "step": 165185
    },
    {
      "epoch": 18.172717271727173,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.000756476062560456,
      "loss": 0.2308,
      "num_input_tokens_seen": 34862208,
      "step": 165190
    },
    {
      "epoch": 18.173267326732674,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0007560245885028888,
      "loss": 0.2309,
      "num_input_tokens_seen": 34863232,
      "step": 165195
    },
    {
      "epoch": 18.173817381738175,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0007555732457252051,
      "loss": 0.2319,
      "num_input_tokens_seen": 34864256,
      "step": 165200
    },
    {
      "epoch": 18.174367436743676,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.00075512203423157,
      "loss": 0.2314,
      "num_input_tokens_seen": 34865312,
      "step": 165205
    },
    {
      "epoch": 18.174917491749174,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0007546709540261404,
      "loss": 0.2314,
      "num_input_tokens_seen": 34866400,
      "step": 165210
    },
    {
      "epoch": 18.175467546754675,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.000754220005113071,
      "loss": 0.2314,
      "num_input_tokens_seen": 34867456,
      "step": 165215
    },
    {
      "epoch": 18.176017601760176,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0007537691874965186,
      "loss": 0.2319,
      "num_input_tokens_seen": 34868480,
      "step": 165220
    },
    {
      "epoch": 18.176567656765677,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0007533185011806415,
      "loss": 0.2304,
      "num_input_tokens_seen": 34869568,
      "step": 165225
    },
    {
      "epoch": 18.177117711771178,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.000752867946169593,
      "loss": 0.2293,
      "num_input_tokens_seen": 34870624,
      "step": 165230
    },
    {
      "epoch": 18.17766776677668,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0007524175224675234,
      "loss": 0.2314,
      "num_input_tokens_seen": 34871680,
      "step": 165235
    },
    {
      "epoch": 18.178217821782177,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.000751967230078584,
      "loss": 0.2319,
      "num_input_tokens_seen": 34872736,
      "step": 165240
    },
    {
      "epoch": 18.178767876787678,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.000751517069006925,
      "loss": 0.2304,
      "num_input_tokens_seen": 34873856,
      "step": 165245
    },
    {
      "epoch": 18.17931793179318,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.0007510670392566982,
      "loss": 0.2319,
      "num_input_tokens_seen": 34874880,
      "step": 165250
    },
    {
      "epoch": 18.17986798679868,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0007506171408320466,
      "loss": 0.2319,
      "num_input_tokens_seen": 34875936,
      "step": 165255
    },
    {
      "epoch": 18.18041804180418,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0007501673737371222,
      "loss": 0.2319,
      "num_input_tokens_seen": 34876928,
      "step": 165260
    },
    {
      "epoch": 18.180968096809682,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.000749717737976065,
      "loss": 0.2319,
      "num_input_tokens_seen": 34877984,
      "step": 165265
    },
    {
      "epoch": 18.181518151815183,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0007492682335530231,
      "loss": 0.2308,
      "num_input_tokens_seen": 34879008,
      "step": 165270
    },
    {
      "epoch": 18.18206820682068,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0007488188604721368,
      "loss": 0.2309,
      "num_input_tokens_seen": 34880000,
      "step": 165275
    },
    {
      "epoch": 18.182618261826182,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.0007483696187375477,
      "loss": 0.2314,
      "num_input_tokens_seen": 34881088,
      "step": 165280
    },
    {
      "epoch": 18.183168316831683,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0007479205083533991,
      "loss": 0.2314,
      "num_input_tokens_seen": 34882176,
      "step": 165285
    },
    {
      "epoch": 18.183718371837184,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0007474715293238276,
      "loss": 0.2314,
      "num_input_tokens_seen": 34883200,
      "step": 165290
    },
    {
      "epoch": 18.184268426842685,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0007470226816529734,
      "loss": 0.2313,
      "num_input_tokens_seen": 34884256,
      "step": 165295
    },
    {
      "epoch": 18.184818481848186,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0007465739653449699,
      "loss": 0.2314,
      "num_input_tokens_seen": 34885312,
      "step": 165300
    },
    {
      "epoch": 18.185368536853684,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0007461253804039536,
      "loss": 0.2303,
      "num_input_tokens_seen": 34886336,
      "step": 165305
    },
    {
      "epoch": 18.185918591859185,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.000745676926834063,
      "loss": 0.2324,
      "num_input_tokens_seen": 34887328,
      "step": 165310
    },
    {
      "epoch": 18.186468646864686,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.0007452286046394263,
      "loss": 0.2308,
      "num_input_tokens_seen": 34888384,
      "step": 165315
    },
    {
      "epoch": 18.187018701870187,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0007447804138241803,
      "loss": 0.2319,
      "num_input_tokens_seen": 34889568,
      "step": 165320
    },
    {
      "epoch": 18.187568756875688,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0007443323543924518,
      "loss": 0.2308,
      "num_input_tokens_seen": 34890560,
      "step": 165325
    },
    {
      "epoch": 18.18811881188119,
      "grad_norm": 0.000492095947265625,
      "learning_rate": 0.000743884426348369,
      "loss": 0.2298,
      "num_input_tokens_seen": 34891584,
      "step": 165330
    },
    {
      "epoch": 18.18866886688669,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0007434366296960637,
      "loss": 0.2309,
      "num_input_tokens_seen": 34892704,
      "step": 165335
    },
    {
      "epoch": 18.189218921892188,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0007429889644396608,
      "loss": 0.2314,
      "num_input_tokens_seen": 34893728,
      "step": 165340
    },
    {
      "epoch": 18.18976897689769,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0007425414305832906,
      "loss": 0.2324,
      "num_input_tokens_seen": 34894720,
      "step": 165345
    },
    {
      "epoch": 18.19031903190319,
      "grad_norm": 0.002777099609375,
      "learning_rate": 0.0007420940281310745,
      "loss": 0.2303,
      "num_input_tokens_seen": 34895808,
      "step": 165350
    },
    {
      "epoch": 18.19086908690869,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0007416467570871343,
      "loss": 0.2324,
      "num_input_tokens_seen": 34896800,
      "step": 165355
    },
    {
      "epoch": 18.191419141914192,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0007411996174555935,
      "loss": 0.234,
      "num_input_tokens_seen": 34897888,
      "step": 165360
    },
    {
      "epoch": 18.191969196919693,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0007407526092405736,
      "loss": 0.2324,
      "num_input_tokens_seen": 34898976,
      "step": 165365
    },
    {
      "epoch": 18.19251925192519,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0007403057324461981,
      "loss": 0.2324,
      "num_input_tokens_seen": 34899968,
      "step": 165370
    },
    {
      "epoch": 18.193069306930692,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0007398589870765803,
      "loss": 0.2293,
      "num_input_tokens_seen": 34900960,
      "step": 165375
    },
    {
      "epoch": 18.193619361936193,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0007394123731358387,
      "loss": 0.2308,
      "num_input_tokens_seen": 34901984,
      "step": 165380
    },
    {
      "epoch": 18.194169416941694,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0007389658906280933,
      "loss": 0.2303,
      "num_input_tokens_seen": 34903008,
      "step": 165385
    },
    {
      "epoch": 18.194719471947195,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0007385195395574523,
      "loss": 0.2298,
      "num_input_tokens_seen": 34904096,
      "step": 165390
    },
    {
      "epoch": 18.195269526952696,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0007380733199280343,
      "loss": 0.2309,
      "num_input_tokens_seen": 34905184,
      "step": 165395
    },
    {
      "epoch": 18.195819581958197,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0007376272317439525,
      "loss": 0.2308,
      "num_input_tokens_seen": 34906272,
      "step": 165400
    },
    {
      "epoch": 18.196369636963695,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0007371812750093154,
      "loss": 0.2314,
      "num_input_tokens_seen": 34907328,
      "step": 165405
    },
    {
      "epoch": 18.196919691969196,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0007367354497282347,
      "loss": 0.2319,
      "num_input_tokens_seen": 34908480,
      "step": 165410
    },
    {
      "epoch": 18.197469746974697,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.0007362897559048187,
      "loss": 0.2288,
      "num_input_tokens_seen": 34909600,
      "step": 165415
    },
    {
      "epoch": 18.198019801980198,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0007358441935431759,
      "loss": 0.2329,
      "num_input_tokens_seen": 34910688,
      "step": 165420
    },
    {
      "epoch": 18.1985698569857,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0007353987626474146,
      "loss": 0.2303,
      "num_input_tokens_seen": 34911712,
      "step": 165425
    },
    {
      "epoch": 18.1991199119912,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0007349534632216348,
      "loss": 0.2324,
      "num_input_tokens_seen": 34912768,
      "step": 165430
    },
    {
      "epoch": 18.199669966996698,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0007345082952699467,
      "loss": 0.2313,
      "num_input_tokens_seen": 34913856,
      "step": 165435
    },
    {
      "epoch": 18.2002200220022,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.0007340632587964518,
      "loss": 0.2319,
      "num_input_tokens_seen": 34914880,
      "step": 165440
    },
    {
      "epoch": 18.2007700770077,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.000733618353805247,
      "loss": 0.2308,
      "num_input_tokens_seen": 34915936,
      "step": 165445
    },
    {
      "epoch": 18.2013201320132,
      "grad_norm": 0.000873565673828125,
      "learning_rate": 0.0007331735803004358,
      "loss": 0.2324,
      "num_input_tokens_seen": 34917024,
      "step": 165450
    },
    {
      "epoch": 18.201870187018702,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.000732728938286118,
      "loss": 0.2319,
      "num_input_tokens_seen": 34918048,
      "step": 165455
    },
    {
      "epoch": 18.202420242024203,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0007322844277663936,
      "loss": 0.2324,
      "num_input_tokens_seen": 34919168,
      "step": 165460
    },
    {
      "epoch": 18.202970297029704,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.000731840048745358,
      "loss": 0.2298,
      "num_input_tokens_seen": 34920224,
      "step": 165465
    },
    {
      "epoch": 18.203520352035202,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0007313958012271027,
      "loss": 0.2319,
      "num_input_tokens_seen": 34921216,
      "step": 165470
    },
    {
      "epoch": 18.204070407040703,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.0007309516852157294,
      "loss": 0.2314,
      "num_input_tokens_seen": 34922272,
      "step": 165475
    },
    {
      "epoch": 18.204620462046204,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0007305077007153215,
      "loss": 0.2319,
      "num_input_tokens_seen": 34923296,
      "step": 165480
    },
    {
      "epoch": 18.205170517051705,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.0007300638477299825,
      "loss": 0.234,
      "num_input_tokens_seen": 34924288,
      "step": 165485
    },
    {
      "epoch": 18.205720572057206,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0007296201262637958,
      "loss": 0.2298,
      "num_input_tokens_seen": 34925408,
      "step": 165490
    },
    {
      "epoch": 18.206270627062707,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.000729176536320853,
      "loss": 0.2314,
      "num_input_tokens_seen": 34926528,
      "step": 165495
    },
    {
      "epoch": 18.206820682068205,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0007287330779052426,
      "loss": 0.2314,
      "num_input_tokens_seen": 34927584,
      "step": 165500
    },
    {
      "epoch": 18.207370737073706,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0007282897510210511,
      "loss": 0.2319,
      "num_input_tokens_seen": 34928640,
      "step": 165505
    },
    {
      "epoch": 18.207920792079207,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.000727846555672364,
      "loss": 0.2314,
      "num_input_tokens_seen": 34929760,
      "step": 165510
    },
    {
      "epoch": 18.20847084708471,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0007274034918632676,
      "loss": 0.2314,
      "num_input_tokens_seen": 34930816,
      "step": 165515
    },
    {
      "epoch": 18.20902090209021,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.000726960559597844,
      "loss": 0.2319,
      "num_input_tokens_seen": 34931904,
      "step": 165520
    },
    {
      "epoch": 18.20957095709571,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0007265177588801763,
      "loss": 0.2308,
      "num_input_tokens_seen": 34932960,
      "step": 165525
    },
    {
      "epoch": 18.21012101210121,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0007260750897143447,
      "loss": 0.2319,
      "num_input_tokens_seen": 34934048,
      "step": 165530
    },
    {
      "epoch": 18.21067106710671,
      "grad_norm": 0.000835418701171875,
      "learning_rate": 0.0007256325521044326,
      "loss": 0.2298,
      "num_input_tokens_seen": 34935104,
      "step": 165535
    },
    {
      "epoch": 18.21122112211221,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0007251901460545118,
      "loss": 0.233,
      "num_input_tokens_seen": 34936224,
      "step": 165540
    },
    {
      "epoch": 18.21177117711771,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0007247478715686655,
      "loss": 0.2298,
      "num_input_tokens_seen": 34937248,
      "step": 165545
    },
    {
      "epoch": 18.212321232123212,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0007243057286509691,
      "loss": 0.2303,
      "num_input_tokens_seen": 34938304,
      "step": 165550
    },
    {
      "epoch": 18.212871287128714,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.000723863717305494,
      "loss": 0.2319,
      "num_input_tokens_seen": 34939360,
      "step": 165555
    },
    {
      "epoch": 18.213421342134215,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0007234218375363205,
      "loss": 0.2329,
      "num_input_tokens_seen": 34940416,
      "step": 165560
    },
    {
      "epoch": 18.213971397139716,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0007229800893475136,
      "loss": 0.2303,
      "num_input_tokens_seen": 34941408,
      "step": 165565
    },
    {
      "epoch": 18.214521452145213,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.0007225384727431482,
      "loss": 0.2309,
      "num_input_tokens_seen": 34942432,
      "step": 165570
    },
    {
      "epoch": 18.215071507150714,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.0007220969877272982,
      "loss": 0.2319,
      "num_input_tokens_seen": 34943456,
      "step": 165575
    },
    {
      "epoch": 18.215621562156215,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.0007216556343040265,
      "loss": 0.2309,
      "num_input_tokens_seen": 34944576,
      "step": 165580
    },
    {
      "epoch": 18.216171617161717,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.000721214412477405,
      "loss": 0.2303,
      "num_input_tokens_seen": 34945632,
      "step": 165585
    },
    {
      "epoch": 18.216721672167218,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0007207733222515005,
      "loss": 0.2298,
      "num_input_tokens_seen": 34946624,
      "step": 165590
    },
    {
      "epoch": 18.21727172717272,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0007203323636303699,
      "loss": 0.2314,
      "num_input_tokens_seen": 34947680,
      "step": 165595
    },
    {
      "epoch": 18.217821782178216,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0007198915366180897,
      "loss": 0.2308,
      "num_input_tokens_seen": 34948672,
      "step": 165600
    },
    {
      "epoch": 18.218371837183717,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0007194508412187133,
      "loss": 0.2324,
      "num_input_tokens_seen": 34949728,
      "step": 165605
    },
    {
      "epoch": 18.21892189218922,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0007190102774363093,
      "loss": 0.2314,
      "num_input_tokens_seen": 34950752,
      "step": 165610
    },
    {
      "epoch": 18.21947194719472,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.0007185698452749345,
      "loss": 0.2314,
      "num_input_tokens_seen": 34951840,
      "step": 165615
    },
    {
      "epoch": 18.22002200220022,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0007181295447386471,
      "loss": 0.2324,
      "num_input_tokens_seen": 34952800,
      "step": 165620
    },
    {
      "epoch": 18.22057205720572,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.0007176893758315055,
      "loss": 0.2314,
      "num_input_tokens_seen": 34953888,
      "step": 165625
    },
    {
      "epoch": 18.221122112211223,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0007172493385575701,
      "loss": 0.2308,
      "num_input_tokens_seen": 34954976,
      "step": 165630
    },
    {
      "epoch": 18.22167216721672,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0007168094329208924,
      "loss": 0.2324,
      "num_input_tokens_seen": 34956000,
      "step": 165635
    },
    {
      "epoch": 18.22222222222222,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.0007163696589255292,
      "loss": 0.2303,
      "num_input_tokens_seen": 34957056,
      "step": 165640
    },
    {
      "epoch": 18.222772277227723,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0007159300165755322,
      "loss": 0.2329,
      "num_input_tokens_seen": 34958112,
      "step": 165645
    },
    {
      "epoch": 18.223322332233224,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.0007154905058749549,
      "loss": 0.2335,
      "num_input_tokens_seen": 34959232,
      "step": 165650
    },
    {
      "epoch": 18.223872387238725,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0007150511268278476,
      "loss": 0.2319,
      "num_input_tokens_seen": 34960320,
      "step": 165655
    },
    {
      "epoch": 18.224422442244226,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0007146118794382567,
      "loss": 0.2309,
      "num_input_tokens_seen": 34961344,
      "step": 165660
    },
    {
      "epoch": 18.224972497249723,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0007141727637102374,
      "loss": 0.2309,
      "num_input_tokens_seen": 34962336,
      "step": 165665
    },
    {
      "epoch": 18.225522552255224,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0007137337796478298,
      "loss": 0.2314,
      "num_input_tokens_seen": 34963328,
      "step": 165670
    },
    {
      "epoch": 18.226072607260726,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0007132949272550842,
      "loss": 0.2314,
      "num_input_tokens_seen": 34964352,
      "step": 165675
    },
    {
      "epoch": 18.226622662266227,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0007128562065360422,
      "loss": 0.2308,
      "num_input_tokens_seen": 34965440,
      "step": 165680
    },
    {
      "epoch": 18.227172717271728,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0007124176174947488,
      "loss": 0.2303,
      "num_input_tokens_seen": 34966496,
      "step": 165685
    },
    {
      "epoch": 18.22772277227723,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0007119791601352493,
      "loss": 0.2319,
      "num_input_tokens_seen": 34967584,
      "step": 165690
    },
    {
      "epoch": 18.22827282728273,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0007115408344615787,
      "loss": 0.2309,
      "num_input_tokens_seen": 34968672,
      "step": 165695
    },
    {
      "epoch": 18.228822882288227,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.000711102640477782,
      "loss": 0.2298,
      "num_input_tokens_seen": 34969728,
      "step": 165700
    },
    {
      "epoch": 18.22937293729373,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0007106645781878962,
      "loss": 0.2308,
      "num_input_tokens_seen": 34970752,
      "step": 165705
    },
    {
      "epoch": 18.22992299229923,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0007102266475959562,
      "loss": 0.2324,
      "num_input_tokens_seen": 34971840,
      "step": 165710
    },
    {
      "epoch": 18.23047304730473,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0007097888487060005,
      "loss": 0.2319,
      "num_input_tokens_seen": 34972896,
      "step": 165715
    },
    {
      "epoch": 18.231023102310232,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0007093511815220643,
      "loss": 0.2324,
      "num_input_tokens_seen": 34973984,
      "step": 165720
    },
    {
      "epoch": 18.231573157315733,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0007089136460481827,
      "loss": 0.2308,
      "num_input_tokens_seen": 34975008,
      "step": 165725
    },
    {
      "epoch": 18.23212321232123,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0007084762422883855,
      "loss": 0.2314,
      "num_input_tokens_seen": 34976064,
      "step": 165730
    },
    {
      "epoch": 18.23267326732673,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0007080389702467032,
      "loss": 0.2319,
      "num_input_tokens_seen": 34977152,
      "step": 165735
    },
    {
      "epoch": 18.233223322332233,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0007076018299271691,
      "loss": 0.2324,
      "num_input_tokens_seen": 34978144,
      "step": 165740
    },
    {
      "epoch": 18.233773377337734,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.0007071648213338099,
      "loss": 0.2298,
      "num_input_tokens_seen": 34979264,
      "step": 165745
    },
    {
      "epoch": 18.234323432343235,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0007067279444706559,
      "loss": 0.2319,
      "num_input_tokens_seen": 34980288,
      "step": 165750
    },
    {
      "epoch": 18.234873487348736,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.000706291199341732,
      "loss": 0.2314,
      "num_input_tokens_seen": 34981312,
      "step": 165755
    },
    {
      "epoch": 18.235423542354237,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0007058545859510618,
      "loss": 0.2308,
      "num_input_tokens_seen": 34982336,
      "step": 165760
    },
    {
      "epoch": 18.235973597359735,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0007054181043026719,
      "loss": 0.2314,
      "num_input_tokens_seen": 34983360,
      "step": 165765
    },
    {
      "epoch": 18.236523652365236,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.0007049817544005826,
      "loss": 0.2309,
      "num_input_tokens_seen": 34984448,
      "step": 165770
    },
    {
      "epoch": 18.237073707370737,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.0007045455362488173,
      "loss": 0.2314,
      "num_input_tokens_seen": 34985504,
      "step": 165775
    },
    {
      "epoch": 18.237623762376238,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.0007041094498513995,
      "loss": 0.2319,
      "num_input_tokens_seen": 34986560,
      "step": 165780
    },
    {
      "epoch": 18.23817381738174,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.0007036734952123408,
      "loss": 0.2314,
      "num_input_tokens_seen": 34987552,
      "step": 165785
    },
    {
      "epoch": 18.23872387238724,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0007032376723356664,
      "loss": 0.2314,
      "num_input_tokens_seen": 34988608,
      "step": 165790
    },
    {
      "epoch": 18.239273927392738,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 0.0007028019812253882,
      "loss": 0.2303,
      "num_input_tokens_seen": 34989632,
      "step": 165795
    },
    {
      "epoch": 18.23982398239824,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.000702366421885523,
      "loss": 0.2319,
      "num_input_tokens_seen": 34990656,
      "step": 165800
    },
    {
      "epoch": 18.24037403740374,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0007019309943200891,
      "loss": 0.2298,
      "num_input_tokens_seen": 34991680,
      "step": 165805
    },
    {
      "epoch": 18.24092409240924,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.0007014956985330934,
      "loss": 0.2314,
      "num_input_tokens_seen": 34992736,
      "step": 165810
    },
    {
      "epoch": 18.241474147414742,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0007010605345285525,
      "loss": 0.2303,
      "num_input_tokens_seen": 34993760,
      "step": 165815
    },
    {
      "epoch": 18.242024202420243,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0007006255023104768,
      "loss": 0.2298,
      "num_input_tokens_seen": 34994784,
      "step": 165820
    },
    {
      "epoch": 18.242574257425744,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0007001906018828713,
      "loss": 0.2303,
      "num_input_tokens_seen": 34995872,
      "step": 165825
    },
    {
      "epoch": 18.24312431243124,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.000699755833249746,
      "loss": 0.2298,
      "num_input_tokens_seen": 34996960,
      "step": 165830
    },
    {
      "epoch": 18.243674367436743,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.0006993211964151113,
      "loss": 0.2314,
      "num_input_tokens_seen": 34998016,
      "step": 165835
    },
    {
      "epoch": 18.244224422442244,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.0006988866913829722,
      "loss": 0.2303,
      "num_input_tokens_seen": 34999072,
      "step": 165840
    },
    {
      "epoch": 18.244774477447745,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0006984523181573321,
      "loss": 0.2298,
      "num_input_tokens_seen": 35000096,
      "step": 165845
    },
    {
      "epoch": 18.245324532453246,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.0006980180767421928,
      "loss": 0.2314,
      "num_input_tokens_seen": 35001120,
      "step": 165850
    },
    {
      "epoch": 18.245874587458747,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0006975839671415579,
      "loss": 0.2309,
      "num_input_tokens_seen": 35002144,
      "step": 165855
    },
    {
      "epoch": 18.246424642464245,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.000697149989359429,
      "loss": 0.2335,
      "num_input_tokens_seen": 35003200,
      "step": 165860
    },
    {
      "epoch": 18.246974697469746,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.0006967161433998065,
      "loss": 0.2303,
      "num_input_tokens_seen": 35004288,
      "step": 165865
    },
    {
      "epoch": 18.247524752475247,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0006962824292666886,
      "loss": 0.2308,
      "num_input_tokens_seen": 35005312,
      "step": 165870
    },
    {
      "epoch": 18.248074807480748,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0006958488469640705,
      "loss": 0.2319,
      "num_input_tokens_seen": 35006400,
      "step": 165875
    },
    {
      "epoch": 18.24862486248625,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0006954153964959509,
      "loss": 0.2324,
      "num_input_tokens_seen": 35007424,
      "step": 165880
    },
    {
      "epoch": 18.24917491749175,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.0006949820778663213,
      "loss": 0.2329,
      "num_input_tokens_seen": 35008480,
      "step": 165885
    },
    {
      "epoch": 18.24972497249725,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.0006945488910791786,
      "loss": 0.2308,
      "num_input_tokens_seen": 35009504,
      "step": 165890
    },
    {
      "epoch": 18.25027502750275,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0006941158361385163,
      "loss": 0.2303,
      "num_input_tokens_seen": 35010528,
      "step": 165895
    },
    {
      "epoch": 18.25082508250825,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0006936829130483212,
      "loss": 0.2319,
      "num_input_tokens_seen": 35011552,
      "step": 165900
    },
    {
      "epoch": 18.25137513751375,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.0006932501218125869,
      "loss": 0.2303,
      "num_input_tokens_seen": 35012640,
      "step": 165905
    },
    {
      "epoch": 18.251925192519252,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0006928174624353017,
      "loss": 0.2314,
      "num_input_tokens_seen": 35013696,
      "step": 165910
    },
    {
      "epoch": 18.252475247524753,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0006923849349204508,
      "loss": 0.2308,
      "num_input_tokens_seen": 35014752,
      "step": 165915
    },
    {
      "epoch": 18.253025302530254,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0006919525392720244,
      "loss": 0.2293,
      "num_input_tokens_seen": 35015776,
      "step": 165920
    },
    {
      "epoch": 18.253575357535752,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0006915202754940042,
      "loss": 0.2324,
      "num_input_tokens_seen": 35016832,
      "step": 165925
    },
    {
      "epoch": 18.254125412541253,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0006910881435903787,
      "loss": 0.2319,
      "num_input_tokens_seen": 35017856,
      "step": 165930
    },
    {
      "epoch": 18.254675467546754,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0006906561435651248,
      "loss": 0.2314,
      "num_input_tokens_seen": 35019008,
      "step": 165935
    },
    {
      "epoch": 18.255225522552255,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0006902242754222293,
      "loss": 0.2303,
      "num_input_tokens_seen": 35020096,
      "step": 165940
    },
    {
      "epoch": 18.255775577557756,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0006897925391656672,
      "loss": 0.2314,
      "num_input_tokens_seen": 35021152,
      "step": 165945
    },
    {
      "epoch": 18.256325632563257,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0006893609347994223,
      "loss": 0.2324,
      "num_input_tokens_seen": 35022240,
      "step": 165950
    },
    {
      "epoch": 18.25687568756876,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.000688929462327471,
      "loss": 0.2303,
      "num_input_tokens_seen": 35023328,
      "step": 165955
    },
    {
      "epoch": 18.257425742574256,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0006884981217537889,
      "loss": 0.2314,
      "num_input_tokens_seen": 35024416,
      "step": 165960
    },
    {
      "epoch": 18.257975797579757,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0006880669130823524,
      "loss": 0.2314,
      "num_input_tokens_seen": 35025472,
      "step": 165965
    },
    {
      "epoch": 18.258525852585258,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0006876358363171353,
      "loss": 0.2324,
      "num_input_tokens_seen": 35026528,
      "step": 165970
    },
    {
      "epoch": 18.25907590759076,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0006872048914621109,
      "loss": 0.2309,
      "num_input_tokens_seen": 35027584,
      "step": 165975
    },
    {
      "epoch": 18.25962596259626,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.0006867740785212528,
      "loss": 0.2319,
      "num_input_tokens_seen": 35028576,
      "step": 165980
    },
    {
      "epoch": 18.26017601760176,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0006863433974985278,
      "loss": 0.2314,
      "num_input_tokens_seen": 35029600,
      "step": 165985
    },
    {
      "epoch": 18.260726072607262,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0006859128483979093,
      "loss": 0.2329,
      "num_input_tokens_seen": 35030624,
      "step": 165990
    },
    {
      "epoch": 18.26127612761276,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0006854824312233643,
      "loss": 0.2314,
      "num_input_tokens_seen": 35031680,
      "step": 165995
    },
    {
      "epoch": 18.26182618261826,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0006850521459788561,
      "loss": 0.2329,
      "num_input_tokens_seen": 35032736,
      "step": 166000
    },
    {
      "epoch": 18.262376237623762,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0006846219926683533,
      "loss": 0.2324,
      "num_input_tokens_seen": 35033760,
      "step": 166005
    },
    {
      "epoch": 18.262926292629263,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0006841919712958228,
      "loss": 0.2319,
      "num_input_tokens_seen": 35034784,
      "step": 166010
    },
    {
      "epoch": 18.263476347634764,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0006837620818652229,
      "loss": 0.2319,
      "num_input_tokens_seen": 35035840,
      "step": 166015
    },
    {
      "epoch": 18.264026402640265,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0006833323243805223,
      "loss": 0.2319,
      "num_input_tokens_seen": 35036928,
      "step": 166020
    },
    {
      "epoch": 18.264576457645763,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0006829026988456744,
      "loss": 0.2335,
      "num_input_tokens_seen": 35038080,
      "step": 166025
    },
    {
      "epoch": 18.265126512651264,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.000682473205264641,
      "loss": 0.2314,
      "num_input_tokens_seen": 35039232,
      "step": 166030
    },
    {
      "epoch": 18.265676567656765,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0006820438436413856,
      "loss": 0.2303,
      "num_input_tokens_seen": 35040256,
      "step": 166035
    },
    {
      "epoch": 18.266226622662266,
      "grad_norm": 0.0006561279296875,
      "learning_rate": 0.0006816146139798585,
      "loss": 0.2324,
      "num_input_tokens_seen": 35041376,
      "step": 166040
    },
    {
      "epoch": 18.266776677667767,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.0006811855162840213,
      "loss": 0.2314,
      "num_input_tokens_seen": 35042432,
      "step": 166045
    },
    {
      "epoch": 18.26732673267327,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0006807565505578244,
      "loss": 0.2298,
      "num_input_tokens_seen": 35043456,
      "step": 166050
    },
    {
      "epoch": 18.26787678767877,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.0006803277168052263,
      "loss": 0.2314,
      "num_input_tokens_seen": 35044480,
      "step": 166055
    },
    {
      "epoch": 18.268426842684267,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0006798990150301737,
      "loss": 0.2314,
      "num_input_tokens_seen": 35045568,
      "step": 166060
    },
    {
      "epoch": 18.268976897689768,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0006794704452366201,
      "loss": 0.2329,
      "num_input_tokens_seen": 35046592,
      "step": 166065
    },
    {
      "epoch": 18.26952695269527,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0006790420074285175,
      "loss": 0.2308,
      "num_input_tokens_seen": 35047648,
      "step": 166070
    },
    {
      "epoch": 18.27007700770077,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0006786137016098126,
      "loss": 0.2293,
      "num_input_tokens_seen": 35048672,
      "step": 166075
    },
    {
      "epoch": 18.27062706270627,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0006781855277844523,
      "loss": 0.2335,
      "num_input_tokens_seen": 35049728,
      "step": 166080
    },
    {
      "epoch": 18.271177117711773,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0006777574859563834,
      "loss": 0.2314,
      "num_input_tokens_seen": 35050816,
      "step": 166085
    },
    {
      "epoch": 18.27172717271727,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0006773295761295511,
      "loss": 0.2324,
      "num_input_tokens_seen": 35051904,
      "step": 166090
    },
    {
      "epoch": 18.27227722772277,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0006769017983079022,
      "loss": 0.2314,
      "num_input_tokens_seen": 35053024,
      "step": 166095
    },
    {
      "epoch": 18.272827282728272,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0006764741524953738,
      "loss": 0.2308,
      "num_input_tokens_seen": 35054048,
      "step": 166100
    },
    {
      "epoch": 18.273377337733773,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0006760466386959124,
      "loss": 0.2329,
      "num_input_tokens_seen": 35055136,
      "step": 166105
    },
    {
      "epoch": 18.273927392739274,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.000675619256913455,
      "loss": 0.2319,
      "num_input_tokens_seen": 35056192,
      "step": 166110
    },
    {
      "epoch": 18.274477447744776,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0006751920071519401,
      "loss": 0.2314,
      "num_input_tokens_seen": 35057184,
      "step": 166115
    },
    {
      "epoch": 18.275027502750277,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0006747648894153063,
      "loss": 0.2303,
      "num_input_tokens_seen": 35058208,
      "step": 166120
    },
    {
      "epoch": 18.275577557755774,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0006743379037074904,
      "loss": 0.2319,
      "num_input_tokens_seen": 35059296,
      "step": 166125
    },
    {
      "epoch": 18.276127612761275,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0006739110500324308,
      "loss": 0.2314,
      "num_input_tokens_seen": 35060352,
      "step": 166130
    },
    {
      "epoch": 18.276677667766776,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0006734843283940577,
      "loss": 0.2324,
      "num_input_tokens_seen": 35061376,
      "step": 166135
    },
    {
      "epoch": 18.277227722772277,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.0006730577387963032,
      "loss": 0.2324,
      "num_input_tokens_seen": 35062464,
      "step": 166140
    },
    {
      "epoch": 18.27777777777778,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.0006726312812431023,
      "loss": 0.2298,
      "num_input_tokens_seen": 35063520,
      "step": 166145
    },
    {
      "epoch": 18.27832783278328,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.0006722049557383819,
      "loss": 0.2314,
      "num_input_tokens_seen": 35064640,
      "step": 166150
    },
    {
      "epoch": 18.278877887788777,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.0006717787622860771,
      "loss": 0.2324,
      "num_input_tokens_seen": 35065664,
      "step": 166155
    },
    {
      "epoch": 18.27942794279428,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.0006713527008901115,
      "loss": 0.2319,
      "num_input_tokens_seen": 35066720,
      "step": 166160
    },
    {
      "epoch": 18.27997799779978,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0006709267715544087,
      "loss": 0.2319,
      "num_input_tokens_seen": 35067808,
      "step": 166165
    },
    {
      "epoch": 18.28052805280528,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0006705009742829021,
      "loss": 0.2303,
      "num_input_tokens_seen": 35068832,
      "step": 166170
    },
    {
      "epoch": 18.28107810781078,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0006700753090795103,
      "loss": 0.2319,
      "num_input_tokens_seen": 35069920,
      "step": 166175
    },
    {
      "epoch": 18.281628162816283,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0006696497759481568,
      "loss": 0.2309,
      "num_input_tokens_seen": 35071008,
      "step": 166180
    },
    {
      "epoch": 18.282178217821784,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0006692243748927684,
      "loss": 0.2324,
      "num_input_tokens_seen": 35072128,
      "step": 166185
    },
    {
      "epoch": 18.28272827282728,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0006687991059172588,
      "loss": 0.2308,
      "num_input_tokens_seen": 35073184,
      "step": 166190
    },
    {
      "epoch": 18.283278327832782,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.000668373969025553,
      "loss": 0.2324,
      "num_input_tokens_seen": 35074240,
      "step": 166195
    },
    {
      "epoch": 18.283828382838283,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0006679489642215697,
      "loss": 0.2308,
      "num_input_tokens_seen": 35075296,
      "step": 166200
    },
    {
      "epoch": 18.284378437843785,
      "grad_norm": 0.001861572265625,
      "learning_rate": 0.0006675240915092173,
      "loss": 0.2319,
      "num_input_tokens_seen": 35076384,
      "step": 166205
    },
    {
      "epoch": 18.284928492849286,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0006670993508924227,
      "loss": 0.2324,
      "num_input_tokens_seen": 35077472,
      "step": 166210
    },
    {
      "epoch": 18.285478547854787,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0006666747423750946,
      "loss": 0.2335,
      "num_input_tokens_seen": 35078528,
      "step": 166215
    },
    {
      "epoch": 18.286028602860284,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.000666250265961148,
      "loss": 0.2308,
      "num_input_tokens_seen": 35079552,
      "step": 166220
    },
    {
      "epoch": 18.286578657865785,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.0006658259216544948,
      "loss": 0.2319,
      "num_input_tokens_seen": 35080608,
      "step": 166225
    },
    {
      "epoch": 18.287128712871286,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0006654017094590453,
      "loss": 0.2293,
      "num_input_tokens_seen": 35081664,
      "step": 166230
    },
    {
      "epoch": 18.287678767876788,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0006649776293787096,
      "loss": 0.2303,
      "num_input_tokens_seen": 35082720,
      "step": 166235
    },
    {
      "epoch": 18.28822882288229,
      "grad_norm": 0.001708984375,
      "learning_rate": 0.0006645536814173963,
      "loss": 0.2324,
      "num_input_tokens_seen": 35083776,
      "step": 166240
    },
    {
      "epoch": 18.28877887788779,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0006641298655790139,
      "loss": 0.2298,
      "num_input_tokens_seen": 35084864,
      "step": 166245
    },
    {
      "epoch": 18.28932893289329,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0006637061818674677,
      "loss": 0.2303,
      "num_input_tokens_seen": 35085888,
      "step": 166250
    },
    {
      "epoch": 18.28987898789879,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0006632826302866613,
      "loss": 0.2303,
      "num_input_tokens_seen": 35086912,
      "step": 166255
    },
    {
      "epoch": 18.29042904290429,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0006628592108405012,
      "loss": 0.2298,
      "num_input_tokens_seen": 35087936,
      "step": 166260
    },
    {
      "epoch": 18.29097909790979,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.000662435923532883,
      "loss": 0.2335,
      "num_input_tokens_seen": 35088992,
      "step": 166265
    },
    {
      "epoch": 18.29152915291529,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0006620127683677184,
      "loss": 0.2314,
      "num_input_tokens_seen": 35090144,
      "step": 166270
    },
    {
      "epoch": 18.292079207920793,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0006615897453489028,
      "loss": 0.2319,
      "num_input_tokens_seen": 35091232,
      "step": 166275
    },
    {
      "epoch": 18.292629262926294,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0006611668544803311,
      "loss": 0.2329,
      "num_input_tokens_seen": 35092224,
      "step": 166280
    },
    {
      "epoch": 18.293179317931795,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0006607440957659055,
      "loss": 0.2293,
      "num_input_tokens_seen": 35093216,
      "step": 166285
    },
    {
      "epoch": 18.293729372937293,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.0006603214692095194,
      "loss": 0.2309,
      "num_input_tokens_seen": 35094336,
      "step": 166290
    },
    {
      "epoch": 18.294279427942794,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0006598989748150696,
      "loss": 0.2303,
      "num_input_tokens_seen": 35095360,
      "step": 166295
    },
    {
      "epoch": 18.294829482948295,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0006594766125864531,
      "loss": 0.2319,
      "num_input_tokens_seen": 35096448,
      "step": 166300
    },
    {
      "epoch": 18.295379537953796,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0006590543825275552,
      "loss": 0.2335,
      "num_input_tokens_seen": 35097504,
      "step": 166305
    },
    {
      "epoch": 18.295929592959297,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0006586322846422743,
      "loss": 0.2308,
      "num_input_tokens_seen": 35098560,
      "step": 166310
    },
    {
      "epoch": 18.296479647964798,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0006582103189344973,
      "loss": 0.2303,
      "num_input_tokens_seen": 35099680,
      "step": 166315
    },
    {
      "epoch": 18.297029702970296,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0006577884854081145,
      "loss": 0.2319,
      "num_input_tokens_seen": 35100736,
      "step": 166320
    },
    {
      "epoch": 18.297579757975797,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0006573667840670128,
      "loss": 0.2293,
      "num_input_tokens_seen": 35101760,
      "step": 166325
    },
    {
      "epoch": 18.298129812981298,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0006569452149150773,
      "loss": 0.2309,
      "num_input_tokens_seen": 35102816,
      "step": 166330
    },
    {
      "epoch": 18.2986798679868,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0006565237779561983,
      "loss": 0.2314,
      "num_input_tokens_seen": 35103840,
      "step": 166335
    },
    {
      "epoch": 18.2992299229923,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0006561024731942544,
      "loss": 0.2298,
      "num_input_tokens_seen": 35104928,
      "step": 166340
    },
    {
      "epoch": 18.2997799779978,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0006556813006331341,
      "loss": 0.2319,
      "num_input_tokens_seen": 35106080,
      "step": 166345
    },
    {
      "epoch": 18.300330033003302,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.0006552602602767127,
      "loss": 0.2314,
      "num_input_tokens_seen": 35107168,
      "step": 166350
    },
    {
      "epoch": 18.3008800880088,
      "grad_norm": 0.00060272216796875,
      "learning_rate": 0.0006548393521288753,
      "loss": 0.2319,
      "num_input_tokens_seen": 35108192,
      "step": 166355
    },
    {
      "epoch": 18.3014301430143,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.0006544185761935022,
      "loss": 0.2314,
      "num_input_tokens_seen": 35109248,
      "step": 166360
    },
    {
      "epoch": 18.301980198019802,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.0006539979324744671,
      "loss": 0.2309,
      "num_input_tokens_seen": 35110304,
      "step": 166365
    },
    {
      "epoch": 18.302530253025303,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0006535774209756501,
      "loss": 0.2309,
      "num_input_tokens_seen": 35111328,
      "step": 166370
    },
    {
      "epoch": 18.303080308030804,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.000653157041700928,
      "loss": 0.2293,
      "num_input_tokens_seen": 35112384,
      "step": 166375
    },
    {
      "epoch": 18.303630363036305,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0006527367946541679,
      "loss": 0.2314,
      "num_input_tokens_seen": 35113472,
      "step": 166380
    },
    {
      "epoch": 18.304180418041803,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.0006523166798392515,
      "loss": 0.2334,
      "num_input_tokens_seen": 35114496,
      "step": 166385
    },
    {
      "epoch": 18.304730473047304,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0006518966972600493,
      "loss": 0.2314,
      "num_input_tokens_seen": 35115520,
      "step": 166390
    },
    {
      "epoch": 18.305280528052805,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.0006514768469204279,
      "loss": 0.2309,
      "num_input_tokens_seen": 35116608,
      "step": 166395
    },
    {
      "epoch": 18.305830583058306,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.0006510571288242611,
      "loss": 0.2324,
      "num_input_tokens_seen": 35117664,
      "step": 166400
    },
    {
      "epoch": 18.306380638063807,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0006506375429754124,
      "loss": 0.2319,
      "num_input_tokens_seen": 35118720,
      "step": 166405
    },
    {
      "epoch": 18.306930693069308,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.0006502180893777537,
      "loss": 0.2319,
      "num_input_tokens_seen": 35119776,
      "step": 166410
    },
    {
      "epoch": 18.30748074807481,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0006497987680351502,
      "loss": 0.2329,
      "num_input_tokens_seen": 35120832,
      "step": 166415
    },
    {
      "epoch": 18.308030803080307,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0006493795789514639,
      "loss": 0.2314,
      "num_input_tokens_seen": 35121856,
      "step": 166420
    },
    {
      "epoch": 18.308580858085808,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0006489605221305616,
      "loss": 0.2303,
      "num_input_tokens_seen": 35122912,
      "step": 166425
    },
    {
      "epoch": 18.30913091309131,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.000648541597576302,
      "loss": 0.2303,
      "num_input_tokens_seen": 35123968,
      "step": 166430
    },
    {
      "epoch": 18.30968096809681,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0006481228052925503,
      "loss": 0.2298,
      "num_input_tokens_seen": 35125024,
      "step": 166435
    },
    {
      "epoch": 18.31023102310231,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.0006477041452831616,
      "loss": 0.2314,
      "num_input_tokens_seen": 35126048,
      "step": 166440
    },
    {
      "epoch": 18.310781078107812,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0006472856175519964,
      "loss": 0.2314,
      "num_input_tokens_seen": 35127104,
      "step": 166445
    },
    {
      "epoch": 18.31133113311331,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0006468672221029148,
      "loss": 0.2293,
      "num_input_tokens_seen": 35128224,
      "step": 166450
    },
    {
      "epoch": 18.31188118811881,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0006464489589397704,
      "loss": 0.2314,
      "num_input_tokens_seen": 35129280,
      "step": 166455
    },
    {
      "epoch": 18.312431243124312,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0006460308280664184,
      "loss": 0.2335,
      "num_input_tokens_seen": 35130368,
      "step": 166460
    },
    {
      "epoch": 18.312981298129813,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0006456128294867108,
      "loss": 0.2319,
      "num_input_tokens_seen": 35131456,
      "step": 166465
    },
    {
      "epoch": 18.313531353135314,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0006451949632045029,
      "loss": 0.2324,
      "num_input_tokens_seen": 35132576,
      "step": 166470
    },
    {
      "epoch": 18.314081408140815,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0006447772292236464,
      "loss": 0.2303,
      "num_input_tokens_seen": 35133632,
      "step": 166475
    },
    {
      "epoch": 18.314631463146316,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0006443596275479885,
      "loss": 0.2303,
      "num_input_tokens_seen": 35134656,
      "step": 166480
    },
    {
      "epoch": 18.315181518151814,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0006439421581813809,
      "loss": 0.2319,
      "num_input_tokens_seen": 35135680,
      "step": 166485
    },
    {
      "epoch": 18.315731573157315,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0006435248211276689,
      "loss": 0.2319,
      "num_input_tokens_seen": 35136800,
      "step": 166490
    },
    {
      "epoch": 18.316281628162816,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.0006431076163906979,
      "loss": 0.2314,
      "num_input_tokens_seen": 35137856,
      "step": 166495
    },
    {
      "epoch": 18.316831683168317,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0006426905439743163,
      "loss": 0.2303,
      "num_input_tokens_seen": 35138880,
      "step": 166500
    },
    {
      "epoch": 18.317381738173818,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0006422736038823662,
      "loss": 0.2314,
      "num_input_tokens_seen": 35140000,
      "step": 166505
    },
    {
      "epoch": 18.31793179317932,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.000641856796118691,
      "loss": 0.2329,
      "num_input_tokens_seen": 35141024,
      "step": 166510
    },
    {
      "epoch": 18.318481848184817,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.0006414401206871345,
      "loss": 0.2329,
      "num_input_tokens_seen": 35142080,
      "step": 166515
    },
    {
      "epoch": 18.319031903190318,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.0006410235775915302,
      "loss": 0.2314,
      "num_input_tokens_seen": 35143072,
      "step": 166520
    },
    {
      "epoch": 18.31958195819582,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0006406071668357233,
      "loss": 0.2319,
      "num_input_tokens_seen": 35144096,
      "step": 166525
    },
    {
      "epoch": 18.32013201320132,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.000640190888423549,
      "loss": 0.2329,
      "num_input_tokens_seen": 35145120,
      "step": 166530
    },
    {
      "epoch": 18.32068206820682,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0006397747423588462,
      "loss": 0.2299,
      "num_input_tokens_seen": 35146144,
      "step": 166535
    },
    {
      "epoch": 18.321232123212322,
      "grad_norm": 0.00201416015625,
      "learning_rate": 0.0006393587286454499,
      "loss": 0.2303,
      "num_input_tokens_seen": 35147232,
      "step": 166540
    },
    {
      "epoch": 18.321782178217823,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0006389428472871922,
      "loss": 0.2298,
      "num_input_tokens_seen": 35148288,
      "step": 166545
    },
    {
      "epoch": 18.32233223322332,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.0006385270982879065,
      "loss": 0.2308,
      "num_input_tokens_seen": 35149408,
      "step": 166550
    },
    {
      "epoch": 18.322882288228822,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0006381114816514249,
      "loss": 0.2319,
      "num_input_tokens_seen": 35150496,
      "step": 166555
    },
    {
      "epoch": 18.323432343234323,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.0006376959973815793,
      "loss": 0.2303,
      "num_input_tokens_seen": 35151520,
      "step": 166560
    },
    {
      "epoch": 18.323982398239824,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0006372806454821983,
      "loss": 0.2308,
      "num_input_tokens_seen": 35152576,
      "step": 166565
    },
    {
      "epoch": 18.324532453245325,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0006368654259571072,
      "loss": 0.2319,
      "num_input_tokens_seen": 35153632,
      "step": 166570
    },
    {
      "epoch": 18.325082508250826,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0006364503388101394,
      "loss": 0.2324,
      "num_input_tokens_seen": 35154720,
      "step": 166575
    },
    {
      "epoch": 18.325632563256324,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.0006360353840451138,
      "loss": 0.2314,
      "num_input_tokens_seen": 35155744,
      "step": 166580
    },
    {
      "epoch": 18.326182618261825,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0006356205616658572,
      "loss": 0.2319,
      "num_input_tokens_seen": 35156768,
      "step": 166585
    },
    {
      "epoch": 18.326732673267326,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0006352058716761949,
      "loss": 0.2329,
      "num_input_tokens_seen": 35157824,
      "step": 166590
    },
    {
      "epoch": 18.327282728272827,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0006347913140799454,
      "loss": 0.2303,
      "num_input_tokens_seen": 35158880,
      "step": 166595
    },
    {
      "epoch": 18.32783278327833,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0006343768888809326,
      "loss": 0.2319,
      "num_input_tokens_seen": 35159936,
      "step": 166600
    },
    {
      "epoch": 18.32838283828383,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0006339625960829764,
      "loss": 0.2308,
      "num_input_tokens_seen": 35160992,
      "step": 166605
    },
    {
      "epoch": 18.32893289328933,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.0006335484356898907,
      "loss": 0.2309,
      "num_input_tokens_seen": 35162016,
      "step": 166610
    },
    {
      "epoch": 18.329482948294828,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.000633134407705494,
      "loss": 0.233,
      "num_input_tokens_seen": 35163040,
      "step": 166615
    },
    {
      "epoch": 18.33003300330033,
      "grad_norm": 0.0115966796875,
      "learning_rate": 0.0006327205121336049,
      "loss": 0.2303,
      "num_input_tokens_seen": 35164096,
      "step": 166620
    },
    {
      "epoch": 18.33058305830583,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.0006323067489780387,
      "loss": 0.2293,
      "num_input_tokens_seen": 35165216,
      "step": 166625
    },
    {
      "epoch": 18.33113311331133,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.0006318931182426073,
      "loss": 0.2303,
      "num_input_tokens_seen": 35166272,
      "step": 166630
    },
    {
      "epoch": 18.331683168316832,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.0006314796199311196,
      "loss": 0.233,
      "num_input_tokens_seen": 35167328,
      "step": 166635
    },
    {
      "epoch": 18.332233223322334,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0006310662540473904,
      "loss": 0.2288,
      "num_input_tokens_seen": 35168416,
      "step": 166640
    },
    {
      "epoch": 18.33278327832783,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0006306530205952288,
      "loss": 0.2309,
      "num_input_tokens_seen": 35169472,
      "step": 166645
    },
    {
      "epoch": 18.333333333333332,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0006302399195784447,
      "loss": 0.2309,
      "num_input_tokens_seen": 35170496,
      "step": 166650
    },
    {
      "epoch": 18.333883388338833,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0006298269510008436,
      "loss": 0.2314,
      "num_input_tokens_seen": 35171520,
      "step": 166655
    },
    {
      "epoch": 18.334433443344334,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0006294141148662308,
      "loss": 0.2329,
      "num_input_tokens_seen": 35172544,
      "step": 166660
    },
    {
      "epoch": 18.334983498349835,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0006290014111784131,
      "loss": 0.233,
      "num_input_tokens_seen": 35173632,
      "step": 166665
    },
    {
      "epoch": 18.335533553355337,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0006285888399411926,
      "loss": 0.2319,
      "num_input_tokens_seen": 35174624,
      "step": 166670
    },
    {
      "epoch": 18.336083608360838,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0006281764011583729,
      "loss": 0.2319,
      "num_input_tokens_seen": 35175712,
      "step": 166675
    },
    {
      "epoch": 18.336633663366335,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0006277640948337559,
      "loss": 0.2314,
      "num_input_tokens_seen": 35176768,
      "step": 166680
    },
    {
      "epoch": 18.337183718371836,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.0006273519209711403,
      "loss": 0.2314,
      "num_input_tokens_seen": 35177856,
      "step": 166685
    },
    {
      "epoch": 18.337733773377337,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0006269398795743264,
      "loss": 0.2324,
      "num_input_tokens_seen": 35178848,
      "step": 166690
    },
    {
      "epoch": 18.33828382838284,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0006265279706471094,
      "loss": 0.2314,
      "num_input_tokens_seen": 35179936,
      "step": 166695
    },
    {
      "epoch": 18.33883388338834,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0006261161941932864,
      "loss": 0.2324,
      "num_input_tokens_seen": 35180992,
      "step": 166700
    },
    {
      "epoch": 18.33938393839384,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.0006257045502166542,
      "loss": 0.2324,
      "num_input_tokens_seen": 35181984,
      "step": 166705
    },
    {
      "epoch": 18.33993399339934,
      "grad_norm": 0.0002956390380859375,
      "learning_rate": 0.0006252930387210048,
      "loss": 0.233,
      "num_input_tokens_seen": 35182944,
      "step": 166710
    },
    {
      "epoch": 18.34048404840484,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0006248816597101352,
      "loss": 0.2324,
      "num_input_tokens_seen": 35184032,
      "step": 166715
    },
    {
      "epoch": 18.34103410341034,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0006244704131878292,
      "loss": 0.2324,
      "num_input_tokens_seen": 35185056,
      "step": 166720
    },
    {
      "epoch": 18.34158415841584,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0006240592991578853,
      "loss": 0.2324,
      "num_input_tokens_seen": 35186112,
      "step": 166725
    },
    {
      "epoch": 18.342134213421343,
      "grad_norm": 0.00069427490234375,
      "learning_rate": 0.0006236483176240853,
      "loss": 0.2298,
      "num_input_tokens_seen": 35187136,
      "step": 166730
    },
    {
      "epoch": 18.342684268426844,
      "grad_norm": 0.000499725341796875,
      "learning_rate": 0.0006232374685902198,
      "loss": 0.2329,
      "num_input_tokens_seen": 35188128,
      "step": 166735
    },
    {
      "epoch": 18.343234323432345,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0006228267520600789,
      "loss": 0.2299,
      "num_input_tokens_seen": 35189152,
      "step": 166740
    },
    {
      "epoch": 18.343784378437842,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0006224161680374429,
      "loss": 0.2319,
      "num_input_tokens_seen": 35190176,
      "step": 166745
    },
    {
      "epoch": 18.344334433443343,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0006220057165261006,
      "loss": 0.2314,
      "num_input_tokens_seen": 35191232,
      "step": 166750
    },
    {
      "epoch": 18.344884488448844,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.0006215953975298288,
      "loss": 0.2319,
      "num_input_tokens_seen": 35192320,
      "step": 166755
    },
    {
      "epoch": 18.345434543454346,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0006211852110524146,
      "loss": 0.2329,
      "num_input_tokens_seen": 35193376,
      "step": 166760
    },
    {
      "epoch": 18.345984598459847,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0006207751570976382,
      "loss": 0.2319,
      "num_input_tokens_seen": 35194464,
      "step": 166765
    },
    {
      "epoch": 18.346534653465348,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0006203652356692751,
      "loss": 0.2314,
      "num_input_tokens_seen": 35195552,
      "step": 166770
    },
    {
      "epoch": 18.34708470847085,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0006199554467711071,
      "loss": 0.2298,
      "num_input_tokens_seen": 35196576,
      "step": 166775
    },
    {
      "epoch": 18.347634763476346,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0006195457904069096,
      "loss": 0.2324,
      "num_input_tokens_seen": 35197632,
      "step": 166780
    },
    {
      "epoch": 18.348184818481847,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0006191362665804561,
      "loss": 0.2308,
      "num_input_tokens_seen": 35198720,
      "step": 166785
    },
    {
      "epoch": 18.34873487348735,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0006187268752955238,
      "loss": 0.2319,
      "num_input_tokens_seen": 35199840,
      "step": 166790
    },
    {
      "epoch": 18.34928492849285,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.0006183176165558863,
      "loss": 0.2303,
      "num_input_tokens_seen": 35200864,
      "step": 166795
    },
    {
      "epoch": 18.34983498349835,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0006179084903653137,
      "loss": 0.2309,
      "num_input_tokens_seen": 35201920,
      "step": 166800
    },
    {
      "epoch": 18.350385038503852,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0006174994967275782,
      "loss": 0.2314,
      "num_input_tokens_seen": 35203040,
      "step": 166805
    },
    {
      "epoch": 18.35093509350935,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.000617090635646445,
      "loss": 0.2319,
      "num_input_tokens_seen": 35204096,
      "step": 166810
    },
    {
      "epoch": 18.35148514851485,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.0006166819071256879,
      "loss": 0.2283,
      "num_input_tokens_seen": 35205216,
      "step": 166815
    },
    {
      "epoch": 18.35203520352035,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0006162733111690738,
      "loss": 0.2298,
      "num_input_tokens_seen": 35206272,
      "step": 166820
    },
    {
      "epoch": 18.352585258525853,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0006158648477803646,
      "loss": 0.2314,
      "num_input_tokens_seen": 35207424,
      "step": 166825
    },
    {
      "epoch": 18.353135313531354,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.0006154565169633274,
      "loss": 0.2314,
      "num_input_tokens_seen": 35208448,
      "step": 166830
    },
    {
      "epoch": 18.353685368536855,
      "grad_norm": 0.0003643035888671875,
      "learning_rate": 0.0006150483187217242,
      "loss": 0.2329,
      "num_input_tokens_seen": 35209504,
      "step": 166835
    },
    {
      "epoch": 18.354235423542356,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0006146402530593203,
      "loss": 0.2308,
      "num_input_tokens_seen": 35210592,
      "step": 166840
    },
    {
      "epoch": 18.354785478547853,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.0006142323199798727,
      "loss": 0.2314,
      "num_input_tokens_seen": 35211584,
      "step": 166845
    },
    {
      "epoch": 18.355335533553355,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.0006138245194871417,
      "loss": 0.2309,
      "num_input_tokens_seen": 35212640,
      "step": 166850
    },
    {
      "epoch": 18.355885588558856,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0006134168515848909,
      "loss": 0.2319,
      "num_input_tokens_seen": 35213696,
      "step": 166855
    },
    {
      "epoch": 18.356435643564357,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.000613009316276869,
      "loss": 0.2303,
      "num_input_tokens_seen": 35214752,
      "step": 166860
    },
    {
      "epoch": 18.356985698569858,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0006126019135668397,
      "loss": 0.2308,
      "num_input_tokens_seen": 35215744,
      "step": 166865
    },
    {
      "epoch": 18.35753575357536,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.0006121946434585534,
      "loss": 0.2309,
      "num_input_tokens_seen": 35216800,
      "step": 166870
    },
    {
      "epoch": 18.358085808580856,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.0006117875059557654,
      "loss": 0.2303,
      "num_input_tokens_seen": 35217920,
      "step": 166875
    },
    {
      "epoch": 18.358635863586358,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.0006113805010622292,
      "loss": 0.2324,
      "num_input_tokens_seen": 35218880,
      "step": 166880
    },
    {
      "epoch": 18.35918591859186,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0006109736287816919,
      "loss": 0.2298,
      "num_input_tokens_seen": 35219968,
      "step": 166885
    },
    {
      "epoch": 18.35973597359736,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.0006105668891179089,
      "loss": 0.234,
      "num_input_tokens_seen": 35221088,
      "step": 166890
    },
    {
      "epoch": 18.36028602860286,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0006101602820746271,
      "loss": 0.2319,
      "num_input_tokens_seen": 35222144,
      "step": 166895
    },
    {
      "epoch": 18.360836083608362,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.0006097538076555903,
      "loss": 0.2319,
      "num_input_tokens_seen": 35223136,
      "step": 166900
    },
    {
      "epoch": 18.361386138613863,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.000609347465864547,
      "loss": 0.2324,
      "num_input_tokens_seen": 35224192,
      "step": 166905
    },
    {
      "epoch": 18.36193619361936,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0006089412567052427,
      "loss": 0.2319,
      "num_input_tokens_seen": 35225216,
      "step": 166910
    },
    {
      "epoch": 18.36248624862486,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.0006085351801814243,
      "loss": 0.2309,
      "num_input_tokens_seen": 35226336,
      "step": 166915
    },
    {
      "epoch": 18.363036303630363,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0006081292362968304,
      "loss": 0.2303,
      "num_input_tokens_seen": 35227392,
      "step": 166920
    },
    {
      "epoch": 18.363586358635864,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.0006077234250552015,
      "loss": 0.2298,
      "num_input_tokens_seen": 35228416,
      "step": 166925
    },
    {
      "epoch": 18.364136413641365,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0006073177464602813,
      "loss": 0.2308,
      "num_input_tokens_seen": 35229472,
      "step": 166930
    },
    {
      "epoch": 18.364686468646866,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0006069122005158067,
      "loss": 0.2303,
      "num_input_tokens_seen": 35230560,
      "step": 166935
    },
    {
      "epoch": 18.365236523652364,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0006065067872255164,
      "loss": 0.2314,
      "num_input_tokens_seen": 35231584,
      "step": 166940
    },
    {
      "epoch": 18.365786578657865,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0006061015065931474,
      "loss": 0.2314,
      "num_input_tokens_seen": 35232608,
      "step": 166945
    },
    {
      "epoch": 18.366336633663366,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0006056963586224317,
      "loss": 0.2329,
      "num_input_tokens_seen": 35233728,
      "step": 166950
    },
    {
      "epoch": 18.366886688668867,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0006052913433171081,
      "loss": 0.2288,
      "num_input_tokens_seen": 35234752,
      "step": 166955
    },
    {
      "epoch": 18.367436743674368,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0006048864606809035,
      "loss": 0.2329,
      "num_input_tokens_seen": 35235840,
      "step": 166960
    },
    {
      "epoch": 18.36798679867987,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0006044817107175548,
      "loss": 0.2308,
      "num_input_tokens_seen": 35236928,
      "step": 166965
    },
    {
      "epoch": 18.36853685368537,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.000604077093430791,
      "loss": 0.2324,
      "num_input_tokens_seen": 35237984,
      "step": 166970
    },
    {
      "epoch": 18.369086908690868,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0006036726088243388,
      "loss": 0.2335,
      "num_input_tokens_seen": 35239008,
      "step": 166975
    },
    {
      "epoch": 18.36963696369637,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.000603268256901932,
      "loss": 0.2308,
      "num_input_tokens_seen": 35240096,
      "step": 166980
    },
    {
      "epoch": 18.37018701870187,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.0006028640376672911,
      "loss": 0.2329,
      "num_input_tokens_seen": 35241120,
      "step": 166985
    },
    {
      "epoch": 18.37073707370737,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0006024599511241413,
      "loss": 0.2303,
      "num_input_tokens_seen": 35242208,
      "step": 166990
    },
    {
      "epoch": 18.371287128712872,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0006020559972762129,
      "loss": 0.2314,
      "num_input_tokens_seen": 35243264,
      "step": 166995
    },
    {
      "epoch": 18.371837183718373,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0006016521761272247,
      "loss": 0.2324,
      "num_input_tokens_seen": 35244320,
      "step": 167000
    },
    {
      "epoch": 18.37238723872387,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0006012484876809004,
      "loss": 0.2303,
      "num_input_tokens_seen": 35245440,
      "step": 167005
    },
    {
      "epoch": 18.372937293729372,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.0006008449319409603,
      "loss": 0.2303,
      "num_input_tokens_seen": 35246496,
      "step": 167010
    },
    {
      "epoch": 18.373487348734873,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.0006004415089111198,
      "loss": 0.2314,
      "num_input_tokens_seen": 35247552,
      "step": 167015
    },
    {
      "epoch": 18.374037403740374,
      "grad_norm": 0.00067901611328125,
      "learning_rate": 0.0006000382185951025,
      "loss": 0.233,
      "num_input_tokens_seen": 35248608,
      "step": 167020
    },
    {
      "epoch": 18.374587458745875,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0005996350609966206,
      "loss": 0.2329,
      "num_input_tokens_seen": 35249600,
      "step": 167025
    },
    {
      "epoch": 18.375137513751376,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0005992320361193959,
      "loss": 0.2303,
      "num_input_tokens_seen": 35250624,
      "step": 167030
    },
    {
      "epoch": 18.375687568756877,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0005988291439671389,
      "loss": 0.2319,
      "num_input_tokens_seen": 35251712,
      "step": 167035
    },
    {
      "epoch": 18.376237623762375,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.0005984263845435617,
      "loss": 0.2319,
      "num_input_tokens_seen": 35252800,
      "step": 167040
    },
    {
      "epoch": 18.376787678767876,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0005980237578523794,
      "loss": 0.2309,
      "num_input_tokens_seen": 35253824,
      "step": 167045
    },
    {
      "epoch": 18.377337733773377,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0005976212638972977,
      "loss": 0.2308,
      "num_input_tokens_seen": 35254848,
      "step": 167050
    },
    {
      "epoch": 18.377887788778878,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0005972189026820351,
      "loss": 0.2324,
      "num_input_tokens_seen": 35255872,
      "step": 167055
    },
    {
      "epoch": 18.37843784378438,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.000596816674210292,
      "loss": 0.2319,
      "num_input_tokens_seen": 35256864,
      "step": 167060
    },
    {
      "epoch": 18.37898789878988,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0005964145784857788,
      "loss": 0.2314,
      "num_input_tokens_seen": 35257824,
      "step": 167065
    },
    {
      "epoch": 18.379537953795378,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.0005960126155122026,
      "loss": 0.2319,
      "num_input_tokens_seen": 35258880,
      "step": 167070
    },
    {
      "epoch": 18.38008800880088,
      "grad_norm": 0.000675201416015625,
      "learning_rate": 0.0005956107852932635,
      "loss": 0.2308,
      "num_input_tokens_seen": 35259872,
      "step": 167075
    },
    {
      "epoch": 18.38063806380638,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0005952090878326671,
      "loss": 0.2319,
      "num_input_tokens_seen": 35260896,
      "step": 167080
    },
    {
      "epoch": 18.38118811881188,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.0005948075231341204,
      "loss": 0.2303,
      "num_input_tokens_seen": 35261952,
      "step": 167085
    },
    {
      "epoch": 18.381738173817382,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005944060912013171,
      "loss": 0.2319,
      "num_input_tokens_seen": 35262944,
      "step": 167090
    },
    {
      "epoch": 18.382288228822883,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.0005940047920379626,
      "loss": 0.2319,
      "num_input_tokens_seen": 35263968,
      "step": 167095
    },
    {
      "epoch": 18.382838283828384,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0005936036256477522,
      "loss": 0.2319,
      "num_input_tokens_seen": 35265056,
      "step": 167100
    },
    {
      "epoch": 18.383388338833882,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0005932025920343864,
      "loss": 0.2324,
      "num_input_tokens_seen": 35266176,
      "step": 167105
    },
    {
      "epoch": 18.383938393839383,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0005928016912015554,
      "loss": 0.2314,
      "num_input_tokens_seen": 35267328,
      "step": 167110
    },
    {
      "epoch": 18.384488448844884,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0005924009231529598,
      "loss": 0.2319,
      "num_input_tokens_seen": 35268288,
      "step": 167115
    },
    {
      "epoch": 18.385038503850385,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0005920002878922931,
      "loss": 0.2313,
      "num_input_tokens_seen": 35269344,
      "step": 167120
    },
    {
      "epoch": 18.385588558855886,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0005915997854232424,
      "loss": 0.2324,
      "num_input_tokens_seen": 35270368,
      "step": 167125
    },
    {
      "epoch": 18.386138613861387,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0005911994157495065,
      "loss": 0.2314,
      "num_input_tokens_seen": 35271456,
      "step": 167130
    },
    {
      "epoch": 18.38668866886689,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0005907991788747691,
      "loss": 0.2319,
      "num_input_tokens_seen": 35272544,
      "step": 167135
    },
    {
      "epoch": 18.387238723872386,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.0005903990748027221,
      "loss": 0.2298,
      "num_input_tokens_seen": 35273568,
      "step": 167140
    },
    {
      "epoch": 18.387788778877887,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0005899991035370528,
      "loss": 0.2308,
      "num_input_tokens_seen": 35274656,
      "step": 167145
    },
    {
      "epoch": 18.388338833883388,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0005895992650814463,
      "loss": 0.2298,
      "num_input_tokens_seen": 35275744,
      "step": 167150
    },
    {
      "epoch": 18.38888888888889,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.0005891995594395915,
      "loss": 0.2319,
      "num_input_tokens_seen": 35276832,
      "step": 167155
    },
    {
      "epoch": 18.38943894389439,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.0005887999866151672,
      "loss": 0.2324,
      "num_input_tokens_seen": 35277920,
      "step": 167160
    },
    {
      "epoch": 18.38998899889989,
      "grad_norm": 0.0018310546875,
      "learning_rate": 0.0005884005466118553,
      "loss": 0.2329,
      "num_input_tokens_seen": 35279072,
      "step": 167165
    },
    {
      "epoch": 18.39053905390539,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.0005880012394333445,
      "loss": 0.2309,
      "num_input_tokens_seen": 35280160,
      "step": 167170
    },
    {
      "epoch": 18.39108910891089,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.0005876020650833103,
      "loss": 0.2329,
      "num_input_tokens_seen": 35281216,
      "step": 167175
    },
    {
      "epoch": 18.39163916391639,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0005872030235654313,
      "loss": 0.2329,
      "num_input_tokens_seen": 35282272,
      "step": 167180
    },
    {
      "epoch": 18.392189218921892,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0005868041148833863,
      "loss": 0.2308,
      "num_input_tokens_seen": 35283328,
      "step": 167185
    },
    {
      "epoch": 18.392739273927393,
      "grad_norm": 0.000644683837890625,
      "learning_rate": 0.0005864053390408508,
      "loss": 0.2308,
      "num_input_tokens_seen": 35284352,
      "step": 167190
    },
    {
      "epoch": 18.393289328932894,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0005860066960415,
      "loss": 0.2303,
      "num_input_tokens_seen": 35285440,
      "step": 167195
    },
    {
      "epoch": 18.393839383938396,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.0005856081858890128,
      "loss": 0.2319,
      "num_input_tokens_seen": 35286528,
      "step": 167200
    },
    {
      "epoch": 18.394389438943893,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0005852098085870544,
      "loss": 0.2298,
      "num_input_tokens_seen": 35287584,
      "step": 167205
    },
    {
      "epoch": 18.394939493949394,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005848115641393037,
      "loss": 0.2319,
      "num_input_tokens_seen": 35288704,
      "step": 167210
    },
    {
      "epoch": 18.395489548954895,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0005844134525494244,
      "loss": 0.2304,
      "num_input_tokens_seen": 35289696,
      "step": 167215
    },
    {
      "epoch": 18.396039603960396,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0005840154738210917,
      "loss": 0.2314,
      "num_input_tokens_seen": 35290720,
      "step": 167220
    },
    {
      "epoch": 18.396589658965897,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0005836176279579696,
      "loss": 0.2335,
      "num_input_tokens_seen": 35291776,
      "step": 167225
    },
    {
      "epoch": 18.3971397139714,
      "grad_norm": 0.00054168701171875,
      "learning_rate": 0.0005832199149637252,
      "loss": 0.2314,
      "num_input_tokens_seen": 35292800,
      "step": 167230
    },
    {
      "epoch": 18.397689768976896,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0005828223348420269,
      "loss": 0.2308,
      "num_input_tokens_seen": 35293888,
      "step": 167235
    },
    {
      "epoch": 18.398239823982397,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.0005824248875965354,
      "loss": 0.2319,
      "num_input_tokens_seen": 35294944,
      "step": 167240
    },
    {
      "epoch": 18.3987898789879,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0005820275732309176,
      "loss": 0.2319,
      "num_input_tokens_seen": 35296032,
      "step": 167245
    },
    {
      "epoch": 18.3993399339934,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005816303917488308,
      "loss": 0.2309,
      "num_input_tokens_seen": 35297152,
      "step": 167250
    },
    {
      "epoch": 18.3998899889989,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0005812333431539384,
      "loss": 0.2309,
      "num_input_tokens_seen": 35298208,
      "step": 167255
    },
    {
      "epoch": 18.4004400440044,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.0005808364274499028,
      "loss": 0.2303,
      "num_input_tokens_seen": 35299296,
      "step": 167260
    },
    {
      "epoch": 18.400990099009903,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0005804396446403743,
      "loss": 0.2314,
      "num_input_tokens_seen": 35300384,
      "step": 167265
    },
    {
      "epoch": 18.4015401540154,
      "grad_norm": 0.000518798828125,
      "learning_rate": 0.000580042994729018,
      "loss": 0.2299,
      "num_input_tokens_seen": 35301440,
      "step": 167270
    },
    {
      "epoch": 18.4020902090209,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0005796464777194865,
      "loss": 0.2303,
      "num_input_tokens_seen": 35302528,
      "step": 167275
    },
    {
      "epoch": 18.402640264026402,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0005792500936154299,
      "loss": 0.234,
      "num_input_tokens_seen": 35303584,
      "step": 167280
    },
    {
      "epoch": 18.403190319031903,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.000578853842420507,
      "loss": 0.234,
      "num_input_tokens_seen": 35304576,
      "step": 167285
    },
    {
      "epoch": 18.403740374037405,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0005784577241383682,
      "loss": 0.2298,
      "num_input_tokens_seen": 35305664,
      "step": 167290
    },
    {
      "epoch": 18.404290429042906,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0005780617387726672,
      "loss": 0.2303,
      "num_input_tokens_seen": 35306720,
      "step": 167295
    },
    {
      "epoch": 18.404840484048403,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0005776658863270495,
      "loss": 0.2309,
      "num_input_tokens_seen": 35307808,
      "step": 167300
    },
    {
      "epoch": 18.405390539053904,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.0005772701668051622,
      "loss": 0.2309,
      "num_input_tokens_seen": 35308896,
      "step": 167305
    },
    {
      "epoch": 18.405940594059405,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0005768745802106573,
      "loss": 0.2308,
      "num_input_tokens_seen": 35310016,
      "step": 167310
    },
    {
      "epoch": 18.406490649064907,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0005764791265471769,
      "loss": 0.2319,
      "num_input_tokens_seen": 35311040,
      "step": 167315
    },
    {
      "epoch": 18.407040704070408,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0005760838058183698,
      "loss": 0.2309,
      "num_input_tokens_seen": 35312128,
      "step": 167320
    },
    {
      "epoch": 18.40759075907591,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0005756886180278764,
      "loss": 0.2319,
      "num_input_tokens_seen": 35313152,
      "step": 167325
    },
    {
      "epoch": 18.40814081408141,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0005752935631793371,
      "loss": 0.2319,
      "num_input_tokens_seen": 35314240,
      "step": 167330
    },
    {
      "epoch": 18.408690869086907,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 0.000574898641276399,
      "loss": 0.2329,
      "num_input_tokens_seen": 35315264,
      "step": 167335
    },
    {
      "epoch": 18.40924092409241,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.0005745038523226958,
      "loss": 0.2293,
      "num_input_tokens_seen": 35316320,
      "step": 167340
    },
    {
      "epoch": 18.40979097909791,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.0005741091963218681,
      "loss": 0.2314,
      "num_input_tokens_seen": 35317408,
      "step": 167345
    },
    {
      "epoch": 18.41034103410341,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.0005737146732775561,
      "loss": 0.2308,
      "num_input_tokens_seen": 35318368,
      "step": 167350
    },
    {
      "epoch": 18.41089108910891,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0005733202831933903,
      "loss": 0.2314,
      "num_input_tokens_seen": 35319424,
      "step": 167355
    },
    {
      "epoch": 18.411441144114413,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0005729260260730129,
      "loss": 0.2314,
      "num_input_tokens_seen": 35320480,
      "step": 167360
    },
    {
      "epoch": 18.41199119911991,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0005725319019200508,
      "loss": 0.2309,
      "num_input_tokens_seen": 35321536,
      "step": 167365
    },
    {
      "epoch": 18.41254125412541,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.000572137910738138,
      "loss": 0.2308,
      "num_input_tokens_seen": 35322592,
      "step": 167370
    },
    {
      "epoch": 18.413091309130913,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0005717440525309098,
      "loss": 0.2314,
      "num_input_tokens_seen": 35323680,
      "step": 167375
    },
    {
      "epoch": 18.413641364136414,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.0005713503273019915,
      "loss": 0.2293,
      "num_input_tokens_seen": 35324704,
      "step": 167380
    },
    {
      "epoch": 18.414191419141915,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005709567350550154,
      "loss": 0.2308,
      "num_input_tokens_seen": 35325824,
      "step": 167385
    },
    {
      "epoch": 18.414741474147416,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0005705632757936085,
      "loss": 0.2319,
      "num_input_tokens_seen": 35326848,
      "step": 167390
    },
    {
      "epoch": 18.415291529152917,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.000570169949521393,
      "loss": 0.2298,
      "num_input_tokens_seen": 35327968,
      "step": 167395
    },
    {
      "epoch": 18.415841584158414,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0005697767562419958,
      "loss": 0.2308,
      "num_input_tokens_seen": 35328992,
      "step": 167400
    },
    {
      "epoch": 18.416391639163916,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0005693836959590442,
      "loss": 0.2309,
      "num_input_tokens_seen": 35330048,
      "step": 167405
    },
    {
      "epoch": 18.416941694169417,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0005689907686761585,
      "loss": 0.2308,
      "num_input_tokens_seen": 35331104,
      "step": 167410
    },
    {
      "epoch": 18.417491749174918,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.000568597974396961,
      "loss": 0.2298,
      "num_input_tokens_seen": 35332160,
      "step": 167415
    },
    {
      "epoch": 18.41804180418042,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005682053131250703,
      "loss": 0.2303,
      "num_input_tokens_seen": 35333216,
      "step": 167420
    },
    {
      "epoch": 18.41859185918592,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0005678127848641051,
      "loss": 0.2308,
      "num_input_tokens_seen": 35334304,
      "step": 167425
    },
    {
      "epoch": 18.419141914191417,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0005674203896176844,
      "loss": 0.2298,
      "num_input_tokens_seen": 35335360,
      "step": 167430
    },
    {
      "epoch": 18.41969196919692,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0005670281273894267,
      "loss": 0.2324,
      "num_input_tokens_seen": 35336416,
      "step": 167435
    },
    {
      "epoch": 18.42024202420242,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0005666359981829444,
      "loss": 0.2314,
      "num_input_tokens_seen": 35337440,
      "step": 167440
    },
    {
      "epoch": 18.42079207920792,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0005662440020018511,
      "loss": 0.2303,
      "num_input_tokens_seen": 35338496,
      "step": 167445
    },
    {
      "epoch": 18.421342134213422,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005658521388497622,
      "loss": 0.2324,
      "num_input_tokens_seen": 35339488,
      "step": 167450
    },
    {
      "epoch": 18.421892189218923,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0005654604087302866,
      "loss": 0.2314,
      "num_input_tokens_seen": 35340608,
      "step": 167455
    },
    {
      "epoch": 18.422442244224424,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0005650688116470364,
      "loss": 0.2293,
      "num_input_tokens_seen": 35341632,
      "step": 167460
    },
    {
      "epoch": 18.42299229922992,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.0005646773476036204,
      "loss": 0.2324,
      "num_input_tokens_seen": 35342720,
      "step": 167465
    },
    {
      "epoch": 18.423542354235423,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.0005642860166036473,
      "loss": 0.2324,
      "num_input_tokens_seen": 35343744,
      "step": 167470
    },
    {
      "epoch": 18.424092409240924,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.0005638948186507225,
      "loss": 0.2319,
      "num_input_tokens_seen": 35344768,
      "step": 167475
    },
    {
      "epoch": 18.424642464246425,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0005635037537484516,
      "loss": 0.2309,
      "num_input_tokens_seen": 35345792,
      "step": 167480
    },
    {
      "epoch": 18.425192519251926,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00056311282190044,
      "loss": 0.2308,
      "num_input_tokens_seen": 35346848,
      "step": 167485
    },
    {
      "epoch": 18.425742574257427,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0005627220231102897,
      "loss": 0.2319,
      "num_input_tokens_seen": 35347936,
      "step": 167490
    },
    {
      "epoch": 18.426292629262925,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0005623313573816029,
      "loss": 0.2324,
      "num_input_tokens_seen": 35348928,
      "step": 167495
    },
    {
      "epoch": 18.426842684268426,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0005619408247179802,
      "loss": 0.2314,
      "num_input_tokens_seen": 35349984,
      "step": 167500
    },
    {
      "epoch": 18.427392739273927,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.0005615504251230202,
      "loss": 0.2314,
      "num_input_tokens_seen": 35351040,
      "step": 167505
    },
    {
      "epoch": 18.427942794279428,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0005611601586003234,
      "loss": 0.2314,
      "num_input_tokens_seen": 35352096,
      "step": 167510
    },
    {
      "epoch": 18.42849284928493,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.0005607700251534836,
      "loss": 0.2314,
      "num_input_tokens_seen": 35353120,
      "step": 167515
    },
    {
      "epoch": 18.42904290429043,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.000560380024786098,
      "loss": 0.2319,
      "num_input_tokens_seen": 35354144,
      "step": 167520
    },
    {
      "epoch": 18.42959295929593,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0005599901575017619,
      "loss": 0.2324,
      "num_input_tokens_seen": 35355200,
      "step": 167525
    },
    {
      "epoch": 18.43014301430143,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0005596004233040658,
      "loss": 0.2303,
      "num_input_tokens_seen": 35356224,
      "step": 167530
    },
    {
      "epoch": 18.43069306930693,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.0005592108221966052,
      "loss": 0.2314,
      "num_input_tokens_seen": 35357280,
      "step": 167535
    },
    {
      "epoch": 18.43124312431243,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0005588213541829673,
      "loss": 0.2319,
      "num_input_tokens_seen": 35358304,
      "step": 167540
    },
    {
      "epoch": 18.431793179317932,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005584320192667425,
      "loss": 0.2303,
      "num_input_tokens_seen": 35359328,
      "step": 167545
    },
    {
      "epoch": 18.432343234323433,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0005580428174515228,
      "loss": 0.2324,
      "num_input_tokens_seen": 35360288,
      "step": 167550
    },
    {
      "epoch": 18.432893289328934,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.0005576537487408939,
      "loss": 0.2314,
      "num_input_tokens_seen": 35361280,
      "step": 167555
    },
    {
      "epoch": 18.433443344334435,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0005572648131384361,
      "loss": 0.2319,
      "num_input_tokens_seen": 35362336,
      "step": 167560
    },
    {
      "epoch": 18.433993399339933,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0005568760106477416,
      "loss": 0.2304,
      "num_input_tokens_seen": 35363392,
      "step": 167565
    },
    {
      "epoch": 18.434543454345434,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005564873412723875,
      "loss": 0.2324,
      "num_input_tokens_seen": 35364416,
      "step": 167570
    },
    {
      "epoch": 18.435093509350935,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.000556098805015961,
      "loss": 0.2303,
      "num_input_tokens_seen": 35365504,
      "step": 167575
    },
    {
      "epoch": 18.435643564356436,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.0005557104018820407,
      "loss": 0.2308,
      "num_input_tokens_seen": 35366624,
      "step": 167580
    },
    {
      "epoch": 18.436193619361937,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005553221318742057,
      "loss": 0.2314,
      "num_input_tokens_seen": 35367680,
      "step": 167585
    },
    {
      "epoch": 18.436743674367438,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.0005549339949960364,
      "loss": 0.2329,
      "num_input_tokens_seen": 35368768,
      "step": 167590
    },
    {
      "epoch": 18.437293729372936,
      "grad_norm": 0.005859375,
      "learning_rate": 0.000554545991251108,
      "loss": 0.2314,
      "num_input_tokens_seen": 35369888,
      "step": 167595
    },
    {
      "epoch": 18.437843784378437,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.000554158120642998,
      "loss": 0.2303,
      "num_input_tokens_seen": 35370912,
      "step": 167600
    },
    {
      "epoch": 18.438393839383938,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0005537703831752816,
      "loss": 0.2288,
      "num_input_tokens_seen": 35371936,
      "step": 167605
    },
    {
      "epoch": 18.43894389438944,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.0005533827788515294,
      "loss": 0.2293,
      "num_input_tokens_seen": 35372992,
      "step": 167610
    },
    {
      "epoch": 18.43949394939494,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0005529953076753185,
      "loss": 0.2308,
      "num_input_tokens_seen": 35374016,
      "step": 167615
    },
    {
      "epoch": 18.44004400440044,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0005526079696502161,
      "loss": 0.2314,
      "num_input_tokens_seen": 35375008,
      "step": 167620
    },
    {
      "epoch": 18.440594059405942,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005522207647797944,
      "loss": 0.2314,
      "num_input_tokens_seen": 35376064,
      "step": 167625
    },
    {
      "epoch": 18.44114411441144,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0005518336930676204,
      "loss": 0.2319,
      "num_input_tokens_seen": 35377184,
      "step": 167630
    },
    {
      "epoch": 18.44169416941694,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0005514467545172613,
      "loss": 0.2298,
      "num_input_tokens_seen": 35378240,
      "step": 167635
    },
    {
      "epoch": 18.442244224422442,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0005510599491322876,
      "loss": 0.2319,
      "num_input_tokens_seen": 35379264,
      "step": 167640
    },
    {
      "epoch": 18.442794279427943,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.000550673276916258,
      "loss": 0.2319,
      "num_input_tokens_seen": 35380320,
      "step": 167645
    },
    {
      "epoch": 18.443344334433444,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0005502867378727416,
      "loss": 0.2308,
      "num_input_tokens_seen": 35381344,
      "step": 167650
    },
    {
      "epoch": 18.443894389438945,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005499003320052969,
      "loss": 0.2309,
      "num_input_tokens_seen": 35382432,
      "step": 167655
    },
    {
      "epoch": 18.444444444444443,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0005495140593174863,
      "loss": 0.2303,
      "num_input_tokens_seen": 35383456,
      "step": 167660
    },
    {
      "epoch": 18.444994499449944,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0005491279198128734,
      "loss": 0.2298,
      "num_input_tokens_seen": 35384512,
      "step": 167665
    },
    {
      "epoch": 18.445544554455445,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005487419134950122,
      "loss": 0.2293,
      "num_input_tokens_seen": 35385568,
      "step": 167670
    },
    {
      "epoch": 18.446094609460946,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0005483560403674647,
      "loss": 0.2314,
      "num_input_tokens_seen": 35386592,
      "step": 167675
    },
    {
      "epoch": 18.446644664466447,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0005479703004337833,
      "loss": 0.2309,
      "num_input_tokens_seen": 35387616,
      "step": 167680
    },
    {
      "epoch": 18.44719471947195,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0005475846936975248,
      "loss": 0.2308,
      "num_input_tokens_seen": 35388704,
      "step": 167685
    },
    {
      "epoch": 18.44774477447745,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005471992201622416,
      "loss": 0.2309,
      "num_input_tokens_seen": 35389696,
      "step": 167690
    },
    {
      "epoch": 18.448294829482947,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.0005468138798314892,
      "loss": 0.2329,
      "num_input_tokens_seen": 35390688,
      "step": 167695
    },
    {
      "epoch": 18.448844884488448,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0005464286727088197,
      "loss": 0.2329,
      "num_input_tokens_seen": 35391680,
      "step": 167700
    },
    {
      "epoch": 18.44939493949395,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0005460435987977818,
      "loss": 0.2303,
      "num_input_tokens_seen": 35392864,
      "step": 167705
    },
    {
      "epoch": 18.44994499449945,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0005456586581019229,
      "loss": 0.2314,
      "num_input_tokens_seen": 35393920,
      "step": 167710
    },
    {
      "epoch": 18.45049504950495,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.0005452738506247917,
      "loss": 0.2324,
      "num_input_tokens_seen": 35394976,
      "step": 167715
    },
    {
      "epoch": 18.451045104510452,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0005448891763699354,
      "loss": 0.2309,
      "num_input_tokens_seen": 35396064,
      "step": 167720
    },
    {
      "epoch": 18.45159515951595,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.0005445046353409011,
      "loss": 0.2319,
      "num_input_tokens_seen": 35397088,
      "step": 167725
    },
    {
      "epoch": 18.45214521452145,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.000544120227541231,
      "loss": 0.2324,
      "num_input_tokens_seen": 35398112,
      "step": 167730
    },
    {
      "epoch": 18.452695269526952,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0005437359529744657,
      "loss": 0.2314,
      "num_input_tokens_seen": 35399232,
      "step": 167735
    },
    {
      "epoch": 18.453245324532453,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 0.0005433518116441521,
      "loss": 0.2334,
      "num_input_tokens_seen": 35400352,
      "step": 167740
    },
    {
      "epoch": 18.453795379537954,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.000542967803553826,
      "loss": 0.2324,
      "num_input_tokens_seen": 35401408,
      "step": 167745
    },
    {
      "epoch": 18.454345434543455,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0005425839287070294,
      "loss": 0.2314,
      "num_input_tokens_seen": 35402432,
      "step": 167750
    },
    {
      "epoch": 18.454895489548957,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.0005422001871072995,
      "loss": 0.2329,
      "num_input_tokens_seen": 35403488,
      "step": 167755
    },
    {
      "epoch": 18.455445544554454,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0005418165787581719,
      "loss": 0.2313,
      "num_input_tokens_seen": 35404544,
      "step": 167760
    },
    {
      "epoch": 18.455995599559955,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0005414331036631853,
      "loss": 0.2309,
      "num_input_tokens_seen": 35405568,
      "step": 167765
    },
    {
      "epoch": 18.456545654565456,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.000541049761825872,
      "loss": 0.2314,
      "num_input_tokens_seen": 35406688,
      "step": 167770
    },
    {
      "epoch": 18.457095709570957,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.0005406665532497607,
      "loss": 0.2308,
      "num_input_tokens_seen": 35407712,
      "step": 167775
    },
    {
      "epoch": 18.45764576457646,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0005402834779383903,
      "loss": 0.2309,
      "num_input_tokens_seen": 35408768,
      "step": 167780
    },
    {
      "epoch": 18.45819581958196,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.000539900535895288,
      "loss": 0.2319,
      "num_input_tokens_seen": 35409792,
      "step": 167785
    },
    {
      "epoch": 18.458745874587457,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0005395177271239859,
      "loss": 0.2319,
      "num_input_tokens_seen": 35410816,
      "step": 167790
    },
    {
      "epoch": 18.459295929592958,
      "grad_norm": 0.0005950927734375,
      "learning_rate": 0.0005391350516280096,
      "loss": 0.233,
      "num_input_tokens_seen": 35411872,
      "step": 167795
    },
    {
      "epoch": 18.45984598459846,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0005387525094108847,
      "loss": 0.2329,
      "num_input_tokens_seen": 35412992,
      "step": 167800
    },
    {
      "epoch": 18.46039603960396,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.0005383701004761382,
      "loss": 0.2314,
      "num_input_tokens_seen": 35414144,
      "step": 167805
    },
    {
      "epoch": 18.46094609460946,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.0005379878248272956,
      "loss": 0.2319,
      "num_input_tokens_seen": 35415136,
      "step": 167810
    },
    {
      "epoch": 18.461496149614963,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0005376056824678826,
      "loss": 0.2309,
      "num_input_tokens_seen": 35416224,
      "step": 167815
    },
    {
      "epoch": 18.462046204620464,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0005372236734014163,
      "loss": 0.2314,
      "num_input_tokens_seen": 35417248,
      "step": 167820
    },
    {
      "epoch": 18.46259625962596,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0005368417976314188,
      "loss": 0.2324,
      "num_input_tokens_seen": 35418304,
      "step": 167825
    },
    {
      "epoch": 18.463146314631462,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0005364600551614123,
      "loss": 0.2313,
      "num_input_tokens_seen": 35419328,
      "step": 167830
    },
    {
      "epoch": 18.463696369636963,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0005360784459949091,
      "loss": 0.2319,
      "num_input_tokens_seen": 35420416,
      "step": 167835
    },
    {
      "epoch": 18.464246424642464,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0005356969701354347,
      "loss": 0.2324,
      "num_input_tokens_seen": 35421472,
      "step": 167840
    },
    {
      "epoch": 18.464796479647966,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005353156275864996,
      "loss": 0.2324,
      "num_input_tokens_seen": 35422528,
      "step": 167845
    },
    {
      "epoch": 18.465346534653467,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0005349344183516191,
      "loss": 0.2314,
      "num_input_tokens_seen": 35423584,
      "step": 167850
    },
    {
      "epoch": 18.465896589658964,
      "grad_norm": 0.0008392333984375,
      "learning_rate": 0.0005345533424343074,
      "loss": 0.2309,
      "num_input_tokens_seen": 35424704,
      "step": 167855
    },
    {
      "epoch": 18.466446644664465,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0005341723998380765,
      "loss": 0.2309,
      "num_input_tokens_seen": 35425760,
      "step": 167860
    },
    {
      "epoch": 18.466996699669966,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0005337915905664353,
      "loss": 0.2303,
      "num_input_tokens_seen": 35426816,
      "step": 167865
    },
    {
      "epoch": 18.467546754675467,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.0005334109146228977,
      "loss": 0.233,
      "num_input_tokens_seen": 35427872,
      "step": 167870
    },
    {
      "epoch": 18.46809680968097,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0005330303720109691,
      "loss": 0.2298,
      "num_input_tokens_seen": 35428928,
      "step": 167875
    },
    {
      "epoch": 18.46864686468647,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.0005326499627341585,
      "loss": 0.2309,
      "num_input_tokens_seen": 35429984,
      "step": 167880
    },
    {
      "epoch": 18.46919691969197,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.0005322696867959697,
      "loss": 0.2308,
      "num_input_tokens_seen": 35431072,
      "step": 167885
    },
    {
      "epoch": 18.46974697469747,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.00053188954419991,
      "loss": 0.2324,
      "num_input_tokens_seen": 35432160,
      "step": 167890
    },
    {
      "epoch": 18.47029702970297,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0005315095349494814,
      "loss": 0.2303,
      "num_input_tokens_seen": 35433152,
      "step": 167895
    },
    {
      "epoch": 18.47084708470847,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0005311296590481845,
      "loss": 0.2335,
      "num_input_tokens_seen": 35434272,
      "step": 167900
    },
    {
      "epoch": 18.47139713971397,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005307499164995266,
      "loss": 0.2329,
      "num_input_tokens_seen": 35435296,
      "step": 167905
    },
    {
      "epoch": 18.471947194719473,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0005303703073070015,
      "loss": 0.2319,
      "num_input_tokens_seen": 35436288,
      "step": 167910
    },
    {
      "epoch": 18.472497249724974,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0005299908314741114,
      "loss": 0.2298,
      "num_input_tokens_seen": 35437344,
      "step": 167915
    },
    {
      "epoch": 18.47304730473047,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0005296114890043502,
      "loss": 0.2304,
      "num_input_tokens_seen": 35438432,
      "step": 167920
    },
    {
      "epoch": 18.473597359735972,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0005292322799012183,
      "loss": 0.2309,
      "num_input_tokens_seen": 35439456,
      "step": 167925
    },
    {
      "epoch": 18.474147414741473,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0005288532041682081,
      "loss": 0.2335,
      "num_input_tokens_seen": 35440512,
      "step": 167930
    },
    {
      "epoch": 18.474697469746975,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.0005284742618088151,
      "loss": 0.2308,
      "num_input_tokens_seen": 35441536,
      "step": 167935
    },
    {
      "epoch": 18.475247524752476,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0005280954528265297,
      "loss": 0.2314,
      "num_input_tokens_seen": 35442624,
      "step": 167940
    },
    {
      "epoch": 18.475797579757977,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.0005277167772248442,
      "loss": 0.2313,
      "num_input_tokens_seen": 35443616,
      "step": 167945
    },
    {
      "epoch": 18.476347634763478,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.0005273382350072459,
      "loss": 0.2324,
      "num_input_tokens_seen": 35444640,
      "step": 167950
    },
    {
      "epoch": 18.476897689768975,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0005269598261772318,
      "loss": 0.2303,
      "num_input_tokens_seen": 35445728,
      "step": 167955
    },
    {
      "epoch": 18.477447744774476,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.000526581550738281,
      "loss": 0.2319,
      "num_input_tokens_seen": 35446848,
      "step": 167960
    },
    {
      "epoch": 18.477997799779978,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0005262034086938838,
      "loss": 0.2308,
      "num_input_tokens_seen": 35447904,
      "step": 167965
    },
    {
      "epoch": 18.47854785478548,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0005258254000475244,
      "loss": 0.2293,
      "num_input_tokens_seen": 35448992,
      "step": 167970
    },
    {
      "epoch": 18.47909790979098,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0005254475248026863,
      "loss": 0.2308,
      "num_input_tokens_seen": 35449984,
      "step": 167975
    },
    {
      "epoch": 18.47964796479648,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0005250697829628537,
      "loss": 0.2314,
      "num_input_tokens_seen": 35451072,
      "step": 167980
    },
    {
      "epoch": 18.480198019801982,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.000524692174531507,
      "loss": 0.2314,
      "num_input_tokens_seen": 35452096,
      "step": 167985
    },
    {
      "epoch": 18.48074807480748,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.000524314699512125,
      "loss": 0.2308,
      "num_input_tokens_seen": 35453120,
      "step": 167990
    },
    {
      "epoch": 18.48129812981298,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 0.0005239373579081918,
      "loss": 0.2314,
      "num_input_tokens_seen": 35454176,
      "step": 167995
    },
    {
      "epoch": 18.48184818481848,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0005235601497231778,
      "loss": 0.2329,
      "num_input_tokens_seen": 35455264,
      "step": 168000
    },
    {
      "epoch": 18.482398239823983,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.000523183074960567,
      "loss": 0.2303,
      "num_input_tokens_seen": 35456320,
      "step": 168005
    },
    {
      "epoch": 18.482948294829484,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0005228061336238282,
      "loss": 0.2293,
      "num_input_tokens_seen": 35457376,
      "step": 168010
    },
    {
      "epoch": 18.483498349834985,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0005224293257164387,
      "loss": 0.2309,
      "num_input_tokens_seen": 35458336,
      "step": 168015
    },
    {
      "epoch": 18.484048404840483,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.0005220526512418739,
      "loss": 0.2309,
      "num_input_tokens_seen": 35459424,
      "step": 168020
    },
    {
      "epoch": 18.484598459845984,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0005216761102035994,
      "loss": 0.2298,
      "num_input_tokens_seen": 35460416,
      "step": 168025
    },
    {
      "epoch": 18.485148514851485,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0005212997026050908,
      "loss": 0.2314,
      "num_input_tokens_seen": 35461440,
      "step": 168030
    },
    {
      "epoch": 18.485698569856986,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0005209234284498138,
      "loss": 0.2308,
      "num_input_tokens_seen": 35462400,
      "step": 168035
    },
    {
      "epoch": 18.486248624862487,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0005205472877412387,
      "loss": 0.2324,
      "num_input_tokens_seen": 35463520,
      "step": 168040
    },
    {
      "epoch": 18.486798679867988,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.0005201712804828329,
      "loss": 0.2319,
      "num_input_tokens_seen": 35464544,
      "step": 168045
    },
    {
      "epoch": 18.48734873487349,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0005197954066780569,
      "loss": 0.2308,
      "num_input_tokens_seen": 35465568,
      "step": 168050
    },
    {
      "epoch": 18.487898789878987,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0005194196663303813,
      "loss": 0.2303,
      "num_input_tokens_seen": 35466656,
      "step": 168055
    },
    {
      "epoch": 18.488448844884488,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0005190440594432666,
      "loss": 0.2309,
      "num_input_tokens_seen": 35467680,
      "step": 168060
    },
    {
      "epoch": 18.48899889988999,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0005186685860201717,
      "loss": 0.2293,
      "num_input_tokens_seen": 35468768,
      "step": 168065
    },
    {
      "epoch": 18.48954895489549,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0005182932460645589,
      "loss": 0.2309,
      "num_input_tokens_seen": 35469760,
      "step": 168070
    },
    {
      "epoch": 18.49009900990099,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0005179180395798905,
      "loss": 0.2314,
      "num_input_tokens_seen": 35470880,
      "step": 168075
    },
    {
      "epoch": 18.490649064906492,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0005175429665696218,
      "loss": 0.2308,
      "num_input_tokens_seen": 35471968,
      "step": 168080
    },
    {
      "epoch": 18.49119911991199,
      "grad_norm": 0.005859375,
      "learning_rate": 0.000517168027037212,
      "loss": 0.2298,
      "num_input_tokens_seen": 35473024,
      "step": 168085
    },
    {
      "epoch": 18.49174917491749,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0005167932209861114,
      "loss": 0.2319,
      "num_input_tokens_seen": 35474080,
      "step": 168090
    },
    {
      "epoch": 18.492299229922992,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0005164185484197791,
      "loss": 0.2308,
      "num_input_tokens_seen": 35475072,
      "step": 168095
    },
    {
      "epoch": 18.492849284928493,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.0005160440093416674,
      "loss": 0.2319,
      "num_input_tokens_seen": 35476160,
      "step": 168100
    },
    {
      "epoch": 18.493399339933994,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0005156696037552283,
      "loss": 0.2314,
      "num_input_tokens_seen": 35477152,
      "step": 168105
    },
    {
      "epoch": 18.493949394939495,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0005152953316639141,
      "loss": 0.2314,
      "num_input_tokens_seen": 35478272,
      "step": 168110
    },
    {
      "epoch": 18.494499449944996,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.0005149211930711689,
      "loss": 0.2293,
      "num_input_tokens_seen": 35479296,
      "step": 168115
    },
    {
      "epoch": 18.495049504950494,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.000514547187980448,
      "loss": 0.2309,
      "num_input_tokens_seen": 35480320,
      "step": 168120
    },
    {
      "epoch": 18.495599559955995,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005141733163951922,
      "loss": 0.2314,
      "num_input_tokens_seen": 35481408,
      "step": 168125
    },
    {
      "epoch": 18.496149614961496,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0005137995783188487,
      "loss": 0.234,
      "num_input_tokens_seen": 35482432,
      "step": 168130
    },
    {
      "epoch": 18.496699669966997,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0005134259737548663,
      "loss": 0.2314,
      "num_input_tokens_seen": 35483520,
      "step": 168135
    },
    {
      "epoch": 18.497249724972498,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.000513052502706684,
      "loss": 0.2309,
      "num_input_tokens_seen": 35484576,
      "step": 168140
    },
    {
      "epoch": 18.497799779978,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.0005126791651777474,
      "loss": 0.2319,
      "num_input_tokens_seen": 35485600,
      "step": 168145
    },
    {
      "epoch": 18.498349834983497,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0005123059611714919,
      "loss": 0.2303,
      "num_input_tokens_seen": 35486656,
      "step": 168150
    },
    {
      "epoch": 18.498899889988998,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0005119328906913617,
      "loss": 0.2308,
      "num_input_tokens_seen": 35487712,
      "step": 168155
    },
    {
      "epoch": 18.4994499449945,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0005115599537407955,
      "loss": 0.2309,
      "num_input_tokens_seen": 35488736,
      "step": 168160
    },
    {
      "epoch": 18.5,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0005111871503232274,
      "loss": 0.2309,
      "num_input_tokens_seen": 35489792,
      "step": 168165
    },
    {
      "epoch": 18.5005500550055,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0005108144804420961,
      "loss": 0.2324,
      "num_input_tokens_seen": 35490880,
      "step": 168170
    },
    {
      "epoch": 18.501100110011002,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0005104419441008356,
      "loss": 0.2319,
      "num_input_tokens_seen": 35492064,
      "step": 168175
    },
    {
      "epoch": 18.501650165016503,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0005100695413028766,
      "loss": 0.2308,
      "num_input_tokens_seen": 35493056,
      "step": 168180
    },
    {
      "epoch": 18.502200220022,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.0005096972720516529,
      "loss": 0.2324,
      "num_input_tokens_seen": 35494112,
      "step": 168185
    },
    {
      "epoch": 18.502750275027502,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0005093251363505985,
      "loss": 0.2314,
      "num_input_tokens_seen": 35495168,
      "step": 168190
    },
    {
      "epoch": 18.503300330033003,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.0005089531342031422,
      "loss": 0.2314,
      "num_input_tokens_seen": 35496224,
      "step": 168195
    },
    {
      "epoch": 18.503850385038504,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0005085812656127098,
      "loss": 0.2314,
      "num_input_tokens_seen": 35497280,
      "step": 168200
    },
    {
      "epoch": 18.504400440044005,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.00050820953058273,
      "loss": 0.2309,
      "num_input_tokens_seen": 35498368,
      "step": 168205
    },
    {
      "epoch": 18.504950495049506,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.0005078379291166285,
      "loss": 0.2303,
      "num_input_tokens_seen": 35499456,
      "step": 168210
    },
    {
      "epoch": 18.505500550055004,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.0005074664612178309,
      "loss": 0.2308,
      "num_input_tokens_seen": 35500512,
      "step": 168215
    },
    {
      "epoch": 18.506050605060505,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.0005070951268897627,
      "loss": 0.2303,
      "num_input_tokens_seen": 35501536,
      "step": 168220
    },
    {
      "epoch": 18.506600660066006,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005067239261358447,
      "loss": 0.2324,
      "num_input_tokens_seen": 35502560,
      "step": 168225
    },
    {
      "epoch": 18.507150715071507,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0005063528589594956,
      "loss": 0.2309,
      "num_input_tokens_seen": 35503552,
      "step": 168230
    },
    {
      "epoch": 18.507700770077008,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0005059819253641396,
      "loss": 0.2324,
      "num_input_tokens_seen": 35504576,
      "step": 168235
    },
    {
      "epoch": 18.50825082508251,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0005056111253531903,
      "loss": 0.2303,
      "num_input_tokens_seen": 35505632,
      "step": 168240
    },
    {
      "epoch": 18.50880088008801,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005052404589300685,
      "loss": 0.2303,
      "num_input_tokens_seen": 35506656,
      "step": 168245
    },
    {
      "epoch": 18.509350935093508,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0005048699260981932,
      "loss": 0.2303,
      "num_input_tokens_seen": 35507680,
      "step": 168250
    },
    {
      "epoch": 18.50990099009901,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.0005044995268609731,
      "loss": 0.2314,
      "num_input_tokens_seen": 35508800,
      "step": 168255
    },
    {
      "epoch": 18.51045104510451,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0005041292612218273,
      "loss": 0.2314,
      "num_input_tokens_seen": 35509760,
      "step": 168260
    },
    {
      "epoch": 18.51100110011001,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0005037591291841664,
      "loss": 0.2308,
      "num_input_tokens_seen": 35510848,
      "step": 168265
    },
    {
      "epoch": 18.511551155115512,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.0005033891307513993,
      "loss": 0.2319,
      "num_input_tokens_seen": 35512032,
      "step": 168270
    },
    {
      "epoch": 18.512101210121013,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.0005030192659269417,
      "loss": 0.2303,
      "num_input_tokens_seen": 35513120,
      "step": 168275
    },
    {
      "epoch": 18.51265126512651,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0005026495347141956,
      "loss": 0.2319,
      "num_input_tokens_seen": 35514112,
      "step": 168280
    },
    {
      "epoch": 18.513201320132012,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 0.0005022799371165753,
      "loss": 0.2293,
      "num_input_tokens_seen": 35515136,
      "step": 168285
    },
    {
      "epoch": 18.513751375137513,
      "grad_norm": 0.01141357421875,
      "learning_rate": 0.0005019104731374829,
      "loss": 0.2313,
      "num_input_tokens_seen": 35516192,
      "step": 168290
    },
    {
      "epoch": 18.514301430143014,
      "grad_norm": 0.00048828125,
      "learning_rate": 0.0005015411427803256,
      "loss": 0.2303,
      "num_input_tokens_seen": 35517216,
      "step": 168295
    },
    {
      "epoch": 18.514851485148515,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.0005011719460485042,
      "loss": 0.2303,
      "num_input_tokens_seen": 35518304,
      "step": 168300
    },
    {
      "epoch": 18.515401540154016,
      "grad_norm": 0.00087738037109375,
      "learning_rate": 0.0005008028829454258,
      "loss": 0.2313,
      "num_input_tokens_seen": 35519360,
      "step": 168305
    },
    {
      "epoch": 18.515951595159517,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0005004339534744895,
      "loss": 0.2329,
      "num_input_tokens_seen": 35520416,
      "step": 168310
    },
    {
      "epoch": 18.516501650165015,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0005000651576390974,
      "loss": 0.2303,
      "num_input_tokens_seen": 35521472,
      "step": 168315
    },
    {
      "epoch": 18.517051705170516,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 0.0004996964954426452,
      "loss": 0.2303,
      "num_input_tokens_seen": 35522560,
      "step": 168320
    },
    {
      "epoch": 18.517601760176017,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.000499327966888532,
      "loss": 0.2309,
      "num_input_tokens_seen": 35523584,
      "step": 168325
    },
    {
      "epoch": 18.51815181518152,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0004989595719801549,
      "loss": 0.2309,
      "num_input_tokens_seen": 35524608,
      "step": 168330
    },
    {
      "epoch": 18.51870187018702,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0004985913107209094,
      "loss": 0.2314,
      "num_input_tokens_seen": 35525696,
      "step": 168335
    },
    {
      "epoch": 18.51925192519252,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0004982231831141915,
      "loss": 0.2304,
      "num_input_tokens_seen": 35526752,
      "step": 168340
    },
    {
      "epoch": 18.519801980198018,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.0004978551891633898,
      "loss": 0.2324,
      "num_input_tokens_seen": 35527776,
      "step": 168345
    },
    {
      "epoch": 18.52035203520352,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.0004974873288718984,
      "loss": 0.2293,
      "num_input_tokens_seen": 35528864,
      "step": 168350
    },
    {
      "epoch": 18.52090209020902,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0004971196022431063,
      "loss": 0.2298,
      "num_input_tokens_seen": 35529856,
      "step": 168355
    },
    {
      "epoch": 18.52145214521452,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.000496752009280404,
      "loss": 0.2308,
      "num_input_tokens_seen": 35530944,
      "step": 168360
    },
    {
      "epoch": 18.522002200220022,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0004963845499871805,
      "loss": 0.2324,
      "num_input_tokens_seen": 35532064,
      "step": 168365
    },
    {
      "epoch": 18.522552255225524,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0004960172243668198,
      "loss": 0.2319,
      "num_input_tokens_seen": 35533120,
      "step": 168370
    },
    {
      "epoch": 18.523102310231025,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0004956500324227108,
      "loss": 0.2309,
      "num_input_tokens_seen": 35534144,
      "step": 168375
    },
    {
      "epoch": 18.523652365236522,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0004952829741582326,
      "loss": 0.2319,
      "num_input_tokens_seen": 35535168,
      "step": 168380
    },
    {
      "epoch": 18.524202420242023,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0004949160495767724,
      "loss": 0.2308,
      "num_input_tokens_seen": 35536224,
      "step": 168385
    },
    {
      "epoch": 18.524752475247524,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0004945492586817108,
      "loss": 0.2303,
      "num_input_tokens_seen": 35537312,
      "step": 168390
    },
    {
      "epoch": 18.525302530253025,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0004941826014764284,
      "loss": 0.2329,
      "num_input_tokens_seen": 35538368,
      "step": 168395
    },
    {
      "epoch": 18.525852585258527,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.0004938160779643042,
      "loss": 0.2309,
      "num_input_tokens_seen": 35539456,
      "step": 168400
    },
    {
      "epoch": 18.526402640264028,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0004934496881487171,
      "loss": 0.2309,
      "num_input_tokens_seen": 35540480,
      "step": 168405
    },
    {
      "epoch": 18.52695269526953,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.000493083432033043,
      "loss": 0.233,
      "num_input_tokens_seen": 35541568,
      "step": 168410
    },
    {
      "epoch": 18.527502750275026,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.0004927173096206572,
      "loss": 0.2314,
      "num_input_tokens_seen": 35542656,
      "step": 168415
    },
    {
      "epoch": 18.528052805280527,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0004923513209149338,
      "loss": 0.2293,
      "num_input_tokens_seen": 35543680,
      "step": 168420
    },
    {
      "epoch": 18.52860286028603,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.0004919854659192501,
      "loss": 0.2308,
      "num_input_tokens_seen": 35544800,
      "step": 168425
    },
    {
      "epoch": 18.52915291529153,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0004916197446369719,
      "loss": 0.2303,
      "num_input_tokens_seen": 35545888,
      "step": 168430
    },
    {
      "epoch": 18.52970297029703,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0004912541570714745,
      "loss": 0.2329,
      "num_input_tokens_seen": 35546912,
      "step": 168435
    },
    {
      "epoch": 18.53025302530253,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.0004908887032261239,
      "loss": 0.2329,
      "num_input_tokens_seen": 35547968,
      "step": 168440
    },
    {
      "epoch": 18.53080308030803,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0004905233831042887,
      "loss": 0.2298,
      "num_input_tokens_seen": 35549056,
      "step": 168445
    },
    {
      "epoch": 18.53135313531353,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0004901581967093416,
      "loss": 0.2298,
      "num_input_tokens_seen": 35550144,
      "step": 168450
    },
    {
      "epoch": 18.53190319031903,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0004897931440446396,
      "loss": 0.2303,
      "num_input_tokens_seen": 35551232,
      "step": 168455
    },
    {
      "epoch": 18.532453245324533,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.000489428225113555,
      "loss": 0.2309,
      "num_input_tokens_seen": 35552320,
      "step": 168460
    },
    {
      "epoch": 18.533003300330034,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.0004890634399194471,
      "loss": 0.2314,
      "num_input_tokens_seen": 35553344,
      "step": 168465
    },
    {
      "epoch": 18.533553355335535,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0004886987884656763,
      "loss": 0.2319,
      "num_input_tokens_seen": 35554336,
      "step": 168470
    },
    {
      "epoch": 18.534103410341036,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.0004883342707556065,
      "loss": 0.2319,
      "num_input_tokens_seen": 35555328,
      "step": 168475
    },
    {
      "epoch": 18.534653465346533,
      "grad_norm": 0.005218505859375,
      "learning_rate": 0.00048796988679259524,
      "loss": 0.2314,
      "num_input_tokens_seen": 35556320,
      "step": 168480
    },
    {
      "epoch": 18.535203520352034,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.00048760563658000464,
      "loss": 0.2303,
      "num_input_tokens_seen": 35557376,
      "step": 168485
    },
    {
      "epoch": 18.535753575357536,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0004872415201211888,
      "loss": 0.2298,
      "num_input_tokens_seen": 35558400,
      "step": 168490
    },
    {
      "epoch": 18.536303630363037,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0004868775374195017,
      "loss": 0.2324,
      "num_input_tokens_seen": 35559424,
      "step": 168495
    },
    {
      "epoch": 18.536853685368538,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00048651368847830057,
      "loss": 0.2319,
      "num_input_tokens_seen": 35560512,
      "step": 168500
    },
    {
      "epoch": 18.53740374037404,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0004861499733009378,
      "loss": 0.2298,
      "num_input_tokens_seen": 35561568,
      "step": 168505
    },
    {
      "epoch": 18.537953795379536,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.00048578639189076895,
      "loss": 0.2298,
      "num_input_tokens_seen": 35562656,
      "step": 168510
    },
    {
      "epoch": 18.538503850385037,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.00048542294425114315,
      "loss": 0.2314,
      "num_input_tokens_seen": 35563712,
      "step": 168515
    },
    {
      "epoch": 18.53905390539054,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.00048505963038540596,
      "loss": 0.2308,
      "num_input_tokens_seen": 35564704,
      "step": 168520
    },
    {
      "epoch": 18.53960396039604,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.0004846964502969114,
      "loss": 0.2314,
      "num_input_tokens_seen": 35565728,
      "step": 168525
    },
    {
      "epoch": 18.54015401540154,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0004843334039890035,
      "loss": 0.2324,
      "num_input_tokens_seen": 35566784,
      "step": 168530
    },
    {
      "epoch": 18.540704070407042,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00048397049146502944,
      "loss": 0.2298,
      "num_input_tokens_seen": 35567808,
      "step": 168535
    },
    {
      "epoch": 18.541254125412543,
      "grad_norm": 0.001220703125,
      "learning_rate": 0.000483607712728335,
      "loss": 0.2309,
      "num_input_tokens_seen": 35568832,
      "step": 168540
    },
    {
      "epoch": 18.54180418041804,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0004832450677822608,
      "loss": 0.2314,
      "num_input_tokens_seen": 35569888,
      "step": 168545
    },
    {
      "epoch": 18.54235423542354,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0004828825566301525,
      "loss": 0.2319,
      "num_input_tokens_seen": 35570944,
      "step": 168550
    },
    {
      "epoch": 18.542904290429043,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.0004825201792753508,
      "loss": 0.2303,
      "num_input_tokens_seen": 35572000,
      "step": 168555
    },
    {
      "epoch": 18.543454345434544,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0004821579357211897,
      "loss": 0.2319,
      "num_input_tokens_seen": 35573120,
      "step": 168560
    },
    {
      "epoch": 18.544004400440045,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.00048179582597101644,
      "loss": 0.2303,
      "num_input_tokens_seen": 35574176,
      "step": 168565
    },
    {
      "epoch": 18.544554455445546,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0004814338500281634,
      "loss": 0.2324,
      "num_input_tokens_seen": 35575232,
      "step": 168570
    },
    {
      "epoch": 18.545104510451043,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0004810720078959696,
      "loss": 0.2314,
      "num_input_tokens_seen": 35576288,
      "step": 168575
    },
    {
      "epoch": 18.545654565456545,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00048071029957776735,
      "loss": 0.2319,
      "num_input_tokens_seen": 35577312,
      "step": 168580
    },
    {
      "epoch": 18.546204620462046,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.00048034872507689063,
      "loss": 0.2314,
      "num_input_tokens_seen": 35578336,
      "step": 168585
    },
    {
      "epoch": 18.546754675467547,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0004799872843966718,
      "loss": 0.2329,
      "num_input_tokens_seen": 35579360,
      "step": 168590
    },
    {
      "epoch": 18.547304730473048,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0004796259775404432,
      "loss": 0.2324,
      "num_input_tokens_seen": 35580448,
      "step": 168595
    },
    {
      "epoch": 18.54785478547855,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.0004792648045115355,
      "loss": 0.2324,
      "num_input_tokens_seen": 35581472,
      "step": 168600
    },
    {
      "epoch": 18.54840484048405,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00047890376531327606,
      "loss": 0.2309,
      "num_input_tokens_seen": 35582592,
      "step": 168605
    },
    {
      "epoch": 18.548954895489548,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.00047854285994899216,
      "loss": 0.2309,
      "num_input_tokens_seen": 35583712,
      "step": 168610
    },
    {
      "epoch": 18.54950495049505,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00047818208842201115,
      "loss": 0.2303,
      "num_input_tokens_seen": 35584800,
      "step": 168615
    },
    {
      "epoch": 18.55005500550055,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.0004778214507356554,
      "loss": 0.2319,
      "num_input_tokens_seen": 35585824,
      "step": 168620
    },
    {
      "epoch": 18.55060506050605,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0004774609468932539,
      "loss": 0.2319,
      "num_input_tokens_seen": 35586944,
      "step": 168625
    },
    {
      "epoch": 18.551155115511552,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0004771005768981257,
      "loss": 0.2314,
      "num_input_tokens_seen": 35587968,
      "step": 168630
    },
    {
      "epoch": 18.551705170517053,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0004767403407535914,
      "loss": 0.2303,
      "num_input_tokens_seen": 35589024,
      "step": 168635
    },
    {
      "epoch": 18.55225522552255,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00047638023846297505,
      "loss": 0.2303,
      "num_input_tokens_seen": 35590080,
      "step": 168640
    },
    {
      "epoch": 18.55280528052805,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.000476020270029589,
      "loss": 0.2308,
      "num_input_tokens_seen": 35591104,
      "step": 168645
    },
    {
      "epoch": 18.553355335533553,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.0004756604354567573,
      "loss": 0.2308,
      "num_input_tokens_seen": 35592160,
      "step": 168650
    },
    {
      "epoch": 18.553905390539054,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00047530073474779386,
      "loss": 0.2314,
      "num_input_tokens_seen": 35593184,
      "step": 168655
    },
    {
      "epoch": 18.554455445544555,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0004749411679060128,
      "loss": 0.2308,
      "num_input_tokens_seen": 35594208,
      "step": 168660
    },
    {
      "epoch": 18.555005500550056,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.0004745817349347314,
      "loss": 0.2319,
      "num_input_tokens_seen": 35595328,
      "step": 168665
    },
    {
      "epoch": 18.555555555555557,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0004742224358372571,
      "loss": 0.2314,
      "num_input_tokens_seen": 35596416,
      "step": 168670
    },
    {
      "epoch": 18.556105610561055,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0004738632706169071,
      "loss": 0.2319,
      "num_input_tokens_seen": 35597536,
      "step": 168675
    },
    {
      "epoch": 18.556655665566556,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.00047350423927698893,
      "loss": 0.2283,
      "num_input_tokens_seen": 35598624,
      "step": 168680
    },
    {
      "epoch": 18.557205720572057,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0004731453418208098,
      "loss": 0.2309,
      "num_input_tokens_seen": 35599616,
      "step": 168685
    },
    {
      "epoch": 18.557755775577558,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0004727865782516805,
      "loss": 0.2329,
      "num_input_tokens_seen": 35600640,
      "step": 168690
    },
    {
      "epoch": 18.55830583058306,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00047242794857290656,
      "loss": 0.2335,
      "num_input_tokens_seen": 35601696,
      "step": 168695
    },
    {
      "epoch": 18.55885588558856,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0004720694527877939,
      "loss": 0.2324,
      "num_input_tokens_seen": 35602752,
      "step": 168700
    },
    {
      "epoch": 18.55940594059406,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.000471711090899643,
      "loss": 0.2303,
      "num_input_tokens_seen": 35603776,
      "step": 168705
    },
    {
      "epoch": 18.55995599559956,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00047135286291176133,
      "loss": 0.2319,
      "num_input_tokens_seen": 35604864,
      "step": 168710
    },
    {
      "epoch": 18.56050605060506,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.00047099476882745127,
      "loss": 0.2308,
      "num_input_tokens_seen": 35605920,
      "step": 168715
    },
    {
      "epoch": 18.56105610561056,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.0004706368086500084,
      "loss": 0.2319,
      "num_input_tokens_seen": 35606976,
      "step": 168720
    },
    {
      "epoch": 18.561606160616062,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.00047027898238273356,
      "loss": 0.2308,
      "num_input_tokens_seen": 35608032,
      "step": 168725
    },
    {
      "epoch": 18.562156215621563,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.00046992129002892735,
      "loss": 0.2319,
      "num_input_tokens_seen": 35609088,
      "step": 168730
    },
    {
      "epoch": 18.562706270627064,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0004695637315918821,
      "loss": 0.2314,
      "num_input_tokens_seen": 35610176,
      "step": 168735
    },
    {
      "epoch": 18.563256325632562,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.00046920630707489693,
      "loss": 0.2324,
      "num_input_tokens_seen": 35611264,
      "step": 168740
    },
    {
      "epoch": 18.563806380638063,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0004688490164812642,
      "loss": 0.2314,
      "num_input_tokens_seen": 35612320,
      "step": 168745
    },
    {
      "epoch": 18.564356435643564,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00046849185981427785,
      "loss": 0.2314,
      "num_input_tokens_seen": 35613408,
      "step": 168750
    },
    {
      "epoch": 18.564906490649065,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.0004681348370772287,
      "loss": 0.2298,
      "num_input_tokens_seen": 35614496,
      "step": 168755
    },
    {
      "epoch": 18.565456545654566,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00046777794827340734,
      "loss": 0.2314,
      "num_input_tokens_seen": 35615552,
      "step": 168760
    },
    {
      "epoch": 18.566006600660067,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00046742119340610454,
      "loss": 0.2324,
      "num_input_tokens_seen": 35616608,
      "step": 168765
    },
    {
      "epoch": 18.566556655665565,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00046706457247860597,
      "loss": 0.2324,
      "num_input_tokens_seen": 35617664,
      "step": 168770
    },
    {
      "epoch": 18.567106710671066,
      "grad_norm": 0.0004062652587890625,
      "learning_rate": 0.0004667080854942007,
      "loss": 0.2314,
      "num_input_tokens_seen": 35618656,
      "step": 168775
    },
    {
      "epoch": 18.567656765676567,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00046635173245617275,
      "loss": 0.2314,
      "num_input_tokens_seen": 35619744,
      "step": 168780
    },
    {
      "epoch": 18.568206820682068,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.0004659955133678062,
      "loss": 0.2308,
      "num_input_tokens_seen": 35620800,
      "step": 168785
    },
    {
      "epoch": 18.56875687568757,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.0004656394282323867,
      "loss": 0.2303,
      "num_input_tokens_seen": 35621856,
      "step": 168790
    },
    {
      "epoch": 18.56930693069307,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00046528347705319326,
      "loss": 0.2314,
      "num_input_tokens_seen": 35622912,
      "step": 168795
    },
    {
      "epoch": 18.56985698569857,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0004649276598335067,
      "loss": 0.2314,
      "num_input_tokens_seen": 35624000,
      "step": 168800
    },
    {
      "epoch": 18.57040704070407,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0004645719765766093,
      "loss": 0.234,
      "num_input_tokens_seen": 35625024,
      "step": 168805
    },
    {
      "epoch": 18.57095709570957,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 0.00046421642728577516,
      "loss": 0.2335,
      "num_input_tokens_seen": 35626112,
      "step": 168810
    },
    {
      "epoch": 18.57150715071507,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.00046386101196428496,
      "loss": 0.2324,
      "num_input_tokens_seen": 35627200,
      "step": 168815
    },
    {
      "epoch": 18.572057205720572,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0004635057306154111,
      "loss": 0.2309,
      "num_input_tokens_seen": 35628256,
      "step": 168820
    },
    {
      "epoch": 18.572607260726073,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.000463150583242431,
      "loss": 0.2308,
      "num_input_tokens_seen": 35629312,
      "step": 168825
    },
    {
      "epoch": 18.573157315731574,
      "grad_norm": 0.00054931640625,
      "learning_rate": 0.0004627955698486169,
      "loss": 0.2303,
      "num_input_tokens_seen": 35630368,
      "step": 168830
    },
    {
      "epoch": 18.573707370737075,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.000462440690437238,
      "loss": 0.2303,
      "num_input_tokens_seen": 35631392,
      "step": 168835
    },
    {
      "epoch": 18.574257425742573,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.00046208594501156997,
      "loss": 0.2324,
      "num_input_tokens_seen": 35632480,
      "step": 168840
    },
    {
      "epoch": 18.574807480748074,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0004617313335748801,
      "loss": 0.2319,
      "num_input_tokens_seen": 35633568,
      "step": 168845
    },
    {
      "epoch": 18.575357535753575,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00046137685613043434,
      "loss": 0.2324,
      "num_input_tokens_seen": 35634592,
      "step": 168850
    },
    {
      "epoch": 18.575907590759076,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0004610225126815015,
      "loss": 0.2324,
      "num_input_tokens_seen": 35635680,
      "step": 168855
    },
    {
      "epoch": 18.576457645764577,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00046066830323134744,
      "loss": 0.2298,
      "num_input_tokens_seen": 35636768,
      "step": 168860
    },
    {
      "epoch": 18.57700770077008,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00046031422778323945,
      "loss": 0.2314,
      "num_input_tokens_seen": 35637824,
      "step": 168865
    },
    {
      "epoch": 18.577557755775576,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0004599602863404384,
      "loss": 0.2309,
      "num_input_tokens_seen": 35638880,
      "step": 168870
    },
    {
      "epoch": 18.578107810781077,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.00045960647890620483,
      "loss": 0.2303,
      "num_input_tokens_seen": 35639936,
      "step": 168875
    },
    {
      "epoch": 18.578657865786578,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.0004592528054838013,
      "loss": 0.2303,
      "num_input_tokens_seen": 35640960,
      "step": 168880
    },
    {
      "epoch": 18.57920792079208,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 0.00045889926607648677,
      "loss": 0.2313,
      "num_input_tokens_seen": 35641952,
      "step": 168885
    },
    {
      "epoch": 18.57975797579758,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.000458545860687522,
      "loss": 0.2308,
      "num_input_tokens_seen": 35643008,
      "step": 168890
    },
    {
      "epoch": 18.58030803080308,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0004581925893201627,
      "loss": 0.2314,
      "num_input_tokens_seen": 35644032,
      "step": 168895
    },
    {
      "epoch": 18.580858085808583,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.0004578394519776613,
      "loss": 0.2314,
      "num_input_tokens_seen": 35645088,
      "step": 168900
    },
    {
      "epoch": 18.58140814081408,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.00045748644866327846,
      "loss": 0.2319,
      "num_input_tokens_seen": 35646080,
      "step": 168905
    },
    {
      "epoch": 18.58195819581958,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0004571335793802633,
      "loss": 0.2319,
      "num_input_tokens_seen": 35647136,
      "step": 168910
    },
    {
      "epoch": 18.582508250825082,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00045678084413186993,
      "loss": 0.2308,
      "num_input_tokens_seen": 35648224,
      "step": 168915
    },
    {
      "epoch": 18.583058305830583,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.0004564282429213506,
      "loss": 0.2304,
      "num_input_tokens_seen": 35649280,
      "step": 168920
    },
    {
      "epoch": 18.583608360836084,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.00045607577575195124,
      "loss": 0.2324,
      "num_input_tokens_seen": 35650368,
      "step": 168925
    },
    {
      "epoch": 18.584158415841586,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.0004557234426269241,
      "loss": 0.2314,
      "num_input_tokens_seen": 35651392,
      "step": 168930
    },
    {
      "epoch": 18.584708470847083,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.0004553712435495133,
      "loss": 0.2288,
      "num_input_tokens_seen": 35652448,
      "step": 168935
    },
    {
      "epoch": 18.585258525852584,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.0004550191785229679,
      "loss": 0.2324,
      "num_input_tokens_seen": 35653408,
      "step": 168940
    },
    {
      "epoch": 18.585808580858085,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.000454667247550532,
      "loss": 0.2329,
      "num_input_tokens_seen": 35654432,
      "step": 168945
    },
    {
      "epoch": 18.586358635863586,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0004543154506354463,
      "loss": 0.2319,
      "num_input_tokens_seen": 35655456,
      "step": 168950
    },
    {
      "epoch": 18.586908690869087,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00045396378778095824,
      "loss": 0.2308,
      "num_input_tokens_seen": 35656608,
      "step": 168955
    },
    {
      "epoch": 18.58745874587459,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0004536122589903052,
      "loss": 0.2319,
      "num_input_tokens_seen": 35657696,
      "step": 168960
    },
    {
      "epoch": 18.58800880088009,
      "grad_norm": 0.000469207763671875,
      "learning_rate": 0.00045326086426672625,
      "loss": 0.2319,
      "num_input_tokens_seen": 35658816,
      "step": 168965
    },
    {
      "epoch": 18.588558855885587,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.0004529096036134622,
      "loss": 0.2298,
      "num_input_tokens_seen": 35659840,
      "step": 168970
    },
    {
      "epoch": 18.58910891089109,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0004525584770337504,
      "loss": 0.2309,
      "num_input_tokens_seen": 35660864,
      "step": 168975
    },
    {
      "epoch": 18.58965896589659,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00045220748453082826,
      "loss": 0.2319,
      "num_input_tokens_seen": 35661888,
      "step": 168980
    },
    {
      "epoch": 18.59020902090209,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0004518566261079282,
      "loss": 0.2303,
      "num_input_tokens_seen": 35662912,
      "step": 168985
    },
    {
      "epoch": 18.59075907590759,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00045150590176828265,
      "loss": 0.2298,
      "num_input_tokens_seen": 35664032,
      "step": 168990
    },
    {
      "epoch": 18.591309130913093,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.0004511553115151273,
      "loss": 0.2309,
      "num_input_tokens_seen": 35665056,
      "step": 168995
    },
    {
      "epoch": 18.59185918591859,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.00045080485535169135,
      "loss": 0.2314,
      "num_input_tokens_seen": 35666112,
      "step": 169000
    },
    {
      "epoch": 18.59240924092409,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0004504545332812088,
      "loss": 0.2319,
      "num_input_tokens_seen": 35667168,
      "step": 169005
    },
    {
      "epoch": 18.592959295929592,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0004501043453069037,
      "loss": 0.2319,
      "num_input_tokens_seen": 35668256,
      "step": 169010
    },
    {
      "epoch": 18.593509350935093,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0004497542914320035,
      "loss": 0.2324,
      "num_input_tokens_seen": 35669312,
      "step": 169015
    },
    {
      "epoch": 18.594059405940595,
      "grad_norm": 0.0008697509765625,
      "learning_rate": 0.000449404371659739,
      "loss": 0.2288,
      "num_input_tokens_seen": 35670368,
      "step": 169020
    },
    {
      "epoch": 18.594609460946096,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0004490545859933292,
      "loss": 0.2309,
      "num_input_tokens_seen": 35671456,
      "step": 169025
    },
    {
      "epoch": 18.595159515951597,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0004487049344360017,
      "loss": 0.2308,
      "num_input_tokens_seen": 35672480,
      "step": 169030
    },
    {
      "epoch": 18.595709570957094,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.0004483554169909803,
      "loss": 0.2324,
      "num_input_tokens_seen": 35673504,
      "step": 169035
    },
    {
      "epoch": 18.596259625962595,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0004480060336614827,
      "loss": 0.2309,
      "num_input_tokens_seen": 35674560,
      "step": 169040
    },
    {
      "epoch": 18.596809680968097,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00044765678445073283,
      "loss": 0.2298,
      "num_input_tokens_seen": 35675584,
      "step": 169045
    },
    {
      "epoch": 18.597359735973598,
      "grad_norm": 0.00274658203125,
      "learning_rate": 0.00044730766936194486,
      "loss": 0.2314,
      "num_input_tokens_seen": 35676608,
      "step": 169050
    },
    {
      "epoch": 18.5979097909791,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.0004469586883983395,
      "loss": 0.2319,
      "num_input_tokens_seen": 35677696,
      "step": 169055
    },
    {
      "epoch": 18.5984598459846,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0004466098415631342,
      "loss": 0.2324,
      "num_input_tokens_seen": 35678720,
      "step": 169060
    },
    {
      "epoch": 18.599009900990097,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.00044626112885954135,
      "loss": 0.2345,
      "num_input_tokens_seen": 35679776,
      "step": 169065
    },
    {
      "epoch": 18.5995599559956,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.00044591255029077846,
      "loss": 0.2319,
      "num_input_tokens_seen": 35680832,
      "step": 169070
    },
    {
      "epoch": 18.6001100110011,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0004455641058600529,
      "loss": 0.2319,
      "num_input_tokens_seen": 35681984,
      "step": 169075
    },
    {
      "epoch": 18.6006600660066,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.00044521579557058044,
      "loss": 0.2329,
      "num_input_tokens_seen": 35683072,
      "step": 169080
    },
    {
      "epoch": 18.6012101210121,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.0004448676194255702,
      "loss": 0.2309,
      "num_input_tokens_seen": 35684064,
      "step": 169085
    },
    {
      "epoch": 18.601760176017603,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00044451957742822957,
      "loss": 0.2303,
      "num_input_tokens_seen": 35685088,
      "step": 169090
    },
    {
      "epoch": 18.602310231023104,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0004441716695817693,
      "loss": 0.2319,
      "num_input_tokens_seen": 35686144,
      "step": 169095
    },
    {
      "epoch": 18.6028602860286,
      "grad_norm": 0.000732421875,
      "learning_rate": 0.00044382389588939363,
      "loss": 0.234,
      "num_input_tokens_seen": 35687168,
      "step": 169100
    },
    {
      "epoch": 18.603410341034103,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 0.00044347625635430816,
      "loss": 0.2314,
      "num_input_tokens_seen": 35688224,
      "step": 169105
    },
    {
      "epoch": 18.603960396039604,
      "grad_norm": 0.00054931640625,
      "learning_rate": 0.00044312875097971547,
      "loss": 0.2288,
      "num_input_tokens_seen": 35689248,
      "step": 169110
    },
    {
      "epoch": 18.604510451045105,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.00044278137976882126,
      "loss": 0.2314,
      "num_input_tokens_seen": 35690304,
      "step": 169115
    },
    {
      "epoch": 18.605060506050606,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0004424341427248263,
      "loss": 0.2319,
      "num_input_tokens_seen": 35691360,
      "step": 169120
    },
    {
      "epoch": 18.605610561056107,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00044208703985093134,
      "loss": 0.2329,
      "num_input_tokens_seen": 35692448,
      "step": 169125
    },
    {
      "epoch": 18.606160616061608,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.00044174007115033387,
      "loss": 0.2314,
      "num_input_tokens_seen": 35693536,
      "step": 169130
    },
    {
      "epoch": 18.606710671067106,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.000441393236626233,
      "loss": 0.2314,
      "num_input_tokens_seen": 35694624,
      "step": 169135
    },
    {
      "epoch": 18.607260726072607,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0004410465362818228,
      "loss": 0.2335,
      "num_input_tokens_seen": 35695680,
      "step": 169140
    },
    {
      "epoch": 18.607810781078108,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0004406999701203007,
      "loss": 0.2319,
      "num_input_tokens_seen": 35696768,
      "step": 169145
    },
    {
      "epoch": 18.60836083608361,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00044035353814486254,
      "loss": 0.2319,
      "num_input_tokens_seen": 35697824,
      "step": 169150
    },
    {
      "epoch": 18.60891089108911,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.00044000724035869905,
      "loss": 0.2314,
      "num_input_tokens_seen": 35698912,
      "step": 169155
    },
    {
      "epoch": 18.60946094609461,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00043966107676500277,
      "loss": 0.2319,
      "num_input_tokens_seen": 35699968,
      "step": 169160
    },
    {
      "epoch": 18.61001100110011,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.000439315047366961,
      "loss": 0.2329,
      "num_input_tokens_seen": 35700992,
      "step": 169165
    },
    {
      "epoch": 18.61056105610561,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00043896915216776633,
      "loss": 0.2324,
      "num_input_tokens_seen": 35702080,
      "step": 169170
    },
    {
      "epoch": 18.61111111111111,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.00043862339117060775,
      "loss": 0.2303,
      "num_input_tokens_seen": 35703104,
      "step": 169175
    },
    {
      "epoch": 18.611661166116612,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0004382777643786695,
      "loss": 0.2319,
      "num_input_tokens_seen": 35704160,
      "step": 169180
    },
    {
      "epoch": 18.612211221122113,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.00043793227179513727,
      "loss": 0.2303,
      "num_input_tokens_seen": 35705184,
      "step": 169185
    },
    {
      "epoch": 18.612761276127614,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0004375869134231952,
      "loss": 0.2303,
      "num_input_tokens_seen": 35706208,
      "step": 169190
    },
    {
      "epoch": 18.61331133113311,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.0004372416892660291,
      "loss": 0.2309,
      "num_input_tokens_seen": 35707264,
      "step": 169195
    },
    {
      "epoch": 18.613861386138613,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.00043689659932681646,
      "loss": 0.2314,
      "num_input_tokens_seen": 35708320,
      "step": 169200
    },
    {
      "epoch": 18.614411441144114,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.000436551643608738,
      "loss": 0.2314,
      "num_input_tokens_seen": 35709376,
      "step": 169205
    },
    {
      "epoch": 18.614961496149615,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.00043620682211497784,
      "loss": 0.2303,
      "num_input_tokens_seen": 35710464,
      "step": 169210
    },
    {
      "epoch": 18.615511551155116,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.00043586213484871013,
      "loss": 0.2314,
      "num_input_tokens_seen": 35711488,
      "step": 169215
    },
    {
      "epoch": 18.616061606160617,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.00043551758181311403,
      "loss": 0.2324,
      "num_input_tokens_seen": 35712576,
      "step": 169220
    },
    {
      "epoch": 18.616611661166118,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.00043517316301136034,
      "loss": 0.2335,
      "num_input_tokens_seen": 35713568,
      "step": 169225
    },
    {
      "epoch": 18.617161716171616,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0004348288784466281,
      "loss": 0.2308,
      "num_input_tokens_seen": 35714592,
      "step": 169230
    },
    {
      "epoch": 18.617711771177117,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0004344847281220915,
      "loss": 0.2303,
      "num_input_tokens_seen": 35715648,
      "step": 169235
    },
    {
      "epoch": 18.618261826182618,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0004341407120409163,
      "loss": 0.2288,
      "num_input_tokens_seen": 35716672,
      "step": 169240
    },
    {
      "epoch": 18.61881188118812,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.00043379683020628,
      "loss": 0.2314,
      "num_input_tokens_seen": 35717728,
      "step": 169245
    },
    {
      "epoch": 18.61936193619362,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.0004334530826213484,
      "loss": 0.2335,
      "num_input_tokens_seen": 35718848,
      "step": 169250
    },
    {
      "epoch": 18.61991199119912,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.00043310946928928727,
      "loss": 0.2335,
      "num_input_tokens_seen": 35719872,
      "step": 169255
    },
    {
      "epoch": 18.620462046204622,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00043276599021326744,
      "loss": 0.2324,
      "num_input_tokens_seen": 35720960,
      "step": 169260
    },
    {
      "epoch": 18.62101210121012,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00043242264539645304,
      "loss": 0.2319,
      "num_input_tokens_seen": 35721984,
      "step": 169265
    },
    {
      "epoch": 18.62156215621562,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0004320794348420115,
      "loss": 0.2308,
      "num_input_tokens_seen": 35723072,
      "step": 169270
    },
    {
      "epoch": 18.622112211221122,
      "grad_norm": 0.0004177093505859375,
      "learning_rate": 0.000431736358553102,
      "loss": 0.2314,
      "num_input_tokens_seen": 35724160,
      "step": 169275
    },
    {
      "epoch": 18.622662266226623,
      "grad_norm": 0.002655029296875,
      "learning_rate": 0.000431393416532887,
      "loss": 0.234,
      "num_input_tokens_seen": 35725248,
      "step": 169280
    },
    {
      "epoch": 18.623212321232124,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0004310506087845289,
      "loss": 0.2303,
      "num_input_tokens_seen": 35726272,
      "step": 169285
    },
    {
      "epoch": 18.623762376237625,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.000430707935311187,
      "loss": 0.2309,
      "num_input_tokens_seen": 35727328,
      "step": 169290
    },
    {
      "epoch": 18.624312431243123,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0004303653961160186,
      "loss": 0.2309,
      "num_input_tokens_seen": 35728384,
      "step": 169295
    },
    {
      "epoch": 18.624862486248624,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0004300229912021813,
      "loss": 0.2324,
      "num_input_tokens_seen": 35729472,
      "step": 169300
    },
    {
      "epoch": 18.625412541254125,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0004296807205728292,
      "loss": 0.2324,
      "num_input_tokens_seen": 35730528,
      "step": 169305
    },
    {
      "epoch": 18.625962596259626,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00042933858423112146,
      "loss": 0.2308,
      "num_input_tokens_seen": 35731520,
      "step": 169310
    },
    {
      "epoch": 18.626512651265127,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00042899658218020553,
      "loss": 0.2329,
      "num_input_tokens_seen": 35732576,
      "step": 169315
    },
    {
      "epoch": 18.627062706270628,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0004286547144232355,
      "loss": 0.2303,
      "num_input_tokens_seen": 35733568,
      "step": 169320
    },
    {
      "epoch": 18.62761276127613,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.00042831298096336563,
      "loss": 0.2319,
      "num_input_tokens_seen": 35734624,
      "step": 169325
    },
    {
      "epoch": 18.628162816281627,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.0004279713818037417,
      "loss": 0.2303,
      "num_input_tokens_seen": 35735712,
      "step": 169330
    },
    {
      "epoch": 18.628712871287128,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.0004276299169475145,
      "loss": 0.2298,
      "num_input_tokens_seen": 35736768,
      "step": 169335
    },
    {
      "epoch": 18.62926292629263,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0004272885863978298,
      "loss": 0.2314,
      "num_input_tokens_seen": 35737856,
      "step": 169340
    },
    {
      "epoch": 18.62981298129813,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0004269473901578302,
      "loss": 0.2303,
      "num_input_tokens_seen": 35738912,
      "step": 169345
    },
    {
      "epoch": 18.63036303630363,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00042660632823066644,
      "loss": 0.2324,
      "num_input_tokens_seen": 35739936,
      "step": 169350
    },
    {
      "epoch": 18.630913091309132,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.0004262654006194777,
      "loss": 0.2298,
      "num_input_tokens_seen": 35740960,
      "step": 169355
    },
    {
      "epoch": 18.63146314631463,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.00042592460732740977,
      "loss": 0.2319,
      "num_input_tokens_seen": 35742048,
      "step": 169360
    },
    {
      "epoch": 18.63201320132013,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0004255839483576018,
      "loss": 0.2298,
      "num_input_tokens_seen": 35743168,
      "step": 169365
    },
    {
      "epoch": 18.632563256325632,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.00042524342371318965,
      "loss": 0.2309,
      "num_input_tokens_seen": 35744192,
      "step": 169370
    },
    {
      "epoch": 18.633113311331133,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.0004249030333973175,
      "loss": 0.2314,
      "num_input_tokens_seen": 35745280,
      "step": 169375
    },
    {
      "epoch": 18.633663366336634,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00042456277741311773,
      "loss": 0.233,
      "num_input_tokens_seen": 35746368,
      "step": 169380
    },
    {
      "epoch": 18.634213421342135,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00042422265576373295,
      "loss": 0.2298,
      "num_input_tokens_seen": 35747392,
      "step": 169385
    },
    {
      "epoch": 18.634763476347636,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00042388266845229226,
      "loss": 0.2335,
      "num_input_tokens_seen": 35748448,
      "step": 169390
    },
    {
      "epoch": 18.635313531353134,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0004235428154819298,
      "loss": 0.2298,
      "num_input_tokens_seen": 35749504,
      "step": 169395
    },
    {
      "epoch": 18.635863586358635,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00042320309685577983,
      "loss": 0.2304,
      "num_input_tokens_seen": 35750560,
      "step": 169400
    },
    {
      "epoch": 18.636413641364136,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00042286351257696974,
      "loss": 0.2304,
      "num_input_tokens_seen": 35751680,
      "step": 169405
    },
    {
      "epoch": 18.636963696369637,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0004225240626486354,
      "loss": 0.2329,
      "num_input_tokens_seen": 35752672,
      "step": 169410
    },
    {
      "epoch": 18.63751375137514,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00042218474707390093,
      "loss": 0.2314,
      "num_input_tokens_seen": 35753696,
      "step": 169415
    },
    {
      "epoch": 18.63806380638064,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00042184556585589226,
      "loss": 0.2319,
      "num_input_tokens_seen": 35754752,
      "step": 169420
    },
    {
      "epoch": 18.638613861386137,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00042150651899774017,
      "loss": 0.2298,
      "num_input_tokens_seen": 35755776,
      "step": 169425
    },
    {
      "epoch": 18.639163916391638,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00042116760650256547,
      "loss": 0.2314,
      "num_input_tokens_seen": 35756800,
      "step": 169430
    },
    {
      "epoch": 18.63971397139714,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.000420828828373494,
      "loss": 0.2319,
      "num_input_tokens_seen": 35757792,
      "step": 169435
    },
    {
      "epoch": 18.64026402640264,
      "grad_norm": 0.00177001953125,
      "learning_rate": 0.0004204901846136483,
      "loss": 0.2324,
      "num_input_tokens_seen": 35758880,
      "step": 169440
    },
    {
      "epoch": 18.64081408140814,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00042015167522614747,
      "loss": 0.2303,
      "num_input_tokens_seen": 35759936,
      "step": 169445
    },
    {
      "epoch": 18.641364136413642,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.0004198133002141141,
      "loss": 0.2303,
      "num_input_tokens_seen": 35760960,
      "step": 169450
    },
    {
      "epoch": 18.641914191419144,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.00041947505958066234,
      "loss": 0.2309,
      "num_input_tokens_seen": 35762080,
      "step": 169455
    },
    {
      "epoch": 18.64246424642464,
      "grad_norm": 0.001953125,
      "learning_rate": 0.0004191369533289163,
      "loss": 0.2314,
      "num_input_tokens_seen": 35763136,
      "step": 169460
    },
    {
      "epoch": 18.643014301430142,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0004187989814619852,
      "loss": 0.2309,
      "num_input_tokens_seen": 35764128,
      "step": 169465
    },
    {
      "epoch": 18.643564356435643,
      "grad_norm": 0.002655029296875,
      "learning_rate": 0.00041846114398298826,
      "loss": 0.2324,
      "num_input_tokens_seen": 35765216,
      "step": 169470
    },
    {
      "epoch": 18.644114411441144,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00041812344089503793,
      "loss": 0.2324,
      "num_input_tokens_seen": 35766208,
      "step": 169475
    },
    {
      "epoch": 18.644664466446645,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0004177858722012484,
      "loss": 0.2308,
      "num_input_tokens_seen": 35767232,
      "step": 169480
    },
    {
      "epoch": 18.645214521452147,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0004174484379047272,
      "loss": 0.2308,
      "num_input_tokens_seen": 35768320,
      "step": 169485
    },
    {
      "epoch": 18.645764576457644,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0004171111380085868,
      "loss": 0.2303,
      "num_input_tokens_seen": 35769408,
      "step": 169490
    },
    {
      "epoch": 18.646314631463145,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0004167739725159347,
      "loss": 0.2309,
      "num_input_tokens_seen": 35770432,
      "step": 169495
    },
    {
      "epoch": 18.646864686468646,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0004164369414298802,
      "loss": 0.2324,
      "num_input_tokens_seen": 35771488,
      "step": 169500
    },
    {
      "epoch": 18.647414741474147,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00041610004475353067,
      "loss": 0.2308,
      "num_input_tokens_seen": 35772544,
      "step": 169505
    },
    {
      "epoch": 18.64796479647965,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0004157632824899854,
      "loss": 0.2319,
      "num_input_tokens_seen": 35773632,
      "step": 169510
    },
    {
      "epoch": 18.64851485148515,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0004154266546423535,
      "loss": 0.234,
      "num_input_tokens_seen": 35774688,
      "step": 169515
    },
    {
      "epoch": 18.64906490649065,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00041509016121373584,
      "loss": 0.2308,
      "num_input_tokens_seen": 35775776,
      "step": 169520
    },
    {
      "epoch": 18.649614961496148,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00041475380220723165,
      "loss": 0.2308,
      "num_input_tokens_seen": 35776864,
      "step": 169525
    },
    {
      "epoch": 18.65016501650165,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00041441757762594673,
      "loss": 0.2314,
      "num_input_tokens_seen": 35777888,
      "step": 169530
    },
    {
      "epoch": 18.65071507150715,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00041408148747297367,
      "loss": 0.2308,
      "num_input_tokens_seen": 35778944,
      "step": 169535
    },
    {
      "epoch": 18.65126512651265,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.0004137455317514132,
      "loss": 0.2324,
      "num_input_tokens_seen": 35779936,
      "step": 169540
    },
    {
      "epoch": 18.651815181518153,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 0.0004134097104643597,
      "loss": 0.2288,
      "num_input_tokens_seen": 35781024,
      "step": 169545
    },
    {
      "epoch": 18.652365236523654,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0004130740236149105,
      "loss": 0.2314,
      "num_input_tokens_seen": 35782016,
      "step": 169550
    },
    {
      "epoch": 18.652915291529155,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.00041273847120615824,
      "loss": 0.2314,
      "num_input_tokens_seen": 35783040,
      "step": 169555
    },
    {
      "epoch": 18.653465346534652,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 0.0004124030532411971,
      "loss": 0.2319,
      "num_input_tokens_seen": 35784064,
      "step": 169560
    },
    {
      "epoch": 18.654015401540153,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.0004120677697231162,
      "loss": 0.2308,
      "num_input_tokens_seen": 35785120,
      "step": 169565
    },
    {
      "epoch": 18.654565456545654,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0004117326206550065,
      "loss": 0.2298,
      "num_input_tokens_seen": 35786176,
      "step": 169570
    },
    {
      "epoch": 18.655115511551156,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00041139760603995875,
      "loss": 0.2308,
      "num_input_tokens_seen": 35787232,
      "step": 169575
    },
    {
      "epoch": 18.655665566556657,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 0.00041106272588105564,
      "loss": 0.2298,
      "num_input_tokens_seen": 35788256,
      "step": 169580
    },
    {
      "epoch": 18.656215621562158,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.00041072798018138956,
      "loss": 0.2319,
      "num_input_tokens_seen": 35789280,
      "step": 169585
    },
    {
      "epoch": 18.656765676567655,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00041039336894404307,
      "loss": 0.2329,
      "num_input_tokens_seen": 35790304,
      "step": 169590
    },
    {
      "epoch": 18.657315731573156,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.00041005889217209877,
      "loss": 0.2319,
      "num_input_tokens_seen": 35791328,
      "step": 169595
    },
    {
      "epoch": 18.657865786578657,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00040972454986864246,
      "loss": 0.2314,
      "num_input_tokens_seen": 35792384,
      "step": 169600
    },
    {
      "epoch": 18.65841584158416,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0004093903420367517,
      "loss": 0.2308,
      "num_input_tokens_seen": 35793440,
      "step": 169605
    },
    {
      "epoch": 18.65896589658966,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00040905626867950907,
      "loss": 0.2319,
      "num_input_tokens_seen": 35794464,
      "step": 169610
    },
    {
      "epoch": 18.65951595159516,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.00040872232979999535,
      "loss": 0.2324,
      "num_input_tokens_seen": 35795584,
      "step": 169615
    },
    {
      "epoch": 18.66006600660066,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0004083885254012848,
      "loss": 0.2308,
      "num_input_tokens_seen": 35796544,
      "step": 169620
    },
    {
      "epoch": 18.66061606160616,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.00040805485548645656,
      "loss": 0.2324,
      "num_input_tokens_seen": 35797504,
      "step": 169625
    },
    {
      "epoch": 18.66116611661166,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0004077213200585866,
      "loss": 0.2324,
      "num_input_tokens_seen": 35798496,
      "step": 169630
    },
    {
      "epoch": 18.66171617161716,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.00040738791912074413,
      "loss": 0.2324,
      "num_input_tokens_seen": 35799520,
      "step": 169635
    },
    {
      "epoch": 18.662266226622663,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00040705465267600494,
      "loss": 0.2298,
      "num_input_tokens_seen": 35800544,
      "step": 169640
    },
    {
      "epoch": 18.662816281628164,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0004067215207274399,
      "loss": 0.2309,
      "num_input_tokens_seen": 35801536,
      "step": 169645
    },
    {
      "epoch": 18.663366336633665,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00040638852327812335,
      "loss": 0.2319,
      "num_input_tokens_seen": 35802560,
      "step": 169650
    },
    {
      "epoch": 18.663916391639162,
      "grad_norm": 0.006439208984375,
      "learning_rate": 0.0004060556603311194,
      "loss": 0.2324,
      "num_input_tokens_seen": 35803552,
      "step": 169655
    },
    {
      "epoch": 18.664466446644663,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.0004057229318894972,
      "loss": 0.2303,
      "num_input_tokens_seen": 35804608,
      "step": 169660
    },
    {
      "epoch": 18.665016501650165,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0004053903379563228,
      "loss": 0.2314,
      "num_input_tokens_seen": 35805760,
      "step": 169665
    },
    {
      "epoch": 18.665566556655666,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.0004050578785346653,
      "loss": 0.2324,
      "num_input_tokens_seen": 35806816,
      "step": 169670
    },
    {
      "epoch": 18.666116611661167,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.000404725553627584,
      "loss": 0.2308,
      "num_input_tokens_seen": 35807936,
      "step": 169675
    },
    {
      "epoch": 18.666666666666668,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.00040439336323814466,
      "loss": 0.2309,
      "num_input_tokens_seen": 35809024,
      "step": 169680
    },
    {
      "epoch": 18.66721672167217,
      "grad_norm": 0.001312255859375,
      "learning_rate": 0.00040406130736940656,
      "loss": 0.2303,
      "num_input_tokens_seen": 35810080,
      "step": 169685
    },
    {
      "epoch": 18.667766776677666,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00040372938602443396,
      "loss": 0.2319,
      "num_input_tokens_seen": 35811104,
      "step": 169690
    },
    {
      "epoch": 18.668316831683168,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.000403397599206281,
      "loss": 0.2329,
      "num_input_tokens_seen": 35812192,
      "step": 169695
    },
    {
      "epoch": 18.66886688668867,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.00040306594691801034,
      "loss": 0.2303,
      "num_input_tokens_seen": 35813248,
      "step": 169700
    },
    {
      "epoch": 18.66941694169417,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00040273442916267607,
      "loss": 0.2309,
      "num_input_tokens_seen": 35814336,
      "step": 169705
    },
    {
      "epoch": 18.66996699669967,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00040240304594333254,
      "loss": 0.2319,
      "num_input_tokens_seen": 35815456,
      "step": 169710
    },
    {
      "epoch": 18.670517051705172,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00040207179726303884,
      "loss": 0.2319,
      "num_input_tokens_seen": 35816576,
      "step": 169715
    },
    {
      "epoch": 18.67106710671067,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.000401740683124841,
      "loss": 0.2303,
      "num_input_tokens_seen": 35817632,
      "step": 169720
    },
    {
      "epoch": 18.67161716171617,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0004014097035317948,
      "loss": 0.2303,
      "num_input_tokens_seen": 35818688,
      "step": 169725
    },
    {
      "epoch": 18.67216721672167,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0004010788584869529,
      "loss": 0.2324,
      "num_input_tokens_seen": 35819712,
      "step": 169730
    },
    {
      "epoch": 18.672717271727173,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0004007481479933594,
      "loss": 0.233,
      "num_input_tokens_seen": 35820800,
      "step": 169735
    },
    {
      "epoch": 18.673267326732674,
      "grad_norm": 0.00045013427734375,
      "learning_rate": 0.000400417572054067,
      "loss": 0.2335,
      "num_input_tokens_seen": 35821856,
      "step": 169740
    },
    {
      "epoch": 18.673817381738175,
      "grad_norm": 0.0005035400390625,
      "learning_rate": 0.00040008713067211986,
      "loss": 0.2298,
      "num_input_tokens_seen": 35822880,
      "step": 169745
    },
    {
      "epoch": 18.674367436743676,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0003997568238505622,
      "loss": 0.2324,
      "num_input_tokens_seen": 35823968,
      "step": 169750
    },
    {
      "epoch": 18.674917491749174,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0003994266515924416,
      "loss": 0.2314,
      "num_input_tokens_seen": 35824992,
      "step": 169755
    },
    {
      "epoch": 18.675467546754675,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.0003990966139007973,
      "loss": 0.2319,
      "num_input_tokens_seen": 35826048,
      "step": 169760
    },
    {
      "epoch": 18.676017601760176,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.00039876671077867517,
      "loss": 0.2329,
      "num_input_tokens_seen": 35827136,
      "step": 169765
    },
    {
      "epoch": 18.676567656765677,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0003984369422291145,
      "loss": 0.2314,
      "num_input_tokens_seen": 35828192,
      "step": 169770
    },
    {
      "epoch": 18.677117711771178,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.0003981073082551528,
      "loss": 0.2314,
      "num_input_tokens_seen": 35829216,
      "step": 169775
    },
    {
      "epoch": 18.67766776677668,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.00039777780885982936,
      "loss": 0.2319,
      "num_input_tokens_seen": 35830272,
      "step": 169780
    },
    {
      "epoch": 18.678217821782177,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 0.0003974484440461817,
      "loss": 0.2308,
      "num_input_tokens_seen": 35831360,
      "step": 169785
    },
    {
      "epoch": 18.678767876787678,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0003971192138172441,
      "loss": 0.2293,
      "num_input_tokens_seen": 35832416,
      "step": 169790
    },
    {
      "epoch": 18.67931793179318,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.0003967901181760541,
      "loss": 0.2319,
      "num_input_tokens_seen": 35833504,
      "step": 169795
    },
    {
      "epoch": 18.67986798679868,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.00039646115712563935,
      "loss": 0.2314,
      "num_input_tokens_seen": 35834624,
      "step": 169800
    },
    {
      "epoch": 18.68041804180418,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0003961323306690356,
      "loss": 0.2319,
      "num_input_tokens_seen": 35835744,
      "step": 169805
    },
    {
      "epoch": 18.680968096809682,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00039580363880927226,
      "loss": 0.2303,
      "num_input_tokens_seen": 35836736,
      "step": 169810
    },
    {
      "epoch": 18.681518151815183,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.0003954750815493785,
      "loss": 0.2293,
      "num_input_tokens_seen": 35837824,
      "step": 169815
    },
    {
      "epoch": 18.68206820682068,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0003951466588923835,
      "loss": 0.2314,
      "num_input_tokens_seen": 35838944,
      "step": 169820
    },
    {
      "epoch": 18.682618261826182,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.0003948183708413133,
      "loss": 0.2314,
      "num_input_tokens_seen": 35840032,
      "step": 169825
    },
    {
      "epoch": 18.683168316831683,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.00039449021739919543,
      "loss": 0.2314,
      "num_input_tokens_seen": 35841088,
      "step": 169830
    },
    {
      "epoch": 18.683718371837184,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.0003941621985690508,
      "loss": 0.2303,
      "num_input_tokens_seen": 35842144,
      "step": 169835
    },
    {
      "epoch": 18.684268426842685,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00039383431435390524,
      "loss": 0.2314,
      "num_input_tokens_seen": 35843200,
      "step": 169840
    },
    {
      "epoch": 18.684818481848186,
      "grad_norm": 0.000667572021484375,
      "learning_rate": 0.00039350656475678144,
      "loss": 0.2314,
      "num_input_tokens_seen": 35844224,
      "step": 169845
    },
    {
      "epoch": 18.685368536853684,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.000393178949780697,
      "loss": 0.2308,
      "num_input_tokens_seen": 35845248,
      "step": 169850
    },
    {
      "epoch": 18.685918591859185,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0003928514694286744,
      "loss": 0.2324,
      "num_input_tokens_seen": 35846240,
      "step": 169855
    },
    {
      "epoch": 18.686468646864686,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00039252412370373134,
      "loss": 0.2314,
      "num_input_tokens_seen": 35847232,
      "step": 169860
    },
    {
      "epoch": 18.687018701870187,
      "grad_norm": 0.000652313232421875,
      "learning_rate": 0.000392196912608882,
      "loss": 0.2314,
      "num_input_tokens_seen": 35848288,
      "step": 169865
    },
    {
      "epoch": 18.687568756875688,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00039186983614714397,
      "loss": 0.2308,
      "num_input_tokens_seen": 35849376,
      "step": 169870
    },
    {
      "epoch": 18.68811881188119,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0003915428943215332,
      "loss": 0.2314,
      "num_input_tokens_seen": 35850368,
      "step": 169875
    },
    {
      "epoch": 18.68866886688669,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00039121608713506224,
      "loss": 0.2303,
      "num_input_tokens_seen": 35851392,
      "step": 169880
    },
    {
      "epoch": 18.689218921892188,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0003908894145907421,
      "loss": 0.2303,
      "num_input_tokens_seen": 35852416,
      "step": 169885
    },
    {
      "epoch": 18.68976897689769,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.0003905628766915836,
      "loss": 0.234,
      "num_input_tokens_seen": 35853504,
      "step": 169890
    },
    {
      "epoch": 18.69031903190319,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.00039023647344059773,
      "loss": 0.2293,
      "num_input_tokens_seen": 35854560,
      "step": 169895
    },
    {
      "epoch": 18.69086908690869,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.0003899102048407904,
      "loss": 0.2319,
      "num_input_tokens_seen": 35855616,
      "step": 169900
    },
    {
      "epoch": 18.691419141914192,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.00038958407089517087,
      "loss": 0.2309,
      "num_input_tokens_seen": 35856640,
      "step": 169905
    },
    {
      "epoch": 18.691969196919693,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.0003892580716067451,
      "loss": 0.2309,
      "num_input_tokens_seen": 35857664,
      "step": 169910
    },
    {
      "epoch": 18.69251925192519,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0003889322069785156,
      "loss": 0.2319,
      "num_input_tokens_seen": 35858656,
      "step": 169915
    },
    {
      "epoch": 18.693069306930692,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0003886064770134884,
      "loss": 0.2314,
      "num_input_tokens_seen": 35859744,
      "step": 169920
    },
    {
      "epoch": 18.693619361936193,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00038828088171466267,
      "loss": 0.2314,
      "num_input_tokens_seen": 35860736,
      "step": 169925
    },
    {
      "epoch": 18.694169416941694,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00038795542108503944,
      "loss": 0.2304,
      "num_input_tokens_seen": 35861792,
      "step": 169930
    },
    {
      "epoch": 18.694719471947195,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.00038763009512762293,
      "loss": 0.2303,
      "num_input_tokens_seen": 35862912,
      "step": 169935
    },
    {
      "epoch": 18.695269526952696,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0003873049038454057,
      "loss": 0.234,
      "num_input_tokens_seen": 35864000,
      "step": 169940
    },
    {
      "epoch": 18.695819581958197,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0003869798472413888,
      "loss": 0.2329,
      "num_input_tokens_seen": 35865088,
      "step": 169945
    },
    {
      "epoch": 18.696369636963695,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0003866549253185647,
      "loss": 0.2314,
      "num_input_tokens_seen": 35866112,
      "step": 169950
    },
    {
      "epoch": 18.696919691969196,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 0.00038633013807992943,
      "loss": 0.2309,
      "num_input_tokens_seen": 35867136,
      "step": 169955
    },
    {
      "epoch": 18.697469746974697,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00038600548552848057,
      "loss": 0.2303,
      "num_input_tokens_seen": 35868160,
      "step": 169960
    },
    {
      "epoch": 18.698019801980198,
      "grad_norm": 0.005859375,
      "learning_rate": 0.000385680967667204,
      "loss": 0.2313,
      "num_input_tokens_seen": 35869152,
      "step": 169965
    },
    {
      "epoch": 18.6985698569857,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.00038535658449909415,
      "loss": 0.2319,
      "num_input_tokens_seen": 35870176,
      "step": 169970
    },
    {
      "epoch": 18.6991199119912,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.00038503233602714014,
      "loss": 0.2345,
      "num_input_tokens_seen": 35871328,
      "step": 169975
    },
    {
      "epoch": 18.6996699669967,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0003847082222543296,
      "loss": 0.2329,
      "num_input_tokens_seen": 35872288,
      "step": 169980
    },
    {
      "epoch": 18.7002200220022,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 0.0003843842431836503,
      "loss": 0.2314,
      "num_input_tokens_seen": 35873376,
      "step": 169985
    },
    {
      "epoch": 18.7007700770077,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00038406039881808796,
      "loss": 0.2345,
      "num_input_tokens_seen": 35874528,
      "step": 169990
    },
    {
      "epoch": 18.7013201320132,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.0003837366891606286,
      "loss": 0.2298,
      "num_input_tokens_seen": 35875552,
      "step": 169995
    },
    {
      "epoch": 18.701870187018702,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0003834131142142516,
      "loss": 0.2324,
      "num_input_tokens_seen": 35876640,
      "step": 170000
    },
    {
      "epoch": 18.702420242024203,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.00038308967398194613,
      "loss": 0.2314,
      "num_input_tokens_seen": 35877696,
      "step": 170005
    },
    {
      "epoch": 18.702970297029704,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0003827663684666865,
      "loss": 0.2293,
      "num_input_tokens_seen": 35878720,
      "step": 170010
    },
    {
      "epoch": 18.703520352035202,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0003824431976714554,
      "loss": 0.2314,
      "num_input_tokens_seen": 35879808,
      "step": 170015
    },
    {
      "epoch": 18.704070407040703,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.0003821201615992337,
      "loss": 0.2309,
      "num_input_tokens_seen": 35880768,
      "step": 170020
    },
    {
      "epoch": 18.704620462046204,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.000381797260252994,
      "loss": 0.2303,
      "num_input_tokens_seen": 35881824,
      "step": 170025
    },
    {
      "epoch": 18.705170517051705,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0003814744936357156,
      "loss": 0.2303,
      "num_input_tokens_seen": 35882848,
      "step": 170030
    },
    {
      "epoch": 18.705720572057206,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.00038115186175037287,
      "loss": 0.2319,
      "num_input_tokens_seen": 35883936,
      "step": 170035
    },
    {
      "epoch": 18.706270627062707,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.000380829364599935,
      "loss": 0.2303,
      "num_input_tokens_seen": 35885056,
      "step": 170040
    },
    {
      "epoch": 18.706820682068205,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.00038050700218737965,
      "loss": 0.2298,
      "num_input_tokens_seen": 35886112,
      "step": 170045
    },
    {
      "epoch": 18.707370737073706,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0003801847745156761,
      "loss": 0.2324,
      "num_input_tokens_seen": 35887168,
      "step": 170050
    },
    {
      "epoch": 18.707920792079207,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00037986268158779533,
      "loss": 0.233,
      "num_input_tokens_seen": 35888224,
      "step": 170055
    },
    {
      "epoch": 18.70847084708471,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00037954072340670497,
      "loss": 0.2314,
      "num_input_tokens_seen": 35889248,
      "step": 170060
    },
    {
      "epoch": 18.70902090209021,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00037921889997537093,
      "loss": 0.2288,
      "num_input_tokens_seen": 35890304,
      "step": 170065
    },
    {
      "epoch": 18.70957095709571,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0003788972112967609,
      "loss": 0.2335,
      "num_input_tokens_seen": 35891360,
      "step": 170070
    },
    {
      "epoch": 18.71012101210121,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0003785756573738408,
      "loss": 0.2298,
      "num_input_tokens_seen": 35892480,
      "step": 170075
    },
    {
      "epoch": 18.71067106710671,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0003782542382095699,
      "loss": 0.2308,
      "num_input_tokens_seen": 35893600,
      "step": 170080
    },
    {
      "epoch": 18.71122112211221,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0003779329538069159,
      "loss": 0.2329,
      "num_input_tokens_seen": 35894656,
      "step": 170085
    },
    {
      "epoch": 18.71177117711771,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0003776118041688364,
      "loss": 0.2319,
      "num_input_tokens_seen": 35895744,
      "step": 170090
    },
    {
      "epoch": 18.712321232123212,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.00037729078929829406,
      "loss": 0.2309,
      "num_input_tokens_seen": 35896864,
      "step": 170095
    },
    {
      "epoch": 18.712871287128714,
      "grad_norm": 0.0024566650390625,
      "learning_rate": 0.0003769699091982448,
      "loss": 0.2298,
      "num_input_tokens_seen": 35897984,
      "step": 170100
    },
    {
      "epoch": 18.713421342134215,
      "grad_norm": 0.00060272216796875,
      "learning_rate": 0.0003766491638716446,
      "loss": 0.2314,
      "num_input_tokens_seen": 35899040,
      "step": 170105
    },
    {
      "epoch": 18.713971397139716,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.0003763285533214561,
      "loss": 0.2314,
      "num_input_tokens_seen": 35900064,
      "step": 170110
    },
    {
      "epoch": 18.714521452145213,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.00037600807755062694,
      "loss": 0.2314,
      "num_input_tokens_seen": 35901152,
      "step": 170115
    },
    {
      "epoch": 18.715071507150714,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0003756877365621164,
      "loss": 0.2314,
      "num_input_tokens_seen": 35902208,
      "step": 170120
    },
    {
      "epoch": 18.715621562156215,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00037536753035887545,
      "loss": 0.2304,
      "num_input_tokens_seen": 35903264,
      "step": 170125
    },
    {
      "epoch": 18.716171617161717,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00037504745894385014,
      "loss": 0.2324,
      "num_input_tokens_seen": 35904320,
      "step": 170130
    },
    {
      "epoch": 18.716721672167218,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.00037472752231999806,
      "loss": 0.2298,
      "num_input_tokens_seen": 35905408,
      "step": 170135
    },
    {
      "epoch": 18.71727172717272,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0003744077204902635,
      "loss": 0.2314,
      "num_input_tokens_seen": 35906464,
      "step": 170140
    },
    {
      "epoch": 18.717821782178216,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.00037408805345759573,
      "loss": 0.2293,
      "num_input_tokens_seen": 35907552,
      "step": 170145
    },
    {
      "epoch": 18.718371837183717,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.00037376852122494083,
      "loss": 0.233,
      "num_input_tokens_seen": 35908672,
      "step": 170150
    },
    {
      "epoch": 18.71892189218922,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0003734491237952397,
      "loss": 0.2308,
      "num_input_tokens_seen": 35909760,
      "step": 170155
    },
    {
      "epoch": 18.71947194719472,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00037312986117144176,
      "loss": 0.2308,
      "num_input_tokens_seen": 35910848,
      "step": 170160
    },
    {
      "epoch": 18.72002200220022,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00037281073335648616,
      "loss": 0.2303,
      "num_input_tokens_seen": 35911872,
      "step": 170165
    },
    {
      "epoch": 18.72057205720572,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.0003724917403533173,
      "loss": 0.2303,
      "num_input_tokens_seen": 35912896,
      "step": 170170
    },
    {
      "epoch": 18.721122112211223,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0003721728821648729,
      "loss": 0.2308,
      "num_input_tokens_seen": 35913984,
      "step": 170175
    },
    {
      "epoch": 18.72167216721672,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0003718541587940904,
      "loss": 0.2298,
      "num_input_tokens_seen": 35915072,
      "step": 170180
    },
    {
      "epoch": 18.72222222222222,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00037153557024390937,
      "loss": 0.2309,
      "num_input_tokens_seen": 35916032,
      "step": 170185
    },
    {
      "epoch": 18.722772277227723,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00037121711651726396,
      "loss": 0.2303,
      "num_input_tokens_seen": 35917056,
      "step": 170190
    },
    {
      "epoch": 18.723322332233224,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0003708987976170935,
      "loss": 0.2319,
      "num_input_tokens_seen": 35918112,
      "step": 170195
    },
    {
      "epoch": 18.723872387238725,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0003705806135463291,
      "loss": 0.2319,
      "num_input_tokens_seen": 35919136,
      "step": 170200
    },
    {
      "epoch": 18.724422442244226,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.0003702625643079016,
      "loss": 0.2309,
      "num_input_tokens_seen": 35920192,
      "step": 170205
    },
    {
      "epoch": 18.724972497249723,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0003699446499047454,
      "loss": 0.2303,
      "num_input_tokens_seen": 35921280,
      "step": 170210
    },
    {
      "epoch": 18.725522552255224,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0003696268703397881,
      "loss": 0.2319,
      "num_input_tokens_seen": 35922400,
      "step": 170215
    },
    {
      "epoch": 18.726072607260726,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00036930922561595914,
      "loss": 0.2314,
      "num_input_tokens_seen": 35923456,
      "step": 170220
    },
    {
      "epoch": 18.726622662266227,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0003689917157361877,
      "loss": 0.2319,
      "num_input_tokens_seen": 35924544,
      "step": 170225
    },
    {
      "epoch": 18.727172717271728,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0003686743407033982,
      "loss": 0.2303,
      "num_input_tokens_seen": 35925600,
      "step": 170230
    },
    {
      "epoch": 18.72772277227723,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00036835710052051827,
      "loss": 0.2313,
      "num_input_tokens_seen": 35926656,
      "step": 170235
    },
    {
      "epoch": 18.72827282728273,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.00036803999519046727,
      "loss": 0.2324,
      "num_input_tokens_seen": 35927712,
      "step": 170240
    },
    {
      "epoch": 18.728822882288227,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00036772302471617277,
      "loss": 0.2324,
      "num_input_tokens_seen": 35928736,
      "step": 170245
    },
    {
      "epoch": 18.72937293729373,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00036740618910055257,
      "loss": 0.2314,
      "num_input_tokens_seen": 35929760,
      "step": 170250
    },
    {
      "epoch": 18.72992299229923,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0003670894883465292,
      "loss": 0.2309,
      "num_input_tokens_seen": 35930912,
      "step": 170255
    },
    {
      "epoch": 18.73047304730473,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.0003667729224570204,
      "loss": 0.233,
      "num_input_tokens_seen": 35931936,
      "step": 170260
    },
    {
      "epoch": 18.731023102310232,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.0003664564914349438,
      "loss": 0.2309,
      "num_input_tokens_seen": 35933024,
      "step": 170265
    },
    {
      "epoch": 18.731573157315733,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0003661401952832155,
      "loss": 0.2314,
      "num_input_tokens_seen": 35934048,
      "step": 170270
    },
    {
      "epoch": 18.73212321232123,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.0003658240340047497,
      "loss": 0.2308,
      "num_input_tokens_seen": 35935104,
      "step": 170275
    },
    {
      "epoch": 18.73267326732673,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.00036550800760246245,
      "loss": 0.2308,
      "num_input_tokens_seen": 35936160,
      "step": 170280
    },
    {
      "epoch": 18.733223322332233,
      "grad_norm": 0.000766754150390625,
      "learning_rate": 0.0003651921160792665,
      "loss": 0.2324,
      "num_input_tokens_seen": 35937248,
      "step": 170285
    },
    {
      "epoch": 18.733773377337734,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.0003648763594380727,
      "loss": 0.2324,
      "num_input_tokens_seen": 35938304,
      "step": 170290
    },
    {
      "epoch": 18.734323432343235,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.00036456073768179054,
      "loss": 0.2319,
      "num_input_tokens_seen": 35939296,
      "step": 170295
    },
    {
      "epoch": 18.734873487348736,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.0003642452508133292,
      "loss": 0.2304,
      "num_input_tokens_seen": 35940384,
      "step": 170300
    },
    {
      "epoch": 18.735423542354237,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0003639298988355949,
      "loss": 0.2335,
      "num_input_tokens_seen": 35941408,
      "step": 170305
    },
    {
      "epoch": 18.735973597359735,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0003636146817514968,
      "loss": 0.2303,
      "num_input_tokens_seen": 35942496,
      "step": 170310
    },
    {
      "epoch": 18.736523652365236,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0003632995995639393,
      "loss": 0.2309,
      "num_input_tokens_seen": 35943584,
      "step": 170315
    },
    {
      "epoch": 18.737073707370737,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0003629846522758251,
      "loss": 0.2314,
      "num_input_tokens_seen": 35944576,
      "step": 170320
    },
    {
      "epoch": 18.737623762376238,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.00036266983989006015,
      "loss": 0.2293,
      "num_input_tokens_seen": 35945696,
      "step": 170325
    },
    {
      "epoch": 18.73817381738174,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00036235516240954054,
      "loss": 0.233,
      "num_input_tokens_seen": 35946752,
      "step": 170330
    },
    {
      "epoch": 18.73872387238724,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 0.00036204061983717063,
      "loss": 0.2288,
      "num_input_tokens_seen": 35947840,
      "step": 170335
    },
    {
      "epoch": 18.739273927392738,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00036172621217584963,
      "loss": 0.2304,
      "num_input_tokens_seen": 35948896,
      "step": 170340
    },
    {
      "epoch": 18.73982398239824,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.00036141193942847204,
      "loss": 0.2303,
      "num_input_tokens_seen": 35949920,
      "step": 170345
    },
    {
      "epoch": 18.74037403740374,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0003610978015979388,
      "loss": 0.2303,
      "num_input_tokens_seen": 35950912,
      "step": 170350
    },
    {
      "epoch": 18.74092409240924,
      "grad_norm": 0.00089263916015625,
      "learning_rate": 0.00036078379868713926,
      "loss": 0.2303,
      "num_input_tokens_seen": 35952032,
      "step": 170355
    },
    {
      "epoch": 18.741474147414742,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00036046993069897447,
      "loss": 0.2314,
      "num_input_tokens_seen": 35953056,
      "step": 170360
    },
    {
      "epoch": 18.742024202420243,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.0003601561976363304,
      "loss": 0.2314,
      "num_input_tokens_seen": 35954048,
      "step": 170365
    },
    {
      "epoch": 18.742574257425744,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.00035984259950210315,
      "loss": 0.2309,
      "num_input_tokens_seen": 35955136,
      "step": 170370
    },
    {
      "epoch": 18.74312431243124,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.00035952913629918367,
      "loss": 0.2314,
      "num_input_tokens_seen": 35956160,
      "step": 170375
    },
    {
      "epoch": 18.743674367436743,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.00035921580803045637,
      "loss": 0.2314,
      "num_input_tokens_seen": 35957152,
      "step": 170380
    },
    {
      "epoch": 18.744224422442244,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.0003589026146988122,
      "loss": 0.2314,
      "num_input_tokens_seen": 35958240,
      "step": 170385
    },
    {
      "epoch": 18.744774477447745,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00035858955630713727,
      "loss": 0.2303,
      "num_input_tokens_seen": 35959296,
      "step": 170390
    },
    {
      "epoch": 18.745324532453246,
      "grad_norm": 0.000667572021484375,
      "learning_rate": 0.00035827663285831587,
      "loss": 0.2308,
      "num_input_tokens_seen": 35960352,
      "step": 170395
    },
    {
      "epoch": 18.745874587458747,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0003579638443552341,
      "loss": 0.2314,
      "num_input_tokens_seen": 35961408,
      "step": 170400
    },
    {
      "epoch": 18.746424642464248,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00035765119080077287,
      "loss": 0.2319,
      "num_input_tokens_seen": 35962400,
      "step": 170405
    },
    {
      "epoch": 18.746974697469746,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 0.0003573386721978167,
      "loss": 0.2298,
      "num_input_tokens_seen": 35963424,
      "step": 170410
    },
    {
      "epoch": 18.747524752475247,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0003570262885492431,
      "loss": 0.2308,
      "num_input_tokens_seen": 35964448,
      "step": 170415
    },
    {
      "epoch": 18.748074807480748,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00035671403985792995,
      "loss": 0.2314,
      "num_input_tokens_seen": 35965504,
      "step": 170420
    },
    {
      "epoch": 18.74862486248625,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00035640192612675813,
      "loss": 0.2319,
      "num_input_tokens_seen": 35966496,
      "step": 170425
    },
    {
      "epoch": 18.74917491749175,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0003560899473586021,
      "loss": 0.2298,
      "num_input_tokens_seen": 35967488,
      "step": 170430
    },
    {
      "epoch": 18.74972497249725,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0003557781035563395,
      "loss": 0.2319,
      "num_input_tokens_seen": 35968576,
      "step": 170435
    },
    {
      "epoch": 18.75027502750275,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00035546639472284314,
      "loss": 0.2303,
      "num_input_tokens_seen": 35969568,
      "step": 170440
    },
    {
      "epoch": 18.75082508250825,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.00035515482086098394,
      "loss": 0.2309,
      "num_input_tokens_seen": 35970656,
      "step": 170445
    },
    {
      "epoch": 18.75137513751375,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.0003548433819736363,
      "loss": 0.2324,
      "num_input_tokens_seen": 35971712,
      "step": 170450
    },
    {
      "epoch": 18.751925192519252,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.00035453207806367126,
      "loss": 0.2319,
      "num_input_tokens_seen": 35972832,
      "step": 170455
    },
    {
      "epoch": 18.752475247524753,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00035422090913395484,
      "loss": 0.2308,
      "num_input_tokens_seen": 35973856,
      "step": 170460
    },
    {
      "epoch": 18.753025302530254,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00035390987518735815,
      "loss": 0.2319,
      "num_input_tokens_seen": 35974880,
      "step": 170465
    },
    {
      "epoch": 18.753575357535752,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.0003535989762267455,
      "loss": 0.2319,
      "num_input_tokens_seen": 35976032,
      "step": 170470
    },
    {
      "epoch": 18.754125412541253,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00035328821225498295,
      "loss": 0.2314,
      "num_input_tokens_seen": 35977120,
      "step": 170475
    },
    {
      "epoch": 18.754675467546754,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0003529775832749349,
      "loss": 0.2304,
      "num_input_tokens_seen": 35978176,
      "step": 170480
    },
    {
      "epoch": 18.755225522552255,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.000352667089289464,
      "loss": 0.2313,
      "num_input_tokens_seen": 35979264,
      "step": 170485
    },
    {
      "epoch": 18.755775577557756,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0003523567303014313,
      "loss": 0.2309,
      "num_input_tokens_seen": 35980352,
      "step": 170490
    },
    {
      "epoch": 18.756325632563257,
      "grad_norm": 0.0028228759765625,
      "learning_rate": 0.0003520465063136996,
      "loss": 0.234,
      "num_input_tokens_seen": 35981440,
      "step": 170495
    },
    {
      "epoch": 18.75687568756876,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0003517364173291265,
      "loss": 0.2283,
      "num_input_tokens_seen": 35982464,
      "step": 170500
    },
    {
      "epoch": 18.757425742574256,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.00035142646335056815,
      "loss": 0.2293,
      "num_input_tokens_seen": 35983456,
      "step": 170505
    },
    {
      "epoch": 18.757975797579757,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00035111664438088215,
      "loss": 0.2298,
      "num_input_tokens_seen": 35984448,
      "step": 170510
    },
    {
      "epoch": 18.758525852585258,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00035080696042292804,
      "loss": 0.2303,
      "num_input_tokens_seen": 35985536,
      "step": 170515
    },
    {
      "epoch": 18.75907590759076,
      "grad_norm": 0.0062255859375,
      "learning_rate": 0.00035049741147955503,
      "loss": 0.2319,
      "num_input_tokens_seen": 35986624,
      "step": 170520
    },
    {
      "epoch": 18.75962596259626,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00035018799755361926,
      "loss": 0.2314,
      "num_input_tokens_seen": 35987712,
      "step": 170525
    },
    {
      "epoch": 18.76017601760176,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00034987871864797015,
      "loss": 0.2314,
      "num_input_tokens_seen": 35988800,
      "step": 170530
    },
    {
      "epoch": 18.760726072607262,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.0003495695747654587,
      "loss": 0.2329,
      "num_input_tokens_seen": 35989824,
      "step": 170535
    },
    {
      "epoch": 18.76127612761276,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.0003492605659089343,
      "loss": 0.2303,
      "num_input_tokens_seen": 35990944,
      "step": 170540
    },
    {
      "epoch": 18.76182618261826,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.00034895169208124465,
      "loss": 0.2303,
      "num_input_tokens_seen": 35991936,
      "step": 170545
    },
    {
      "epoch": 18.762376237623762,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0003486429532852392,
      "loss": 0.2298,
      "num_input_tokens_seen": 35992960,
      "step": 170550
    },
    {
      "epoch": 18.762926292629263,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.00034833434952376063,
      "loss": 0.2319,
      "num_input_tokens_seen": 35994016,
      "step": 170555
    },
    {
      "epoch": 18.763476347634764,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00034802588079965166,
      "loss": 0.2309,
      "num_input_tokens_seen": 35995072,
      "step": 170560
    },
    {
      "epoch": 18.764026402640265,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0003477175471157584,
      "loss": 0.2303,
      "num_input_tokens_seen": 35996160,
      "step": 170565
    },
    {
      "epoch": 18.764576457645763,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0003474093484749219,
      "loss": 0.2319,
      "num_input_tokens_seen": 35997184,
      "step": 170570
    },
    {
      "epoch": 18.765126512651264,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0003471012848799848,
      "loss": 0.2319,
      "num_input_tokens_seen": 35998240,
      "step": 170575
    },
    {
      "epoch": 18.765676567656765,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00034679335633378157,
      "loss": 0.2304,
      "num_input_tokens_seen": 35999264,
      "step": 170580
    },
    {
      "epoch": 18.766226622662266,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 0.0003464855628391533,
      "loss": 0.2303,
      "num_input_tokens_seen": 36000320,
      "step": 170585
    },
    {
      "epoch": 18.766776677667767,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.00034617790439893603,
      "loss": 0.2309,
      "num_input_tokens_seen": 36001440,
      "step": 170590
    },
    {
      "epoch": 18.76732673267327,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00034587038101596576,
      "loss": 0.2314,
      "num_input_tokens_seen": 36002464,
      "step": 170595
    },
    {
      "epoch": 18.76787678767877,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0003455629926930753,
      "loss": 0.2314,
      "num_input_tokens_seen": 36003520,
      "step": 170600
    },
    {
      "epoch": 18.768426842684267,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00034525573943310073,
      "loss": 0.2329,
      "num_input_tokens_seen": 36004608,
      "step": 170605
    },
    {
      "epoch": 18.768976897689768,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0003449486212388697,
      "loss": 0.2335,
      "num_input_tokens_seen": 36005696,
      "step": 170610
    },
    {
      "epoch": 18.76952695269527,
      "grad_norm": 0.005157470703125,
      "learning_rate": 0.0003446416381132167,
      "loss": 0.2314,
      "num_input_tokens_seen": 36006656,
      "step": 170615
    },
    {
      "epoch": 18.77007700770077,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0003443347900589677,
      "loss": 0.234,
      "num_input_tokens_seen": 36007680,
      "step": 170620
    },
    {
      "epoch": 18.77062706270627,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00034402807707895386,
      "loss": 0.2298,
      "num_input_tokens_seen": 36008736,
      "step": 170625
    },
    {
      "epoch": 18.771177117711773,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0003437214991760012,
      "loss": 0.2324,
      "num_input_tokens_seen": 36009824,
      "step": 170630
    },
    {
      "epoch": 18.77172717271727,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00034341505635293254,
      "loss": 0.2308,
      "num_input_tokens_seen": 36010880,
      "step": 170635
    },
    {
      "epoch": 18.77227722772277,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0003431087486125772,
      "loss": 0.2329,
      "num_input_tokens_seen": 36011968,
      "step": 170640
    },
    {
      "epoch": 18.772827282728272,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00034280257595775465,
      "loss": 0.2308,
      "num_input_tokens_seen": 36012992,
      "step": 170645
    },
    {
      "epoch": 18.773377337733773,
      "grad_norm": 0.006622314453125,
      "learning_rate": 0.0003424965383912859,
      "loss": 0.2319,
      "num_input_tokens_seen": 36014080,
      "step": 170650
    },
    {
      "epoch": 18.773927392739274,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0003421906359159937,
      "loss": 0.2309,
      "num_input_tokens_seen": 36015104,
      "step": 170655
    },
    {
      "epoch": 18.774477447744776,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00034188486853469754,
      "loss": 0.2293,
      "num_input_tokens_seen": 36016224,
      "step": 170660
    },
    {
      "epoch": 18.775027502750277,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.00034157923625021514,
      "loss": 0.2324,
      "num_input_tokens_seen": 36017216,
      "step": 170665
    },
    {
      "epoch": 18.775577557755774,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.00034127373906536415,
      "loss": 0.2303,
      "num_input_tokens_seen": 36018272,
      "step": 170670
    },
    {
      "epoch": 18.776127612761275,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0003409683769829574,
      "loss": 0.2324,
      "num_input_tokens_seen": 36019392,
      "step": 170675
    },
    {
      "epoch": 18.776677667766776,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00034066315000581105,
      "loss": 0.2314,
      "num_input_tokens_seen": 36020448,
      "step": 170680
    },
    {
      "epoch": 18.777227722772277,
      "grad_norm": 0.0004711151123046875,
      "learning_rate": 0.0003403580581367377,
      "loss": 0.2298,
      "num_input_tokens_seen": 36021504,
      "step": 170685
    },
    {
      "epoch": 18.77777777777778,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.0003400531013785518,
      "loss": 0.2308,
      "num_input_tokens_seen": 36022592,
      "step": 170690
    },
    {
      "epoch": 18.77832783278328,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.0003397482797340612,
      "loss": 0.2324,
      "num_input_tokens_seen": 36023680,
      "step": 170695
    },
    {
      "epoch": 18.778877887788777,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00033944359320607517,
      "loss": 0.2329,
      "num_input_tokens_seen": 36024704,
      "step": 170700
    },
    {
      "epoch": 18.77942794279428,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0003391390417974033,
      "loss": 0.2303,
      "num_input_tokens_seen": 36025792,
      "step": 170705
    },
    {
      "epoch": 18.77997799779978,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0003388346255108515,
      "loss": 0.2335,
      "num_input_tokens_seen": 36026848,
      "step": 170710
    },
    {
      "epoch": 18.78052805280528,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.000338530344349226,
      "loss": 0.2319,
      "num_input_tokens_seen": 36027936,
      "step": 170715
    },
    {
      "epoch": 18.78107810781078,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0003382261983153312,
      "loss": 0.2309,
      "num_input_tokens_seen": 36029056,
      "step": 170720
    },
    {
      "epoch": 18.781628162816283,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0003379221874119698,
      "loss": 0.2329,
      "num_input_tokens_seen": 36030048,
      "step": 170725
    },
    {
      "epoch": 18.782178217821784,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.00033761831164194454,
      "loss": 0.2314,
      "num_input_tokens_seen": 36031104,
      "step": 170730
    },
    {
      "epoch": 18.78272827282728,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.00033731457100805493,
      "loss": 0.234,
      "num_input_tokens_seen": 36032128,
      "step": 170735
    },
    {
      "epoch": 18.783278327832782,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.00033701096551310203,
      "loss": 0.2319,
      "num_input_tokens_seen": 36033184,
      "step": 170740
    },
    {
      "epoch": 18.783828382838283,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.00033670749515988364,
      "loss": 0.2293,
      "num_input_tokens_seen": 36034144,
      "step": 170745
    },
    {
      "epoch": 18.784378437843785,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 0.0003364041599511941,
      "loss": 0.2335,
      "num_input_tokens_seen": 36035200,
      "step": 170750
    },
    {
      "epoch": 18.784928492849286,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0003361009598898329,
      "loss": 0.2314,
      "num_input_tokens_seen": 36036256,
      "step": 170755
    },
    {
      "epoch": 18.785478547854787,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0003357978949785911,
      "loss": 0.2309,
      "num_input_tokens_seen": 36037344,
      "step": 170760
    },
    {
      "epoch": 18.786028602860284,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00033549496522026486,
      "loss": 0.2324,
      "num_input_tokens_seen": 36038400,
      "step": 170765
    },
    {
      "epoch": 18.786578657865785,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0003351921706176436,
      "loss": 0.2319,
      "num_input_tokens_seen": 36039456,
      "step": 170770
    },
    {
      "epoch": 18.787128712871286,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.00033488951117352007,
      "loss": 0.2308,
      "num_input_tokens_seen": 36040544,
      "step": 170775
    },
    {
      "epoch": 18.787678767876788,
      "grad_norm": 0.00083160400390625,
      "learning_rate": 0.00033458698689068365,
      "loss": 0.2308,
      "num_input_tokens_seen": 36041568,
      "step": 170780
    },
    {
      "epoch": 18.78822882288229,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0003342845977719205,
      "loss": 0.2314,
      "num_input_tokens_seen": 36042592,
      "step": 170785
    },
    {
      "epoch": 18.78877887788779,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0003339823438200201,
      "loss": 0.2314,
      "num_input_tokens_seen": 36043648,
      "step": 170790
    },
    {
      "epoch": 18.78932893289329,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00033368022503776684,
      "loss": 0.2314,
      "num_input_tokens_seen": 36044704,
      "step": 170795
    },
    {
      "epoch": 18.78987898789879,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00033337824142794345,
      "loss": 0.2303,
      "num_input_tokens_seen": 36045728,
      "step": 170800
    },
    {
      "epoch": 18.79042904290429,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00033307639299333776,
      "loss": 0.2309,
      "num_input_tokens_seen": 36046784,
      "step": 170805
    },
    {
      "epoch": 18.79097909790979,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.0003327746797367259,
      "loss": 0.2324,
      "num_input_tokens_seen": 36047936,
      "step": 170810
    },
    {
      "epoch": 18.79152915291529,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0003324731016608956,
      "loss": 0.2314,
      "num_input_tokens_seen": 36048928,
      "step": 170815
    },
    {
      "epoch": 18.792079207920793,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0003321716587686213,
      "loss": 0.2314,
      "num_input_tokens_seen": 36049920,
      "step": 170820
    },
    {
      "epoch": 18.792629262926294,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0003318703510626808,
      "loss": 0.2304,
      "num_input_tokens_seen": 36050976,
      "step": 170825
    },
    {
      "epoch": 18.793179317931795,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.00033156917854585354,
      "loss": 0.2319,
      "num_input_tokens_seen": 36052064,
      "step": 170830
    },
    {
      "epoch": 18.793729372937293,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0003312681412209156,
      "loss": 0.2308,
      "num_input_tokens_seen": 36053056,
      "step": 170835
    },
    {
      "epoch": 18.794279427942794,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.00033096723909063986,
      "loss": 0.2319,
      "num_input_tokens_seen": 36054208,
      "step": 170840
    },
    {
      "epoch": 18.794829482948295,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0003306664721578006,
      "loss": 0.2303,
      "num_input_tokens_seen": 36055232,
      "step": 170845
    },
    {
      "epoch": 18.795379537953796,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00033036584042516745,
      "loss": 0.2303,
      "num_input_tokens_seen": 36056288,
      "step": 170850
    },
    {
      "epoch": 18.795929592959297,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00033006534389551476,
      "loss": 0.2319,
      "num_input_tokens_seen": 36057312,
      "step": 170855
    },
    {
      "epoch": 18.796479647964798,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.000329764982571612,
      "loss": 0.2299,
      "num_input_tokens_seen": 36058400,
      "step": 170860
    },
    {
      "epoch": 18.797029702970296,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00032946475645622363,
      "loss": 0.2309,
      "num_input_tokens_seen": 36059424,
      "step": 170865
    },
    {
      "epoch": 18.797579757975797,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.00032916466555212075,
      "loss": 0.2329,
      "num_input_tokens_seen": 36060448,
      "step": 170870
    },
    {
      "epoch": 18.798129812981298,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.0003288647098620645,
      "loss": 0.2324,
      "num_input_tokens_seen": 36061440,
      "step": 170875
    },
    {
      "epoch": 18.7986798679868,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.000328564889388826,
      "loss": 0.2309,
      "num_input_tokens_seen": 36062496,
      "step": 170880
    },
    {
      "epoch": 18.7992299229923,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00032826520413516136,
      "loss": 0.2335,
      "num_input_tokens_seen": 36063520,
      "step": 170885
    },
    {
      "epoch": 18.7997799779978,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00032796565410383836,
      "loss": 0.2308,
      "num_input_tokens_seen": 36064576,
      "step": 170890
    },
    {
      "epoch": 18.8003300330033,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0003276662392976148,
      "loss": 0.2314,
      "num_input_tokens_seen": 36065632,
      "step": 170895
    },
    {
      "epoch": 18.8008800880088,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0003273669597192519,
      "loss": 0.2335,
      "num_input_tokens_seen": 36066624,
      "step": 170900
    },
    {
      "epoch": 18.8014301430143,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 0.0003270678153715073,
      "loss": 0.2303,
      "num_input_tokens_seen": 36067712,
      "step": 170905
    },
    {
      "epoch": 18.801980198019802,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0003267688062571372,
      "loss": 0.2309,
      "num_input_tokens_seen": 36068736,
      "step": 170910
    },
    {
      "epoch": 18.802530253025303,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.00032646993237889606,
      "loss": 0.2313,
      "num_input_tokens_seen": 36069792,
      "step": 170915
    },
    {
      "epoch": 18.803080308030804,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.0003261711937395434,
      "loss": 0.2314,
      "num_input_tokens_seen": 36070816,
      "step": 170920
    },
    {
      "epoch": 18.803630363036305,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0003258725903418286,
      "loss": 0.2319,
      "num_input_tokens_seen": 36071808,
      "step": 170925
    },
    {
      "epoch": 18.804180418041803,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00032557412218850446,
      "loss": 0.2308,
      "num_input_tokens_seen": 36072864,
      "step": 170930
    },
    {
      "epoch": 18.804730473047304,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0003252757892823238,
      "loss": 0.2324,
      "num_input_tokens_seen": 36073824,
      "step": 170935
    },
    {
      "epoch": 18.805280528052805,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.0003249775916260328,
      "loss": 0.2314,
      "num_input_tokens_seen": 36074848,
      "step": 170940
    },
    {
      "epoch": 18.805830583058306,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.0003246795292223825,
      "loss": 0.2319,
      "num_input_tokens_seen": 36075872,
      "step": 170945
    },
    {
      "epoch": 18.806380638063807,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00032438160207411745,
      "loss": 0.2329,
      "num_input_tokens_seen": 36076960,
      "step": 170950
    },
    {
      "epoch": 18.806930693069308,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00032408381018398714,
      "loss": 0.2309,
      "num_input_tokens_seen": 36077984,
      "step": 170955
    },
    {
      "epoch": 18.80748074807481,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 0.00032378615355473426,
      "loss": 0.2329,
      "num_input_tokens_seen": 36079040,
      "step": 170960
    },
    {
      "epoch": 18.808030803080307,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0003234886321891017,
      "loss": 0.2314,
      "num_input_tokens_seen": 36080128,
      "step": 170965
    },
    {
      "epoch": 18.808580858085808,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0003231912460898323,
      "loss": 0.2309,
      "num_input_tokens_seen": 36081184,
      "step": 170970
    },
    {
      "epoch": 18.80913091309131,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.0003228939952596638,
      "loss": 0.2308,
      "num_input_tokens_seen": 36082272,
      "step": 170975
    },
    {
      "epoch": 18.80968096809681,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00032259687970134066,
      "loss": 0.2309,
      "num_input_tokens_seen": 36083328,
      "step": 170980
    },
    {
      "epoch": 18.81023102310231,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.00032229989941760074,
      "loss": 0.2293,
      "num_input_tokens_seen": 36084448,
      "step": 170985
    },
    {
      "epoch": 18.810781078107812,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0003220030544111768,
      "loss": 0.2329,
      "num_input_tokens_seen": 36085472,
      "step": 170990
    },
    {
      "epoch": 18.81133113311331,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00032170634468481004,
      "loss": 0.2293,
      "num_input_tokens_seen": 36086528,
      "step": 170995
    },
    {
      "epoch": 18.81188118811881,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00032140977024122994,
      "loss": 0.2314,
      "num_input_tokens_seen": 36087552,
      "step": 171000
    },
    {
      "epoch": 18.812431243124312,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00032111333108317264,
      "loss": 0.2329,
      "num_input_tokens_seen": 36088640,
      "step": 171005
    },
    {
      "epoch": 18.812981298129813,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.00032081702721337255,
      "loss": 0.2303,
      "num_input_tokens_seen": 36089696,
      "step": 171010
    },
    {
      "epoch": 18.813531353135314,
      "grad_norm": 0.0023651123046875,
      "learning_rate": 0.0003205208586345559,
      "loss": 0.2324,
      "num_input_tokens_seen": 36090816,
      "step": 171015
    },
    {
      "epoch": 18.814081408140815,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00032022482534945716,
      "loss": 0.2308,
      "num_input_tokens_seen": 36091872,
      "step": 171020
    },
    {
      "epoch": 18.814631463146316,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0003199289273608008,
      "loss": 0.2308,
      "num_input_tokens_seen": 36092960,
      "step": 171025
    },
    {
      "epoch": 18.815181518151814,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.0003196331646713146,
      "loss": 0.234,
      "num_input_tokens_seen": 36094048,
      "step": 171030
    },
    {
      "epoch": 18.815731573157315,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.00031933753728372483,
      "loss": 0.2298,
      "num_input_tokens_seen": 36095072,
      "step": 171035
    },
    {
      "epoch": 18.816281628162816,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0003190420452007575,
      "loss": 0.2324,
      "num_input_tokens_seen": 36096128,
      "step": 171040
    },
    {
      "epoch": 18.816831683168317,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00031874668842513554,
      "loss": 0.2319,
      "num_input_tokens_seen": 36097184,
      "step": 171045
    },
    {
      "epoch": 18.817381738173818,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0003184514669595817,
      "loss": 0.2314,
      "num_input_tokens_seen": 36098240,
      "step": 171050
    },
    {
      "epoch": 18.81793179317932,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.0003181563808068122,
      "loss": 0.2293,
      "num_input_tokens_seen": 36099296,
      "step": 171055
    },
    {
      "epoch": 18.818481848184817,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 0.00031786142996955146,
      "loss": 0.2314,
      "num_input_tokens_seen": 36100384,
      "step": 171060
    },
    {
      "epoch": 18.819031903190318,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00031756661445051735,
      "loss": 0.2314,
      "num_input_tokens_seen": 36101504,
      "step": 171065
    },
    {
      "epoch": 18.81958195819582,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.0003172719342524277,
      "loss": 0.2308,
      "num_input_tokens_seen": 36102656,
      "step": 171070
    },
    {
      "epoch": 18.82013201320132,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.000316977389377997,
      "loss": 0.2309,
      "num_input_tokens_seen": 36103776,
      "step": 171075
    },
    {
      "epoch": 18.82068206820682,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0003166829798299381,
      "loss": 0.2309,
      "num_input_tokens_seen": 36104832,
      "step": 171080
    },
    {
      "epoch": 18.821232123212322,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00031638870561096876,
      "loss": 0.2319,
      "num_input_tokens_seen": 36105888,
      "step": 171085
    },
    {
      "epoch": 18.821782178217823,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00031609456672379685,
      "loss": 0.2314,
      "num_input_tokens_seen": 36106976,
      "step": 171090
    },
    {
      "epoch": 18.82233223322332,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00031580056317113525,
      "loss": 0.2298,
      "num_input_tokens_seen": 36108032,
      "step": 171095
    },
    {
      "epoch": 18.822882288228822,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00031550669495569505,
      "loss": 0.2319,
      "num_input_tokens_seen": 36109024,
      "step": 171100
    },
    {
      "epoch": 18.823432343234323,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.00031521296208018244,
      "loss": 0.2309,
      "num_input_tokens_seen": 36110048,
      "step": 171105
    },
    {
      "epoch": 18.823982398239824,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 0.00031491936454730693,
      "loss": 0.2309,
      "num_input_tokens_seen": 36111104,
      "step": 171110
    },
    {
      "epoch": 18.824532453245325,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0003146259023597714,
      "loss": 0.2309,
      "num_input_tokens_seen": 36112096,
      "step": 171115
    },
    {
      "epoch": 18.825082508250826,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.00031433257552028194,
      "loss": 0.2319,
      "num_input_tokens_seen": 36113120,
      "step": 171120
    },
    {
      "epoch": 18.825632563256324,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00031403938403154306,
      "loss": 0.2335,
      "num_input_tokens_seen": 36114176,
      "step": 171125
    },
    {
      "epoch": 18.826182618261825,
      "grad_norm": 0.002349853515625,
      "learning_rate": 0.000313746327896256,
      "loss": 0.2314,
      "num_input_tokens_seen": 36115136,
      "step": 171130
    },
    {
      "epoch": 18.826732673267326,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00031345340711712354,
      "loss": 0.2303,
      "num_input_tokens_seen": 36116160,
      "step": 171135
    },
    {
      "epoch": 18.827282728272827,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0003131606216968419,
      "loss": 0.2293,
      "num_input_tokens_seen": 36117248,
      "step": 171140
    },
    {
      "epoch": 18.82783278327833,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0003128679716381122,
      "loss": 0.2324,
      "num_input_tokens_seen": 36118336,
      "step": 171145
    },
    {
      "epoch": 18.82838283828383,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.000312575456943629,
      "loss": 0.2314,
      "num_input_tokens_seen": 36119424,
      "step": 171150
    },
    {
      "epoch": 18.82893289328933,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00031228307761609173,
      "loss": 0.2303,
      "num_input_tokens_seen": 36120480,
      "step": 171155
    },
    {
      "epoch": 18.829482948294828,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00031199083365819337,
      "loss": 0.2303,
      "num_input_tokens_seen": 36121568,
      "step": 171160
    },
    {
      "epoch": 18.83003300330033,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0003116987250726266,
      "loss": 0.2308,
      "num_input_tokens_seen": 36122688,
      "step": 171165
    },
    {
      "epoch": 18.83058305830583,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0003114067518620861,
      "loss": 0.2329,
      "num_input_tokens_seen": 36123712,
      "step": 171170
    },
    {
      "epoch": 18.83113311331133,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0003111149140292596,
      "loss": 0.2303,
      "num_input_tokens_seen": 36124832,
      "step": 171175
    },
    {
      "epoch": 18.831683168316832,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00031082321157684,
      "loss": 0.2309,
      "num_input_tokens_seen": 36125920,
      "step": 171180
    },
    {
      "epoch": 18.832233223322334,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.00031053164450751345,
      "loss": 0.2329,
      "num_input_tokens_seen": 36126944,
      "step": 171185
    },
    {
      "epoch": 18.83278327832783,
      "grad_norm": 0.0009002685546875,
      "learning_rate": 0.00031024021282396785,
      "loss": 0.2324,
      "num_input_tokens_seen": 36128032,
      "step": 171190
    },
    {
      "epoch": 18.833333333333332,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00030994891652889097,
      "loss": 0.2324,
      "num_input_tokens_seen": 36129056,
      "step": 171195
    },
    {
      "epoch": 18.833883388338833,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.0003096577556249658,
      "loss": 0.2324,
      "num_input_tokens_seen": 36130080,
      "step": 171200
    },
    {
      "epoch": 18.834433443344334,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.000309366730114875,
      "loss": 0.2309,
      "num_input_tokens_seen": 36131072,
      "step": 171205
    },
    {
      "epoch": 18.834983498349835,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0003090758400012999,
      "loss": 0.2324,
      "num_input_tokens_seen": 36132096,
      "step": 171210
    },
    {
      "epoch": 18.835533553355337,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0003087850852869267,
      "loss": 0.2314,
      "num_input_tokens_seen": 36133152,
      "step": 171215
    },
    {
      "epoch": 18.836083608360838,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00030849446597442986,
      "loss": 0.2319,
      "num_input_tokens_seen": 36134208,
      "step": 171220
    },
    {
      "epoch": 18.836633663366335,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00030820398206649226,
      "loss": 0.2319,
      "num_input_tokens_seen": 36135200,
      "step": 171225
    },
    {
      "epoch": 18.837183718371836,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00030791363356578506,
      "loss": 0.2298,
      "num_input_tokens_seen": 36136288,
      "step": 171230
    },
    {
      "epoch": 18.837733773377337,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.00030762342047498944,
      "loss": 0.2335,
      "num_input_tokens_seen": 36137376,
      "step": 171235
    },
    {
      "epoch": 18.83828382838284,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.00030733334279678,
      "loss": 0.234,
      "num_input_tokens_seen": 36138400,
      "step": 171240
    },
    {
      "epoch": 18.83883388338834,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00030704340053382795,
      "loss": 0.2303,
      "num_input_tokens_seen": 36139360,
      "step": 171245
    },
    {
      "epoch": 18.83938393839384,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.00030675359368880607,
      "loss": 0.2298,
      "num_input_tokens_seen": 36140416,
      "step": 171250
    },
    {
      "epoch": 18.83993399339934,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.00030646392226438724,
      "loss": 0.2314,
      "num_input_tokens_seen": 36141472,
      "step": 171255
    },
    {
      "epoch": 18.84048404840484,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 0.0003061743862632393,
      "loss": 0.2308,
      "num_input_tokens_seen": 36142528,
      "step": 171260
    },
    {
      "epoch": 18.84103410341034,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00030588498568803025,
      "loss": 0.2329,
      "num_input_tokens_seen": 36143616,
      "step": 171265
    },
    {
      "epoch": 18.84158415841584,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00030559572054142777,
      "loss": 0.2314,
      "num_input_tokens_seen": 36144736,
      "step": 171270
    },
    {
      "epoch": 18.842134213421343,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0003053065908260999,
      "loss": 0.2308,
      "num_input_tokens_seen": 36145760,
      "step": 171275
    },
    {
      "epoch": 18.842684268426844,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.0003050175965447094,
      "loss": 0.2298,
      "num_input_tokens_seen": 36146784,
      "step": 171280
    },
    {
      "epoch": 18.843234323432345,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.00030472873769992246,
      "loss": 0.2319,
      "num_input_tokens_seen": 36147872,
      "step": 171285
    },
    {
      "epoch": 18.843784378437842,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 0.00030444001429439537,
      "loss": 0.2308,
      "num_input_tokens_seen": 36148992,
      "step": 171290
    },
    {
      "epoch": 18.844334433443343,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 0.0003041514263307959,
      "loss": 0.2324,
      "num_input_tokens_seen": 36150080,
      "step": 171295
    },
    {
      "epoch": 18.844884488448844,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.0003038629738117804,
      "loss": 0.2309,
      "num_input_tokens_seen": 36151168,
      "step": 171300
    },
    {
      "epoch": 18.845434543454346,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.0003035746567400066,
      "loss": 0.2329,
      "num_input_tokens_seen": 36152256,
      "step": 171305
    },
    {
      "epoch": 18.845984598459847,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00030328647511813575,
      "loss": 0.2309,
      "num_input_tokens_seen": 36153280,
      "step": 171310
    },
    {
      "epoch": 18.846534653465348,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.00030299842894882076,
      "loss": 0.2335,
      "num_input_tokens_seen": 36154304,
      "step": 171315
    },
    {
      "epoch": 18.847084708470845,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0003027105182347162,
      "loss": 0.2324,
      "num_input_tokens_seen": 36155360,
      "step": 171320
    },
    {
      "epoch": 18.847634763476346,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.0003024227429784748,
      "loss": 0.2335,
      "num_input_tokens_seen": 36156416,
      "step": 171325
    },
    {
      "epoch": 18.848184818481847,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0003021351031827513,
      "loss": 0.2319,
      "num_input_tokens_seen": 36157440,
      "step": 171330
    },
    {
      "epoch": 18.84873487348735,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.00030184759885019675,
      "loss": 0.2319,
      "num_input_tokens_seen": 36158528,
      "step": 171335
    },
    {
      "epoch": 18.84928492849285,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00030156022998346075,
      "loss": 0.2314,
      "num_input_tokens_seen": 36159584,
      "step": 171340
    },
    {
      "epoch": 18.84983498349835,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0003012729965851896,
      "loss": 0.2324,
      "num_input_tokens_seen": 36160672,
      "step": 171345
    },
    {
      "epoch": 18.850385038503852,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00030098589865803105,
      "loss": 0.2303,
      "num_input_tokens_seen": 36161728,
      "step": 171350
    },
    {
      "epoch": 18.85093509350935,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00030069893620463304,
      "loss": 0.2324,
      "num_input_tokens_seen": 36162816,
      "step": 171355
    },
    {
      "epoch": 18.85148514851485,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 0.0003004121092276418,
      "loss": 0.2303,
      "num_input_tokens_seen": 36163840,
      "step": 171360
    },
    {
      "epoch": 18.85203520352035,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0003001254177296969,
      "loss": 0.2324,
      "num_input_tokens_seen": 36164864,
      "step": 171365
    },
    {
      "epoch": 18.852585258525853,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00029983886171344287,
      "loss": 0.2313,
      "num_input_tokens_seen": 36166016,
      "step": 171370
    },
    {
      "epoch": 18.853135313531354,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.0002995524411815209,
      "loss": 0.2303,
      "num_input_tokens_seen": 36166976,
      "step": 171375
    },
    {
      "epoch": 18.853685368536855,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.00029926615613656723,
      "loss": 0.2324,
      "num_input_tokens_seen": 36168000,
      "step": 171380
    },
    {
      "epoch": 18.854235423542356,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0002989800065812248,
      "loss": 0.2309,
      "num_input_tokens_seen": 36169152,
      "step": 171385
    },
    {
      "epoch": 18.854785478547853,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00029869399251812974,
      "loss": 0.2314,
      "num_input_tokens_seen": 36170240,
      "step": 171390
    },
    {
      "epoch": 18.855335533553355,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0002984081139499184,
      "loss": 0.2308,
      "num_input_tokens_seen": 36171296,
      "step": 171395
    },
    {
      "epoch": 18.855885588558856,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.00029812237087922355,
      "loss": 0.2319,
      "num_input_tokens_seen": 36172320,
      "step": 171400
    },
    {
      "epoch": 18.856435643564357,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00029783676330868145,
      "loss": 0.2329,
      "num_input_tokens_seen": 36173344,
      "step": 171405
    },
    {
      "epoch": 18.856985698569858,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.0002975512912409217,
      "loss": 0.2319,
      "num_input_tokens_seen": 36174368,
      "step": 171410
    },
    {
      "epoch": 18.85753575357536,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 0.0002972659546785789,
      "loss": 0.2329,
      "num_input_tokens_seen": 36175424,
      "step": 171415
    },
    {
      "epoch": 18.858085808580856,
      "grad_norm": 0.0009765625,
      "learning_rate": 0.0002969807536242774,
      "loss": 0.2324,
      "num_input_tokens_seen": 36176608,
      "step": 171420
    },
    {
      "epoch": 18.858635863586358,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00029669568808065206,
      "loss": 0.2319,
      "num_input_tokens_seen": 36177664,
      "step": 171425
    },
    {
      "epoch": 18.85918591859186,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00029641075805032725,
      "loss": 0.2319,
      "num_input_tokens_seen": 36178656,
      "step": 171430
    },
    {
      "epoch": 18.85973597359736,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.0002961259635359276,
      "loss": 0.2324,
      "num_input_tokens_seen": 36179808,
      "step": 171435
    },
    {
      "epoch": 18.86028602860286,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0002958413045400793,
      "loss": 0.2308,
      "num_input_tokens_seen": 36180896,
      "step": 171440
    },
    {
      "epoch": 18.860836083608362,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.00029555678106540695,
      "loss": 0.2314,
      "num_input_tokens_seen": 36181888,
      "step": 171445
    },
    {
      "epoch": 18.861386138613863,
      "grad_norm": 0.002532958984375,
      "learning_rate": 0.0002952723931145318,
      "loss": 0.2319,
      "num_input_tokens_seen": 36182976,
      "step": 171450
    },
    {
      "epoch": 18.86193619361936,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.00029498814069007503,
      "loss": 0.2293,
      "num_input_tokens_seen": 36184032,
      "step": 171455
    },
    {
      "epoch": 18.86248624862486,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00029470402379465463,
      "loss": 0.2319,
      "num_input_tokens_seen": 36185024,
      "step": 171460
    },
    {
      "epoch": 18.863036303630363,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0002944200424308918,
      "loss": 0.2329,
      "num_input_tokens_seen": 36186048,
      "step": 171465
    },
    {
      "epoch": 18.863586358635864,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.00029413619660140276,
      "loss": 0.2298,
      "num_input_tokens_seen": 36187104,
      "step": 171470
    },
    {
      "epoch": 18.864136413641365,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00029385248630880545,
      "loss": 0.2324,
      "num_input_tokens_seen": 36188128,
      "step": 171475
    },
    {
      "epoch": 18.864686468646866,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0002935689115557144,
      "loss": 0.2309,
      "num_input_tokens_seen": 36189216,
      "step": 171480
    },
    {
      "epoch": 18.865236523652364,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00029328547234473755,
      "loss": 0.2298,
      "num_input_tokens_seen": 36190208,
      "step": 171485
    },
    {
      "epoch": 18.865786578657865,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.00029300216867849446,
      "loss": 0.2303,
      "num_input_tokens_seen": 36191264,
      "step": 171490
    },
    {
      "epoch": 18.866336633663366,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 0.0002927190005595931,
      "loss": 0.2329,
      "num_input_tokens_seen": 36192320,
      "step": 171495
    },
    {
      "epoch": 18.866886688668867,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0002924359679906413,
      "loss": 0.2308,
      "num_input_tokens_seen": 36193344,
      "step": 171500
    },
    {
      "epoch": 18.867436743674368,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.000292153070974252,
      "loss": 0.2324,
      "num_input_tokens_seen": 36194368,
      "step": 171505
    },
    {
      "epoch": 18.86798679867987,
      "grad_norm": 0.002105712890625,
      "learning_rate": 0.0002918703095130298,
      "loss": 0.2314,
      "num_input_tokens_seen": 36195488,
      "step": 171510
    },
    {
      "epoch": 18.86853685368537,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.000291587683609581,
      "loss": 0.2308,
      "num_input_tokens_seen": 36196576,
      "step": 171515
    },
    {
      "epoch": 18.869086908690868,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0002913051932665117,
      "loss": 0.2319,
      "num_input_tokens_seen": 36197600,
      "step": 171520
    },
    {
      "epoch": 18.86963696369637,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0002910228384864233,
      "loss": 0.2314,
      "num_input_tokens_seen": 36198720,
      "step": 171525
    },
    {
      "epoch": 18.87018701870187,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0002907406192719203,
      "loss": 0.2329,
      "num_input_tokens_seen": 36199776,
      "step": 171530
    },
    {
      "epoch": 18.87073707370737,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0002904585356256006,
      "loss": 0.2319,
      "num_input_tokens_seen": 36200864,
      "step": 171535
    },
    {
      "epoch": 18.871287128712872,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.00029017658755006894,
      "loss": 0.2303,
      "num_input_tokens_seen": 36201984,
      "step": 171540
    },
    {
      "epoch": 18.871837183718373,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0002898947750479197,
      "loss": 0.2319,
      "num_input_tokens_seen": 36203008,
      "step": 171545
    },
    {
      "epoch": 18.87238723872387,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0002896130981217543,
      "loss": 0.2335,
      "num_input_tokens_seen": 36204064,
      "step": 171550
    },
    {
      "epoch": 18.872937293729372,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0002893315567741622,
      "loss": 0.2309,
      "num_input_tokens_seen": 36205056,
      "step": 171555
    },
    {
      "epoch": 18.873487348734873,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.0002890501510077448,
      "loss": 0.2293,
      "num_input_tokens_seen": 36206112,
      "step": 171560
    },
    {
      "epoch": 18.874037403740374,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0002887688808250932,
      "loss": 0.2303,
      "num_input_tokens_seen": 36207072,
      "step": 171565
    },
    {
      "epoch": 18.874587458745875,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0002884877462287988,
      "loss": 0.2283,
      "num_input_tokens_seen": 36208160,
      "step": 171570
    },
    {
      "epoch": 18.875137513751376,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.0002882067472214544,
      "loss": 0.2329,
      "num_input_tokens_seen": 36209120,
      "step": 171575
    },
    {
      "epoch": 18.875687568756877,
      "grad_norm": 0.000972747802734375,
      "learning_rate": 0.0002879258838056498,
      "loss": 0.2329,
      "num_input_tokens_seen": 36210112,
      "step": 171580
    },
    {
      "epoch": 18.876237623762375,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.00028764515598396934,
      "loss": 0.2314,
      "num_input_tokens_seen": 36211232,
      "step": 171585
    },
    {
      "epoch": 18.876787678767876,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.0002873645637590061,
      "loss": 0.2314,
      "num_input_tokens_seen": 36212320,
      "step": 171590
    },
    {
      "epoch": 18.877337733773377,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0002870841071333446,
      "loss": 0.2308,
      "num_input_tokens_seen": 36213440,
      "step": 171595
    },
    {
      "epoch": 18.877887788778878,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00028680378610956793,
      "loss": 0.2314,
      "num_input_tokens_seen": 36214464,
      "step": 171600
    },
    {
      "epoch": 18.87843784378438,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 0.0002865236006902605,
      "loss": 0.2293,
      "num_input_tokens_seen": 36215520,
      "step": 171605
    },
    {
      "epoch": 18.87898789878988,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.0002862435508780053,
      "loss": 0.2308,
      "num_input_tokens_seen": 36216576,
      "step": 171610
    },
    {
      "epoch": 18.879537953795378,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00028596363667538204,
      "loss": 0.2308,
      "num_input_tokens_seen": 36217664,
      "step": 171615
    },
    {
      "epoch": 18.88008800880088,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.00028568385808497353,
      "loss": 0.2314,
      "num_input_tokens_seen": 36218720,
      "step": 171620
    },
    {
      "epoch": 18.88063806380638,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.00028540421510935444,
      "loss": 0.2314,
      "num_input_tokens_seen": 36219840,
      "step": 171625
    },
    {
      "epoch": 18.88118811881188,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0002851247077511043,
      "loss": 0.2308,
      "num_input_tokens_seen": 36220896,
      "step": 171630
    },
    {
      "epoch": 18.881738173817382,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0002848453360127995,
      "loss": 0.2288,
      "num_input_tokens_seen": 36221952,
      "step": 171635
    },
    {
      "epoch": 18.882288228822883,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00028456609989701285,
      "loss": 0.2324,
      "num_input_tokens_seen": 36222912,
      "step": 171640
    },
    {
      "epoch": 18.882838283828384,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00028428699940632074,
      "loss": 0.2314,
      "num_input_tokens_seen": 36223968,
      "step": 171645
    },
    {
      "epoch": 18.883388338833882,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.0002840080345432927,
      "loss": 0.2303,
      "num_input_tokens_seen": 36225024,
      "step": 171650
    },
    {
      "epoch": 18.883938393839383,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0002837292053105034,
      "loss": 0.2314,
      "num_input_tokens_seen": 36226048,
      "step": 171655
    },
    {
      "epoch": 18.884488448844884,
      "grad_norm": 0.000736236572265625,
      "learning_rate": 0.00028345051171051736,
      "loss": 0.2314,
      "num_input_tokens_seen": 36227072,
      "step": 171660
    },
    {
      "epoch": 18.885038503850385,
      "grad_norm": 0.0031280517578125,
      "learning_rate": 0.0002831719537459093,
      "loss": 0.2329,
      "num_input_tokens_seen": 36228160,
      "step": 171665
    },
    {
      "epoch": 18.885588558855886,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0002828935314192421,
      "loss": 0.2309,
      "num_input_tokens_seen": 36229216,
      "step": 171670
    },
    {
      "epoch": 18.886138613861387,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00028261524473308206,
      "loss": 0.2303,
      "num_input_tokens_seen": 36230304,
      "step": 171675
    },
    {
      "epoch": 18.88668866886689,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.00028233709368999713,
      "loss": 0.2303,
      "num_input_tokens_seen": 36231328,
      "step": 171680
    },
    {
      "epoch": 18.887238723872386,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0002820590782925486,
      "loss": 0.2345,
      "num_input_tokens_seen": 36232416,
      "step": 171685
    },
    {
      "epoch": 18.887788778877887,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0002817811985432994,
      "loss": 0.2308,
      "num_input_tokens_seen": 36233440,
      "step": 171690
    },
    {
      "epoch": 18.888338833883388,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00028150345444481083,
      "loss": 0.2324,
      "num_input_tokens_seen": 36234432,
      "step": 171695
    },
    {
      "epoch": 18.88888888888889,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0002812258459996392,
      "loss": 0.2293,
      "num_input_tokens_seen": 36235520,
      "step": 171700
    },
    {
      "epoch": 18.88943894389439,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.000280948373210349,
      "loss": 0.2314,
      "num_input_tokens_seen": 36236608,
      "step": 171705
    },
    {
      "epoch": 18.88998899889989,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.00028067103607949505,
      "loss": 0.2324,
      "num_input_tokens_seen": 36237632,
      "step": 171710
    },
    {
      "epoch": 18.89053905390539,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0002803938346096318,
      "loss": 0.2324,
      "num_input_tokens_seen": 36238688,
      "step": 171715
    },
    {
      "epoch": 18.89108910891089,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.00028011676880331735,
      "loss": 0.2288,
      "num_input_tokens_seen": 36239744,
      "step": 171720
    },
    {
      "epoch": 18.89163916391639,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.00027983983866310125,
      "loss": 0.2303,
      "num_input_tokens_seen": 36240800,
      "step": 171725
    },
    {
      "epoch": 18.892189218921892,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0002795630441915381,
      "loss": 0.2319,
      "num_input_tokens_seen": 36241824,
      "step": 171730
    },
    {
      "epoch": 18.892739273927393,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0002792863853911792,
      "loss": 0.2303,
      "num_input_tokens_seen": 36242848,
      "step": 171735
    },
    {
      "epoch": 18.893289328932894,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00027900986226457423,
      "loss": 0.2329,
      "num_input_tokens_seen": 36243904,
      "step": 171740
    },
    {
      "epoch": 18.893839383938392,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 0.0002787334748142728,
      "loss": 0.2324,
      "num_input_tokens_seen": 36244928,
      "step": 171745
    },
    {
      "epoch": 18.894389438943893,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.00027845722304281785,
      "loss": 0.2298,
      "num_input_tokens_seen": 36245984,
      "step": 171750
    },
    {
      "epoch": 18.894939493949394,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0002781811069527623,
      "loss": 0.2314,
      "num_input_tokens_seen": 36247040,
      "step": 171755
    },
    {
      "epoch": 18.895489548954895,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 0.0002779051265466426,
      "loss": 0.2309,
      "num_input_tokens_seen": 36248032,
      "step": 171760
    },
    {
      "epoch": 18.896039603960396,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0002776292818270115,
      "loss": 0.2303,
      "num_input_tokens_seen": 36249088,
      "step": 171765
    },
    {
      "epoch": 18.896589658965897,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00027735357279640545,
      "loss": 0.2329,
      "num_input_tokens_seen": 36250112,
      "step": 171770
    },
    {
      "epoch": 18.8971397139714,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00027707799945736575,
      "loss": 0.2308,
      "num_input_tokens_seen": 36251168,
      "step": 171775
    },
    {
      "epoch": 18.897689768976896,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0002768025618124353,
      "loss": 0.2319,
      "num_input_tokens_seen": 36252224,
      "step": 171780
    },
    {
      "epoch": 18.898239823982397,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.0002765272598641488,
      "loss": 0.2309,
      "num_input_tokens_seen": 36253280,
      "step": 171785
    },
    {
      "epoch": 18.8987898789879,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.0002762520936150459,
      "loss": 0.2319,
      "num_input_tokens_seen": 36254368,
      "step": 171790
    },
    {
      "epoch": 18.8993399339934,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00027597706306766275,
      "loss": 0.2329,
      "num_input_tokens_seen": 36255392,
      "step": 171795
    },
    {
      "epoch": 18.8998899889989,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0002757021682245342,
      "loss": 0.2319,
      "num_input_tokens_seen": 36256352,
      "step": 171800
    },
    {
      "epoch": 18.9004400440044,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0002754274090881931,
      "loss": 0.2308,
      "num_input_tokens_seen": 36257440,
      "step": 171805
    },
    {
      "epoch": 18.900990099009903,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0002751527856611724,
      "loss": 0.2314,
      "num_input_tokens_seen": 36258560,
      "step": 171810
    },
    {
      "epoch": 18.9015401540154,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0002748782979460018,
      "loss": 0.2314,
      "num_input_tokens_seen": 36259648,
      "step": 171815
    },
    {
      "epoch": 18.9020902090209,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0002746039459452126,
      "loss": 0.2319,
      "num_input_tokens_seen": 36260704,
      "step": 171820
    },
    {
      "epoch": 18.902640264026402,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00027432972966133116,
      "loss": 0.2314,
      "num_input_tokens_seen": 36261728,
      "step": 171825
    },
    {
      "epoch": 18.903190319031903,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0002740556490968887,
      "loss": 0.2293,
      "num_input_tokens_seen": 36262784,
      "step": 171830
    },
    {
      "epoch": 18.903740374037405,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 0.00027378170425440996,
      "loss": 0.2324,
      "num_input_tokens_seen": 36263776,
      "step": 171835
    },
    {
      "epoch": 18.904290429042906,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0002735078951364178,
      "loss": 0.2309,
      "num_input_tokens_seen": 36264864,
      "step": 171840
    },
    {
      "epoch": 18.904840484048403,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0002732342217454353,
      "loss": 0.2314,
      "num_input_tokens_seen": 36265920,
      "step": 171845
    },
    {
      "epoch": 18.905390539053904,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0002729606840839871,
      "loss": 0.2303,
      "num_input_tokens_seen": 36266912,
      "step": 171850
    },
    {
      "epoch": 18.905940594059405,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.00027268728215459616,
      "loss": 0.2309,
      "num_input_tokens_seen": 36267968,
      "step": 171855
    },
    {
      "epoch": 18.906490649064907,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00027241401595977876,
      "loss": 0.2314,
      "num_input_tokens_seen": 36269056,
      "step": 171860
    },
    {
      "epoch": 18.907040704070408,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.00027214088550205296,
      "loss": 0.2309,
      "num_input_tokens_seen": 36270080,
      "step": 171865
    },
    {
      "epoch": 18.90759075907591,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.00027186789078393833,
      "loss": 0.2329,
      "num_input_tokens_seen": 36271168,
      "step": 171870
    },
    {
      "epoch": 18.90814081408141,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00027159503180794963,
      "loss": 0.2319,
      "num_input_tokens_seen": 36272192,
      "step": 171875
    },
    {
      "epoch": 18.908690869086907,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00027132230857660145,
      "loss": 0.2303,
      "num_input_tokens_seen": 36273216,
      "step": 171880
    },
    {
      "epoch": 18.90924092409241,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.0002710497210924084,
      "loss": 0.2303,
      "num_input_tokens_seen": 36274304,
      "step": 171885
    },
    {
      "epoch": 18.90979097909791,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.0002707772693578836,
      "loss": 0.2314,
      "num_input_tokens_seen": 36275360,
      "step": 171890
    },
    {
      "epoch": 18.91034103410341,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0002705049533755366,
      "loss": 0.2329,
      "num_input_tokens_seen": 36276416,
      "step": 171895
    },
    {
      "epoch": 18.91089108910891,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00027023277314787707,
      "loss": 0.2324,
      "num_input_tokens_seen": 36277408,
      "step": 171900
    },
    {
      "epoch": 18.911441144114413,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00026996072867741303,
      "loss": 0.2314,
      "num_input_tokens_seen": 36278496,
      "step": 171905
    },
    {
      "epoch": 18.91199119911991,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00026968881996665417,
      "loss": 0.2324,
      "num_input_tokens_seen": 36279552,
      "step": 171910
    },
    {
      "epoch": 18.91254125412541,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0002694170470181051,
      "loss": 0.2309,
      "num_input_tokens_seen": 36280640,
      "step": 171915
    },
    {
      "epoch": 18.913091309130913,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.00026914540983427046,
      "loss": 0.2298,
      "num_input_tokens_seen": 36281696,
      "step": 171920
    },
    {
      "epoch": 18.913641364136414,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.000268873908417655,
      "loss": 0.2303,
      "num_input_tokens_seen": 36282688,
      "step": 171925
    },
    {
      "epoch": 18.914191419141915,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.00026860254277075835,
      "loss": 0.2314,
      "num_input_tokens_seen": 36283680,
      "step": 171930
    },
    {
      "epoch": 18.914741474147416,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.00026833131289608346,
      "loss": 0.2308,
      "num_input_tokens_seen": 36284736,
      "step": 171935
    },
    {
      "epoch": 18.915291529152917,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.00026806021879613006,
      "loss": 0.2324,
      "num_input_tokens_seen": 36285760,
      "step": 171940
    },
    {
      "epoch": 18.915841584158414,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 0.0002677892604733978,
      "loss": 0.2324,
      "num_input_tokens_seen": 36286784,
      "step": 171945
    },
    {
      "epoch": 18.916391639163916,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0002675184379303813,
      "loss": 0.2314,
      "num_input_tokens_seen": 36287840,
      "step": 171950
    },
    {
      "epoch": 18.916941694169417,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00026724775116958034,
      "loss": 0.2314,
      "num_input_tokens_seen": 36288896,
      "step": 171955
    },
    {
      "epoch": 18.917491749174918,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00026697720019348446,
      "loss": 0.2324,
      "num_input_tokens_seen": 36289952,
      "step": 171960
    },
    {
      "epoch": 18.91804180418042,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.0002667067850045918,
      "loss": 0.2319,
      "num_input_tokens_seen": 36291040,
      "step": 171965
    },
    {
      "epoch": 18.91859185918592,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00026643650560539186,
      "loss": 0.2313,
      "num_input_tokens_seen": 36292096,
      "step": 171970
    },
    {
      "epoch": 18.919141914191417,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0002661663619983778,
      "loss": 0.2314,
      "num_input_tokens_seen": 36293184,
      "step": 171975
    },
    {
      "epoch": 18.91969196919692,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0002658963541860393,
      "loss": 0.2324,
      "num_input_tokens_seen": 36294208,
      "step": 171980
    },
    {
      "epoch": 18.92024202420242,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00026562648217086257,
      "loss": 0.2303,
      "num_input_tokens_seen": 36295200,
      "step": 171985
    },
    {
      "epoch": 18.92079207920792,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.00026535674595533576,
      "loss": 0.2314,
      "num_input_tokens_seen": 36296256,
      "step": 171990
    },
    {
      "epoch": 18.921342134213422,
      "grad_norm": 0.000732421875,
      "learning_rate": 0.00026508714554194677,
      "loss": 0.2319,
      "num_input_tokens_seen": 36297376,
      "step": 171995
    },
    {
      "epoch": 18.921892189218923,
      "grad_norm": 0.001434326171875,
      "learning_rate": 0.0002648176809331787,
      "loss": 0.2309,
      "num_input_tokens_seen": 36298432,
      "step": 172000
    },
    {
      "epoch": 18.922442244224424,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0002645483521315145,
      "loss": 0.2303,
      "num_input_tokens_seen": 36299488,
      "step": 172005
    },
    {
      "epoch": 18.92299229922992,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.0002642791591394389,
      "loss": 0.2313,
      "num_input_tokens_seen": 36300576,
      "step": 172010
    },
    {
      "epoch": 18.923542354235423,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.00026401010195942985,
      "loss": 0.2308,
      "num_input_tokens_seen": 36301632,
      "step": 172015
    },
    {
      "epoch": 18.924092409240924,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00026374118059396877,
      "loss": 0.2319,
      "num_input_tokens_seen": 36302688,
      "step": 172020
    },
    {
      "epoch": 18.924642464246425,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.0002634723950455353,
      "loss": 0.2303,
      "num_input_tokens_seen": 36303808,
      "step": 172025
    },
    {
      "epoch": 18.925192519251926,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00026320374531660584,
      "loss": 0.2319,
      "num_input_tokens_seen": 36304832,
      "step": 172030
    },
    {
      "epoch": 18.925742574257427,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.000262935231409655,
      "loss": 0.2308,
      "num_input_tokens_seen": 36305856,
      "step": 172035
    },
    {
      "epoch": 18.926292629262925,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0002626668533271592,
      "loss": 0.2314,
      "num_input_tokens_seen": 36306880,
      "step": 172040
    },
    {
      "epoch": 18.926842684268426,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.00026239861107159144,
      "loss": 0.2314,
      "num_input_tokens_seen": 36307968,
      "step": 172045
    },
    {
      "epoch": 18.927392739273927,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00026213050464542306,
      "loss": 0.2314,
      "num_input_tokens_seen": 36309056,
      "step": 172050
    },
    {
      "epoch": 18.927942794279428,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00026186253405112545,
      "loss": 0.2324,
      "num_input_tokens_seen": 36310080,
      "step": 172055
    },
    {
      "epoch": 18.92849284928493,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00026159469929117,
      "loss": 0.2308,
      "num_input_tokens_seen": 36311136,
      "step": 172060
    },
    {
      "epoch": 18.92904290429043,
      "grad_norm": 0.00080108642578125,
      "learning_rate": 0.0002613270003680229,
      "loss": 0.2324,
      "num_input_tokens_seen": 36312192,
      "step": 172065
    },
    {
      "epoch": 18.92959295929593,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00026105943728415405,
      "loss": 0.2324,
      "num_input_tokens_seen": 36313216,
      "step": 172070
    },
    {
      "epoch": 18.93014301430143,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.0002607920100420263,
      "loss": 0.2309,
      "num_input_tokens_seen": 36314208,
      "step": 172075
    },
    {
      "epoch": 18.93069306930693,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0002605247186441062,
      "loss": 0.2308,
      "num_input_tokens_seen": 36315296,
      "step": 172080
    },
    {
      "epoch": 18.93124312431243,
      "grad_norm": 0.0028076171875,
      "learning_rate": 0.00026025756309285827,
      "loss": 0.233,
      "num_input_tokens_seen": 36316352,
      "step": 172085
    },
    {
      "epoch": 18.931793179317932,
      "grad_norm": 0.0052490234375,
      "learning_rate": 0.0002599905433907423,
      "loss": 0.2293,
      "num_input_tokens_seen": 36317376,
      "step": 172090
    },
    {
      "epoch": 18.932343234323433,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.000259723659540223,
      "loss": 0.2309,
      "num_input_tokens_seen": 36318432,
      "step": 172095
    },
    {
      "epoch": 18.932893289328934,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0002594569115437567,
      "loss": 0.2314,
      "num_input_tokens_seen": 36319488,
      "step": 172100
    },
    {
      "epoch": 18.933443344334435,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.00025919029940380146,
      "loss": 0.2319,
      "num_input_tokens_seen": 36320544,
      "step": 172105
    },
    {
      "epoch": 18.933993399339933,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00025892382312281523,
      "loss": 0.2324,
      "num_input_tokens_seen": 36321536,
      "step": 172110
    },
    {
      "epoch": 18.934543454345434,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.00025865748270325614,
      "loss": 0.2303,
      "num_input_tokens_seen": 36322560,
      "step": 172115
    },
    {
      "epoch": 18.935093509350935,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0002583912781475772,
      "loss": 0.2303,
      "num_input_tokens_seen": 36323680,
      "step": 172120
    },
    {
      "epoch": 18.935643564356436,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00025812520945823315,
      "loss": 0.2324,
      "num_input_tokens_seen": 36324736,
      "step": 172125
    },
    {
      "epoch": 18.936193619361937,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0002578592766376736,
      "loss": 0.2298,
      "num_input_tokens_seen": 36325760,
      "step": 172130
    },
    {
      "epoch": 18.936743674367438,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0002575934796883517,
      "loss": 0.2309,
      "num_input_tokens_seen": 36326816,
      "step": 172135
    },
    {
      "epoch": 18.937293729372936,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.000257327818612717,
      "loss": 0.2319,
      "num_input_tokens_seen": 36327904,
      "step": 172140
    },
    {
      "epoch": 18.937843784378437,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0002570622934132177,
      "loss": 0.2293,
      "num_input_tokens_seen": 36328960,
      "step": 172145
    },
    {
      "epoch": 18.938393839383938,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.00025679690409230016,
      "loss": 0.2319,
      "num_input_tokens_seen": 36329952,
      "step": 172150
    },
    {
      "epoch": 18.93894389438944,
      "grad_norm": 0.0018463134765625,
      "learning_rate": 0.0002565316506524107,
      "loss": 0.2329,
      "num_input_tokens_seen": 36330976,
      "step": 172155
    },
    {
      "epoch": 18.93949394939494,
      "grad_norm": 0.000946044921875,
      "learning_rate": 0.0002562665330959973,
      "loss": 0.2309,
      "num_input_tokens_seen": 36332032,
      "step": 172160
    },
    {
      "epoch": 18.94004400440044,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0002560015514254965,
      "loss": 0.2303,
      "num_input_tokens_seen": 36333024,
      "step": 172165
    },
    {
      "epoch": 18.94059405940594,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.00025573670564335625,
      "loss": 0.2314,
      "num_input_tokens_seen": 36334016,
      "step": 172170
    },
    {
      "epoch": 18.94114411441144,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.0002554719957520163,
      "loss": 0.2319,
      "num_input_tokens_seen": 36335040,
      "step": 172175
    },
    {
      "epoch": 18.94169416941694,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00025520742175391464,
      "loss": 0.2314,
      "num_input_tokens_seen": 36336064,
      "step": 172180
    },
    {
      "epoch": 18.942244224422442,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0002549429836514927,
      "loss": 0.2324,
      "num_input_tokens_seen": 36337088,
      "step": 172185
    },
    {
      "epoch": 18.942794279427943,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0002546786814471852,
      "loss": 0.2308,
      "num_input_tokens_seen": 36338144,
      "step": 172190
    },
    {
      "epoch": 18.943344334433444,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.00025441451514342846,
      "loss": 0.2303,
      "num_input_tokens_seen": 36339136,
      "step": 172195
    },
    {
      "epoch": 18.943894389438945,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.00025415048474265733,
      "loss": 0.2319,
      "num_input_tokens_seen": 36340224,
      "step": 172200
    },
    {
      "epoch": 18.944444444444443,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0002538865902473064,
      "loss": 0.2304,
      "num_input_tokens_seen": 36341280,
      "step": 172205
    },
    {
      "epoch": 18.944994499449944,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.0002536228316598055,
      "loss": 0.2298,
      "num_input_tokens_seen": 36342368,
      "step": 172210
    },
    {
      "epoch": 18.945544554455445,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00025335920898258925,
      "loss": 0.2319,
      "num_input_tokens_seen": 36343392,
      "step": 172215
    },
    {
      "epoch": 18.946094609460946,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 0.0002530957222180824,
      "loss": 0.2319,
      "num_input_tokens_seen": 36344416,
      "step": 172220
    },
    {
      "epoch": 18.946644664466447,
      "grad_norm": 0.000774383544921875,
      "learning_rate": 0.00025283237136871636,
      "loss": 0.2314,
      "num_input_tokens_seen": 36345408,
      "step": 172225
    },
    {
      "epoch": 18.94719471947195,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 0.0002525691564369192,
      "loss": 0.2314,
      "num_input_tokens_seen": 36346464,
      "step": 172230
    },
    {
      "epoch": 18.94774477447745,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.0002523060774251157,
      "loss": 0.2314,
      "num_input_tokens_seen": 36347488,
      "step": 172235
    },
    {
      "epoch": 18.948294829482947,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00025204313433573045,
      "loss": 0.2314,
      "num_input_tokens_seen": 36348480,
      "step": 172240
    },
    {
      "epoch": 18.948844884488448,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.0002517803271711866,
      "loss": 0.2319,
      "num_input_tokens_seen": 36349504,
      "step": 172245
    },
    {
      "epoch": 18.94939493949395,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.0002515176559339055,
      "loss": 0.2319,
      "num_input_tokens_seen": 36350624,
      "step": 172250
    },
    {
      "epoch": 18.94994499449945,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.0002512551206263086,
      "loss": 0.2308,
      "num_input_tokens_seen": 36351680,
      "step": 172255
    },
    {
      "epoch": 18.95049504950495,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00025099272125081896,
      "loss": 0.2303,
      "num_input_tokens_seen": 36352768,
      "step": 172260
    },
    {
      "epoch": 18.951045104510452,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.00025073045780985123,
      "loss": 0.2314,
      "num_input_tokens_seen": 36353760,
      "step": 172265
    },
    {
      "epoch": 18.95159515951595,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.0002504683303058219,
      "loss": 0.2303,
      "num_input_tokens_seen": 36354880,
      "step": 172270
    },
    {
      "epoch": 18.95214521452145,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.00025020633874115074,
      "loss": 0.2308,
      "num_input_tokens_seen": 36355904,
      "step": 172275
    },
    {
      "epoch": 18.952695269526952,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00024994448311824733,
      "loss": 0.2335,
      "num_input_tokens_seen": 36356928,
      "step": 172280
    },
    {
      "epoch": 18.953245324532453,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0002496827634395299,
      "loss": 0.2319,
      "num_input_tokens_seen": 36357952,
      "step": 172285
    },
    {
      "epoch": 18.953795379537954,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.0002494211797074081,
      "loss": 0.2308,
      "num_input_tokens_seen": 36359040,
      "step": 172290
    },
    {
      "epoch": 18.954345434543455,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0002491597319242933,
      "loss": 0.2314,
      "num_input_tokens_seen": 36360096,
      "step": 172295
    },
    {
      "epoch": 18.954895489548957,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.0002488984200925953,
      "loss": 0.2314,
      "num_input_tokens_seen": 36361216,
      "step": 172300
    },
    {
      "epoch": 18.955445544554454,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.0002486372442147222,
      "loss": 0.2303,
      "num_input_tokens_seen": 36362336,
      "step": 172305
    },
    {
      "epoch": 18.955995599559955,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00024837620429308036,
      "loss": 0.2314,
      "num_input_tokens_seen": 36363424,
      "step": 172310
    },
    {
      "epoch": 18.956545654565456,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.0002481153003300779,
      "loss": 0.2303,
      "num_input_tokens_seen": 36364416,
      "step": 172315
    },
    {
      "epoch": 18.957095709570957,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.00024785453232811615,
      "loss": 0.2309,
      "num_input_tokens_seen": 36365504,
      "step": 172320
    },
    {
      "epoch": 18.95764576457646,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.00024759390028960323,
      "loss": 0.2324,
      "num_input_tokens_seen": 36366592,
      "step": 172325
    },
    {
      "epoch": 18.95819581958196,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0002473334042169356,
      "loss": 0.2303,
      "num_input_tokens_seen": 36367680,
      "step": 172330
    },
    {
      "epoch": 18.958745874587457,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.0002470730441125196,
      "loss": 0.2319,
      "num_input_tokens_seen": 36368768,
      "step": 172335
    },
    {
      "epoch": 18.959295929592958,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00024681281997875006,
      "loss": 0.2308,
      "num_input_tokens_seen": 36369760,
      "step": 172340
    },
    {
      "epoch": 18.95984598459846,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.00024655273181802825,
      "loss": 0.2303,
      "num_input_tokens_seen": 36370816,
      "step": 172345
    },
    {
      "epoch": 18.96039603960396,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.0002462927796327507,
      "loss": 0.2314,
      "num_input_tokens_seen": 36371840,
      "step": 172350
    },
    {
      "epoch": 18.96094609460946,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0002460329634253122,
      "loss": 0.2319,
      "num_input_tokens_seen": 36372896,
      "step": 172355
    },
    {
      "epoch": 18.961496149614963,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0002457732831981091,
      "loss": 0.2324,
      "num_input_tokens_seen": 36373888,
      "step": 172360
    },
    {
      "epoch": 18.962046204620464,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0002455137389535344,
      "loss": 0.2314,
      "num_input_tokens_seen": 36374976,
      "step": 172365
    },
    {
      "epoch": 18.96259625962596,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.00024525433069397804,
      "loss": 0.2293,
      "num_input_tokens_seen": 36376000,
      "step": 172370
    },
    {
      "epoch": 18.963146314631462,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.00024499505842183467,
      "loss": 0.2303,
      "num_input_tokens_seen": 36376992,
      "step": 172375
    },
    {
      "epoch": 18.963696369636963,
      "grad_norm": 0.002685546875,
      "learning_rate": 0.0002447359221394907,
      "loss": 0.2324,
      "num_input_tokens_seen": 36378048,
      "step": 172380
    },
    {
      "epoch": 18.964246424642464,
      "grad_norm": 0.000804901123046875,
      "learning_rate": 0.0002444769218493342,
      "loss": 0.2288,
      "num_input_tokens_seen": 36379040,
      "step": 172385
    },
    {
      "epoch": 18.964796479647966,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00024421805755375504,
      "loss": 0.2324,
      "num_input_tokens_seen": 36380096,
      "step": 172390
    },
    {
      "epoch": 18.965346534653467,
      "grad_norm": 0.00091552734375,
      "learning_rate": 0.00024395932925513618,
      "loss": 0.2319,
      "num_input_tokens_seen": 36381120,
      "step": 172395
    },
    {
      "epoch": 18.965896589658964,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00024370073695586413,
      "loss": 0.2329,
      "num_input_tokens_seen": 36382208,
      "step": 172400
    },
    {
      "epoch": 18.966446644664465,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.0002434422806583236,
      "loss": 0.2308,
      "num_input_tokens_seen": 36383264,
      "step": 172405
    },
    {
      "epoch": 18.966996699669966,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00024318396036489276,
      "loss": 0.2314,
      "num_input_tokens_seen": 36384352,
      "step": 172410
    },
    {
      "epoch": 18.967546754675467,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.00024292577607795462,
      "loss": 0.234,
      "num_input_tokens_seen": 36385376,
      "step": 172415
    },
    {
      "epoch": 18.96809680968097,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00024266772779988732,
      "loss": 0.2303,
      "num_input_tokens_seen": 36386432,
      "step": 172420
    },
    {
      "epoch": 18.96864686468647,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00024240981553307227,
      "loss": 0.2308,
      "num_input_tokens_seen": 36387520,
      "step": 172425
    },
    {
      "epoch": 18.96919691969197,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 0.00024215203927988426,
      "loss": 0.2314,
      "num_input_tokens_seen": 36388640,
      "step": 172430
    },
    {
      "epoch": 18.96974697469747,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00024189439904269804,
      "loss": 0.2309,
      "num_input_tokens_seen": 36389664,
      "step": 172435
    },
    {
      "epoch": 18.97029702970297,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 0.0002416368948238917,
      "loss": 0.2314,
      "num_input_tokens_seen": 36390752,
      "step": 172440
    },
    {
      "epoch": 18.97084708470847,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0002413795266258334,
      "loss": 0.233,
      "num_input_tokens_seen": 36391808,
      "step": 172445
    },
    {
      "epoch": 18.97139713971397,
      "grad_norm": 0.01068115234375,
      "learning_rate": 0.00024112229445090116,
      "loss": 0.2314,
      "num_input_tokens_seen": 36392928,
      "step": 172450
    },
    {
      "epoch": 18.971947194719473,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.0002408651983014598,
      "loss": 0.2319,
      "num_input_tokens_seen": 36393984,
      "step": 172455
    },
    {
      "epoch": 18.972497249724974,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00024060823817988408,
      "loss": 0.2309,
      "num_input_tokens_seen": 36394976,
      "step": 172460
    },
    {
      "epoch": 18.97304730473047,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00024035141408853877,
      "loss": 0.2319,
      "num_input_tokens_seen": 36396096,
      "step": 172465
    },
    {
      "epoch": 18.973597359735972,
      "grad_norm": 0.00075531005859375,
      "learning_rate": 0.00024009472602979197,
      "loss": 0.2303,
      "num_input_tokens_seen": 36397120,
      "step": 172470
    },
    {
      "epoch": 18.974147414741473,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0002398381740060118,
      "loss": 0.2319,
      "num_input_tokens_seen": 36398272,
      "step": 172475
    },
    {
      "epoch": 18.974697469746975,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00023958175801955971,
      "loss": 0.2309,
      "num_input_tokens_seen": 36399296,
      "step": 172480
    },
    {
      "epoch": 18.975247524752476,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00023932547807279712,
      "loss": 0.2303,
      "num_input_tokens_seen": 36400288,
      "step": 172485
    },
    {
      "epoch": 18.975797579757977,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 0.0002390693341680905,
      "loss": 0.2309,
      "num_input_tokens_seen": 36401312,
      "step": 172490
    },
    {
      "epoch": 18.976347634763478,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0002388133263077996,
      "loss": 0.2314,
      "num_input_tokens_seen": 36402336,
      "step": 172495
    },
    {
      "epoch": 18.976897689768975,
      "grad_norm": 0.0029296875,
      "learning_rate": 0.00023855745449428254,
      "loss": 0.2319,
      "num_input_tokens_seen": 36403456,
      "step": 172500
    },
    {
      "epoch": 18.977447744774476,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00023830171872989912,
      "loss": 0.2314,
      "num_input_tokens_seen": 36404544,
      "step": 172505
    },
    {
      "epoch": 18.977997799779978,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0002380461190170041,
      "loss": 0.2308,
      "num_input_tokens_seen": 36405632,
      "step": 172510
    },
    {
      "epoch": 18.97854785478548,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.00023779065535795395,
      "loss": 0.2319,
      "num_input_tokens_seen": 36406656,
      "step": 172515
    },
    {
      "epoch": 18.97909790979098,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0002375353277551051,
      "loss": 0.2329,
      "num_input_tokens_seen": 36407680,
      "step": 172520
    },
    {
      "epoch": 18.97964796479648,
      "grad_norm": 0.005035400390625,
      "learning_rate": 0.000237280136210809,
      "loss": 0.2314,
      "num_input_tokens_seen": 36408672,
      "step": 172525
    },
    {
      "epoch": 18.980198019801982,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00023702508072741878,
      "loss": 0.2314,
      "num_input_tokens_seen": 36409760,
      "step": 172530
    },
    {
      "epoch": 18.98074807480748,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00023677016130728423,
      "loss": 0.2314,
      "num_input_tokens_seen": 36410816,
      "step": 172535
    },
    {
      "epoch": 18.98129812981298,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 0.0002365153779527551,
      "loss": 0.2288,
      "num_input_tokens_seen": 36411840,
      "step": 172540
    },
    {
      "epoch": 18.98184818481848,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00023626073066617624,
      "loss": 0.2319,
      "num_input_tokens_seen": 36412928,
      "step": 172545
    },
    {
      "epoch": 18.982398239823983,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0002360062194499024,
      "loss": 0.2298,
      "num_input_tokens_seen": 36413952,
      "step": 172550
    },
    {
      "epoch": 18.982948294829484,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0002357518443062734,
      "loss": 0.2309,
      "num_input_tokens_seen": 36415008,
      "step": 172555
    },
    {
      "epoch": 18.983498349834985,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00023549760523763395,
      "loss": 0.2319,
      "num_input_tokens_seen": 36416032,
      "step": 172560
    },
    {
      "epoch": 18.984048404840483,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00023524350224632894,
      "loss": 0.2314,
      "num_input_tokens_seen": 36417088,
      "step": 172565
    },
    {
      "epoch": 18.984598459845984,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00023498953533469978,
      "loss": 0.2314,
      "num_input_tokens_seen": 36418144,
      "step": 172570
    },
    {
      "epoch": 18.985148514851485,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.0002347357045050863,
      "loss": 0.2335,
      "num_input_tokens_seen": 36419232,
      "step": 172575
    },
    {
      "epoch": 18.985698569856986,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00023448200975983156,
      "loss": 0.2329,
      "num_input_tokens_seen": 36420256,
      "step": 172580
    },
    {
      "epoch": 18.986248624862487,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0002342284511012671,
      "loss": 0.2335,
      "num_input_tokens_seen": 36421344,
      "step": 172585
    },
    {
      "epoch": 18.986798679867988,
      "grad_norm": 0.0031585693359375,
      "learning_rate": 0.00023397502853173766,
      "loss": 0.2303,
      "num_input_tokens_seen": 36422400,
      "step": 172590
    },
    {
      "epoch": 18.98734873487349,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0002337217420535731,
      "loss": 0.2299,
      "num_input_tokens_seen": 36423488,
      "step": 172595
    },
    {
      "epoch": 18.987898789878987,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0002334685916691098,
      "loss": 0.2304,
      "num_input_tokens_seen": 36424544,
      "step": 172600
    },
    {
      "epoch": 18.988448844884488,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00023321557738067931,
      "loss": 0.2319,
      "num_input_tokens_seen": 36425600,
      "step": 172605
    },
    {
      "epoch": 18.98899889988999,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.0002329626991906164,
      "loss": 0.2329,
      "num_input_tokens_seen": 36426752,
      "step": 172610
    },
    {
      "epoch": 18.98954895489549,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00023270995710125084,
      "loss": 0.2324,
      "num_input_tokens_seen": 36427808,
      "step": 172615
    },
    {
      "epoch": 18.99009900990099,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00023245735111491084,
      "loss": 0.2308,
      "num_input_tokens_seen": 36428896,
      "step": 172620
    },
    {
      "epoch": 18.990649064906492,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00023220488123392612,
      "loss": 0.2324,
      "num_input_tokens_seen": 36429984,
      "step": 172625
    },
    {
      "epoch": 18.99119911991199,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.00023195254746062155,
      "loss": 0.2319,
      "num_input_tokens_seen": 36431040,
      "step": 172630
    },
    {
      "epoch": 18.99174917491749,
      "grad_norm": 0.001678466796875,
      "learning_rate": 0.00023170034979732356,
      "loss": 0.2314,
      "num_input_tokens_seen": 36432032,
      "step": 172635
    },
    {
      "epoch": 18.992299229922992,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.00023144828824635865,
      "loss": 0.2309,
      "num_input_tokens_seen": 36433024,
      "step": 172640
    },
    {
      "epoch": 18.992849284928493,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00023119636281004828,
      "loss": 0.2308,
      "num_input_tokens_seen": 36434112,
      "step": 172645
    },
    {
      "epoch": 18.993399339933994,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0002309445734907123,
      "loss": 0.2314,
      "num_input_tokens_seen": 36435136,
      "step": 172650
    },
    {
      "epoch": 18.993949394939495,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.00023069292029067543,
      "loss": 0.2288,
      "num_input_tokens_seen": 36436288,
      "step": 172655
    },
    {
      "epoch": 18.994499449944996,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00023044140321225426,
      "loss": 0.2314,
      "num_input_tokens_seen": 36437344,
      "step": 172660
    },
    {
      "epoch": 18.995049504950494,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00023019002225776517,
      "loss": 0.2303,
      "num_input_tokens_seen": 36438336,
      "step": 172665
    },
    {
      "epoch": 18.995599559955995,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00022993877742953138,
      "loss": 0.2324,
      "num_input_tokens_seen": 36439360,
      "step": 172670
    },
    {
      "epoch": 18.996149614961496,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 0.000229687668729861,
      "loss": 0.2335,
      "num_input_tokens_seen": 36440416,
      "step": 172675
    },
    {
      "epoch": 18.996699669966997,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00022943669616107553,
      "loss": 0.2314,
      "num_input_tokens_seen": 36441408,
      "step": 172680
    },
    {
      "epoch": 18.997249724972498,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00022918585972548145,
      "loss": 0.2324,
      "num_input_tokens_seen": 36442496,
      "step": 172685
    },
    {
      "epoch": 18.997799779978,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00022893515942539356,
      "loss": 0.2303,
      "num_input_tokens_seen": 36443552,
      "step": 172690
    },
    {
      "epoch": 18.998349834983497,
      "grad_norm": 0.005859375,
      "learning_rate": 0.000228684595263125,
      "loss": 0.2298,
      "num_input_tokens_seen": 36444608,
      "step": 172695
    },
    {
      "epoch": 18.998899889988998,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00022843416724097897,
      "loss": 0.2314,
      "num_input_tokens_seen": 36445632,
      "step": 172700
    },
    {
      "epoch": 18.9994499449945,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00022818387536127025,
      "loss": 0.2288,
      "num_input_tokens_seen": 36446624,
      "step": 172705
    },
    {
      "epoch": 19.0,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.00022793371962630038,
      "loss": 0.2324,
      "num_input_tokens_seen": 36447600,
      "step": 172710
    },
    {
      "epoch": 19.0,
      "eval_loss": 0.23148144781589508,
      "eval_runtime": 60.5106,
      "eval_samples_per_second": 66.765,
      "eval_steps_per_second": 16.691,
      "num_input_tokens_seen": 36447600,
      "step": 172710
    },
    {
      "epoch": 19.0005500550055,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.0002276837000383791,
      "loss": 0.2293,
      "num_input_tokens_seen": 36448624,
      "step": 172715
    },
    {
      "epoch": 19.001100110011002,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 0.00022743381659980799,
      "loss": 0.2308,
      "num_input_tokens_seen": 36449712,
      "step": 172720
    },
    {
      "epoch": 19.001650165016503,
      "grad_norm": 0.00213623046875,
      "learning_rate": 0.00022718406931289013,
      "loss": 0.2324,
      "num_input_tokens_seen": 36450768,
      "step": 172725
    },
    {
      "epoch": 19.002200220022,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00022693445817992874,
      "loss": 0.2335,
      "num_input_tokens_seen": 36451792,
      "step": 172730
    },
    {
      "epoch": 19.002750275027502,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.0002266849832032236,
      "loss": 0.2308,
      "num_input_tokens_seen": 36452880,
      "step": 172735
    },
    {
      "epoch": 19.003300330033003,
      "grad_norm": 0.006195068359375,
      "learning_rate": 0.00022643564438507458,
      "loss": 0.2314,
      "num_input_tokens_seen": 36453904,
      "step": 172740
    },
    {
      "epoch": 19.003850385038504,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.00022618644172777645,
      "loss": 0.2314,
      "num_input_tokens_seen": 36454928,
      "step": 172745
    },
    {
      "epoch": 19.004400440044005,
      "grad_norm": 0.0064697265625,
      "learning_rate": 0.00022593737523363076,
      "loss": 0.2298,
      "num_input_tokens_seen": 36455920,
      "step": 172750
    },
    {
      "epoch": 19.004950495049506,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00022568844490493232,
      "loss": 0.2319,
      "num_input_tokens_seen": 36457008,
      "step": 172755
    },
    {
      "epoch": 19.005500550055004,
      "grad_norm": 0.006500244140625,
      "learning_rate": 0.00022543965074397265,
      "loss": 0.234,
      "num_input_tokens_seen": 36458096,
      "step": 172760
    },
    {
      "epoch": 19.006050605060505,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00022519099275304655,
      "loss": 0.2293,
      "num_input_tokens_seen": 36459152,
      "step": 172765
    },
    {
      "epoch": 19.006600660066006,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00022494247093444552,
      "loss": 0.2298,
      "num_input_tokens_seen": 36460144,
      "step": 172770
    },
    {
      "epoch": 19.007150715071507,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00022469408529045775,
      "loss": 0.2288,
      "num_input_tokens_seen": 36461200,
      "step": 172775
    },
    {
      "epoch": 19.007700770077008,
      "grad_norm": 0.001556396484375,
      "learning_rate": 0.0002244458358233764,
      "loss": 0.2335,
      "num_input_tokens_seen": 36462288,
      "step": 172780
    },
    {
      "epoch": 19.00825082508251,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.00022419772253548965,
      "loss": 0.2324,
      "num_input_tokens_seen": 36463408,
      "step": 172785
    },
    {
      "epoch": 19.00880088008801,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.00022394974542907896,
      "loss": 0.2324,
      "num_input_tokens_seen": 36464432,
      "step": 172790
    },
    {
      "epoch": 19.009350935093508,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0002237019045064359,
      "loss": 0.2309,
      "num_input_tokens_seen": 36465552,
      "step": 172795
    },
    {
      "epoch": 19.00990099009901,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.00022345419976984026,
      "loss": 0.2324,
      "num_input_tokens_seen": 36466608,
      "step": 172800
    },
    {
      "epoch": 19.01045104510451,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0002232066312215769,
      "loss": 0.2314,
      "num_input_tokens_seen": 36467632,
      "step": 172805
    },
    {
      "epoch": 19.01100110011001,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.000222959198863929,
      "loss": 0.2298,
      "num_input_tokens_seen": 36468656,
      "step": 172810
    },
    {
      "epoch": 19.011551155115512,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00022271190269917472,
      "loss": 0.2314,
      "num_input_tokens_seen": 36469680,
      "step": 172815
    },
    {
      "epoch": 19.012101210121013,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.00022246474272959394,
      "loss": 0.2308,
      "num_input_tokens_seen": 36470736,
      "step": 172820
    },
    {
      "epoch": 19.01265126512651,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00022221771895746477,
      "loss": 0.2298,
      "num_input_tokens_seen": 36471760,
      "step": 172825
    },
    {
      "epoch": 19.013201320132012,
      "grad_norm": 0.0015869140625,
      "learning_rate": 0.00022197083138506378,
      "loss": 0.2314,
      "num_input_tokens_seen": 36472848,
      "step": 172830
    },
    {
      "epoch": 19.013751375137513,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.0002217240800146658,
      "loss": 0.2324,
      "num_input_tokens_seen": 36473904,
      "step": 172835
    },
    {
      "epoch": 19.014301430143014,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00022147746484854568,
      "loss": 0.2308,
      "num_input_tokens_seen": 36474960,
      "step": 172840
    },
    {
      "epoch": 19.014851485148515,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.0002212309858889766,
      "loss": 0.2298,
      "num_input_tokens_seen": 36475952,
      "step": 172845
    },
    {
      "epoch": 19.015401540154016,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00022098464313823006,
      "loss": 0.2303,
      "num_input_tokens_seen": 36476976,
      "step": 172850
    },
    {
      "epoch": 19.015951595159517,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.0002207384365985776,
      "loss": 0.2309,
      "num_input_tokens_seen": 36478032,
      "step": 172855
    },
    {
      "epoch": 19.016501650165015,
      "grad_norm": 0.000545501708984375,
      "learning_rate": 0.00022049236627228572,
      "loss": 0.2308,
      "num_input_tokens_seen": 36479024,
      "step": 172860
    },
    {
      "epoch": 19.017051705170516,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.00022024643216162265,
      "loss": 0.2308,
      "num_input_tokens_seen": 36480080,
      "step": 172865
    },
    {
      "epoch": 19.017601760176017,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00022000063426885817,
      "loss": 0.2308,
      "num_input_tokens_seen": 36481200,
      "step": 172870
    },
    {
      "epoch": 19.01815181518152,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00021975497259625552,
      "loss": 0.2324,
      "num_input_tokens_seen": 36482160,
      "step": 172875
    },
    {
      "epoch": 19.01870187018702,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00021950944714607955,
      "loss": 0.2319,
      "num_input_tokens_seen": 36483248,
      "step": 172880
    },
    {
      "epoch": 19.01925192519252,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00021926405792059176,
      "loss": 0.2314,
      "num_input_tokens_seen": 36484336,
      "step": 172885
    },
    {
      "epoch": 19.019801980198018,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00021901880492205538,
      "loss": 0.2314,
      "num_input_tokens_seen": 36485456,
      "step": 172890
    },
    {
      "epoch": 19.02035203520352,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00021877368815272855,
      "loss": 0.2319,
      "num_input_tokens_seen": 36486480,
      "step": 172895
    },
    {
      "epoch": 19.02090209020902,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00021852870761487284,
      "loss": 0.2329,
      "num_input_tokens_seen": 36487632,
      "step": 172900
    },
    {
      "epoch": 19.02145214521452,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00021828386331074477,
      "loss": 0.2314,
      "num_input_tokens_seen": 36488720,
      "step": 172905
    },
    {
      "epoch": 19.022002200220022,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00021803915524260252,
      "loss": 0.2319,
      "num_input_tokens_seen": 36489840,
      "step": 172910
    },
    {
      "epoch": 19.022552255225524,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 0.0002177945834126993,
      "loss": 0.2303,
      "num_input_tokens_seen": 36490896,
      "step": 172915
    },
    {
      "epoch": 19.023102310231025,
      "grad_norm": 0.0019989013671875,
      "learning_rate": 0.00021755014782328996,
      "loss": 0.2324,
      "num_input_tokens_seen": 36492016,
      "step": 172920
    },
    {
      "epoch": 19.023652365236522,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0002173058484766277,
      "loss": 0.2314,
      "num_input_tokens_seen": 36493040,
      "step": 172925
    },
    {
      "epoch": 19.024202420242023,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0002170616853749657,
      "loss": 0.2324,
      "num_input_tokens_seen": 36494064,
      "step": 172930
    },
    {
      "epoch": 19.024752475247524,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00021681765852055222,
      "loss": 0.2309,
      "num_input_tokens_seen": 36495184,
      "step": 172935
    },
    {
      "epoch": 19.025302530253025,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00021657376791563377,
      "loss": 0.2314,
      "num_input_tokens_seen": 36496208,
      "step": 172940
    },
    {
      "epoch": 19.025852585258527,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00021633001356246515,
      "loss": 0.2319,
      "num_input_tokens_seen": 36497264,
      "step": 172945
    },
    {
      "epoch": 19.026402640264028,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 0.00021608639546328632,
      "loss": 0.2314,
      "num_input_tokens_seen": 36498288,
      "step": 172950
    },
    {
      "epoch": 19.02695269526953,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00021584291362034547,
      "loss": 0.2319,
      "num_input_tokens_seen": 36499344,
      "step": 172955
    },
    {
      "epoch": 19.027502750275026,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00021559956803588742,
      "loss": 0.2309,
      "num_input_tokens_seen": 36500400,
      "step": 172960
    },
    {
      "epoch": 19.028052805280527,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00021535635871215207,
      "loss": 0.2324,
      "num_input_tokens_seen": 36501424,
      "step": 172965
    },
    {
      "epoch": 19.02860286028603,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0002151132856513843,
      "loss": 0.2313,
      "num_input_tokens_seen": 36502480,
      "step": 172970
    },
    {
      "epoch": 19.02915291529153,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.00021487034885582233,
      "loss": 0.2314,
      "num_input_tokens_seen": 36503568,
      "step": 172975
    },
    {
      "epoch": 19.02970297029703,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00021462754832770602,
      "loss": 0.2314,
      "num_input_tokens_seen": 36504592,
      "step": 172980
    },
    {
      "epoch": 19.03025302530253,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00021438488406927192,
      "loss": 0.2319,
      "num_input_tokens_seen": 36505680,
      "step": 172985
    },
    {
      "epoch": 19.03080308030803,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.0002141423560827582,
      "loss": 0.2308,
      "num_input_tokens_seen": 36506704,
      "step": 172990
    },
    {
      "epoch": 19.03135313531353,
      "grad_norm": 0.00653076171875,
      "learning_rate": 0.0002138999643703998,
      "loss": 0.2308,
      "num_input_tokens_seen": 36507760,
      "step": 172995
    },
    {
      "epoch": 19.03190319031903,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00021365770893443158,
      "loss": 0.2303,
      "num_input_tokens_seen": 36508912,
      "step": 173000
    },
    {
      "epoch": 19.032453245324533,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0002134155897770834,
      "loss": 0.2329,
      "num_input_tokens_seen": 36509968,
      "step": 173005
    },
    {
      "epoch": 19.033003300330034,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.0002131736069005885,
      "loss": 0.2303,
      "num_input_tokens_seen": 36511056,
      "step": 173010
    },
    {
      "epoch": 19.033553355335535,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00021293176030717675,
      "loss": 0.2309,
      "num_input_tokens_seen": 36512144,
      "step": 173015
    },
    {
      "epoch": 19.034103410341036,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.00021269004999907804,
      "loss": 0.2319,
      "num_input_tokens_seen": 36513200,
      "step": 173020
    },
    {
      "epoch": 19.034653465346533,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.00021244847597852055,
      "loss": 0.2314,
      "num_input_tokens_seen": 36514288,
      "step": 173025
    },
    {
      "epoch": 19.035203520352034,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 0.00021220703824772756,
      "loss": 0.2329,
      "num_input_tokens_seen": 36515408,
      "step": 173030
    },
    {
      "epoch": 19.035753575357536,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00021196573680892726,
      "loss": 0.2334,
      "num_input_tokens_seen": 36516464,
      "step": 173035
    },
    {
      "epoch": 19.036303630363037,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.0002117245716643429,
      "loss": 0.2309,
      "num_input_tokens_seen": 36517552,
      "step": 173040
    },
    {
      "epoch": 19.036853685368538,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00021148354281619764,
      "loss": 0.2324,
      "num_input_tokens_seen": 36518544,
      "step": 173045
    },
    {
      "epoch": 19.03740374037404,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0002112426502667114,
      "loss": 0.2324,
      "num_input_tokens_seen": 36519568,
      "step": 173050
    },
    {
      "epoch": 19.037953795379536,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00021100189401810574,
      "loss": 0.2319,
      "num_input_tokens_seen": 36520624,
      "step": 173055
    },
    {
      "epoch": 19.038503850385037,
      "grad_norm": 0.0107421875,
      "learning_rate": 0.0002107612740726006,
      "loss": 0.2293,
      "num_input_tokens_seen": 36521680,
      "step": 173060
    },
    {
      "epoch": 19.03905390539054,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 0.00021052079043240912,
      "loss": 0.2319,
      "num_input_tokens_seen": 36522672,
      "step": 173065
    },
    {
      "epoch": 19.03960396039604,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0002102804430997529,
      "loss": 0.2313,
      "num_input_tokens_seen": 36523696,
      "step": 173070
    },
    {
      "epoch": 19.04015401540154,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00021004023207684517,
      "loss": 0.2324,
      "num_input_tokens_seen": 36524848,
      "step": 173075
    },
    {
      "epoch": 19.040704070407042,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.00020980015736589918,
      "loss": 0.2314,
      "num_input_tokens_seen": 36525872,
      "step": 173080
    },
    {
      "epoch": 19.041254125412543,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.00020956021896912811,
      "loss": 0.2329,
      "num_input_tokens_seen": 36526928,
      "step": 173085
    },
    {
      "epoch": 19.04180418041804,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0002093204168887419,
      "loss": 0.2309,
      "num_input_tokens_seen": 36527952,
      "step": 173090
    },
    {
      "epoch": 19.04235423542354,
      "grad_norm": 0.000629425048828125,
      "learning_rate": 0.0002090807511269538,
      "loss": 0.2319,
      "num_input_tokens_seen": 36528976,
      "step": 173095
    },
    {
      "epoch": 19.042904290429043,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00020884122168597196,
      "loss": 0.2314,
      "num_input_tokens_seen": 36530000,
      "step": 173100
    },
    {
      "epoch": 19.043454345434544,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00020860182856800134,
      "loss": 0.2329,
      "num_input_tokens_seen": 36531088,
      "step": 173105
    },
    {
      "epoch": 19.044004400440045,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.00020836257177525184,
      "loss": 0.2298,
      "num_input_tokens_seen": 36532112,
      "step": 173110
    },
    {
      "epoch": 19.044554455445546,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00020812345130992503,
      "loss": 0.2319,
      "num_input_tokens_seen": 36533136,
      "step": 173115
    },
    {
      "epoch": 19.045104510451043,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00020788446717422748,
      "loss": 0.2303,
      "num_input_tokens_seen": 36534224,
      "step": 173120
    },
    {
      "epoch": 19.045654565456545,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 0.00020764561937035907,
      "loss": 0.2298,
      "num_input_tokens_seen": 36535280,
      "step": 173125
    },
    {
      "epoch": 19.046204620462046,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00020740690790052474,
      "loss": 0.2308,
      "num_input_tokens_seen": 36536368,
      "step": 173130
    },
    {
      "epoch": 19.046754675467547,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0002071683327669227,
      "loss": 0.2329,
      "num_input_tokens_seen": 36537424,
      "step": 173135
    },
    {
      "epoch": 19.047304730473048,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.00020692989397175286,
      "loss": 0.2319,
      "num_input_tokens_seen": 36538512,
      "step": 173140
    },
    {
      "epoch": 19.04785478547855,
      "grad_norm": 0.0009307861328125,
      "learning_rate": 0.00020669159151721016,
      "loss": 0.2309,
      "num_input_tokens_seen": 36539632,
      "step": 173145
    },
    {
      "epoch": 19.04840484048405,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0002064534254054928,
      "loss": 0.2319,
      "num_input_tokens_seen": 36540720,
      "step": 173150
    },
    {
      "epoch": 19.048954895489548,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.0002062153956387941,
      "loss": 0.2319,
      "num_input_tokens_seen": 36541808,
      "step": 173155
    },
    {
      "epoch": 19.04950495049505,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0002059775022193122,
      "loss": 0.2334,
      "num_input_tokens_seen": 36542864,
      "step": 173160
    },
    {
      "epoch": 19.05005500550055,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.00020573974514923542,
      "loss": 0.2314,
      "num_input_tokens_seen": 36543888,
      "step": 173165
    },
    {
      "epoch": 19.05060506050605,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.000205502124430757,
      "loss": 0.2314,
      "num_input_tokens_seen": 36544912,
      "step": 173170
    },
    {
      "epoch": 19.051155115511552,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.00020526464006606515,
      "loss": 0.2298,
      "num_input_tokens_seen": 36546032,
      "step": 173175
    },
    {
      "epoch": 19.051705170517053,
      "grad_norm": 0.0106201171875,
      "learning_rate": 0.0002050272920573515,
      "loss": 0.2298,
      "num_input_tokens_seen": 36547024,
      "step": 173180
    },
    {
      "epoch": 19.05225522552255,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00020479008040679924,
      "loss": 0.2314,
      "num_input_tokens_seen": 36548112,
      "step": 173185
    },
    {
      "epoch": 19.05280528052805,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.00020455300511660002,
      "loss": 0.2324,
      "num_input_tokens_seen": 36549200,
      "step": 173190
    },
    {
      "epoch": 19.053355335533553,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.00020431606618893537,
      "loss": 0.2324,
      "num_input_tokens_seen": 36550320,
      "step": 173195
    },
    {
      "epoch": 19.053905390539054,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00020407926362599026,
      "loss": 0.2298,
      "num_input_tokens_seen": 36551408,
      "step": 173200
    },
    {
      "epoch": 19.054455445544555,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00020384259742994626,
      "loss": 0.2308,
      "num_input_tokens_seen": 36552432,
      "step": 173205
    },
    {
      "epoch": 19.055005500550056,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00020360606760298493,
      "loss": 0.2314,
      "num_input_tokens_seen": 36553456,
      "step": 173210
    },
    {
      "epoch": 19.055555555555557,
      "grad_norm": 0.000926971435546875,
      "learning_rate": 0.0002033696741472879,
      "loss": 0.2314,
      "num_input_tokens_seen": 36554480,
      "step": 173215
    },
    {
      "epoch": 19.056105610561055,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00020313341706503174,
      "loss": 0.2298,
      "num_input_tokens_seen": 36555568,
      "step": 173220
    },
    {
      "epoch": 19.056655665566556,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0002028972963583947,
      "loss": 0.2303,
      "num_input_tokens_seen": 36556656,
      "step": 173225
    },
    {
      "epoch": 19.057205720572057,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00020266131202955172,
      "loss": 0.2324,
      "num_input_tokens_seen": 36557744,
      "step": 173230
    },
    {
      "epoch": 19.057755775577558,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.00020242546408068106,
      "loss": 0.2303,
      "num_input_tokens_seen": 36558800,
      "step": 173235
    },
    {
      "epoch": 19.05830583058306,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.00020218975251395265,
      "loss": 0.2314,
      "num_input_tokens_seen": 36559888,
      "step": 173240
    },
    {
      "epoch": 19.05885588558856,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00020195417733154142,
      "loss": 0.2319,
      "num_input_tokens_seen": 36560944,
      "step": 173245
    },
    {
      "epoch": 19.059405940594058,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 0.00020171873853561727,
      "loss": 0.2319,
      "num_input_tokens_seen": 36561968,
      "step": 173250
    },
    {
      "epoch": 19.05995599559956,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.00020148343612835017,
      "loss": 0.2308,
      "num_input_tokens_seen": 36563056,
      "step": 173255
    },
    {
      "epoch": 19.06050605060506,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00020124827011191003,
      "loss": 0.2298,
      "num_input_tokens_seen": 36564080,
      "step": 173260
    },
    {
      "epoch": 19.06105610561056,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00020101324048846513,
      "loss": 0.2324,
      "num_input_tokens_seen": 36565104,
      "step": 173265
    },
    {
      "epoch": 19.061606160616062,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0002007783472601754,
      "loss": 0.2293,
      "num_input_tokens_seen": 36566160,
      "step": 173270
    },
    {
      "epoch": 19.062156215621563,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 0.00020054359042921409,
      "loss": 0.2308,
      "num_input_tokens_seen": 36567184,
      "step": 173275
    },
    {
      "epoch": 19.062706270627064,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0002003089699977395,
      "loss": 0.2309,
      "num_input_tokens_seen": 36568272,
      "step": 173280
    },
    {
      "epoch": 19.063256325632562,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 0.00020007448596791655,
      "loss": 0.2308,
      "num_input_tokens_seen": 36569296,
      "step": 173285
    },
    {
      "epoch": 19.063806380638063,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00019984013834190516,
      "loss": 0.2314,
      "num_input_tokens_seen": 36570256,
      "step": 173290
    },
    {
      "epoch": 19.064356435643564,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.0001996059271218653,
      "loss": 0.2303,
      "num_input_tokens_seen": 36571312,
      "step": 173295
    },
    {
      "epoch": 19.064906490649065,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.00019937185230995522,
      "loss": 0.2308,
      "num_input_tokens_seen": 36572432,
      "step": 173300
    },
    {
      "epoch": 19.065456545654566,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00019913791390833158,
      "loss": 0.2298,
      "num_input_tokens_seen": 36573424,
      "step": 173305
    },
    {
      "epoch": 19.066006600660067,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00019890411191915425,
      "loss": 0.2314,
      "num_input_tokens_seen": 36574480,
      "step": 173310
    },
    {
      "epoch": 19.066556655665565,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00019867044634457487,
      "loss": 0.2308,
      "num_input_tokens_seen": 36575504,
      "step": 173315
    },
    {
      "epoch": 19.067106710671066,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0001984369171867467,
      "loss": 0.2329,
      "num_input_tokens_seen": 36576560,
      "step": 173320
    },
    {
      "epoch": 19.067656765676567,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00019820352444782307,
      "loss": 0.2308,
      "num_input_tokens_seen": 36577616,
      "step": 173325
    },
    {
      "epoch": 19.068206820682068,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00019797026812995387,
      "loss": 0.2303,
      "num_input_tokens_seen": 36578704,
      "step": 173330
    },
    {
      "epoch": 19.06875687568757,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.00019773714823529242,
      "loss": 0.2314,
      "num_input_tokens_seen": 36579760,
      "step": 173335
    },
    {
      "epoch": 19.06930693069307,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00019750416476598364,
      "loss": 0.234,
      "num_input_tokens_seen": 36580784,
      "step": 173340
    },
    {
      "epoch": 19.06985698569857,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 0.00019727131772417583,
      "loss": 0.2324,
      "num_input_tokens_seen": 36581808,
      "step": 173345
    },
    {
      "epoch": 19.07040704070407,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.00019703860711201725,
      "loss": 0.2329,
      "num_input_tokens_seen": 36582832,
      "step": 173350
    },
    {
      "epoch": 19.07095709570957,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.00019680603293164789,
      "loss": 0.2324,
      "num_input_tokens_seen": 36583920,
      "step": 173355
    },
    {
      "epoch": 19.07150715071507,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00019657359518521434,
      "loss": 0.2308,
      "num_input_tokens_seen": 36584912,
      "step": 173360
    },
    {
      "epoch": 19.072057205720572,
      "grad_norm": 0.0008544921875,
      "learning_rate": 0.00019634129387485988,
      "loss": 0.2298,
      "num_input_tokens_seen": 36586000,
      "step": 173365
    },
    {
      "epoch": 19.072607260726073,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00019610912900272447,
      "loss": 0.2309,
      "num_input_tokens_seen": 36587024,
      "step": 173370
    },
    {
      "epoch": 19.073157315731574,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0001958771005709464,
      "loss": 0.233,
      "num_input_tokens_seen": 36588080,
      "step": 173375
    },
    {
      "epoch": 19.073707370737075,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.00019564520858166733,
      "loss": 0.2314,
      "num_input_tokens_seen": 36589168,
      "step": 173380
    },
    {
      "epoch": 19.074257425742573,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00019541345303702052,
      "loss": 0.2309,
      "num_input_tokens_seen": 36590224,
      "step": 173385
    },
    {
      "epoch": 19.074807480748074,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00019518183393914422,
      "loss": 0.2319,
      "num_input_tokens_seen": 36591312,
      "step": 173390
    },
    {
      "epoch": 19.075357535753575,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00019495035129017346,
      "loss": 0.2303,
      "num_input_tokens_seen": 36592368,
      "step": 173395
    },
    {
      "epoch": 19.075907590759076,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00019471900509224148,
      "loss": 0.2319,
      "num_input_tokens_seen": 36593392,
      "step": 173400
    },
    {
      "epoch": 19.076457645764577,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.00019448779534747996,
      "loss": 0.2303,
      "num_input_tokens_seen": 36594448,
      "step": 173405
    },
    {
      "epoch": 19.07700770077008,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0001942567220580188,
      "loss": 0.2314,
      "num_input_tokens_seen": 36595504,
      "step": 173410
    },
    {
      "epoch": 19.077557755775576,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00019402578522598966,
      "loss": 0.2303,
      "num_input_tokens_seen": 36596592,
      "step": 173415
    },
    {
      "epoch": 19.078107810781077,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.0001937949848535192,
      "loss": 0.2319,
      "num_input_tokens_seen": 36597648,
      "step": 173420
    },
    {
      "epoch": 19.078657865786578,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.000193564320942739,
      "loss": 0.2324,
      "num_input_tokens_seen": 36598704,
      "step": 173425
    },
    {
      "epoch": 19.07920792079208,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 0.00019333379349576905,
      "loss": 0.2324,
      "num_input_tokens_seen": 36599760,
      "step": 173430
    },
    {
      "epoch": 19.07975797579758,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00019310340251473766,
      "loss": 0.2314,
      "num_input_tokens_seen": 36600784,
      "step": 173435
    },
    {
      "epoch": 19.08030803080308,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00019287314800176645,
      "loss": 0.2319,
      "num_input_tokens_seen": 36601840,
      "step": 173440
    },
    {
      "epoch": 19.080858085808583,
      "grad_norm": 0.0030364990234375,
      "learning_rate": 0.00019264302995897873,
      "loss": 0.2319,
      "num_input_tokens_seen": 36602896,
      "step": 173445
    },
    {
      "epoch": 19.08140814081408,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0001924130483884945,
      "loss": 0.2308,
      "num_input_tokens_seen": 36603920,
      "step": 173450
    },
    {
      "epoch": 19.08195819581958,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00019218320329243365,
      "loss": 0.2298,
      "num_input_tokens_seen": 36605008,
      "step": 173455
    },
    {
      "epoch": 19.082508250825082,
      "grad_norm": 0.00092315673828125,
      "learning_rate": 0.00019195349467291455,
      "loss": 0.2309,
      "num_input_tokens_seen": 36606032,
      "step": 173460
    },
    {
      "epoch": 19.083058305830583,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0001917239225320555,
      "loss": 0.2303,
      "num_input_tokens_seen": 36607056,
      "step": 173465
    },
    {
      "epoch": 19.083608360836084,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00019149448687197146,
      "loss": 0.2319,
      "num_input_tokens_seen": 36608144,
      "step": 173470
    },
    {
      "epoch": 19.084158415841586,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00019126518769477574,
      "loss": 0.2319,
      "num_input_tokens_seen": 36609264,
      "step": 173475
    },
    {
      "epoch": 19.084708470847083,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00019103602500258332,
      "loss": 0.2308,
      "num_input_tokens_seen": 36610320,
      "step": 173480
    },
    {
      "epoch": 19.085258525852584,
      "grad_norm": 0.0018768310546875,
      "learning_rate": 0.00019080699879750585,
      "loss": 0.2303,
      "num_input_tokens_seen": 36611344,
      "step": 173485
    },
    {
      "epoch": 19.085808580858085,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00019057810908165495,
      "loss": 0.2324,
      "num_input_tokens_seen": 36612400,
      "step": 173490
    },
    {
      "epoch": 19.086358635863586,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00019034935585713895,
      "loss": 0.2319,
      "num_input_tokens_seen": 36613488,
      "step": 173495
    },
    {
      "epoch": 19.086908690869087,
      "grad_norm": 0.00604248046875,
      "learning_rate": 0.00019012073912606618,
      "loss": 0.234,
      "num_input_tokens_seen": 36614672,
      "step": 173500
    },
    {
      "epoch": 19.08745874587459,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00018989225889054495,
      "loss": 0.2329,
      "num_input_tokens_seen": 36615728,
      "step": 173505
    },
    {
      "epoch": 19.08800880088009,
      "grad_norm": 0.000885009765625,
      "learning_rate": 0.00018966391515267854,
      "loss": 0.2308,
      "num_input_tokens_seen": 36616816,
      "step": 173510
    },
    {
      "epoch": 19.088558855885587,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00018943570791457365,
      "loss": 0.2303,
      "num_input_tokens_seen": 36617872,
      "step": 173515
    },
    {
      "epoch": 19.08910891089109,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 0.00018920763717833355,
      "loss": 0.2314,
      "num_input_tokens_seen": 36618896,
      "step": 173520
    },
    {
      "epoch": 19.08965896589659,
      "grad_norm": 0.001495361328125,
      "learning_rate": 0.00018897970294605993,
      "loss": 0.2303,
      "num_input_tokens_seen": 36619952,
      "step": 173525
    },
    {
      "epoch": 19.09020902090209,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 0.00018875190521985274,
      "loss": 0.2319,
      "num_input_tokens_seen": 36621104,
      "step": 173530
    },
    {
      "epoch": 19.09075907590759,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.000188524244001812,
      "loss": 0.2314,
      "num_input_tokens_seen": 36622096,
      "step": 173535
    },
    {
      "epoch": 19.091309130913093,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00018829671929403768,
      "loss": 0.2293,
      "num_input_tokens_seen": 36623248,
      "step": 173540
    },
    {
      "epoch": 19.09185918591859,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.00018806933109862477,
      "loss": 0.2309,
      "num_input_tokens_seen": 36624336,
      "step": 173545
    },
    {
      "epoch": 19.09240924092409,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.00018784207941766826,
      "loss": 0.2314,
      "num_input_tokens_seen": 36625360,
      "step": 173550
    },
    {
      "epoch": 19.092959295929592,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.00018761496425326484,
      "loss": 0.2319,
      "num_input_tokens_seen": 36626416,
      "step": 173555
    },
    {
      "epoch": 19.093509350935093,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.00018738798560750445,
      "loss": 0.2319,
      "num_input_tokens_seen": 36627408,
      "step": 173560
    },
    {
      "epoch": 19.094059405940595,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00018716114348248214,
      "loss": 0.2309,
      "num_input_tokens_seen": 36628464,
      "step": 173565
    },
    {
      "epoch": 19.094609460946096,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00018693443788028952,
      "loss": 0.2314,
      "num_input_tokens_seen": 36629488,
      "step": 173570
    },
    {
      "epoch": 19.095159515951597,
      "grad_norm": 0.000843048095703125,
      "learning_rate": 0.0001867078688030116,
      "loss": 0.2324,
      "num_input_tokens_seen": 36630544,
      "step": 173575
    },
    {
      "epoch": 19.095709570957094,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0001864814362527417,
      "loss": 0.2319,
      "num_input_tokens_seen": 36631600,
      "step": 173580
    },
    {
      "epoch": 19.096259625962595,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.00018625514023156152,
      "loss": 0.2309,
      "num_input_tokens_seen": 36632656,
      "step": 173585
    },
    {
      "epoch": 19.096809680968097,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.00018602898074156104,
      "loss": 0.2314,
      "num_input_tokens_seen": 36633680,
      "step": 173590
    },
    {
      "epoch": 19.097359735973598,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 0.0001858029577848236,
      "loss": 0.2314,
      "num_input_tokens_seen": 36634768,
      "step": 173595
    },
    {
      "epoch": 19.0979097909791,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00018557707136343082,
      "loss": 0.2329,
      "num_input_tokens_seen": 36635856,
      "step": 173600
    },
    {
      "epoch": 19.0984598459846,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0001853513214794661,
      "loss": 0.233,
      "num_input_tokens_seen": 36636944,
      "step": 173605
    },
    {
      "epoch": 19.099009900990097,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00018512570813500773,
      "loss": 0.2324,
      "num_input_tokens_seen": 36638032,
      "step": 173610
    },
    {
      "epoch": 19.0995599559956,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00018490023133213907,
      "loss": 0.2308,
      "num_input_tokens_seen": 36639152,
      "step": 173615
    },
    {
      "epoch": 19.1001100110011,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 0.0001846748910729351,
      "loss": 0.2329,
      "num_input_tokens_seen": 36640240,
      "step": 173620
    },
    {
      "epoch": 19.1006600660066,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00018444968735947253,
      "loss": 0.2314,
      "num_input_tokens_seen": 36641232,
      "step": 173625
    },
    {
      "epoch": 19.1012101210121,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00018422462019382967,
      "loss": 0.2314,
      "num_input_tokens_seen": 36642288,
      "step": 173630
    },
    {
      "epoch": 19.101760176017603,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00018399968957807822,
      "loss": 0.2329,
      "num_input_tokens_seen": 36643344,
      "step": 173635
    },
    {
      "epoch": 19.102310231023104,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00018377489551429314,
      "loss": 0.2303,
      "num_input_tokens_seen": 36644368,
      "step": 173640
    },
    {
      "epoch": 19.1028602860286,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.00018355023800454285,
      "loss": 0.2303,
      "num_input_tokens_seen": 36645424,
      "step": 173645
    },
    {
      "epoch": 19.103410341034103,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00018332571705090227,
      "loss": 0.2288,
      "num_input_tokens_seen": 36646480,
      "step": 173650
    },
    {
      "epoch": 19.103960396039604,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.0001831013326554398,
      "loss": 0.2314,
      "num_input_tokens_seen": 36647440,
      "step": 173655
    },
    {
      "epoch": 19.104510451045105,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00018287708482022047,
      "loss": 0.2309,
      "num_input_tokens_seen": 36648432,
      "step": 173660
    },
    {
      "epoch": 19.105060506050606,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00018265297354731424,
      "loss": 0.2303,
      "num_input_tokens_seen": 36649520,
      "step": 173665
    },
    {
      "epoch": 19.105610561056107,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.00018242899883878615,
      "loss": 0.2319,
      "num_input_tokens_seen": 36650544,
      "step": 173670
    },
    {
      "epoch": 19.106160616061604,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00018220516069669955,
      "loss": 0.2319,
      "num_input_tokens_seen": 36651632,
      "step": 173675
    },
    {
      "epoch": 19.106710671067106,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00018198145912311613,
      "loss": 0.2298,
      "num_input_tokens_seen": 36652688,
      "step": 173680
    },
    {
      "epoch": 19.107260726072607,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00018175789412009922,
      "loss": 0.2298,
      "num_input_tokens_seen": 36653744,
      "step": 173685
    },
    {
      "epoch": 19.107810781078108,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 0.0001815344656897122,
      "loss": 0.2298,
      "num_input_tokens_seen": 36654768,
      "step": 173690
    },
    {
      "epoch": 19.10836083608361,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00018131117383401007,
      "loss": 0.2308,
      "num_input_tokens_seen": 36655824,
      "step": 173695
    },
    {
      "epoch": 19.10891089108911,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00018108801855505117,
      "loss": 0.2314,
      "num_input_tokens_seen": 36656912,
      "step": 173700
    },
    {
      "epoch": 19.10946094609461,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00018086499985489222,
      "loss": 0.2303,
      "num_input_tokens_seen": 36658064,
      "step": 173705
    },
    {
      "epoch": 19.11001100110011,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 0.00018064211773559158,
      "loss": 0.2314,
      "num_input_tokens_seen": 36659152,
      "step": 173710
    },
    {
      "epoch": 19.11056105610561,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.0001804193721992009,
      "loss": 0.2308,
      "num_input_tokens_seen": 36660176,
      "step": 173715
    },
    {
      "epoch": 19.11111111111111,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0001801967632477752,
      "loss": 0.2314,
      "num_input_tokens_seen": 36661264,
      "step": 173720
    },
    {
      "epoch": 19.111661166116612,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00017997429088336292,
      "loss": 0.2324,
      "num_input_tokens_seen": 36662288,
      "step": 173725
    },
    {
      "epoch": 19.112211221122113,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.00017975195510801733,
      "loss": 0.2314,
      "num_input_tokens_seen": 36663312,
      "step": 173730
    },
    {
      "epoch": 19.112761276127614,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00017952975592378518,
      "loss": 0.2309,
      "num_input_tokens_seen": 36664336,
      "step": 173735
    },
    {
      "epoch": 19.11331133113311,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00017930769333271646,
      "loss": 0.2319,
      "num_input_tokens_seen": 36665456,
      "step": 173740
    },
    {
      "epoch": 19.113861386138613,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00017908576733685787,
      "loss": 0.2309,
      "num_input_tokens_seen": 36666544,
      "step": 173745
    },
    {
      "epoch": 19.114411441144114,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0001788639779382528,
      "loss": 0.2314,
      "num_input_tokens_seen": 36667632,
      "step": 173750
    },
    {
      "epoch": 19.114961496149615,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00017864232513894795,
      "loss": 0.2319,
      "num_input_tokens_seen": 36668752,
      "step": 173755
    },
    {
      "epoch": 19.115511551155116,
      "grad_norm": 0.000766754150390625,
      "learning_rate": 0.00017842080894098333,
      "loss": 0.2314,
      "num_input_tokens_seen": 36669712,
      "step": 173760
    },
    {
      "epoch": 19.116061606160617,
      "grad_norm": 0.000621795654296875,
      "learning_rate": 0.00017819942934640232,
      "loss": 0.2335,
      "num_input_tokens_seen": 36670768,
      "step": 173765
    },
    {
      "epoch": 19.116611661166118,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.00017797818635724493,
      "loss": 0.2314,
      "num_input_tokens_seen": 36671728,
      "step": 173770
    },
    {
      "epoch": 19.117161716171616,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00017775707997554955,
      "loss": 0.2303,
      "num_input_tokens_seen": 36672784,
      "step": 173775
    },
    {
      "epoch": 19.117711771177117,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.00017753611020335624,
      "loss": 0.2309,
      "num_input_tokens_seen": 36673840,
      "step": 173780
    },
    {
      "epoch": 19.118261826182618,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00017731527704270165,
      "loss": 0.2314,
      "num_input_tokens_seen": 36674896,
      "step": 173785
    },
    {
      "epoch": 19.11881188118812,
      "grad_norm": 0.0030975341796875,
      "learning_rate": 0.00017709458049561589,
      "loss": 0.2314,
      "num_input_tokens_seen": 36675984,
      "step": 173790
    },
    {
      "epoch": 19.11936193619362,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.00017687402056413892,
      "loss": 0.2324,
      "num_input_tokens_seen": 36677104,
      "step": 173795
    },
    {
      "epoch": 19.11991199119912,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00017665359725029916,
      "loss": 0.2314,
      "num_input_tokens_seen": 36678160,
      "step": 173800
    },
    {
      "epoch": 19.120462046204622,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0001764333105561333,
      "loss": 0.2314,
      "num_input_tokens_seen": 36679216,
      "step": 173805
    },
    {
      "epoch": 19.12101210121012,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 0.0001762131604836681,
      "loss": 0.2309,
      "num_input_tokens_seen": 36680272,
      "step": 173810
    },
    {
      "epoch": 19.12156215621562,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.00017599314703493018,
      "loss": 0.2319,
      "num_input_tokens_seen": 36681328,
      "step": 173815
    },
    {
      "epoch": 19.122112211221122,
      "grad_norm": 0.00244140625,
      "learning_rate": 0.0001757732702119513,
      "loss": 0.2308,
      "num_input_tokens_seen": 36682320,
      "step": 173820
    },
    {
      "epoch": 19.122662266226623,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00017555353001675822,
      "loss": 0.2303,
      "num_input_tokens_seen": 36683344,
      "step": 173825
    },
    {
      "epoch": 19.123212321232124,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00017533392645137423,
      "loss": 0.2319,
      "num_input_tokens_seen": 36684368,
      "step": 173830
    },
    {
      "epoch": 19.123762376237625,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00017511445951782443,
      "loss": 0.2293,
      "num_input_tokens_seen": 36685456,
      "step": 173835
    },
    {
      "epoch": 19.124312431243123,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00017489512921813055,
      "loss": 0.2303,
      "num_input_tokens_seen": 36686512,
      "step": 173840
    },
    {
      "epoch": 19.124862486248624,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.00017467593555431425,
      "loss": 0.2303,
      "num_input_tokens_seen": 36687568,
      "step": 173845
    },
    {
      "epoch": 19.125412541254125,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.00017445687852839565,
      "loss": 0.2293,
      "num_input_tokens_seen": 36688560,
      "step": 173850
    },
    {
      "epoch": 19.125962596259626,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.00017423795814239305,
      "loss": 0.2324,
      "num_input_tokens_seen": 36689616,
      "step": 173855
    },
    {
      "epoch": 19.126512651265127,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0001740191743983266,
      "loss": 0.2324,
      "num_input_tokens_seen": 36690640,
      "step": 173860
    },
    {
      "epoch": 19.127062706270628,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.00017380052729821128,
      "loss": 0.2304,
      "num_input_tokens_seen": 36691696,
      "step": 173865
    },
    {
      "epoch": 19.12761276127613,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00017358201684406215,
      "loss": 0.2335,
      "num_input_tokens_seen": 36692688,
      "step": 173870
    },
    {
      "epoch": 19.128162816281627,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00017336364303789263,
      "loss": 0.2319,
      "num_input_tokens_seen": 36693680,
      "step": 173875
    },
    {
      "epoch": 19.128712871287128,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.0001731454058817161,
      "loss": 0.2319,
      "num_input_tokens_seen": 36694736,
      "step": 173880
    },
    {
      "epoch": 19.12926292629263,
      "grad_norm": 0.001190185546875,
      "learning_rate": 0.0001729273053775443,
      "loss": 0.2324,
      "num_input_tokens_seen": 36695760,
      "step": 173885
    },
    {
      "epoch": 19.12981298129813,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00017270934152738559,
      "loss": 0.2309,
      "num_input_tokens_seen": 36696816,
      "step": 173890
    },
    {
      "epoch": 19.13036303630363,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0001724915143332517,
      "loss": 0.2314,
      "num_input_tokens_seen": 36697840,
      "step": 173895
    },
    {
      "epoch": 19.130913091309132,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.00017227382379714773,
      "loss": 0.233,
      "num_input_tokens_seen": 36698928,
      "step": 173900
    },
    {
      "epoch": 19.13146314631463,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.00017205626992108035,
      "loss": 0.2314,
      "num_input_tokens_seen": 36699920,
      "step": 173905
    },
    {
      "epoch": 19.13201320132013,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00017183885270705634,
      "loss": 0.2314,
      "num_input_tokens_seen": 36700976,
      "step": 173910
    },
    {
      "epoch": 19.132563256325632,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0001716215721570774,
      "loss": 0.2309,
      "num_input_tokens_seen": 36702000,
      "step": 173915
    },
    {
      "epoch": 19.133113311331133,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00017140442827314693,
      "loss": 0.2303,
      "num_input_tokens_seen": 36703088,
      "step": 173920
    },
    {
      "epoch": 19.133663366336634,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 0.00017118742105726837,
      "loss": 0.2303,
      "num_input_tokens_seen": 36704144,
      "step": 173925
    },
    {
      "epoch": 19.134213421342135,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00017097055051143673,
      "loss": 0.2324,
      "num_input_tokens_seen": 36705168,
      "step": 173930
    },
    {
      "epoch": 19.134763476347636,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0001707538166376571,
      "loss": 0.2314,
      "num_input_tokens_seen": 36706128,
      "step": 173935
    },
    {
      "epoch": 19.135313531353134,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00017053721943791953,
      "loss": 0.2314,
      "num_input_tokens_seen": 36707184,
      "step": 173940
    },
    {
      "epoch": 19.135863586358635,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.00017032075891422916,
      "loss": 0.2303,
      "num_input_tokens_seen": 36708176,
      "step": 173945
    },
    {
      "epoch": 19.136413641364136,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00017010443506857432,
      "loss": 0.2293,
      "num_input_tokens_seen": 36709296,
      "step": 173950
    },
    {
      "epoch": 19.136963696369637,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.00016988824790295008,
      "loss": 0.2329,
      "num_input_tokens_seen": 36710352,
      "step": 173955
    },
    {
      "epoch": 19.13751375137514,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00016967219741935156,
      "loss": 0.2308,
      "num_input_tokens_seen": 36711408,
      "step": 173960
    },
    {
      "epoch": 19.13806380638064,
      "grad_norm": 0.000827789306640625,
      "learning_rate": 0.0001694562836197655,
      "loss": 0.2309,
      "num_input_tokens_seen": 36712400,
      "step": 173965
    },
    {
      "epoch": 19.138613861386137,
      "grad_norm": 0.0030670166015625,
      "learning_rate": 0.00016924050650618693,
      "loss": 0.2308,
      "num_input_tokens_seen": 36713456,
      "step": 173970
    },
    {
      "epoch": 19.139163916391638,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00016902486608060097,
      "loss": 0.2324,
      "num_input_tokens_seen": 36714480,
      "step": 173975
    },
    {
      "epoch": 19.13971397139714,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.00016880936234499598,
      "loss": 0.2309,
      "num_input_tokens_seen": 36715504,
      "step": 173980
    },
    {
      "epoch": 19.14026402640264,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00016859399530135875,
      "loss": 0.2308,
      "num_input_tokens_seen": 36716496,
      "step": 173985
    },
    {
      "epoch": 19.14081408140814,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00016837876495167436,
      "loss": 0.2303,
      "num_input_tokens_seen": 36717584,
      "step": 173990
    },
    {
      "epoch": 19.141364136413642,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00016816367129792454,
      "loss": 0.2303,
      "num_input_tokens_seen": 36718608,
      "step": 173995
    },
    {
      "epoch": 19.141914191419144,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.000167948714342096,
      "loss": 0.2308,
      "num_input_tokens_seen": 36719664,
      "step": 174000
    },
    {
      "epoch": 19.14246424642464,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.0001677338940861639,
      "loss": 0.2329,
      "num_input_tokens_seen": 36720688,
      "step": 174005
    },
    {
      "epoch": 19.143014301430142,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00016751921053211326,
      "loss": 0.2335,
      "num_input_tokens_seen": 36721808,
      "step": 174010
    },
    {
      "epoch": 19.143564356435643,
      "grad_norm": 0.00058746337890625,
      "learning_rate": 0.0001673046636819192,
      "loss": 0.2329,
      "num_input_tokens_seen": 36722832,
      "step": 174015
    },
    {
      "epoch": 19.144114411441144,
      "grad_norm": 0.006378173828125,
      "learning_rate": 0.00016709025353756345,
      "loss": 0.2335,
      "num_input_tokens_seen": 36723856,
      "step": 174020
    },
    {
      "epoch": 19.144664466446645,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0001668759801010161,
      "loss": 0.2329,
      "num_input_tokens_seen": 36724912,
      "step": 174025
    },
    {
      "epoch": 19.145214521452147,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.0001666618433742556,
      "loss": 0.2303,
      "num_input_tokens_seen": 36725904,
      "step": 174030
    },
    {
      "epoch": 19.145764576457644,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00016644784335925698,
      "loss": 0.2309,
      "num_input_tokens_seen": 36726960,
      "step": 174035
    },
    {
      "epoch": 19.146314631463145,
      "grad_norm": 0.006134033203125,
      "learning_rate": 0.0001662339800579904,
      "loss": 0.2324,
      "num_input_tokens_seen": 36727984,
      "step": 174040
    },
    {
      "epoch": 19.146864686468646,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00016602025347242755,
      "loss": 0.2324,
      "num_input_tokens_seen": 36729104,
      "step": 174045
    },
    {
      "epoch": 19.147414741474147,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00016580666360453689,
      "loss": 0.2308,
      "num_input_tokens_seen": 36730160,
      "step": 174050
    },
    {
      "epoch": 19.14796479647965,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 0.00016559321045628682,
      "loss": 0.2314,
      "num_input_tokens_seen": 36731184,
      "step": 174055
    },
    {
      "epoch": 19.14851485148515,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.00016537989402964914,
      "loss": 0.2303,
      "num_input_tokens_seen": 36732176,
      "step": 174060
    },
    {
      "epoch": 19.14906490649065,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00016516671432658392,
      "loss": 0.234,
      "num_input_tokens_seen": 36733200,
      "step": 174065
    },
    {
      "epoch": 19.149614961496148,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0001649536713490596,
      "loss": 0.2288,
      "num_input_tokens_seen": 36734224,
      "step": 174070
    },
    {
      "epoch": 19.15016501650165,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00016474076509903957,
      "loss": 0.2314,
      "num_input_tokens_seen": 36735280,
      "step": 174075
    },
    {
      "epoch": 19.15071507150715,
      "grad_norm": 0.0108642578125,
      "learning_rate": 0.00016452799557848397,
      "loss": 0.2309,
      "num_input_tokens_seen": 36736336,
      "step": 174080
    },
    {
      "epoch": 19.15126512651265,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 0.0001643153627893562,
      "loss": 0.2314,
      "num_input_tokens_seen": 36737424,
      "step": 174085
    },
    {
      "epoch": 19.151815181518153,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00016410286673361306,
      "loss": 0.2303,
      "num_input_tokens_seen": 36738480,
      "step": 174090
    },
    {
      "epoch": 19.152365236523654,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0001638905074132163,
      "loss": 0.2319,
      "num_input_tokens_seen": 36739568,
      "step": 174095
    },
    {
      "epoch": 19.15291529152915,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0001636782848301227,
      "loss": 0.2303,
      "num_input_tokens_seen": 36740656,
      "step": 174100
    },
    {
      "epoch": 19.153465346534652,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 0.00016346619898628567,
      "loss": 0.2329,
      "num_input_tokens_seen": 36741744,
      "step": 174105
    },
    {
      "epoch": 19.154015401540153,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.00016325424988366199,
      "loss": 0.2298,
      "num_input_tokens_seen": 36742800,
      "step": 174110
    },
    {
      "epoch": 19.154565456545654,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00016304243752420343,
      "loss": 0.2314,
      "num_input_tokens_seen": 36743856,
      "step": 174115
    },
    {
      "epoch": 19.155115511551156,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.00016283076190986345,
      "loss": 0.2308,
      "num_input_tokens_seen": 36744912,
      "step": 174120
    },
    {
      "epoch": 19.155665566556657,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.0001626192230425938,
      "loss": 0.2303,
      "num_input_tokens_seen": 36745968,
      "step": 174125
    },
    {
      "epoch": 19.156215621562158,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0001624078209243429,
      "loss": 0.2303,
      "num_input_tokens_seen": 36747088,
      "step": 174130
    },
    {
      "epoch": 19.156765676567655,
      "grad_norm": 0.00616455078125,
      "learning_rate": 0.00016219655555705924,
      "loss": 0.2308,
      "num_input_tokens_seen": 36748080,
      "step": 174135
    },
    {
      "epoch": 19.157315731573156,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00016198542694268958,
      "loss": 0.2324,
      "num_input_tokens_seen": 36749136,
      "step": 174140
    },
    {
      "epoch": 19.157865786578657,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00016177443508318066,
      "loss": 0.2303,
      "num_input_tokens_seen": 36750224,
      "step": 174145
    },
    {
      "epoch": 19.15841584158416,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00016156357998047764,
      "loss": 0.2303,
      "num_input_tokens_seen": 36751280,
      "step": 174150
    },
    {
      "epoch": 19.15896589658966,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00016135286163652062,
      "loss": 0.2324,
      "num_input_tokens_seen": 36752336,
      "step": 174155
    },
    {
      "epoch": 19.15951595159516,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00016114228005325635,
      "loss": 0.2324,
      "num_input_tokens_seen": 36753392,
      "step": 174160
    },
    {
      "epoch": 19.16006600660066,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00016093183523262333,
      "loss": 0.2329,
      "num_input_tokens_seen": 36754448,
      "step": 174165
    },
    {
      "epoch": 19.16061606160616,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00016072152717655996,
      "loss": 0.2313,
      "num_input_tokens_seen": 36755504,
      "step": 174170
    },
    {
      "epoch": 19.16116611661166,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.00016051135588700472,
      "loss": 0.233,
      "num_input_tokens_seen": 36756528,
      "step": 174175
    },
    {
      "epoch": 19.16171617161716,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 0.00016030132136589603,
      "loss": 0.2314,
      "num_input_tokens_seen": 36757520,
      "step": 174180
    },
    {
      "epoch": 19.162266226622663,
      "grad_norm": 0.0020751953125,
      "learning_rate": 0.0001600914236151707,
      "loss": 0.234,
      "num_input_tokens_seen": 36758544,
      "step": 174185
    },
    {
      "epoch": 19.162816281628164,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00015988166263676218,
      "loss": 0.2324,
      "num_input_tokens_seen": 36759504,
      "step": 174190
    },
    {
      "epoch": 19.163366336633665,
      "grad_norm": 0.006103515625,
      "learning_rate": 0.00015967203843260223,
      "loss": 0.2308,
      "num_input_tokens_seen": 36760496,
      "step": 174195
    },
    {
      "epoch": 19.163916391639162,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0001594625510046227,
      "loss": 0.2319,
      "num_input_tokens_seen": 36761520,
      "step": 174200
    },
    {
      "epoch": 19.164466446644663,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00015925320035475697,
      "loss": 0.2329,
      "num_input_tokens_seen": 36762640,
      "step": 174205
    },
    {
      "epoch": 19.165016501650165,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.00015904398648493355,
      "loss": 0.2319,
      "num_input_tokens_seen": 36763696,
      "step": 174210
    },
    {
      "epoch": 19.165566556655666,
      "grad_norm": 0.00189208984375,
      "learning_rate": 0.0001588349093970809,
      "loss": 0.2319,
      "num_input_tokens_seen": 36764848,
      "step": 174215
    },
    {
      "epoch": 19.166116611661167,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0001586259690931241,
      "loss": 0.2303,
      "num_input_tokens_seen": 36765872,
      "step": 174220
    },
    {
      "epoch": 19.166666666666668,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 0.00015841716557499162,
      "loss": 0.2314,
      "num_input_tokens_seen": 36766928,
      "step": 174225
    },
    {
      "epoch": 19.16721672167217,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 0.0001582084988446053,
      "loss": 0.2335,
      "num_input_tokens_seen": 36767920,
      "step": 174230
    },
    {
      "epoch": 19.167766776677666,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00015799996890388855,
      "loss": 0.2298,
      "num_input_tokens_seen": 36768912,
      "step": 174235
    },
    {
      "epoch": 19.168316831683168,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.00015779157575476654,
      "loss": 0.2309,
      "num_input_tokens_seen": 36769936,
      "step": 174240
    },
    {
      "epoch": 19.16886688668867,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00015758331939915603,
      "loss": 0.2324,
      "num_input_tokens_seen": 36770992,
      "step": 174245
    },
    {
      "epoch": 19.16941694169417,
      "grad_norm": 0.00070953369140625,
      "learning_rate": 0.00015737519983897885,
      "loss": 0.2319,
      "num_input_tokens_seen": 36771952,
      "step": 174250
    },
    {
      "epoch": 19.16996699669967,
      "grad_norm": 0.00128173828125,
      "learning_rate": 0.00015716721707615178,
      "loss": 0.2314,
      "num_input_tokens_seen": 36772976,
      "step": 174255
    },
    {
      "epoch": 19.170517051705172,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0001569593711125916,
      "loss": 0.2324,
      "num_input_tokens_seen": 36774000,
      "step": 174260
    },
    {
      "epoch": 19.17106710671067,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.00015675166195021516,
      "loss": 0.2303,
      "num_input_tokens_seen": 36774992,
      "step": 174265
    },
    {
      "epoch": 19.17161716171617,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 0.0001565440895909359,
      "loss": 0.2324,
      "num_input_tokens_seen": 36776080,
      "step": 174270
    },
    {
      "epoch": 19.17216721672167,
      "grad_norm": 0.00115966796875,
      "learning_rate": 0.0001563366540366673,
      "loss": 0.2319,
      "num_input_tokens_seen": 36777072,
      "step": 174275
    },
    {
      "epoch": 19.172717271727173,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.00015612935528932115,
      "loss": 0.2319,
      "num_input_tokens_seen": 36778128,
      "step": 174280
    },
    {
      "epoch": 19.173267326732674,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00015592219335080758,
      "loss": 0.2309,
      "num_input_tokens_seen": 36779184,
      "step": 174285
    },
    {
      "epoch": 19.173817381738175,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00015571516822303677,
      "loss": 0.2324,
      "num_input_tokens_seen": 36780176,
      "step": 174290
    },
    {
      "epoch": 19.174367436743676,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00015550827990791548,
      "loss": 0.2303,
      "num_input_tokens_seen": 36781200,
      "step": 174295
    },
    {
      "epoch": 19.174917491749174,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.0001553015284073522,
      "loss": 0.2309,
      "num_input_tokens_seen": 36782192,
      "step": 174300
    },
    {
      "epoch": 19.175467546754675,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00015509491372325213,
      "loss": 0.2308,
      "num_input_tokens_seen": 36783280,
      "step": 174305
    },
    {
      "epoch": 19.176017601760176,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.000154888435857517,
      "loss": 0.2319,
      "num_input_tokens_seen": 36784336,
      "step": 174310
    },
    {
      "epoch": 19.176567656765677,
      "grad_norm": 0.00152587890625,
      "learning_rate": 0.00015468209481205364,
      "loss": 0.2319,
      "num_input_tokens_seen": 36785360,
      "step": 174315
    },
    {
      "epoch": 19.177117711771178,
      "grad_norm": 0.00070953369140625,
      "learning_rate": 0.00015447589058876054,
      "loss": 0.2314,
      "num_input_tokens_seen": 36786448,
      "step": 174320
    },
    {
      "epoch": 19.17766776677668,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00015426982318954117,
      "loss": 0.2324,
      "num_input_tokens_seen": 36787504,
      "step": 174325
    },
    {
      "epoch": 19.178217821782177,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00015406389261629237,
      "loss": 0.2324,
      "num_input_tokens_seen": 36788656,
      "step": 174330
    },
    {
      "epoch": 19.178767876787678,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0001538580988709126,
      "loss": 0.2314,
      "num_input_tokens_seen": 36789680,
      "step": 174335
    },
    {
      "epoch": 19.17931793179318,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0001536524419552987,
      "loss": 0.2329,
      "num_input_tokens_seen": 36790704,
      "step": 174340
    },
    {
      "epoch": 19.17986798679868,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 0.0001534469218713458,
      "loss": 0.2309,
      "num_input_tokens_seen": 36791792,
      "step": 174345
    },
    {
      "epoch": 19.18041804180418,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.000153241538620949,
      "loss": 0.2309,
      "num_input_tokens_seen": 36792816,
      "step": 174350
    },
    {
      "epoch": 19.180968096809682,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.00015303629220600188,
      "loss": 0.2356,
      "num_input_tokens_seen": 36793936,
      "step": 174355
    },
    {
      "epoch": 19.181518151815183,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.00015283118262839456,
      "loss": 0.2324,
      "num_input_tokens_seen": 36794992,
      "step": 174360
    },
    {
      "epoch": 19.18206820682068,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00015262620989001718,
      "loss": 0.2319,
      "num_input_tokens_seen": 36796016,
      "step": 174365
    },
    {
      "epoch": 19.182618261826182,
      "grad_norm": 0.0030364990234375,
      "learning_rate": 0.0001524213739927599,
      "loss": 0.2319,
      "num_input_tokens_seen": 36797072,
      "step": 174370
    },
    {
      "epoch": 19.183168316831683,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00015221667493850954,
      "loss": 0.2283,
      "num_input_tokens_seen": 36798192,
      "step": 174375
    },
    {
      "epoch": 19.183718371837184,
      "grad_norm": 0.006317138671875,
      "learning_rate": 0.0001520121127291546,
      "loss": 0.2283,
      "num_input_tokens_seen": 36799376,
      "step": 174380
    },
    {
      "epoch": 19.184268426842685,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.00015180768736657856,
      "loss": 0.2298,
      "num_input_tokens_seen": 36800400,
      "step": 174385
    },
    {
      "epoch": 19.184818481848186,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0001516033988526666,
      "loss": 0.2314,
      "num_input_tokens_seen": 36801424,
      "step": 174390
    },
    {
      "epoch": 19.185368536853684,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.00015139924718930218,
      "loss": 0.2319,
      "num_input_tokens_seen": 36802448,
      "step": 174395
    },
    {
      "epoch": 19.185918591859185,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00015119523237836386,
      "loss": 0.2324,
      "num_input_tokens_seen": 36803504,
      "step": 174400
    },
    {
      "epoch": 19.186468646864686,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.00015099135442173505,
      "loss": 0.2309,
      "num_input_tokens_seen": 36804560,
      "step": 174405
    },
    {
      "epoch": 19.187018701870187,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00015078761332129264,
      "loss": 0.2314,
      "num_input_tokens_seen": 36805616,
      "step": 174410
    },
    {
      "epoch": 19.187568756875688,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00015058400907891678,
      "loss": 0.2319,
      "num_input_tokens_seen": 36806672,
      "step": 174415
    },
    {
      "epoch": 19.18811881188119,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00015038054169648263,
      "loss": 0.2319,
      "num_input_tokens_seen": 36807632,
      "step": 174420
    },
    {
      "epoch": 19.18866886688669,
      "grad_norm": 0.0025634765625,
      "learning_rate": 0.00015017721117586535,
      "loss": 0.2314,
      "num_input_tokens_seen": 36808752,
      "step": 174425
    },
    {
      "epoch": 19.189218921892188,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00014997401751893678,
      "loss": 0.2319,
      "num_input_tokens_seen": 36809776,
      "step": 174430
    },
    {
      "epoch": 19.18976897689769,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00014977096072757377,
      "loss": 0.2309,
      "num_input_tokens_seen": 36810832,
      "step": 174435
    },
    {
      "epoch": 19.19031903190319,
      "grad_norm": 0.0018157958984375,
      "learning_rate": 0.0001495680408036465,
      "loss": 0.2308,
      "num_input_tokens_seen": 36811856,
      "step": 174440
    },
    {
      "epoch": 19.19086908690869,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00014936525774902508,
      "loss": 0.2319,
      "num_input_tokens_seen": 36812912,
      "step": 174445
    },
    {
      "epoch": 19.191419141914192,
      "grad_norm": 0.001983642578125,
      "learning_rate": 0.0001491626115655764,
      "loss": 0.234,
      "num_input_tokens_seen": 36813968,
      "step": 174450
    },
    {
      "epoch": 19.191969196919693,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.0001489601022551723,
      "loss": 0.2303,
      "num_input_tokens_seen": 36815088,
      "step": 174455
    },
    {
      "epoch": 19.19251925192519,
      "grad_norm": 0.000850677490234375,
      "learning_rate": 0.0001487577298196746,
      "loss": 0.2314,
      "num_input_tokens_seen": 36816080,
      "step": 174460
    },
    {
      "epoch": 19.193069306930692,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00014855549426095015,
      "loss": 0.2319,
      "num_input_tokens_seen": 36817040,
      "step": 174465
    },
    {
      "epoch": 19.193619361936193,
      "grad_norm": 0.002197265625,
      "learning_rate": 0.00014835339558086412,
      "loss": 0.2329,
      "num_input_tokens_seen": 36818096,
      "step": 174470
    },
    {
      "epoch": 19.194169416941694,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00014815143378127838,
      "loss": 0.2329,
      "num_input_tokens_seen": 36819184,
      "step": 174475
    },
    {
      "epoch": 19.194719471947195,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00014794960886405472,
      "loss": 0.2309,
      "num_input_tokens_seen": 36820208,
      "step": 174480
    },
    {
      "epoch": 19.195269526952696,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0001477479208310517,
      "loss": 0.2293,
      "num_input_tokens_seen": 36821328,
      "step": 174485
    },
    {
      "epoch": 19.195819581958197,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.00014754636968412947,
      "loss": 0.2329,
      "num_input_tokens_seen": 36822352,
      "step": 174490
    },
    {
      "epoch": 19.196369636963695,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00014734495542514657,
      "loss": 0.2314,
      "num_input_tokens_seen": 36823408,
      "step": 174495
    },
    {
      "epoch": 19.196919691969196,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.0001471436780559565,
      "loss": 0.2303,
      "num_input_tokens_seen": 36824464,
      "step": 174500
    },
    {
      "epoch": 19.197469746974697,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0001469425375784178,
      "loss": 0.2324,
      "num_input_tokens_seen": 36825584,
      "step": 174505
    },
    {
      "epoch": 19.198019801980198,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.00014674153399438228,
      "loss": 0.2319,
      "num_input_tokens_seen": 36826672,
      "step": 174510
    },
    {
      "epoch": 19.1985698569857,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.00014654066730570181,
      "loss": 0.2319,
      "num_input_tokens_seen": 36827632,
      "step": 174515
    },
    {
      "epoch": 19.1991199119912,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 0.0001463399375142299,
      "loss": 0.2309,
      "num_input_tokens_seen": 36828720,
      "step": 174520
    },
    {
      "epoch": 19.199669966996698,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.00014613934462181344,
      "loss": 0.2314,
      "num_input_tokens_seen": 36829808,
      "step": 174525
    },
    {
      "epoch": 19.2002200220022,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.00014593888863030424,
      "loss": 0.2329,
      "num_input_tokens_seen": 36830832,
      "step": 174530
    },
    {
      "epoch": 19.2007700770077,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00014573856954154917,
      "loss": 0.2314,
      "num_input_tokens_seen": 36831824,
      "step": 174535
    },
    {
      "epoch": 19.2013201320132,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 0.00014553838735739344,
      "loss": 0.2314,
      "num_input_tokens_seen": 36832848,
      "step": 174540
    },
    {
      "epoch": 19.201870187018702,
      "grad_norm": 0.002166748046875,
      "learning_rate": 0.00014533834207968222,
      "loss": 0.2288,
      "num_input_tokens_seen": 36833904,
      "step": 174545
    },
    {
      "epoch": 19.202420242024203,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00014513843371025904,
      "loss": 0.2329,
      "num_input_tokens_seen": 36834928,
      "step": 174550
    },
    {
      "epoch": 19.202970297029704,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.00014493866225096907,
      "loss": 0.2314,
      "num_input_tokens_seen": 36835920,
      "step": 174555
    },
    {
      "epoch": 19.203520352035202,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.0001447390277036492,
      "loss": 0.2298,
      "num_input_tokens_seen": 36836976,
      "step": 174560
    },
    {
      "epoch": 19.204070407040703,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.0001445395300701413,
      "loss": 0.2304,
      "num_input_tokens_seen": 36837968,
      "step": 174565
    },
    {
      "epoch": 19.204620462046204,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00014434016935228555,
      "loss": 0.2324,
      "num_input_tokens_seen": 36838928,
      "step": 174570
    },
    {
      "epoch": 19.205170517051705,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00014414094555191713,
      "loss": 0.2309,
      "num_input_tokens_seen": 36840016,
      "step": 174575
    },
    {
      "epoch": 19.205720572057206,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00014394185867087295,
      "loss": 0.2308,
      "num_input_tokens_seen": 36841072,
      "step": 174580
    },
    {
      "epoch": 19.206270627062707,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 0.00014374290871098816,
      "loss": 0.2303,
      "num_input_tokens_seen": 36842096,
      "step": 174585
    },
    {
      "epoch": 19.206820682068205,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00014354409567409796,
      "loss": 0.2303,
      "num_input_tokens_seen": 36843184,
      "step": 174590
    },
    {
      "epoch": 19.207370737073706,
      "grad_norm": 0.00046539306640625,
      "learning_rate": 0.00014334541956203094,
      "loss": 0.2308,
      "num_input_tokens_seen": 36844304,
      "step": 174595
    },
    {
      "epoch": 19.207920792079207,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 0.0001431468803766206,
      "loss": 0.2298,
      "num_input_tokens_seen": 36845360,
      "step": 174600
    },
    {
      "epoch": 19.20847084708471,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.00014294847811969713,
      "loss": 0.2324,
      "num_input_tokens_seen": 36846416,
      "step": 174605
    },
    {
      "epoch": 19.20902090209021,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0001427502127930874,
      "loss": 0.2319,
      "num_input_tokens_seen": 36847536,
      "step": 174610
    },
    {
      "epoch": 19.20957095709571,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.00014255208439862166,
      "loss": 0.2329,
      "num_input_tokens_seen": 36848624,
      "step": 174615
    },
    {
      "epoch": 19.21012101210121,
      "grad_norm": 0.005859375,
      "learning_rate": 0.0001423540929381234,
      "loss": 0.2309,
      "num_input_tokens_seen": 36849712,
      "step": 174620
    },
    {
      "epoch": 19.21067106710671,
      "grad_norm": 0.00093841552734375,
      "learning_rate": 0.00014215623841341618,
      "loss": 0.2319,
      "num_input_tokens_seen": 36850832,
      "step": 174625
    },
    {
      "epoch": 19.21122112211221,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00014195852082632686,
      "loss": 0.2319,
      "num_input_tokens_seen": 36851952,
      "step": 174630
    },
    {
      "epoch": 19.21177117711771,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00014176094017867734,
      "loss": 0.2309,
      "num_input_tokens_seen": 36853072,
      "step": 174635
    },
    {
      "epoch": 19.212321232123212,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0001415634964722845,
      "loss": 0.2324,
      "num_input_tokens_seen": 36854192,
      "step": 174640
    },
    {
      "epoch": 19.212871287128714,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00014136618970897352,
      "loss": 0.2324,
      "num_input_tokens_seen": 36855280,
      "step": 174645
    },
    {
      "epoch": 19.213421342134215,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 0.00014116901989055962,
      "loss": 0.2288,
      "num_input_tokens_seen": 36856400,
      "step": 174650
    },
    {
      "epoch": 19.213971397139716,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.00014097198701886303,
      "loss": 0.2298,
      "num_input_tokens_seen": 36857392,
      "step": 174655
    },
    {
      "epoch": 19.214521452145213,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00014077509109569562,
      "loss": 0.2314,
      "num_input_tokens_seen": 36858448,
      "step": 174660
    },
    {
      "epoch": 19.215071507150714,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00014057833212287261,
      "loss": 0.234,
      "num_input_tokens_seen": 36859600,
      "step": 174665
    },
    {
      "epoch": 19.215621562156215,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 0.0001403817101022109,
      "loss": 0.2319,
      "num_input_tokens_seen": 36860656,
      "step": 174670
    },
    {
      "epoch": 19.216171617161717,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.00014018522503552065,
      "loss": 0.2314,
      "num_input_tokens_seen": 36861744,
      "step": 174675
    },
    {
      "epoch": 19.216721672167218,
      "grad_norm": 0.0021209716796875,
      "learning_rate": 0.00013998887692461215,
      "loss": 0.2308,
      "num_input_tokens_seen": 36862768,
      "step": 174680
    },
    {
      "epoch": 19.21727172717272,
      "grad_norm": 0.001129150390625,
      "learning_rate": 0.00013979266577129722,
      "loss": 0.2329,
      "num_input_tokens_seen": 36863824,
      "step": 174685
    },
    {
      "epoch": 19.217821782178216,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.00013959659157738113,
      "loss": 0.2303,
      "num_input_tokens_seen": 36864944,
      "step": 174690
    },
    {
      "epoch": 19.218371837183717,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0001394006543446724,
      "loss": 0.2298,
      "num_input_tokens_seen": 36866000,
      "step": 174695
    },
    {
      "epoch": 19.21892189218922,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00013920485407497627,
      "loss": 0.2308,
      "num_input_tokens_seen": 36866992,
      "step": 174700
    },
    {
      "epoch": 19.21947194719472,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00013900919077009964,
      "loss": 0.2319,
      "num_input_tokens_seen": 36868016,
      "step": 174705
    },
    {
      "epoch": 19.22002200220022,
      "grad_norm": 0.000942230224609375,
      "learning_rate": 0.0001388136644318444,
      "loss": 0.2308,
      "num_input_tokens_seen": 36869040,
      "step": 174710
    },
    {
      "epoch": 19.22057205720572,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.0001386182750620124,
      "loss": 0.2303,
      "num_input_tokens_seen": 36870064,
      "step": 174715
    },
    {
      "epoch": 19.221122112211223,
      "grad_norm": 0.005859375,
      "learning_rate": 0.00013842302266240392,
      "loss": 0.2319,
      "num_input_tokens_seen": 36871152,
      "step": 174720
    },
    {
      "epoch": 19.22167216721672,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00013822790723481748,
      "loss": 0.2319,
      "num_input_tokens_seen": 36872272,
      "step": 174725
    },
    {
      "epoch": 19.22222222222222,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00013803292878105666,
      "loss": 0.2314,
      "num_input_tokens_seen": 36873328,
      "step": 174730
    },
    {
      "epoch": 19.222772277227723,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00013783808730291334,
      "loss": 0.2324,
      "num_input_tokens_seen": 36874384,
      "step": 174735
    },
    {
      "epoch": 19.223322332233224,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.00013764338280218447,
      "loss": 0.2308,
      "num_input_tokens_seen": 36875472,
      "step": 174740
    },
    {
      "epoch": 19.223872387238725,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00013744881528066522,
      "loss": 0.2329,
      "num_input_tokens_seen": 36876528,
      "step": 174745
    },
    {
      "epoch": 19.224422442244226,
      "grad_norm": 0.00115203857421875,
      "learning_rate": 0.0001372543847401475,
      "loss": 0.2319,
      "num_input_tokens_seen": 36877616,
      "step": 174750
    },
    {
      "epoch": 19.224972497249723,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0001370600911824249,
      "loss": 0.2324,
      "num_input_tokens_seen": 36878736,
      "step": 174755
    },
    {
      "epoch": 19.225522552255224,
      "grad_norm": 0.0010986328125,
      "learning_rate": 0.00013686593460928762,
      "loss": 0.2314,
      "num_input_tokens_seen": 36879728,
      "step": 174760
    },
    {
      "epoch": 19.226072607260726,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00013667191502252595,
      "loss": 0.2335,
      "num_input_tokens_seen": 36880784,
      "step": 174765
    },
    {
      "epoch": 19.226622662266227,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.00013647803242392676,
      "loss": 0.2324,
      "num_input_tokens_seen": 36881776,
      "step": 174770
    },
    {
      "epoch": 19.227172717271728,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 0.00013628428681527525,
      "loss": 0.2298,
      "num_input_tokens_seen": 36882832,
      "step": 174775
    },
    {
      "epoch": 19.22772277227723,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.00013609067819836173,
      "loss": 0.2324,
      "num_input_tokens_seen": 36883920,
      "step": 174780
    },
    {
      "epoch": 19.22827282728273,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.0001358972065749664,
      "loss": 0.2314,
      "num_input_tokens_seen": 36885008,
      "step": 174785
    },
    {
      "epoch": 19.228822882288227,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0001357038719468745,
      "loss": 0.2319,
      "num_input_tokens_seen": 36886064,
      "step": 174790
    },
    {
      "epoch": 19.22937293729373,
      "grad_norm": 0.001373291015625,
      "learning_rate": 0.00013551067431586794,
      "loss": 0.2309,
      "num_input_tokens_seen": 36887120,
      "step": 174795
    },
    {
      "epoch": 19.22992299229923,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 0.00013531761368372697,
      "loss": 0.2324,
      "num_input_tokens_seen": 36888208,
      "step": 174800
    },
    {
      "epoch": 19.23047304730473,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00013512469005223015,
      "loss": 0.2324,
      "num_input_tokens_seen": 36889264,
      "step": 174805
    },
    {
      "epoch": 19.231023102310232,
      "grad_norm": 0.01116943359375,
      "learning_rate": 0.00013493190342315608,
      "loss": 0.2314,
      "num_input_tokens_seen": 36890256,
      "step": 174810
    },
    {
      "epoch": 19.231573157315733,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00013473925379828167,
      "loss": 0.2309,
      "num_input_tokens_seen": 36891280,
      "step": 174815
    },
    {
      "epoch": 19.23212321232123,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0001345467411793838,
      "loss": 0.2309,
      "num_input_tokens_seen": 36892304,
      "step": 174820
    },
    {
      "epoch": 19.23267326732673,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 0.00013435436556823277,
      "loss": 0.234,
      "num_input_tokens_seen": 36893424,
      "step": 174825
    },
    {
      "epoch": 19.233223322332233,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0001341621269666071,
      "loss": 0.2303,
      "num_input_tokens_seen": 36894512,
      "step": 174830
    },
    {
      "epoch": 19.233773377337734,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00013397002537627546,
      "loss": 0.2309,
      "num_input_tokens_seen": 36895568,
      "step": 174835
    },
    {
      "epoch": 19.234323432343235,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.00013377806079900632,
      "loss": 0.2319,
      "num_input_tokens_seen": 36896624,
      "step": 174840
    },
    {
      "epoch": 19.234873487348736,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.00013358623323657336,
      "loss": 0.2314,
      "num_input_tokens_seen": 36897744,
      "step": 174845
    },
    {
      "epoch": 19.235423542354237,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0001333945426907418,
      "loss": 0.2303,
      "num_input_tokens_seen": 36898832,
      "step": 174850
    },
    {
      "epoch": 19.235973597359735,
      "grad_norm": 0.00069427490234375,
      "learning_rate": 0.00013320298916328022,
      "loss": 0.2308,
      "num_input_tokens_seen": 36899824,
      "step": 174855
    },
    {
      "epoch": 19.236523652365236,
      "grad_norm": 0.01153564453125,
      "learning_rate": 0.0001330115726559522,
      "loss": 0.2309,
      "num_input_tokens_seen": 36900912,
      "step": 174860
    },
    {
      "epoch": 19.237073707370737,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00013282029317052135,
      "loss": 0.2329,
      "num_input_tokens_seen": 36901968,
      "step": 174865
    },
    {
      "epoch": 19.237623762376238,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.00013262915070875293,
      "loss": 0.2308,
      "num_input_tokens_seen": 36903024,
      "step": 174870
    },
    {
      "epoch": 19.23817381738174,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00013243814527240882,
      "loss": 0.2335,
      "num_input_tokens_seen": 36904112,
      "step": 174875
    },
    {
      "epoch": 19.23872387238724,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.000132247276863246,
      "loss": 0.2319,
      "num_input_tokens_seen": 36905104,
      "step": 174880
    },
    {
      "epoch": 19.239273927392738,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.00013205654548302802,
      "loss": 0.2319,
      "num_input_tokens_seen": 36906192,
      "step": 174885
    },
    {
      "epoch": 19.23982398239824,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.0001318659511335085,
      "loss": 0.2335,
      "num_input_tokens_seen": 36907248,
      "step": 174890
    },
    {
      "epoch": 19.24037403740374,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.0001316754938164477,
      "loss": 0.2329,
      "num_input_tokens_seen": 36908368,
      "step": 174895
    },
    {
      "epoch": 19.24092409240924,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.00013148517353359757,
      "loss": 0.2319,
      "num_input_tokens_seen": 36909424,
      "step": 174900
    },
    {
      "epoch": 19.241474147414742,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.00013129499028671497,
      "loss": 0.2309,
      "num_input_tokens_seen": 36910480,
      "step": 174905
    },
    {
      "epoch": 19.242024202420243,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00013110494407755192,
      "loss": 0.2319,
      "num_input_tokens_seen": 36911536,
      "step": 174910
    },
    {
      "epoch": 19.242574257425744,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.00013091503490785862,
      "loss": 0.2324,
      "num_input_tokens_seen": 36912624,
      "step": 174915
    },
    {
      "epoch": 19.24312431243124,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.00013072526277938868,
      "loss": 0.2309,
      "num_input_tokens_seen": 36913744,
      "step": 174920
    },
    {
      "epoch": 19.243674367436743,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00013053562769388738,
      "loss": 0.2319,
      "num_input_tokens_seen": 36914800,
      "step": 174925
    },
    {
      "epoch": 19.244224422442244,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00013034612965310332,
      "loss": 0.2309,
      "num_input_tokens_seen": 36915856,
      "step": 174930
    },
    {
      "epoch": 19.244774477447745,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00013015676865878677,
      "loss": 0.2314,
      "num_input_tokens_seen": 36916944,
      "step": 174935
    },
    {
      "epoch": 19.245324532453246,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00012996754471267802,
      "loss": 0.2308,
      "num_input_tokens_seen": 36917968,
      "step": 174940
    },
    {
      "epoch": 19.245874587458747,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.00012977845781652397,
      "loss": 0.2329,
      "num_input_tokens_seen": 36919024,
      "step": 174945
    },
    {
      "epoch": 19.246424642464245,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.0001295895079720649,
      "loss": 0.2314,
      "num_input_tokens_seen": 36920112,
      "step": 174950
    },
    {
      "epoch": 19.246974697469746,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00012940069518104446,
      "loss": 0.2308,
      "num_input_tokens_seen": 36921168,
      "step": 174955
    },
    {
      "epoch": 19.247524752475247,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00012921201944520288,
      "loss": 0.2319,
      "num_input_tokens_seen": 36922320,
      "step": 174960
    },
    {
      "epoch": 19.248074807480748,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.0001290234807662771,
      "loss": 0.2324,
      "num_input_tokens_seen": 36923344,
      "step": 174965
    },
    {
      "epoch": 19.24862486248625,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0001288350791460091,
      "loss": 0.2308,
      "num_input_tokens_seen": 36924400,
      "step": 174970
    },
    {
      "epoch": 19.24917491749175,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0001286468145861308,
      "loss": 0.2308,
      "num_input_tokens_seen": 36925456,
      "step": 174975
    },
    {
      "epoch": 19.24972497249725,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.00012845868708837748,
      "loss": 0.2319,
      "num_input_tokens_seen": 36926544,
      "step": 174980
    },
    {
      "epoch": 19.25027502750275,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00012827069665448442,
      "loss": 0.2319,
      "num_input_tokens_seen": 36927600,
      "step": 174985
    },
    {
      "epoch": 19.25082508250825,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00012808284328618525,
      "loss": 0.2319,
      "num_input_tokens_seen": 36928624,
      "step": 174990
    },
    {
      "epoch": 19.25137513751375,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.0001278951269852102,
      "loss": 0.234,
      "num_input_tokens_seen": 36929712,
      "step": 174995
    },
    {
      "epoch": 19.251925192519252,
      "grad_norm": 0.00167083740234375,
      "learning_rate": 0.00012770754775328963,
      "loss": 0.2335,
      "num_input_tokens_seen": 36930768,
      "step": 175000
    },
    {
      "epoch": 19.252475247524753,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.0001275201055921521,
      "loss": 0.2293,
      "num_input_tokens_seen": 36931888,
      "step": 175005
    },
    {
      "epoch": 19.253025302530254,
      "grad_norm": 0.00164031982421875,
      "learning_rate": 0.00012733280050352457,
      "loss": 0.2309,
      "num_input_tokens_seen": 36932880,
      "step": 175010
    },
    {
      "epoch": 19.253575357535752,
      "grad_norm": 0.001068115234375,
      "learning_rate": 0.00012714563248913401,
      "loss": 0.2324,
      "num_input_tokens_seen": 36933904,
      "step": 175015
    },
    {
      "epoch": 19.254125412541253,
      "grad_norm": 0.005950927734375,
      "learning_rate": 0.0001269586015507057,
      "loss": 0.2314,
      "num_input_tokens_seen": 36935088,
      "step": 175020
    },
    {
      "epoch": 19.254675467546754,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00012677170768996492,
      "loss": 0.2303,
      "num_input_tokens_seen": 36936144,
      "step": 175025
    },
    {
      "epoch": 19.255225522552255,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0001265849509086303,
      "loss": 0.2314,
      "num_input_tokens_seen": 36937168,
      "step": 175030
    },
    {
      "epoch": 19.255775577557756,
      "grad_norm": 0.001617431640625,
      "learning_rate": 0.00012639833120842546,
      "loss": 0.2319,
      "num_input_tokens_seen": 36938224,
      "step": 175035
    },
    {
      "epoch": 19.256325632563257,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0001262118485910707,
      "loss": 0.2319,
      "num_input_tokens_seen": 36939216,
      "step": 175040
    },
    {
      "epoch": 19.25687568756876,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00012602550305828298,
      "loss": 0.2303,
      "num_input_tokens_seen": 36940336,
      "step": 175045
    },
    {
      "epoch": 19.257425742574256,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00012583929461178255,
      "loss": 0.2303,
      "num_input_tokens_seen": 36941360,
      "step": 175050
    },
    {
      "epoch": 19.257975797579757,
      "grad_norm": 0.00173187255859375,
      "learning_rate": 0.00012565322325328309,
      "loss": 0.233,
      "num_input_tokens_seen": 36942384,
      "step": 175055
    },
    {
      "epoch": 19.258525852585258,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00012546728898450154,
      "loss": 0.2313,
      "num_input_tokens_seen": 36943504,
      "step": 175060
    },
    {
      "epoch": 19.25907590759076,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 0.00012528149180714986,
      "loss": 0.2319,
      "num_input_tokens_seen": 36944496,
      "step": 175065
    },
    {
      "epoch": 19.25962596259626,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0001250958317229417,
      "loss": 0.2314,
      "num_input_tokens_seen": 36945552,
      "step": 175070
    },
    {
      "epoch": 19.26017601760176,
      "grad_norm": 0.01080322265625,
      "learning_rate": 0.00012491030873358565,
      "loss": 0.2308,
      "num_input_tokens_seen": 36946608,
      "step": 175075
    },
    {
      "epoch": 19.260726072607262,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0001247249228407954,
      "loss": 0.2308,
      "num_input_tokens_seen": 36947664,
      "step": 175080
    },
    {
      "epoch": 19.26127612761276,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00012453967404627787,
      "loss": 0.2319,
      "num_input_tokens_seen": 36948720,
      "step": 175085
    },
    {
      "epoch": 19.26182618261826,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.00012435456235173836,
      "loss": 0.2329,
      "num_input_tokens_seen": 36949744,
      "step": 175090
    },
    {
      "epoch": 19.262376237623762,
      "grad_norm": 0.000568389892578125,
      "learning_rate": 0.00012416958775888554,
      "loss": 0.2319,
      "num_input_tokens_seen": 36950832,
      "step": 175095
    },
    {
      "epoch": 19.262926292629263,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 0.00012398475026942468,
      "loss": 0.2324,
      "num_input_tokens_seen": 36951888,
      "step": 175100
    },
    {
      "epoch": 19.263476347634764,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.0001238000498850561,
      "loss": 0.2319,
      "num_input_tokens_seen": 36952944,
      "step": 175105
    },
    {
      "epoch": 19.264026402640265,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.00012361548660748677,
      "loss": 0.2329,
      "num_input_tokens_seen": 36953968,
      "step": 175110
    },
    {
      "epoch": 19.264576457645763,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.00012343106043841368,
      "loss": 0.2314,
      "num_input_tokens_seen": 36955024,
      "step": 175115
    },
    {
      "epoch": 19.265126512651264,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.0001232467713795371,
      "loss": 0.2335,
      "num_input_tokens_seen": 36956080,
      "step": 175120
    },
    {
      "epoch": 19.265676567656765,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.00012306261943255902,
      "loss": 0.2303,
      "num_input_tokens_seen": 36957072,
      "step": 175125
    },
    {
      "epoch": 19.266226622662266,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00012287860459917144,
      "loss": 0.2309,
      "num_input_tokens_seen": 36958128,
      "step": 175130
    },
    {
      "epoch": 19.266776677667767,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 0.00012269472688107463,
      "loss": 0.2298,
      "num_input_tokens_seen": 36959120,
      "step": 175135
    },
    {
      "epoch": 19.26732673267327,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 0.0001225109862799606,
      "loss": 0.2314,
      "num_input_tokens_seen": 36960144,
      "step": 175140
    },
    {
      "epoch": 19.26787678767877,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.00012232738279752465,
      "loss": 0.2304,
      "num_input_tokens_seen": 36961136,
      "step": 175145
    },
    {
      "epoch": 19.268426842684267,
      "grad_norm": 0.002227783203125,
      "learning_rate": 0.00012214391643545873,
      "loss": 0.2303,
      "num_input_tokens_seen": 36962160,
      "step": 175150
    },
    {
      "epoch": 19.268976897689768,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00012196058719545321,
      "loss": 0.2319,
      "num_input_tokens_seen": 36963216,
      "step": 175155
    },
    {
      "epoch": 19.26952695269527,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00012177739507919672,
      "loss": 0.234,
      "num_input_tokens_seen": 36964240,
      "step": 175160
    },
    {
      "epoch": 19.27007700770077,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.00012159434008837954,
      "loss": 0.234,
      "num_input_tokens_seen": 36965264,
      "step": 175165
    },
    {
      "epoch": 19.27062706270627,
      "grad_norm": 0.0059814453125,
      "learning_rate": 0.00012141142222468704,
      "loss": 0.2319,
      "num_input_tokens_seen": 36966352,
      "step": 175170
    },
    {
      "epoch": 19.271177117711773,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 0.00012122864148980617,
      "loss": 0.2314,
      "num_input_tokens_seen": 36967440,
      "step": 175175
    },
    {
      "epoch": 19.27172717271727,
      "grad_norm": 0.000911712646484375,
      "learning_rate": 0.00012104599788542225,
      "loss": 0.2319,
      "num_input_tokens_seen": 36968464,
      "step": 175180
    },
    {
      "epoch": 19.27227722772277,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 0.00012086349141321727,
      "loss": 0.2324,
      "num_input_tokens_seen": 36969488,
      "step": 175185
    },
    {
      "epoch": 19.272827282728272,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00012068112207487325,
      "loss": 0.2314,
      "num_input_tokens_seen": 36970608,
      "step": 175190
    },
    {
      "epoch": 19.273377337733773,
      "grad_norm": 0.001251220703125,
      "learning_rate": 0.00012049888987207213,
      "loss": 0.2308,
      "num_input_tokens_seen": 36971600,
      "step": 175195
    },
    {
      "epoch": 19.273927392739274,
      "grad_norm": 0.00164794921875,
      "learning_rate": 0.0001203167948064926,
      "loss": 0.2314,
      "num_input_tokens_seen": 36972752,
      "step": 175200
    },
    {
      "epoch": 19.274477447744776,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00012013483687981329,
      "loss": 0.2298,
      "num_input_tokens_seen": 36973808,
      "step": 175205
    },
    {
      "epoch": 19.275027502750277,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.00011995301609371289,
      "loss": 0.2319,
      "num_input_tokens_seen": 36974832,
      "step": 175210
    },
    {
      "epoch": 19.275577557755774,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.00011977133244986338,
      "loss": 0.2303,
      "num_input_tokens_seen": 36975920,
      "step": 175215
    },
    {
      "epoch": 19.276127612761275,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00011958978594994007,
      "loss": 0.2314,
      "num_input_tokens_seen": 36977008,
      "step": 175220
    },
    {
      "epoch": 19.276677667766776,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00011940837659561997,
      "loss": 0.234,
      "num_input_tokens_seen": 36978064,
      "step": 175225
    },
    {
      "epoch": 19.277227722772277,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00011922710438857175,
      "loss": 0.2293,
      "num_input_tokens_seen": 36979120,
      "step": 175230
    },
    {
      "epoch": 19.27777777777778,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 0.0001190459693304674,
      "loss": 0.2304,
      "num_input_tokens_seen": 36980176,
      "step": 175235
    },
    {
      "epoch": 19.27832783278328,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00011886497142297558,
      "loss": 0.2313,
      "num_input_tokens_seen": 36981232,
      "step": 175240
    },
    {
      "epoch": 19.278877887788777,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 0.00011868411066776329,
      "loss": 0.2313,
      "num_input_tokens_seen": 36982256,
      "step": 175245
    },
    {
      "epoch": 19.27942794279428,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 0.0001185033870664992,
      "loss": 0.2309,
      "num_input_tokens_seen": 36983344,
      "step": 175250
    },
    {
      "epoch": 19.27997799779978,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00011832280062084865,
      "loss": 0.2314,
      "num_input_tokens_seen": 36984432,
      "step": 175255
    },
    {
      "epoch": 19.28052805280528,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.0001181423513324753,
      "loss": 0.2314,
      "num_input_tokens_seen": 36985488,
      "step": 175260
    },
    {
      "epoch": 19.28107810781078,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.00011796203920304449,
      "loss": 0.2329,
      "num_input_tokens_seen": 36986640,
      "step": 175265
    },
    {
      "epoch": 19.281628162816283,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 0.00011778186423421654,
      "loss": 0.2319,
      "num_input_tokens_seen": 36987696,
      "step": 175270
    },
    {
      "epoch": 19.282178217821784,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 0.00011760182642765015,
      "loss": 0.2329,
      "num_input_tokens_seen": 36988784,
      "step": 175275
    },
    {
      "epoch": 19.28272827282728,
      "grad_norm": 0.005889892578125,
      "learning_rate": 0.00011742192578500732,
      "loss": 0.2314,
      "num_input_tokens_seen": 36989904,
      "step": 175280
    },
    {
      "epoch": 19.283278327832782,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 0.00011724216230794504,
      "loss": 0.2319,
      "num_input_tokens_seen": 36990992,
      "step": 175285
    },
    {
      "epoch": 19.283828382838283,
      "grad_norm": 0.006011962890625,
      "learning_rate": 0.000117062535998122,
      "loss": 0.2324,
      "num_input_tokens_seen": 36992080,
      "step": 175290
    },
    {
      "epoch": 19.284378437843785,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00011688304685719019,
      "loss": 0.2324,
      "num_input_tokens_seen": 36993104,
      "step": 175295
    },
    {
      "epoch": 19.284928492849286,
      "grad_norm": 0.005340576171875,
      "learning_rate": 0.00011670369488680665,
      "loss": 0.2304,
      "num_input_tokens_seen": 36994224,
      "step": 175300
    },
    {
      "epoch": 19.285478547854787,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 0.0001165244800886217,
      "loss": 0.2308,
      "num_input_tokens_seen": 36995248,
      "step": 175305
    },
    {
      "epoch": 19.286028602860284,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00011634540246429069,
      "loss": 0.2319,
      "num_input_tokens_seen": 36996368,
      "step": 175310
    },
    {
      "epoch": 19.286578657865785,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 0.00011616646201546232,
      "loss": 0.2319,
      "num_input_tokens_seen": 36997456,
      "step": 175315
    },
    {
      "epoch": 19.287128712871286,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.00011598765874378524,
      "loss": 0.2335,
      "num_input_tokens_seen": 36998544,
      "step": 175320
    },
    {
      "epoch": 19.287678767876788,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.00011580899265090815,
      "loss": 0.2303,
      "num_input_tokens_seen": 36999536,
      "step": 175325
    },
    {
      "epoch": 19.28822882288229,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.0001156304637384764,
      "loss": 0.2319,
      "num_input_tokens_seen": 37000592,
      "step": 175330
    },
    {
      "epoch": 19.28877887788779,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00011545207200813701,
      "loss": 0.2319,
      "num_input_tokens_seen": 37001616,
      "step": 175335
    },
    {
      "epoch": 19.28932893289329,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.00011527381746153364,
      "loss": 0.2319,
      "num_input_tokens_seen": 37002704,
      "step": 175340
    },
    {
      "epoch": 19.28987898789879,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 0.00011509570010030834,
      "loss": 0.2324,
      "num_input_tokens_seen": 37003760,
      "step": 175345
    },
    {
      "epoch": 19.29042904290429,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00011491771992610477,
      "loss": 0.2303,
      "num_input_tokens_seen": 37004816,
      "step": 175350
    },
    {
      "epoch": 19.29097909790979,
      "grad_norm": 0.00098419189453125,
      "learning_rate": 0.00011473987694056164,
      "loss": 0.2308,
      "num_input_tokens_seen": 37005872,
      "step": 175355
    },
    {
      "epoch": 19.29152915291529,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00011456217114531764,
      "loss": 0.2319,
      "num_input_tokens_seen": 37006896,
      "step": 175360
    },
    {
      "epoch": 19.292079207920793,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 0.0001143846025420131,
      "loss": 0.2308,
      "num_input_tokens_seen": 37007984,
      "step": 175365
    },
    {
      "epoch": 19.292629262926294,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00011420717113228007,
      "loss": 0.2324,
      "num_input_tokens_seen": 37008976,
      "step": 175370
    },
    {
      "epoch": 19.293179317931795,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.0001140298769177589,
      "loss": 0.2314,
      "num_input_tokens_seen": 37010096,
      "step": 175375
    },
    {
      "epoch": 19.293729372937293,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00011385271990008161,
      "loss": 0.2314,
      "num_input_tokens_seen": 37011184,
      "step": 175380
    },
    {
      "epoch": 19.294279427942794,
      "grad_norm": 0.011474609375,
      "learning_rate": 0.00011367570008088023,
      "loss": 0.2303,
      "num_input_tokens_seen": 37012208,
      "step": 175385
    },
    {
      "epoch": 19.294829482948295,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00011349881746178513,
      "loss": 0.2319,
      "num_input_tokens_seen": 37013296,
      "step": 175390
    },
    {
      "epoch": 19.295379537953796,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00011332207204443,
      "loss": 0.2314,
      "num_input_tokens_seen": 37014352,
      "step": 175395
    },
    {
      "epoch": 19.295929592959297,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 0.00011314546383044188,
      "loss": 0.2324,
      "num_input_tokens_seen": 37015344,
      "step": 175400
    },
    {
      "epoch": 19.296479647964798,
      "grad_norm": 0.00567626953125,
      "learning_rate": 0.00011296899282144945,
      "loss": 0.2319,
      "num_input_tokens_seen": 37016400,
      "step": 175405
    },
    {
      "epoch": 19.297029702970296,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00011279265901907642,
      "loss": 0.2303,
      "num_input_tokens_seen": 37017424,
      "step": 175410
    },
    {
      "epoch": 19.297579757975797,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00011261646242495149,
      "loss": 0.2309,
      "num_input_tokens_seen": 37018480,
      "step": 175415
    },
    {
      "epoch": 19.298129812981298,
      "grad_norm": 0.00640869140625,
      "learning_rate": 0.00011244040304069502,
      "loss": 0.2298,
      "num_input_tokens_seen": 37019536,
      "step": 175420
    },
    {
      "epoch": 19.2986798679868,
      "grad_norm": 0.006072998046875,
      "learning_rate": 0.00011226448086793238,
      "loss": 0.2314,
      "num_input_tokens_seen": 37020592,
      "step": 175425
    },
    {
      "epoch": 19.2992299229923,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 0.00011208869590828563,
      "loss": 0.2298,
      "num_input_tokens_seen": 37021616,
      "step": 175430
    },
    {
      "epoch": 19.2997799779978,
      "grad_norm": 0.00531005859375,
      "learning_rate": 0.00011191304816337177,
      "loss": 0.2288,
      "num_input_tokens_seen": 37022704,
      "step": 175435
    },
    {
      "epoch": 19.300330033003302,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.0001117375376348112,
      "loss": 0.2319,
      "num_input_tokens_seen": 37023760,
      "step": 175440
    },
    {
      "epoch": 19.3008800880088,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.0001115621643242226,
      "loss": 0.2319,
      "num_input_tokens_seen": 37024784,
      "step": 175445
    },
    {
      "epoch": 19.3014301430143,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.00011138692823322137,
      "loss": 0.2303,
      "num_input_tokens_seen": 37025808,
      "step": 175450
    },
    {
      "epoch": 19.301980198019802,
      "grad_norm": 0.005828857421875,
      "learning_rate": 0.00011121182936342288,
      "loss": 0.2324,
      "num_input_tokens_seen": 37026832,
      "step": 175455
    },
    {
      "epoch": 19.302530253025303,
      "grad_norm": 0.0013427734375,
      "learning_rate": 0.00011103686771643916,
      "loss": 0.2303,
      "num_input_tokens_seen": 37027952,
      "step": 175460
    },
    {
      "epoch": 19.303080308030804,
      "grad_norm": 0.005523681640625,
      "learning_rate": 0.0001108620432938856,
      "loss": 0.2314,
      "num_input_tokens_seen": 37029040,
      "step": 175465
    },
    {
      "epoch": 19.303630363036305,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 0.00011068735609737257,
      "loss": 0.2288,
      "num_input_tokens_seen": 37030064,
      "step": 175470
    },
    {
      "epoch": 19.304180418041803,
      "grad_norm": 0.00579833984375,
      "learning_rate": 0.0001105128061285071,
      "loss": 0.2335,
      "num_input_tokens_seen": 37031120,
      "step": 175475
    },
    {
      "epoch": 19.304730473047304,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 0.00011033839338890294,
      "loss": 0.2304,
      "num_input_tokens_seen": 37032240,
      "step": 175480
    },
    {
      "epoch": 19.305280528052805,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00011016411788016377,
      "loss": 0.2314,
      "num_input_tokens_seen": 37033264,
      "step": 175485
    },
    {
      "epoch": 19.305830583058306,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 0.00010998997960389666,
      "loss": 0.2314,
      "num_input_tokens_seen": 37034288,
      "step": 175490
    },
    {
      "epoch": 19.306380638063807,
      "grad_norm": 0.01104736328125,
      "learning_rate": 0.00010981597856170865,
      "loss": 0.2319,
      "num_input_tokens_seen": 37035312,
      "step": 175495
    },
    {
      "epoch": 19.306930693069308,
      "grad_norm": 0.005645751953125,
      "learning_rate": 0.00010964211475520014,
      "loss": 0.2314,
      "num_input_tokens_seen": 37036400,
      "step": 175500
    },
    {
      "epoch": 19.30748074807481,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00010946838818597648,
      "loss": 0.2309,
      "num_input_tokens_seen": 37037424,
      "step": 175505
    },
    {
      "epoch": 19.308030803080307,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00010929479885563641,
      "loss": 0.2319,
      "num_input_tokens_seen": 37038512,
      "step": 175510
    },
    {
      "epoch": 19.308580858085808,
      "grad_norm": 0.001922607421875,
      "learning_rate": 0.000109121346765782,
      "loss": 0.2319,
      "num_input_tokens_seen": 37039600,
      "step": 175515
    },
    {
      "epoch": 19.30913091309131,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.00010894803191801028,
      "loss": 0.2329,
      "num_input_tokens_seen": 37040656,
      "step": 175520
    },
    {
      "epoch": 19.30968096809681,
      "grad_norm": 0.001800537109375,
      "learning_rate": 0.00010877485431391831,
      "loss": 0.2293,
      "num_input_tokens_seen": 37041744,
      "step": 175525
    },
    {
      "epoch": 19.31023102310231,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00010860181395510481,
      "loss": 0.2303,
      "num_input_tokens_seen": 37042800,
      "step": 175530
    },
    {
      "epoch": 19.310781078107812,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.0001084289108431602,
      "loss": 0.2293,
      "num_input_tokens_seen": 37043888,
      "step": 175535
    },
    {
      "epoch": 19.31133113311331,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00010825614497968317,
      "loss": 0.2335,
      "num_input_tokens_seen": 37044976,
      "step": 175540
    },
    {
      "epoch": 19.31188118811881,
      "grad_norm": 0.010986328125,
      "learning_rate": 0.00010808351636626245,
      "loss": 0.2329,
      "num_input_tokens_seen": 37046032,
      "step": 175545
    },
    {
      "epoch": 19.312431243124312,
      "grad_norm": 0.00140380859375,
      "learning_rate": 0.00010791102500449011,
      "loss": 0.2335,
      "num_input_tokens_seen": 37047088,
      "step": 175550
    },
    {
      "epoch": 19.312981298129813,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00010773867089595656,
      "loss": 0.2303,
      "num_input_tokens_seen": 37048208,
      "step": 175555
    },
    {
      "epoch": 19.313531353135314,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00010756645404224885,
      "loss": 0.2314,
      "num_input_tokens_seen": 37049232,
      "step": 175560
    },
    {
      "epoch": 19.314081408140815,
      "grad_norm": 0.0113525390625,
      "learning_rate": 0.0001073943744449557,
      "loss": 0.2288,
      "num_input_tokens_seen": 37050256,
      "step": 175565
    },
    {
      "epoch": 19.314631463146316,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00010722243210566085,
      "loss": 0.2324,
      "num_input_tokens_seen": 37051312,
      "step": 175570
    },
    {
      "epoch": 19.315181518151814,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00010705062702595136,
      "loss": 0.2288,
      "num_input_tokens_seen": 37052336,
      "step": 175575
    },
    {
      "epoch": 19.315731573157315,
      "grad_norm": 0.01092529296875,
      "learning_rate": 0.00010687895920741097,
      "loss": 0.2298,
      "num_input_tokens_seen": 37053392,
      "step": 175580
    },
    {
      "epoch": 19.316281628162816,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 0.00010670742865162008,
      "loss": 0.2309,
      "num_input_tokens_seen": 37054416,
      "step": 175585
    },
    {
      "epoch": 19.316831683168317,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00010653603536016076,
      "loss": 0.2324,
      "num_input_tokens_seen": 37055408,
      "step": 175590
    },
    {
      "epoch": 19.317381738173818,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00010636477933461174,
      "loss": 0.2324,
      "num_input_tokens_seen": 37056400,
      "step": 175595
    },
    {
      "epoch": 19.31793179317932,
      "grad_norm": 0.0026397705078125,
      "learning_rate": 0.00010619366057655177,
      "loss": 0.2319,
      "num_input_tokens_seen": 37057424,
      "step": 175600
    },
    {
      "epoch": 19.318481848184817,
      "grad_norm": 0.01123046875,
      "learning_rate": 0.00010602267908755958,
      "loss": 0.2324,
      "num_input_tokens_seen": 37058544,
      "step": 175605
    },
    {
      "epoch": 19.319031903190318,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 0.00010585183486920724,
      "loss": 0.2319,
      "num_input_tokens_seen": 37059632,
      "step": 175610
    },
    {
      "epoch": 19.31958195819582,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.0001056811279230735,
      "loss": 0.2319,
      "num_input_tokens_seen": 37060752,
      "step": 175615
    },
    {
      "epoch": 19.32013201320132,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.00010551055825072875,
      "loss": 0.2324,
      "num_input_tokens_seen": 37061808,
      "step": 175620
    },
    {
      "epoch": 19.32068206820682,
      "grad_norm": 0.00250244140625,
      "learning_rate": 0.00010534012585374507,
      "loss": 0.2308,
      "num_input_tokens_seen": 37062800,
      "step": 175625
    },
    {
      "epoch": 19.321232123212322,
      "grad_norm": 0.005584716796875,
      "learning_rate": 0.00010516983073369622,
      "loss": 0.2303,
      "num_input_tokens_seen": 37063888,
      "step": 175630
    },
    {
      "epoch": 19.321782178217823,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.00010499967289214928,
      "loss": 0.2308,
      "num_input_tokens_seen": 37064880,
      "step": 175635
    },
    {
      "epoch": 19.32233223322332,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00010482965233067298,
      "loss": 0.2308,
      "num_input_tokens_seen": 37065968,
      "step": 175640
    },
    {
      "epoch": 19.322882288228822,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.0001046597690508344,
      "loss": 0.2314,
      "num_input_tokens_seen": 37067056,
      "step": 175645
    },
    {
      "epoch": 19.323432343234323,
      "grad_norm": 0.00628662109375,
      "learning_rate": 0.00010449002305419897,
      "loss": 0.2298,
      "num_input_tokens_seen": 37068144,
      "step": 175650
    },
    {
      "epoch": 19.323982398239824,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00010432041434233208,
      "loss": 0.2303,
      "num_input_tokens_seen": 37069168,
      "step": 175655
    },
    {
      "epoch": 19.324532453245325,
      "grad_norm": 0.0007781982421875,
      "learning_rate": 0.00010415094291679583,
      "loss": 0.2324,
      "num_input_tokens_seen": 37070192,
      "step": 175660
    },
    {
      "epoch": 19.325082508250826,
      "grad_norm": 0.002044677734375,
      "learning_rate": 0.0001039816087791523,
      "loss": 0.2303,
      "num_input_tokens_seen": 37071248,
      "step": 175665
    },
    {
      "epoch": 19.325632563256324,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00010381241193096357,
      "loss": 0.2304,
      "num_input_tokens_seen": 37072368,
      "step": 175670
    },
    {
      "epoch": 19.326182618261825,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00010364335237378674,
      "loss": 0.2319,
      "num_input_tokens_seen": 37073392,
      "step": 175675
    },
    {
      "epoch": 19.326732673267326,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00010347443010918222,
      "loss": 0.2314,
      "num_input_tokens_seen": 37074416,
      "step": 175680
    },
    {
      "epoch": 19.327282728272827,
      "grad_norm": 0.0021514892578125,
      "learning_rate": 0.00010330564513870544,
      "loss": 0.2314,
      "num_input_tokens_seen": 37075472,
      "step": 175685
    },
    {
      "epoch": 19.32783278327833,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.0001031369974639118,
      "loss": 0.2293,
      "num_input_tokens_seen": 37076560,
      "step": 175690
    },
    {
      "epoch": 19.32838283828383,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 0.00010296848708635674,
      "loss": 0.2309,
      "num_input_tokens_seen": 37077648,
      "step": 175695
    },
    {
      "epoch": 19.32893289328933,
      "grad_norm": 0.005462646484375,
      "learning_rate": 0.00010280011400759237,
      "loss": 0.2298,
      "num_input_tokens_seen": 37078672,
      "step": 175700
    },
    {
      "epoch": 19.329482948294828,
      "grad_norm": 0.00543212890625,
      "learning_rate": 0.00010263187822917241,
      "loss": 0.2298,
      "num_input_tokens_seen": 37079760,
      "step": 175705
    },
    {
      "epoch": 19.33003300330033,
      "grad_norm": 0.00592041015625,
      "learning_rate": 0.00010246377975264398,
      "loss": 0.2335,
      "num_input_tokens_seen": 37080848,
      "step": 175710
    },
    {
      "epoch": 19.33058305830583,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.00010229581857955915,
      "loss": 0.2308,
      "num_input_tokens_seen": 37081936,
      "step": 175715
    },
    {
      "epoch": 19.33113311331133,
      "grad_norm": 0.001739501953125,
      "learning_rate": 0.00010212799471146505,
      "loss": 0.2308,
      "num_input_tokens_seen": 37082992,
      "step": 175720
    },
    {
      "epoch": 19.331683168316832,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00010196030814990875,
      "loss": 0.2309,
      "num_input_tokens_seen": 37084080,
      "step": 175725
    },
    {
      "epoch": 19.332233223322334,
      "grad_norm": 0.0057373046875,
      "learning_rate": 0.00010179275889643569,
      "loss": 0.2309,
      "num_input_tokens_seen": 37085104,
      "step": 175730
    },
    {
      "epoch": 19.33278327832783,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.00010162534695259128,
      "loss": 0.2293,
      "num_input_tokens_seen": 37086160,
      "step": 175735
    },
    {
      "epoch": 19.333333333333332,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 0.00010145807231991433,
      "loss": 0.2324,
      "num_input_tokens_seen": 37087248,
      "step": 175740
    },
    {
      "epoch": 19.333883388338833,
      "grad_norm": 0.005615234375,
      "learning_rate": 0.00010129093499995023,
      "loss": 0.2329,
      "num_input_tokens_seen": 37088304,
      "step": 175745
    },
    {
      "epoch": 19.334433443344334,
      "grad_norm": 0.0111083984375,
      "learning_rate": 0.00010112393499423944,
      "loss": 0.2319,
      "num_input_tokens_seen": 37089392,
      "step": 175750
    },
    {
      "epoch": 19.334983498349835,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 0.00010095707230431904,
      "loss": 0.2324,
      "num_input_tokens_seen": 37090384,
      "step": 175755
    },
    {
      "epoch": 19.335533553355337,
      "grad_norm": 0.0054931640625,
      "learning_rate": 0.0001007903469317295,
      "loss": 0.2319,
      "num_input_tokens_seen": 37091472,
      "step": 175760
    },
    {
      "epoch": 19.336083608360838,
      "grad_norm": 0.00537109375,
      "learning_rate": 0.00010062375887800456,
      "loss": 0.2293,
      "num_input_tokens_seen": 37092496,
      "step": 175765
    },
    {
      "epoch": 19.336633663366335,
      "grad_norm": 0.00146484375,
      "learning_rate": 0.00010045730814468134,
      "loss": 0.2303,
      "num_input_tokens_seen": 37093520,
      "step": 175770
    },
    {
      "epoch": 19.337183718371836,
      "grad_norm": 0.00555419921875,
      "learning_rate": 0.00010029099473329361,
      "loss": 0.2314,
      "num_input_tokens_seen": 37094576,
      "step": 175775
    },
    {
      "epoch": 19.337733773377337,
      "grad_norm": 0.005767822265625,
      "learning_rate": 0.00010012481864537514,
      "loss": 0.2319,
      "num_input_tokens_seen": 37095632,
      "step": 175780
    },
    {
      "epoch": 19.33828382838284,
      "grad_norm": 0.002288818359375,
      "learning_rate": 9.995877988245638e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37096656,
      "step": 175785
    },
    {
      "epoch": 19.33883388338834,
      "grad_norm": 0.0107421875,
      "learning_rate": 9.97928784460661e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37097744,
      "step": 175790
    },
    {
      "epoch": 19.33938393839384,
      "grad_norm": 0.00171661376953125,
      "learning_rate": 9.962711433773641e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37098832,
      "step": 175795
    },
    {
      "epoch": 19.33993399339934,
      "grad_norm": 0.00555419921875,
      "learning_rate": 9.946148755899275e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37099888,
      "step": 175800
    },
    {
      "epoch": 19.34048404840484,
      "grad_norm": 0.00244140625,
      "learning_rate": 9.929599811136391e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37100976,
      "step": 175805
    },
    {
      "epoch": 19.34103410341034,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 9.913064599637366e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37102032,
      "step": 175810
    },
    {
      "epoch": 19.34158415841584,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 9.896543121554579e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37103120,
      "step": 175815
    },
    {
      "epoch": 19.342134213421343,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 9.88003537704024e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37104208,
      "step": 175820
    },
    {
      "epoch": 19.342684268426844,
      "grad_norm": 0.00592041015625,
      "learning_rate": 9.863541366246731e-05,
      "loss": 0.234,
      "num_input_tokens_seen": 37105200,
      "step": 175825
    },
    {
      "epoch": 19.343234323432345,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 9.84706108932576e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37106256,
      "step": 175830
    },
    {
      "epoch": 19.343784378437842,
      "grad_norm": 0.01092529296875,
      "learning_rate": 9.830594546429538e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37107248,
      "step": 175835
    },
    {
      "epoch": 19.344334433443343,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 9.814141737709614e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37108304,
      "step": 175840
    },
    {
      "epoch": 19.344884488448844,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 9.797702663317697e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37109360,
      "step": 175845
    },
    {
      "epoch": 19.345434543454346,
      "grad_norm": 0.001220703125,
      "learning_rate": 9.781277323405334e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37110416,
      "step": 175850
    },
    {
      "epoch": 19.345984598459847,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 9.764865718123738e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37111376,
      "step": 175855
    },
    {
      "epoch": 19.346534653465348,
      "grad_norm": 0.0057373046875,
      "learning_rate": 9.748467847624454e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37112400,
      "step": 175860
    },
    {
      "epoch": 19.34708470847085,
      "grad_norm": 0.005706787109375,
      "learning_rate": 9.732083712058193e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37113488,
      "step": 175865
    },
    {
      "epoch": 19.347634763476346,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 9.715713311576501e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37114576,
      "step": 175870
    },
    {
      "epoch": 19.348184818481847,
      "grad_norm": 0.005889892578125,
      "learning_rate": 9.699356646329927e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37115600,
      "step": 175875
    },
    {
      "epoch": 19.34873487348735,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 9.683013716469179e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37116560,
      "step": 175880
    },
    {
      "epoch": 19.34928492849285,
      "grad_norm": 0.0022430419921875,
      "learning_rate": 9.666684522144807e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37117584,
      "step": 175885
    },
    {
      "epoch": 19.34983498349835,
      "grad_norm": 0.006103515625,
      "learning_rate": 9.650369063507691e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37118608,
      "step": 175890
    },
    {
      "epoch": 19.350385038503852,
      "grad_norm": 0.0108642578125,
      "learning_rate": 9.634067340708041e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37119664,
      "step": 175895
    },
    {
      "epoch": 19.35093509350935,
      "grad_norm": 0.00567626953125,
      "learning_rate": 9.617779353895905e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37120720,
      "step": 175900
    },
    {
      "epoch": 19.35148514851485,
      "grad_norm": 0.005615234375,
      "learning_rate": 9.601505103221496e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37121776,
      "step": 175905
    },
    {
      "epoch": 19.35203520352035,
      "grad_norm": 0.005828857421875,
      "learning_rate": 9.58524458883503e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37122832,
      "step": 175910
    },
    {
      "epoch": 19.352585258525853,
      "grad_norm": 0.005645751953125,
      "learning_rate": 9.568997810886048e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37123856,
      "step": 175915
    },
    {
      "epoch": 19.353135313531354,
      "grad_norm": 0.0013427734375,
      "learning_rate": 9.552764769524436e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37124912,
      "step": 175920
    },
    {
      "epoch": 19.353685368536855,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 9.536545464899904e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37126000,
      "step": 175925
    },
    {
      "epoch": 19.354235423542356,
      "grad_norm": 0.002288818359375,
      "learning_rate": 9.520339897161833e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37127088,
      "step": 175930
    },
    {
      "epoch": 19.354785478547853,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 9.504148066459606e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37128208,
      "step": 175935
    },
    {
      "epoch": 19.355335533553355,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 9.487969972942268e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37129264,
      "step": 175940
    },
    {
      "epoch": 19.355885588558856,
      "grad_norm": 0.005859375,
      "learning_rate": 9.471805616759199e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37130352,
      "step": 175945
    },
    {
      "epoch": 19.356435643564357,
      "grad_norm": 0.005950927734375,
      "learning_rate": 9.455654998059281e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37131376,
      "step": 175950
    },
    {
      "epoch": 19.356985698569858,
      "grad_norm": 0.000659942626953125,
      "learning_rate": 9.43951811699123e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37132432,
      "step": 175955
    },
    {
      "epoch": 19.35753575357536,
      "grad_norm": 0.0111083984375,
      "learning_rate": 9.423394973704091e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37133520,
      "step": 175960
    },
    {
      "epoch": 19.358085808580856,
      "grad_norm": 0.005767822265625,
      "learning_rate": 9.407285568346246e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37134608,
      "step": 175965
    },
    {
      "epoch": 19.358635863586358,
      "grad_norm": 0.010986328125,
      "learning_rate": 9.391189901066244e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37135760,
      "step": 175970
    },
    {
      "epoch": 19.35918591859186,
      "grad_norm": 0.005645751953125,
      "learning_rate": 9.375107972012131e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37136816,
      "step": 175975
    },
    {
      "epoch": 19.35973597359736,
      "grad_norm": 0.005706787109375,
      "learning_rate": 9.359039781332623e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37137840,
      "step": 175980
    },
    {
      "epoch": 19.36028602860286,
      "grad_norm": 0.005767822265625,
      "learning_rate": 9.342985329175434e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37138896,
      "step": 175985
    },
    {
      "epoch": 19.360836083608362,
      "grad_norm": 0.01092529296875,
      "learning_rate": 9.326944615688614e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37139952,
      "step": 175990
    },
    {
      "epoch": 19.361386138613863,
      "grad_norm": 0.005889892578125,
      "learning_rate": 9.310917641020211e-05,
      "loss": 0.2304,
      "num_input_tokens_seen": 37141104,
      "step": 175995
    },
    {
      "epoch": 19.36193619361936,
      "grad_norm": 0.00115966796875,
      "learning_rate": 9.294904405317771e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37142160,
      "step": 176000
    },
    {
      "epoch": 19.36248624862486,
      "grad_norm": 0.00567626953125,
      "learning_rate": 9.27890490872868e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37143216,
      "step": 176005
    },
    {
      "epoch": 19.363036303630363,
      "grad_norm": 0.00555419921875,
      "learning_rate": 9.262919151400817e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37144240,
      "step": 176010
    },
    {
      "epoch": 19.363586358635864,
      "grad_norm": 0.00262451171875,
      "learning_rate": 9.246947133481231e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37145264,
      "step": 176015
    },
    {
      "epoch": 19.364136413641365,
      "grad_norm": 0.005462646484375,
      "learning_rate": 9.230988855117306e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37146256,
      "step": 176020
    },
    {
      "epoch": 19.364686468646866,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 9.215044316455923e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37147376,
      "step": 176025
    },
    {
      "epoch": 19.365236523652364,
      "grad_norm": 0.002197265625,
      "learning_rate": 9.199113517643964e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37148496,
      "step": 176030
    },
    {
      "epoch": 19.365786578657865,
      "grad_norm": 0.001312255859375,
      "learning_rate": 9.183196458828646e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37149520,
      "step": 176035
    },
    {
      "epoch": 19.366336633663366,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 9.167293140156351e-05,
      "loss": 0.2345,
      "num_input_tokens_seen": 37150608,
      "step": 176040
    },
    {
      "epoch": 19.366886688668867,
      "grad_norm": 0.005645751953125,
      "learning_rate": 9.151403561773796e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37151632,
      "step": 176045
    },
    {
      "epoch": 19.367436743674368,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 9.135527723827363e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37152656,
      "step": 176050
    },
    {
      "epoch": 19.36798679867987,
      "grad_norm": 0.00567626953125,
      "learning_rate": 9.119665626463268e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37153680,
      "step": 176055
    },
    {
      "epoch": 19.36853685368537,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 9.103817269827895e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37154736,
      "step": 176060
    },
    {
      "epoch": 19.369086908690868,
      "grad_norm": 0.005645751953125,
      "learning_rate": 9.087982654067295e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37155760,
      "step": 176065
    },
    {
      "epoch": 19.36963696369637,
      "grad_norm": 0.00567626953125,
      "learning_rate": 9.072161779327348e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37156752,
      "step": 176070
    },
    {
      "epoch": 19.37018701870187,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 9.056354645753938e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37157840,
      "step": 176075
    },
    {
      "epoch": 19.37073707370737,
      "grad_norm": 0.002349853515625,
      "learning_rate": 9.040561253492451e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37158864,
      "step": 176080
    },
    {
      "epoch": 19.371287128712872,
      "grad_norm": 0.01116943359375,
      "learning_rate": 9.024781602688936e-05,
      "loss": 0.2304,
      "num_input_tokens_seen": 37159920,
      "step": 176085
    },
    {
      "epoch": 19.371837183718373,
      "grad_norm": 0.0108642578125,
      "learning_rate": 9.009015693488443e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37160976,
      "step": 176090
    },
    {
      "epoch": 19.37238723872387,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 8.993263526036354e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37162064,
      "step": 176095
    },
    {
      "epoch": 19.372937293729372,
      "grad_norm": 0.001434326171875,
      "learning_rate": 8.977525100477889e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37163152,
      "step": 176100
    },
    {
      "epoch": 19.373487348734873,
      "grad_norm": 0.00555419921875,
      "learning_rate": 8.961800416958266e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37164176,
      "step": 176105
    },
    {
      "epoch": 19.374037403740374,
      "grad_norm": 0.005859375,
      "learning_rate": 8.946089475622198e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37165328,
      "step": 176110
    },
    {
      "epoch": 19.374587458745875,
      "grad_norm": 0.005859375,
      "learning_rate": 8.930392276614406e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37166352,
      "step": 176115
    },
    {
      "epoch": 19.375137513751376,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 8.914708820079775e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37167440,
      "step": 176120
    },
    {
      "epoch": 19.375687568756877,
      "grad_norm": 0.0020751953125,
      "learning_rate": 8.899039106162853e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37168464,
      "step": 176125
    },
    {
      "epoch": 19.376237623762375,
      "grad_norm": 0.00616455078125,
      "learning_rate": 8.883383135008027e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37169488,
      "step": 176130
    },
    {
      "epoch": 19.376787678767876,
      "grad_norm": 0.01080322265625,
      "learning_rate": 8.867740906759514e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37170448,
      "step": 176135
    },
    {
      "epoch": 19.377337733773377,
      "grad_norm": 0.00286865234375,
      "learning_rate": 8.852112421561531e-05,
      "loss": 0.234,
      "num_input_tokens_seen": 37171536,
      "step": 176140
    },
    {
      "epoch": 19.377887788778878,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 8.836497679557964e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37172592,
      "step": 176145
    },
    {
      "epoch": 19.37843784378438,
      "grad_norm": 0.0057373046875,
      "learning_rate": 8.820896680892864e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37173712,
      "step": 176150
    },
    {
      "epoch": 19.37898789878988,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 8.805309425710117e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37174704,
      "step": 176155
    },
    {
      "epoch": 19.379537953795378,
      "grad_norm": 0.005401611328125,
      "learning_rate": 8.789735914153273e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37175760,
      "step": 176160
    },
    {
      "epoch": 19.38008800880088,
      "grad_norm": 0.00567626953125,
      "learning_rate": 8.774176146366053e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37176752,
      "step": 176165
    },
    {
      "epoch": 19.38063806380638,
      "grad_norm": 0.005615234375,
      "learning_rate": 8.758630122491506e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37177872,
      "step": 176170
    },
    {
      "epoch": 19.38118811881188,
      "grad_norm": 0.005706787109375,
      "learning_rate": 8.74309784267302e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37178928,
      "step": 176175
    },
    {
      "epoch": 19.381738173817382,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 8.727579307053979e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37179984,
      "step": 176180
    },
    {
      "epoch": 19.382288228822883,
      "grad_norm": 0.00567626953125,
      "learning_rate": 8.712074515777268e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37181040,
      "step": 176185
    },
    {
      "epoch": 19.382838283828384,
      "grad_norm": 0.00592041015625,
      "learning_rate": 8.696583468985774e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37182000,
      "step": 176190
    },
    {
      "epoch": 19.383388338833882,
      "grad_norm": 0.00567626953125,
      "learning_rate": 8.681106166822216e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37183088,
      "step": 176195
    },
    {
      "epoch": 19.383938393839383,
      "grad_norm": 0.010986328125,
      "learning_rate": 8.665642609429313e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37184144,
      "step": 176200
    },
    {
      "epoch": 19.384488448844884,
      "grad_norm": 0.005615234375,
      "learning_rate": 8.650192796949452e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37185200,
      "step": 176205
    },
    {
      "epoch": 19.385038503850385,
      "grad_norm": 0.011474609375,
      "learning_rate": 8.634756729525184e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37186224,
      "step": 176210
    },
    {
      "epoch": 19.385588558855886,
      "grad_norm": 0.001129150390625,
      "learning_rate": 8.619334407298895e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37187280,
      "step": 176215
    },
    {
      "epoch": 19.386138613861387,
      "grad_norm": 0.01116943359375,
      "learning_rate": 8.603925830412473e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37188336,
      "step": 176220
    },
    {
      "epoch": 19.38668866886689,
      "grad_norm": 0.005584716796875,
      "learning_rate": 8.588530999008137e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37189392,
      "step": 176225
    },
    {
      "epoch": 19.387238723872386,
      "grad_norm": 0.000553131103515625,
      "learning_rate": 8.573149913227606e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37190480,
      "step": 176230
    },
    {
      "epoch": 19.387788778877887,
      "grad_norm": 0.005828857421875,
      "learning_rate": 8.557782573212436e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37191504,
      "step": 176235
    },
    {
      "epoch": 19.388338833883388,
      "grad_norm": 0.00579833984375,
      "learning_rate": 8.542428979104843e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37192528,
      "step": 176240
    },
    {
      "epoch": 19.38888888888889,
      "grad_norm": 0.005645751953125,
      "learning_rate": 8.527089131045883e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37193584,
      "step": 176245
    },
    {
      "epoch": 19.38943894389439,
      "grad_norm": 0.00567626953125,
      "learning_rate": 8.51176302917711e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37194608,
      "step": 176250
    },
    {
      "epoch": 19.38998899889989,
      "grad_norm": 0.01092529296875,
      "learning_rate": 8.496450673639744e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37195664,
      "step": 176255
    },
    {
      "epoch": 19.39053905390539,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 8.481152064574837e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37196720,
      "step": 176260
    },
    {
      "epoch": 19.39108910891089,
      "grad_norm": 0.005401611328125,
      "learning_rate": 8.465867202123445e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37197840,
      "step": 176265
    },
    {
      "epoch": 19.39163916391639,
      "grad_norm": 0.00579833984375,
      "learning_rate": 8.450596086426288e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37198896,
      "step": 176270
    },
    {
      "epoch": 19.392189218921892,
      "grad_norm": 0.0019073486328125,
      "learning_rate": 8.435338717624585e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37199888,
      "step": 176275
    },
    {
      "epoch": 19.392739273927393,
      "grad_norm": 0.00555419921875,
      "learning_rate": 8.42009509585856e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37200976,
      "step": 176280
    },
    {
      "epoch": 19.393289328932894,
      "grad_norm": 0.00170135498046875,
      "learning_rate": 8.404865221268764e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37202064,
      "step": 176285
    },
    {
      "epoch": 19.393839383938396,
      "grad_norm": 0.002685546875,
      "learning_rate": 8.389649093995588e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37203120,
      "step": 176290
    },
    {
      "epoch": 19.394389438943893,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 8.374446714179084e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37204208,
      "step": 176295
    },
    {
      "epoch": 19.394939493949394,
      "grad_norm": 0.006256103515625,
      "learning_rate": 8.359258081959807e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37205264,
      "step": 176300
    },
    {
      "epoch": 19.395489548954895,
      "grad_norm": 0.00567626953125,
      "learning_rate": 8.34408319747748e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37206352,
      "step": 176305
    },
    {
      "epoch": 19.396039603960396,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 8.328922060871823e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37207440,
      "step": 176310
    },
    {
      "epoch": 19.396589658965897,
      "grad_norm": 0.005523681640625,
      "learning_rate": 8.313774672282725e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37208432,
      "step": 176315
    },
    {
      "epoch": 19.3971397139714,
      "grad_norm": 0.005218505859375,
      "learning_rate": 8.298641031849906e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37209520,
      "step": 176320
    },
    {
      "epoch": 19.397689768976896,
      "grad_norm": 0.01123046875,
      "learning_rate": 8.283521139712591e-05,
      "loss": 0.2304,
      "num_input_tokens_seen": 37210576,
      "step": 176325
    },
    {
      "epoch": 19.398239823982397,
      "grad_norm": 0.005950927734375,
      "learning_rate": 8.268414996010497e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37211664,
      "step": 176330
    },
    {
      "epoch": 19.3987898789879,
      "grad_norm": 0.005706787109375,
      "learning_rate": 8.25332260088235e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37212720,
      "step": 176335
    },
    {
      "epoch": 19.3993399339934,
      "grad_norm": 0.01129150390625,
      "learning_rate": 8.238243954467705e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37213808,
      "step": 176340
    },
    {
      "epoch": 19.3998899889989,
      "grad_norm": 0.005828857421875,
      "learning_rate": 8.223179056905283e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37214864,
      "step": 176345
    },
    {
      "epoch": 19.4004400440044,
      "grad_norm": 0.00592041015625,
      "learning_rate": 8.208127908333973e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37215952,
      "step": 176350
    },
    {
      "epoch": 19.400990099009903,
      "grad_norm": 0.0015869140625,
      "learning_rate": 8.193090508892497e-05,
      "loss": 0.2267,
      "num_input_tokens_seen": 37217072,
      "step": 176355
    },
    {
      "epoch": 19.4015401540154,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 8.178066858719412e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37218192,
      "step": 176360
    },
    {
      "epoch": 19.4020902090209,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 8.16305695795344e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37219280,
      "step": 176365
    },
    {
      "epoch": 19.402640264026402,
      "grad_norm": 0.005645751953125,
      "learning_rate": 8.148060806732638e-05,
      "loss": 0.2313,
      "num_input_tokens_seen": 37220368,
      "step": 176370
    },
    {
      "epoch": 19.403190319031903,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 8.133078405195226e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37221424,
      "step": 176375
    },
    {
      "epoch": 19.403740374037405,
      "grad_norm": 0.005828857421875,
      "learning_rate": 8.11810975347943e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37222544,
      "step": 176380
    },
    {
      "epoch": 19.404290429042906,
      "grad_norm": 0.005584716796875,
      "learning_rate": 8.103154851723137e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37223600,
      "step": 176385
    },
    {
      "epoch": 19.404840484048403,
      "grad_norm": 0.01104736328125,
      "learning_rate": 8.088213700064073e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37224656,
      "step": 176390
    },
    {
      "epoch": 19.405390539053904,
      "grad_norm": 0.005950927734375,
      "learning_rate": 8.073286298640125e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37225712,
      "step": 176395
    },
    {
      "epoch": 19.405940594059405,
      "grad_norm": 0.005584716796875,
      "learning_rate": 8.058372647589018e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37226736,
      "step": 176400
    },
    {
      "epoch": 19.406490649064907,
      "grad_norm": 0.005615234375,
      "learning_rate": 8.043472747047809e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37227728,
      "step": 176405
    },
    {
      "epoch": 19.407040704070408,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 8.028586597153886e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37228784,
      "step": 176410
    },
    {
      "epoch": 19.40759075907591,
      "grad_norm": 0.0111083984375,
      "learning_rate": 8.013714198044641e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37229840,
      "step": 176415
    },
    {
      "epoch": 19.40814081408141,
      "grad_norm": 0.00061798095703125,
      "learning_rate": 7.99885554985713e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37230832,
      "step": 176420
    },
    {
      "epoch": 19.408690869086907,
      "grad_norm": 0.0062255859375,
      "learning_rate": 7.984010652728246e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37231888,
      "step": 176425
    },
    {
      "epoch": 19.40924092409241,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 7.96917950679471e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37232976,
      "step": 176430
    },
    {
      "epoch": 19.40979097909791,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 7.954362112193247e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37234032,
      "step": 176435
    },
    {
      "epoch": 19.41034103410341,
      "grad_norm": 0.00616455078125,
      "learning_rate": 7.939558469060581e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37235152,
      "step": 176440
    },
    {
      "epoch": 19.41089108910891,
      "grad_norm": 0.00555419921875,
      "learning_rate": 7.924768577533103e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37236176,
      "step": 176445
    },
    {
      "epoch": 19.411441144114413,
      "grad_norm": 0.005584716796875,
      "learning_rate": 7.90999243774687e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37237200,
      "step": 176450
    },
    {
      "epoch": 19.41199119911991,
      "grad_norm": 0.0111083984375,
      "learning_rate": 7.895230049838442e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37238256,
      "step": 176455
    },
    {
      "epoch": 19.41254125412541,
      "grad_norm": 0.005523681640625,
      "learning_rate": 7.880481413943707e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37239376,
      "step": 176460
    },
    {
      "epoch": 19.413091309130913,
      "grad_norm": 0.005462646484375,
      "learning_rate": 7.865746530198558e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37240464,
      "step": 176465
    },
    {
      "epoch": 19.413641364136414,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 7.85102539873872e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37241552,
      "step": 176470
    },
    {
      "epoch": 19.414191419141915,
      "grad_norm": 0.0024871826171875,
      "learning_rate": 7.836318019700083e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37242640,
      "step": 176475
    },
    {
      "epoch": 19.414741474147416,
      "grad_norm": 0.00567626953125,
      "learning_rate": 7.82162439321804e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37243696,
      "step": 176480
    },
    {
      "epoch": 19.415291529152917,
      "grad_norm": 0.0054931640625,
      "learning_rate": 7.806944519427983e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37244752,
      "step": 176485
    },
    {
      "epoch": 19.415841584158414,
      "grad_norm": 0.006103515625,
      "learning_rate": 7.792278398465468e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37245872,
      "step": 176490
    },
    {
      "epoch": 19.416391639163916,
      "grad_norm": 0.0054931640625,
      "learning_rate": 7.77762603046539e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37246864,
      "step": 176495
    },
    {
      "epoch": 19.416941694169417,
      "grad_norm": 0.0017242431640625,
      "learning_rate": 7.762987415562972e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37247920,
      "step": 176500
    },
    {
      "epoch": 19.417491749174918,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 7.74836255389294e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37248912,
      "step": 176505
    },
    {
      "epoch": 19.41804180418042,
      "grad_norm": 0.005615234375,
      "learning_rate": 7.733751445590353e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37249904,
      "step": 176510
    },
    {
      "epoch": 19.41859185918592,
      "grad_norm": 0.005828857421875,
      "learning_rate": 7.719154090789604e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37251024,
      "step": 176515
    },
    {
      "epoch": 19.419141914191417,
      "grad_norm": 0.005615234375,
      "learning_rate": 7.704570489625418e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37252112,
      "step": 176520
    },
    {
      "epoch": 19.41969196919692,
      "grad_norm": 0.01080322265625,
      "learning_rate": 7.69000064223202e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37253200,
      "step": 176525
    },
    {
      "epoch": 19.42024202420242,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 7.675444548743971e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37254320,
      "step": 176530
    },
    {
      "epoch": 19.42079207920792,
      "grad_norm": 0.005523681640625,
      "learning_rate": 7.660902209294994e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37255376,
      "step": 176535
    },
    {
      "epoch": 19.421342134213422,
      "grad_norm": 0.00555419921875,
      "learning_rate": 7.646373624019654e-05,
      "loss": 0.2288,
      "num_input_tokens_seen": 37256400,
      "step": 176540
    },
    {
      "epoch": 19.421892189218923,
      "grad_norm": 0.0108642578125,
      "learning_rate": 7.631858793051505e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37257488,
      "step": 176545
    },
    {
      "epoch": 19.422442244224424,
      "grad_norm": 0.005950927734375,
      "learning_rate": 7.617357716524442e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37258608,
      "step": 176550
    },
    {
      "epoch": 19.42299229922992,
      "grad_norm": 0.005340576171875,
      "learning_rate": 7.602870394572025e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37259728,
      "step": 176555
    },
    {
      "epoch": 19.423542354235423,
      "grad_norm": 0.005645751953125,
      "learning_rate": 7.588396827327814e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37260720,
      "step": 176560
    },
    {
      "epoch": 19.424092409240924,
      "grad_norm": 0.005401611328125,
      "learning_rate": 7.573937014925203e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37261744,
      "step": 176565
    },
    {
      "epoch": 19.424642464246425,
      "grad_norm": 0.0018310546875,
      "learning_rate": 7.559490957497583e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37262736,
      "step": 176570
    },
    {
      "epoch": 19.425192519251926,
      "grad_norm": 0.00095367431640625,
      "learning_rate": 7.545058655177849e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37263792,
      "step": 176575
    },
    {
      "epoch": 19.425742574257427,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 7.530640108099395e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37264784,
      "step": 176580
    },
    {
      "epoch": 19.426292629262925,
      "grad_norm": 0.00604248046875,
      "learning_rate": 7.516235316394615e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37265872,
      "step": 176585
    },
    {
      "epoch": 19.426842684268426,
      "grad_norm": 0.005706787109375,
      "learning_rate": 7.501844280196567e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37266992,
      "step": 176590
    },
    {
      "epoch": 19.427392739273927,
      "grad_norm": 0.005828857421875,
      "learning_rate": 7.487466999637982e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37267984,
      "step": 176595
    },
    {
      "epoch": 19.427942794279428,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 7.473103474851083e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37269008,
      "step": 176600
    },
    {
      "epoch": 19.42849284928493,
      "grad_norm": 0.001190185546875,
      "learning_rate": 7.458753705968601e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37270032,
      "step": 176605
    },
    {
      "epoch": 19.42904290429043,
      "grad_norm": 0.005828857421875,
      "learning_rate": 7.444417693122429e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37271088,
      "step": 176610
    },
    {
      "epoch": 19.42959295929593,
      "grad_norm": 0.005645751953125,
      "learning_rate": 7.430095436444961e-05,
      "loss": 0.2313,
      "num_input_tokens_seen": 37272176,
      "step": 176615
    },
    {
      "epoch": 19.43014301430143,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 7.415786936068091e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37273328,
      "step": 176620
    },
    {
      "epoch": 19.43069306930693,
      "grad_norm": 0.0020599365234375,
      "learning_rate": 7.401492192123548e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37274384,
      "step": 176625
    },
    {
      "epoch": 19.43124312431243,
      "grad_norm": 0.005767822265625,
      "learning_rate": 7.387211204743393e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37275376,
      "step": 176630
    },
    {
      "epoch": 19.431793179317932,
      "grad_norm": 0.005340576171875,
      "learning_rate": 7.372943974059187e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37276432,
      "step": 176635
    },
    {
      "epoch": 19.432343234323433,
      "grad_norm": 0.00592041015625,
      "learning_rate": 7.358690500202159e-05,
      "loss": 0.2313,
      "num_input_tokens_seen": 37277552,
      "step": 176640
    },
    {
      "epoch": 19.432893289328934,
      "grad_norm": 0.0054931640625,
      "learning_rate": 7.344450783303701e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37278704,
      "step": 176645
    },
    {
      "epoch": 19.433443344334435,
      "grad_norm": 0.00109100341796875,
      "learning_rate": 7.330224823495379e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37279728,
      "step": 176650
    },
    {
      "epoch": 19.433993399339933,
      "grad_norm": 0.00128173828125,
      "learning_rate": 7.316012620908252e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37280720,
      "step": 176655
    },
    {
      "epoch": 19.434543454345434,
      "grad_norm": 0.005340576171875,
      "learning_rate": 7.301814175673048e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37281744,
      "step": 176660
    },
    {
      "epoch": 19.435093509350935,
      "grad_norm": 0.006103515625,
      "learning_rate": 7.287629487920666e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37282832,
      "step": 176665
    },
    {
      "epoch": 19.435643564356436,
      "grad_norm": 0.005340576171875,
      "learning_rate": 7.273458557781997e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37283856,
      "step": 176670
    },
    {
      "epoch": 19.436193619361937,
      "grad_norm": 0.005859375,
      "learning_rate": 7.25930138538744e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37284880,
      "step": 176675
    },
    {
      "epoch": 19.436743674367438,
      "grad_norm": 0.01080322265625,
      "learning_rate": 7.245157970867888e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37285904,
      "step": 176680
    },
    {
      "epoch": 19.437293729372936,
      "grad_norm": 0.006011962890625,
      "learning_rate": 7.231028314353239e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37287024,
      "step": 176685
    },
    {
      "epoch": 19.437843784378437,
      "grad_norm": 0.00555419921875,
      "learning_rate": 7.216912415974052e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37288080,
      "step": 176690
    },
    {
      "epoch": 19.438393839383938,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 7.202810275860227e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37289168,
      "step": 176695
    },
    {
      "epoch": 19.43894389438944,
      "grad_norm": 0.00180816650390625,
      "learning_rate": 7.188721894141659e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37290192,
      "step": 176700
    },
    {
      "epoch": 19.43949394939494,
      "grad_norm": 0.001953125,
      "learning_rate": 7.174647270948409e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37291184,
      "step": 176705
    },
    {
      "epoch": 19.44004400440044,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 7.160586406410041e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37292240,
      "step": 176710
    },
    {
      "epoch": 19.440594059405942,
      "grad_norm": 0.010986328125,
      "learning_rate": 7.146539300656285e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37293328,
      "step": 176715
    },
    {
      "epoch": 19.44114411441144,
      "grad_norm": 0.005767822265625,
      "learning_rate": 7.132505953816537e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37294352,
      "step": 176720
    },
    {
      "epoch": 19.44169416941694,
      "grad_norm": 0.00567626953125,
      "learning_rate": 7.118486366020194e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37295440,
      "step": 176725
    },
    {
      "epoch": 19.442244224422442,
      "grad_norm": 0.005523681640625,
      "learning_rate": 7.104480537396318e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37296560,
      "step": 176730
    },
    {
      "epoch": 19.442794279427943,
      "grad_norm": 0.006103515625,
      "learning_rate": 7.090488468074141e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37297680,
      "step": 176735
    },
    {
      "epoch": 19.443344334433444,
      "grad_norm": 0.005645751953125,
      "learning_rate": 7.076510158182392e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37298704,
      "step": 176740
    },
    {
      "epoch": 19.443894389438945,
      "grad_norm": 0.005401611328125,
      "learning_rate": 7.062545607850302e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37299728,
      "step": 176745
    },
    {
      "epoch": 19.444444444444443,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 7.048594817206266e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37300752,
      "step": 176750
    },
    {
      "epoch": 19.444994499449944,
      "grad_norm": 0.005584716796875,
      "learning_rate": 7.03465778637885e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37301840,
      "step": 176755
    },
    {
      "epoch": 19.445544554455445,
      "grad_norm": 0.006011962890625,
      "learning_rate": 7.020734515496619e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37302928,
      "step": 176760
    },
    {
      "epoch": 19.446094609460946,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 7.006825004687967e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37303952,
      "step": 176765
    },
    {
      "epoch": 19.446644664466447,
      "grad_norm": 0.005615234375,
      "learning_rate": 6.99292925408096e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37305040,
      "step": 176770
    },
    {
      "epoch": 19.44719471947195,
      "grad_norm": 0.005645751953125,
      "learning_rate": 6.979047263803828e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37306064,
      "step": 176775
    },
    {
      "epoch": 19.44774477447745,
      "grad_norm": 0.001922607421875,
      "learning_rate": 6.965179033984304e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37307120,
      "step": 176780
    },
    {
      "epoch": 19.448294829482947,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 6.951324564750282e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37308112,
      "step": 176785
    },
    {
      "epoch": 19.448844884488448,
      "grad_norm": 0.001495361328125,
      "learning_rate": 6.937483856229331e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37309168,
      "step": 176790
    },
    {
      "epoch": 19.44939493949395,
      "grad_norm": 0.001220703125,
      "learning_rate": 6.923656908549346e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37310224,
      "step": 176795
    },
    {
      "epoch": 19.44994499449945,
      "grad_norm": 0.005462646484375,
      "learning_rate": 6.909843721837561e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37311280,
      "step": 176800
    },
    {
      "epoch": 19.45049504950495,
      "grad_norm": 0.005584716796875,
      "learning_rate": 6.896044296221204e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37312336,
      "step": 176805
    },
    {
      "epoch": 19.451045104510452,
      "grad_norm": 0.005615234375,
      "learning_rate": 6.882258631827675e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37313328,
      "step": 176810
    },
    {
      "epoch": 19.45159515951595,
      "grad_norm": 0.005523681640625,
      "learning_rate": 6.868486728783707e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37314352,
      "step": 176815
    },
    {
      "epoch": 19.45214521452145,
      "grad_norm": 0.001129150390625,
      "learning_rate": 6.85472858721653e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37315344,
      "step": 176820
    },
    {
      "epoch": 19.452695269526952,
      "grad_norm": 0.005523681640625,
      "learning_rate": 6.840984207252876e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37316336,
      "step": 176825
    },
    {
      "epoch": 19.453245324532453,
      "grad_norm": 0.005584716796875,
      "learning_rate": 6.82725358901931e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37317392,
      "step": 176830
    },
    {
      "epoch": 19.453795379537954,
      "grad_norm": 0.00579833984375,
      "learning_rate": 6.813536732642566e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37318384,
      "step": 176835
    },
    {
      "epoch": 19.454345434543455,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 6.799833638248709e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37319376,
      "step": 176840
    },
    {
      "epoch": 19.454895489548957,
      "grad_norm": 0.01092529296875,
      "learning_rate": 6.78614430596447e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37320400,
      "step": 176845
    },
    {
      "epoch": 19.455445544554454,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 6.772468735915748e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37321456,
      "step": 176850
    },
    {
      "epoch": 19.455995599559955,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 6.758806928228777e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37322544,
      "step": 176855
    },
    {
      "epoch": 19.456545654565456,
      "grad_norm": 0.005615234375,
      "learning_rate": 6.745158883029289e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37323664,
      "step": 176860
    },
    {
      "epoch": 19.457095709570957,
      "grad_norm": 0.000949859619140625,
      "learning_rate": 6.731524600443183e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37324720,
      "step": 176865
    },
    {
      "epoch": 19.45764576457646,
      "grad_norm": 0.0107421875,
      "learning_rate": 6.717904080596027e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37325744,
      "step": 176870
    },
    {
      "epoch": 19.45819581958196,
      "grad_norm": 0.005767822265625,
      "learning_rate": 6.704297323613384e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37326832,
      "step": 176875
    },
    {
      "epoch": 19.458745874587457,
      "grad_norm": 0.00074005126953125,
      "learning_rate": 6.690704329620656e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37327856,
      "step": 176880
    },
    {
      "epoch": 19.459295929592958,
      "grad_norm": 0.00567626953125,
      "learning_rate": 6.677125098743242e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37328912,
      "step": 176885
    },
    {
      "epoch": 19.45984598459846,
      "grad_norm": 0.00555419921875,
      "learning_rate": 6.663559631106041e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37329968,
      "step": 176890
    },
    {
      "epoch": 19.46039603960396,
      "grad_norm": 0.001739501953125,
      "learning_rate": 6.650007926834455e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37331024,
      "step": 176895
    },
    {
      "epoch": 19.46094609460946,
      "grad_norm": 0.00604248046875,
      "learning_rate": 6.636469986053051e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37332080,
      "step": 176900
    },
    {
      "epoch": 19.461496149614963,
      "grad_norm": 0.0018310546875,
      "learning_rate": 6.622945808886726e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37333168,
      "step": 176905
    },
    {
      "epoch": 19.462046204620464,
      "grad_norm": 0.005645751953125,
      "learning_rate": 6.60943539546005e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37334256,
      "step": 176910
    },
    {
      "epoch": 19.46259625962596,
      "grad_norm": 0.005645751953125,
      "learning_rate": 6.59593874589759e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37335312,
      "step": 176915
    },
    {
      "epoch": 19.463146314631462,
      "grad_norm": 0.005615234375,
      "learning_rate": 6.58245586032391e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37336400,
      "step": 176920
    },
    {
      "epoch": 19.463696369636963,
      "grad_norm": 0.005401611328125,
      "learning_rate": 6.568986738863081e-05,
      "loss": 0.2304,
      "num_input_tokens_seen": 37337424,
      "step": 176925
    },
    {
      "epoch": 19.464246424642464,
      "grad_norm": 0.005401611328125,
      "learning_rate": 6.555531381639334e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37338480,
      "step": 176930
    },
    {
      "epoch": 19.464796479647966,
      "grad_norm": 0.005462646484375,
      "learning_rate": 6.542089788776407e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37339568,
      "step": 176935
    },
    {
      "epoch": 19.465346534653467,
      "grad_norm": 0.0054931640625,
      "learning_rate": 6.528661960398529e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37340560,
      "step": 176940
    },
    {
      "epoch": 19.465896589658964,
      "grad_norm": 0.00555419921875,
      "learning_rate": 6.515247896629439e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37341680,
      "step": 176945
    },
    {
      "epoch": 19.466446644664465,
      "grad_norm": 0.006103515625,
      "learning_rate": 6.501847597592703e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37342704,
      "step": 176950
    },
    {
      "epoch": 19.466996699669966,
      "grad_norm": 0.01104736328125,
      "learning_rate": 6.488461063411555e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37343792,
      "step": 176955
    },
    {
      "epoch": 19.467546754675467,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 6.47508829420973e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37344784,
      "step": 176960
    },
    {
      "epoch": 19.46809680968097,
      "grad_norm": 0.01104736328125,
      "learning_rate": 6.461729290110296e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37345840,
      "step": 176965
    },
    {
      "epoch": 19.46864686468647,
      "grad_norm": 0.005615234375,
      "learning_rate": 6.448384051236488e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37346832,
      "step": 176970
    },
    {
      "epoch": 19.46919691969197,
      "grad_norm": 0.005706787109375,
      "learning_rate": 6.435052577711375e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37347920,
      "step": 176975
    },
    {
      "epoch": 19.46974697469747,
      "grad_norm": 0.005523681640625,
      "learning_rate": 6.421734869657525e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37348912,
      "step": 176980
    },
    {
      "epoch": 19.47029702970297,
      "grad_norm": 0.005523681640625,
      "learning_rate": 6.40843092719784e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37350032,
      "step": 176985
    },
    {
      "epoch": 19.47084708470847,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 6.395140750455053e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37351088,
      "step": 176990
    },
    {
      "epoch": 19.47139713971397,
      "grad_norm": 0.005462646484375,
      "learning_rate": 6.381864339551402e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37352176,
      "step": 176995
    },
    {
      "epoch": 19.471947194719473,
      "grad_norm": 0.00634765625,
      "learning_rate": 6.36860169460962e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37353232,
      "step": 177000
    },
    {
      "epoch": 19.472497249724974,
      "grad_norm": 0.00579833984375,
      "learning_rate": 6.355352815751614e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37354224,
      "step": 177005
    },
    {
      "epoch": 19.47304730473047,
      "grad_norm": 0.005523681640625,
      "learning_rate": 6.342117703099615e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37355280,
      "step": 177010
    },
    {
      "epoch": 19.473597359735972,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 6.328896356775526e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37356336,
      "step": 177015
    },
    {
      "epoch": 19.474147414741473,
      "grad_norm": 0.005615234375,
      "learning_rate": 6.315688776901251e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37357456,
      "step": 177020
    },
    {
      "epoch": 19.474697469746975,
      "grad_norm": 0.01116943359375,
      "learning_rate": 6.302494963598692e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37358512,
      "step": 177025
    },
    {
      "epoch": 19.475247524752476,
      "grad_norm": 0.005401611328125,
      "learning_rate": 6.289314916989086e-05,
      "loss": 0.233,
      "num_input_tokens_seen": 37359568,
      "step": 177030
    },
    {
      "epoch": 19.475797579757977,
      "grad_norm": 0.005828857421875,
      "learning_rate": 6.276148637194333e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37360624,
      "step": 177035
    },
    {
      "epoch": 19.476347634763478,
      "grad_norm": 0.0009765625,
      "learning_rate": 6.262996124335507e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37361680,
      "step": 177040
    },
    {
      "epoch": 19.476897689768975,
      "grad_norm": 0.005584716796875,
      "learning_rate": 6.249857378533675e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37362736,
      "step": 177045
    },
    {
      "epoch": 19.477447744774476,
      "grad_norm": 0.005340576171875,
      "learning_rate": 6.236732399910404e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37363728,
      "step": 177050
    },
    {
      "epoch": 19.477997799779978,
      "grad_norm": 0.01080322265625,
      "learning_rate": 6.223621188586103e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37364816,
      "step": 177055
    },
    {
      "epoch": 19.47854785478548,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 6.210523744682173e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37365872,
      "step": 177060
    },
    {
      "epoch": 19.47909790979098,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 6.197440068319016e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37366864,
      "step": 177065
    },
    {
      "epoch": 19.47964796479648,
      "grad_norm": 0.0025177001953125,
      "learning_rate": 6.184370159617203e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37367952,
      "step": 177070
    },
    {
      "epoch": 19.480198019801982,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 6.171314018697304e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37369040,
      "step": 177075
    },
    {
      "epoch": 19.48074807480748,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 6.158271645679391e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37370064,
      "step": 177080
    },
    {
      "epoch": 19.48129812981298,
      "grad_norm": 0.0057373046875,
      "learning_rate": 6.145243040684034e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37371184,
      "step": 177085
    },
    {
      "epoch": 19.48184818481848,
      "grad_norm": 0.005767822265625,
      "learning_rate": 6.132228203831136e-05,
      "loss": 0.2304,
      "num_input_tokens_seen": 37372240,
      "step": 177090
    },
    {
      "epoch": 19.482398239823983,
      "grad_norm": 0.005645751953125,
      "learning_rate": 6.1192271352406e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37373328,
      "step": 177095
    },
    {
      "epoch": 19.482948294829484,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 6.106239835032501e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37374352,
      "step": 177100
    },
    {
      "epoch": 19.483498349834985,
      "grad_norm": 0.002685546875,
      "learning_rate": 6.0932663033260726e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37375440,
      "step": 177105
    },
    {
      "epoch": 19.484048404840483,
      "grad_norm": 0.01129150390625,
      "learning_rate": 6.0803065402413866e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37376464,
      "step": 177110
    },
    {
      "epoch": 19.484598459845984,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 6.067360545897515e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37377552,
      "step": 177115
    },
    {
      "epoch": 19.485148514851485,
      "grad_norm": 0.005706787109375,
      "learning_rate": 6.054428320414029e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37378608,
      "step": 177120
    },
    {
      "epoch": 19.485698569856986,
      "grad_norm": 0.00592041015625,
      "learning_rate": 6.041509863909999e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37379632,
      "step": 177125
    },
    {
      "epoch": 19.486248624862487,
      "grad_norm": 0.005615234375,
      "learning_rate": 6.0286051765044974e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37380624,
      "step": 177130
    },
    {
      "epoch": 19.486798679867988,
      "grad_norm": 0.0054931640625,
      "learning_rate": 6.015714258316429e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37381712,
      "step": 177135
    },
    {
      "epoch": 19.48734873487349,
      "grad_norm": 0.00537109375,
      "learning_rate": 6.002837109464698e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37382768,
      "step": 177140
    },
    {
      "epoch": 19.487898789878987,
      "grad_norm": 0.01104736328125,
      "learning_rate": 5.9899737300678766e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37383856,
      "step": 177145
    },
    {
      "epoch": 19.488448844884488,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 5.977124120244537e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37384880,
      "step": 177150
    },
    {
      "epoch": 19.48899889988999,
      "grad_norm": 0.005462646484375,
      "learning_rate": 5.96428828011325e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37385968,
      "step": 177155
    },
    {
      "epoch": 19.48954895489549,
      "grad_norm": 0.00103759765625,
      "learning_rate": 5.951466209792089e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37386992,
      "step": 177160
    },
    {
      "epoch": 19.49009900990099,
      "grad_norm": 0.01080322265625,
      "learning_rate": 5.938657909399458e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37388112,
      "step": 177165
    },
    {
      "epoch": 19.490649064906492,
      "grad_norm": 0.005706787109375,
      "learning_rate": 5.9258633790530977e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37389200,
      "step": 177170
    },
    {
      "epoch": 19.49119911991199,
      "grad_norm": 0.005615234375,
      "learning_rate": 5.913082618871412e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37390192,
      "step": 177175
    },
    {
      "epoch": 19.49174917491749,
      "grad_norm": 0.005523681640625,
      "learning_rate": 5.9003156289716416e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37391152,
      "step": 177180
    },
    {
      "epoch": 19.492299229922992,
      "grad_norm": 0.000553131103515625,
      "learning_rate": 5.887562409472025e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37392208,
      "step": 177185
    },
    {
      "epoch": 19.492849284928493,
      "grad_norm": 0.005615234375,
      "learning_rate": 5.8748229604896335e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37393264,
      "step": 177190
    },
    {
      "epoch": 19.493399339933994,
      "grad_norm": 0.01104736328125,
      "learning_rate": 5.8620972821418756e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37394288,
      "step": 177195
    },
    {
      "epoch": 19.493949394939495,
      "grad_norm": 0.0019683837890625,
      "learning_rate": 5.849385374546489e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37395376,
      "step": 177200
    },
    {
      "epoch": 19.494499449944996,
      "grad_norm": 0.005859375,
      "learning_rate": 5.836687237820215e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37396496,
      "step": 177205
    },
    {
      "epoch": 19.495049504950494,
      "grad_norm": 0.005706787109375,
      "learning_rate": 5.8240028720802915e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37397584,
      "step": 177210
    },
    {
      "epoch": 19.495599559955995,
      "grad_norm": 0.006195068359375,
      "learning_rate": 5.811332277443626e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37398608,
      "step": 177215
    },
    {
      "epoch": 19.496149614961496,
      "grad_norm": 0.0106201171875,
      "learning_rate": 5.798675454026791e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37399632,
      "step": 177220
    },
    {
      "epoch": 19.496699669966997,
      "grad_norm": 0.00555419921875,
      "learning_rate": 5.7860324019465276e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37400592,
      "step": 177225
    },
    {
      "epoch": 19.497249724972498,
      "grad_norm": 0.005767822265625,
      "learning_rate": 5.773403121319576e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37401584,
      "step": 177230
    },
    {
      "epoch": 19.497799779978,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 5.760787612262008e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37402704,
      "step": 177235
    },
    {
      "epoch": 19.498349834983497,
      "grad_norm": 0.00543212890625,
      "learning_rate": 5.748185874890399e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37403728,
      "step": 177240
    },
    {
      "epoch": 19.498899889988998,
      "grad_norm": 0.005706787109375,
      "learning_rate": 5.735597909320655e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37404784,
      "step": 177245
    },
    {
      "epoch": 19.4994499449945,
      "grad_norm": 0.005615234375,
      "learning_rate": 5.723023715669017e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37405808,
      "step": 177250
    },
    {
      "epoch": 19.5,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 5.71046329405106e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37406832,
      "step": 177255
    },
    {
      "epoch": 19.5005500550055,
      "grad_norm": 0.005767822265625,
      "learning_rate": 5.697916644582857e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37407856,
      "step": 177260
    },
    {
      "epoch": 19.501100110011002,
      "grad_norm": 0.005462646484375,
      "learning_rate": 5.6853837673798144e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37408912,
      "step": 177265
    },
    {
      "epoch": 19.501650165016503,
      "grad_norm": 0.001129150390625,
      "learning_rate": 5.672864662557675e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37410064,
      "step": 177270
    },
    {
      "epoch": 19.502200220022,
      "grad_norm": 0.01129150390625,
      "learning_rate": 5.660359330231679e-05,
      "loss": 0.2345,
      "num_input_tokens_seen": 37411120,
      "step": 177275
    },
    {
      "epoch": 19.502750275027502,
      "grad_norm": 0.006134033203125,
      "learning_rate": 5.647867770516901e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37412176,
      "step": 177280
    },
    {
      "epoch": 19.503300330033003,
      "grad_norm": 0.00567626953125,
      "learning_rate": 5.6353899835289154e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37413264,
      "step": 177285
    },
    {
      "epoch": 19.503850385038504,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 5.622925969382297e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37414384,
      "step": 177290
    },
    {
      "epoch": 19.504400440044005,
      "grad_norm": 0.01092529296875,
      "learning_rate": 5.610475728192121e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37415504,
      "step": 177295
    },
    {
      "epoch": 19.504950495049506,
      "grad_norm": 0.006103515625,
      "learning_rate": 5.598039260073295e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37416528,
      "step": 177300
    },
    {
      "epoch": 19.505500550055004,
      "grad_norm": 0.00567626953125,
      "learning_rate": 5.585616565140061e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37417552,
      "step": 177305
    },
    {
      "epoch": 19.506050605060505,
      "grad_norm": 0.00194549560546875,
      "learning_rate": 5.5732076435071605e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37418576,
      "step": 177310
    },
    {
      "epoch": 19.506600660066006,
      "grad_norm": 0.0057373046875,
      "learning_rate": 5.560812495289002e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37419600,
      "step": 177315
    },
    {
      "epoch": 19.507150715071507,
      "grad_norm": 0.0010223388671875,
      "learning_rate": 5.5484311205996614e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37420656,
      "step": 177320
    },
    {
      "epoch": 19.507700770077008,
      "grad_norm": 0.01104736328125,
      "learning_rate": 5.53606351955338e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37421680,
      "step": 177325
    },
    {
      "epoch": 19.50825082508251,
      "grad_norm": 0.005767822265625,
      "learning_rate": 5.523709692264067e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37422768,
      "step": 177330
    },
    {
      "epoch": 19.50880088008801,
      "grad_norm": 0.00164794921875,
      "learning_rate": 5.5113696388454646e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37423760,
      "step": 177335
    },
    {
      "epoch": 19.509350935093508,
      "grad_norm": 0.005706787109375,
      "learning_rate": 5.4990433594114816e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37424784,
      "step": 177340
    },
    {
      "epoch": 19.50990099009901,
      "grad_norm": 0.00567626953125,
      "learning_rate": 5.4867308540756944e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37425840,
      "step": 177345
    },
    {
      "epoch": 19.51045104510451,
      "grad_norm": 0.01080322265625,
      "learning_rate": 5.474432122951678e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37426896,
      "step": 177350
    },
    {
      "epoch": 19.51100110011001,
      "grad_norm": 0.0017547607421875,
      "learning_rate": 5.4621471661526755e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37427952,
      "step": 177355
    },
    {
      "epoch": 19.511551155115512,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 5.4498759837919295e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37428976,
      "step": 177360
    },
    {
      "epoch": 19.512101210121013,
      "grad_norm": 0.005615234375,
      "learning_rate": 5.43761857598235e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37430000,
      "step": 177365
    },
    {
      "epoch": 19.51265126512651,
      "grad_norm": 0.010986328125,
      "learning_rate": 5.42537494283718e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37431024,
      "step": 177370
    },
    {
      "epoch": 19.513201320132012,
      "grad_norm": 0.00543212890625,
      "learning_rate": 5.413145084469162e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37432016,
      "step": 177375
    },
    {
      "epoch": 19.513751375137513,
      "grad_norm": 0.0057373046875,
      "learning_rate": 5.400929000991039e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37433040,
      "step": 177380
    },
    {
      "epoch": 19.514301430143014,
      "grad_norm": 0.0108642578125,
      "learning_rate": 5.388726692515388e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37434032,
      "step": 177385
    },
    {
      "epoch": 19.514851485148515,
      "grad_norm": 0.0059814453125,
      "learning_rate": 5.37653815915462e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37435088,
      "step": 177390
    },
    {
      "epoch": 19.515401540154016,
      "grad_norm": 0.00555419921875,
      "learning_rate": 5.364363401021144e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37436112,
      "step": 177395
    },
    {
      "epoch": 19.515951595159517,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 5.352202418227036e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37437168,
      "step": 177400
    },
    {
      "epoch": 19.516501650165015,
      "grad_norm": 0.0057373046875,
      "learning_rate": 5.340055210884542e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37438256,
      "step": 177405
    },
    {
      "epoch": 19.517051705170516,
      "grad_norm": 0.01092529296875,
      "learning_rate": 5.327921779105571e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37439344,
      "step": 177410
    },
    {
      "epoch": 19.517601760176017,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 5.315802123001867e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37440400,
      "step": 177415
    },
    {
      "epoch": 19.51815181518152,
      "grad_norm": 0.00537109375,
      "learning_rate": 5.303696242685341e-05,
      "loss": 0.2304,
      "num_input_tokens_seen": 37441520,
      "step": 177420
    },
    {
      "epoch": 19.51870187018702,
      "grad_norm": 0.005706787109375,
      "learning_rate": 5.291604138267236e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37442544,
      "step": 177425
    },
    {
      "epoch": 19.51925192519252,
      "grad_norm": 0.00119781494140625,
      "learning_rate": 5.279525809859298e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37443664,
      "step": 177430
    },
    {
      "epoch": 19.519801980198018,
      "grad_norm": 0.005523681640625,
      "learning_rate": 5.2674612575727696e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37444784,
      "step": 177435
    },
    {
      "epoch": 19.52035203520352,
      "grad_norm": 0.005279541015625,
      "learning_rate": 5.25541048151873e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37445840,
      "step": 177440
    },
    {
      "epoch": 19.52090209020902,
      "grad_norm": 0.001739501953125,
      "learning_rate": 5.243373481808422e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37446896,
      "step": 177445
    },
    {
      "epoch": 19.52145214521452,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 5.2313502585525916e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37447952,
      "step": 177450
    },
    {
      "epoch": 19.522002200220022,
      "grad_norm": 0.005615234375,
      "learning_rate": 5.219340811861983e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37449040,
      "step": 177455
    },
    {
      "epoch": 19.522552255225524,
      "grad_norm": 0.01092529296875,
      "learning_rate": 5.2073451418476744e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37450064,
      "step": 177460
    },
    {
      "epoch": 19.523102310231025,
      "grad_norm": 0.00567626953125,
      "learning_rate": 5.1953632486199105e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37451120,
      "step": 177465
    },
    {
      "epoch": 19.523652365236522,
      "grad_norm": 0.00555419921875,
      "learning_rate": 5.18339513228927e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37452208,
      "step": 177470
    },
    {
      "epoch": 19.524202420242023,
      "grad_norm": 0.005767822265625,
      "learning_rate": 5.1714407929659974e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37453296,
      "step": 177475
    },
    {
      "epoch": 19.524752475247524,
      "grad_norm": 0.005462646484375,
      "learning_rate": 5.1595002307603385e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37454416,
      "step": 177480
    },
    {
      "epoch": 19.525302530253025,
      "grad_norm": 0.005645751953125,
      "learning_rate": 5.1475734457820385e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37455440,
      "step": 177485
    },
    {
      "epoch": 19.525852585258527,
      "grad_norm": 0.002288818359375,
      "learning_rate": 5.135660438141343e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37456464,
      "step": 177490
    },
    {
      "epoch": 19.526402640264028,
      "grad_norm": 0.00543212890625,
      "learning_rate": 5.1237612079479966e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37457488,
      "step": 177495
    },
    {
      "epoch": 19.52695269526953,
      "grad_norm": 0.00128173828125,
      "learning_rate": 5.111875755311745e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37458576,
      "step": 177500
    },
    {
      "epoch": 19.527502750275026,
      "grad_norm": 0.005889892578125,
      "learning_rate": 5.1000040803418356e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37459600,
      "step": 177505
    },
    {
      "epoch": 19.528052805280527,
      "grad_norm": 0.005767822265625,
      "learning_rate": 5.088146183148012e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37460688,
      "step": 177510
    },
    {
      "epoch": 19.52860286028603,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 5.076302063839188e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37461712,
      "step": 177515
    },
    {
      "epoch": 19.52915291529153,
      "grad_norm": 0.005859375,
      "learning_rate": 5.064471722524943e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37462800,
      "step": 177520
    },
    {
      "epoch": 19.52970297029703,
      "grad_norm": 0.005462646484375,
      "learning_rate": 5.05265515931419e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37463792,
      "step": 177525
    },
    {
      "epoch": 19.53025302530253,
      "grad_norm": 0.01092529296875,
      "learning_rate": 5.0408523743156736e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37464880,
      "step": 177530
    },
    {
      "epoch": 19.53080308030803,
      "grad_norm": 0.005584716796875,
      "learning_rate": 5.029063367638142e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37465872,
      "step": 177535
    },
    {
      "epoch": 19.53135313531353,
      "grad_norm": 0.005279541015625,
      "learning_rate": 5.017288139390507e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37466896,
      "step": 177540
    },
    {
      "epoch": 19.53190319031903,
      "grad_norm": 0.00058746337890625,
      "learning_rate": 5.0055266896811836e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37467984,
      "step": 177545
    },
    {
      "epoch": 19.532453245324533,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 4.993779018618416e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37469104,
      "step": 177550
    },
    {
      "epoch": 19.533003300330034,
      "grad_norm": 0.0107421875,
      "learning_rate": 4.9820451263107855e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37470096,
      "step": 177555
    },
    {
      "epoch": 19.533553355335535,
      "grad_norm": 0.00592041015625,
      "learning_rate": 4.9703250128662054e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37471184,
      "step": 177560
    },
    {
      "epoch": 19.534103410341036,
      "grad_norm": 0.005889892578125,
      "learning_rate": 4.958618678392756e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37472336,
      "step": 177565
    },
    {
      "epoch": 19.534653465346533,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 4.946926122998352e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37473456,
      "step": 177570
    },
    {
      "epoch": 19.535203520352034,
      "grad_norm": 0.00128173828125,
      "learning_rate": 4.935247346790572e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37474480,
      "step": 177575
    },
    {
      "epoch": 19.535753575357536,
      "grad_norm": 0.005462646484375,
      "learning_rate": 4.9235823498774975e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37475472,
      "step": 177580
    },
    {
      "epoch": 19.536303630363037,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 4.9119311323662097e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37476560,
      "step": 177585
    },
    {
      "epoch": 19.536853685368538,
      "grad_norm": 0.001434326171875,
      "learning_rate": 4.900293694364122e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37477648,
      "step": 177590
    },
    {
      "epoch": 19.53740374037404,
      "grad_norm": 0.00213623046875,
      "learning_rate": 4.888670035978815e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37478704,
      "step": 177595
    },
    {
      "epoch": 19.537953795379536,
      "grad_norm": 0.00555419921875,
      "learning_rate": 4.877060157317037e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37479792,
      "step": 177600
    },
    {
      "epoch": 19.538503850385037,
      "grad_norm": 0.00543212890625,
      "learning_rate": 4.865464058486035e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37480880,
      "step": 177605
    },
    {
      "epoch": 19.53905390539054,
      "grad_norm": 0.005828857421875,
      "learning_rate": 4.8538817395927245e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37482064,
      "step": 177610
    },
    {
      "epoch": 19.53960396039604,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 4.8423132007435196e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37483152,
      "step": 177615
    },
    {
      "epoch": 19.54015401540154,
      "grad_norm": 0.00152587890625,
      "learning_rate": 4.830758442045502e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37484240,
      "step": 177620
    },
    {
      "epoch": 19.540704070407042,
      "grad_norm": 0.00555419921875,
      "learning_rate": 4.819217463604752e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37485264,
      "step": 177625
    },
    {
      "epoch": 19.541254125412543,
      "grad_norm": 0.005889892578125,
      "learning_rate": 4.807690265527853e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37486256,
      "step": 177630
    },
    {
      "epoch": 19.54180418041804,
      "grad_norm": 0.005615234375,
      "learning_rate": 4.796176847921218e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37487280,
      "step": 177635
    },
    {
      "epoch": 19.54235423542354,
      "grad_norm": 0.0057373046875,
      "learning_rate": 4.784677210890431e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37488368,
      "step": 177640
    },
    {
      "epoch": 19.542904290429043,
      "grad_norm": 0.001434326171875,
      "learning_rate": 4.773191354542072e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37489424,
      "step": 177645
    },
    {
      "epoch": 19.543454345434544,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 4.7617192789815574e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37490544,
      "step": 177650
    },
    {
      "epoch": 19.544004400440045,
      "grad_norm": 0.010986328125,
      "learning_rate": 4.750260984314802e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37491600,
      "step": 177655
    },
    {
      "epoch": 19.544554455445546,
      "grad_norm": 0.001190185546875,
      "learning_rate": 4.738816470647389e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37492592,
      "step": 177660
    },
    {
      "epoch": 19.545104510451043,
      "grad_norm": 0.00084686279296875,
      "learning_rate": 4.7273857380848995e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37493648,
      "step": 177665
    },
    {
      "epoch": 19.545654565456545,
      "grad_norm": 0.005584716796875,
      "learning_rate": 4.715968786732582e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37494672,
      "step": 177670
    },
    {
      "epoch": 19.546204620462046,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 4.7045656166956884e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37495728,
      "step": 177675
    },
    {
      "epoch": 19.546754675467547,
      "grad_norm": 0.00567626953125,
      "learning_rate": 4.693176228079299e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37496752,
      "step": 177680
    },
    {
      "epoch": 19.547304730473048,
      "grad_norm": 0.002105712890625,
      "learning_rate": 4.68180062098833e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37497776,
      "step": 177685
    },
    {
      "epoch": 19.54785478547855,
      "grad_norm": 0.0025787353515625,
      "learning_rate": 4.670438795527532e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37498768,
      "step": 177690
    },
    {
      "epoch": 19.54840484048405,
      "grad_norm": 0.005523681640625,
      "learning_rate": 4.6590907518019866e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37499888,
      "step": 177695
    },
    {
      "epoch": 19.548954895489548,
      "grad_norm": 0.005401611328125,
      "learning_rate": 4.64775648991611e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37500912,
      "step": 177700
    },
    {
      "epoch": 19.54950495049505,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 4.636436009974154e-05,
      "loss": 0.2345,
      "num_input_tokens_seen": 37501968,
      "step": 177705
    },
    {
      "epoch": 19.55005500550055,
      "grad_norm": 0.001556396484375,
      "learning_rate": 4.6251293120807e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37503088,
      "step": 177710
    },
    {
      "epoch": 19.55060506050605,
      "grad_norm": 0.01116943359375,
      "learning_rate": 4.613836396339832e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37504208,
      "step": 177715
    },
    {
      "epoch": 19.551155115511552,
      "grad_norm": 0.005950927734375,
      "learning_rate": 4.6025572628557995e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37505264,
      "step": 177720
    },
    {
      "epoch": 19.551705170517053,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 4.5912919117323534e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37506320,
      "step": 177725
    },
    {
      "epoch": 19.55225522552255,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 4.58004034307341e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37507408,
      "step": 177730
    },
    {
      "epoch": 19.55280528052805,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 4.568802556982721e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37508528,
      "step": 177735
    },
    {
      "epoch": 19.553355335533553,
      "grad_norm": 0.005706787109375,
      "learning_rate": 4.5575785535637016e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37509648,
      "step": 177740
    },
    {
      "epoch": 19.553905390539054,
      "grad_norm": 0.005523681640625,
      "learning_rate": 4.5463683329199365e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37510704,
      "step": 177745
    },
    {
      "epoch": 19.554455445544555,
      "grad_norm": 0.00099945068359375,
      "learning_rate": 4.535171895154677e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37511824,
      "step": 177750
    },
    {
      "epoch": 19.555005500550056,
      "grad_norm": 0.006134033203125,
      "learning_rate": 4.523989240371173e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37512912,
      "step": 177755
    },
    {
      "epoch": 19.555555555555557,
      "grad_norm": 0.005645751953125,
      "learning_rate": 4.51282036867251e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37513936,
      "step": 177760
    },
    {
      "epoch": 19.556105610561055,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 4.501665280161438e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37515024,
      "step": 177765
    },
    {
      "epoch": 19.556655665566556,
      "grad_norm": 0.005523681640625,
      "learning_rate": 4.4905239749410406e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37516016,
      "step": 177770
    },
    {
      "epoch": 19.557205720572057,
      "grad_norm": 0.006134033203125,
      "learning_rate": 4.4793964531137376e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37517104,
      "step": 177775
    },
    {
      "epoch": 19.557755775577558,
      "grad_norm": 0.00579833984375,
      "learning_rate": 4.4682827147822786e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37518224,
      "step": 177780
    },
    {
      "epoch": 19.55830583058306,
      "grad_norm": 0.001373291015625,
      "learning_rate": 4.457182760049083e-05,
      "loss": 0.2288,
      "num_input_tokens_seen": 37519312,
      "step": 177785
    },
    {
      "epoch": 19.55885588558856,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 4.446096589016402e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37520464,
      "step": 177790
    },
    {
      "epoch": 19.55940594059406,
      "grad_norm": 0.0020751953125,
      "learning_rate": 4.4350242017864856e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37521520,
      "step": 177795
    },
    {
      "epoch": 19.55995599559956,
      "grad_norm": 0.005462646484375,
      "learning_rate": 4.423965598461088e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37522608,
      "step": 177800
    },
    {
      "epoch": 19.56050605060506,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 4.4129207791424594e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37523600,
      "step": 177805
    },
    {
      "epoch": 19.56105610561056,
      "grad_norm": 0.00567626953125,
      "learning_rate": 4.401889743932352e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37524656,
      "step": 177810
    },
    {
      "epoch": 19.561606160616062,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 4.3908724929323514e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37525776,
      "step": 177815
    },
    {
      "epoch": 19.562156215621563,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 4.379869026243876e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37526832,
      "step": 177820
    },
    {
      "epoch": 19.562706270627064,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 4.368879343968679e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37527888,
      "step": 177825
    },
    {
      "epoch": 19.563256325632562,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 4.357903446207678e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37528912,
      "step": 177830
    },
    {
      "epoch": 19.563806380638063,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 4.346941333062293e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37529968,
      "step": 177835
    },
    {
      "epoch": 19.564356435643564,
      "grad_norm": 0.00531005859375,
      "learning_rate": 4.3359930046334423e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37531024,
      "step": 177840
    },
    {
      "epoch": 19.564906490649065,
      "grad_norm": 0.01123046875,
      "learning_rate": 4.3250584610220464e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37532080,
      "step": 177845
    },
    {
      "epoch": 19.565456545654566,
      "grad_norm": 0.01092529296875,
      "learning_rate": 4.314137702329024e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37533104,
      "step": 177850
    },
    {
      "epoch": 19.566006600660067,
      "grad_norm": 0.005401611328125,
      "learning_rate": 4.303230728654794e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37534128,
      "step": 177855
    },
    {
      "epoch": 19.566556655665565,
      "grad_norm": 0.01068115234375,
      "learning_rate": 4.292337540099944e-05,
      "loss": 0.2304,
      "num_input_tokens_seen": 37535248,
      "step": 177860
    },
    {
      "epoch": 19.567106710671066,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 4.281458136764893e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37536304,
      "step": 177865
    },
    {
      "epoch": 19.567656765676567,
      "grad_norm": 0.005523681640625,
      "learning_rate": 4.270592518749894e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37537360,
      "step": 177870
    },
    {
      "epoch": 19.568206820682068,
      "grad_norm": 0.005340576171875,
      "learning_rate": 4.2597406861551996e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37538352,
      "step": 177875
    },
    {
      "epoch": 19.56875687568757,
      "grad_norm": 0.00250244140625,
      "learning_rate": 4.248902639080732e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37539408,
      "step": 177880
    },
    {
      "epoch": 19.56930693069307,
      "grad_norm": 0.005859375,
      "learning_rate": 4.238078377626575e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37540432,
      "step": 177885
    },
    {
      "epoch": 19.56985698569857,
      "grad_norm": 0.005279541015625,
      "learning_rate": 4.2272679018919845e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37541392,
      "step": 177890
    },
    {
      "epoch": 19.57040704070407,
      "grad_norm": 0.01092529296875,
      "learning_rate": 4.2164712119770465e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37542448,
      "step": 177895
    },
    {
      "epoch": 19.57095709570957,
      "grad_norm": 0.0057373046875,
      "learning_rate": 4.2056883079813476e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37543472,
      "step": 177900
    },
    {
      "epoch": 19.57150715071507,
      "grad_norm": 0.01104736328125,
      "learning_rate": 4.19491919000381e-05,
      "loss": 0.2313,
      "num_input_tokens_seen": 37544528,
      "step": 177905
    },
    {
      "epoch": 19.572057205720572,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 4.184163858144019e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37545584,
      "step": 177910
    },
    {
      "epoch": 19.572607260726073,
      "grad_norm": 0.0108642578125,
      "learning_rate": 4.1734223125010625e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37546640,
      "step": 177915
    },
    {
      "epoch": 19.573157315731574,
      "grad_norm": 0.00128936767578125,
      "learning_rate": 4.162694553174029e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37547728,
      "step": 177920
    },
    {
      "epoch": 19.573707370737075,
      "grad_norm": 0.002685546875,
      "learning_rate": 4.151980580261505e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37548816,
      "step": 177925
    },
    {
      "epoch": 19.574257425742573,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 4.1412803938624117e-05,
      "loss": 0.2304,
      "num_input_tokens_seen": 37549968,
      "step": 177930
    },
    {
      "epoch": 19.574807480748074,
      "grad_norm": 0.006103515625,
      "learning_rate": 4.1305939940753376e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37551088,
      "step": 177935
    },
    {
      "epoch": 19.575357535753575,
      "grad_norm": 0.005523681640625,
      "learning_rate": 4.11992138099887e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37552144,
      "step": 177940
    },
    {
      "epoch": 19.575907590759076,
      "grad_norm": 0.01092529296875,
      "learning_rate": 4.109262554731263e-05,
      "loss": 0.234,
      "num_input_tokens_seen": 37553168,
      "step": 177945
    },
    {
      "epoch": 19.576457645764577,
      "grad_norm": 0.005706787109375,
      "learning_rate": 4.098617515370606e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37554224,
      "step": 177950
    },
    {
      "epoch": 19.57700770077008,
      "grad_norm": 0.00604248046875,
      "learning_rate": 4.0879862630153194e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37555312,
      "step": 177955
    },
    {
      "epoch": 19.577557755775576,
      "grad_norm": 0.005340576171875,
      "learning_rate": 4.0773687977633254e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37556368,
      "step": 177960
    },
    {
      "epoch": 19.578107810781077,
      "grad_norm": 0.005401611328125,
      "learning_rate": 4.0667651197125454e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37557424,
      "step": 177965
    },
    {
      "epoch": 19.578657865786578,
      "grad_norm": 0.005645751953125,
      "learning_rate": 4.0561752289602344e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37558544,
      "step": 177970
    },
    {
      "epoch": 19.57920792079208,
      "grad_norm": 0.001739501953125,
      "learning_rate": 4.045599125604648e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37559568,
      "step": 177975
    },
    {
      "epoch": 19.57975797579758,
      "grad_norm": 0.0108642578125,
      "learning_rate": 4.035036809742709e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37560624,
      "step": 177980
    },
    {
      "epoch": 19.58030803080308,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 4.024488281472005e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37561648,
      "step": 177985
    },
    {
      "epoch": 19.580858085808583,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 4.0139535408899585e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37562736,
      "step": 177990
    },
    {
      "epoch": 19.58140814081408,
      "grad_norm": 0.01104736328125,
      "learning_rate": 4.003432588093325e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37563760,
      "step": 177995
    },
    {
      "epoch": 19.58195819581958,
      "grad_norm": 0.005401611328125,
      "learning_rate": 3.992925423179194e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37564816,
      "step": 178000
    },
    {
      "epoch": 19.582508250825082,
      "grad_norm": 0.005615234375,
      "learning_rate": 3.982432046244322e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37565872,
      "step": 178005
    },
    {
      "epoch": 19.583058305830583,
      "grad_norm": 0.005767822265625,
      "learning_rate": 3.97195245738563e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37566960,
      "step": 178010
    },
    {
      "epoch": 19.583608360836084,
      "grad_norm": 0.00555419921875,
      "learning_rate": 3.9614866566997085e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37568016,
      "step": 178015
    },
    {
      "epoch": 19.584158415841586,
      "grad_norm": 0.00604248046875,
      "learning_rate": 3.951034644282647e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37569040,
      "step": 178020
    },
    {
      "epoch": 19.584708470847083,
      "grad_norm": 0.01116943359375,
      "learning_rate": 3.9405964202312014e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37570064,
      "step": 178025
    },
    {
      "epoch": 19.585258525852584,
      "grad_norm": 0.0015869140625,
      "learning_rate": 3.930171984641295e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37571056,
      "step": 178030
    },
    {
      "epoch": 19.585808580858085,
      "grad_norm": 0.0108642578125,
      "learning_rate": 3.9197613376091845e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37572112,
      "step": 178035
    },
    {
      "epoch": 19.586358635863586,
      "grad_norm": 0.0010986328125,
      "learning_rate": 3.9093644792307926e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37573136,
      "step": 178040
    },
    {
      "epoch": 19.586908690869087,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 3.8989814096018756e-05,
      "loss": 0.2288,
      "num_input_tokens_seen": 37574192,
      "step": 178045
    },
    {
      "epoch": 19.58745874587459,
      "grad_norm": 0.005523681640625,
      "learning_rate": 3.888612128818025e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37575280,
      "step": 178050
    },
    {
      "epoch": 19.58800880088009,
      "grad_norm": 0.005584716796875,
      "learning_rate": 3.878256636974997e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37576272,
      "step": 178055
    },
    {
      "epoch": 19.588558855885587,
      "grad_norm": 0.005706787109375,
      "learning_rate": 3.867914934168215e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37577328,
      "step": 178060
    },
    {
      "epoch": 19.58910891089109,
      "grad_norm": 0.010986328125,
      "learning_rate": 3.8575870204929365e-05,
      "loss": 0.234,
      "num_input_tokens_seen": 37578416,
      "step": 178065
    },
    {
      "epoch": 19.58965896589659,
      "grad_norm": 0.00177764892578125,
      "learning_rate": 3.847272896044251e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37579504,
      "step": 178070
    },
    {
      "epoch": 19.59020902090209,
      "grad_norm": 0.005401611328125,
      "learning_rate": 3.836972560917418e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37580592,
      "step": 178075
    },
    {
      "epoch": 19.59075907590759,
      "grad_norm": 0.005767822265625,
      "learning_rate": 3.826686015207192e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37581616,
      "step": 178080
    },
    {
      "epoch": 19.591309130913093,
      "grad_norm": 0.00579833984375,
      "learning_rate": 3.8164132590086665e-05,
      "loss": 0.233,
      "num_input_tokens_seen": 37582640,
      "step": 178085
    },
    {
      "epoch": 19.59185918591859,
      "grad_norm": 0.01104736328125,
      "learning_rate": 3.806154292416264e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37583696,
      "step": 178090
    },
    {
      "epoch": 19.59240924092409,
      "grad_norm": 0.00567626953125,
      "learning_rate": 3.795909115524409e-05,
      "loss": 0.233,
      "num_input_tokens_seen": 37584720,
      "step": 178095
    },
    {
      "epoch": 19.592959295929592,
      "grad_norm": 0.005462646484375,
      "learning_rate": 3.7856777284278606e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37585744,
      "step": 178100
    },
    {
      "epoch": 19.593509350935093,
      "grad_norm": 0.005828857421875,
      "learning_rate": 3.775460131220709e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37586832,
      "step": 178105
    },
    {
      "epoch": 19.594059405940595,
      "grad_norm": 0.005615234375,
      "learning_rate": 3.765256323997212e-05,
      "loss": 0.2313,
      "num_input_tokens_seen": 37587856,
      "step": 178110
    },
    {
      "epoch": 19.594609460946096,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 3.755066306851462e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37588944,
      "step": 178115
    },
    {
      "epoch": 19.595159515951597,
      "grad_norm": 0.0054931640625,
      "learning_rate": 3.744890079877217e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37589968,
      "step": 178120
    },
    {
      "epoch": 19.595709570957094,
      "grad_norm": 0.001434326171875,
      "learning_rate": 3.734727643168234e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37591024,
      "step": 178125
    },
    {
      "epoch": 19.596259625962595,
      "grad_norm": 0.00555419921875,
      "learning_rate": 3.724578996818439e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37592112,
      "step": 178130
    },
    {
      "epoch": 19.596809680968097,
      "grad_norm": 0.001922607421875,
      "learning_rate": 3.714444140921258e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37593200,
      "step": 178135
    },
    {
      "epoch": 19.597359735973598,
      "grad_norm": 0.005523681640625,
      "learning_rate": 3.7043230755699485e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37594224,
      "step": 178140
    },
    {
      "epoch": 19.5979097909791,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 3.694215800857936e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37595312,
      "step": 178145
    },
    {
      "epoch": 19.5984598459846,
      "grad_norm": 0.000583648681640625,
      "learning_rate": 3.684122316878313e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37596336,
      "step": 178150
    },
    {
      "epoch": 19.599009900990097,
      "grad_norm": 0.0108642578125,
      "learning_rate": 3.674042623724172e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37597424,
      "step": 178155
    },
    {
      "epoch": 19.5995599559956,
      "grad_norm": 0.00115966796875,
      "learning_rate": 3.663976721488271e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37598448,
      "step": 178160
    },
    {
      "epoch": 19.6001100110011,
      "grad_norm": 0.001068115234375,
      "learning_rate": 3.653924610263703e-05,
      "loss": 0.2299,
      "num_input_tokens_seen": 37599504,
      "step": 178165
    },
    {
      "epoch": 19.6006600660066,
      "grad_norm": 0.001373291015625,
      "learning_rate": 3.643886290142728e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37600624,
      "step": 178170
    },
    {
      "epoch": 19.6012101210121,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 3.6338617612181046e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37601616,
      "step": 178175
    },
    {
      "epoch": 19.601760176017603,
      "grad_norm": 0.005767822265625,
      "learning_rate": 3.623851023582092e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37602576,
      "step": 178180
    },
    {
      "epoch": 19.602310231023104,
      "grad_norm": 0.005523681640625,
      "learning_rate": 3.61385407732695e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37603632,
      "step": 178185
    },
    {
      "epoch": 19.6028602860286,
      "grad_norm": 0.0054931640625,
      "learning_rate": 3.603870922544938e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37604688,
      "step": 178190
    },
    {
      "epoch": 19.603410341034103,
      "grad_norm": 0.00579833984375,
      "learning_rate": 3.59390155932815e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37605776,
      "step": 178195
    },
    {
      "epoch": 19.603960396039604,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 3.583945987768178e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37606832,
      "step": 178200
    },
    {
      "epoch": 19.604510451045105,
      "grad_norm": 0.005401611328125,
      "learning_rate": 3.574004207956949e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37607920,
      "step": 178205
    },
    {
      "epoch": 19.605060506050606,
      "grad_norm": 0.00060272216796875,
      "learning_rate": 3.5640762199860563e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37608912,
      "step": 178210
    },
    {
      "epoch": 19.605610561056107,
      "grad_norm": 0.005889892578125,
      "learning_rate": 3.554162023947094e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37610032,
      "step": 178215
    },
    {
      "epoch": 19.606160616061608,
      "grad_norm": 0.01116943359375,
      "learning_rate": 3.544261619931155e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37611120,
      "step": 178220
    },
    {
      "epoch": 19.606710671067106,
      "grad_norm": 0.01116943359375,
      "learning_rate": 3.534375008029833e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37612240,
      "step": 178225
    },
    {
      "epoch": 19.607260726072607,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 3.524502188334222e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37613264,
      "step": 178230
    },
    {
      "epoch": 19.607810781078108,
      "grad_norm": 0.001861572265625,
      "learning_rate": 3.5146431609350824e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37614352,
      "step": 178235
    },
    {
      "epoch": 19.60836083608361,
      "grad_norm": 0.001495361328125,
      "learning_rate": 3.504797925923342e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37615440,
      "step": 178240
    },
    {
      "epoch": 19.60891089108911,
      "grad_norm": 0.0007476806640625,
      "learning_rate": 3.494966483389761e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37616528,
      "step": 178245
    },
    {
      "epoch": 19.60946094609461,
      "grad_norm": 0.005615234375,
      "learning_rate": 3.4851488334251e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37617552,
      "step": 178250
    },
    {
      "epoch": 19.61001100110011,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 3.4753449761196206e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37618576,
      "step": 178255
    },
    {
      "epoch": 19.61056105610561,
      "grad_norm": 0.005523681640625,
      "learning_rate": 3.465554911563917e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37619664,
      "step": 178260
    },
    {
      "epoch": 19.61111111111111,
      "grad_norm": 0.0021820068359375,
      "learning_rate": 3.4557786398479174e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37620848,
      "step": 178265
    },
    {
      "epoch": 19.611661166116612,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 3.446016161062049e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37621872,
      "step": 178270
    },
    {
      "epoch": 19.612211221122113,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 3.436267475296073e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37622928,
      "step": 178275
    },
    {
      "epoch": 19.612761276127614,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 3.4265325826399184e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37623984,
      "step": 178280
    },
    {
      "epoch": 19.61331133113311,
      "grad_norm": 0.006011962890625,
      "learning_rate": 3.4168114831831797e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37624976,
      "step": 178285
    },
    {
      "epoch": 19.613861386138613,
      "grad_norm": 0.005889892578125,
      "learning_rate": 3.407104177015618e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37626064,
      "step": 178290
    },
    {
      "epoch": 19.614411441144114,
      "grad_norm": 0.0054931640625,
      "learning_rate": 3.397410664226663e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37627120,
      "step": 178295
    },
    {
      "epoch": 19.614961496149615,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 3.3877309449057424e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37628272,
      "step": 178300
    },
    {
      "epoch": 19.615511551155116,
      "grad_norm": 0.001220703125,
      "learning_rate": 3.378065019141785e-05,
      "loss": 0.234,
      "num_input_tokens_seen": 37629328,
      "step": 178305
    },
    {
      "epoch": 19.616061606160617,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 3.368412887024219e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37630352,
      "step": 178310
    },
    {
      "epoch": 19.616611661166118,
      "grad_norm": 0.01104736328125,
      "learning_rate": 3.358774548641641e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37631472,
      "step": 178315
    },
    {
      "epoch": 19.617161716171616,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 3.3491500040831456e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37632528,
      "step": 178320
    },
    {
      "epoch": 19.617711771177117,
      "grad_norm": 0.010986328125,
      "learning_rate": 3.3395392534374953e-05,
      "loss": 0.2288,
      "num_input_tokens_seen": 37633584,
      "step": 178325
    },
    {
      "epoch": 19.618261826182618,
      "grad_norm": 0.00140380859375,
      "learning_rate": 3.32994229679312e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37634640,
      "step": 178330
    },
    {
      "epoch": 19.61881188118812,
      "grad_norm": 0.01141357421875,
      "learning_rate": 3.320359134238282e-05,
      "loss": 0.233,
      "num_input_tokens_seen": 37635696,
      "step": 178335
    },
    {
      "epoch": 19.61936193619362,
      "grad_norm": 0.005615234375,
      "learning_rate": 3.3107897658617435e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37636784,
      "step": 178340
    },
    {
      "epoch": 19.61991199119912,
      "grad_norm": 0.005645751953125,
      "learning_rate": 3.301234191751434e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37637808,
      "step": 178345
    },
    {
      "epoch": 19.620462046204622,
      "grad_norm": 0.005462646484375,
      "learning_rate": 3.2916924119954504e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37638896,
      "step": 178350
    },
    {
      "epoch": 19.62101210121012,
      "grad_norm": 0.005584716796875,
      "learning_rate": 3.282164426681722e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37640016,
      "step": 178355
    },
    {
      "epoch": 19.62156215621562,
      "grad_norm": 0.005615234375,
      "learning_rate": 3.272650235898011e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37640976,
      "step": 178360
    },
    {
      "epoch": 19.622112211221122,
      "grad_norm": 0.00579833984375,
      "learning_rate": 3.263149839732249e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37642032,
      "step": 178365
    },
    {
      "epoch": 19.622662266226623,
      "grad_norm": 0.00131988525390625,
      "learning_rate": 3.253663238271698e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37643120,
      "step": 178370
    },
    {
      "epoch": 19.623212321232124,
      "grad_norm": 0.000701904296875,
      "learning_rate": 3.244190431603955e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37644112,
      "step": 178375
    },
    {
      "epoch": 19.623762376237625,
      "grad_norm": 0.005462646484375,
      "learning_rate": 3.2347314198162836e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37645168,
      "step": 178380
    },
    {
      "epoch": 19.624312431243123,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 3.2252862029959474e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37646256,
      "step": 178385
    },
    {
      "epoch": 19.624862486248624,
      "grad_norm": 0.001922607421875,
      "learning_rate": 3.2158547812298764e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37647376,
      "step": 178390
    },
    {
      "epoch": 19.625412541254125,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 3.206437154605002e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37648496,
      "step": 178395
    },
    {
      "epoch": 19.625962596259626,
      "grad_norm": 0.0054931640625,
      "learning_rate": 3.197033323208253e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37649520,
      "step": 178400
    },
    {
      "epoch": 19.626512651265127,
      "grad_norm": 0.006011962890625,
      "learning_rate": 3.187643287126229e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37650576,
      "step": 178405
    },
    {
      "epoch": 19.627062706270628,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 3.17826704644536e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37651696,
      "step": 178410
    },
    {
      "epoch": 19.62761276127613,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 3.168904601252076e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37652752,
      "step": 178415
    },
    {
      "epoch": 19.628162816281627,
      "grad_norm": 0.005523681640625,
      "learning_rate": 3.159555951632975e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37653808,
      "step": 178420
    },
    {
      "epoch": 19.628712871287128,
      "grad_norm": 0.005706787109375,
      "learning_rate": 3.150221097673822e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37654896,
      "step": 178425
    },
    {
      "epoch": 19.62926292629263,
      "grad_norm": 0.0013427734375,
      "learning_rate": 3.140900039460881e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37655984,
      "step": 178430
    },
    {
      "epoch": 19.62981298129813,
      "grad_norm": 0.005523681640625,
      "learning_rate": 3.131592777080083e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37657008,
      "step": 178435
    },
    {
      "epoch": 19.63036303630363,
      "grad_norm": 0.0013885498046875,
      "learning_rate": 3.122299310617027e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37658032,
      "step": 178440
    },
    {
      "epoch": 19.630913091309132,
      "grad_norm": 0.01123046875,
      "learning_rate": 3.1130196401574774e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37659088,
      "step": 178445
    },
    {
      "epoch": 19.63146314631463,
      "grad_norm": 0.002197265625,
      "learning_rate": 3.1037537657870314e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37660176,
      "step": 178450
    },
    {
      "epoch": 19.63201320132013,
      "grad_norm": 0.005523681640625,
      "learning_rate": 3.094501687590956e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37661232,
      "step": 178455
    },
    {
      "epoch": 19.632563256325632,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 3.0852634056545144e-05,
      "loss": 0.233,
      "num_input_tokens_seen": 37662256,
      "step": 178460
    },
    {
      "epoch": 19.633113311331133,
      "grad_norm": 0.005523681640625,
      "learning_rate": 3.0760389200629736e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37663280,
      "step": 178465
    },
    {
      "epoch": 19.633663366336634,
      "grad_norm": 0.005401611328125,
      "learning_rate": 3.066828230901264e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37664368,
      "step": 178470
    },
    {
      "epoch": 19.634213421342135,
      "grad_norm": 0.006103515625,
      "learning_rate": 3.057631338254152e-05,
      "loss": 0.2288,
      "num_input_tokens_seen": 37665520,
      "step": 178475
    },
    {
      "epoch": 19.634763476347636,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 3.048448242206736e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37666640,
      "step": 178480
    },
    {
      "epoch": 19.635313531353134,
      "grad_norm": 0.01104736328125,
      "learning_rate": 3.0392789428434484e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37667696,
      "step": 178485
    },
    {
      "epoch": 19.635863586358635,
      "grad_norm": 0.005523681640625,
      "learning_rate": 3.0301234402488884e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37668688,
      "step": 178490
    },
    {
      "epoch": 19.636413641364136,
      "grad_norm": 0.005889892578125,
      "learning_rate": 3.020981734507322e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37669808,
      "step": 178495
    },
    {
      "epoch": 19.636963696369637,
      "grad_norm": 0.00555419921875,
      "learning_rate": 3.0118538257028482e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37670864,
      "step": 178500
    },
    {
      "epoch": 19.63751375137514,
      "grad_norm": 0.0111083984375,
      "learning_rate": 3.002739713920066e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37671920,
      "step": 178505
    },
    {
      "epoch": 19.63806380638064,
      "grad_norm": 0.0057373046875,
      "learning_rate": 2.993639399242409e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37673104,
      "step": 178510
    },
    {
      "epoch": 19.638613861386137,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 2.9845528817543096e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37674160,
      "step": 178515
    },
    {
      "epoch": 19.639163916391638,
      "grad_norm": 0.00543212890625,
      "learning_rate": 2.9754801615392013e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37675312,
      "step": 178520
    },
    {
      "epoch": 19.63971397139714,
      "grad_norm": 0.00186920166015625,
      "learning_rate": 2.966421238680683e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37676400,
      "step": 178525
    },
    {
      "epoch": 19.64026402640264,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 2.9573761132623555e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37677520,
      "step": 178530
    },
    {
      "epoch": 19.64081408140814,
      "grad_norm": 0.005767822265625,
      "learning_rate": 2.948344785367485e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37678608,
      "step": 178535
    },
    {
      "epoch": 19.641364136413642,
      "grad_norm": 0.005584716796875,
      "learning_rate": 2.939327255079338e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37679568,
      "step": 178540
    },
    {
      "epoch": 19.641914191419144,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 2.9303235224811818e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37680624,
      "step": 178545
    },
    {
      "epoch": 19.64246424642464,
      "grad_norm": 0.00628662109375,
      "learning_rate": 2.9213335876557832e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37681712,
      "step": 178550
    },
    {
      "epoch": 19.643014301430142,
      "grad_norm": 0.005584716796875,
      "learning_rate": 2.912357450686076e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37682832,
      "step": 178555
    },
    {
      "epoch": 19.643564356435643,
      "grad_norm": 0.01092529296875,
      "learning_rate": 2.9033951116548274e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37683792,
      "step": 178560
    },
    {
      "epoch": 19.644114411441144,
      "grad_norm": 0.005523681640625,
      "learning_rate": 2.8944465706446374e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37684816,
      "step": 178565
    },
    {
      "epoch": 19.644664466446645,
      "grad_norm": 0.000457763671875,
      "learning_rate": 2.8855118277379407e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37685936,
      "step": 178570
    },
    {
      "epoch": 19.645214521452147,
      "grad_norm": 0.002227783203125,
      "learning_rate": 2.876590883017005e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37686960,
      "step": 178575
    },
    {
      "epoch": 19.645764576457644,
      "grad_norm": 0.0009765625,
      "learning_rate": 2.8676837365642637e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37688080,
      "step": 178580
    },
    {
      "epoch": 19.646314631463145,
      "grad_norm": 0.001190185546875,
      "learning_rate": 2.8587903884616517e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37689072,
      "step": 178585
    },
    {
      "epoch": 19.646864686468646,
      "grad_norm": 0.005401611328125,
      "learning_rate": 2.8499108387911032e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37690096,
      "step": 178590
    },
    {
      "epoch": 19.647414741474147,
      "grad_norm": 0.005859375,
      "learning_rate": 2.8410450876345527e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37691120,
      "step": 178595
    },
    {
      "epoch": 19.64796479647965,
      "grad_norm": 0.006134033203125,
      "learning_rate": 2.8321931350737684e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37692144,
      "step": 178600
    },
    {
      "epoch": 19.64851485148515,
      "grad_norm": 0.00537109375,
      "learning_rate": 2.8233549811901846e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37693136,
      "step": 178605
    },
    {
      "epoch": 19.64906490649065,
      "grad_norm": 0.005615234375,
      "learning_rate": 2.8145306260654033e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37694224,
      "step": 178610
    },
    {
      "epoch": 19.649614961496148,
      "grad_norm": 0.00238037109375,
      "learning_rate": 2.805720069780526e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37695280,
      "step": 178615
    },
    {
      "epoch": 19.65016501650165,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 2.7969233124169877e-05,
      "loss": 0.2299,
      "num_input_tokens_seen": 37696368,
      "step": 178620
    },
    {
      "epoch": 19.65071507150715,
      "grad_norm": 0.005584716796875,
      "learning_rate": 2.7881403540557235e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37697456,
      "step": 178625
    },
    {
      "epoch": 19.65126512651265,
      "grad_norm": 0.005584716796875,
      "learning_rate": 2.7793711947778353e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37698512,
      "step": 178630
    },
    {
      "epoch": 19.651815181518153,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 2.7706158346639253e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37699568,
      "step": 178635
    },
    {
      "epoch": 19.652365236523654,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 2.7618742737949285e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37700560,
      "step": 178640
    },
    {
      "epoch": 19.652915291529155,
      "grad_norm": 0.005706787109375,
      "learning_rate": 2.753146512251281e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37701616,
      "step": 178645
    },
    {
      "epoch": 19.653465346534652,
      "grad_norm": 0.000579833984375,
      "learning_rate": 2.7444325501132514e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37702640,
      "step": 178650
    },
    {
      "epoch": 19.654015401540153,
      "grad_norm": 0.00555419921875,
      "learning_rate": 2.7357323874614423e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37703696,
      "step": 178655
    },
    {
      "epoch": 19.654565456545654,
      "grad_norm": 0.0010986328125,
      "learning_rate": 2.727046024376123e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37704720,
      "step": 178660
    },
    {
      "epoch": 19.655115511551156,
      "grad_norm": 0.005584716796875,
      "learning_rate": 2.7183734609368958e-05,
      "loss": 0.233,
      "num_input_tokens_seen": 37705808,
      "step": 178665
    },
    {
      "epoch": 19.655665566556657,
      "grad_norm": 0.01092529296875,
      "learning_rate": 2.7097146972240305e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37706800,
      "step": 178670
    },
    {
      "epoch": 19.656215621562158,
      "grad_norm": 0.005615234375,
      "learning_rate": 2.701069733317296e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37707824,
      "step": 178675
    },
    {
      "epoch": 19.656765676567655,
      "grad_norm": 0.002197265625,
      "learning_rate": 2.6924385692962958e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37708880,
      "step": 178680
    },
    {
      "epoch": 19.657315731573156,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 2.6838212052406328e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37709968,
      "step": 178685
    },
    {
      "epoch": 19.657865786578657,
      "grad_norm": 0.01104736328125,
      "learning_rate": 2.6752176412297433e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37710992,
      "step": 178690
    },
    {
      "epoch": 19.65841584158416,
      "grad_norm": 0.005828857421875,
      "learning_rate": 2.6666278773428975e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37712080,
      "step": 178695
    },
    {
      "epoch": 19.65896589658966,
      "grad_norm": 0.010986328125,
      "learning_rate": 2.6580519136591984e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37713168,
      "step": 178700
    },
    {
      "epoch": 19.65951595159516,
      "grad_norm": 0.0107421875,
      "learning_rate": 2.6494897502577497e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37714224,
      "step": 178705
    },
    {
      "epoch": 19.66006600660066,
      "grad_norm": 0.005615234375,
      "learning_rate": 2.6409413872173214e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37715312,
      "step": 178710
    },
    {
      "epoch": 19.66061606160616,
      "grad_norm": 0.005401611328125,
      "learning_rate": 2.6324068246170173e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37716368,
      "step": 178715
    },
    {
      "epoch": 19.66116611661166,
      "grad_norm": 0.000736236572265625,
      "learning_rate": 2.6238860625352744e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37717424,
      "step": 178720
    },
    {
      "epoch": 19.66171617161716,
      "grad_norm": 0.0010528564453125,
      "learning_rate": 2.61537910105053e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37718512,
      "step": 178725
    },
    {
      "epoch": 19.662266226622663,
      "grad_norm": 0.00604248046875,
      "learning_rate": 2.6068859402415543e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37719536,
      "step": 178730
    },
    {
      "epoch": 19.662816281628164,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 2.5984065801862854e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37720528,
      "step": 178735
    },
    {
      "epoch": 19.663366336633665,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 2.589941020962827e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37721552,
      "step": 178740
    },
    {
      "epoch": 19.663916391639162,
      "grad_norm": 0.005615234375,
      "learning_rate": 2.5814892626496167e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37722672,
      "step": 178745
    },
    {
      "epoch": 19.664466446644663,
      "grad_norm": 0.005615234375,
      "learning_rate": 2.5730513053240923e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37723728,
      "step": 178750
    },
    {
      "epoch": 19.665016501650165,
      "grad_norm": 0.00115966796875,
      "learning_rate": 2.564627149064358e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37724752,
      "step": 178755
    },
    {
      "epoch": 19.665566556655666,
      "grad_norm": 0.010986328125,
      "learning_rate": 2.556216793947852e-05,
      "loss": 0.2356,
      "num_input_tokens_seen": 37725840,
      "step": 178760
    },
    {
      "epoch": 19.666116611661167,
      "grad_norm": 0.0062255859375,
      "learning_rate": 2.5478202400521787e-05,
      "loss": 0.2313,
      "num_input_tokens_seen": 37726960,
      "step": 178765
    },
    {
      "epoch": 19.666666666666668,
      "grad_norm": 0.005767822265625,
      "learning_rate": 2.53943748745461e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37728048,
      "step": 178770
    },
    {
      "epoch": 19.66721672167217,
      "grad_norm": 0.01092529296875,
      "learning_rate": 2.5310685362325835e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37729136,
      "step": 178775
    },
    {
      "epoch": 19.667766776677666,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 2.522713386463038e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37730096,
      "step": 178780
    },
    {
      "epoch": 19.668316831683168,
      "grad_norm": 0.01092529296875,
      "learning_rate": 2.5143720382232446e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37731120,
      "step": 178785
    },
    {
      "epoch": 19.66886688668867,
      "grad_norm": 0.00543212890625,
      "learning_rate": 2.5060444915898093e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37732144,
      "step": 178790
    },
    {
      "epoch": 19.66941694169417,
      "grad_norm": 0.005767822265625,
      "learning_rate": 2.49773074663967e-05,
      "loss": 0.233,
      "num_input_tokens_seen": 37733264,
      "step": 178795
    },
    {
      "epoch": 19.66996699669967,
      "grad_norm": 0.006195068359375,
      "learning_rate": 2.489430803449266e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37734320,
      "step": 178800
    },
    {
      "epoch": 19.670517051705172,
      "grad_norm": 0.005767822265625,
      "learning_rate": 2.4811446620952027e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37735440,
      "step": 178805
    },
    {
      "epoch": 19.67106710671067,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 2.4728723226539184e-05,
      "loss": 0.233,
      "num_input_tokens_seen": 37736496,
      "step": 178810
    },
    {
      "epoch": 19.67161716171617,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 2.4646137852016858e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37737616,
      "step": 178815
    },
    {
      "epoch": 19.67216721672167,
      "grad_norm": 0.0016326904296875,
      "learning_rate": 2.456369049814444e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37738640,
      "step": 178820
    },
    {
      "epoch": 19.672717271727173,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 2.4481381165681325e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37739728,
      "step": 178825
    },
    {
      "epoch": 19.673267326732674,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 2.439920985538857e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37740784,
      "step": 178830
    },
    {
      "epoch": 19.673817381738175,
      "grad_norm": 0.00543212890625,
      "learning_rate": 2.4317176568022235e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37741840,
      "step": 178835
    },
    {
      "epoch": 19.674367436743676,
      "grad_norm": 0.005767822265625,
      "learning_rate": 2.4235281304338385e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37742864,
      "step": 178840
    },
    {
      "epoch": 19.674917491749174,
      "grad_norm": 0.0054931640625,
      "learning_rate": 2.4153524065091413e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37743920,
      "step": 178845
    },
    {
      "epoch": 19.675467546754675,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 2.4071904851035717e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37745008,
      "step": 178850
    },
    {
      "epoch": 19.676017601760176,
      "grad_norm": 0.001495361328125,
      "learning_rate": 2.3990423662922367e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37746096,
      "step": 178855
    },
    {
      "epoch": 19.676567656765677,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 2.3909080501504088e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37747216,
      "step": 178860
    },
    {
      "epoch": 19.677117711771178,
      "grad_norm": 0.005584716796875,
      "learning_rate": 2.3827875367528617e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37748176,
      "step": 178865
    },
    {
      "epoch": 19.67766776677668,
      "grad_norm": 0.00555419921875,
      "learning_rate": 2.374680826174702e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37749264,
      "step": 178870
    },
    {
      "epoch": 19.678217821782177,
      "grad_norm": 0.0054931640625,
      "learning_rate": 2.3665879184903703e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37750320,
      "step": 178875
    },
    {
      "epoch": 19.678767876787678,
      "grad_norm": 0.005645751953125,
      "learning_rate": 2.35850881377464e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37751344,
      "step": 178880
    },
    {
      "epoch": 19.67931793179318,
      "grad_norm": 0.010986328125,
      "learning_rate": 2.350443512101785e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37752368,
      "step": 178885
    },
    {
      "epoch": 19.67986798679868,
      "grad_norm": 0.00592041015625,
      "learning_rate": 2.342392013546246e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37753392,
      "step": 178890
    },
    {
      "epoch": 19.68041804180418,
      "grad_norm": 0.006011962890625,
      "learning_rate": 2.3343543181822966e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37754416,
      "step": 178895
    },
    {
      "epoch": 19.680968096809682,
      "grad_norm": 0.00543212890625,
      "learning_rate": 2.3263304260838777e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37755472,
      "step": 178900
    },
    {
      "epoch": 19.681518151815183,
      "grad_norm": 0.00543212890625,
      "learning_rate": 2.318320337325097e-05,
      "loss": 0.2288,
      "num_input_tokens_seen": 37756496,
      "step": 178905
    },
    {
      "epoch": 19.68206820682068,
      "grad_norm": 0.005828857421875,
      "learning_rate": 2.3103240519797284e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37757520,
      "step": 178910
    },
    {
      "epoch": 19.682618261826182,
      "grad_norm": 0.005889892578125,
      "learning_rate": 2.30234157012138e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37758576,
      "step": 178915
    },
    {
      "epoch": 19.683168316831683,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 2.2943728918236595e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37759600,
      "step": 178920
    },
    {
      "epoch": 19.683718371837184,
      "grad_norm": 0.00182342529296875,
      "learning_rate": 2.2864180171601742e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37760656,
      "step": 178925
    },
    {
      "epoch": 19.684268426842685,
      "grad_norm": 0.005615234375,
      "learning_rate": 2.2784769462038666e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37761680,
      "step": 178930
    },
    {
      "epoch": 19.684818481848186,
      "grad_norm": 0.005859375,
      "learning_rate": 2.270549679028344e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37762736,
      "step": 178935
    },
    {
      "epoch": 19.685368536853684,
      "grad_norm": 0.00543212890625,
      "learning_rate": 2.2626362157063815e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37763792,
      "step": 178940
    },
    {
      "epoch": 19.685918591859185,
      "grad_norm": 0.005523681640625,
      "learning_rate": 2.2547365563110876e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37764816,
      "step": 178945
    },
    {
      "epoch": 19.686468646864686,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 2.246850700915237e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37765808,
      "step": 178950
    },
    {
      "epoch": 19.687018701870187,
      "grad_norm": 0.0059814453125,
      "learning_rate": 2.2389786495914387e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37766832,
      "step": 178955
    },
    {
      "epoch": 19.687568756875688,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 2.2311204024121345e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37767856,
      "step": 178960
    },
    {
      "epoch": 19.68811881188119,
      "grad_norm": 0.00616455078125,
      "learning_rate": 2.2232759594500993e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37768880,
      "step": 178965
    },
    {
      "epoch": 19.68866886688669,
      "grad_norm": 0.005584716796875,
      "learning_rate": 2.215445320777276e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37769968,
      "step": 178970
    },
    {
      "epoch": 19.689218921892188,
      "grad_norm": 0.0108642578125,
      "learning_rate": 2.2076284864661065e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37770992,
      "step": 178975
    },
    {
      "epoch": 19.68976897689769,
      "grad_norm": 0.00604248046875,
      "learning_rate": 2.1998254565883667e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37772048,
      "step": 178980
    },
    {
      "epoch": 19.69031903190319,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 2.192036231216332e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37773104,
      "step": 178985
    },
    {
      "epoch": 19.69086908690869,
      "grad_norm": 0.0054931640625,
      "learning_rate": 2.1842608104214457e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37774288,
      "step": 178990
    },
    {
      "epoch": 19.691419141914192,
      "grad_norm": 0.0013427734375,
      "learning_rate": 2.176499194275483e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37775344,
      "step": 178995
    },
    {
      "epoch": 19.691969196919693,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 2.1687513828500538e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37776432,
      "step": 179000
    },
    {
      "epoch": 19.69251925192519,
      "grad_norm": 0.005706787109375,
      "learning_rate": 2.161017376216434e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37777456,
      "step": 179005
    },
    {
      "epoch": 19.693069306930692,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 2.1532971744460672e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37778512,
      "step": 179010
    },
    {
      "epoch": 19.693619361936193,
      "grad_norm": 0.0108642578125,
      "learning_rate": 2.1455907776100624e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37779568,
      "step": 179015
    },
    {
      "epoch": 19.694169416941694,
      "grad_norm": 0.00579833984375,
      "learning_rate": 2.137898185779363e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37780624,
      "step": 179020
    },
    {
      "epoch": 19.694719471947195,
      "grad_norm": 0.005584716796875,
      "learning_rate": 2.130219399024913e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37781680,
      "step": 179025
    },
    {
      "epoch": 19.695269526952696,
      "grad_norm": 0.0020751953125,
      "learning_rate": 2.1225544174173216e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37782768,
      "step": 179030
    },
    {
      "epoch": 19.695819581958197,
      "grad_norm": 0.005615234375,
      "learning_rate": 2.114903241027699e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37783792,
      "step": 179035
    },
    {
      "epoch": 19.696369636963695,
      "grad_norm": 0.00151824951171875,
      "learning_rate": 2.1072658699259892e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37784816,
      "step": 179040
    },
    {
      "epoch": 19.696919691969196,
      "grad_norm": 0.005584716796875,
      "learning_rate": 2.0996423041829693e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37785872,
      "step": 179045
    },
    {
      "epoch": 19.697469746974697,
      "grad_norm": 0.0016632080078125,
      "learning_rate": 2.0920325438689157e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37786896,
      "step": 179050
    },
    {
      "epoch": 19.698019801980198,
      "grad_norm": 0.005584716796875,
      "learning_rate": 2.0844365890536066e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37787984,
      "step": 179055
    },
    {
      "epoch": 19.6985698569857,
      "grad_norm": 0.005645751953125,
      "learning_rate": 2.0768544398073186e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37789104,
      "step": 179060
    },
    {
      "epoch": 19.6991199119912,
      "grad_norm": 0.00103759765625,
      "learning_rate": 2.0692860961999957e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37790128,
      "step": 179065
    },
    {
      "epoch": 19.6996699669967,
      "grad_norm": 0.005615234375,
      "learning_rate": 2.0617315583012497e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37791120,
      "step": 179070
    },
    {
      "epoch": 19.7002200220022,
      "grad_norm": 0.0006866455078125,
      "learning_rate": 2.0541908261806906e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37792176,
      "step": 179075
    },
    {
      "epoch": 19.7007700770077,
      "grad_norm": 0.01123046875,
      "learning_rate": 2.04666389990793e-05,
      "loss": 0.2345,
      "num_input_tokens_seen": 37793328,
      "step": 179080
    },
    {
      "epoch": 19.7013201320132,
      "grad_norm": 0.00189208984375,
      "learning_rate": 2.0391507795522457e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37794384,
      "step": 179085
    },
    {
      "epoch": 19.701870187018702,
      "grad_norm": 0.005462646484375,
      "learning_rate": 2.0316514651829152e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37795440,
      "step": 179090
    },
    {
      "epoch": 19.702420242024203,
      "grad_norm": 0.0018310546875,
      "learning_rate": 2.0241659568690505e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37796528,
      "step": 179095
    },
    {
      "epoch": 19.702970297029704,
      "grad_norm": 0.00555419921875,
      "learning_rate": 2.0166942546795963e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37797616,
      "step": 179100
    },
    {
      "epoch": 19.703520352035202,
      "grad_norm": 0.005462646484375,
      "learning_rate": 2.0092363586836636e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37798672,
      "step": 179105
    },
    {
      "epoch": 19.704070407040703,
      "grad_norm": 0.0111083984375,
      "learning_rate": 2.0017922689495316e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37799696,
      "step": 179110
    },
    {
      "epoch": 19.704620462046204,
      "grad_norm": 0.005859375,
      "learning_rate": 1.9943619855463112e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37800720,
      "step": 179115
    },
    {
      "epoch": 19.705170517051705,
      "grad_norm": 0.005828857421875,
      "learning_rate": 1.986945508542115e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37801776,
      "step": 179120
    },
    {
      "epoch": 19.705720572057206,
      "grad_norm": 0.005584716796875,
      "learning_rate": 1.9795428380053882e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37802800,
      "step": 179125
    },
    {
      "epoch": 19.706270627062707,
      "grad_norm": 0.00555419921875,
      "learning_rate": 1.9721539740045755e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37803824,
      "step": 179130
    },
    {
      "epoch": 19.706820682068205,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 1.964778916607457e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37804944,
      "step": 179135
    },
    {
      "epoch": 19.707370737073706,
      "grad_norm": 0.01123046875,
      "learning_rate": 1.957417665882144e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37806000,
      "step": 179140
    },
    {
      "epoch": 19.707920792079207,
      "grad_norm": 0.002197265625,
      "learning_rate": 1.950070221896416e-05,
      "loss": 0.2313,
      "num_input_tokens_seen": 37807024,
      "step": 179145
    },
    {
      "epoch": 19.70847084708471,
      "grad_norm": 0.00537109375,
      "learning_rate": 1.942736584718219e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37808080,
      "step": 179150
    },
    {
      "epoch": 19.70902090209021,
      "grad_norm": 0.005462646484375,
      "learning_rate": 1.9354167544148316e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37809104,
      "step": 179155
    },
    {
      "epoch": 19.70957095709571,
      "grad_norm": 0.006317138671875,
      "learning_rate": 1.9281107310540334e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37810224,
      "step": 179160
    },
    {
      "epoch": 19.71012101210121,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 1.920818514702771e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37811312,
      "step": 179165
    },
    {
      "epoch": 19.71067106710671,
      "grad_norm": 0.005462646484375,
      "learning_rate": 1.913540105428657e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37812336,
      "step": 179170
    },
    {
      "epoch": 19.71122112211221,
      "grad_norm": 0.005645751953125,
      "learning_rate": 1.9062755032984713e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37813296,
      "step": 179175
    },
    {
      "epoch": 19.71177117711771,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 1.899024708379493e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37814320,
      "step": 179180
    },
    {
      "epoch": 19.712321232123212,
      "grad_norm": 0.00567626953125,
      "learning_rate": 1.8917877207381693e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37815312,
      "step": 179185
    },
    {
      "epoch": 19.712871287128714,
      "grad_norm": 0.00579833984375,
      "learning_rate": 1.884564540441447e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37816336,
      "step": 179190
    },
    {
      "epoch": 19.713421342134215,
      "grad_norm": 0.0027008056640625,
      "learning_rate": 1.877355167555772e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37817456,
      "step": 179195
    },
    {
      "epoch": 19.713971397139716,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 1.8701596021475917e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37818512,
      "step": 179200
    },
    {
      "epoch": 19.714521452145213,
      "grad_norm": 0.01123046875,
      "learning_rate": 1.8629778442833533e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37819568,
      "step": 179205
    },
    {
      "epoch": 19.715071507150714,
      "grad_norm": 0.005615234375,
      "learning_rate": 1.85580989402917e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37820624,
      "step": 179210
    },
    {
      "epoch": 19.715621562156215,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 1.848655751451156e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37821616,
      "step": 179215
    },
    {
      "epoch": 19.716171617161717,
      "grad_norm": 0.005706787109375,
      "learning_rate": 1.8415154166150914e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37822672,
      "step": 179220
    },
    {
      "epoch": 19.716721672167218,
      "grad_norm": 0.00141143798828125,
      "learning_rate": 1.834388889586924e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37823696,
      "step": 179225
    },
    {
      "epoch": 19.71727172717272,
      "grad_norm": 0.01092529296875,
      "learning_rate": 1.8272761704324347e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37824784,
      "step": 179230
    },
    {
      "epoch": 19.717821782178216,
      "grad_norm": 0.006103515625,
      "learning_rate": 1.820177259216904e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37825904,
      "step": 179235
    },
    {
      "epoch": 19.718371837183717,
      "grad_norm": 0.005706787109375,
      "learning_rate": 1.8130921560059466e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37826928,
      "step": 179240
    },
    {
      "epoch": 19.71892189218922,
      "grad_norm": 0.0111083984375,
      "learning_rate": 1.8060208608648432e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37827952,
      "step": 179245
    },
    {
      "epoch": 19.71947194719472,
      "grad_norm": 0.005828857421875,
      "learning_rate": 1.7989633738587083e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37829008,
      "step": 179250
    },
    {
      "epoch": 19.72002200220022,
      "grad_norm": 0.001373291015625,
      "learning_rate": 1.7919196950526572e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37830064,
      "step": 179255
    },
    {
      "epoch": 19.72057205720572,
      "grad_norm": 0.0024261474609375,
      "learning_rate": 1.784889824511471e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37831088,
      "step": 179260
    },
    {
      "epoch": 19.721122112211223,
      "grad_norm": 0.00604248046875,
      "learning_rate": 1.7778737623000974e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37832112,
      "step": 179265
    },
    {
      "epoch": 19.72167216721672,
      "grad_norm": 0.006378173828125,
      "learning_rate": 1.7708715084831515e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37833200,
      "step": 179270
    },
    {
      "epoch": 19.72222222222222,
      "grad_norm": 0.005645751953125,
      "learning_rate": 1.763883063125249e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37834192,
      "step": 179275
    },
    {
      "epoch": 19.722772277227723,
      "grad_norm": 0.0108642578125,
      "learning_rate": 1.7569084262906708e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37835248,
      "step": 179280
    },
    {
      "epoch": 19.723322332233224,
      "grad_norm": 0.005767822265625,
      "learning_rate": 1.7499475980436997e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37836240,
      "step": 179285
    },
    {
      "epoch": 19.723872387238725,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 1.7430005784484502e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37837296,
      "step": 179290
    },
    {
      "epoch": 19.724422442244226,
      "grad_norm": 0.006256103515625,
      "learning_rate": 1.736067367569205e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37838320,
      "step": 179295
    },
    {
      "epoch": 19.724972497249723,
      "grad_norm": 0.0057373046875,
      "learning_rate": 1.7291479654695795e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37839376,
      "step": 179300
    },
    {
      "epoch": 19.725522552255224,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 1.7222423722135228e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37840400,
      "step": 179305
    },
    {
      "epoch": 19.726072607260726,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 1.7153505878644835e-05,
      "loss": 0.233,
      "num_input_tokens_seen": 37841552,
      "step": 179310
    },
    {
      "epoch": 19.726622662266227,
      "grad_norm": 0.00555419921875,
      "learning_rate": 1.708472612486245e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37842608,
      "step": 179315
    },
    {
      "epoch": 19.727172717271728,
      "grad_norm": 0.0054931640625,
      "learning_rate": 1.7016084461420887e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37843600,
      "step": 179320
    },
    {
      "epoch": 19.72772277227723,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 1.694758088895132e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37844688,
      "step": 179325
    },
    {
      "epoch": 19.72827282728273,
      "grad_norm": 0.005889892578125,
      "learning_rate": 1.6879215408086566e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37845680,
      "step": 179330
    },
    {
      "epoch": 19.728822882288227,
      "grad_norm": 0.00174713134765625,
      "learning_rate": 1.6810988019457796e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37846768,
      "step": 179335
    },
    {
      "epoch": 19.72937293729373,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 1.6742898723691167e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37847760,
      "step": 179340
    },
    {
      "epoch": 19.72992299229923,
      "grad_norm": 0.005401611328125,
      "learning_rate": 1.6674947521416183e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37848848,
      "step": 179345
    },
    {
      "epoch": 19.73047304730473,
      "grad_norm": 0.00543212890625,
      "learning_rate": 1.660713441325734e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37849904,
      "step": 179350
    },
    {
      "epoch": 19.731023102310232,
      "grad_norm": 0.005584716796875,
      "learning_rate": 1.653945939984247e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37850960,
      "step": 179355
    },
    {
      "epoch": 19.731573157315733,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 1.6471922481792744e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37852080,
      "step": 179360
    },
    {
      "epoch": 19.73212321232123,
      "grad_norm": 0.00133514404296875,
      "learning_rate": 1.6404523659730996e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37853136,
      "step": 179365
    },
    {
      "epoch": 19.73267326732673,
      "grad_norm": 0.005859375,
      "learning_rate": 1.633726293428006e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37854192,
      "step": 179370
    },
    {
      "epoch": 19.733223322332233,
      "grad_norm": 0.0057373046875,
      "learning_rate": 1.627014030605778e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37855280,
      "step": 179375
    },
    {
      "epoch": 19.733773377337734,
      "grad_norm": 0.005767822265625,
      "learning_rate": 1.6203155775683653e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37856368,
      "step": 179380
    },
    {
      "epoch": 19.734323432343235,
      "grad_norm": 0.01055908203125,
      "learning_rate": 1.6136309343775523e-05,
      "loss": 0.2283,
      "num_input_tokens_seen": 37857456,
      "step": 179385
    },
    {
      "epoch": 19.734873487348736,
      "grad_norm": 0.00176239013671875,
      "learning_rate": 1.6069601010949563e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37858480,
      "step": 179390
    },
    {
      "epoch": 19.735423542354237,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 1.600303077782028e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37859472,
      "step": 179395
    },
    {
      "epoch": 19.735973597359735,
      "grad_norm": 0.00244140625,
      "learning_rate": 1.593659864500052e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37860560,
      "step": 179400
    },
    {
      "epoch": 19.736523652365236,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 1.5870304613103125e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37861680,
      "step": 179405
    },
    {
      "epoch": 19.737073707370737,
      "grad_norm": 0.001678466796875,
      "learning_rate": 1.580414868273927e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37862704,
      "step": 179410
    },
    {
      "epoch": 19.737623762376238,
      "grad_norm": 0.005523681640625,
      "learning_rate": 1.5738130854516808e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37863728,
      "step": 179415
    },
    {
      "epoch": 19.73817381738174,
      "grad_norm": 0.000934600830078125,
      "learning_rate": 1.5672251129048575e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37864816,
      "step": 179420
    },
    {
      "epoch": 19.73872387238724,
      "grad_norm": 0.000621795654296875,
      "learning_rate": 1.5606509506937427e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37865872,
      "step": 179425
    },
    {
      "epoch": 19.739273927392738,
      "grad_norm": 0.01116943359375,
      "learning_rate": 1.5540905988791208e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37866960,
      "step": 179430
    },
    {
      "epoch": 19.73982398239824,
      "grad_norm": 0.005767822265625,
      "learning_rate": 1.5475440575212773e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37867952,
      "step": 179435
    },
    {
      "epoch": 19.74037403740374,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 1.54101132668083e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37869040,
      "step": 179440
    },
    {
      "epoch": 19.74092409240924,
      "grad_norm": 0.0054931640625,
      "learning_rate": 1.5344924064178977e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37870064,
      "step": 179445
    },
    {
      "epoch": 19.741474147414742,
      "grad_norm": 0.001495361328125,
      "learning_rate": 1.5279872967924325e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37871056,
      "step": 179450
    },
    {
      "epoch": 19.742024202420243,
      "grad_norm": 0.005706787109375,
      "learning_rate": 1.5214959978645525e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37872048,
      "step": 179455
    },
    {
      "epoch": 19.742574257425744,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 1.5150185096940437e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37873136,
      "step": 179460
    },
    {
      "epoch": 19.74312431243124,
      "grad_norm": 0.0054931640625,
      "learning_rate": 1.5085548323403586e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37874256,
      "step": 179465
    },
    {
      "epoch": 19.743674367436743,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 1.5021049658634488e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37875344,
      "step": 179470
    },
    {
      "epoch": 19.744224422442244,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 1.4956689103224341e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37876432,
      "step": 179475
    },
    {
      "epoch": 19.744774477447745,
      "grad_norm": 0.005584716796875,
      "learning_rate": 1.4892466657769331e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37877520,
      "step": 179480
    },
    {
      "epoch": 19.745324532453246,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 1.482838232285899e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37878544,
      "step": 179485
    },
    {
      "epoch": 19.745874587458747,
      "grad_norm": 0.00567626953125,
      "learning_rate": 1.4764436099086176e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37879536,
      "step": 179490
    },
    {
      "epoch": 19.746424642464248,
      "grad_norm": 0.006011962890625,
      "learning_rate": 1.470062798703875e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37880592,
      "step": 179495
    },
    {
      "epoch": 19.746974697469746,
      "grad_norm": 0.002044677734375,
      "learning_rate": 1.463695798730291e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37881616,
      "step": 179500
    },
    {
      "epoch": 19.747524752475247,
      "grad_norm": 0.005859375,
      "learning_rate": 1.4573426100469854e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37882704,
      "step": 179505
    },
    {
      "epoch": 19.748074807480748,
      "grad_norm": 0.0111083984375,
      "learning_rate": 1.4510032327122446e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37883824,
      "step": 179510
    },
    {
      "epoch": 19.74862486248625,
      "grad_norm": 0.0054931640625,
      "learning_rate": 1.4446776667843552e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37884816,
      "step": 179515
    },
    {
      "epoch": 19.74917491749175,
      "grad_norm": 0.00128173828125,
      "learning_rate": 1.4383659123219372e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37885872,
      "step": 179520
    },
    {
      "epoch": 19.74972497249725,
      "grad_norm": 0.00543212890625,
      "learning_rate": 1.4320679693831105e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37886928,
      "step": 179525
    },
    {
      "epoch": 19.75027502750275,
      "grad_norm": 0.005767822265625,
      "learning_rate": 1.4257838380256626e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37888016,
      "step": 179530
    },
    {
      "epoch": 19.75082508250825,
      "grad_norm": 0.005706787109375,
      "learning_rate": 1.4195135183077133e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37889104,
      "step": 179535
    },
    {
      "epoch": 19.75137513751375,
      "grad_norm": 0.0108642578125,
      "learning_rate": 1.41325701028705e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37890160,
      "step": 179540
    },
    {
      "epoch": 19.751925192519252,
      "grad_norm": 0.005645751953125,
      "learning_rate": 1.4070143140212932e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37891216,
      "step": 179545
    },
    {
      "epoch": 19.752475247524753,
      "grad_norm": 0.002105712890625,
      "learning_rate": 1.4007854295680633e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37892272,
      "step": 179550
    },
    {
      "epoch": 19.753025302530254,
      "grad_norm": 0.005462646484375,
      "learning_rate": 1.3945703569846478e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37893328,
      "step": 179555
    },
    {
      "epoch": 19.753575357535752,
      "grad_norm": 0.0015716552734375,
      "learning_rate": 1.3883690963283345e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37894416,
      "step": 179560
    },
    {
      "epoch": 19.754125412541253,
      "grad_norm": 0.00604248046875,
      "learning_rate": 1.3821816476562443e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37895440,
      "step": 179565
    },
    {
      "epoch": 19.754675467546754,
      "grad_norm": 0.0004730224609375,
      "learning_rate": 1.3760080110256645e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37896560,
      "step": 179570
    },
    {
      "epoch": 19.755225522552255,
      "grad_norm": 0.0054931640625,
      "learning_rate": 1.3698481864932165e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37897616,
      "step": 179575
    },
    {
      "epoch": 19.755775577557756,
      "grad_norm": 0.005523681640625,
      "learning_rate": 1.363702174115855e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37898672,
      "step": 179580
    },
    {
      "epoch": 19.756325632563257,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 1.3575699739500346e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37899696,
      "step": 179585
    },
    {
      "epoch": 19.75687568756876,
      "grad_norm": 0.0057373046875,
      "learning_rate": 1.3514515860523768e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37900688,
      "step": 179590
    },
    {
      "epoch": 19.757425742574256,
      "grad_norm": 0.0057373046875,
      "learning_rate": 1.3453470104793363e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37901776,
      "step": 179595
    },
    {
      "epoch": 19.757975797579757,
      "grad_norm": 0.005889892578125,
      "learning_rate": 1.339256247287035e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37902800,
      "step": 179600
    },
    {
      "epoch": 19.758525852585258,
      "grad_norm": 0.005706787109375,
      "learning_rate": 1.3331792965315946e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37903824,
      "step": 179605
    },
    {
      "epoch": 19.75907590759076,
      "grad_norm": 0.001373291015625,
      "learning_rate": 1.3271161582693035e-05,
      "loss": 0.2345,
      "num_input_tokens_seen": 37904848,
      "step": 179610
    },
    {
      "epoch": 19.75962596259626,
      "grad_norm": 0.005462646484375,
      "learning_rate": 1.321066832555784e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37905968,
      "step": 179615
    },
    {
      "epoch": 19.76017601760176,
      "grad_norm": 0.0023040771484375,
      "learning_rate": 1.3150313194468244e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37907088,
      "step": 179620
    },
    {
      "epoch": 19.760726072607262,
      "grad_norm": 0.00567626953125,
      "learning_rate": 1.309009618998047e-05,
      "loss": 0.2304,
      "num_input_tokens_seen": 37908208,
      "step": 179625
    },
    {
      "epoch": 19.76127612761276,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 1.3030017312650742e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37909328,
      "step": 179630
    },
    {
      "epoch": 19.76182618261826,
      "grad_norm": 0.005523681640625,
      "learning_rate": 1.2970076563030285e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37910320,
      "step": 179635
    },
    {
      "epoch": 19.762376237623762,
      "grad_norm": 0.005615234375,
      "learning_rate": 1.291027394167532e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37911440,
      "step": 179640
    },
    {
      "epoch": 19.762926292629263,
      "grad_norm": 0.00555419921875,
      "learning_rate": 1.2850609449133743e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37912464,
      "step": 179645
    },
    {
      "epoch": 19.763476347634764,
      "grad_norm": 0.005523681640625,
      "learning_rate": 1.2791083085956777e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 37913488,
      "step": 179650
    },
    {
      "epoch": 19.764026402640265,
      "grad_norm": 0.01068115234375,
      "learning_rate": 1.273169485269232e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37914576,
      "step": 179655
    },
    {
      "epoch": 19.764576457645763,
      "grad_norm": 0.005584716796875,
      "learning_rate": 1.2672444749889933e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37915696,
      "step": 179660
    },
    {
      "epoch": 19.765126512651264,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 1.2613332778092511e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37916784,
      "step": 179665
    },
    {
      "epoch": 19.765676567656765,
      "grad_norm": 0.01092529296875,
      "learning_rate": 1.2554358937846288e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37917808,
      "step": 179670
    },
    {
      "epoch": 19.766226622662266,
      "grad_norm": 0.005584716796875,
      "learning_rate": 1.249552322969416e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37918832,
      "step": 179675
    },
    {
      "epoch": 19.766776677667767,
      "grad_norm": 0.01116943359375,
      "learning_rate": 1.2436825654180693e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37919792,
      "step": 179680
    },
    {
      "epoch": 19.76732673267327,
      "grad_norm": 0.01123046875,
      "learning_rate": 1.2378266211845456e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37920816,
      "step": 179685
    },
    {
      "epoch": 19.76787678767877,
      "grad_norm": 0.00154876708984375,
      "learning_rate": 1.231984490322635e-05,
      "loss": 0.2283,
      "num_input_tokens_seen": 37921840,
      "step": 179690
    },
    {
      "epoch": 19.768426842684267,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 1.2261561728864612e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37922864,
      "step": 179695
    },
    {
      "epoch": 19.768976897689768,
      "grad_norm": 0.005523681640625,
      "learning_rate": 1.2203416689296476e-05,
      "loss": 0.233,
      "num_input_tokens_seen": 37923824,
      "step": 179700
    },
    {
      "epoch": 19.76952695269527,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 1.2145409785056516e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37924944,
      "step": 179705
    },
    {
      "epoch": 19.77007700770077,
      "grad_norm": 0.00077056884765625,
      "learning_rate": 1.2087541016680968e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37925968,
      "step": 179710
    },
    {
      "epoch": 19.77062706270627,
      "grad_norm": 0.000499725341796875,
      "learning_rate": 1.2029810384702743e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37926992,
      "step": 179715
    },
    {
      "epoch": 19.771177117711773,
      "grad_norm": 0.000896453857421875,
      "learning_rate": 1.197221788965308e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37928080,
      "step": 179720
    },
    {
      "epoch": 19.77172717271727,
      "grad_norm": 0.01092529296875,
      "learning_rate": 1.1914763532064886e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37929104,
      "step": 179725
    },
    {
      "epoch": 19.77227722772277,
      "grad_norm": 0.01092529296875,
      "learning_rate": 1.1857447312466073e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 37930128,
      "step": 179730
    },
    {
      "epoch": 19.772827282728272,
      "grad_norm": 0.005615234375,
      "learning_rate": 1.1800269231384552e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37931152,
      "step": 179735
    },
    {
      "epoch": 19.773377337733773,
      "grad_norm": 0.00567626953125,
      "learning_rate": 1.1743229289348233e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37932304,
      "step": 179740
    },
    {
      "epoch": 19.773927392739274,
      "grad_norm": 0.00628662109375,
      "learning_rate": 1.1686327486881698e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37933360,
      "step": 179745
    },
    {
      "epoch": 19.774477447744776,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 1.162956382451119e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37934416,
      "step": 179750
    },
    {
      "epoch": 19.775027502750277,
      "grad_norm": 0.005706787109375,
      "learning_rate": 1.1572938302757962e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37935504,
      "step": 179755
    },
    {
      "epoch": 19.775577557755774,
      "grad_norm": 0.005767822265625,
      "learning_rate": 1.1516450922146593e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37936528,
      "step": 179760
    },
    {
      "epoch": 19.776127612761275,
      "grad_norm": 0.005706787109375,
      "learning_rate": 1.1460101683193334e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37937584,
      "step": 179765
    },
    {
      "epoch": 19.776677667766776,
      "grad_norm": 0.005462646484375,
      "learning_rate": 1.1403890586422772e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37938672,
      "step": 179770
    },
    {
      "epoch": 19.777227722772277,
      "grad_norm": 0.00543212890625,
      "learning_rate": 1.1347817632347822e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37939760,
      "step": 179775
    },
    {
      "epoch": 19.77777777777778,
      "grad_norm": 0.001312255859375,
      "learning_rate": 1.129188282148974e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37940848,
      "step": 179780
    },
    {
      "epoch": 19.77832783278328,
      "grad_norm": 0.00128173828125,
      "learning_rate": 1.1236086154361445e-05,
      "loss": 0.2299,
      "num_input_tokens_seen": 37941808,
      "step": 179785
    },
    {
      "epoch": 19.778877887788777,
      "grad_norm": 0.005523681640625,
      "learning_rate": 1.1180427631477529e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37942864,
      "step": 179790
    },
    {
      "epoch": 19.77942794279428,
      "grad_norm": 0.005523681640625,
      "learning_rate": 1.1124907253350912e-05,
      "loss": 0.2298,
      "num_input_tokens_seen": 37943888,
      "step": 179795
    },
    {
      "epoch": 19.77997799779978,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 1.106952502049452e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37944944,
      "step": 179800
    },
    {
      "epoch": 19.78052805280528,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 1.101428093341794e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37946032,
      "step": 179805
    },
    {
      "epoch": 19.78107810781078,
      "grad_norm": 0.00592041015625,
      "learning_rate": 1.0959174992630771e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 37947088,
      "step": 179810
    },
    {
      "epoch": 19.781628162816283,
      "grad_norm": 0.010986328125,
      "learning_rate": 1.090420719863927e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37948112,
      "step": 179815
    },
    {
      "epoch": 19.782178217821784,
      "grad_norm": 0.005889892578125,
      "learning_rate": 1.0849377551951366e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37949232,
      "step": 179820
    },
    {
      "epoch": 19.78272827282728,
      "grad_norm": 0.0108642578125,
      "learning_rate": 1.0794686053073321e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37950320,
      "step": 179825
    },
    {
      "epoch": 19.783278327832782,
      "grad_norm": 0.005706787109375,
      "learning_rate": 1.0740132702506399e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37951344,
      "step": 179830
    },
    {
      "epoch": 19.783828382838283,
      "grad_norm": 0.001251220703125,
      "learning_rate": 1.0685717500756864e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 37952432,
      "step": 179835
    },
    {
      "epoch": 19.784378437843785,
      "grad_norm": 0.005615234375,
      "learning_rate": 1.0631440448322649e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37953456,
      "step": 179840
    },
    {
      "epoch": 19.784928492849286,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 1.0577301545708351e-05,
      "loss": 0.2288,
      "num_input_tokens_seen": 37954448,
      "step": 179845
    },
    {
      "epoch": 19.785478547854787,
      "grad_norm": 0.005340576171875,
      "learning_rate": 1.0523300793408573e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 37955472,
      "step": 179850
    },
    {
      "epoch": 19.786028602860284,
      "grad_norm": 0.010986328125,
      "learning_rate": 1.046943819192292e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 37956528,
      "step": 179855
    },
    {
      "epoch": 19.786578657865785,
      "grad_norm": 0.002349853515625,
      "learning_rate": 1.0415713741749322e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37957520,
      "step": 179860
    },
    {
      "epoch": 19.787128712871286,
      "grad_norm": 0.00555419921875,
      "learning_rate": 1.0362127443379055e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37958544,
      "step": 179865
    },
    {
      "epoch": 19.787678767876788,
      "grad_norm": 0.00592041015625,
      "learning_rate": 1.0308679297310053e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 37959568,
      "step": 179870
    },
    {
      "epoch": 19.78822882288229,
      "grad_norm": 0.005767822265625,
      "learning_rate": 1.0255369304031924e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37960624,
      "step": 179875
    },
    {
      "epoch": 19.78877887788779,
      "grad_norm": 0.00579833984375,
      "learning_rate": 1.0202197464039276e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 37961744,
      "step": 179880
    },
    {
      "epoch": 19.78932893289329,
      "grad_norm": 0.00592041015625,
      "learning_rate": 1.0149163777818382e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37962832,
      "step": 179885
    },
    {
      "epoch": 19.78987898789879,
      "grad_norm": 0.0057373046875,
      "learning_rate": 1.0096268245860518e-05,
      "loss": 0.2334,
      "num_input_tokens_seen": 37963856,
      "step": 179890
    },
    {
      "epoch": 19.79042904290429,
      "grad_norm": 0.00555419921875,
      "learning_rate": 1.0043510868651961e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 37964880,
      "step": 179895
    },
    {
      "epoch": 19.79097909790979,
      "grad_norm": 0.00579833984375,
      "learning_rate": 9.990891646680654e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 37965904,
      "step": 179900
    },
    {
      "epoch": 19.79152915291529,
      "grad_norm": 0.000865936279296875,
      "learning_rate": 9.938410580429546e-06,
      "loss": 0.2288,
      "num_input_tokens_seen": 37966928,
      "step": 179905
    },
    {
      "epoch": 19.792079207920793,
      "grad_norm": 0.0054931640625,
      "learning_rate": 9.886067670383247e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 37968016,
      "step": 179910
    },
    {
      "epoch": 19.792629262926294,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 9.83386291702304e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 37969072,
      "step": 179915
    },
    {
      "epoch": 19.793179317931795,
      "grad_norm": 0.005615234375,
      "learning_rate": 9.781796320833535e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 37970128,
      "step": 179920
    },
    {
      "epoch": 19.793729372937293,
      "grad_norm": 0.005615234375,
      "learning_rate": 9.729867882289355e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 37971184,
      "step": 179925
    },
    {
      "epoch": 19.794279427942794,
      "grad_norm": 0.005401611328125,
      "learning_rate": 9.678077601873447e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 37972208,
      "step": 179930
    },
    {
      "epoch": 19.794829482948295,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 9.626425480062094e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 37973296,
      "step": 179935
    },
    {
      "epoch": 19.795379537953796,
      "grad_norm": 0.00555419921875,
      "learning_rate": 9.574911517331585e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 37974320,
      "step": 179940
    },
    {
      "epoch": 19.795929592959297,
      "grad_norm": 0.00555419921875,
      "learning_rate": 9.523535714154874e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 37975408,
      "step": 179945
    },
    {
      "epoch": 19.796479647964798,
      "grad_norm": 0.00579833984375,
      "learning_rate": 9.472298071006579e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 37976464,
      "step": 179950
    },
    {
      "epoch": 19.797029702970296,
      "grad_norm": 0.005584716796875,
      "learning_rate": 9.421198588357993e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 37977520,
      "step": 179955
    },
    {
      "epoch": 19.797579757975797,
      "grad_norm": 0.005584716796875,
      "learning_rate": 9.370237266682069e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 37978576,
      "step": 179960
    },
    {
      "epoch": 19.798129812981298,
      "grad_norm": 0.006011962890625,
      "learning_rate": 9.31941410644843e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 37979664,
      "step": 179965
    },
    {
      "epoch": 19.7986798679868,
      "grad_norm": 0.005767822265625,
      "learning_rate": 9.268729108123374e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 37980784,
      "step": 179970
    },
    {
      "epoch": 19.7992299229923,
      "grad_norm": 0.00579833984375,
      "learning_rate": 9.218182272174857e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 37981904,
      "step": 179975
    },
    {
      "epoch": 19.7997799779978,
      "grad_norm": 0.005645751953125,
      "learning_rate": 9.167773599069173e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 37982960,
      "step": 179980
    },
    {
      "epoch": 19.8003300330033,
      "grad_norm": 0.000621795654296875,
      "learning_rate": 9.11750308927095e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 37983952,
      "step": 179985
    },
    {
      "epoch": 19.8008800880088,
      "grad_norm": 0.0057373046875,
      "learning_rate": 9.067370743244818e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 37985040,
      "step": 179990
    },
    {
      "epoch": 19.8014301430143,
      "grad_norm": 0.010986328125,
      "learning_rate": 9.017376561448742e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 37986096,
      "step": 179995
    },
    {
      "epoch": 19.801980198019802,
      "grad_norm": 0.0052490234375,
      "learning_rate": 8.967520544349017e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 37987056,
      "step": 180000
    },
    {
      "epoch": 19.802530253025303,
      "grad_norm": 0.00145721435546875,
      "learning_rate": 8.917802692400278e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 37988080,
      "step": 180005
    },
    {
      "epoch": 19.803080308030804,
      "grad_norm": 0.00125885009765625,
      "learning_rate": 8.868223006063825e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 37989168,
      "step": 180010
    },
    {
      "epoch": 19.803630363036305,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 8.81878148579429e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 37990192,
      "step": 180015
    },
    {
      "epoch": 19.804180418041803,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 8.769478132049646e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 37991248,
      "step": 180020
    },
    {
      "epoch": 19.804730473047304,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 8.72031294528286e-06,
      "loss": 0.2304,
      "num_input_tokens_seen": 37992336,
      "step": 180025
    },
    {
      "epoch": 19.805280528052805,
      "grad_norm": 0.005645751953125,
      "learning_rate": 8.671285925946902e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 37993424,
      "step": 180030
    },
    {
      "epoch": 19.805830583058306,
      "grad_norm": 0.00616455078125,
      "learning_rate": 8.622397074494748e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 37994480,
      "step": 180035
    },
    {
      "epoch": 19.806380638063807,
      "grad_norm": 0.0054931640625,
      "learning_rate": 8.573646391376033e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 37995536,
      "step": 180040
    },
    {
      "epoch": 19.806930693069308,
      "grad_norm": 0.002288818359375,
      "learning_rate": 8.5250338770404e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 37996560,
      "step": 180045
    },
    {
      "epoch": 19.80748074807481,
      "grad_norm": 0.00567626953125,
      "learning_rate": 8.476559531937488e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 37997552,
      "step": 180050
    },
    {
      "epoch": 19.808030803080307,
      "grad_norm": 0.0020751953125,
      "learning_rate": 8.428223356510279e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 37998608,
      "step": 180055
    },
    {
      "epoch": 19.808580858085808,
      "grad_norm": 0.00160980224609375,
      "learning_rate": 8.380025351208408e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 37999696,
      "step": 180060
    },
    {
      "epoch": 19.80913091309131,
      "grad_norm": 0.005767822265625,
      "learning_rate": 8.331965516473193e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 38000752,
      "step": 180065
    },
    {
      "epoch": 19.80968096809681,
      "grad_norm": 0.00159454345703125,
      "learning_rate": 8.284043852747613e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38001840,
      "step": 180070
    },
    {
      "epoch": 19.81023102310231,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 8.23626036047631e-06,
      "loss": 0.2335,
      "num_input_tokens_seen": 38002832,
      "step": 180075
    },
    {
      "epoch": 19.810781078107812,
      "grad_norm": 0.00543212890625,
      "learning_rate": 8.188615040095603e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38003920,
      "step": 180080
    },
    {
      "epoch": 19.81133113311331,
      "grad_norm": 0.006439208984375,
      "learning_rate": 8.14110789204847e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38005008,
      "step": 180085
    },
    {
      "epoch": 19.81188118811881,
      "grad_norm": 0.00153350830078125,
      "learning_rate": 8.093738916769565e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38006000,
      "step": 180090
    },
    {
      "epoch": 19.812431243124312,
      "grad_norm": 0.0004215240478515625,
      "learning_rate": 8.046508114698536e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38007152,
      "step": 180095
    },
    {
      "epoch": 19.812981298129813,
      "grad_norm": 0.005706787109375,
      "learning_rate": 7.999415486266703e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38008240,
      "step": 180100
    },
    {
      "epoch": 19.813531353135314,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 7.95246103191205e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 38009296,
      "step": 180105
    },
    {
      "epoch": 19.814081408140815,
      "grad_norm": 0.01104736328125,
      "learning_rate": 7.905644752064233e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38010320,
      "step": 180110
    },
    {
      "epoch": 19.814631463146316,
      "grad_norm": 0.005584716796875,
      "learning_rate": 7.858966647157906e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38011408,
      "step": 180115
    },
    {
      "epoch": 19.815181518151814,
      "grad_norm": 0.00193023681640625,
      "learning_rate": 7.812426717621057e-06,
      "loss": 0.2335,
      "num_input_tokens_seen": 38012496,
      "step": 180120
    },
    {
      "epoch": 19.815731573157315,
      "grad_norm": 0.00086212158203125,
      "learning_rate": 7.766024963883343e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38013616,
      "step": 180125
    },
    {
      "epoch": 19.816281628162816,
      "grad_norm": 0.00113677978515625,
      "learning_rate": 7.719761386371092e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38014640,
      "step": 180130
    },
    {
      "epoch": 19.816831683168317,
      "grad_norm": 0.005645751953125,
      "learning_rate": 7.673635985512295e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38015728,
      "step": 180135
    },
    {
      "epoch": 19.817381738173818,
      "grad_norm": 0.01080322265625,
      "learning_rate": 7.627648761731609e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38016784,
      "step": 180140
    },
    {
      "epoch": 19.81793179317932,
      "grad_norm": 0.0054931640625,
      "learning_rate": 7.581799715453696e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38017808,
      "step": 180145
    },
    {
      "epoch": 19.818481848184817,
      "grad_norm": 0.001953125,
      "learning_rate": 7.536088847098221e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38018864,
      "step": 180150
    },
    {
      "epoch": 19.819031903190318,
      "grad_norm": 0.0057373046875,
      "learning_rate": 7.490516157091509e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38019952,
      "step": 180155
    },
    {
      "epoch": 19.81958195819582,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 7.4450816458482284e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38021040,
      "step": 180160
    },
    {
      "epoch": 19.82013201320132,
      "grad_norm": 0.005950927734375,
      "learning_rate": 7.3997853137897086e-06,
      "loss": 0.2329,
      "num_input_tokens_seen": 38022128,
      "step": 180165
    },
    {
      "epoch": 19.82068206820682,
      "grad_norm": 0.01104736328125,
      "learning_rate": 7.354627161333948e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38023184,
      "step": 180170
    },
    {
      "epoch": 19.821232123212322,
      "grad_norm": 0.005645751953125,
      "learning_rate": 7.309607188895617e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38024208,
      "step": 180175
    },
    {
      "epoch": 19.821782178217823,
      "grad_norm": 0.0054931640625,
      "learning_rate": 7.264725396891047e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38025264,
      "step": 180180
    },
    {
      "epoch": 19.82233223322332,
      "grad_norm": 0.0016937255859375,
      "learning_rate": 7.219981785733242e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38026256,
      "step": 180185
    },
    {
      "epoch": 19.822882288228822,
      "grad_norm": 0.001312255859375,
      "learning_rate": 7.175376355835206e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38027312,
      "step": 180190
    },
    {
      "epoch": 19.823432343234323,
      "grad_norm": 0.0004596710205078125,
      "learning_rate": 7.1309091076066085e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38028304,
      "step": 180195
    },
    {
      "epoch": 19.823982398239824,
      "grad_norm": 0.0111083984375,
      "learning_rate": 7.0865800414587894e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38029392,
      "step": 180200
    },
    {
      "epoch": 19.824532453245325,
      "grad_norm": 0.005859375,
      "learning_rate": 7.042389157799755e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38030448,
      "step": 180205
    },
    {
      "epoch": 19.825082508250826,
      "grad_norm": 0.0054931640625,
      "learning_rate": 6.998336457035847e-06,
      "loss": 0.2329,
      "num_input_tokens_seen": 38031472,
      "step": 180210
    },
    {
      "epoch": 19.825632563256324,
      "grad_norm": 0.005584716796875,
      "learning_rate": 6.954421939575073e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38032560,
      "step": 180215
    },
    {
      "epoch": 19.826182618261825,
      "grad_norm": 0.005615234375,
      "learning_rate": 6.9106456058187765e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38033616,
      "step": 180220
    },
    {
      "epoch": 19.826732673267326,
      "grad_norm": 0.0009613037109375,
      "learning_rate": 6.867007456174967e-06,
      "loss": 0.2335,
      "num_input_tokens_seen": 38034736,
      "step": 180225
    },
    {
      "epoch": 19.827282728272827,
      "grad_norm": 0.005279541015625,
      "learning_rate": 6.823507491043323e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38035760,
      "step": 180230
    },
    {
      "epoch": 19.82783278327833,
      "grad_norm": 0.005859375,
      "learning_rate": 6.780145710823526e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38036848,
      "step": 180235
    },
    {
      "epoch": 19.82838283828383,
      "grad_norm": 0.010986328125,
      "learning_rate": 6.736922115918587e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38037872,
      "step": 180240
    },
    {
      "epoch": 19.82893289328933,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 6.693836706723188e-06,
      "loss": 0.2283,
      "num_input_tokens_seen": 38038928,
      "step": 180245
    },
    {
      "epoch": 19.829482948294828,
      "grad_norm": 0.00592041015625,
      "learning_rate": 6.650889483637012e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38039952,
      "step": 180250
    },
    {
      "epoch": 19.83003300330033,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 6.608080447054742e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38041008,
      "step": 180255
    },
    {
      "epoch": 19.83058305830583,
      "grad_norm": 0.005462646484375,
      "learning_rate": 6.565409597371063e-06,
      "loss": 0.2304,
      "num_input_tokens_seen": 38042096,
      "step": 180260
    },
    {
      "epoch": 19.83113311331133,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 6.522876934980659e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38043152,
      "step": 180265
    },
    {
      "epoch": 19.831683168316832,
      "grad_norm": 0.005615234375,
      "learning_rate": 6.480482460273218e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38044176,
      "step": 180270
    },
    {
      "epoch": 19.832233223322334,
      "grad_norm": 0.005889892578125,
      "learning_rate": 6.438226173641758e-06,
      "loss": 0.233,
      "num_input_tokens_seen": 38045200,
      "step": 180275
    },
    {
      "epoch": 19.83278327832783,
      "grad_norm": 0.01092529296875,
      "learning_rate": 6.396108075474305e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38046256,
      "step": 180280
    },
    {
      "epoch": 19.833333333333332,
      "grad_norm": 0.005767822265625,
      "learning_rate": 6.3541281661588786e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38047376,
      "step": 180285
    },
    {
      "epoch": 19.833883388338833,
      "grad_norm": 0.00567626953125,
      "learning_rate": 6.3122864460818385e-06,
      "loss": 0.2335,
      "num_input_tokens_seen": 38048496,
      "step": 180290
    },
    {
      "epoch": 19.834433443344334,
      "grad_norm": 0.010986328125,
      "learning_rate": 6.270582915631206e-06,
      "loss": 0.2304,
      "num_input_tokens_seen": 38049488,
      "step": 180295
    },
    {
      "epoch": 19.834983498349835,
      "grad_norm": 0.005584716796875,
      "learning_rate": 6.2290175751883445e-06,
      "loss": 0.2335,
      "num_input_tokens_seen": 38050448,
      "step": 180300
    },
    {
      "epoch": 19.835533553355337,
      "grad_norm": 0.010986328125,
      "learning_rate": 6.187590425139611e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38051504,
      "step": 180305
    },
    {
      "epoch": 19.836083608360838,
      "grad_norm": 0.005859375,
      "learning_rate": 6.146301465864701e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38052560,
      "step": 180310
    },
    {
      "epoch": 19.836633663366335,
      "grad_norm": 0.005828857421875,
      "learning_rate": 6.105150697743311e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38053648,
      "step": 180315
    },
    {
      "epoch": 19.837183718371836,
      "grad_norm": 0.0054931640625,
      "learning_rate": 6.064138121158469e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38054736,
      "step": 180320
    },
    {
      "epoch": 19.837733773377337,
      "grad_norm": 0.00543212890625,
      "learning_rate": 6.023263736483209e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38055792,
      "step": 180325
    },
    {
      "epoch": 19.83828382838284,
      "grad_norm": 0.01080322265625,
      "learning_rate": 5.982527544098892e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38056816,
      "step": 180330
    },
    {
      "epoch": 19.83883388338834,
      "grad_norm": 0.001129150390625,
      "learning_rate": 5.941929544376889e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38057808,
      "step": 180335
    },
    {
      "epoch": 19.83938393839384,
      "grad_norm": 0.005645751953125,
      "learning_rate": 5.901469737693565e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38058768,
      "step": 180340
    },
    {
      "epoch": 19.83993399339934,
      "grad_norm": 0.01104736328125,
      "learning_rate": 5.8611481244219556e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38059856,
      "step": 180345
    },
    {
      "epoch": 19.84048404840484,
      "grad_norm": 0.00567626953125,
      "learning_rate": 5.820964704933429e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38060944,
      "step": 180350
    },
    {
      "epoch": 19.84103410341034,
      "grad_norm": 0.005645751953125,
      "learning_rate": 5.78091947959769e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38062032,
      "step": 180355
    },
    {
      "epoch": 19.84158415841584,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 5.741012448784444e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 38063088,
      "step": 180360
    },
    {
      "epoch": 19.842134213421343,
      "grad_norm": 0.005523681640625,
      "learning_rate": 5.701243612861728e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 38064144,
      "step": 180365
    },
    {
      "epoch": 19.842684268426844,
      "grad_norm": 0.0054931640625,
      "learning_rate": 5.661612972195917e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38065200,
      "step": 180370
    },
    {
      "epoch": 19.843234323432345,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 5.622120527151719e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38066256,
      "step": 180375
    },
    {
      "epoch": 19.843784378437842,
      "grad_norm": 0.005950927734375,
      "learning_rate": 5.582766278093842e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38067312,
      "step": 180380
    },
    {
      "epoch": 19.844334433443343,
      "grad_norm": 0.005706787109375,
      "learning_rate": 5.543550225383664e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38068432,
      "step": 180385
    },
    {
      "epoch": 19.844884488448844,
      "grad_norm": 0.0054931640625,
      "learning_rate": 5.504472369384227e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 38069552,
      "step": 180390
    },
    {
      "epoch": 19.845434543454346,
      "grad_norm": 0.00152587890625,
      "learning_rate": 5.46553271045358e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38070576,
      "step": 180395
    },
    {
      "epoch": 19.845984598459847,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 5.426731248953098e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38071600,
      "step": 180400
    },
    {
      "epoch": 19.846534653465348,
      "grad_norm": 0.01080322265625,
      "learning_rate": 5.388067985239164e-06,
      "loss": 0.2304,
      "num_input_tokens_seen": 38072624,
      "step": 180405
    },
    {
      "epoch": 19.847084708470845,
      "grad_norm": 0.01129150390625,
      "learning_rate": 5.349542919669825e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38073680,
      "step": 180410
    },
    {
      "epoch": 19.847634763476346,
      "grad_norm": 0.002288818359375,
      "learning_rate": 5.311156052596466e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38074704,
      "step": 180415
    },
    {
      "epoch": 19.848184818481847,
      "grad_norm": 0.005645751953125,
      "learning_rate": 5.272907384375469e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38075760,
      "step": 180420
    },
    {
      "epoch": 19.84873487348735,
      "grad_norm": 0.00543212890625,
      "learning_rate": 5.234796915359885e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38076816,
      "step": 180425
    },
    {
      "epoch": 19.84928492849285,
      "grad_norm": 0.005462646484375,
      "learning_rate": 5.196824645899433e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38077904,
      "step": 180430
    },
    {
      "epoch": 19.84983498349835,
      "grad_norm": 0.0057373046875,
      "learning_rate": 5.158990576343836e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38078960,
      "step": 180435
    },
    {
      "epoch": 19.850385038503852,
      "grad_norm": 0.00164794921875,
      "learning_rate": 5.121294707044477e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38080048,
      "step": 180440
    },
    {
      "epoch": 19.85093509350935,
      "grad_norm": 0.005523681640625,
      "learning_rate": 5.083737038346081e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 38081072,
      "step": 180445
    },
    {
      "epoch": 19.85148514851485,
      "grad_norm": 0.005828857421875,
      "learning_rate": 5.0463175705950375e-06,
      "loss": 0.2329,
      "num_input_tokens_seen": 38082128,
      "step": 180450
    },
    {
      "epoch": 19.85203520352035,
      "grad_norm": 0.00177001953125,
      "learning_rate": 5.009036304136072e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38083152,
      "step": 180455
    },
    {
      "epoch": 19.852585258525853,
      "grad_norm": 0.005828857421875,
      "learning_rate": 4.971893239315572e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38084240,
      "step": 180460
    },
    {
      "epoch": 19.853135313531354,
      "grad_norm": 0.005462646484375,
      "learning_rate": 4.934888376473267e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38085264,
      "step": 180465
    },
    {
      "epoch": 19.853685368536855,
      "grad_norm": 0.005706787109375,
      "learning_rate": 4.898021715950551e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38086320,
      "step": 180470
    },
    {
      "epoch": 19.854235423542356,
      "grad_norm": 0.000789642333984375,
      "learning_rate": 4.8612932580871515e-06,
      "loss": 0.2335,
      "num_input_tokens_seen": 38087312,
      "step": 180475
    },
    {
      "epoch": 19.854785478547853,
      "grad_norm": 0.005615234375,
      "learning_rate": 4.824703003221131e-06,
      "loss": 0.2335,
      "num_input_tokens_seen": 38088336,
      "step": 180480
    },
    {
      "epoch": 19.855335533553355,
      "grad_norm": 0.005706787109375,
      "learning_rate": 4.788250951690554e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 38089328,
      "step": 180485
    },
    {
      "epoch": 19.855885588558856,
      "grad_norm": 0.005859375,
      "learning_rate": 4.751937103831816e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38090416,
      "step": 180490
    },
    {
      "epoch": 19.856435643564357,
      "grad_norm": 0.005889892578125,
      "learning_rate": 4.71576145997965e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38091472,
      "step": 180495
    },
    {
      "epoch": 19.856985698569858,
      "grad_norm": 0.00121307373046875,
      "learning_rate": 4.679724020467124e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38092528,
      "step": 180500
    },
    {
      "epoch": 19.85753575357536,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 4.643824785625639e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38093552,
      "step": 180505
    },
    {
      "epoch": 19.858085808580856,
      "grad_norm": 0.005859375,
      "learning_rate": 4.60806375578493e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38094608,
      "step": 180510
    },
    {
      "epoch": 19.858635863586358,
      "grad_norm": 0.005615234375,
      "learning_rate": 4.572440931278066e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38095632,
      "step": 180515
    },
    {
      "epoch": 19.85918591859186,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 4.5369563124314505e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38096752,
      "step": 180520
    },
    {
      "epoch": 19.85973597359736,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 4.50160989957149e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38097776,
      "step": 180525
    },
    {
      "epoch": 19.86028602860286,
      "grad_norm": 0.00136566162109375,
      "learning_rate": 4.4664016930245905e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38098864,
      "step": 180530
    },
    {
      "epoch": 19.860836083608362,
      "grad_norm": 0.005767822265625,
      "learning_rate": 4.431331693115492e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38099920,
      "step": 180535
    },
    {
      "epoch": 19.861386138613863,
      "grad_norm": 0.0011444091796875,
      "learning_rate": 4.3963999001672695e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38100944,
      "step": 180540
    },
    {
      "epoch": 19.86193619361936,
      "grad_norm": 0.0108642578125,
      "learning_rate": 4.361606314502997e-06,
      "loss": 0.2313,
      "num_input_tokens_seen": 38102032,
      "step": 180545
    },
    {
      "epoch": 19.86248624862486,
      "grad_norm": 0.006134033203125,
      "learning_rate": 4.326950936440755e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38103088,
      "step": 180550
    },
    {
      "epoch": 19.863036303630363,
      "grad_norm": 0.005584716796875,
      "learning_rate": 4.292433766301951e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38104144,
      "step": 180555
    },
    {
      "epoch": 19.863586358635864,
      "grad_norm": 0.00104522705078125,
      "learning_rate": 4.2580548044029995e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38105200,
      "step": 180560
    },
    {
      "epoch": 19.864136413641365,
      "grad_norm": 0.00567626953125,
      "learning_rate": 4.223814051063646e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38106192,
      "step": 180565
    },
    {
      "epoch": 19.864686468646866,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 4.189711506596971e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38107216,
      "step": 180570
    },
    {
      "epoch": 19.865236523652364,
      "grad_norm": 0.005767822265625,
      "learning_rate": 4.1557471713177254e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38108240,
      "step": 180575
    },
    {
      "epoch": 19.865786578657865,
      "grad_norm": 0.0052490234375,
      "learning_rate": 4.121921045538989e-06,
      "loss": 0.2304,
      "num_input_tokens_seen": 38109264,
      "step": 180580
    },
    {
      "epoch": 19.866336633663366,
      "grad_norm": 0.005706787109375,
      "learning_rate": 4.088233129573848e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38110320,
      "step": 180585
    },
    {
      "epoch": 19.866886688668867,
      "grad_norm": 0.005889892578125,
      "learning_rate": 4.054683423730387e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38111312,
      "step": 180590
    },
    {
      "epoch": 19.867436743674368,
      "grad_norm": 0.00213623046875,
      "learning_rate": 4.021271928320025e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38112368,
      "step": 180595
    },
    {
      "epoch": 19.86798679867987,
      "grad_norm": 0.005584716796875,
      "learning_rate": 3.9879986436508474e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38113392,
      "step": 180600
    },
    {
      "epoch": 19.86853685368537,
      "grad_norm": 0.00130462646484375,
      "learning_rate": 3.954863570027611e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38114544,
      "step": 180605
    },
    {
      "epoch": 19.869086908690868,
      "grad_norm": 0.005706787109375,
      "learning_rate": 3.921866707755072e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38115600,
      "step": 180610
    },
    {
      "epoch": 19.86963696369637,
      "grad_norm": 0.01104736328125,
      "learning_rate": 3.889008057141318e-06,
      "loss": 0.2329,
      "num_input_tokens_seen": 38116656,
      "step": 180615
    },
    {
      "epoch": 19.87018701870187,
      "grad_norm": 0.006134033203125,
      "learning_rate": 3.856287618484444e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38117776,
      "step": 180620
    },
    {
      "epoch": 19.87073707370737,
      "grad_norm": 0.000553131103515625,
      "learning_rate": 3.82370539209087e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38118832,
      "step": 180625
    },
    {
      "epoch": 19.871287128712872,
      "grad_norm": 0.00653076171875,
      "learning_rate": 3.7912613782553617e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 38119856,
      "step": 180630
    },
    {
      "epoch": 19.871837183718373,
      "grad_norm": 0.0054931640625,
      "learning_rate": 3.7589555772826744e-06,
      "loss": 0.2313,
      "num_input_tokens_seen": 38120912,
      "step": 180635
    },
    {
      "epoch": 19.87238723872387,
      "grad_norm": 0.00579833984375,
      "learning_rate": 3.726787989465907e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38121936,
      "step": 180640
    },
    {
      "epoch": 19.872937293729372,
      "grad_norm": 0.0057373046875,
      "learning_rate": 3.6947586151048204e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38122992,
      "step": 180645
    },
    {
      "epoch": 19.873487348734873,
      "grad_norm": 0.0054931640625,
      "learning_rate": 3.6628674544925132e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 38124112,
      "step": 180650
    },
    {
      "epoch": 19.874037403740374,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 3.631114507923749e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38125168,
      "step": 180655
    },
    {
      "epoch": 19.874587458745875,
      "grad_norm": 0.00537109375,
      "learning_rate": 3.5994997756899627e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38126224,
      "step": 180660
    },
    {
      "epoch": 19.875137513751376,
      "grad_norm": 0.005615234375,
      "learning_rate": 3.568023258085917e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38127280,
      "step": 180665
    },
    {
      "epoch": 19.875687568756877,
      "grad_norm": 0.000965118408203125,
      "learning_rate": 3.5366849553980505e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38128304,
      "step": 180670
    },
    {
      "epoch": 19.876237623762375,
      "grad_norm": 0.0025634765625,
      "learning_rate": 3.505484867916131e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38129360,
      "step": 180675
    },
    {
      "epoch": 19.876787678767876,
      "grad_norm": 0.00537109375,
      "learning_rate": 3.474422995928261e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38130416,
      "step": 180680
    },
    {
      "epoch": 19.877337733773377,
      "grad_norm": 0.0059814453125,
      "learning_rate": 3.443499339720879e-06,
      "loss": 0.2335,
      "num_input_tokens_seen": 38131472,
      "step": 180685
    },
    {
      "epoch": 19.877887788778878,
      "grad_norm": 0.0057373046875,
      "learning_rate": 3.4127138995787565e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38132528,
      "step": 180690
    },
    {
      "epoch": 19.87843784378438,
      "grad_norm": 0.01092529296875,
      "learning_rate": 3.3820666757866654e-06,
      "loss": 0.2313,
      "num_input_tokens_seen": 38133584,
      "step": 180695
    },
    {
      "epoch": 19.87898789878988,
      "grad_norm": 0.00555419921875,
      "learning_rate": 3.351557668624383e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38134608,
      "step": 180700
    },
    {
      "epoch": 19.879537953795378,
      "grad_norm": 0.000904083251953125,
      "learning_rate": 3.321186878375015e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38135632,
      "step": 180705
    },
    {
      "epoch": 19.88008800880088,
      "grad_norm": 0.005706787109375,
      "learning_rate": 3.2909543053183388e-06,
      "loss": 0.233,
      "num_input_tokens_seen": 38136752,
      "step": 180710
    },
    {
      "epoch": 19.88063806380638,
      "grad_norm": 0.006378173828125,
      "learning_rate": 3.260859949732464e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38137744,
      "step": 180715
    },
    {
      "epoch": 19.88118811881188,
      "grad_norm": 0.005859375,
      "learning_rate": 3.2309038118955025e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38138800,
      "step": 180720
    },
    {
      "epoch": 19.881738173817382,
      "grad_norm": 0.005462646484375,
      "learning_rate": 3.201085892083899e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38139824,
      "step": 180725
    },
    {
      "epoch": 19.882288228822883,
      "grad_norm": 0.005462646484375,
      "learning_rate": 3.171406190570769e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38140816,
      "step": 180730
    },
    {
      "epoch": 19.882838283828384,
      "grad_norm": 0.0057373046875,
      "learning_rate": 3.1418647076308925e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38141872,
      "step": 180735
    },
    {
      "epoch": 19.883388338833882,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 3.112461443535719e-06,
      "loss": 0.2334,
      "num_input_tokens_seen": 38142928,
      "step": 180740
    },
    {
      "epoch": 19.883938393839383,
      "grad_norm": 0.01080322265625,
      "learning_rate": 3.083196398558363e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38143984,
      "step": 180745
    },
    {
      "epoch": 19.884488448844884,
      "grad_norm": 0.006256103515625,
      "learning_rate": 3.0540695729652786e-06,
      "loss": 0.233,
      "num_input_tokens_seen": 38145040,
      "step": 180750
    },
    {
      "epoch": 19.885038503850385,
      "grad_norm": 0.005523681640625,
      "learning_rate": 3.0250809670279154e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38146096,
      "step": 180755
    },
    {
      "epoch": 19.885588558855886,
      "grad_norm": 0.005645751953125,
      "learning_rate": 2.996230581011061e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38147152,
      "step": 180760
    },
    {
      "epoch": 19.886138613861387,
      "grad_norm": 0.0023193359375,
      "learning_rate": 2.9675184151811693e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38148304,
      "step": 180765
    },
    {
      "epoch": 19.88668866886689,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 2.9389444698046938e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38149328,
      "step": 180770
    },
    {
      "epoch": 19.887238723872386,
      "grad_norm": 0.005523681640625,
      "learning_rate": 2.910508745143092e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38150352,
      "step": 180775
    },
    {
      "epoch": 19.887788778877887,
      "grad_norm": 0.005706787109375,
      "learning_rate": 2.8822112414594867e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38151440,
      "step": 180780
    },
    {
      "epoch": 19.888338833883388,
      "grad_norm": 0.01092529296875,
      "learning_rate": 2.85405195901367e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38152496,
      "step": 180785
    },
    {
      "epoch": 19.88888888888889,
      "grad_norm": 0.001678466796875,
      "learning_rate": 2.8260308980654348e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38153552,
      "step": 180790
    },
    {
      "epoch": 19.88943894389439,
      "grad_norm": 0.0023956298828125,
      "learning_rate": 2.798148058874572e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38154672,
      "step": 180795
    },
    {
      "epoch": 19.88998899889989,
      "grad_norm": 0.00323486328125,
      "learning_rate": 2.770403441695879e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38155696,
      "step": 180800
    },
    {
      "epoch": 19.89053905390539,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 2.7427970467858165e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38156752,
      "step": 180805
    },
    {
      "epoch": 19.89108910891089,
      "grad_norm": 0.00115966796875,
      "learning_rate": 2.715328874400846e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38157776,
      "step": 180810
    },
    {
      "epoch": 19.89163916391639,
      "grad_norm": 0.005706787109375,
      "learning_rate": 2.6879989247907686e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38158832,
      "step": 180815
    },
    {
      "epoch": 19.892189218921892,
      "grad_norm": 0.00118255615234375,
      "learning_rate": 2.6608071982087145e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38159856,
      "step": 180820
    },
    {
      "epoch": 19.892739273927393,
      "grad_norm": 0.005645751953125,
      "learning_rate": 2.633753694906149e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38160912,
      "step": 180825
    },
    {
      "epoch": 19.893289328932894,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 2.6068384151312074e-06,
      "loss": 0.2335,
      "num_input_tokens_seen": 38162000,
      "step": 180830
    },
    {
      "epoch": 19.893839383938392,
      "grad_norm": 0.00156402587890625,
      "learning_rate": 2.5800613591336895e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38163056,
      "step": 180835
    },
    {
      "epoch": 19.894389438943893,
      "grad_norm": 0.00543212890625,
      "learning_rate": 2.5534225271584e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38164080,
      "step": 180840
    },
    {
      "epoch": 19.894939493949394,
      "grad_norm": 0.005615234375,
      "learning_rate": 2.5269219194518077e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38165104,
      "step": 180845
    },
    {
      "epoch": 19.895489548954895,
      "grad_norm": 0.0107421875,
      "learning_rate": 2.5005595362587173e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38166160,
      "step": 180850
    },
    {
      "epoch": 19.896039603960396,
      "grad_norm": 0.0108642578125,
      "learning_rate": 2.474335377820602e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38167248,
      "step": 180855
    },
    {
      "epoch": 19.896589658965897,
      "grad_norm": 0.005401611328125,
      "learning_rate": 2.448249444380601e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 38168272,
      "step": 180860
    },
    {
      "epoch": 19.8971397139714,
      "grad_norm": 0.005523681640625,
      "learning_rate": 2.4223017361785224e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38169232,
      "step": 180865
    },
    {
      "epoch": 19.897689768976896,
      "grad_norm": 0.0007171630859375,
      "learning_rate": 2.396492253452509e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38170256,
      "step": 180870
    },
    {
      "epoch": 19.898239823982397,
      "grad_norm": 0.01129150390625,
      "learning_rate": 2.3708209964423686e-06,
      "loss": 0.2329,
      "num_input_tokens_seen": 38171344,
      "step": 180875
    },
    {
      "epoch": 19.8987898789879,
      "grad_norm": 0.00185394287109375,
      "learning_rate": 2.3452879653845793e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38172496,
      "step": 180880
    },
    {
      "epoch": 19.8993399339934,
      "grad_norm": 0.005584716796875,
      "learning_rate": 2.3198931605122874e-06,
      "loss": 0.234,
      "num_input_tokens_seen": 38173584,
      "step": 180885
    },
    {
      "epoch": 19.8998899889989,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 2.2946365820619706e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38174608,
      "step": 180890
    },
    {
      "epoch": 19.9004400440044,
      "grad_norm": 0.000820159912109375,
      "learning_rate": 2.2695182302651106e-06,
      "loss": 0.233,
      "num_input_tokens_seen": 38175568,
      "step": 180895
    },
    {
      "epoch": 19.900990099009903,
      "grad_norm": 0.005645751953125,
      "learning_rate": 2.2445381053531885e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38176624,
      "step": 180900
    },
    {
      "epoch": 19.9015401540154,
      "grad_norm": 0.00543212890625,
      "learning_rate": 2.2196962075560212e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38177616,
      "step": 180905
    },
    {
      "epoch": 19.9020902090209,
      "grad_norm": 0.005706787109375,
      "learning_rate": 2.1949925371050893e-06,
      "loss": 0.2335,
      "num_input_tokens_seen": 38178704,
      "step": 180910
    },
    {
      "epoch": 19.902640264026402,
      "grad_norm": 0.005645751953125,
      "learning_rate": 2.1704270942252136e-06,
      "loss": 0.2335,
      "num_input_tokens_seen": 38179760,
      "step": 180915
    },
    {
      "epoch": 19.903190319031903,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 2.1459998791445444e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38180848,
      "step": 180920
    },
    {
      "epoch": 19.903740374037405,
      "grad_norm": 0.005523681640625,
      "learning_rate": 2.121710892086237e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 38181904,
      "step": 180925
    },
    {
      "epoch": 19.904290429042906,
      "grad_norm": 0.0111083984375,
      "learning_rate": 2.0975601332767766e-06,
      "loss": 0.2329,
      "num_input_tokens_seen": 38182864,
      "step": 180930
    },
    {
      "epoch": 19.904840484048403,
      "grad_norm": 0.005584716796875,
      "learning_rate": 2.0735476029359877e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38183856,
      "step": 180935
    },
    {
      "epoch": 19.905390539053904,
      "grad_norm": 0.005584716796875,
      "learning_rate": 2.04967330128869e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38184880,
      "step": 180940
    },
    {
      "epoch": 19.905940594059405,
      "grad_norm": 0.00150299072265625,
      "learning_rate": 2.0259372285513777e-06,
      "loss": 0.2329,
      "num_input_tokens_seen": 38185904,
      "step": 180945
    },
    {
      "epoch": 19.906490649064907,
      "grad_norm": 0.005645751953125,
      "learning_rate": 2.0023393849438742e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38186896,
      "step": 180950
    },
    {
      "epoch": 19.907040704070408,
      "grad_norm": 0.00592041015625,
      "learning_rate": 1.9788797706843386e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38187984,
      "step": 180955
    },
    {
      "epoch": 19.90759075907591,
      "grad_norm": 0.005767822265625,
      "learning_rate": 1.9555583859892643e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38189040,
      "step": 180960
    },
    {
      "epoch": 19.90814081408141,
      "grad_norm": 0.00107574462890625,
      "learning_rate": 1.9323752310734796e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38190064,
      "step": 180965
    },
    {
      "epoch": 19.908690869086907,
      "grad_norm": 0.005859375,
      "learning_rate": 1.9093303061484825e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38191120,
      "step": 180970
    },
    {
      "epoch": 19.90924092409241,
      "grad_norm": 0.00604248046875,
      "learning_rate": 1.8864236114307652e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38192080,
      "step": 180975
    },
    {
      "epoch": 19.90979097909791,
      "grad_norm": 0.006072998046875,
      "learning_rate": 1.8636551471268303e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38193200,
      "step": 180980
    },
    {
      "epoch": 19.91034103410341,
      "grad_norm": 0.0011749267578125,
      "learning_rate": 1.8410249134498402e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38194288,
      "step": 180985
    },
    {
      "epoch": 19.91089108910891,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 1.8185329106079617e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38195440,
      "step": 180990
    },
    {
      "epoch": 19.911441144114413,
      "grad_norm": 0.002410888671875,
      "learning_rate": 1.7961791388076963e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38196432,
      "step": 180995
    },
    {
      "epoch": 19.91199119911991,
      "grad_norm": 0.005615234375,
      "learning_rate": 1.7739635982538803e-06,
      "loss": 0.2313,
      "num_input_tokens_seen": 38197456,
      "step": 181000
    },
    {
      "epoch": 19.91254125412541,
      "grad_norm": 0.0057373046875,
      "learning_rate": 1.751886289153015e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 38198448,
      "step": 181005
    },
    {
      "epoch": 19.913091309130913,
      "grad_norm": 0.01092529296875,
      "learning_rate": 1.7299472117082714e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38199504,
      "step": 181010
    },
    {
      "epoch": 19.913641364136414,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 1.7081463661228203e-06,
      "loss": 0.2313,
      "num_input_tokens_seen": 38200592,
      "step": 181015
    },
    {
      "epoch": 19.914191419141915,
      "grad_norm": 0.00115966796875,
      "learning_rate": 1.686483752594836e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38201616,
      "step": 181020
    },
    {
      "epoch": 19.914741474147416,
      "grad_norm": 0.005523681640625,
      "learning_rate": 1.66495937132749e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38202576,
      "step": 181025
    },
    {
      "epoch": 19.915291529152917,
      "grad_norm": 0.00096893310546875,
      "learning_rate": 1.6435732225156263e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38203696,
      "step": 181030
    },
    {
      "epoch": 19.915841584158414,
      "grad_norm": 0.0057373046875,
      "learning_rate": 1.6223253063590848e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38204784,
      "step": 181035
    },
    {
      "epoch": 19.916391639163916,
      "grad_norm": 0.00168609619140625,
      "learning_rate": 1.6012156230527096e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38205840,
      "step": 181040
    },
    {
      "epoch": 19.916941694169417,
      "grad_norm": 0.0011138916015625,
      "learning_rate": 1.5802441727896798e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38206896,
      "step": 181045
    },
    {
      "epoch": 19.917491749174918,
      "grad_norm": 0.01092529296875,
      "learning_rate": 1.5594109557648395e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38207888,
      "step": 181050
    },
    {
      "epoch": 19.91804180418042,
      "grad_norm": 0.00628662109375,
      "learning_rate": 1.5387159721713672e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38208944,
      "step": 181055
    },
    {
      "epoch": 19.91859185918592,
      "grad_norm": 0.005615234375,
      "learning_rate": 1.518159222197446e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38209968,
      "step": 181060
    },
    {
      "epoch": 19.919141914191417,
      "grad_norm": 0.0010833740234375,
      "learning_rate": 1.4977407060345892e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38210992,
      "step": 181065
    },
    {
      "epoch": 19.91969196919692,
      "grad_norm": 0.01068115234375,
      "learning_rate": 1.4774604238693145e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38212080,
      "step": 181070
    },
    {
      "epoch": 19.92024202420242,
      "grad_norm": 0.01104736328125,
      "learning_rate": 1.4573183758898044e-06,
      "loss": 0.2288,
      "num_input_tokens_seen": 38213136,
      "step": 181075
    },
    {
      "epoch": 19.92079207920792,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 1.4373145622809114e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38214224,
      "step": 181080
    },
    {
      "epoch": 19.921342134213422,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 1.417448983227487e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38215248,
      "step": 181085
    },
    {
      "epoch": 19.921892189218923,
      "grad_norm": 0.000919342041015625,
      "learning_rate": 1.3977216389110535e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38216272,
      "step": 181090
    },
    {
      "epoch": 19.922442244224424,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 1.3781325295164626e-06,
      "loss": 0.2288,
      "num_input_tokens_seen": 38217360,
      "step": 181095
    },
    {
      "epoch": 19.92299229922992,
      "grad_norm": 0.005523681640625,
      "learning_rate": 1.358681655221905e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38218416,
      "step": 181100
    },
    {
      "epoch": 19.923542354235423,
      "grad_norm": 0.00189971923828125,
      "learning_rate": 1.3393690162072369e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38219504,
      "step": 181105
    },
    {
      "epoch": 19.924092409240924,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 1.3201946126506492e-06,
      "loss": 0.2329,
      "num_input_tokens_seen": 38220592,
      "step": 181110
    },
    {
      "epoch": 19.924642464246425,
      "grad_norm": 0.005828857421875,
      "learning_rate": 1.3011584447286672e-06,
      "loss": 0.2319,
      "num_input_tokens_seen": 38221712,
      "step": 181115
    },
    {
      "epoch": 19.925192519251926,
      "grad_norm": 0.0013275146484375,
      "learning_rate": 1.2822605126178166e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38222736,
      "step": 181120
    },
    {
      "epoch": 19.925742574257427,
      "grad_norm": 0.01092529296875,
      "learning_rate": 1.2635008164896266e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38223824,
      "step": 181125
    },
    {
      "epoch": 19.926292629262925,
      "grad_norm": 0.001800537109375,
      "learning_rate": 1.2448793565206229e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38224912,
      "step": 181130
    },
    {
      "epoch": 19.926842684268426,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 1.226396132879004e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38225936,
      "step": 181135
    },
    {
      "epoch": 19.927392739273927,
      "grad_norm": 0.005584716796875,
      "learning_rate": 1.2080511457362997e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 38227056,
      "step": 181140
    },
    {
      "epoch": 19.927942794279428,
      "grad_norm": 0.005828857421875,
      "learning_rate": 1.1898443952623738e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38228080,
      "step": 181145
    },
    {
      "epoch": 19.92849284928493,
      "grad_norm": 0.005462646484375,
      "learning_rate": 1.1717758816254252e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 38229104,
      "step": 181150
    },
    {
      "epoch": 19.92904290429043,
      "grad_norm": 0.005401611328125,
      "learning_rate": 1.1538456049903222e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 38230160,
      "step": 181155
    },
    {
      "epoch": 19.92959295929593,
      "grad_norm": 0.01123046875,
      "learning_rate": 1.136053565523598e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 38231248,
      "step": 181160
    },
    {
      "epoch": 19.93014301430143,
      "grad_norm": 0.0012359619140625,
      "learning_rate": 1.1183997633884557e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 38232304,
      "step": 181165
    },
    {
      "epoch": 19.93069306930693,
      "grad_norm": 0.005523681640625,
      "learning_rate": 1.1008841987464323e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 38233296,
      "step": 181170
    },
    {
      "epoch": 19.93124312431243,
      "grad_norm": 0.005859375,
      "learning_rate": 1.0835068717623963e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38234416,
      "step": 181175
    },
    {
      "epoch": 19.931793179317932,
      "grad_norm": 0.00165557861328125,
      "learning_rate": 1.066267782592889e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38235504,
      "step": 181180
    },
    {
      "epoch": 19.932343234323433,
      "grad_norm": 0.005645751953125,
      "learning_rate": 1.0491669313994479e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 38236688,
      "step": 181185
    },
    {
      "epoch": 19.932893289328934,
      "grad_norm": 0.005706787109375,
      "learning_rate": 1.0322043183386142e-06,
      "loss": 0.2303,
      "num_input_tokens_seen": 38237712,
      "step": 181190
    },
    {
      "epoch": 19.933443344334435,
      "grad_norm": 0.00106048583984375,
      "learning_rate": 1.0153799435669298e-06,
      "loss": 0.2329,
      "num_input_tokens_seen": 38238800,
      "step": 181195
    },
    {
      "epoch": 19.933993399339933,
      "grad_norm": 0.005584716796875,
      "learning_rate": 9.98693807237605e-07,
      "loss": 0.2314,
      "num_input_tokens_seen": 38239792,
      "step": 181200
    },
    {
      "epoch": 19.934543454345434,
      "grad_norm": 0.00139617919921875,
      "learning_rate": 9.821459095088469e-07,
      "loss": 0.2303,
      "num_input_tokens_seen": 38240848,
      "step": 181205
    },
    {
      "epoch": 19.935093509350935,
      "grad_norm": 0.002685546875,
      "learning_rate": 9.657362505288701e-07,
      "loss": 0.2298,
      "num_input_tokens_seen": 38241872,
      "step": 181210
    },
    {
      "epoch": 19.935643564356436,
      "grad_norm": 0.01092529296875,
      "learning_rate": 9.494648304492203e-07,
      "loss": 0.2303,
      "num_input_tokens_seen": 38242832,
      "step": 181215
    },
    {
      "epoch": 19.936193619361937,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 9.33331649423108e-07,
      "loss": 0.2314,
      "num_input_tokens_seen": 38243888,
      "step": 181220
    },
    {
      "epoch": 19.936743674367438,
      "grad_norm": 0.005645751953125,
      "learning_rate": 9.173367075970828e-07,
      "loss": 0.2324,
      "num_input_tokens_seen": 38244976,
      "step": 181225
    },
    {
      "epoch": 19.937293729372936,
      "grad_norm": 0.00164794921875,
      "learning_rate": 9.01480005117694e-07,
      "loss": 0.2309,
      "num_input_tokens_seen": 38246064,
      "step": 181230
    },
    {
      "epoch": 19.937843784378437,
      "grad_norm": 0.000957489013671875,
      "learning_rate": 8.857615421314913e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38247184,
      "step": 181235
    },
    {
      "epoch": 19.938393839383938,
      "grad_norm": 0.010986328125,
      "learning_rate": 8.701813187850237e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38248176,
      "step": 181240
    },
    {
      "epoch": 19.93894389438944,
      "grad_norm": 0.0014190673828125,
      "learning_rate": 8.547393352215105e-07,
      "loss": 0.2308,
      "num_input_tokens_seen": 38249296,
      "step": 181245
    },
    {
      "epoch": 19.93949394939494,
      "grad_norm": 0.00101470947265625,
      "learning_rate": 8.394355915825047e-07,
      "loss": 0.2309,
      "num_input_tokens_seen": 38250320,
      "step": 181250
    },
    {
      "epoch": 19.94004400440044,
      "grad_norm": 0.005767822265625,
      "learning_rate": 8.242700880078946e-07,
      "loss": 0.2314,
      "num_input_tokens_seen": 38251408,
      "step": 181255
    },
    {
      "epoch": 19.94059405940594,
      "grad_norm": 0.00555419921875,
      "learning_rate": 8.092428246392336e-07,
      "loss": 0.2308,
      "num_input_tokens_seen": 38252496,
      "step": 181260
    },
    {
      "epoch": 19.94114411441144,
      "grad_norm": 0.0025482177734375,
      "learning_rate": 7.943538016147444e-07,
      "loss": 0.2335,
      "num_input_tokens_seen": 38253552,
      "step": 181265
    },
    {
      "epoch": 19.94169416941694,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 7.796030190709846e-07,
      "loss": 0.2303,
      "num_input_tokens_seen": 38254544,
      "step": 181270
    },
    {
      "epoch": 19.942244224422442,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 7.649904771445115e-07,
      "loss": 0.2329,
      "num_input_tokens_seen": 38255632,
      "step": 181275
    },
    {
      "epoch": 19.942794279427943,
      "grad_norm": 0.00579833984375,
      "learning_rate": 7.505161759702172e-07,
      "loss": 0.2293,
      "num_input_tokens_seen": 38256752,
      "step": 181280
    },
    {
      "epoch": 19.943344334433444,
      "grad_norm": 0.005615234375,
      "learning_rate": 7.361801156796632e-07,
      "loss": 0.2324,
      "num_input_tokens_seen": 38257744,
      "step": 181285
    },
    {
      "epoch": 19.943894389438945,
      "grad_norm": 0.00144195556640625,
      "learning_rate": 7.219822964077415e-07,
      "loss": 0.2308,
      "num_input_tokens_seen": 38258832,
      "step": 181290
    },
    {
      "epoch": 19.944444444444443,
      "grad_norm": 0.005615234375,
      "learning_rate": 7.07922718282683e-07,
      "loss": 0.2303,
      "num_input_tokens_seen": 38259984,
      "step": 181295
    },
    {
      "epoch": 19.944994499449944,
      "grad_norm": 0.001129150390625,
      "learning_rate": 6.94001381436049e-07,
      "loss": 0.2303,
      "num_input_tokens_seen": 38261136,
      "step": 181300
    },
    {
      "epoch": 19.945544554455445,
      "grad_norm": 0.005645751953125,
      "learning_rate": 6.802182859960703e-07,
      "loss": 0.2298,
      "num_input_tokens_seen": 38262160,
      "step": 181305
    },
    {
      "epoch": 19.946094609460946,
      "grad_norm": 0.005859375,
      "learning_rate": 6.665734320893124e-07,
      "loss": 0.2324,
      "num_input_tokens_seen": 38263184,
      "step": 181310
    },
    {
      "epoch": 19.946644664466447,
      "grad_norm": 0.005645751953125,
      "learning_rate": 6.530668198406753e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38264208,
      "step": 181315
    },
    {
      "epoch": 19.94719471947195,
      "grad_norm": 0.002166748046875,
      "learning_rate": 6.39698449375059e-07,
      "loss": 0.2309,
      "num_input_tokens_seen": 38265328,
      "step": 181320
    },
    {
      "epoch": 19.94774477447745,
      "grad_norm": 0.0054931640625,
      "learning_rate": 6.264683208156985e-07,
      "loss": 0.2314,
      "num_input_tokens_seen": 38266384,
      "step": 181325
    },
    {
      "epoch": 19.948294829482947,
      "grad_norm": 0.00115966796875,
      "learning_rate": 6.133764342858283e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38267472,
      "step": 181330
    },
    {
      "epoch": 19.948844884488448,
      "grad_norm": 0.005584716796875,
      "learning_rate": 6.004227899053526e-07,
      "loss": 0.2303,
      "num_input_tokens_seen": 38268528,
      "step": 181335
    },
    {
      "epoch": 19.94939493949395,
      "grad_norm": 0.0013580322265625,
      "learning_rate": 5.876073877925103e-07,
      "loss": 0.2309,
      "num_input_tokens_seen": 38269584,
      "step": 181340
    },
    {
      "epoch": 19.94994499449945,
      "grad_norm": 0.005462646484375,
      "learning_rate": 5.7493022806554e-07,
      "loss": 0.2303,
      "num_input_tokens_seen": 38270608,
      "step": 181345
    },
    {
      "epoch": 19.95049504950495,
      "grad_norm": 0.0054931640625,
      "learning_rate": 5.623913108426803e-07,
      "loss": 0.2335,
      "num_input_tokens_seen": 38271696,
      "step": 181350
    },
    {
      "epoch": 19.951045104510452,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 5.499906362388396e-07,
      "loss": 0.234,
      "num_input_tokens_seen": 38272720,
      "step": 181355
    },
    {
      "epoch": 19.95159515951595,
      "grad_norm": 0.00555419921875,
      "learning_rate": 5.377282043689257e-07,
      "loss": 0.2298,
      "num_input_tokens_seen": 38273776,
      "step": 181360
    },
    {
      "epoch": 19.95214521452145,
      "grad_norm": 0.010986328125,
      "learning_rate": 5.256040153445163e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38274864,
      "step": 181365
    },
    {
      "epoch": 19.952695269526952,
      "grad_norm": 0.00090789794921875,
      "learning_rate": 5.136180692788539e-07,
      "loss": 0.2298,
      "num_input_tokens_seen": 38275920,
      "step": 181370
    },
    {
      "epoch": 19.953245324532453,
      "grad_norm": 0.0111083984375,
      "learning_rate": 5.017703662818506e-07,
      "loss": 0.2314,
      "num_input_tokens_seen": 38276944,
      "step": 181375
    },
    {
      "epoch": 19.953795379537954,
      "grad_norm": 0.005615234375,
      "learning_rate": 4.900609064617534e-07,
      "loss": 0.2335,
      "num_input_tokens_seen": 38278000,
      "step": 181380
    },
    {
      "epoch": 19.954345434543455,
      "grad_norm": 0.001800537109375,
      "learning_rate": 4.784896899268088e-07,
      "loss": 0.2303,
      "num_input_tokens_seen": 38279056,
      "step": 181385
    },
    {
      "epoch": 19.954895489548957,
      "grad_norm": 0.0015411376953125,
      "learning_rate": 4.670567167852635e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38280080,
      "step": 181390
    },
    {
      "epoch": 19.955445544554454,
      "grad_norm": 0.0115966796875,
      "learning_rate": 4.557619871420337e-07,
      "loss": 0.234,
      "num_input_tokens_seen": 38281136,
      "step": 181395
    },
    {
      "epoch": 19.955995599559955,
      "grad_norm": 0.01092529296875,
      "learning_rate": 4.4460550109870485e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38282192,
      "step": 181400
    },
    {
      "epoch": 19.956545654565456,
      "grad_norm": 0.0023345947265625,
      "learning_rate": 4.3358725876019297e-07,
      "loss": 0.2308,
      "num_input_tokens_seen": 38283248,
      "step": 181405
    },
    {
      "epoch": 19.957095709570957,
      "grad_norm": 0.00579833984375,
      "learning_rate": 4.227072602297488e-07,
      "loss": 0.2314,
      "num_input_tokens_seen": 38284336,
      "step": 181410
    },
    {
      "epoch": 19.95764576457646,
      "grad_norm": 0.00201416015625,
      "learning_rate": 4.119655056039617e-07,
      "loss": 0.2298,
      "num_input_tokens_seen": 38285424,
      "step": 181415
    },
    {
      "epoch": 19.95819581958196,
      "grad_norm": 0.00124359130859375,
      "learning_rate": 4.0136199498441716e-07,
      "loss": 0.2309,
      "num_input_tokens_seen": 38286544,
      "step": 181420
    },
    {
      "epoch": 19.958745874587457,
      "grad_norm": 0.01104736328125,
      "learning_rate": 3.908967284677045e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38287600,
      "step": 181425
    },
    {
      "epoch": 19.959295929592958,
      "grad_norm": 0.0057373046875,
      "learning_rate": 3.8056970615041316e-07,
      "loss": 0.2335,
      "num_input_tokens_seen": 38288656,
      "step": 181430
    },
    {
      "epoch": 19.95984598459846,
      "grad_norm": 0.00112152099609375,
      "learning_rate": 3.703809281274672e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38289712,
      "step": 181435
    },
    {
      "epoch": 19.96039603960396,
      "grad_norm": 0.005889892578125,
      "learning_rate": 3.6033039449379077e-07,
      "loss": 0.2309,
      "num_input_tokens_seen": 38290736,
      "step": 181440
    },
    {
      "epoch": 19.96094609460946,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 3.504181053409772e-07,
      "loss": 0.2314,
      "num_input_tokens_seen": 38291856,
      "step": 181445
    },
    {
      "epoch": 19.961496149614963,
      "grad_norm": 0.010986328125,
      "learning_rate": 3.4064406076228515e-07,
      "loss": 0.233,
      "num_input_tokens_seen": 38292944,
      "step": 181450
    },
    {
      "epoch": 19.962046204620464,
      "grad_norm": 0.005859375,
      "learning_rate": 3.310082608443121e-07,
      "loss": 0.2314,
      "num_input_tokens_seen": 38293968,
      "step": 181455
    },
    {
      "epoch": 19.96259625962596,
      "grad_norm": 0.00152587890625,
      "learning_rate": 3.215107056803168e-07,
      "loss": 0.2314,
      "num_input_tokens_seen": 38295120,
      "step": 181460
    },
    {
      "epoch": 19.963146314631462,
      "grad_norm": 0.00567626953125,
      "learning_rate": 3.1215139535356593e-07,
      "loss": 0.2335,
      "num_input_tokens_seen": 38296144,
      "step": 181465
    },
    {
      "epoch": 19.963696369636963,
      "grad_norm": 0.00116729736328125,
      "learning_rate": 3.0293032995232227e-07,
      "loss": 0.2324,
      "num_input_tokens_seen": 38297200,
      "step": 181470
    },
    {
      "epoch": 19.964246424642464,
      "grad_norm": 0.005462646484375,
      "learning_rate": 2.9384750956151785e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38298192,
      "step": 181475
    },
    {
      "epoch": 19.964796479647966,
      "grad_norm": 0.001953125,
      "learning_rate": 2.8490293426441935e-07,
      "loss": 0.2303,
      "num_input_tokens_seen": 38299248,
      "step": 181480
    },
    {
      "epoch": 19.965346534653467,
      "grad_norm": 0.0014801025390625,
      "learning_rate": 2.760966041442936e-07,
      "loss": 0.2308,
      "num_input_tokens_seen": 38300272,
      "step": 181485
    },
    {
      "epoch": 19.965896589658964,
      "grad_norm": 0.005523681640625,
      "learning_rate": 2.674285192827419e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38301264,
      "step": 181490
    },
    {
      "epoch": 19.966446644664465,
      "grad_norm": 0.005218505859375,
      "learning_rate": 2.5889867975803503e-07,
      "loss": 0.2309,
      "num_input_tokens_seen": 38302352,
      "step": 181495
    },
    {
      "epoch": 19.966996699669966,
      "grad_norm": 0.00110626220703125,
      "learning_rate": 2.505070856484437e-07,
      "loss": 0.2303,
      "num_input_tokens_seen": 38303408,
      "step": 181500
    },
    {
      "epoch": 19.967546754675467,
      "grad_norm": 0.00127410888671875,
      "learning_rate": 2.4225373703390397e-07,
      "loss": 0.2298,
      "num_input_tokens_seen": 38304496,
      "step": 181505
    },
    {
      "epoch": 19.96809680968097,
      "grad_norm": 0.005615234375,
      "learning_rate": 2.3413863398769052e-07,
      "loss": 0.2303,
      "num_input_tokens_seen": 38305552,
      "step": 181510
    },
    {
      "epoch": 19.96864686468647,
      "grad_norm": 0.00604248046875,
      "learning_rate": 2.261617765864088e-07,
      "loss": 0.234,
      "num_input_tokens_seen": 38306608,
      "step": 181515
    },
    {
      "epoch": 19.96919691969197,
      "grad_norm": 0.005615234375,
      "learning_rate": 2.1832316490333348e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38307600,
      "step": 181520
    },
    {
      "epoch": 19.96974697469747,
      "grad_norm": 0.0012664794921875,
      "learning_rate": 2.1062279901007395e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38308656,
      "step": 181525
    },
    {
      "epoch": 19.97029702970297,
      "grad_norm": 0.0054931640625,
      "learning_rate": 2.0306067897657432e-07,
      "loss": 0.2314,
      "num_input_tokens_seen": 38309648,
      "step": 181530
    },
    {
      "epoch": 19.97084708470847,
      "grad_norm": 0.005523681640625,
      "learning_rate": 1.9563680487610923e-07,
      "loss": 0.2314,
      "num_input_tokens_seen": 38310704,
      "step": 181535
    },
    {
      "epoch": 19.97139713971397,
      "grad_norm": 0.005767822265625,
      "learning_rate": 1.8835117677362676e-07,
      "loss": 0.233,
      "num_input_tokens_seen": 38311792,
      "step": 181540
    },
    {
      "epoch": 19.971947194719473,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 1.8120379473740566e-07,
      "loss": 0.2308,
      "num_input_tokens_seen": 38312880,
      "step": 181545
    },
    {
      "epoch": 19.972497249724974,
      "grad_norm": 0.0108642578125,
      "learning_rate": 1.7419465883405927e-07,
      "loss": 0.2288,
      "num_input_tokens_seen": 38313872,
      "step": 181550
    },
    {
      "epoch": 19.97304730473047,
      "grad_norm": 0.005462646484375,
      "learning_rate": 1.673237691268703e-07,
      "loss": 0.233,
      "num_input_tokens_seen": 38314992,
      "step": 181555
    },
    {
      "epoch": 19.973597359735972,
      "grad_norm": 0.00555419921875,
      "learning_rate": 1.605911256807868e-07,
      "loss": 0.2293,
      "num_input_tokens_seen": 38316016,
      "step": 181560
    },
    {
      "epoch": 19.974147414741473,
      "grad_norm": 0.0057373046875,
      "learning_rate": 1.5399672855576086e-07,
      "loss": 0.2309,
      "num_input_tokens_seen": 38317072,
      "step": 181565
    },
    {
      "epoch": 19.974697469746975,
      "grad_norm": 0.00147247314453125,
      "learning_rate": 1.4754057781340977e-07,
      "loss": 0.2314,
      "num_input_tokens_seen": 38318128,
      "step": 181570
    },
    {
      "epoch": 19.975247524752476,
      "grad_norm": 0.01080322265625,
      "learning_rate": 1.41222673515351e-07,
      "loss": 0.2309,
      "num_input_tokens_seen": 38319184,
      "step": 181575
    },
    {
      "epoch": 19.975797579757977,
      "grad_norm": 0.00142669677734375,
      "learning_rate": 1.3504301571654054e-07,
      "loss": 0.2324,
      "num_input_tokens_seen": 38320208,
      "step": 181580
    },
    {
      "epoch": 19.976347634763478,
      "grad_norm": 0.00537109375,
      "learning_rate": 1.2900160447693042e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38321296,
      "step": 181585
    },
    {
      "epoch": 19.976897689768975,
      "grad_norm": 0.0022735595703125,
      "learning_rate": 1.2309843984981139e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38322320,
      "step": 181590
    },
    {
      "epoch": 19.977447744774476,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 1.1733352189013945e-07,
      "loss": 0.2319,
      "num_input_tokens_seen": 38323440,
      "step": 181595
    },
    {
      "epoch": 19.977997799779978,
      "grad_norm": 0.00555419921875,
      "learning_rate": 1.1170685065120533e-07,
      "loss": 0.2309,
      "num_input_tokens_seen": 38324464,
      "step": 181600
    },
    {
      "epoch": 19.97854785478548,
      "grad_norm": 0.005859375,
      "learning_rate": 1.0621842618463439e-07,
      "loss": 0.2324,
      "num_input_tokens_seen": 38325520,
      "step": 181605
    },
    {
      "epoch": 19.97909790979098,
      "grad_norm": 0.002716064453125,
      "learning_rate": 1.00868248542052e-07,
      "loss": 0.2303,
      "num_input_tokens_seen": 38326576,
      "step": 181610
    },
    {
      "epoch": 19.97964796479648,
      "grad_norm": 0.005584716796875,
      "learning_rate": 9.56563177734182e-08,
      "loss": 0.2304,
      "num_input_tokens_seen": 38327664,
      "step": 181615
    },
    {
      "epoch": 19.980198019801982,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 9.058263392369703e-08,
      "loss": 0.2308,
      "num_input_tokens_seen": 38328688,
      "step": 181620
    },
    {
      "epoch": 19.98074807480748,
      "grad_norm": 0.0054931640625,
      "learning_rate": 8.564719704118318e-08,
      "loss": 0.2314,
      "num_input_tokens_seen": 38329744,
      "step": 181625
    },
    {
      "epoch": 19.98129812981298,
      "grad_norm": 0.01104736328125,
      "learning_rate": 8.085000717250601e-08,
      "loss": 0.2345,
      "num_input_tokens_seen": 38330832,
      "step": 181630
    },
    {
      "epoch": 19.98184818481848,
      "grad_norm": 0.00148773193359375,
      "learning_rate": 7.619106436096423e-08,
      "loss": 0.2298,
      "num_input_tokens_seen": 38331856,
      "step": 181635
    },
    {
      "epoch": 19.982398239823983,
      "grad_norm": 0.0054931640625,
      "learning_rate": 7.167036864985655e-08,
      "loss": 0.2314,
      "num_input_tokens_seen": 38332880,
      "step": 181640
    },
    {
      "epoch": 19.982948294829484,
      "grad_norm": 0.001678466796875,
      "learning_rate": 6.728792007915097e-08,
      "loss": 0.2319,
      "num_input_tokens_seen": 38334032,
      "step": 181645
    },
    {
      "epoch": 19.983498349834985,
      "grad_norm": 0.00555419921875,
      "learning_rate": 6.30437186921462e-08,
      "loss": 0.2319,
      "num_input_tokens_seen": 38335024,
      "step": 181650
    },
    {
      "epoch": 19.984048404840483,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 5.8937764525479605e-08,
      "loss": 0.2314,
      "num_input_tokens_seen": 38336112,
      "step": 181655
    },
    {
      "epoch": 19.984598459845984,
      "grad_norm": 0.0059814453125,
      "learning_rate": 5.497005761745388e-08,
      "loss": 0.2329,
      "num_input_tokens_seen": 38337104,
      "step": 181660
    },
    {
      "epoch": 19.985148514851485,
      "grad_norm": 0.005584716796875,
      "learning_rate": 5.114059800637172e-08,
      "loss": 0.2303,
      "num_input_tokens_seen": 38338128,
      "step": 181665
    },
    {
      "epoch": 19.985698569856986,
      "grad_norm": 0.001373291015625,
      "learning_rate": 4.7449385725539806e-08,
      "loss": 0.2329,
      "num_input_tokens_seen": 38339152,
      "step": 181670
    },
    {
      "epoch": 19.986248624862487,
      "grad_norm": 0.005767822265625,
      "learning_rate": 4.389642080993017e-08,
      "loss": 0.2303,
      "num_input_tokens_seen": 38340176,
      "step": 181675
    },
    {
      "epoch": 19.986798679867988,
      "grad_norm": 0.005584716796875,
      "learning_rate": 4.04817032928495e-08,
      "loss": 0.2309,
      "num_input_tokens_seen": 38341264,
      "step": 181680
    },
    {
      "epoch": 19.98734873487349,
      "grad_norm": 0.00537109375,
      "learning_rate": 3.720523320260849e-08,
      "loss": 0.2298,
      "num_input_tokens_seen": 38342320,
      "step": 181685
    },
    {
      "epoch": 19.987898789878987,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 3.406701057417916e-08,
      "loss": 0.2314,
      "num_input_tokens_seen": 38343344,
      "step": 181690
    },
    {
      "epoch": 19.988448844884488,
      "grad_norm": 0.005523681640625,
      "learning_rate": 3.106703543254152e-08,
      "loss": 0.2314,
      "num_input_tokens_seen": 38344400,
      "step": 181695
    },
    {
      "epoch": 19.98899889988999,
      "grad_norm": 0.0016021728515625,
      "learning_rate": 2.820530780767161e-08,
      "loss": 0.2324,
      "num_input_tokens_seen": 38345488,
      "step": 181700
    },
    {
      "epoch": 19.98954895489549,
      "grad_norm": 0.00604248046875,
      "learning_rate": 2.5481827724549433e-08,
      "loss": 0.2308,
      "num_input_tokens_seen": 38346512,
      "step": 181705
    },
    {
      "epoch": 19.99009900990099,
      "grad_norm": 0.00122833251953125,
      "learning_rate": 2.2896595209820348e-08,
      "loss": 0.2303,
      "num_input_tokens_seen": 38347600,
      "step": 181710
    },
    {
      "epoch": 19.990649064906492,
      "grad_norm": 0.00183868408203125,
      "learning_rate": 2.04496102851337e-08,
      "loss": 0.2324,
      "num_input_tokens_seen": 38348624,
      "step": 181715
    },
    {
      "epoch": 19.99119911991199,
      "grad_norm": 0.005615234375,
      "learning_rate": 1.814087297546951e-08,
      "loss": 0.2308,
      "num_input_tokens_seen": 38349744,
      "step": 181720
    },
    {
      "epoch": 19.99174917491749,
      "grad_norm": 0.0020751953125,
      "learning_rate": 1.5970383300811797e-08,
      "loss": 0.2314,
      "num_input_tokens_seen": 38350864,
      "step": 181725
    },
    {
      "epoch": 19.992299229922992,
      "grad_norm": 0.00138092041015625,
      "learning_rate": 1.3938141279479231e-08,
      "loss": 0.2345,
      "num_input_tokens_seen": 38351888,
      "step": 181730
    },
    {
      "epoch": 19.992849284928493,
      "grad_norm": 0.00616455078125,
      "learning_rate": 1.204414693312117e-08,
      "loss": 0.2319,
      "num_input_tokens_seen": 38352880,
      "step": 181735
    },
    {
      "epoch": 19.993399339933994,
      "grad_norm": 0.01104736328125,
      "learning_rate": 1.0288400278390951e-08,
      "loss": 0.2304,
      "num_input_tokens_seen": 38354000,
      "step": 181740
    },
    {
      "epoch": 19.993949394939495,
      "grad_norm": 0.00592041015625,
      "learning_rate": 8.670901331941927e-09,
      "loss": 0.2335,
      "num_input_tokens_seen": 38355088,
      "step": 181745
    },
    {
      "epoch": 19.994499449944996,
      "grad_norm": 0.0111083984375,
      "learning_rate": 7.19165010709677e-09,
      "loss": 0.2324,
      "num_input_tokens_seen": 38356176,
      "step": 181750
    },
    {
      "epoch": 19.995049504950494,
      "grad_norm": 0.000705718994140625,
      "learning_rate": 5.850646617178157e-09,
      "loss": 0.2314,
      "num_input_tokens_seen": 38357264,
      "step": 181755
    },
    {
      "epoch": 19.995599559955995,
      "grad_norm": 0.0057373046875,
      "learning_rate": 4.647890877174099e-09,
      "loss": 0.2319,
      "num_input_tokens_seen": 38358352,
      "step": 181760
    },
    {
      "epoch": 19.996149614961496,
      "grad_norm": 0.00162506103515625,
      "learning_rate": 3.5833828954112687e-09,
      "loss": 0.2303,
      "num_input_tokens_seen": 38359440,
      "step": 181765
    },
    {
      "epoch": 19.996699669966997,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 2.6571226835470083e-09,
      "loss": 0.2298,
      "num_input_tokens_seen": 38360496,
      "step": 181770
    },
    {
      "epoch": 19.997249724972498,
      "grad_norm": 0.00555419921875,
      "learning_rate": 1.8691102499079903e-09,
      "loss": 0.2308,
      "num_input_tokens_seen": 38361520,
      "step": 181775
    },
    {
      "epoch": 19.997799779978,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 1.2193456011555525e-09,
      "loss": 0.2313,
      "num_input_tokens_seen": 38362544,
      "step": 181780
    },
    {
      "epoch": 19.998349834983497,
      "grad_norm": 0.005767822265625,
      "learning_rate": 7.078287422856988e-10,
      "loss": 0.2309,
      "num_input_tokens_seen": 38363632,
      "step": 181785
    },
    {
      "epoch": 19.998899889988998,
      "grad_norm": 0.0017852783203125,
      "learning_rate": 3.345596799597672e-10,
      "loss": 0.2319,
      "num_input_tokens_seen": 38364656,
      "step": 181790
    },
    {
      "epoch": 19.9994499449945,
      "grad_norm": 0.0054931640625,
      "learning_rate": 9.953841750842684e-11,
      "loss": 0.2309,
      "num_input_tokens_seen": 38365680,
      "step": 181795
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.011962890625,
      "learning_rate": 2.764956597012258e-12,
      "loss": 0.2319,
      "num_input_tokens_seen": 38366624,
      "step": 181800
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.23131532967090607,
      "eval_runtime": 60.4483,
      "eval_samples_per_second": 66.834,
      "eval_steps_per_second": 16.708,
      "num_input_tokens_seen": 38366624,
      "step": 181800
    },
    {
      "epoch": 20.0,
      "num_input_tokens_seen": 38366624,
      "step": 181800,
      "total_flos": 1.727631738233684e+18,
      "train_loss": 0.23324829005598366,
      "train_runtime": 25757.3665,
      "train_samples_per_second": 28.231,
      "train_steps_per_second": 7.058
    }
  ],
  "logging_steps": 5,
  "max_steps": 181800,
  "num_input_tokens_seen": 38366624,
  "num_train_epochs": 20,
  "save_steps": 9090,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.727631738233684e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}